Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Host plant variation and population limitation of two introduced insects Morrison, Peter D. S. 1986

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

Item Metadata

Download

Media
831-UBC_1987_A1 M67.pdf [ 9.06MB ]
Metadata
JSON: 831-1.0097476.json
JSON-LD: 831-1.0097476-ld.json
RDF/XML (Pretty): 831-1.0097476-rdf.xml
RDF/JSON: 831-1.0097476-rdf.json
Turtle: 831-1.0097476-turtle.txt
N-Triples: 831-1.0097476-rdf-ntriples.txt
Original Record: 831-1.0097476-source.json
Full Text
831-1.0097476-fulltext.txt
Citation
831-1.0097476.ris

Full Text

HOST PLANT VARIATION AND POPULATION LIMITATION OF TWO INTRODUCED INSECTS by PETER D.S. MORRISON B.S . Stanford Univers i ty 1978 A THESIS SUBMITTED IN PARTIAL FULFILLMENT THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Zoology) We accept th i s thes is as conforming to th required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1986 @ Peter D .S . Morrison, 1986 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of 7 n n - j n z v  The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date February 22, 1987 r>F-fin/ft-n A B S T R A C T T h e r e s p o n s e t o h o s t p l a n t v a r i a t i o n s h a p e s t h e l o n g - t e r m s u c c e s s o f p h y t o p h a g o u s i n s e c t s . T w o g a l l - f o r m i n g t e p h r i t i d f l i e s , U r o p h o r a a f f i n i s a n d U . q u a d r i f a s c i a t a , o v i p o s i t i n f l o w e r b u d s o f C e n t a u r e a d i f f u s a a n d C . m a c u l o s a ( A s t e r a c e a e ) . F e m a l e s o f b o t h f l y s p e c i e s c h o s e a m o n g p l a n t s , a m o n g g r o u p s o f b u d s o n p l a n t s , a n d a m o n g b u d s . A m o n g p l a n t c h o i c e s w e r e c o r r e l a t e d w i t h b u d s p e r p l a n t . A m o n g b u d c h o i c e s c o r r e s p o n d e d t o l a r v a l d e v e l o p m e n t a l r e q u i r e m e n t s . I n s e c t a t t a c k l e d t o g a l l f o r m a t i o n , b u d a b o r t i o n , a n d r e d u c e d s e e d p r o d u c t i o n . B u d a b o r t i o n , c a u s e d b y p r o b i n g f e m a l e s , l i m i t e d g a l l d e n s i t i e s . I n c r e a s e d d e n s i t i e s o f U . a f f i n i s f e m a l e s r e l a t i v e t o o v i p o s i t i o n s i t e s l e d t o m o r e U . a f f i n i s g a l l s , i n c r e a s e d b u d a b o r t i o n , f e w e r U . q u a d r i f a s c i a t a g a l l s , a n d f e w e r s e e d s . A t e m p o r a l r e f u g e f o r s e e d p r o d u c t i o n w a s o b s e r v e d . P l a n t s c o m p e n s a t e d o n l y s l i g h t l y f o r a b o r t e d b u d s . B u d a b o r t i o n m a y i n c r e a s e t h e s e a r c h t i m e b e t w e e n s u c c e s s f u l o v i p o s i t i o n s . A s i m u l a t i o n m o d e l b a s e d o n t h i s p r e m i s e i m p l i e d t h a t b u d a b o r t i o n m a y d r a m a t i c a l l y r e d u c e t o t a l g a l l f o r m a t i o n . P l a n t q u a l i t y w a s m a n i p u l a t e d i n a n a t t e m p t t o s h i f t t h r e e p o p u l a t i o n l i m i t i n g f a c t o r s . P l a n t s r e s p o n d e d t o f e r t i l i z a t i o n a n d w a t e r i n g w i t h a n i n c r e a s e i n b u d n u m b e r s . E x c e p t f o r t w o y e a r - s i t e - t r e a t m e n t c o m b i n a t i o n s , g a l l s p e r d e v e l o p e d b u d d i d n o t d i f f e r s i g n i f i c a n t l y b e t w e e n t r e a t m e n t s . T r e a t e d p l a n t s d i d n o t d i f f e r i n t h e i r p r o p e n s i t y t o a b o r t b u d s . U . a f f i n i s l a r v a e d e v e l o p e d f a s t e r i n f e r t i l i z e d p l a n t s . Among year comparisons showed that the d e n s i t y of buds a v a i l a b l e f o r o v i p o s i t i o n was l i m i t e d by p r e c i p i t a t i o n , non-random i n s e c t a t t a c k , and, i n the longer term, by the r e d u c t i o n i n seed p r o d u c t i o n due to f l y a t t a c k . Bud d e n s i t i e s , i n t u r n , l i m i t e d g a l l d e n s i t i e s . iv TABLE OF CONTENTS ABSTRACT 11 LIST OF TABLES ix LIST OF FIGURES x i i i ACKNOWLEDGEMENTS X V INTRODUCTION 1 ORGANISMS 4 Plants 4 Insects 5 GENERAL METHODS 8 Study s i tes 8 Bud descr ipt ions 11 S t a t i s t i c a l methods 14 I . THE EFFECT OF HOST SELECTION ON THE POPULATION DYNAMICS OF TWO INTRODUCED INSECTS 16 MATERIALS AND METHODS 19 Observation methods 19 Plant c o l l e c t i o n s 21 C a l o r i f i c content 21 RESULTS 23 Var ia t ion in resources 23 Insect choice 23 Among plants 23 Among buds on plants 27 Consequences of insect choice 32 Among plants 32 Among buds on plants 40 DISCUSSION 47 V a r i a t i o n in bud product iv i ty 47 Bud abort ion and population l i m i t a t i o n 48 Basis for choice 49 Insect in terac t ions 51 Summary 52 I I . THE EFFECT OF TIMING OF ATTACK ON THE POPULATION DYNAMICS OF TWO INTRODUCED INSECTS 53 MATERIALS AND METHODS 57 Observation methods 57 Plant c o l l e c t i o n s 58 Insect density manipulation 59 RESULTS 62 Insect attack and bud i n i t i a t i o n 62 Insect attack 62 Bud i n i t i a t i o n 65 Interact ion 65 Changes in plant a l l o c a t i o n 77 Bud growth and development 77 Compensatory reproduction 77 Insect density manipulation 81 DISCUSSION 85 Changes in insect density 85 Seed refuge 87 Compensatory reproduction 88 v i Evolut ionary consequences 89 Summary 91 APPENDIX IIA. EFFECT OF COLLECTION DATE 93 APPENDIX IIB. EFFECT OF DENSITY ENCLOSURES 95 I I I . BUD ABORTION AND POPULATION LIMITATION OF UROPHORA AFFINIS (DIPTERA: TEPHRITIDAE) IN BRITISH COLUMBIA 97 MATERIALS AND METHODS 100 Bud c o l l e c t i o n and d i s sec t ion 100 RESULTS 102 DISCUSSION 104 Ovipos i t ion behaviour 104 Search time between ov ipos i t ions 105 Model formulation 107 Model re su l t s 109 Summary 114 APPENDIX IIIA. LISTING OF THE NUMERICAL MODEL 116 IV. PLANT QUALITY AND THE POPULATION DYNAMICS OF TWO INTRODUCED INSECTS 120 MATERIALS AND METHODS 123 Weather 123 Experimental treatments in 1 979 123 Nutrient analys i s 125 Experimental treatments in 1980 128 Observation of insects in 1979 129 Observation of insects in 1980 132 Plant c o l l e c t i o n s and d i s sec t ion 132 Addi t iona l s t a t i s t i c a l methods 134 v i i RESULTS 135 Response of p l a n t s 135 Response to p l a n t s 139 G a l l f l i e s 139 Other h e r b i v o r e s 145 Change i n i n t e r a c t i o n 146 Attac k l e v e l s 1979 146 Attac k l e v e l s 1980 149 L a r v a l s u r v i v a l and development 152 DISCUSSION 154 Response of p l a n t s 154 Response to p l a n t s 154 G a l l f l i e s 154 Other h e r b i v o r e s 155 Change i n i n t e r a c t i o n 156 E f f e c t of bud a b o r t i o n 156 L a r v a l s u r v i v a l and development 158 E f f e c t of p l a n t q u a l i t y on p o p u l a t i o n dynamics 159 F e r t i l i z a t i o n as a management t o o l 160 Summary 161 V. POPULATION LIMITATION OF TWO INTRODUCED INSECTS: PROCESSES WITHIN AND BETWEEN YEARS 163 MATERIALS AND METHODS 166 Weather 166 Pl a n t c o l l e c t i o n s in 1979 166 Pl a n t c o l l e c t i o n s i n 1980 167 Pl a n t c o l l e c t i o n s i n 1981 168 v i i i Plant d i s sec t ions 168 RESULTS 169 Among year di f ferences 169 Effect of r a i n f a l l 173 DISCUSSION 176 Bud density 176 Effect of bud density on g a l l densi ty 182 G a l l d i s t r i b u t i o n s 183 Interact ion between g a l l f l y species 187 Summary 193 APPENDIX VA. ESTIMATION OF BUD AVAILABILITY 194 Methods 194 Results 195 CONCLUDING DISCUSSION 202 LITERATURE CITED 207 i x LIST OF TABLES Table 1.1 C a l o r i f i c content of developed d i f fuse knapweed buds by branching category 24 Table 1.2 Observed and predicted d i s t r i b u t i o n s of U. a f f i n i s (UA) and U. quadri fasc iata (UQ) adults among branching categories 30 Table 1.3 D i s t r i b u t i o n of U. a f f i n i s (UA) and U. quadr i fasc ia ta (UQ) adults by s ize of d i f fuse knapweed buds 31 Table 2.1 Dates on which buds were f i r s t observed and corresponding bud i n i t i a t i o n categories 58 Table 2.2 Mean day of observation for d i f f erent categories of Urophora f l i e s on di f fuse knapweed p lant s , Robertson's 1980 62 Table 2.3 Counts of Urophora f l i e s observed in density enclosures , Robertson's 1980 81 Table 2.4 Ef fec t of g a l l f l y density manipulations on d i f fuse knapweed c h a r a c t e r i s t i c s , g a l l product ion, seed product ion , and bud abortion 82 Table 2.5 E f f e c t of plant c o l l e c t i o n date on d i f fuse knapweed c h a r a c t e r i s t i c s and g a l l f l y a t tack , Robertson's 1980 94 Table 2.6 Ef fec t of Urophora enclosures on d i f fuse knapweed c h a r a c t e r i s t i c s and g a l l f l y attack 95 Table 3.1 Urophora eggs and larvae and proportion of buds X aborted f o r t e r m i n a l buds of d i f f u s e knapweed, Robertson's 1980 102 Table 3.2 D u r a t i o n of p r o b i n g i n t o s p o t t e d knapweed buds by female U. a f f i n i s 106 Table 3.3 L i s t i n g of the numerical model 117 Table 4.1 S o i l c h a r a c t e r i s t i c s at the study s i t e s 128 Table 4.2 E f f e c t of f e r t i l i z a t i o n and watering on the t o t a l number of d i f f u s e knapweed buds and the number of developed buds, 1979 135 Table 4.3 E f f e c t of f e r t i l i z a t i o n and watering on d i f f u s e knapweed c h a r a c t e r i s t i c s and i n s e c t a t t a c k , Ned's Creek 1979 136 Table 4.4 E f f e c t of f e r t i l i z a t i o n and watering on d i f f u s e knapweed c h a r a c t e r i s t i c s and i n s e c t a t t a c k , Robertson's 1979 137 Table 4.5 E f f e c t of f e r t i l i z a t i o n and watering on d i f f u s e knapweed c h a r a c t e r i s t i c s and i n s e c t a t t a c k , Robertson's 1980 139 Table 4.6 E f f e c t of f e r t i l i z a t i o n and watering on the t o t a l number of s p o t t e d knapweed buds and the number of developed buds, 1979 140 Table 4.7 E f f e c t of f e r t i l i z a t i o n and watering on spotted knapweed c h a r a c t e r i s t i c s and i n s e c t a t t a c k , Chase 1979 .141 Table 4.8 E f f e c t of f e r t i l i z a t i o n and watering on spotted knapweed c h a r a c t e r i s t i c s and i n s e c t a t t a c k , Chase 1980 .144 Table 4.9 E f f e c t of f e r t i l i z a t i o n and watering on the p r o p o r t i o n of knapweed buds chewed, 1979 146 xi Table 4.10 Ef fec t of f e r t i l i z a t i o n and watering on the proport ion of knapweed buds aborted, 1979 147 Table 4.11 Ef fec t of f e r t i l i z a t i o n and watering on the number of U. a f f i n i s g a l l s per developed bud, 1979 148 Table 4.12 Ef fec t of f e r t i l i z a t i o n and watering on the proport ion of knapweed buds developed, 1979 149 Table 4.13 Ef fec t of f e r t i l i z a t i o n and watering on the number of U. quadri fasc iata g a l l s per developed bud, 1979 150 Table 4.14 Contents of Urophora g a l l s from contro l and treated d i f fuse knapweed p lants , Robertson's 1980 152 Table 5.1 Dif fuse knapweed c h a r a c t e r i s t i c s and Urophora attack l e v e l s , Ned's Creek 1979-1980 169 Table 5.2 Diffuse knapweed c h a r a c t e r i s t i c s and Urophora attack l e v e l s , Robertson's 1979-1981 171 Table 5.3 Spotted knapweed c h a r a c t e r i s t i c s and Urophora attack l e v e l s , Chase 1979-1981 172 Table 5.4 Urophora g a l l s per developed bud at the three study s i t e s , 1973-1981 184 Table 5.5 Proportion of d i f fuse knapweed buds unattacked and estimated proportion of buds unavai lable to o v i p o s i t i n g g a l l f l i e s , Ned's Creek 1973-1980 198 Table 5.6 Proportion of d i f fuse knapweed buds unattacked and estimated proportion of buds unavai lable to o v i p o s i t i n g g a l l f l i e s , Robertson's 1977-1981 199 Table 5.7 Proportion of spotted knapweed buds unattacked and estimated proportion of buds unavai lable to x i i o v i p o s i t i n g g a l l f l i e s , C h a s e 1 9 7 3 - 1 9 8 1 2 0 0 x i i i LIST OF FIGURES F i g u r e 0.1 L o c a t i o n of the study s i t e s 9 F i g u r e 0.2 Numbering scheme f o r knapweed buds 12 F i g u r e 1.1 D i s t r i b u t i o n of both s p e c i e s of g a l l f l i e s among d i f f u s e knapweed p l a n t s 25 F i g u r e 1.2 T o t a l g a l l f l i e s and t o t a l buds on d i f f u s e knapweed p l a n t s 28 F i g u r e 1.3 T o t a l U. a f f i n i s a d u l t s and U. a f f i n i s g a l l s on d i f f u s e knapweed p l a n t s 34 F i g u r e 1.4 T o t a l U. quadri fasc i a t a a d u l t s and U. quadr i fasc i a t a g a l l s on d i f f u s e knapweed p l a n t s 36 F i g u r e 1.5 Mean U. a f f i n i s g a l l s per developed bud and p r o p o r t i o n of buds aborted 38 F i g u r e 1.6 D i s t r i b u t i o n of g a l l s and seeds per developed bud a c r o s s branching c a t e g o r i e s 41 F i g u r e 1.7 D i s t r i b u t i o n of bud f a t e s a c r o s s branching c a t e g o r i e s 43 F i g u r e 2.1 Counts on d i f f u s e knapweed p l a n t s of the two s p e c i e s of g a l l f l y at Robertson's in 1980 63 F i g u r e 2.2 Average numbers of buds on d i f f u s e knapweed p l a n t s at Robertson's in 1980 66 F i g u r e 2.3 Mean number of U. a f f i n i s g a l l s per developed bud by bud i n i t i a t i o n category f o r Robertson's i n 1980 . 68 F i g u r e 2.4 P r o b a b i l i t y of bud a b o r t i o n by bud i n i t i a t i o n c a tegory for Robertson's i n 1980 70 x i v F i g u r e 2.5 Mean number of U. q u a d r i f a s c i a t a g a l l s per developed bud by bud i n i t i a t i o n category f o r Robertson's i n 1980 72 F i g u r e 2.6 Mean number of seeds per developed bud by bud i n i t i a t i o n c a t e g o r y f o r Robertson's i n 1980 74 F i g u r e 2.7 D i s t r i b u t i o n of phenology measures across branching c a t e g o r i e s 78 F i g u r e 3.1 Sample output from the simple model of the i n t e r a c t ion 110 F i g u r e 3.2 E f f e c t of v a r y i n g probes per bud i n the numerical model 112 F i g u r e 4.1 Watering and n a t u r a l p r e c i p i t a t i o n i n 1979 126 F i g u r e 4.2 Watering and n a t u r a l p r e c i p i t a t i o n i n 1980 130 F i g u r e 4.3 T o t a l g a l l f l i e s observed per p l a n t for t r e a t e d and c o n t r o l d i f f u s e knapweed p l a n t s at Robertson's i n 1 980 1 42 F i g u r e 5.1 E f f e c t of r a i n f a l l (June 10-July 27) on buds per d i f f u s e knapweed p l a n t 174 F i g u r e 5.2 D e n s i t i e s of d i f f u s e knapweed buds and t o t a l Urophora g a l l s at Ned's Creek 1972-1979 178 F i g u r e 5.3 D e n s i t i e s of s p o t t e d knapweed buds and t o t a l Urophora g a l l s at Chase 1971-1979 180 F i g u r e 5.4 Changes i n the p r o p o r t i o n of U. a f f i n i s g a l l s of a l l Urophora g a l l s 190 F i g u r e 5.5 Observed d i s t r i b u t i o n of U. a f f i n i s g a l l s i n developed buds at Ned's Creek and the f i t t e d d i s t r i b u t i o n s 196 XV ACKNOWLEDGEMENTS Any undertaking such as t h i s one i s not done i n i s o l a t i o n , but b u i l d s on the work of other s c i e n t i s t s and r e l i e s on the i n t e l l e c t u a l and emotional support from many people. In p a r t i c u l a r , I would l i k e to thank my s u p e r v i s o r , Judy Myers, f o r f i l l i n g both these r o l e s and f o r s u f f e r i n g the d e l a y s and d i f f i c u l t i e s with p a t i e n c e and good cheer. Denis Berube and Peter H a r r i s were generous with t h e i r i d e a s , comments, and unpublished data. C h a r l e s Krebs, Don Ludwig, Jeremy McN e i l , Tom Northcote, Michael P i t t , Peter P r i c e , Geoff Scudder, Tony S i n c l a i r , Donald Strong, Roy T u r k i n g t o n , and Norman Wilimovsky p r o v i d e d encouragement, u s e f u l c r i t i c i s m s , and c o n s t r u c t i v e suggestions at v a r i o u s stages i n the genesis of t h i s t h e s i s . S p e c i a l thanks to Dorothy Chen, Jean Hnytka, C h r i s t i n e L a i , and E l l e n Randlesome for h e l p i n g with the mind numbing task of d i s s e c t i n g l a r g e numbers of knapweed heads while p u t t i n g up with my i n s i s t e n c e on p r e c i s i o n and completeness. T h i s r e s e a r c h was supported in part by an A g r i c u l t u r e Canada c o n t r a c t to Judy Myers (No. 0SU79.-00093). The people who p r o v i d e d support and i n s p i r a t i o n to me d u r i n g the l a s t s e v e r a l years c o n t r i b u t e d to the completion of t h i s t h e s i s i n ways too numerous to d e t a i l and with such l i t t l e reward: Alex B r e t t , J a n i Burns, Mag C l a y t o n , Linda Edwards, L i z F u r n i s s , Nancy Hawkins, Jean K i r k , Ken Lertzman, Cindy Lyon, Sue McCormack, C h a r l i e McDermott, Marcel Marceau, Dave Marmorek, L i z Pope, Rob Powell, Andrew P u r v i s , Laura R i c h a r d s , Jens Roland, Joan Sutherland, and Jane Templeman. Thanks a l s o to the xvi management of Cafe Madeleine who refused to charge me rent and to Glen Armstrong who reminded me that knapweed has p o t e n t i a l as a cure for insomnia. This thes is i s dedicated to the memory of Ann V a l l e e . The s c i e n t i f i c d i s c i p l i n e i s about seeking t ru th , yet Ann's death was a very d i f f i c u l t t ru th to face. The truth in her l i f e w i l l l i v e on. 1 INTRODUCTION S c i e n t i s t s continue to s t r u g g l e to make out the dim o u t l i n e s of the e v o l u t i o n a r y p r o c e s s . I t i s slow going; h a r d l y s u r p r i s i n g , given the enormous time s c a l e , the sheer numbers of organisms, and the r a d i c a l environmental changes f o r which minimal i n f o r m a t i o n e x i s t s . Since Darwin's and Wallace's i n s i g h t s and the d i s c o v e r y of the g e n e t i c b a s i s of e v o l u t i o n a r y change, many qu e s t i o n s about the e v o l u t i o n a r y p r o c e s s have been reduced to a much simpler one: why does one organism do b e t t e r than another? T h i s q u e s t i o n has been asked when a s k i n g why species go e x t i n c t , when comparing s p e c i e s w i t h i n groups, p o p u l a t i o n s w i t h i n a s p e c i e s , or i n d i v i d u a l organisms w i t h i n a populat i o n . The e v o l u t i o n a r y success of a s p e c i e s or a p o p u l a t i o n i s u s u a l l y judged by three r e l a t e d c r i t e r i a . The f i r s t , p o p u l a t i o n s i z e , measures how w e l l the p o p u l a t i o n has done i n the past. The second, d i s t r i b u t i o n , i n d i c a t e s the f l e x i b i l i t y of the sp e c i e s i n d e a l i n g with v a r i a t i o n i n i t s environment. These two s t a t i c measures are roughly c o r r e l a t e d with one another f o r i f a species i s widespread through s e v e r a l h a b i t a t s , i t i s more l i k e l y to have a l a r g e p o p u l a t i o n s i z e (Brown, 1984). The f i r s t two measures are a l s o r e l a t e d to the t h i r d c r i t e r i o n , e v o l u t i o n a r y p e r s i s t e n c e . I f a s p e c i e s i s widespread and has a lar g e p o p u l a t i o n s i z e , i t i s l e s s l i k e l y to go e x t i n c t through s t o c h a s t i c p r o c e s s e s . P e r s i s t e n c e may a l s o be thought of as a wide d i s t r i b u t i o n i n time. 2 With an extant s p e c i e s , i t i s p o s s i b l e to gauge i t s p o t e n t i a l f o r e v o l u t i o n a r y success by examining a f o u r t h c r i t e r i o n , i t s p o p u l a t i o n dynamics in response to v a r i a t i o n i n the environment. If the p o p u l a t i o n s i z e drops s h a r p l y under changed c o n d i t i o n s , the p r o b a b i l i t y of e x t i n c t i o n i s much h i g h e r . On the other hand, i f the p o p u l a t i o n s i z e i s r e l a t i v e l y i n s e n s i t i v e to changes i n the environment, the s p e c i e s i s more l i k e l y to p e r s i s t . P o p u l a t i o n dynamics are o f t e n d e s c r i b e d i n terms of a set of more or l e s s independently a c t i n g f a c t o r s or p r o c e s s e s , such as r e p r o d u c t i o n , d i s p e r s a l , and density-dependent p r e d a t i o n (e.g. V a r l e y and G r a d w e l l , 1960). Those processes which l i m i t the p o p u l a t i o n are c e n t r a l to the response of p o p u l a t i o n s i z e to environmental v a r i a t i o n . To evaluate a p o p u l a t i o n ' s p o t e n t i a l f o r e v o l u t i o n a r y success using the f o u r t h c r i t e r i o n , the tasks f a c i n g the p o p u l a t i o n e c o l o g i s t are to develop a d e s c r i p t i o n of environmental v a r i a t i o n , i s o l a t e the processes l i m i t i n g p o p u l a t i o n s i z e , and then to evaluate how the processes s h i f t with changes in the environment. In s e v e r a l ways, t e r r e s t r i a l weeds are i d e a l organisms with which to study environmental v a r i a t i o n . They are widespread, dense, and sedentary. For each p l a n t , many c h a r a c t e r s may be measured e a s i l y . Thus, i t i s p o s s i b l e to q u a n t i f y v a r i a t i o n i n a d e t a i l e d way over q u i t e broad areas. The response of phytophagous i n s e c t s may be set a g a i n s t t h i s background d e s c r i p t i o n . Systems where i n s e c t s have been i n t r o d u c e d as b i o l o g i c a l c o n t r o l agents (Andres, 1977; Andres 3 and Goeden, 1971; Wilson, 1964) may be p a r t i c u l a r l y good f o r addressing i s s u e s of responses to environmental v a r i a t i o n . Because of the economic importance of weeds, h i s t o r i c a l data on the p l a n t s and t h e i r i n t r o d u c e d consumers are u s u a l l y a v a i l a b l e , f a c i l i t a t i n g i d e n t i f i c a t i o n of processes l i m i t i n g p o p u l a t i o n s i z e . M o d i f i c a t i o n s of the processes to i n c r e a s e the p o p u l a t i o n s i z e of the c o n t r o l agents have c l e a r economic b e n e f i t s . F i n a l l y , c o n s t r a i n t s on ma n i p u l a t i o n of the p l a n t p o p u l a t i o n are r a r e . T h i s t h e s i s uses the f o u r t h c r i t e r i o n of e v o l u t i o n a r y success, p o p u l a t i o n dynamics i n response to environmental v a r i a t i o n , to ev a l u a t e the success of two g a l l - f o r m i n g f l i e s i n troduced as b i o l o g i c a l c o n t r o l agents. F o l l o w i n g a d e s c r i p t i o n of the organisms and some g e n e r a l methods, Chapter I d e s c r i b e s the b e h a v i o u r a l response of the i n s e c t s to v a r i a t i o n w i t h i n and among host p l a n t s , d e t a i l s the major consequences of in s e c t a t t a c k , and focusses on two processes l i m i t i n g the i n s e c t p o p u l a t i o n d e n s i t y . Chapter II examines the e f f e c t of v a r i a t i o n in the ti m i n g of i n s e c t a t t a c k and v a r i a t i o n i n i n s e c t d e n s i t y on the outcome of i n s e c t a t t a c k . Chapter I I I d i s c u s s e s the d e t a i l s of bud a b o r t i o n by the p l a n t s , one of the processes l i m i t i n g the i n s e c t p o p u l a t i o n . Chapter IV d e s c r i b e s the response of the i n s e c t s ' a t t a c k to e x p e r i m e n t a l l y manipulated changes i n p l a n t " q u a l i t y " . Chapter V draws on data c o l l e c t e d during the establishment of the g a l l f l i e s and combines i t with o b s e r v a t i o n s over a three year p e r i o d to l i n k the i n t e r a c t i o n with the p o p u l a t i o n dynamics of the system. 4 ORGANISMS PLANTS Two s p e c i e s of knapweed, d i f f u s e knapweed (Centaurea  d i f f u s a Lam.) and spotted knapweed (C. maculosa Lam.) (Asteraceae) , are economically important rangeland weeds i n B r i t i s h Columbia and the northwestern U n i t e d S t a t e s . A c c i d e n t a l l y imported from E u r a s i a , they are now f i r m l y e s t a b l i s h e d i n North America. A c o n s e r v a t i v e estimate p l a c e d the area i n f e s t e d by these weeds at 1.5 m i l l i o n h e c t a r e s i n the U n i t e d S t a t e s (Maddox, 1979). A p o t e n t i a l 1.1 m i l l i o n h e c t a r e s are s u s c e p t i b l e i n B r i t i s h Columbia alone ( H a r r i s and Cranston, 1979). These t h i s t l e - l i k e members of the Asteraceae t h r i v e i n the s e m i - a r i d i n t e r i o r of B.C. Spotted knapweed i s t y p i c a l l y found i n c o o l e r and wetter areas than d i f f u s e knapweed ( H a r r i s and Cranston, 1979). Frankton and M u l l i g a n (1977) d e s c r i b e d the ranges of the two s p e c i e s . Both s p e c i e s are b i e n n i a l s or s h o r t - l i v e d p e r e n n i a l s . The r e p r o d u c t i v e phase of the p l a n t s u s u a l l y begins i n mid May with b o l t i n g stems. As v e r t i c a l growth slows, buds and t h e i r s u p p o r t i n g l a t e r a l branches are i n i t i a t e d . Buds are i n i t i a t e d from the b e g i n n i n g of June u n t i l the p l a n t s begin f l o w e r i n g i n J u l y . Both s p e c i e s of p l a n t are q u i t e p l a s t i c i n t h e i r response to s o i l c o n d i t i o n s , moisture, and d i s t u r b a n c e . The most v a r i a b l e component of seed p r o d u c t i o n i s the number of flower buds i n i t i a t e d (Roze, 1981; Schirman, 1981; Story, 1978a; Watson, 1972). The two s p e c i e s d i f f e r i n growth form and consequently i n 5 the d i s t r i b u t i o n of seed p r o d u c t i o n . D i f f u s e knapweed u s u a l l y produces more buds than s p o t t e d knapweed (100 vs. 15), but fewer seeds per bud (12 vs. 25) (Watson, 1972). The d i f f e r e n c e i n seed p r o d u c t i o n i s o f f s e t by the a b i l i t y of s p o tted knapweed to b o l t more than once from the same root s t o c k . Roze (1981) concluded that the l i f e t i m e r e p r o d u c t i v e output of i n d i v i d u a l s of the two s p e c i e s i s n e a r l y i d e n t i c a l . The b i o l o g y and taxonomy of the two s p e c i e s were d i s c u s s e d by Frankton and M u l l i g a n (1977), H a r r i s (1980a), Roze (1981), and Watson and Renney (1974). The s p e c i e s of knapweed r e f e r r e d to as s p o t t e d knapweed in B r i t i s h Columbia has been d e s c r i b e d as C. maculosa (Moore and Frankton, 1974), however t h i s may not be e q u i v a l e n t to the European C. maculosa ( H a r r i s , pers. comm.). In B r i t i s h Columbia, the s p e c i e s has 2n=36, while the comparable European s p e c i e s have a lower chromosome count (2n=l8), except C. maculosa spp. micranthos (=C. b i e b e r s t e i n i i spp. b i e b e r s t e i n i i D.C.) which i s found i n the southern U.S.S.R, A l b a n i a , B u l g a r i a , C z e c h o s l o v a k i a , Hungary, and Romania. In order to remain c o n s i s t e n t with the North American usage, I have l e f t the s p e c i e s name in t h i s t h e s i s as C. maculosa. INSECTS Two s p e c i e s of t r u e f r u i t f l y (Urophora a f f i n i s F r f l d . and U. quadr i fasc i a t a (Meig.)) ( D i p t e r a : T e p h r i t i d a e ) were i n t r o d u c e d from Europe as p a r t of a b i o l o g i c a l c o n t r o l program a g a i n s t the knapweeds. Dr. Peter H a r r i s of A g r i c u l t u r e Canada r e l e a s e d f l i e s at four s i t e s i n B r i t i s h Columbia in the e a r l y 1970's: two r e l e a s e s on d i f f u s e knapweed (1970, 1972) and 6 two on s p o t t e d knapweed (1970, 1971). The f l i e s e s t a b l i s h e d s u c c e s s f u l l y and the p o p u l a t i o n s grew r a p i d l y ( H a r r i s , 1980a). The n a t u r a l h i s t o r i e s of the two s p e c i e s of f l y have been d i s c u s s e d i n d e t a i l by Roze (1981) and Zwolfer (1970). A d u l t s emerge i n e a r l y summer, u s u a l l y beginning between June 1 and June 15, two to three weeks a f t e r p u p a t i o n . Males emerge on average one day before the females. When the females emerge they mate almost immediately. O v i p o s i t i o n ensues approximately t h r e e days l a t e r and m u l t i p l e matings are observed. The i n s e c t s appear to spend most of t h e i r a d u l t l i v e s on the knapweed p l a n t s . Females l i v e up to three weeks. The t i m i n g of emergence of the g a l l f l i e s appears to be determined by temperature ( c f . Uvarov, 1931). Roze (1981) observed t h a t the g a l l f l i e s emerged much e a r l i e r i n 1977 than i n 1976. Her o b s e r v a t i o n s , and those of Berube (pers. comm.), are c o n s i s t e n t with temperature-dependent developmental r a t e s (MS i n p r e p . ) . Eggs are l a i d on top of the immature knapweed f l o r e t s . When the eggs hatch, 3-4 days a f t e r o v i p o s i t i o n , the l a r v a e burrow i n t o the f l o r e t s and form g a l l s , presumably by chemical i n d u c t i o n of the p l a n t t i s s u e (Dieleman, 1969; H o r i , 1974; Mani, 1964; Osborne, 1972). More than one g a l l may be formed i n a s i n g l e bud. The l a r v a e feed on the n u t r i t i v e l a y e r i n s i d e the g a l l u n t i l they reach the t h i r d i n s t a r s e v e r a l weeks l a t e r . In a l l c a s e s , the l a r v a e overwinter as d i a p a u s i n g t h i r d i n s t a r l a r v a e i n s i d e the g a l l s . Zwolfer (1970) suggests that the s i z e of the immature 7 tubular f l ore t s is the c r i t i c a l factor in ov ipos i t ion choice . Buds above a c e r t a i n s ize threshold w i l l not receive any eggs, despite being the only substrate a v a i l a b l e to a f e r t i l e female. U. a f f i n i s females caged on a s ing le spotted knapweed plant l a i d 93% of the i r eggs in buds in the 4-6 mm size range (Zwolfer, 1970). Based on greenhouse studies of both spotted and d i f fuse knapweed, Berube and Harr i s (1978) concluded that U. a f f i n i s larvae develop in buds in a s i ze range of 2-8 mm, while U. quadr i fasc ia ta larvae develop in larger buds, 6-10 mm long. In U. q u a d r i f a s c i a t a , a second generation is bel ieved to be obl igate (Berube, pers . comm.). In U. a f f i n i s , the second generation is f a c u l t a t i v e (Zwolfer, 1970), presumably determined by a temperature cue. While r e l a t i v e l y few temperate-zone insects have been shown to use temperature as a primary diapause inducing stimulus (Tauber et ajL. , 1986), temperature is one of the few environmental cues a v a i l a b l e to U. a f f i n i s larvae . Photoperiod is u n l i k e l y to be used by larvae inside woody g a l l s ins ide unopened flower buds. 8 GENERAL METHODS STUDY SITES F i e l d work was done at three study s i t e s , two of which, Ned's C r e e k - P r i t c h a r d and Chase, were o r i g i n a l r e l e a s e s i t e s f o r the g a l l f l i e s on d i f f u s e and s p o t t e d knapweed, r e s p e c t i v e l y . (Ned's C r e e k - P r i t c h a r d i s the r e s u l t of the merging of two separate r e l e a s e s approximately 200 m apa r t and w i l l be r e f e r r e d to as Ned's Creek h e r e a f t e r . ) The t h i r d s i t e , Robertson's, was used as an a d d i t i o n a l d i f f u s e knapweed s i t e i n 1979 and became the primary d i f f u s e knapweed s i t e i n 1980 when Ned's Creek was sprayed with h e r b i c i d e . A l l three s i t e s are on t e r r a c e s above the South Thompson R i v e r between Kamloops and Chase, B r i t i s h Columbia ( F i g u r e 0.1). Chase r e c e i v e s more r a i n d u r i n g the p e r i o d from A p r i l to August (3.5 cm on the average) and i s c o o l e r (approximately 1°C d u r i n g the summer) than Kamloops A i r p o r t , roughly 60 km to the west. The Ned's Creek and Robertson's s i t e s are approximately halfway between Chase and Kamloops. Only d i f f u s e knapweed i s found at the Ned's Creek and Robertson's s i t e s and the Chase s i t e i s a pure stand of s p o t t e d knapweed. Based on nearest neighbour d i s t a n c e s measured i n 1979, the d e n s i t y of d i f f u s e knapweed at Robertson's was higher than at Ned's Creek (6.34±0.26 cm (MeantS.E.) at Robertson's vs. 8.08±0.29 cm at Ned's Creek). Knapweed has been pres e n t at a l l s i t e s s i n c e at l e a s t 1965 ( H a r r i s , pers. Comm.; Robertson, p e r s . comm.). The s i t e s are d e s c r i b e d i n more d e t a i l by Berube (1980), H a r r i s (1980a), and Roze (1981). 9 Figure 0.1. Locat ion of the study s i t e s . The two d i f fuse knapweed s i t e s , Ned's Creek and Robertson's, are located at approximately 1 1 9 ° 5 0 ' W , 5 0 ° 4 2 ' N and 119°50 'W, 5 0 ° 4 0 ' N , r e s p e c t i v e l y . The spotted knapweed s i t e , Chase, i s located at approximately 119°42 'W, 5 0 ° 4 8 ' N . The top of the f igure is due north. • 1 Ned's Creek • 2 Robertson's • 3 Chase Metres 0 1000 2000 3000 4000 • 4 11 Chapters I, I I , and I I I present r e s u l t s from Robertson's o n l y ; Chapters IV and V p r e s e n t r e s u l t s from a l l three s i t e s . BUD DESCRIPTIONS The f o l l o w i n g terms and methods are used to d e s c r i b e the developmental s t a t e and l o c a t i o n of knapweed buds. Any bud that i s l a r g e ' enough to be measured (> 2.0 mm) i s d e s c r i b e d as i n i t i a t e d and p l a c e d i n t o one of three c a t e g o r i e s : undeveloped, developed, or a b o r t e d . A l l buds are i n i t i a l l y undeveloped; some remain i n that category. If a bud r e c e i v e s enough resources to mature and by the time of c o l l e c t i o n i s l a r g e enough to c o n t a i n g a l l s or seeds, the bud i s d e s c r i b e d as developed. If the bud would normally be developed, but i s not because of i n s e c t a t t a c k , i t i s d e s c r i b e d as aborted. The p a t t e r n of growth of knapweed s i m p l i f i e s r e c o r d i n g the l o c a t i o n of buds on p l a n t s . The a p i c a l bud i s t y p i c a l l y the f i r s t to be i n i t i a t e d and the f i r s t to flower ( i n the absence of f l y a t t a c k ) . I t i s f o l l o w e d by l a t e r a l buds down the stem. As these l a t e r a l branches e l o n g a t e , l a t e r a l buds develop on them. L a t e r a l buds and branches can a l s o be i d e n t i f i e d by t h e i r subtending b r a c t s . T h i s p a t t e r n repeats i t s e l f r e c u r s i v e l y , b a s i p e t a l l y and c e n t r i p e t a l l y , u n t i l a l l buds are i n i t i a t e d . T h i s r e c u r s i v e and d i r e c t i o n a l growth means that i n d i v i d u a l buds can be given a unique d e s i g n a t i o n (see F i g u r e 0.2). The a p i c a l bud i s denoted "1". The f i r s t l a t e r a l bud i s "2". The f i r s t l a t e r a l bud o f f the f i r s t l a t e r a l branch i s "2-1" and so on. L o c a t i o n s of f l i e s were reco r d e d r e l a t i v e to these buds. Roze (1981) used a s i m i l a r scheme. Buds may be placed in a branching 1 2 Figure 0.2. Numbering scheme for knapweed buds. D e t a i l s are given in the text . 1 4 category based on the number of i n d i c e s r e q u i r e d to d e s c r i b e the l o c a t i o n of the bud (e.g. "2" would be i n the f i r s t b r anching category and "4-1-1" would be i n the t h i r d ) . The determinate p a t t e r n of knapweed growth makes i t p o s s i b l e to i d e n t i f y aborted buds by t h e i r p o s i t i o n on the p l a n t . Aborted buds were determined by t h e i r l o c a t i o n r e l a t i v e to developed (=good) buds. For example, i f buds 4, 4-1, and 4-2 were not developed and 4-3 had flowered, the f i r s t t h ree would be i d e n t i f i e d as aborted. I f 4-1-1 was a l s o not developed, i t c o u l d not be pl a c e d i n the aborted category, s i n c e i t does not n e c e s s a r i l y precede bud 4-3 i n development. More p r e c i s e l y , a borted buds were those that were f o l l o w e d i n the normal developmental sequence by a bud that was e i t h e r l a r g e r by 2.0 mm or more, or the bud under c o n s i d e r a t i o n was not developed when the f o l l o w i n g bud was developed (flowered, or c o n t a i n e d g a l l s ) . The numbers of aborted buds estimated i n t h i s way are prob a b l y underestimates s i n c e aborted buds can only be i d e n t i f i e d i n r e l a t i o n s h i p to developed buds. Aborted buds are e q u i v a l e n t to Roze's (1981) " s u p e r p a r a s i t i z e d " buds. STATISTICAL METHODS Unless otherwise i n d i c a t e d , when means or r e g r e s s i o n c o e f f i c i e n t s are re p o r t e d the standard e r r o r i s giv e n (e.g. Mean±S.E.). When x 2 t e s t s are used to compare d i s t r i b u t i o n s , c e l l s with fewer than f i v e o b s e r v a t i o n s are combined (Sokal and Rohlf, 1969). R e s u l t s of s t a t i s t i c a l t e s t s are given by an exact p r o b a b i l i t y , p, i f p i s g r e a t e r than 0.001, otherwise by the i n e q u a l i t y p<0.00l. I t was not p o s s i b l e 1 5 to a s s i g n exact p r o b a b i l i t i e s to the r e s u l t s of some s t a t i s t i c a l t e s t s . Square root and l o g a r i t h m i c t r a n s f o r m a t i o n s are used when the v a r i a n c e s of groups being compared are s i g n i f i c a n t l y d i f f e r e n t or i f the v a r i a n c e i s a f u n c t i o n of the mean. The degrees of freedom f o r Student's t - t e s t was a d j u s t e d using S a t t e r t h w a i t e ' s formula i f the v a r i a n c e s of the two groups were s i g n i f i c a n t l y d i f f e r e n t by the F - t e s t at a=0.05 (SAS I n s t i t u t e , 1982). 1 6 I. THE EFFECT OF HOST SELECTION ON THE POPULATION DYNAMICS OF TWO INTRODUCED INSECTS Phytophagous i n s e c t s are c o n f r o n t e d with an enormous v a r i a t i o n i n the s u i t a b i l i t y of food resources (e.g. Chew, 1975; Dixon, 1976). I n s e c t s respond to t h i s v a r i a t i o n by s e l e c t i n g food resources on the b a s i s of s p e c i e s (Brues, 1920), chemical composition ( D e t h i e r , 1941; F r a e n k e l , 1959; M c N e i l l and Southwood, 1978), growth form (Bach, 1981), s i z e (Whitham, 1978; 1980), developmental stage ( C a l c o t e , 1975), and combinations of these and other f a c t o r s (e.g. Crawley, 1983; Hare and Futuyma, 1978; Moore, 1978b). The l i n k between host p l a n t v a r i a t i o n and o v i p o s i t i o n c h o i c e i s c r i t i c a l where one i n s e c t l i f e stage chooses f o r another l i f e stage. In some cases, o v i p o s i t i o n c h o i c e s may not r e f l e c t s u i t a b i l i t y f o r l a r v a l s u r v i v a l and development (Rausher, 1979b; Wiklund, 1975), and indeed may run counter to i t (Chew, 1977; Courtney, 1981; though see Rausher and Papaj, 1983). The two in t r o d u c e d i n s e c t s , Urophora a f f i n i s F r f l d . and U. q u a d r i f a s c i a t a (Meig.) ( D i p t e r a : T e p h r i t i d a e ) , l a y eggs i n the immature flower buds of d i f f u s e knapweed, Centaurea d i f f u s a Lam. (As t e r a c e a e ) . The young l a r v a e s t i m u l a t e g a l l formation w i t h i n the buds and feed on n u t r i t i v e t i s s u e produced by the p l a n t . Thus the observed d i s t r i b u t i o n of g a l l s , both among p l a n t s and among buds on p l a n t s , should be c o r r e l a t e d with the ch o i c e s made by the a d u l t females. 1 7 Two p o s s i b l e c o n t r a s t s w i l l c l a r i f y the process of o v i p o s i t i o n s i t e s e l e c t i o n . Larvae of the two i n s e c t s p e c i e s have been found to complete g a l l formation in buds of d i f f e r e n t developmental stages (Berube and H a r r i s , 1978). T h i s should be r e f l e c t e d i n the c h o i c e of buds by o v i p o s i t i n g females. The second c o n t r a s t i s between sexes. If females are making c h o i c e s among o v i p o s i t i o n s i t e s , the males must choose the same buds as females because the g a l l f l i e s mate on or near the e ventual o v i p o s i t i o n s i t e s (Zwolfer-, 1970). However, the males of many t e p h r i t i d s are t e r r i t o r i a l (e.g. Burk, 1981; Parker, 1974; V a r l e y , 1947) and U. a f f i n i s and U. q u a d r i f a s c i a t a males are no e x c e p t i o n (Berube and Myers, MS; Zwolfer, 1974). T e r r i t o r i a l behaviour may l e a d to d i f f e r e n c e s i n the d i s t r i b u t i o n s of male and female f l i e s , or i n the d i s t r i b u t i o n s of males that are mating compared wi t h males that are not mating. One response of the p l a n t s to f l y a t t a c k may reduce the s u i t a b i l i t y of buds as o v i p o s i t i o n s i t e s . Roze (1981) observed that knapweed buds f a i l e d to develop a f t e r they were exposed to high l e v e l s of i n s e c t a t t a c k i n the l a b o r a t o r y . Because aborted buds do not produce any g a l l s , the d i s t r i b u t i o n of a t t a c k among buds w i l l a f f e c t the t o t a l number of g a l l s produced and, hence, the number of a d u l t s i n the next g e n e r a t i o n . A t t a c k by the g a l l f l i e s a l s o has a d e l a y e d impact on f u t u r e i n s e c t d e n s i t y . I f the l e v e l of f l y a t t a c k i s high enough, seed p r o d u c t i o n by the p l a n t s may be s i g n i f i c a n t l y reduced. A drop in seed production may i n turn a f f e c t the d e n s i t y of p l a n t s and hence of o v i p o s i t i o n s i t e s i n the f u t u r e . 18 T h i s Chapter (1) examines the r e l a t i o n s h i p between s e l e c t i o n of o v i p o s i t i o n s i t e s by the two g a l l f l i e s and the s u i t a b i l i t y of those s i t e s f o r l a r v a l growth and development, and (2) assesses the consequences of the s e l e c t i o n and s i t e s u i t a b i l i t y f o r the p o p u l a t i o n dynamics of the i n s e c t s . 19 MATERIALS AND METHODS OBSERVATION METHODS An obs e r v a t i o n p l o t was e s t a b l i s h e d at Robertson's on June 1, 1980. A r e c t a n g u l a r g r i d of f i f t y (5x10) p o i n t s was p l a c e d over a 3 m x 6.5 m p o r t i o n of the f i e l d with a v i s u a l l y u n iform d e n s i t y of d i f f u s e knapweed. The p l a n t which had begun to b o l t nearest each p o i n t on the g r i d was staked. The p o i n t s on the g r i d were f a r enough apart so that staked p l a n t s were r a r e l y nearest neighbours. The staked p l a n t s provided a b a s i s f o r d e t a i l e d o b s e r v a t i o n s of the f l i e s and t h e i r h o s t s . Approximately every two days, I conducted a branch by branch v i s u a l survey of each staked p l a n t . The l o c a t i o n , s p e c i e s , sex, and a c t i v i t y of each a d u l t f l y was recorded at the moment of o b s e r v a t i o n . The two s p e c i e s are e a s i l y d i s t i n g u i s h e d by the banding p a t t e r n on the wings and sexes are d i s t i n g u i s h e d by the prominent o v i s c a p e of the females. The f l i e s are q u i t e d o c i l e and t h e i r behaviour was not o b v i o u s l y a l t e r e d by an observer. The v i s u a l surveys f o r a d u l t f l i e s on p l a n t s were u s u a l l y conducted between 8:30 a.m. and 11:30 a.m., with most begun at 9:30 a.m. The surveys l a s t e d from one h a l f hour to one hour, depending on the number of f l i e s observed. In four cases, i t was necessary to survey a d u l t s i n the e a r l y a f t e r n o o n (June 17 and 27 and J u l y 1 and 7). I t i s u n l i k e l y that d a i l y p a t t e r n s of a c t i v i t y s t r o n g l y a f f e c t e d the counts of U. a f f i n i s . In the one case where counts were made twice i n one day (June 13), the count over a l l f i f t y p l a n t s at 8:40 a.m. was 10 U. a f f i n i s and 20 at 2:00 p.m. was 14 U. a f f i n i s . ( I n s e c t s were not observed on the same p l a n t s i n the two surveys.) The time the surveys were conducted may, however, a f f e c t the counts of U. a f f i n i s r e l a t i v e to U. q u a d r i f a s c i a t a i f the two s p e c i e s have d i f f e r e n t d a i l y a c t i v i t y p a t t e r n s . An i n d i c a t i o n of the extent of t h i s d i f f e r e n c e comes from comparing the s p e c i e s composition recorded from the v i s u a l surveys with the s p e c i e s composition of f l i e s caught on a s t i c k y t r a p . As part of another experiment (MS i n p r e p . ) , a l a r g e (6 m on i t s longest d i a g o n a l ) hexagon of f i b r e g l a s s n e t t i n g was c o n s t r u c t e d . It extended from ground l e v e l to 0.5 m h i g h , supported by wooden posts at the v e r t i c e s and at the midpoints of the s i d e s . The i n t e r i o r of the hexagon was coated with T a n g l e f o o t ( r e g i s t e r e d trademark of The T a n g l e f o o t Company, Grand Rapids, Michigan) which was r e f r e s h e d every three weeks. G a l l f l i e s were removed from the n e t t i n g approximately every two days d u r i n g the f i r s t g e n e r a t i o n . S t i c k y t r a p counts i n t e g r a t e over the e n t i r e d a i l y a c t i v i t y c y c l e . The p r o p o r t i o n of U. quadr i f asc i a t a a d u l t s of a l l g a l l f l i e s was very s i m i l a r on the s t i c k y t r a p (7.1%) and from v i s u a l surveys (7.8%). T h i s suggests that the v i s u a l surveys g i v e a reasonable p i c t u r e of the r e l a t i v e numbers of U. a f f i n i s and U. q u a d r i fasc i a t a . D i f f e r e n c e s i n a c t i v i t y p a t t e r n s w i l l not a f f e c t comparisons among p l a n t s or among o b s e r v a t i o n times f o r the same i n s e c t s p e c i e s . I a l s o recorded the number, l o c a t i o n , and s i z e of a l l the buds on the p l a n t s 19-21 times d u r i n g the p e r i o d June 1 to August 24. The i n t e r v a l between the o b s e r v a t i o n s of buds 21 increased as the number of buds increased during the season. Bud s izes were measured with c a l i p e r s as described by Berube and H a r r i s (1978) to the nearest m i l l i m e t e r . With some prac t i ce i t became poss ib le to v i s u a l l y estimate s i z e s . Spot checks with c a l i p e r s confirmed the accuracy of these estimates. Plant height , developmental state , and evidence of herbivory were a lso noted. One plant died during the period of observat ion, but continued to a t t rac t f l i e s . It was excluded from s t a t i s t i c a l ana lys i s when appropriate . PLANT COLLECTIONS Plants were c o l l e c t e d i n d i v i d u a l l y on August 23, 1980, by c l i p p i n g them off at ground l e v e l . They were then stored in folded and stapled paper bags at room temperature u n t i l d i s s e c t i o n . When they were d i s sec ted , the height of each plant was measured and the s i ze , developmental s tatus , and l o c a t i o n of each bud were recorded. Buds large enough to contain e i ther g a l l s or seeds were i n d i v i d u a l l y d i s sec ted . For these, the number of seeds and g a l l s present were noted. The g a l l s of the two insect species are e a s i l y d i s t ingu i shed; U. af f i n i s g a l l s are hard and woody and U. quadri fasc ia ta g a l l s are th in and papery. CALORIFIC CONTENT One measure of the s u i t a b i l i t y of buds for l a r v a l development, the maximum number of g a l l s a bud could produce, could not be evaluated d i r e c t l y because buds a lso produced seeds. Using H a r r i s ' (1980b) measurements of c a l o r i f i c value of poss ib le bud contents, the "carrying capacity" of 22 developed buds may be c a l c u l a t e d in common uni t s . Diffuse knapweed seeds have an average energy content of 6 . 5 ± 0 . 1 c a l o r i e s . The combination of U. a f f i n i s larva and g a l l has a c a l o r i f i c value of 2 8 . 0 ± 1 . 1 c a l o r i e s ( H a r r i s , 1980b). Harr i s does not give a value for U. quadr i fasc ia ta larva and g a l l , but the value for the larva alone ( 4 . 2 7 ± 0 . 1 4 ca lor i e s ) may be used, since l i t t l e g a l l t i ssue remains a f t er l a r v a l feeding. This measure of s u i t a b i l i t y i s a s t a t i c measure estimated at the end of the season and thus represents the accumulated v a r i a t i o n among buds that females should t ry to match by the ir dynamic choices during the season. 23 RESULTS VARIATION IN RESOURCES P l a n t s v a r i e d c o n s i d e r a b l y i n the number of buds. The range i n buds per p l a n t was 16-180, with a mean of 71.7 and a c o e f f i c i e n t of v a r i a t i o n of 59.5% ( c f . 62% based on Schirman's (1981) d a t a ) . Using H a r r i s ' (1980b) c a l o r i f i c values to estimate the d i s t r i b u t i o n of c a l o r i f i c value among the contents of developed buds g i v e s a mean of 57.3 c a l o r i e s per developed bud with a range of 0-279 c a l o r i e s and a c o e f f i c i e n t of v a r i a t i o n of 69%, again a high degree of v a r i a t i o n . T h i s measure w i l l not be an exact r e p r e s e n t a t i o n of the v a r i a t i o n because g a l l s and seeds are not i d e n t i c a l in weight from bud to bud. Buds i n the primary branching category had s i g n i f i c a n t l y more energy re s o u r c e s than the other three c a t e g o r i e s (Table 1.1). INSECT CHOICE AMONG PLANTS The g a l l f l i e s chose among p l a n t s . A x 2 t e s t of t o t a l numbers of f l i e s observed on i n d i v i d u a l p l a n t s r e v e a l s t h a t the d i s t r i b u t i o n was not uniform among p l a n t s (U. a f f i n i s , X 2=510, df=49, p<0.001; U. q u a d r i f a s c i a t a , x 2=338, df=8, P<0.001). C o n t r a s t i n g the d i s t r i b u t i o n s of the two sexes and the two s p e c i e s w i l l c l a r i f y the b a s i s f o r c h o i c e among p l a n t s . The 24 Table 1.1 - C a l o r i f i c content of developed di f fuse knapweed buds by branching category Branching Calor i f i c Category Content * N 1 7 5 . 6 ± 3 . 1 ** 269 2 5 4 . 3 ± 1 .3 782 3 4 8 . 6 ± 1 .6 448 4 4 4 . 8 ± 5 . 7 34 * Ca lor i e s (MeaniS .E . ) . Based on H a r r i s ' (1980b) measurements. ** Analys is of variance on log transformed data gives F=9.88, df=3,1529, p<0.001. numbers of males and females of the two species observed on plants were highly c o r r e l a t e d (log transformed data, r=0.73, df=48, p<0.00l) . Males and females may use the same (or related) c r i t e r i a for p o s i t i o n i n g themselves. The d i s t r i b u t i o n of a l l observed U. a f f i n i s adults among plants d i f f e r e d from the d i s t r i b u t i o n of a l l observed U. quadr i fasc ia ta adults ( x 2 = l 9 0 . 2 , df = 4, p<0.00l), however the heterogeneity between the two species in the ir response to the plants i s p r i m a r i l y due to a s ing le plant (Figure 1.1). If th i s plant is removed from a n a l y s i s , the heterogeneity p e r s i s t s ( x 2 =48 .26 , df=3, p<0.001), but o v e r a l l the counts of f l i e s per plant are well c o r r e l a t e d (log transformed data, r=0.53, df=47, P<0.001). The anomalous plant i n i t i a t e d buds much e a r l i e r than any of the other plants observed. As a re su l t , the buds were propor t iona l ly larger during the ov ipos i t ion period of the f l i e s , and a t t rac ted almost twice as many U. quadr i fasc iata as U. a f f i n i s . The d i f ference in bud s ize preference between f l y 25 F i g u r e 1.1. T o t a l number of a d u l t f l i e s of the two s p e c i e s of g a l l f l y observed on each of the d i f f u s e knapweed p l a n t s . Each p o i n t represents a s i n g l e p l a n t . T, 25. o > © 2 20. TO CO o (0 CO 15J =! 10J a) 5-E 3 i n . i »JL • •• •• • • • r 10 20 30 40 —r-50 —r— 60 Number of U. affinis observed 27 species (see below) and di f ferences among plants in bud i n i t i a t i o n and growth probably account for the remaining heterogeneity in the d i s t r i b u t i o n s of the two species . O v e r a l l , there is a strong r e l a t i o n s h i p between the t o t a l number of f l i e s observed on plants and the number of buds on that plant (Figure 1.2). Considering the two species separately, the regression for U. a f f i n i s i s : ln(UA+1) = 0 . 9 1 ( ± 0 . 1 1 ) l n ( B U D S + 1 ) - 0 . 9 4 ( ± 0 . 4 4 ) , F=72.32, df = 1,48, p<0.001, r = 0.78. Removing the anomalous plant discussed above, for U. quadr i fasc ia ta the r e l a t i o n s h i p i s : ln(UQ+l) = 0 . 5 9 ( ± 0 . 1 3 ) l n ( B U D S + 1 ) - 1 . 8 6 ( ± 0 . 5 5 ) , F=19.98, df=1,47, p<0.00l, r=0.55. These regressions expla in the c o r r e l a t i o n of U. af f i n i s and U. q u a d r i f a s c i a t a counts observed in Figure 1.1. V a r i a b i l i t y unaccounted for in the regressions may be due to d i f ferences among plants not r e f l e c t e d in the numbers of buds per plant or to the i n a b i l i t y of the g a l l f l i e s to per fec t ly track the number of o v i p o s i t i o n s i t e s per p l a n t . AMONG BUDS ON PLANTS The f l i e s also chose s i t e s within p l a n t s . If buds are d iv ided into the i r branching categories (primary, secondary, e t c . ) , and the numbers of f l i e s observed on buds are compared with the numbers of buds in the respect ive categories adjusted by the d i s t r i b u t i o n of buds at the time the insects were observed on the p lant , both U. a f f i n i s (x 2=21.65, df=2, P<0.001) and U. quadr i fasc ia ta (X 2 =39.51, df=1, p<0.00l) were observed d i sproport ionate ly more often on primary buds (Table 28 Figure 1.2. T o t a l number of g a l l f l i e s and t o t a l number of buds for each of the d i f fuse knapweed p l a n t s . Each point represents a s ingle p l a n t . 60 40J c (0 (0 ® 20J i i i • 1 1 — 20 40 60 80 100 120 — i 1 — 140 160 180 200 Buds per plant 30 1.2). (The ef fect of the temporal d i s t r i b u t i o n of the f l i e s i s Table 1.2 - Observed and predic ted d i s t r i b u t i o n s of U. a f f i n i s (UA) and U. quadr i fasc ia ta (UQ) adults among branching categories BRANCHING CATEGORY 1 2 3 UA Obs. 463 (0.56)a 324 (0.40) 34 (0.04) UA Pred. 411.6 b 350.6 63. 8 UQ Obs. 53 (0.67) 23 (0.29) 3 (0.04) UQ Pred. 26.6 52.4 c a - Tota l observed (Proportion) b - Predicted numbers of f l i e s in branching categories are based on the t o t a l number of f l i e s observed on a l l buds weighted by the d i s t r i b u t i o n s of buds among branching categories as the d i s t r i b u t i o n s change in time. c - Categories 2 and 3 were combined because of the small number of observations in category 3. treated separately in the next Chapter.) There was no s i g n i f i c a n t d i f ference between sexes in the d i s t r i b u t i o n of adult f l i e s among branching categories (U. a f f i n i s , x 2=1.342, df = 2, p=0.511, N=690; U. q u a d r i f a s c i a t a , x 2 = 1 - 117, df=1, p=0.291, N=65). The preference for primary buds is a lso observed i f the sexes of each species are considered separately (p<0.01 in a l l cases) . Though the sample size is small (N=26), probing U. a f f i n i s females a l so preferred primary buds (x 2=4.89, df=1, p=0.027). This preference is consistent with the higher c a l o r i f i c content of primary buds compared to secondary or t e r t i a r y buds (Table 1.1). 31 There was no evidence that the d i s t r i b u t i o n of male f l i e s among branching categories was af fected by t e r r i t o r i a l behaviour. The d i s t r i b u t i o n of mating U. a f f i n i s males was not s i g n i f i c a n t l y d i f f eren t from the d i s t r i b u t i o n of males that were not mating (x 2=0.878, df=2, p=0.645, N=573). The two f l y species a lso d i f f e r e d s i g n i f i c a n t l y in the ir preference for bud s izes (Table 1.3). As expected, U. a f f i n i s Table 1.3 - D i s t r i b u t i o n of U. a f f i n i s (UA) and U. quadr i fasc ia ta (UQ) adults by s ize of d i f fuse knapweed buds Insect Category Bud Size (mm) N UA males 2 . 9 1 ± 0 . 0 7 a 519 UA females 3 . 0 1 ± 0 . 1 3 a 171 UA mating pair 3 . 2 2 ± 0 . 2 3 c 54 UQ males 3 . 6 0 ± 0 . 2 7 b 40 UQ females 4 . 6 6 ± 0 . 3 4 b 25 UQ mating pair 5 . 3 5 ± 0 . 9 8 c 3 Let ters indicate s t a t i s t i c a l comparisons: a - t=0.061, df=582, p=0.952. b - t=1.75, df=57, p=0.085. c - Mann Whitney U tes t , U=139, 0.02<p<0.05. sat on or near buds that were smaller than those chosen by U. quadr i fasc ia ta (overa l l t - t e s t , t=4.96, df=842, p<0.00l) . The bud s izes chosen by male and female U. a f f i n i s d i d not d i f f e r s i g n i f i c a n t l y , but U. quadri fasc ia ta females picked s l i g h t l y bigger buds than conspec i f ic males. The s l i g h t preference for larger buds by U. quadr i fasc ia ta females i s also re f l ec ted in the buds on which mating pa i r s were observed (Table 32 1.3). The bud s izes U. quadr i fasc ia ta females chose are c loser to the s izes pre ferred for ov ipos i t ion in the laboratory (Berube and H a r r i s , 1978) than the bud s izes males chose. The d i s t r i b u t i o n s of mating and non-mating males r e l a t i v e to bud s ize d id not d i f f e r s i g n i f i c a n t l y for e i ther species of g a l l f l y (U. a f f i n i s , t=1.59, df=480, p=0.1l3; U. quadr i fasc i a t a , t=1 .33, df = 36, p = 0 . l 9 l ) . Again, no d i f ference that could be a t t r i b u t e d to t e r r i t o r i a l behaviour was detected. The s izes of buds which I observed being probed by female f l i e s in the f i e l d correspond to the bud s ize preferences observed in the laboratory (Berube and H a r r i s , 1978; Roze, 1981; Zwolfer, 1970). The mean s ize of buds probed by U. a f f i n i s was 3.7 mm (N=32), and by U. quadr i fasc iata 5.1 mm (N=6). The d i f ference between the two species was s i g n i f i c a n t (t=3.34, df = 36, p=0.002). The s ize probed by U. quadr i fasc ia ta was smaller than that previous ly observed, but the sample s ize is small and the females may not have been ov ipos i t ing (see below). CONSEQUENCES OF INSECT CHOICE AMONG PLANTS GALL FORMATION The s ingle most important consequence of o v i p o s i t i o n from the f l i e s ' perspective is g a l l formation. The number of g a l l s formed per plant is strongly corre la ted with the number of females observed on those plants (log transformed data , r=0.70, df=47, p<0.00l for U. a f f i n i s , Figure 1.3; untransformed data, Spearman r=0.52, df=47, p<0.001 33 for U. quadr i fasc i a t a , Figure 1.4). Since the number of f l i e s per plant i s wel l corre la ted with the number of buds, the number of g a l l s per plant should also depend on the number of buds. For the number of U. a f f i n i s g a l l s as a function of the number of buds, the regress ion i s : ln(UAG+1) = 0.80(±0.18)ln(BUDS+1) + 0.06(±0.76), F=19.54, df=1,47, p<0.00l, r=0.54. For U. q u a d r i f a s c i a t a the regression i s : ln(UQG+1) = 0.99(±0.25)ln(BUDS+1) - 2.4(±1.1), F=15.53, df=1,47, p<0.00l, r=0.50. BUD ABORTION Bud abort ion is an a l t e r n a t i v e outcome of f l y attack. If bud abortion is an increas ing funct ion of f l y at tack, then the number of aborted buds per plant w i l l be corre la ted with the number of U. a f f i n i s females. This was observed (log transformed data , F=16.44, df=1,47, p<0.00l, r=0.5l). If the two outcomes of f l y attack are mutually exclus ive , then for a given l e v e l of f l y attack the proportion of buds aborted per plant w i l l be inverse ly propor t iona l to the average number of U. a f f i n i s g a l l s per developed bud: PROP. ABT = -0.054(±0.019)UA GALLS/BUD + 0.346(±0.028), F=8.08, df=1,47, p=0.007, r=-0.38 (Figure 1.5). Figure 1.5 also indicates that there are s i g n i f i c a n t d i f ferences among plants in the propensity to abort buds. SEED REDUCTION The most important consequence of insect choice from the p lant ' s perspective i s the reduction in seeds. Since 34 Figure 1.3. T o t a l number of U. t o t a l number of U. a f f i n i s knapweed p lants . Deta i l s in the text . a f f i n i s adults observed and g a l l s for each of the d i f fuse of the c o r r e l a t i o n are given 35 * • * Lo CO in o </> £> O CO C .o r CO CO o -° CM E z Lo o CM O O O CO o O 5t o CM S||B6 Bujiinsau 36 Figure 1.4. Tota l number of U. quadri fasc ia ta adults observed and t o t a l number of U. quadr i fasc ia ta g a l l s for each of the d i f fuse knapweed p l a n t s . De ta i l s of the c o r r e l a t i o n are given in the text . Note that nine points were recorded at (0 ,0) . •o 25. CD > <D CO •g 20. CO CD o CO CO 15J 10J $ 5-•U E 3 • • • 1 . . t i t 1 - « . « 10 20 30 40 Number of U. affinis observed 38 Figure 1.5. Mean number of U. a f f i n i s g a l l s per developed bud and proportion of buds aborted for each of the d i f f u s e knapweed p lants . De ta i l s of the c o r r e l a t i o n are given in the text . 39 •9 3 T3 3 •a CD a o CD > CD CD CM <n "co O) .Q II CM . - I =1 I • • • UQ —I T 1 1 I IT) rt CO CM T d o d d ^ pajjoqe spnq jo uojjjodojd 40 seeds w i l l be proport ional to the number of developed buds, I regressed the number of seeds per plant against the number of developed buds per plant and the number of U. a f f i n i s g a l l s per p lant : SEEDS = 4 . 7 4 ( ± 0 . 4 8 ) D B U D S - 0 . 9 7 ( ± 0 . 2 8 ) U A G - 3 . 3 ( ± 1 4 . 3 ) , F=51.9, df=2,46, p<0.00l, r=0.83. Thus for any given p lant , the number of seeds produced w i l l be p o s i t i v e l y corre la t ed with the number of developed buds and negatively corre la ted with the number of U. a f f i n i s g a l l s . The inc lus ion of a term for U. quadri fasc ia ta g a l l s in the mult ip le regression d id not s i g n i f i c a n t l y increase the proportion of the variance accounted f o r . Given the average number of U. a f f i n i s g a l l s per plant ( 3 6 . 9 ± 4 . 0 ) , the average number of developed buds per plant ( 3 0 . 7 ± 2 . 4 ) and the regression above, g a l l s w i l l reduce seed production per plant by 27%. As discussed below, th i s value does not take account of the effect of bud abortion or of d i f ferences in bud p r o d u c t i v i t y . AMONG BUDS ON PLANTS The re la t ionsh ips between f l y attack, g a l l formation, bud abort ion , and seed production observed among plants a l so hold across branching categories within plants (Figures 1.6, 1.7). Branching categories d i f f e r in the proport ions of developed buds, aborted buds, and undeveloped buds (Figure 1.7). For comparison, in unattacked plants there are no aborted buds and the proportion of buds developed dec l ines with increas ing branching category number. F ly choice 41 Figure 1.6. D i s t r i b u t i o n of g a l l s and seeds per developed bud across branching ca tegor ie s . V e r t i c a l l ines give ± one standard e r r o r . Unattacked d i f fuse knapweed buds have approximately 12 seeds per developed bud (Watson, 1972). 6.0. TJ 3 A "|5.0. O > o "° 4.0_ L. o a (0 "g3.0. o (0 c (0 2.0-(0 O) OT 'E =1 1.0. • U affinis galls per developed bud Seeds per developed bud ijijijiA quadrifasciata galls per developed bud T " 2 i T 4 Branching category .1.0 3 13 I TJ <D a O •0-8| .0.6 a (0 ro O) ro .0.4-2 OT ro 0.23 Dl ro 43 Figure 1.7. D i s t r i b u t i o n of bud fates across branching categor ies . A l l buds f a l l into one of three fate categor ies . The sample s ize for each branching category is shown in parentheses above the corresponding proport ions. 1 2 3 4 Branching category £ 45 as measured by locat ion of adults was p o s i t i v e l y re lated to g a l l formation and bud abortion and negat ive ly re lated to seed product ion . Fewer g a l l s were obtained from primary buds than expected based on the d i s t r i b u t i o n s of adult f l i e s . This was true for both U. a f f i n i s (x 2=835, df=2, p<0.00l) and for U. quadr i fasc ia ta (x2=846, df=2, p<0.001). Both species preferred primary buds, but g a l l formation per insect was r e l a t i v e l y lower in these buds, presumably because of the higher bud abort ion . Of the 32 probings by female U. a f f i n i s I observed, twenty (63%) were into buds which l a t e r contained U. a f f i n i s g a l l s , eight (25%) were into buds which aborted, and one (3%) was into a bud which matured, but d id not contain a U. a f f i n i s g a l l . (The remainder of the buds (9%) were damaged by grasshoppers.) The s ize of the buds probed by U. af f i n i s that were subsequently aborted was smaller than of buds that subsequently contained g a l l s ( 3 . 4 7 ± 0 . 1 5 mm vs. 4 . 2 1 ± 0 . 1 5 mm; Mann-Whitney U tes t , U=128.5, 0.01<p<0.02). These data suggest that smaller buds are more l i k e l y to abort for a given number of probes. The c o r r e l a t i o n between probing and g a l l production for U. quadr i fasc ia ta was not nearly so good. Of the six probings I observed, only one of the probed buds subsequently contained a U. quadr i fasc ia ta g a l l , none were aborted, and three la ter contained U. a f f i n i s g a l l s . These observations raise the p o s s i b i l i t y that U. quadr i fasc ia ta can detect the presence of U. a f f i n i s eggs or larvae . They are a l so consistent with a 46 r e j e c t i o n of the buds on the basis of unsuitable developmental s ta te . 47 DISCUSSION VARIATION IN BUD PRODUCTIVITY The v a r i a t i o n in "carrying capacity" of buds and the concentrat ion of attack in more productive buds explains the nonlinear impact of g a l l formation on seed production observed by other workers (Story, 1978b; Zwolfer, 1969, 1978). This may a lso account for the nons igni f icant e f fect of U. q u a d r i f a s c i a t a on seed product ion. Story (1978b) noted a h ighly s i g n i f i c a n t , but nonlinear reduction in seed set by spotted knapweed with increasing attack by U. a f f i n i s in f i e l d cages in Montana. Zwolfer (1978) i d e n t i f i e d a s imi lar nonlinear impact on seed production of spotted knapweed in Europe. An analogous pattern was observed for Urophora s iruna-seva a t tack ing Centaurea s o l s t i t i a l i s L . in Europe. The reduction in seed set was 50% for s ingly attacked buds, but less than 50% for mul t ip ly attacked buds (Zwolfer, 1969) . Harr i s (1980b) argues that g a l l s act as a "sink" for plant nutr ients (cf . Fourcroy and Braun, 1967; Jankiewicz e_t a l . , 1970) . He used a comparison of seed production in spotted knapweed as evidence for the sink e f f e c t . He observed that the regression of the e f fect of U. a f f i n i s g a l l s on seed production had a lower intercept (19.9 seeds) than found by Watson (1972) (26.6 seeds) for unattacked p l a n t s . Harr i s suggests that the d i f ference is due to the sequestering of nutr ients by the g a l l f l y larvae away from unattacked buds. Var ia t ion in bud p r o d u c t i v i t y and se l ec t ive attack on more productive buds can 48 a l so explain the dif ference observed by H a r r i s . BUD ABORTION AND POPULATION LIMITATION Bud abort ion in C. d i f f u s a and C. maculosa (Roze, 1981) are the f i r s t documented examples of members of the Asteraceae abort ing flowers or f r u i t s (c f . Stephenson, 1981; Hare and Futuyma (1978) give an example of seed abortion in a cocklebur as the resul t of insect a t t a c k ) . The abortion appears to be functioning as a plant defence. It i s something of a Pyrrhic v i c t o r y for the plant as seed production from aborted buds is a lso e l iminated, however the cost of aborting a bud may be r e l a t i v e l y small compared to the resources required to develop i t (Stephenson, 1981). Bud abortion reduces the reproductive output of the g a l l f l y populat ions . Faeth et a l . (1981) suggest that a s i m i l a r "plant defence", ear ly leaf a b s c i s s i o n , may be an important source of morta l i ty for f o l i o v o r e s . These kinds of plant defence w i l l be p a r t i c u l a r l y e f f ec t ive against sedentary insects l i k e leaf miners and g a l l formers. The reduction in the p o t e n t i a l number of g a l l s due to bud abort ion may be estimated from the d i s t r i b u t i o n of g a l l s and from the proportion of buds aborted. I used the branching category d i s t r i b u t i o n s and assumed that aborted buds would have produced the same number of g a l l s per bud as unaborted buds. The number of U. af f i n i s and U. quadr i fasc ia ta g a l l s per plant would have been an average of 72% and 71% higher, r e s p e c t i v e l y , i f the plants had not aborted buds. These percentages are conservat ive because they do not include the e f fec ts discussed 49 in the next Chapters. If the g a l l f l i e s s i g n i f i c a n t l y reduce seed production per p lant , the density of o v i p o s i t i o n s i t es (and hence insect density) in subsequent years may a lso drop. It is poss ible to estimate the p o t e n t i a l seed production per plant in the absence of f l y attack from the observed c a l o r i f i c content of developed buds by branching category. Assuming that aborted buds would have had the same c a l o r i f i c content as developed buds in the same branching category, the t o t a l number of ca lor i e s that would have been ava i lab l e for seed production in the average plant i s 2986. Div id ing by the average c a l o r i f i c content of d i f fuse knapweed seeds ( 6 . 5 ± 0 . 1 ) , gives an estimated potent ia l seed production of 459 seeds. Compared with the observed number of seeds per plant ( 1 0 6 ± 1 1 ) , gives an estimated reduction in seed production as a resu l t of f l y attack of 77%. BASIS FOR CHOICE The observations are consistent with Zwolfer's (1970) observation that the g a l l f l i e s choose plants on the basis of the phys i ca l s tructure of the p lant . The numbers of f l i e s of both species observed on plants were d i r e c t l y proport ional to the number of buds per p lant . Moore (1978b) also found no r e l a t i o n s h i p between the in tens i ty of seed predation and plant s i z e . Both species of g a l l f l y preferred buds in the f i r s t branching category. Zwolfer's (1970) observations suggest that the proximal basis for s e l ec t ion is the locat ion of these buds at the ends of the branches, however primary buds are also more 50 su i table for l a r v a l growth. They have a higher energy content (Table 1.1), grow faster (Roze, 1981), and are less l i k e l y to abort buds for a given l eve l of f l y at tack. When the se lec t ion of plants and c lasses of buds i s combined with bud s ize se l ec t ion corresponding to l a r v a l requirements, the ov ipos i t ing g a l l f l i e s appear to be making a nearly "ideal" sequence of dec i s ions . Bud abort ion changes that; the sequence may s t i l l be "ideal" for i n d i v i d u a l g a l l f l i e s act ing alone, but i t i s not for the populat ion in aggregate. In the face of th i s density-dependent plant response, there i s a p o s s i b i l i t y for other behaviour to be incorporated into the ov ipos i t i on se lect ion process. Increased t e r r i t o r y s ize of males would reduce the o v e r a l l insect dens i ty . Reduced s e l e c t i v i t y would d i s t r i b u t e the insect attack more evenly (but may not reduce the proportion of buds aborted) . The use of marking pheromones (Prokopy, 1981) on probed buds or buds that had received eggs would also d i s t r i b u t e the insect attack more evenly (Monro, 1967). There may be severe r e s t r i c t i o n s on the number of p o t e n t i a l l y informative inputs the insects can respond to in order to make the "correct" d e c i s i o n , e s p e c i a l l y given the t ight space l imi ta t ions on the ir neural systems (Huber, 1974). For example, Morse and F r i t z (1982) found that not a l l spiders used the best foraging s i t e , p r i m a r i l y because of the simple stimulus employed for s i t e s e l e c t i o n . The apparent use by the g a l l f l i e s of a r e l a t i v e l y simple sequence of decis ions may be a good 51 compromise, d e s p i t e the impact i t has on r e p r o d u c t i v e s u c c e s s . INSECT INTERACTIONS S e l e c t i o n p a t t e r n s may d e v i a t e from the "optimal" i f i n s e c t s i n t e r a c t with one another, as Whitham (1978, 1980) has demonstrated f o r g a l l - f o r m i n g aphids. D e s p i t e the a g g r e s s i v e n e s s of g a l l f l y males towards one another (Berube and Myers, MS; p e r s . o b s . ) , I d e t e c t e d no e f f e c t of a g g r e s s i v e i n t e r a c t i o n s on the d i s t r i b u t i o n of males compared to females. T h i s behaviour i s u n l i k e l y to i n f l u e n c e the d i s t r i b u t i o n of o v i p o s i t i o n s because of the time l a g between mating and o v i p o s i t i o n ( p e r s . obs.; Zwolfer, 1970) and because males do not appear to guard buds which have r e c e i v e d eggs ( p e r s . obs.; Berube, p e r s . comm.). The " t e r r i t o r i a l i t y " of the males i s a very f l u i d one; i t appears to be p r i m a r i l y d i r e c t e d to o b t a i n i n g access to females, r a t h e r than defending p a r t i c u l a r o v i p o s i t i o n s i t e s . There are s e v e r a l p o s s i b l e reasons for t h i s t e r r i t o r i a l behaviour. Males may be c r e a t i n g space around themselves i n order to ensure access to females a t t r a c t e d by pheromone r e l e a s e (Smith and Prokopy, 1980). Borgia's (1981) o b s e r v a t i o n s of Scatophaga s t e r c o r a r i a L. ( D i p t e r a ) r a i s e another p o s s i b i l i t y . The males of that s p e c i e s that s u c c e s s f u l l y defend t e r r i t o r i e s tend to be l a r g e r and are chosen by females i n two ways: females move towards l a r g e males and towards areas where l a r g e males are common. Mate s e l e c t i o n based on c o n t r o l of o v i p o s i t i o n s i t e s has been observed i n b u l l f r o g s (Howard, 1978a, 1978b). D r a g o n f l i e s may a l s o d i s p l a y t e r r i t o r i a l i t y at the optimum 52 mating p e r i o d (Campanella, 1975). If males mating a f t e r another male have a h i g h sperm precedence, as found i n water bugs (Smith, 1979), T r i b o l i u m (Schlager, 1960), and D r o s o p h i l a melanogaster (Gromko and P y l e , 1978), m u l t i p l e mating i s l i k e l y . As a r e s u l t , there would be c o n t i n u i n g competition f o r females (Parker, 1970) and c o n t i n u i n g t e r r i t o r i a l behaviour by Urophora males. SUMMARY Both p l a n t s and buds v a r i e d i n the r e s o u r c e s they p r o v i d e d to the f l i e s . The two s p e c i e s of g a l l f l y , U. a f f i n i s and U. q u a d r i f a s c i a t a , made s i g n i f i c a n t l y d i f f e r e n t c h o i c e s among knapweed p l a n t s , among groups of buds on p l a n t s , and among buds. The c h o i c e s f o r both s p e c i e s at the p l a n t l e v e l were c o r r e l a t e d with the number of buds per p l a n t . W i t h i n p l a n t s , both s p e c i e s p r e f e r r e d buds in the primary branching c a t e g o r y . These were higher q u a l i t y buds and were l e s s l i k e l y to abort at a given l e v e l of f l y a t t a c k . The d i f f e r e n c e i n bud s i z e p r e f e r e n c e of the two g a l l f l y s p e c i e s p r e v i o u s l y noted i n the l a b o r a t o r y was a l s o observed i n the f i e l d . The c h o i c e s l e d to g a l l f ormation, bud a b o r t i o n , and reduced seed p r o d u c t i o n . The r e l a t i o n s h i p between i n s e c t c h o i c e s and these three outcomes h e l d both among and w i t h i n p l a n t s . G a l l p r o d u c t i o n was reduced by bud a b o r t i o n . 53 I I . THE EFFECT OF TIMING OF ATTACK ON THE POPULATION DYNAMICS OF TWO INTRODUCED INSECTS Gather ye rosebuds while ye may, Old Time i s s t i l l a - f l y i n g ; And t h i s same flower that smiles to-day, To-morrow w i l l be d y i n g . - Robert H e r r i c k The t i m i n g of i n s e c t a t t a c k r e l a t i v e to host p l a n t growth and development p l a y s a c r i t i c a l r o l e i n the dynamics of many i n s e c t - p l a n t systems (e.g. Dixon, 1976; Feeny, 1970; Green and Palmblad, 1975; M i l l e r and D i n g l e , 1982; Myers, 1981; Parker, 1984; Solomon, 1981; Sutton, 1984; Thompson and P r i c e , 1977), yet o n l y r a r e l y has i t been p o s s i b l e to e m p i r i c a l l y l i n k a d e t a i l e d d e s c r i p t i o n of the temporal processes with t h e i r impact on p o p u l a t i o n dynamics (e.g. van der Meijden, 1971.; M o r r i s , 1963; Roland, 1986). Females of the two i n t r o d u c e d i n s e c t s , Urophora a f f i n i s F r f l d . and U. quadr i f a s c i a t a (Meig.) ( D i p t e r a : T e p h r i t i d a e ) , lay eggs i n the immature flower buds of d i f f u s e knapweed, Centaurea  d i f f u s a Lam. ( A s t e r a c e a e ) . The young l a r v a e s t i m u l a t e g a l l formation w i t h i n the buds, thereby reducing seed p r o d u c t i o n . The d e n s i t i e s of o v i p o s i t i n g f l i e s and of s u i t a b l e o v i p o s i t i o n s i t e s change d u r i n g the summer. The emergence of a d u l t f l i e s from o v e r w i n t e r i n g g a l l s and t h e i r subsequent m o r t a l i t y d i c t a t e the i n s e c t d e n s i t y . The m o b i l i z a t i o n and a l l o c a t i o n of p l a n t resources determine the number of buds 54 s u i t a b l e f o r o v i p o s i t i o n . The r e l a t i v e d e n s i t y of i n s e c t s per bud should be r e f l e c t e d i n the d i s t r i b u t i o n of g a l l s among buds and hence in the d i s t r i b u t i o n of seeds. The numbers of g a l l s and seeds that r e s u l t i n a given year w i l l a f f e c t p o p u l a t i o n s i z e s i n the f o l l o w i n g y e a r s . The m o b i l i z a t i o n and a l l o c a t i o n of p l a n t r e s o u r c e s i s not independent of f l y a t t a c k . T h i s Chapter c o n s i d e r s three ways that the a l l o c a t i o n of p l a n t r e s o u r c e s may change as a r e s u l t of i n s e c t a t t a c k : reduced bud growth r a t e s , a lower p r o p o r t i o n of buds developed, and compensation f o r a t t a c k e d buds. F l y attack may suppress bud growth (Berube, 1980). Berube suggests that slower bud growth reduces U. q u a d r i fasc i a t a g a l l d e n s i t i e s r e l a t i v e to U. a f f i n i s . Many s p e c i e s of p l a n t s i n i t i a t e more buds than are developed and s e l e c t i v e l y a l l o c a t e r e s o u r c e s to a subset of i n i t i a t e d buds (Stephenson, 1980). Udovic and Aker (1981) argue that e a r l i e r i n i t i a t e d f r u i t s are more l i k e l y to be developed. A s i g n i f i c a n t p r o p o r t i o n of i n i t i a t e d buds on unattacked d i f f u s e knapweed p l a n t s are not developed. From a survey of s e v e r a l s i t e s i n B r i t i s h Columbia, Roze (1981) e s t i m a t e d an average of 14±3% (Mean±S.E.) of i n i t i a t e d buds were not developed. Attack by U. a f f i n i s can cause a b o r t i o n of e a r l y i n i t i a t e d buds that would normally be developed (Chapter I; Roze, 1981). T h i s pla n t response reduces the number of buds capable of producing e i t h e r g a l l s or seeds. Compensatory growth i s a widespread phenomenon i n p l a n t s subject to g r a z i n g or d e f o l i a t i o n (e.g. Bardner and F l e t c h e r , 55 1974; Chapin and Slack, 1979; Harper, 1977; H a r r i s , 1972; Pinthus and M i l l e t , 1978; Stephenson, 1981). Compensation is a s h i f t of plant resources into l a t e r - i n i t i a t e d leaves, buds, or f r u i t s (Stephenson, 1980; e .g . Hendrix, 1979). If d i f fuse knapweed compensates for insect attack, the rate of bud i n i t i a t i o n should increase following insect a t tack . Knapweed may also compensate for aborted buds by increasing resources to unaborted buds, r e s u l t i n g in faster bud growth or an increased p r o b a b i l i t y of f lowering for unaborted buds. The e f fect of the timing of insect attack on g a l l and seed production i s confounded by the effect of poss ib le plant responses. Experimental manipulation of insect dens i t i e s helps i so la te the two e f f e c t s . Increased f l y density should give more g a l l s per bud, increased proportions of buds aborted, and fewer seeds per bud. If U. a f f i n i s has a negative effect on U. quadri f a s c i a t a , as Berube (1980) argues, then increased density of both f l y species should lead to a r e l a t i v e reduction in U. quadr i fasc i a ta density . If plants compensate completely for insect a t tack , seed production per plant should be constant i r r e s p e c t i v e of insect densi ty . This Chapter (1) examines the timing of attack by two g a l l -forming f l i e s r e l a t i v e to the development of the i r host p lant , (2) describes three ways that insect attack may change plant development, (3) describes experimental manipulations of insect density to separate the effect of the r e l a t i v e timing of insect attack from the ef fect of a l tered plant resource a l l o c a t i o n , and (4) evaluates the effect of r e l a t i v e timing on the insects ' population dynamics. 57 MATERIALS AND METHODS OBSERVATION METHODS An observation p lot was es tabl i shed at Robertson's on June 1, 1980. A rectangular g r i d of f i f t y (5x10) points was placed over a 3 m x 6.5 m port ion of the f i e l d with a v i s u a l l y uniform density of d i f fuse knapweed. The plant which had begun to bolt nearest each point on the g r i d was staked. The points on the gr id were far enough apart so that staked plants were r a r e l y nearest neighbours. The staked plants provided a basis for d e t a i l e d observations of the f l i e s and the ir hosts . Approximately every two days, I conducted a branch by branch v i s u a l survey of each staked p l a n t . The loca t ion , species, sex, posture, and a c t i v i t y of each adul t f ly was recorded at the moment of observat ion . (See Chapter I for addi t iona l d e t a i l s of the f l y survey.) The number, l oca t ion , s i ze , and developmental status of a l l the buds on a l l the plants were recorded 19-21 times during the summer. Plant height, developmental s tate , and evidence of herbivory were also noted. The date on which a bud was f i r s t observed was used to place i t in an i n i t i a t i o n category (Table 2 .1) , roughly corresponding to a complete set of measurements of a l l buds. Some of the early sets of measurements are combined to obtain a reasonable sample for s t a t i s t i c a l comparisons. Bud s izes were measured with c a l i p e r s as described by Berube and H a r r i s (1978) to the nearest mi l l imeter . With some p r a c t i c e i t became possible to v i s u a l l y estimate s i ze s . Spot checks with c a l i p e r s confirmed the accuracy of these estimates. 58 Table 2.1 - Dates on which buds were f i r s t observed and corresponding bud i n i t i a t i o n categories , Robertson's 1980 Category Date 1 June 1- 9 2 June 11-13 3 June 15 4 June 1 7 5 June 19 6 June 23-25 7 June 27-29 8 Ju ly 2- 4 9 Ju ly 8-10 1 0 Ju ly 11-13 1 1 Ju ly 26 12 August 4- 7 1 3 August 11-13 1 4 August 20-24 In addi t ion to the method descr ibed in the General Methods (above), aborted buds may a lso be i d e n t i f i e d by the ir developmental h i s tory when that i s known. Range populations of knapweed plants t y p i c a l l y contain a s i g n i f i c a n t proportion of undeveloped buds that are i n i t i a t e d la te in the season. I assumed that any given bud had been aborted by the plant i f i t remained at a constant s ize for a per iod of t h i r t y or more days. PLANT COLLECTIONS Plants were c o l l e c t e d i n d i v i d u a l l y on August 23, 1980 by c l i p p i n g them off at ground l e v e l . They were then stored in folded and stapled paper bags at room temperature u n t i l d i s s e c t i o n . When they were d i s sec ted , the height of each plant was measured and the s i z e , developmental s tatus , and loca t ion of each bud were recorded. Buds large enough to 59 contain e i ther ga l l s or seeds were i n d i v i d u a l l y d i s sec ted . For these, the number of seeds and contents of any g a l l s present were noted. The g a l l s of the two insect species are e a s i l y d i s t ingu i shed; U. a f f i n i s g a l l s are hard and woody and U. quadri fasc iata g a l l s are th in and papery. The c o l l e c t i o n of plants at a s ingle point in time represents a tradeoff between loss of seeds from mature buds and an incomplete second generation of f l i e s . The s e n s i t i v i t y of the conclusions to the time of c o l l e c t i o n was evaluated by a second c o l l e c t i o n of twenty plants on September 12, 1980. These plants were the nearest plants to randomly se lected po in t s . Dif ferences between c o l l e c t i o n s are discussed in Appendix IIA. INSECT DENSITY MANIPULATION The f l y dens i t i es r e l a t i v e to o v i p o s i t i o n s i tes change throughout the summer as does the a l l o c a t i o n of resources by the p l a n t s . To separate these two e f fec t s on g a l l and seed product ion, I constructed enclosures with d i f f eren t dens i t ies of f l i e s at Robertson's in 1980. Nine enclosures were used, three at each of three insect d e n s i t i e s . Each 1.5 m3 enclosure was b u i l t of f ibreg lass net t ing s tretched on a wooden frame. The bottom edge of the nett ing was buried in the ground. Because e a r l i e r experiments using s i m i l a r net t ing had demonstrated a s i g n i f i c a n t ef fect of shading on the growth of d i f fuse knapweed plants (Berube, pers . comm.; pers . o b s . ) , the enclosures were l e f t open at the top. This arrangement d i d not completely prevent changes in plant q u a l i t y as a resu l t of the enclosures (see Appendix IIB) , but minimized such changes 60 while preserving d i f ferences in adult f ly density among enclosures (see below). G a l l f l y dens i t i e s were manipulated by moving dead stem plants from the previous year containing diapausing larvae . A l l v i s i b l e o ld plants and seed heads were removed from the low density enclosures . Each of the three high density enclosures received the o ld plants and seed heads from one of the low density enclosures . The o ld plants in the contro l density enclosures were l e f t in p lace . These manipulations were performed on June 1, 1980, p r i o r to the emergence of any g a l l f l i e s . Relat ive counts of f l i e s in the c e l l s were obtained on three d i f f er e n t days (June 27 and 28 and July 10) to ensure that the movement of stem plants among enclosures had var ied f l y d e n s i t i e s . (The counts of g a l l f l i e s in the low density enclosures suggests that some g a l l f l i e s emerged from seed heads which escaped de tec t ion . ) I recorded the numbers of f l i e s seen in each enclosure during a three minute v i sua l survey. Surveys were begun between 11:30 a.m. and 2:30 p.m. and took a t o t a l of approximately t h i r t y minutes for a l l enclosures. Differences in the timing of surveys among days would not af fect among enclosure comparisons on the same day. Species and sex could not be r e l i a b l y determined through the net t ing , however there was no reason to bel ieve that e i ther species composition or sex r a t i o d i f f e r e d among enclosures . Plants were c o l l e c t e d from each enclosure on August 16, 1980, af ter the f i r s t generation of f l i e s . Seed production was 61 incomplete by th i s date so seed counts are not comparable to unenclosed p lants , however among treatment comparisons are not af fected by the ear ly c o l l e c t i o n date. The e f fect of the enclosures on f l y attack is assessed in Appendix IIB. Plants were stored and dissected as described above. 62 RESULTS INSECT ATTACK AND BUD INITIATION INSECT ATTACK The data from the f i r s t generation of g a l l f l i e s suggest that U. a f f i n i s males emerged two to three days e a r l i e r than females, mating occurred as soon as the females emerge, and ov ipos i t i on followed a few days la ter (Table 2 .2) . These Table 2.2 - Mean day of observation for d i f f erent categories of Urophora f l i e s on d i f fuse knapweed p lant s , Robertson's 1980 Insect Day of Observation N Urophora a f f i n i s Males 24. . 9 ± 0 , .3 * 689 Females 27. , 3 ± 0 , .4 275 Mating f l i e s 27. . 2 ± 0 . .7 74 Probing f l i e s 30, .0±1 , . 1 32 A l l f l i e s 25. , 5 ± 0 . .2 1046 Urophora q u a d r i f a s c i a t a A l l f l i e s 25. , 0 ± 0 , .8 81 - Days from June 1 ( M e a n ± S . E . ) observations on plants in the f i e l d agree with Zwolfer's (1970) observations in the laboratory in terms of the sequence of events, though he found that males emerged only one day before females. The mean dates for U. a f f i n i s and U. quadr i fasc ia ta were s imi lar in the f i r s t generation (t=0.827, df=l049, p=0.408; Figure 2 .1) . In the second generat ion, U. quadr i fasc iata emerged e a r l i e r 63 Figure 2.1. Counts on f i f t y d i f fuse knapweed plants of the two species of g a l l f l y at Robertson's in 1980. Ju ly 25 d iv ides the f i r s t and second generations of g a l l f l i e s . 65 than U. a f f i n i s (Figure 2.1) and reproduced sooner. A l l of the observed probings of buds by U. quadr i fa sc ia ta in the second generation preceded those by U. a f f i n i s (Mann-Whitney, U=0, N=11, p=0.006). Seven out of the eight observed U. quadr i fasc ia ta matings came before any U. a f f i n i s matings (Mann-Whitney, U=13, N=21, 0.002<p<0.02). BUD INITIATION The r e l a t i v e magnitude of the standard errors in Figure 2.2 indicate that the timing of bud i n i t i a t i o n and the f i n a l numbers of buds per plant var ied cons iderably among di f fuse knapweed plants in 1980. Despite sharp changes in the f ly attack (Figure 2.1) and r a i n f a l l (Figure 4 .2) , there was a roughly l inear increase in bud numbers during the f i r s t generation of adult f l i e s . INTERACTION The outcomes of f l y attack on buds i n i t i a t e d throughout the summer are shown in Figures 2 .3-2 .6 . The number of U. a f f i n i s g a l l s per developed bud was greater in e a r l i e r i n i t i a t e d buds and decl ined sooner than would be expected on the basis of f ly abundance (Figure 2 .3) . The d i f ference in the timing of the peak of f l y abundance and the peak of g a l l s per developed bud is p a r t l y due to the length of time between bud i n i t i a t i o n and the point at which buds reach a su i tab le s ize range for o v i p o s i t i o n . The time to reach a su i tab le s i ze for ov ipos i t ion var ied throughout the summer because bud growth rates changed with locat ion of the bud on the plant and with f l y attack (see below). Based on bud growth rates at the peak of Figure 2.2. Average numbers of buds on d i f fuse knapweed plants at Robertson's in 1980. Each point gives the mean number of buds on f i f t y plants ± one standard e r r o r . 68 Figure 2.3. Mean number of U. a f f i n i s g a l l s per developed bud by bud i n i t i a t i o n category for Robertson's in 1980. V e r t i c a l l i n e s give ± one standard error for g a l l s per developed bud. Interpolated counts of U. a f f i n i s adults on the same f i f t y plants in the same i n i t i a t i o n categories are a lso shown. Counts were interpolated from the data in Figure 2.1. Bud i n i t i a t i o n categories are defined in Table 2 .1 . 1.75J Bud initiation category 70 Figure 2.4. P r o b a b i l i t y of bud abort ion by bud i n i t i a t i o n category for Robertson's in 1980. V e r t i c a l l i n e s give ± one standard error for the proport ion of buds aborted. Interpolated counts of U. a f f i n i s adults on the same f i f t y plants in the same i n i t i a t i o n categories are a lso shown. Counts were interpolated from the data in Figure 2 .1 . Bud i n i t i a t i o n categories are defined in Table 2.1. The dotted l i n e between bud i n i t i a t i o n categories 10 and 11 indicate the assumed change in the proportion of buds aborted. It was not poss ib le to d i s t i n g u i s h between aborted and undeveloped buds in i n i t i a t i o n category 11. L.120 Bud initiation category 72 Figure 2.5. Mean number of U. quadr i fasc iata g a l l s per developed bud by bud i n i t i a t i o n category for Robertson's in 1980. V e r t i c a l l ines give ± one standard error for g a l l s per developed bud. Interpolated counts of U. a f f i n i s adults on the same f i f t y plants in the same i n i t i a t i o n categories are also shown. Counts were in terpo la ted from the data in Figure 2.1. Bud i n i t i a t i o n categories are defined in Table 2 .1 . Bud initiation category — i 74 Figure 2.6. Mean number of seeds per developed bud by bud i n i t i a t i o n category for Robertson's in 1980. V e r t i c a l l ines give ± one standard error for seeds per developed bud. Interpolated counts of U. a f f i n i s adults on the same f i f t y plants in the same i n i t i a t i o n categories are also shown. Counts were in terpo la ted from the data in Figure 2.1, Bud i n i t i a t i o n categories are defined in Table 2 .1 . 76 g a l l formation and the s ize of probed buds, the time to reach a su i tab le s ize range for ov ipos i t ion probably corresponds to a d i f ference of between one and two bud i n i t i a t i o n categor ies . This d i f ference is not enough to explain the contrast between the peaks in f l y abundance and g a l l formation. A s i m i l a r time i n t e r v a l to reach a su i tab le s ize for probing does account for the di f ference in the timing of bud abort ion and f l y abundance (Figure 2 .4) . These data on the r e l a t i v e t iming of f l y abundance, g a l l formation, and bud abort ion indicate that changes in bud abortion coincided with changes in f l y abundance and suggests that bud abort ion reduces g a l l formation by U. a f f i n i s . Despite the synchronous f i r s t generation of U. af f i n i s and U. q u a d r i f a s c i a t a adul t s (Table 2.2) , U. quadr i fasc ia ta g a l l production dropped c lose to zero during the peak of U. a f f i n i s adult abundance (Figure 2 .5) . This drop was negatively re la ted to bud abor t ion . The e a r l i e r second generation of U. quadr i fasc ia ta r e l a t i v e to U. a f f i n i s may be explained by the suppression of U. quadr i fasc i a ta g a l l formation during the peak of the f i r s t generat ion. The f l i e s emerging in the second generation were the o f f s p r i n g of the ear ly emerging and ear ly o v i p o s i t i n g U. quadr i fasc ia ta that successful ly produced g a l l s . The number of seeds per developed bud was negatively re la ted to the counts of U. a f f i n i s on staked p l a n t s . Seed production peaked af ter the maximum of U. a f f i n i s abundance (Figure 2 .6 ) . There appears to have been a seed refuge late in 77 the season. CHANGES IN PLANT ALLOCATION BUD GROWTH AND DEVELOPMENT Buds on primary and secondary branches grew more slowly than buds on higher order branches (Figure 2 .7) . The opposite trend holds for unattacked di f fuse knapweed. Roze (1981) found that d i s t a l buds took 3 2 . 9 ± 0 . 5 days to flower compared with 4 5 . 6 ± 0 . 6 days for proximal buds. Thus growth of developed buds is slowed by insect a t tack . COMPENSATORY REPRODUCTION Plants could have compensated for aborted buds in three poss ib le ways: (1) bud i n i t i a t i o n could have increased af ter f ly at tack, (2) for every bud aborted another bud could have been developed, or (3) the buds l a t e r a l to the aborted buds could have grown faster and been more l i k e l y to develop. The f i r s t p o s s i b i l i t y , increased bud i n i t i a t i o n , may be ruled out, since no change in the bud i n i t i a t i o n rate corresponding to changes in f l y attack was observed (Figure 2.2) . To test the second mechanism, I examined the r e l a t i o n s h i p between the proport ion of buds aborted per plant (PA) and the proportion of buds developed per plant (PD) at the end of the summer. Assuming that the t o t a l number of buds per plant was independent of f l y at tack, then buds which aborted as a result of insect attack ( instead of developing) should have been 78 Figure 2 . 7 . D i s t r i b u t i o n of phenology measures across branching categor ies . Date i n i t i a t e d and date flowered are in days from June 1. Time to flower i s the d i f f erence ( in days) between i n i t i a t i o n and f lowering. Date i n i t i a t e d is for a l l buds in a given branching category. Date flowered and time to flower are for only those buds which flowered in a given branching category. V e r t i c a l l ines give ± one standard e r r o r . 80. 80 compensated for by buds which developed ( instead of remaining undeveloped). The resul t would be an increase in PA, no change in PD, and a decrease in the proportion of undeveloped buds per plant (unless t h i s l a t t e r proportion reached zero) . If the second mechanism was operating, a plot of the proport ion of buds aborted (PA) against the proportion of buds developed (PD) should have a slope not s i g n i f i c a n t l y d i f f erent from zero. The observed r e l a t i o n s h i p i s : PD = - 0 . 7 0 2 ( ± 0 . 0 8 2 ) P A + 0 . 6 4 6 ( ± 0 . 0 2 5 ) , F=73.0, df=1,47, p<0.00l, r=-0.78. The slope of the r e l a t i o n s h i p i s s i g n i f i c a n t l y less than zero. Compensatory reproduction does not appear to have been operating in th i s case. If the t h i r d mechanism was ac t ing , then the speed of development and p r o b a b i l i t y of flowering of l a t e r a l buds would have been p o s i t i v e l y affected by abortion of the terminal bud. If the p r o b a b i l i t y of a bud aborting is negatively re la ted to the resources a v a i l a b l e , then l a t e r a l buds would have been less l i k e l y to abort i f the terminal bud aborted. L a t e r a l buds nearest to terminal buds would have been the most d i r e c t l y a f fected by abort ion of the terminal bud since e f fects on other buds would not have been dis t inguishable from other influences on growth. I d iv ided these l a t e r a l buds into those which had an aborted terminal bud and those which d id not. The time to flower was not s i g n i f i c a n t l y d i f f erent between the two groups (13 .91±0 .69 days (N=699) with the terminal bud not aborted vs . 1 3 . 0 7 ± 0 . 7 3 days (N=579) with the terminal bud aborted) . The 81 p r o b a b i l i t y of the l a t e r a l bud abort ing was also not s i g n i f i c a n t l y d i f f erent (x 2=2.23, df=1, p=0.135) and the trend was in the opposite d i r e c t i o n from that predic ted (0.29 vs . 0.33). The p r o b a b i l i t y of the l a t e r a l bud flowering behaved as predicted (x 2=6.47, df=1, P=0.011), but the effect was r e l a t i v e l y small (0.36 vs . 0.43), a d i f ference of only 16%. This was the only evidence for plant compensation. INSECT DENSITY MANIPULATION There were s i g n i f i c a n t d i f ferences in the counts of f l i e s among treatments for the f l y density manipulations (Table 2.3) . The Table 2.3 - Counts of Urophora f l i e s observed in density enclosures, Robertson's 1980 INSECT DENSITY * Date High Contro l Low June 27 9,11,16 4,8 1,3,6 June 28 19,21,23 10,12 5,5,9 July 10 0,2,3 0,3 0,1,1 * Mann-Whitney U-tests were used for pairwise comparisons of sets of counts. Results of tests for June 27 and June 28 were combined using F i s h e r ' s procedure (Sokal and Rohlf, 1969). Low vs. C o n t r o l , X a = 7 . 8 2 , df=4, p=0.098; Control vs . High, ;X.2 = 9.21, df=4, p=0.056; Low vs. High, =11.92, df=4, p=0.0l8. di f ference between the counts at the end of June and the counts in the middle of July agrees with the counts of insects on plants outside of the enclosures (Figure 2 .1 ) . 82 The number of buds per plant d id not vary s i g n i f i c a n t l y among treatments (F=0.68, df=2,54, p=0.511; Table 2.4) . Table 2.4 - Ef fect of g a l l f ly density manipulations on d i f fuse knapweed c h a r a c t e r i s t i c s , g a l l production, seed production, and bud abortion Character High INSECT DENSITY Control Low Number Buds/Plant Dev . /P lant 62 21 22 . 5 ± 7 . 6 . 2 ± 2 . 8 18 6 7 . 3 ± 7 . 3 2 4 . 9 ± 2 . 9 1 7 77 . 8 ± 1 0 . 1 2 7 . 7 ± 4 . 2 UA/Dev. UQ/Dev. Seeds/Dev. 1 . 0. 0. 1 0 ± 0 . 0 6 0 3 ± 0 . 0 1 5 3 ± 0 . 0 7 0 . 8 9 ± 0 . 0 6 0 . 0 5 ± 0 . 0 2 0 . 7 9 ± 0 . 0 8 0 . 7 3 ± 0 . 0 5 0 . 0 8 ± 0 . 0 2 1 . 3 2 ± 0 . 11 Prop Aborted Prop Undev. 0. 0. 1 1 ± 0 . 0 1 2 5 ± 0 . 0 2 0. 1 0 ± 0 . 0 1 0 . 2 8 ± 0 . 0 2 0 . 1 3 ± 0 . 0 2 0 . 2 6 ± 0 . 0 3 Seeds/Plant 11 . 4 ± 3 . 3 1 9 . 6 ± 4 . 0 3 6 . 6 ± 7 . 8 Prop UQ * 0.02 0.06 0.10 Prop Unatt .** .0.43 0.51 0.51 * Proport ion of U. quadr i fasc ia ta g a l l s of a l l g a l l s . ** Proport ion of developed buds unattacked by e i ther species of g a l l f l y . G a l l formation by U. a f f i n i s tracked the change in adult dens i ty . The number of U. a f f i n i s g a l l s per developed bud in the high density enclosures was higher than in e i ther the c o n t r o l (t=2.63, df=9l4, p=0.009) or low density enclosures (t=5.00, df=905, p<0.00l). The number of U. a f f i n i s g a l l s per developed bud in the contro l enclosures was also greater than in the low density enclosures (t=2.!7, df=882, p=0.030). The 83 changes were p r o p o r t i o n a l l y smaller than the changes in the observed adult f l y dens i ty . If male t e r r i t o r i a l i t y and movement of f l i e s among enclosures had resu l ted in a more uniform d i s t r i b u t i o n of males among enclosures than for females, the di f ferences in g a l l s per developed bud among treatments would have been greater than the d i f ferences in adult dens i t i e s . The proport ion of buds aborted was not s i g n i f i c a n t l y d i f f erent among treatments (F=0.96, df=2,54, p=0.389). Possible reasons for th i s include the enclosure ef fects discussed in Appendix IIB. The attack by U. quadr i fasc i a ta and the seed production also responded to the change in t o t a l adult f l y dens i ty . G a l l formation by U. quadr i fasc ia ta dropped with increased adult density (high vs . low; t=2.42, df=723, p=0.U16). The proportion of U. quadr i fasc ia ta g a l l s of a l l g a l l s was s i g n i f i c a n t l y af fected by the change in densi ty (x2=27.68, df=2, p<0.00l). The comparisons of the c o n t r o l density with the two extremes were also s i g n i f i c a n t (x2>6.88 in both cases, df=1, p<0.0l). Plants in the high f l y densi ty enclosures had fewer seeds per developed bud than e i ther the contro l plants (t=2.43, df=898, p=0.0l5) or the plants in the low density enclosures (t=6.25, df=8l9, p<0.00l). The seed production per developed bud for contro l plants was a l so s i g n i f i c a n t l y greater than seed production per developed bud for p lants in the low density enclosures (t=4.06, df=860, p<0.00l). The proportion of buds unattacked by e i ther species of g a l l f l y var ied s i g n i f i c a n t l y with insect density (x2=7.33, df=2, 84 p=0.026). This comparison indicates that the s ize of the seed refuge (Figure 2.6) depends on insect dens i ty . The change in s ize of the seed refuge was not l inear with the change in insect dens i ty . There was no addi t iona l evidence of compensation for insect a t tack . The t o t a l number of seeds per plant was negat ively re la ted to insect density (Table 2 .4) . There was a lso no evidence that plants reduced the proportion of undeveloped buds to compensate for attacked buds (F=0.38, df=2,54, p=0.686; Table 2 .4 ) . 85 DISCUSSION CHANGES IN INSECT DENSITY As a resu l t of manipulation of the g a l l f l y d e n s i t i e s , r e l a t i v e l y small changes in the number of g a l l s per bud were observed. S i m i l a r l y , large changes in insect density during the season had r e l a t i v e l y s l i gh t ef fects on g a l l formation. The same kind of discrepancy between v a r i a t i o n in adult population s ize and v a r i a t i o n in l a r v a l population s ize was observed by Hirose et a l . (1980) for the c i t r u s swal lowta i l . The density manipulation experiment a lso showed that within the density range I observed there was no c e i l i n g on g a l l dens i ty , since an increase in the number of f l i e s led to an increase in the number of g a l l s , despite s i m i l a r numbers of buds per plant and (presumably) s i m i l a r phenologies . The drop in U. a f f i n i s g a l l s per developed bud pr ior to the decl ine in f l y density (Figure 2.3) suggests that some factor reduced the reproductive success of the g a l l f l i e s in the l a t t e r part of the season. The preference of probing females for buds in the primary branching category observed in Chapter I could explain the drop, however the time between the i n i t i a t i o n of the buds that were probed and the time of the observed probes d id not change s i g n i f i c a n t l y during the season; i t was 7 . 9 4 ± 0 . 9 1 days in the f i r s t hal f of the season and 8 . 6 9 ± 1 . 3 7 days in the second hal f of the season. These data suggest that there was a r e l a t i v e preference for e a r l i e r i n i t i a t e d buds, not an absolute preference. The decl ine in U. a f f i n i s g a l l production could also be 86 explained by reduced probing and ov ipos i t i on l a t e r in the summer. This p o s s i b i l i t y may be tested by comparing the proport ion of observed females seen probing in the f i r s t and second halves of the season. The proport ion was a c t u a l l y greater in the second hal f of the season (x 2=8.92, df=1, p=0.003). Thus g a l l formation dropped despite increased probing, because the proportion of probes r e s u l t i n g in ov ipos i t ion dec l ined , because egg morta l i ty was higher , or because larvae from eggs l a i d late in the season were less l i k e l y to produce g a l l s . Because aborted buds remain on the plant and remain in the s ize range for ov ipos i t i on by U. a f f i n i s (Chapter I ) , aborted buds w i l l form an increasing proportion of the populat ion of su i tab ly s ized buds. This cumulative e f fect could" cause the observed drop in g a l l formation over time in two ways. The f i r s t poss ible mechanism is that aborted buds act as an egg "sink" re su l t ing in egg "wastage" (Monro, 1967). The second poss ible mechanism is that the accumulation of aborted buds increases the search time for su i table o v i p o s i t i o n s i t e s ( i . e . unaborted buds). Both mechanisms are consistent with the observed density-dependence of bud abortion and r e l a t i v e l y small changes in g a l l formation with large changes in insect dens i ty . These a l t ernat ive mechanisms are explored further in the next Chapter. This cumulative ef fect of bud abortion w i l l r e s t r i c t the reproduction, not only of U. a f f i n i s , but a lso of U. quadr i fasc i a t a . While aborted buds are not, in general , in 87 the su i table s ize range for ov ipos i t i on by U. q u a d r i f a s c i a t a , by preventing growth to larger s i z e s , bud abortion w i l l reduce the a v a i l a b i l i t y of buds for th i s species . Because U. a f f i n i s and U. quadr i fasc ia ta have had synchronous phenologies in the f ive years in which these have been observed (1976, 1977, Roze (1981); 1978, Berube, pers . comm.; 1979, Berube, pers . comm. and pers . obs . ; 1980, pers . o b s . ) , the suppression of g a l l production during the f i r s t generation of U. quadr i fasc ia ta by bud abort ion and by preemption of ov ipos i t ion s i t es by U. a f f i n i s w i l l be a consistent l i m i t i n g factor on the population density of U. quadri fasc i a t a . Addi t iona l evidence for i n t e r s p e c i f i c competit ion between the two g a l l f l y species i s given in Chapter V. SEED REFUGE A high degree of synchrony exis ts between the g a l l f l i e s and bud i n i t i a t i o n and development by the ir host p lant s . Roze (1981) observed bud i n i t i a t i o n and g a l l f l y emergence patterns in 1976 and 1977 at the two release s i t e s . The la ter bud i n i t i a t i o n in 1977 was p a r a l l e l e d by a la ter peak in f l y emergence. Berube suggests that the development of plants in 1979 was about two weeks in advance of 1978 at both Ned's Creek (dif fuse knapweed) and Chase (spotted knapweed). Again, bud i n i t i a t i o n and f l y emergence sh i f ted in the same d i r e c t i o n between the two years . F i n a l l y , despite a d i f ference of twelve days in the peak of U. a f f i n i s abundance between 1974 and 1975 in Montana, the degree of synchrony between g a l l f l y emergence and spotted knapweed bud i n i t i a t i o n remained the same (Story and 88 Anderson, 1978). This synchrony implies that the s ize and timing of the seed refuge may be r e l a t i v e l y constant. Hence a c e r t a i n proport ion of the t o t a l po tent ia l seed production (pr ior to g a l l f l y attack) w i l l be ava i lab le to the next generation of p lants . Chapter V describes a technique to estimate the s ize of seed refuges from g a l l d i s t r i b u t i o n s and appl ies i t to h i s t o r i c a l data at the o r i g i n a l release s i t e s for the g a l l f l i e s . A refuge for knapweed seed production ex i s t s in several p a r a l l e l systems. Story (pers. comm.) found that at the peak density of U. a f f i n i s in Montana only 62% of spotted knapweed buds were attacked. Zwolfer (1978) observed that in the samples of C. maculosa flower buds he c o l l e c t e d in Europe the percentage of buds attacked by any insect larvae d i d not exceed 48%. This includes the Upper Rhine Va l l ey population where U. a f f i n i s const i tuted greater than 70% of a l l insect larvae , and which was the one of the sources for the Canadian U. aff i n i s populat ions . Varley (1947) observed such a refuge for C. nemoralis attacked by U. jaceana. "Preliminary census work in over t h i r t y d i f ferent l o c a l i t i e s in England and Wales showed no sample with more than 48% of the flower heads conta in ing g a l l s . " U. s o l s t i t i a l i s F . also has a low proport ion of attacked buds (c. 43%) and a highly clumped d i s t r i b u t i o n in Carduus nutans L . in Europe (Zwolfer, 1979). COMPENSATORY REPRODUCTION Based on a comparison among years , Roze and Frazer (1978) argue that d i f fuse knapweed compensates 89 for f l y a t tack . They suggest that proximal undeveloped buds are developed when d i s t a l buds are aborted. Their data do not support t h e i r c l a i m . If for each d i s t a l bud aborted a proximal bud was developed, then the proportion of undeveloped buds ( inc luding d i s t a l aborted buds and proximal undeveloped buds) should remain constant . As the attack rate increased from 1975 to 1977, the proport ion of buds undeveloped increased. My data show that there is some compensation for bud abort ion in the form of a small di f ference in the p r o b a b i l i t y of flowering of l a t e r a l buds. Despite heavy f r u i t losses to insect attack by Haplopappus squarrosus H. and A . , no compensation was observed in that shrub (Louda, 1982). The a b i l i t y to compensate for insect attack may be spec i e s - spec i f i c ; sycamore leaves compensate for aphid attack (Dixon, 1971a), while lime leaves do not (Dixon, 1971b). EVOLUTIONARY CONSEQUENCES The timing of g a l l f l y emergence r e l a t i v e to knapweed bud i n i t i a t i o n may be subject to evolut ionary change. In the t r a d i t i o n a l view of microevolut ionary change, three factors are required: phenotypic v a r i a t i o n , a genetic basis for th i s v a r i a t i o n , and d i f f e r e n t i a l reproduction or s u r v i v a l of the v a r i a n t s . The f i r s t two ingredients are almost c e r t a i n l y present. There is considerable v a r i a t i o n in emergence time of the g a l l f l i e s (Figure 2.1) and emergence time and developmental rate of insects are probably her i tab le (Richards and Myers, 1980; Tay lor , 1981). The t h i r d f a c t o r , d i f f e r e n t i a l reproduction, has at least 90 three components. The f i r s t i s the synchrony with food plants (e .g . Mooney et_ a l . , 1981). Buds i n i t i a t e d ear ly in the season are higher q u a l i t y and can support more g a l l s than buds i n i t i a t e d la ter in the season. The second component is the degree of v o l t i n i s m . Other things being equal , a g a l l f l y with two generations a year w i l l increase in frequency in a population r e l a t i v e to g a l l f l i e s with only a s ing le generation every year. This increase requires su i tab le condit ions for reproduction la te in the season. Kingsolver (1979) concluded that a mixed strategy for v o l t i n i s m in the p i tcher plant mosquito, Wyeomyia smithi i C o q . , was opt imal . This mixed strategy may also be appropriate for the g a l l f l i e s because reproductive p o s s i b i l i t i e s l a t er in the summer may be quite v a r i a b l e . The t h i r d component in d i f f e r e n t i a l reproduction is a frequency-dependent one. Adults which emerge during the peak of f l y abundance w i l l be faced with large numbers of aborted buds which w i l l d i l u t e the population of su i tab le unaborted buds. Disrupt ive se lec t ion would resul t in an increased variance of emergence time. Natural s e l ec t ion may also act on the t iming of resource a l l o c a t i o n by the p l a n t s . I have shown that i n d i v i d u a l plants vary considerably in the timing of bud i n i t i a t i o n and in the number of buds i n i t i a t e d . The plants may be considered to be sampling the i r environment to determine when to i n i t i a t e buds. The number of seeds produced by the plants I followed varied from two to 347. This v a r i a t i o n in reproduction w i l l be corre lated with the v a r i a t i o n in the timing of bud i n i t i a t i o n 91 when the plants are attacked by the f l i e s . If the f l i e s attack ear ly in the summer, plants which i n i t i a t e buds la ter w i l l produce more seeds per bud (assuming equal resource a v a i l a b i l i t y ) . Knapweed may also be selected for the best time to d i sp lay flowers in order to maximize p o l l i n a t i o n (cf . Stapanian, 1982). This system is not one in which se lec t ion on e i ther the f l i e s or plants are constant in d i r e c t i o n or magnitude. Rather, because the other organism is the most important se l ec t ive inf luence , the system w i l l coevolve. This process of coevolut ion has been explored in models by several authors (e .g . Levin and Udovic , 1977; reviewed by S la tk in and Maynard Smith, 1979). Schaffer and Rosenzweig (1978) predic ted that a coevolving predator-prey system could approach eco log i ca l s t a b i l i t y i f the prey turned over more qu ick ly than the predators . This condit ion c l e a r l y does not hold in th i s system where f l i e s may have two generations a year and the plants are at least b i e n n i a l . However, because the f l i e s may reproduce twice a year under very d i f f erent condi t ions , c o n f l i c t i n g se l ec t ion pressures may counteract the ir more rapid turnover. SUMMARY The v a r i a t i o n in r e l a t i v e density of g a l l f l i e s per bud during the summer s i g n i f i c a n t l y changed the outcome of the i n t e r a c t i o n ; increased U. a f f i n i s dens i t ies led to increased U. a f f i n i s g a l l s per developed bud, increased bud abort ion , decreased 92 # U. quadr i fasc ia ta g a l l s per developed bud, and decreased seed product ion. The number of U. a f f i n i s g a l l s per developed bud was lower in the second hal f of the f i r s t generation than in the f i r s t h a l f . The timing of f l y attack in r e l a t i o n to bud i n i t i a t i o n led to a refuge in time for seed product ion. Two ef fects of changing a l l o c a t i o n of plant resources were observed in addi t ion to bud abor t ion . Times from bud i n i t i a t i o n to flowering were s i g n i f i c a n t l y reduced by insect attack. Only a s l i gh t change in the p r o b a b i l i t y of f lowering of buds l a t e r a l to aborted buds was detected. This Chapter also examined the e f fect of changed insect dens i t i es on bud abort ion , g a l l product ion , and seed production. Experimental manipulations of f l y dens i t i e s were confounded by se lec t ive attack by grasshoppers on enclosed p lant s , however the number of U. a f f i n i s g a l l s per developed bud and seed numbers in developed buds followed the expected trends . The number of U. quadr i fasc ia ta g a l l s per developed bud dropped with increased t o t a l adult f l y dens i ty . 93 APPENDIX U A . EFFECT OF COLLECTION DATE The des truct ive sampling of systems that are changing in time creates p o t e n t i a l l y serious problems for understanding the system. The timing of the sample becomes an important methodological issue. In the Urophora- Centaurea system, c o l l e c t i o n s at the end of the summer should be timed to avoid i n t e r r u p t i n g reproduction by the second generation of f l i e s and to minimize loss of seeds from open seed heads. The second c o l l e c t i o n of d i f fuse knapweed was d i f f e r e n t from the e a r l i e r c o l l e c t i o n (Table 2.5) . It had a greater number of buds per plant (though not s i g n i f i c a n t l y so) . The number of U. a f f i n i s ga l l s per developed bud was lower in the second c o l l e c t i o n , possibly a r e f l e c t i o n of the greater number of buds. The proportion of buds attacked by U. a f f i n i s i s lower in the second c o l l e c t i o n (X 2=12.52, df=1, p<0.00l). The number of U. quadr i fasc ia ta g a l l s per developed bud was higher , p r i m a r i l y as a resul t of an increased number of buds attacked by U. q u a d r i f a s c i a t a (x 2=32.64, df=1, p<0.00l), rather than higher g a l l counts in attacked buds. For both insect species , the number of g a l l s per bud did not d i f f e r s i g n i f i c a n t l y between c o l l e c t i o n s , i f just buds attacked by that species were cons idered. Taken together, these observations suggest that some developed buds were attacked by the second generation of U. quadri fasc ia ta after the f i r s t c o l l e c t i o n , at about the same rate as in the f i r s t generation. Table 2.5 - Effect of plant c o l l e c t i o n date on di f fuse knapweed c h a r a c t e r i s t i c s and g a l l f l y attack, Robertson's 1980 Character * F i r s t C o l l e c t i o n Second Co l l ec t i on (August 23) (September 12) Number of Plants 50 20 Buds/Plant 7 1 . 7 ± 6 . 0 7 4 . 9 ± 1 3 . 6 Dev . /P lant 3 0 . 7 ± 2 . 4 3 5 . 2 ± 7 . 1 Dev./Buds 0 . 4 4 ± 0 . 0 2 0 . 4 6 ± 0 . 0 3 UA/Plant 3 6 . 9 ± 4 . 0 3 6 . 0 ± 9 . 8 UA Buds/Plant 16.9+1.5 1 6 . 4 ± 4 . 2 UA/Dev. 1 . 2 0 ± 0 . 0 4 1 . 0 2 ± 0 . 0 6 UA/Attacked 2 . 1 8 ± 0 . 0 5 2 . 2 0 ± 0 . 0 8 Prop Att UA 0 . 5 5 ± 0 . 0 1 0 . 4 7 ± 0 . 0 2 UQ/Plant 8 . 3 ± 1 .3 17.1+3.7 UQ Buds/Plant 4 . 5 ± 2 . 1 8 . 7 ± 2 . 9 UQ/Dev. 0 . 2 7 ± 0 . 0 2 0 . 4 9 ± 0 . 0 4 UQ/Attacked 1 . 8 4 ± 0 . 0 7 1 . 9 6 ± 0 . 1 1 Prop Att UQ 0 . 1 5 ± 0 . 0 1 0 . 2 5 ± 0 . 0 2 Seeds/Plant 106±11 138±22 Seed Heads/Plant 1 7 . 6 ± 1 .6 2 2 . 5 ± 4 . 3 Seeds/Dev. 3 . 4 7 ± 0 . 1 0 3 . 9 3 ± 0 . 1 6 Seeds/Produc ing 6 . 0 3 ± 0 . 12 6 . 1 4 ± 0 . 18 * This Chapter and the fo l lowing Chapters contain several tables with th i s format. The plant characters include: Chewed/Plant (=number of buds damaged by chewing per p l a n t ) , Chewed/Buds (=buds damaged by chewing as a proportion of a l l buds), Dev . /P lant (=number of developed buds per p l a n t ) , UA/Plant (=number of U. a f f i n i s ga l l s per p l a n t ) , UA/Dev. (=number of U. a f f i n i s g a l l s per developed bud), UA/Attacked (=density of U. a f f i n i s g a l l s in a l l buds containing U. aff i n i s g a l l s ) , Prop Att UA (=proportion of developed buds containing U. a f f i n i s g a l l s ) , Seeds/Producing (=density of seeds in a l l buds containing seeds), and Prop. Aborted (=aborted buds as a proport ion of a l l buds). 95 APPENDIX I IB. EFFECT OF DENSITY ENCLOSURES Comparison of the plants in the contro l density enclosures with adjacent unenclosed plants demonstrates that the enclosures had no s i gn i f i can t effect on plant height or number of buds (Table 2 .6) . Table 2.6 - Effect of Urophora enclosures on d i f fuse knapweed c h a r a c t e r i s t i c s and g a l l f l y attack Unenclosed Enclosed Plants Character * Plants (Control Density) Number 50 18 Height 3 0 . 0 ± 0 . 7 3 0 . 0 ± 1 . 9 Buds/Plant 7 1 . 7 ± 6 . 0 6 7 . 3 ± 7 . 3 Chewed/Plant 6 .6±1 .2 1 5 . 3 ± 2 . 5 Chewed/Buds 0 . 0 9 ± 0 . 0 1 0 . 2 1 ± 0 . 0 2 Dev . /P lant 3 0 . 7 ± 2 . 4 2 4 . 9 ± 2 . 9 Dev./Buds 0 . 4 4 ± 0 . 0 2 0 . 3 9 ± 0 . 0 3 UA/Plant 3 6 . 9 ± 4 . 0 2 2 . 1 ± 5 . 6 UA/Dev. 1 . 2 0 ± 0 . 0 4 0 . 8 9 ± 0 . 0 6 UQ/Plant 8 . 3 ± 1 .3 1 . 3 ± 0 . 9 UQ/Dev. 0 . 2 7 ± 0 . 0 2 0 . 0 5 ± 0 . 0 2 Seeds/Plant 106±11 1 9 . 6 ± 4 . 0 Seeds/Dev. 3 . 4 7 ± 0 . 1 0 0 . 7 9 ± 0 . 0 8 Prop Aborted 0 . 2 0 ± 0 . 0 1 0 . 1 0 ± 0 . 0 1 * Deta i l ed descr ipt ions of these characters are given in Table 2.5. The most s i gn i f i can t enclosure effect was that grasshoppers s e l e c t i v e l y attacked the knapweed in the enclosures . The proport ion of buds chewed on enclosed plants was over double the 96 proportion for unenclosed p l a n t s . The greater damage to enclosed plants may be accounted for by the higher q u a l i t y of those p l a n t s . The net t ing probably reduced moisture s tress ; the enclosed plants were v i s i b l y greener. Many grasshoppers were observed on the outside of the enclosure net t ing . A s imi lar response of grasshoppers to higher q u a l i t y plants is described in Chapter IV. Damage by the grasshoppers was concentrated on buds in the primary branching category. Since g a l l f l y attack was d i s t r i b u t e d among branching categories on enclosed plants in a s imi lar way as on unenclosed p lant s , aborted buds and buds with high dens i t i e s of g a l l s and low dens i t i es of seeds were s e l e c t i v e l y removed. The s e l e c t i v i t y of grasshopper damage should not a f fec t the r e l a t i v e d i f ferences among treatments; the d i f ferences among treatments a l so held when the resu l t s are broken down by branching category. The lower proport ions of buds aborted on enclosed plants (even when they are corrected for chewing damage) than on unenclosed plants may a lso be due to a plant q u a l i t y e f f e c t . This hypothesis i s tested in Chapter IV. The o v e r a l l reduction in bud abort ion may be responsible for the lack of s i g n i f i c a n t d i f ferences between density treatments in the proportion of buds aborted. 97 I I I . BUD ABORTION AND POPULATION LIMITATION OF UROPHORA AFFINIS (DIPTERA: TEPHRITIDAE) IN BRITISH COLUMBIA In the f i r s t f i e l d experiments based on Nicholson and B a i l e y ' s (1935) models of population dynamics, Varley (1947) claimed to have i d e n t i f i e d three density-dependent processes which he assumed contro l l ed the population density of Urophora  jaceana Her. in England: attack by two p a r a s i t i c wasps and l a r v a l mor ta l i ty . Several c r i t i c i s m s inva l ida ted V a r l e y ' s o r i g i n a l conclusions, ranging from s t a t i s t i c a l problems with h i s ana lys i s (Andrewartha and B i r c h , 1954; Finney and V a r l e y , 1955) to his a p r i o r i assumption of the a p p l i c a b i l i t y of Nicholson and B a i l e y ' s model (Andrewartha and B i r c h , 1954). Roze (1981) argues that in a s imi lar system, U. a f f i n i s F r f l d . populations in B r i t i s h Columbia, density-dependent l a r v a l morta l i ty regulates population s i z e . She assumed that V a r l e y ' s conclusions were d i r e c t l y appl icab le to the North American system and noted that the paras i tes Varley observed to be responsible for two of the density-dependent processes in England are not present in B r i t i s h Columbia. Hence l a r v a l m o r t a l i t y , the t h i r d process Varley i d e n t i f i e d , must regulate population s i ze . Her argument regarding l a r v a l morta l i ty r e l i e s on the assumption that the bud abort ion by the host p l a n t s , d i f fuse knapweed (Centaurea d i f fusa Lam.) and spotted knapweed (C. maculosa Lam.), is caused by supernumerary larvae . Larvae in aborted buds are unable to form g a l l s and die at an ear ly stage. 98 An a l t e r n a t i v e to the mechanism for bud abortion Roze assumes is suggested by Zwolfer's (1970) laboratory observation that females may probe extens ive ly into buds without lay ing eggs. The mechanical damage caused by th is a c t i v i t y might a lso lead to bud abort ion as Berube (1978b) found in Sonchus arvensis L . buds pr icked with an insect p i n . These two poss ib le mechanisms may be d i s t ingu i shed since they predict d i f f e r e n t numbers of eggs in aborted buds. Roze (1981) and Chapters I and II . have demonstrated that the p r o b a b i l i t y of bud abort ion depends on the in tens i ty of insect attack and that bud abort ion reduces the po tent ia l g a l l formation in attacked p l a n t s . Thus bud abortion l i m i t s the g a l l f l y populat ions , in the sense that , i f i t d id not occur, population dens i t i e s would be higher. How great i s the reduction in population density? On the basis of a simple ana lys i s of the d i s t r i b u t i o n s of g a l l s and aborted buds at the end of the summer, I concluded in Chapter I that g a l l production by U. a f f i n i s would have been 72% greater i f buds had not aborted. Yet insect attack does not occur at a s ingle point in time. Chapter II advanced the hypothesis that the accumulation of aborted buds s i g n i f i c a n t l y a l t ered the attack of the g a l l f l i e s over time. Two mechanisms were suggested: (1) aborted buds act as an egg "sink", and (2) aborted buds increase the search time for sui table o v i p o s i t i o n s i t e s . The counts of eggs in aborted buds w i l l a lso d iscr iminate between these two mechanisms. U. quadr i fasc ia ta (Meig . ) , the congeneric gal l - former that 99 was introduced to North America at the same time as U. a f f i n i s , a lso suffers from the ef fects of bud a b o r t i o n . Because U. a f f i n i s prefers smaller buds than U. quadr i fasc ia ta (Berube and H a r r i s , 1978) and because probed buds which aborted were smaller than those that d id not (Chapter I ) , i t i s u n l i k e l y that U. quadr i fasc iata contributes s i g n i f i c a n t l y to bud a b o r t i o n . U. quadr i fasc ia ta w i l l not be considered further in th i s Chapter. This Chapter (1) tests the hypothesis that d i f fuse knapweed aborts buds because they contain too many l a r v a e , and (2) evaluates the impact of bud abortion on the populat ion dynamics of U. a f f i n i s . 100 MATERIALS AND METHODS BUD COLLECTION AND DISSECTION Diffuse knapweed plants were obtained throughout the ov ipos i t ion period of the f l i e s to determine the number of eggs in aborted and unaborted buds. In 1980, nine to s ixteen randomly selected plants were c o l l e c t e d at Robertson's on each of the f i r s t four dates shown in Table 3 .1 . The remaining c o l l e c t i o n s cons i s t ing of a t o t a l of 82 plants were made on August 1, 8, 16, 22, and September 12. The numbers of eggs, l arvae , or g a l l s and the proportions aborted from these l a t t e r c o l l e c t i o n s were not s i g n i f i c a n t l y d i f f erent (a=0.05) and so they were combined. I d i s sec ted the a p i c a l bud and the terminal buds on the top four branches of each p l a n t . E a r l i e r work (Berube, pers . comm.) had suggested that these buds were among the buds rece iv ing the highest egg loads . I noted the presence of eggs or larvae . A d i s t i n c t i v e brown d i s c o l o r a t i o n of the f l o r e t s caused by probing, o v i p o s i t i o n , or l a r v a l feeding f a c i l i t a t e d locat ion of the immature forms of the insect . Varley (1947) noted a s imi lar response to l a r v a l damage in C. nemoralis J o r d . Since s e r i a l sect ioning of the buds was not done, a cer ta in proport ion of the eggs might have escaped detect ion . The r e l a t i v e prominence of the eggs in the immature f l ore t s and the ir large s ize combined with the c l e a r damage associated with l a r v a l feeding and g a l l formation make i t h ighly un l ike ly that th is proportion exceeded 0.20. Eggs l a i d a f t er g a l l formation was well underway or eggs which f a i l e d to hatch and subsequently died would be more l i k e l y 101 to be overlooked. For comparison, Varley (1947) found that U. jaceana eggs suffered r e l a t i v e l y low morta l i t y p r i o r to hatching: 8.9% in 1935 and 15.3% in 1936. Bud abort ion appeared to be associated with the presence of a dark brown band in the receptacle , poss ib ly caused by the blockage of vascular connections. 1 02 RESULTS O v i p o s i t i o n in the f i r s t f ive terminal buds was corre la ted with f l y abundance. Table 3.1 demonstrates the gradual accumulation of eggs and larvae in these buds. The biggest Table 3.1 - Urophora eggs and larvae and proportion of buds aborted for terminal buds of d i f fuse knapweed, Robertson's 1980 Date Eggs and Larvae N Proport ion N per Bud Aborted June 17 0 . 1 4 ± 0 . 1 0 41 0 . 0 2 ± 0 . 0 2 43 June 21 0 . 8 8 ± 0 . 1 6 a 55 0 . 0 3 ± 0 . 0 2 e 58 June 28 1 . 4 0 ± 0 . 2 4 b 52 0 . 2 9 ± 0 . 0 5 f 75 Ju ly 10 2 . 5 3 ± 0 . 3 1 c 31 0 . 5 9 ± 0 . 0 6 g 78 F i n a l * 3 . 3 7 ± 0 . 1 6 d 185 0.5510.02 h 410 * August 1, 8, 16, 22, and September 12. Le t ters ind ica te s t a t i s t i c a l comparison with preceding date: a. t=3.90, df=88, p<0.001; b. t=1.81, df=9l, p=0.074; c . t=2.89, df=83, p=0.005; d. t=2.11, df=214, p=0.036; e. F i s h e r ' s exact t e s t , p=1.000; f. X*=13.12, df=1, p<0.001; g. X 2 = l 3 - 6 0 , df=1, p<0.00l; h. y^=0.i5r df=1, p=0.505. increase in eggs and larvae per bud, d iv ided by the number of days between c o l l e c t i o n dates, occurred during the period June 17 to June 21, which coincided with the f i r s t s i g n i f i c a n t number of f l i e s (Chapter I I ; Figure 2.1) . The proport ion of aborted buds a lso increased, but carefu l d i s sec t ion revealed that only th i r teen (18%) of the aborted buds from the f i r s t four c o l l e c t i o n s contained eggs. The mean number of eggs in these buds was 1.2. These eggs could be a t t r ibuted to bud abortion 1 03 after o v i p o s i t i o n or mistakes by the o v i p o s i t i n g female, since f l i e s may not be able to d i s t i n g u i s h aborted buds from developed buds on the basis of external appearance. There was no evidence that buds which aborted were overloaded with eggs or larvae . 1 04 DISCUSSION OVIPOSITION BEHAVIOUR The s i g n i f i c a n t s ize (and presumably developmental) di f ference between probed buds which aborted and those which d id not (Chapter I) suggests that the small buds just enter ing the size range of su i table ov ipos i t i on s i t e s may be more l i k e l y to abort . The low egg density in aborted buds implies that aborted buds do not subsequently receive eggs. Since aborted buds do not move out of the s ize range of su i tab le buds, females can detect that a bud is aborted, at least af ter probing i t . This detection may be based on the s ize of the immature f l o r e t s (Zwolfer, 1970) or some other p h y s i o l o g i c a l cue, such as i n t r a c e l l u l a r prote in concentrat ion. Zwolfer's (1970) laboratory studies indicate that females do not neces sar i ly discr iminate between buds on the basis of the presence of eggs. The observed contagion of g a l l d i s t r i b u t i o n s in the f i e l d (Myers and H a r r i s , 1980; Chapter V) supports t h i s idea (cf . Dacus tryoni on loquat; Monro, 1967). The contagion is a lso consistent with Rausher's (1979a) claim that insects w i l l d i scr iminate when the s ize of the host i s small r e l a t i v e to l a r v a l demands. The data I obtained indicate that egg loads are much lower in the f i e l d than when g a l l f l i e s are confined to a s ing le plant in the laboratory (Zwolfer, 1970). E p i d e i c t i c pheromones are widely used by t e p h r i t i d f l i e s to mark t h e i r ov ipos i t ion s i t es (Prokopy, 1981), but there are exceptions (e .g . Berube, 1978a). There is no evidence for such marking by e i ther U. a f f i n i s or U. quadr i fasc iata in North 1 05 America (Berube, pers . comm.; pers . o b s . ) . In a s i m p l i f i e d view of the system, i f females have a f ixed number of eggs to o v i p o s i t , and i f they lay some in buds that eventual ly abort , the ir reproduction w i l l be lower than i t would be i f they d i scr iminated against aborted buds. I f , as the data suggest, bud abortion is caused by probing damage and f l i e s do not lay eggs in aborted buds, then the females re ta in the ir complement of eggs. An a l t e r n a t i v e mechanism is required to explain the reduction in g a l l formation associated with bud abort ion . Such an a l t e r n a t i v e i s an increase in the search time for a su i tab le o v i p o s i t i o n s i t e (e .g . Jones et a l . , 1980). SEARCH TIME BETWEEN OVIPOSITIONS The average search time leading to a successful o v i p o s i t i o n w i l l be the sum of the time to locate a bud of a su i table s ize and the time to probe the bud to determine i t s s u i t a b i l i t y for l a r v a l development div ided by the p r o b a b i l i t y of o v i p o s i t i o n given a bud of a sui table s i ze . Put symbol ica l ly , th i s becomes: AST = (LT + PT)/p where AST is the average search time, LT is the locat ion time, PT is the probing time, and p is the p r o b a b i l i t y of successful o v i p o s i t i o n . This p r o b a b i l i t y i s equal to the proportion of good buds, GB, in the t o t a l populat ion of su i tably s ized buds, SSB (assuming that g a l l f l i e s p e r f e c t l y d i s t i n g u i s h good buds from aborted buds, AB). Thus p = GB/SSB = GB/(GB + AB). A p laus ib l e expression for LT i s an inverse function of the density of buds of a su i tab le s ize (SSB), i . e . 106 LT = k/SSB, where k i s a constant of p r o p o r t i o n a l i t y . The t h e o r e t i c a l r e l a t i o n s h i p between LT and SSB has (0, i n f i n i t y ) and ( i n f i n i t y , 0) on i t . Table 3.2 summarizes Zwolfer's (1970) observations on probing times for U. a f f i n i s in spotted knapweed (C. maculosa) buds in the laboratory . These data suggest that probing i s a Table 3.2 - Duration of probing into spotted knapweed buds by female U. a f f i n i s * Duration Tota l Duration of of Probe N Probing Bout N Eggs deposited 1 0 . 4 ± 2 . 1 ** 23 35 ±10 7 No eggs deposited 12.5+1.9 26 27. 1±5.8 1 2 Combined obs. 11 .5±1 . 4 49 3 0 . 0 ± 5 . 2 1 9 * - Data from Zwolfer (1970). ** - Minutes ( M e a n ± S . E . ) . r e l a t i v e l y time consuming a c t i v i t y and that probing times are not s i g n i f i c a n t l y d i f f erent whether eggs are deposited or not. Subs t i tu t ing the expressions for LT (= k/SSB), p (= GB/SSB), and SSB (= GB + AB), into the expression for AST and s i m p l i f y i n g , the expression for AST becomes: k AB AST = - - + PT(1 + --) . GB GB Assuming constant bud i n i t i a t i o n and growth, such that GB i s constant , the average search time between successful o v i p o s i t i o n s w i l l be a l inear function of the density of aborted 1 07 buds with the probing time as the constant of p r o p o r t i o n a l i t y . The presence of aborted buds w i l l s i g n i f i c a n t l y reduce the a b i l i t y of the g a l l f l i e s to success fu l ly o v i p o s i t , despite a constant rate of encounter with buds of a sui table s i ze . This e f fect may be viewed as a form of "mutual interference" (Hasse l l , 1978). This ana lys i s has hypothesized that the time spent probing as a key constant in the populat ion l i m i t a t i o n of the g a l l f l i e s and has indicated the importance to the f l i e s of the a b i l i t y to detect in the shortest poss ib le time buds that are unsui table , e i ther because they are too small or because they are aborted. The density of aborted buds, which i s a cumulative function of insect at tack, i s one of the key state var iab l e s . MODEL FORMULATION The impact of bud abortion on the net production of g a l l s by U. a f f i n i s was examined by using a simple numerical model. A number of s i m p l i f y i n g assumptions were made about the system. I assumed that the density of ov ipos i t ing f l i e s was normally d i s t r i b u t e d in time around a time i n t e r v a l in the middle of a season of eleven d i scre te time in terva l s (mean -i n t e r v a l 6; s . d . - 2 i n t e r v a l s ; N=3300). The area in the t a i l s of the d i s t r i b u t i o n of f l i e s outside of the defined season was r e d i s t r i b u t e d within the season in proportion to the number of f l i e s per i n t e r v a l . Thus the t o t a l number of f l i e s in the season remained constant even i f the standard deviat ion of the d i s t r i b u t i o n was changed. The number of probes per f l y was assumed to be constant throughout the season, independent of 1 08 whether probes resul ted in ov ipos i t i on or not. I assumed that buds were i n i t i a t e d at a constant rate (300 buds per time i n t e r v a l ; Chapter I I ) , but were only a v a i l a b l e and su i tab le for ov ipos i t ion in the time i n t e r v a l they were i n i t i a t e d . Since buds change in qua l i ty during the summer, I assumed that the p r o b a b i l i t y of not maturing (independent of f l y attack) increased with time from 2.7% in the f i r s t time i n t e r v a l to 29.7% in the eleventh i n t e r v a l . G a l l f l i e s were assumed not to probe • undeveloped buds. This way of t rea t ing undeveloped buds w i l l tend to reduce the ef fect of bud abort ion on g a l l product ion. I omitted the ef fect of d i f ferences in bud growth rates during the season. The t o t a l number of probes in a given i n t e r v a l was assumed to be d i s t r i b u t e d among a l l buds of a su i table s ize according to a Poisson d i s t r i b u t i o n . The mean of the Poisson d i s t r i b u t i o n was ca lcu la ted as the number of probes d iv ided by the sum of the buds i n i t i a t e d in that i n t e r v a l and aborted buds accumulated from previous time i n t e r v a l s . Bud abortion was modelled by using an abort ion thresho ld , corresponding to the number of probes for which 50% of buds aborted (six probes in the runs described below). I assumed that the p r o b a b i l i t y of a bud abort ing was d i r e c t l y propor t iona l to the number of probes per bud. Buds which received more than eleven probes per bud were assumed to abort with p r o b a b i l i t y 1.0. Buds aborted in each time i n t e r v a l were added to the cumulative number of aborted buds. A complete l i s t i n g of the model i s given in Appendix IIIA. 1 09 MODEL RESULTS The resu l t of a sample run of the model i s shown in Figure 3 .1 . A temporal pattern of g a l l production and bud abortion s i m i l a r to that observed for the f i r s t generation of f l i e s was produced. The g a l l s per developed bud peaked before the f l y abundance d id and dec l ined more slowly than i t rose, exactly as observed in the f i e l d . The sh i f t between the f l y abundance and g a l l s per developed bud was not as great as observed in the f i e l d because the time for buds to grow into a su i table s ize for o v i p o s i t i o n was omitted from the model. S i m i l a r l y , the maximum proport ion aborted coincided with the peak in g a l l s per developed bud in the model output; the observed pattern had the maximum proportion aborted sh i f ted to the r ight of the peak number of g a l l s per developed bud, in part because aborted buds are s l i g h t l y smaller than buds which produce g a l l s . The ef fect of varying the number of probes per bud in th i s model i s shown in Figure 3.2. The average number of g a l l s per developed bud increased monotonical ly, but because of bud abort ion , the t o t a l number of g a l l s produced reached a maximum and then d e c l i n e d . The peak number of g a l l s produced for the parameter values used occurred at approximately ten probes per bud. If bud abort ion d i d not occur, the t o t a l number of g a l l s corresponding to the values in Figure 3.2 would have been a l inear function of the probes per bud with a slope of 2765 (the t o t a l number of developed buds in the model). Thus for one probe per bud, bud abort ion reduced the t o t a l number of g a l l s by 81%; for ten probes per bud the reduction is 96%. 1 1 0 Figure 3.1 . Sample output from the model of the in teract ion of the g a l l f l i e s with the ir host plant over the summer. The pattern should be compared with Figures 2.3 and 2.4. 1 1 2 Figure 3.2. Ef fect of varying probes per bud in the numerical model. Ga l l s per bud (xlOOO) i s the s o l i d l i n e . Tota l g a l l s produced is the dotted l i n e . Only the number of probes per bud was var i ed ; a l l other parameters were unchanged. 1 1 4 If the g a l l f l i e s could not detect whether buds were aborted pr ior to o v i p o s i t i o n , then attack l eve l s would be higher in a l l time in terva l s except the f i r s t ; both bud abort ion and the number of g a l l s per developed bud would be greater . As the proportion of buds aborted increased with the average probes per bud, th i s would trans la te into lower t o t a l g a l l production compared with the case where bud abort ion is detected. The model resu l t s demonstrate that increased search time between ov ipos i t ions due to bud abort ion may account for the observed temporal patterns and may d r a s t i c a l l y reduce g a l l formation. The ef fect of a reduced encounter rate with su i tab le buds would carry over into the second generat ion. Chapter IV describes experiments designed to manipulate the propensity of plants to abort buds. SUMMARY Two mechanisms for insect-caused bud abort ion in d i f fuse knapweed were compared. There was no evidence for the mechanism Roze (1981) suggested, l a r v a l feeding. The a l t e r n a t i v e , mechanical damage caused by probing females, was cons is tent with the observed d i s t r i b u t i o n of eggs and larvae among buds. Bud abortion may reduce g a l l production by increas ing the search time between successful o v i p o s i t i o n s . A numerical model based on th i s premise produced a pattern of g a l l formation and bud abortion s imi lar to that observed in Chapter I I . The model implies that even r e l a t i v e l y low l eve l s of bud abort ion may 1 15 dramat ica l ly reduce the t o t a l number of ga l l s formed. 1 1 6 APPENDIX IIIA. LISTING OF THE NUMERICAL MODEL A l i s t i n g of the numerical model used to evaluate the effect of bud abort ion is given in th i s Appendix. PROGRAM FLYMOD C C FORTRAN 77 PROGRAM TO EXPLORE THE PROPOSED MECHANISM FOR THE EFFECT OF BUD C ABORTION ON THE PATTERN OF GALL FORMATION FOR UROPHORA AFFINIS AND ITS C LONG TERM EFFECT ON POPULATION LIMITATION. C C P E T E R MORRISON, INST I TUTE OF ANIMAL R 1984 C INTEGER ABTHR REAL MNFLY '," SDFLV . PRBBUD. PRPABT, PABf REAL FLYINT(II), UNOEV(II). POISSN(12), CUMPSN(IS), TOTGAL(11) REAL ABTBDS(II), AVAIL(H), UNATT(II), ATTBDS(II). c DATA PRBBUD /20.0 / DATA MNFLY / 60.0 / DATA'""sbFilV ' " / 2 o " o " " / ~ C C ABTHR • NUMBER OF PROBES PER BUD THAT CAUSES SOX OF BUDS C TO ABORT C DATA ABTHR f 6 / DATA TOTGAL / 1 1 •0i'6"7 C C DETERMINE DISTRIBUTION OF PROBES OVER INTERVALS - USING NORMAL OIST. "c : TOTPRB • 3300.0*PRBBU0 SUMPRB • 0.0 _ _ D " 0 " " 2 Q " j - ^ Y ' " i " " i " ; " " i " i — " " " ' " "" ORO • ICNT'10.0 FLYINT(ICNT) « EXP(-(ORO-MNFLY)*(ORO-MNFLY)/(2.0*SDFLY»SDFLY)) FLYINTriCNtj " i fbfPRB^ SUMPRB • SUMPRB • FLYINT(ICNT) 20 CONTINUE c " DIFPRB ' TOTPRB - SUMPRB DO 30 ICNT • 1. 11 F L Y i N f T i ^ 30 CONTINUE . C C STEP'THROUGH '"fl ME:'"iNT ERVAIS' UP^f IN6" THE' 'p'ROPORf ION ABORT ED C CUMABT "0.0 C C CALCULATE PROPORTION UNDEVELOPED AS A FUNCTION OF TIME !"c UNOEV(ICNT) • 0.027 • ICNT * 300.0 C 0E'VBUD"''"''3'^  C C CALCULATE THE MEAN NUMBER OF PROBES PER BUD AND DISTRIBUTE THEM C ACCORD C PRBMN • FLYINT(ICNT) / (DEVBUD + CUMABT) c DO SO OCNT • 2, 12 CUMPSN(JCNT) • 0.0 poiSSN(JCNTy - 6:o 50 CONTINUE c POISSN(I) • EXP(-1.O'PRBMN) CUMPSN(I) • POISSN(I) C DO 100 KCNT • 2, 12 POISSN(KCNT) • POISSN(KCNT-I) • PRBMN / FLOAT(KCNT) CUMPSN(KCNT) • CUMPSN(KCNT-I) • POISSN(KCNT) 100 CONTINUE C CALCULATE THE PROPORTION OF BUDS ABORTED c PRPABT • 0.0 DO ISO MCNT • 2, 12 pABT'0"6•(MCNT - " 1 j / FLOATf*BTHRj IF (PABT .GT. 1.0) PABT • 1.0 PRPABT • PRPABT • PABT'POISSNjMCNT) POiSSN(MCNT) • pbiSSN^MCNTWl.6 - PABT) 150 CONTINUE PRPABT • PRPABT • (1.0 - CUMPSN(12)J c C CALCULATE PROPORTIONS IN VARIOUS CATEGORIES OF ATTACK C ABT BOSiH' CNT )"' • DEVBUD • PRPABT AVAIL(ICNT) • 300.0 - UNDEV(ICNT) - ABTBDS(ICNT) UNATTUCNT) • POISSN(I) » DEVBUD ATf BDSTiCNT )' "•" AVAIL(ICNT )~-"UNATf (fcWJ™ C CUMABT • CUMABT + ABTBDS(ICNT) c r DO 200 NCNT - 1, 12 TOTGAL(ICNT) • TOTGALjICNT) • OEVBUO«POISSN(NCNT)«iNCNT-1) 20OCONTINUE C 1000 CONTINUE " c " C CALCULATE STATISTICS AND WRITE OUT RESULTS C WHIITE(G'ri i6oV" MNFLY'r's " " ~ 1100 FORMAT(' MEAN '.F6.2.' SO '.F6.3.6X, • ' PROBES/BUD ',Fe.3,' ABT. THRESH. ',13,, _ • / / • I U N O E V E ' L " A B O R T E D A T T A C K E D ""' + 'UNATTACK. GALLS/DEV GALLS/ATT. FLIES ') C TOTAVL "• 6 "6 TOTATT • 0.0 SUMGAL • 0.0 bb isbo i C N T " • ~ i i i R1 • UNDEV(ICNT)/3CO.O R2 • ABTBDS(ICNT)/300.0 R3''-''AtfBb's"riTOfy/3o6'.''6 R4 • UNATT(ICNT)/300.0 IF (AVAIL(ICNT) .GT. 0.O01) THEN R5"'•i fbfGALfiCNf V/AVAlL(iCNf ) ~ ELSE R5 • 0.0 ENDIF IF (ATTBDS(ICNT) .GT. 0.001) THEN R6 • TOTGAH ICNT)/ATTB0S( ICNT) ELSE R6 - 0.0 ENOIF  H7 • FLYINT(ICNT)»10.0/T0TPRB TOTAVL • TOTAVL • AVAIL(ICNT) TOTATT - TOTATT • ATTBOS(ICNT) SUMGAL • "sUMWL"T'fofGAL(ICNf j C WRITE(6,1200)ICNT,R1,R2,R3, R4, R5 ,R6 ,R7 1200 FORMAT(l3.7Fi6.5) 1500 CONTINUE C c WRITE: bin;'"suMMARv'"sTATmics C UAOEV • SUMGAL / TOTAVL UAATT • ' SUMGAL / TOT AT f PRPATT • TOTATT / TOTAVL PRPABT > CUMABT / 3300.0 WRITE<6.1900) 190O FORMAT(//' GALLS/OEV. GALLS/ATT. PROP ATT. PROP ABT. DEV BUDS') WRITE(6.2000) UAOEV. UAATT, PRPATT, PRPABT, TOTAVL 20O0 FORMAT(4F10.5.F10.2) C STOP 1 20 IV. PLANT QUALITY AND THE POPULATION DYNAMICS OF TWO INTRODUCED INSECTS Plant q u a l i t y may have s t r i k i n g ef fects on phytophagous insect populations (e .g . Dodd, 1940 (c i ted in Wilson, 1960); Myers and Post , 1981; Port and Thompson, 1980; White, 1976). Several mechanisms may lead to larger insect populations on higher q u a l i t y p lant s , ranging from host choice by moving insects (Kaireva , 1983; Myers, 1985; Vince and V a l i e l a , 1981; Chapter I ) , to improved s u r v i v a l , development, and reproduction (Scriber and Slansky, 1981; Vince and V a l i e l a , 1981; White, 1976). Some recent reviews have summarized much of th i s l i t e r a t u r e (e .g . Crawley, 1983; McNei l l and Southwood, 1978; Mattson, 1980; Rhoades, 1983; Scriber and Slansky, 1981). The two ga l l - forming f l i e s , Urophora a f f i n i s F r f l d . and U. q u a d r i f a s c i a t a (Meig.) (Diptera: T e p h r i t i d a e ) , lay eggs in the immature flower buds of d i f fuse knapweed, Centaurea d i f fusa Lam., and spotted knapweed, C. maculosa Lam. (Asteraceae). In B r i t i s h Columbia, g a l l f l y dens i t ies on d i f fuse knapweed are l i m i t e d by two fac tors : the density of ov ipos i t i on s i t e s and bud abort ion in response to probing by female f l i e s (Chapters I , I I , and I I I ) . If plant qua l i ty increases, both of these factors should s h i f t in favour of higher insect d e n s i t i e s . This Chapter considers the e f fect of nitrogen f e r t i l i z a t i o n and watering on these two populat ion l i m i t a t i o n factors and on l a r v a l s u r v i v a l and development. Both water and ni trogen, separately and in combination, 121 should increase the resources a v a i l a b l e to the bo l t ing knapweed plants and to i n d i v i d u a l buds. Plants should respond by increasing the number of buds they i n i t i a t e (c f . Watson, 1972). Based on the re la t ionsh ips observed in Chapter I , the number of f l i e s per plant should be proport iona l to the number of buds per p lant . The r e s u l t i n g l eve l s of attack should be the same for treated and untreated p lant s . If bud abort ion is a function of the resources ava i lab le to the plant (Stephenson, 1981), then increased resources should lead to a drop in the p r o b a b i l i t y of abort ion at a given l e v e l of attack (though Onuf et a l . (1977) found the opposite effect on Rhizophora mangle (G.F.W. Meyer) E n g l e r . ) . A l t e r n a t i v e l y , the same proport ion of buds could be aborted i f the o v e r a l l l eve l of attack on f e r t i l i z e d plants was h igher . Because of the tradeoff between g a l l formation and bud abort ion (Chapter I ) , th i s a l t e r n a t i v e should give more g a l l s per developed bud. Increased resources a l l o c a t e d to buds should increase the ir a b i l i t y to support l a r v a l growth and development. .Both leaf water and nitrogen l eve l s increase the r e l a t i v e growth rate of young insects (Scriber and Slansky, 1981; but see Schroeder and Maimer, 1980). Some poss ible consequences for the g a l l f l y larvae are: improved l a r v a l s u r v i v a l , more successful development to pupae and a d u l t s , and faster development with a resu l t ing change in the timing of the second generation. Personal observations and data presented by Harr i s (1980a,b) and Roze (1981) suggest that the two factors l i m i t i n g g a l l f l y populations on d i f fuse knapweed, densi ty of ov ipos i t ion 1 22 s i t e s and bud abort ion , may also be act ing on f l y populations on spotted knapweed. This Chapter (1) describes the e f fects of nitrogen f e r t i l i z a t i o n and watering on the number of buds per p lant , (2) documents the response of the g a l l f l i e s and other herbivores to the changes in plant q u a l i t y , and (3) assesses changes in the number of g a l l s per developed bud, bud abort ion , and l a r v a l s u r v i v a l and development due to the treatments. 1 23 MATERIALS AND METHODS WEATHER The natural p r e c i p i t a t i o n at the study s i tes provides the c o n t r o l for the watering treatments. Weather data were obtained from Environment Canada. Data from the Kamloops A i r p o r t weather s tat ion were used for s p e c i f i c r a i n f a l l patterns presented in th i s Chapter. The values were consistent with the patterns recorded at Chase, B . C . , at the opposite end of the South Thompson River v a l l e y , and a l so agreed with the days r a i n f a l l was observed at the study s i t e s . EXPERIMENTAL TREATMENTS IN 1979 Nitrogen f e r t i l i z a t i o n and watering may interact s i g n i f i c a n t l y both in terms of the plant response (e .g . Buchner and Sturm, 1971; Decau and P u j o l , -1973) and in terms of surv ivorsh ip of young insects feeding on plants (e .g . Mispagel , 1978). Under more a r i d condi t ions , nitrogen f e r t i l i z a t i o n may increase water s tress to the plant (e .g . crested wheatgrass, Wil l iams et a l . , 1979), even to the extent of causing f e r t i l i z e r "burn". As a resu l t of these cons iderat ions , a set of in terac t ing treatments of water and f e r t i l i z e r was establ ished in 1979 at each of the three study s i t e s : Chase, Ned's Creek, and Robertson's . The treatments at each s i t e consisted of a l l poss ib le combinations of three f e r t i l i z e r l eve l s (high, low, and none) and three watering l eve l s (high, low, and none). Each of the nine treatments was r e p l i c a t e d three times at each s i t e for a t o t a l of 27 p lo ts per s i t e . Each p lo t consisted of one m2 1 24 surrounded by a space of 0.25 m to el iminate s p i l l o v e r e f fec ts and to permit access to the plants for watering and measurement. The nine p lo ts in each rep l i ca te were grouped together in a 3.5 m by 3.5 m area with a v i s u a l l y uniform density of knapweed. The c e n t r a l rep l i ca te was separated from the two r e p l i c a t e s on e i ther side by 0.5 m. A t o t a l of twenty plants were followed for each treatment, f ive each from the two outside rep l i ca te s and ten from the c e n t r a l r e p l i c a t e . Each plant was selected as the nearest b o l t i n g plant to an a r b i t r a r i l y chosen point within the p l o t . F ive of the points in each plot were taken as the center of the p lot and the four points b i sec t ing the s tra ight l i n e s between the center and the four corners of the p l o t . In p lo ts where ten plants were followed, four of the a d d i t i o n a l points were taken as the points b i sec t ing the s tra ight l ine s between the se lected plants near adjacent plot diagonals . The tenth plant was chosen by b l i n d l y f l i p p i n g a penc i l into the p l o t . If the nearest plant to the penc i l t i p was not previous ly se lected and was within the plot boundaries, i t was chosen. Selected plants were r a r e l y nearest neighbours. The f e r t i l i z e r treatment in 1979 consisted of a s ing le a p p l i c a t i o n of ammonium n i t r a t e p e l l e t s on May 28, during b o l t i n g . Ammonium n i t ra te was chosen because i t provides both forms of ionic nitrogen and because i t was not known which ion ic form would have the greatest effect on knapweed growth. Plants may d i f f e r in the ir response to the two ionic forms of ni trogen depending on the ir pH preference (Buchner and Sturm, 1971; Gigon 125 and Rorison, 1972), or the concentrat ion of the opposite ionic form (Cox and Reisnauer, 1973). The rates for the high and low l eve l s were the equivalent of 100 and 50 kg per ha, r e spec t ive ly , of n i t rogen . It rained during the f e r t i l i z e r a p p l i c a t i o n and none of the p e l l e t s were v i s i b l e one hour af ter app l i ca t ion . Watering treatments were 350 and 700 ml per m2 appl ied approximately every four days during June and 250 and 500 ml per m2 during July u n t i l Ju ly 23. Water was appl ied between 8:30 a.m. and 9:30 a.m. to reduce evaporative los s . Watering l eve l s amounted to approximately 7% and 14% of the normal ( i . e . t h i r t y year average) r a i n f a l l recorded at the Kamloops A i r p o r t for June 1 to August 31. Watered plots received to ta l s of 3.45 or 6.9 l i t e r s per m2 at Ned's Creek (low and high watering l e v e l s , re spec t ive ly ) and 3.2 or 6.4 l i t e r s per m2 at Robertson's and Chase (low and high watering l e v e l s , r e s p e c t i v e l y ) . The timing and magnitude of watering added to natural p r e c i p i t a t i o n is shown in Figure 4.1. Because the two d i f fuse knapweed s i t e s were watered on d i f ferent days, between s i t e comparisons must be q u a l i f i e d according ly . NUTRIENT ANALYSIS The i n i t i a l l eve l s of nitrogen and other nutr ients in the s o i l could af fec t the resul ts of the experimental treatments and between s i t e comparisons. Six samples were taken at each of the three s i tes on May 29, 1979 for nutr ient a n a l y s i s . The samples were screened through a #2 screen to remove plant mater ia l and rocks which would not have 1 26 Figure 4.1. Watering and natural p r e c i p i t a t i o n in 1979. The r a i n f a l l recorded at Kamloops A irpor t during the summer is shown in the upper part of the f i g u r e . Data are from Environment Canada. The lower part of the f igure gives the sequence and amount of a d d i t i o n a l water applied to the experimental p lots expressed in mm. The hor izonta l dashes b i sec t ing each watering treatment indicate the two treatment l e v e l s . Ned's Creek Robertson's Chase watering 13 17 22 June 4-4-29 3 7 11 15 ro 25 2  i 1 1 r™ 15 19 23 27 31 T 4 - r 8 - r -12 I I I I 16 20 24 28 July August contained immediately access ib le analyzed by the B . C . Min i s t ry Kelowna. Values for each of the 1 28 n u t r i e n t s . The samples were of A g r i c u l t u r e laboratory in s i t e s are given in Table 4 .1 . Table 4.1 - S o i l c h a r a c t e r i s t i c s at the study s i t e s Ned's Creek SITE Robertson's Chase Organic matter (%) 3 . 9 ± 0 . 1 a 5 . 4 ± 0 . 2 5 . 2 ± 0 . 2 pH 6 . 6 ± 0 . 1 7 . 0 ± 0 . 1 6 . 4 ± 0 . 1 Conduct iv i ty b 0 . 1 8 ± 0 . 0 1 0 . 2 5 ± 0 . 0 1 0 . 2 1 ± 0 . 01 Ni tra tes (ppm) 6 . 2 ± 1 . 3 1 . 3 ± 0 . 2 1 . 5 ± 0 . 2 Phosphorous (ppm) 21 ±2 64±4 129±11 Potassium (ppm) 640±20 950±30 7 3 0 ± 3 0 Calcium (ppm) 2900±80 4 2 0 0 ± 1 0 0 3 7 8 0 ± 8 0 Magnesium (ppm) 535±7 740+14 3 2 0 ± 1 0 a - MeaniS.E. b - Measured in mS/cm at 25 degrees C. S o i l c h a r a c t e r i s t i c s for Ned's Creek and Chase are comparable to Watson's (1972) values. EXPERIMENTAL TREATMENTS IN 1980 The resu l t s from 1979 (see below) showed that both f e r t i l i z e r and water af fected plant .growth and the treatments were approximately add i t ive in terms of numbers of buds per p lant . In 1980, f e r t i l i z e r and water were appl ied at a s ingle higher l e v e l in an attempt to "push" the system further . Four plots of f ive plants each at Robertson's and six p l o t s of f ive plants each at Chase were treated with the s ingle l e v e l of f e r t i l i z e r and water. Plot dimensions were i d e n t i c a l to 1 29 those used in 1979. Plants were selected in the same way. Control p lo t s were es tab l i shed on June 1 at Robertson's and on June 2 at Chase. At each s i t e , a rectangular g r i d of f i f t y (5x10) points was placed over a 3 m x 6.5 m portion of the f i e l d with a v i s u a l l y uniform density of knapweed. The plant which had begun to bolt nearest each point on the g r i d was staked. The points on the g r i d were far enough apart so that staked plants were r a r e l y nearest neighbours. F e r t i l i z e r was appl i ed at 150 kg per ha and water was appl ied at 700 ml per m 2 , usual ly every two days (Figure 4 .2) . F e r t i l i z e r was app l i ed on June 3 at Robertson's and on June 6 at Chase, during the ear ly phase of b o l t i n g . A t o t a l of 23.8 l i t e r s per m2 was appl i ed at Robertson's. The t o t a l at Chase (18.9 l i t e r s per m2) was lower because watering was halted when I s tarted c o l l e c t i n g the spotted knapweed. OBSERVATION OF INSECTS IN 1979 In 1979, s t icky traps (as in Berube, 1980) were used as the primary means of measuring insect abundance over time. Results from th i s method are not reported because subsequent ana lys i s showed that counts on s t i cky traps are very sens i t ive to temperature (MS in p r e p . ) . V i s u a l surveys of staked plants were also done in 1979 as a second method of measuring r e l a t i v e insect abundance. Observations were made at each s i t e every 4-5 days from June 13 to July 13, between 8:30 a.m. and 12:00 noon. The two species of g a l l f l y are e a s i l y d i s t ingu i shed by the banding pattern on the wings and sexes are d i s t ingu i shed by the prominent oviscape 1 30 gure 4.2. Watering and natural p r e c i p i t a t i o n in 1980. The r a i n f a l l .recorded at Kamloops Airport during the summer is shown in the upper part of the f igure . Data are from Environment Canada. Note the break in the v e r t i c a l a x i s . The lower part of the f igure gives the sequence and amount of a d d i t i o n a l water appl ied to the experimental p lo t s expressed in mm. 1 32 of the females. The f l i e s are qui te doc i l e and t h e i r behaviour was not obviously a l t ered by an observer. Because of the r e l a t i v e l y small numbers of f l i e s observed (a t o t a l of 370 over 180 d i f fuse knapweed plants during the ent i re summer at Ned's Creek, the s i t e with the greatest number recorded) , I combined the observations for a l l p lants in a given f e r t i l i z a t i o n and watering treatment. Observations of un ident i f i ed sp i t t l ebugs (Homoptera: Cercopidae) on spotted knapweed at Chase were made at the same time as the surveys for g a l l f l i e s . OBSERVATION OF INSECTS IN 1980 Approximately every two days a l t ernat ing between the two s i t e s , I conducted a branch by branch v i s u a l survey of each staked plant in c o n t r o l and treated p l o t s . The l o c a t i o n , species , and sex of each adult f l y was recorded at the moment of observat ion. The v i s u a l surveys for adult f l i e s on plants were usual ly conducted between 8:30 a.m. and 11:30 a .m. , with most begun at 9:30 a.m. The surveys lasted from one hal f hour to one hour, depending on the number of f l i e s observed. (See Chapter I for a d d i t i o n a l d e t a i l s of the f l y surveys in 1980.) PLANT COLLECTIONS AND DISSECTION In 1979, d i f f u s e knapweed plants were co l l ec t ed on August 22 at Ned's Creek, and on August 25 and 26 at Robertson's . In 1980, d i f fuse knapweed plants at Robertson's were c o l l e c t e d on August 23. The rapid loss of seeds from spotted knapweed seed heads 1 33 (Hubbard, 1971) and the necessity of c o l l e c t i n g a plant at a s ingle point in time meant that there was a tradeoff between los ing seeds and c o l l e c t i n g plants before a l l the seeds had a chance to mature. In 1979, spotted knapweed at Chase was c o l l e c t e d at three day in terva l s from August 8 to 20. Four plants were c o l l e c t e d from each treatment on each date, thus the v a r i a t i o n in seeds per plant due to c o l l e c t i o n date i s included in the within treatment v a r i a t i o n in the summary of treatment e f fects (Table 4.7 below). Analys is of variance on untransformed and log transformed data indicate that c o l l e c t i o n date has no s i g n i f i c a n t (a=0.05) effect on number of seeds per p lant . This resu l t held for both a one way ANOVA and a three way ANOVA ( inc luding f e r t i l i z e r and water treatments). In 1980, spotted knapweed plants were co l l ec t ed just before the f i r s t seed head on each plant shed i t s dr ied f l o r e t s . This c o l l e c t i o n extended over the per iod August 1 to August 24 as seed heads matured. The same procedure was used for both treated and contro l p l a n t s . Both methods w i l l underestimate t o t a l seed product ion, but by inc lud ing seeds that were not f u l l y developed in seed counts, e rrors are probably smal l . Plants were c o l l e c t e d i n d i v i d u a l l y by c l i p p i n g them off at ground l e v e l . They were then stored in folded and stapled paper bags at room temperature u n t i l d i s s e c t i o n . When they were d i s sec ted , the height of each plant was measured and the s i ze , developmental s ta tus , and locat ion of each bud were recorded. Damage r e s u l t i n g from grasshopper feeding was also noted. Bud s izes were measured with c a l i p e r s as described by Berube and 1 34 Harr is (1978) to the nearest m i l l i m e t e r . Buds large enough to contain e i ther g a l l s or seeds ( i . e . developed buds) were i n d i v i d u a l l y d i s s ec t ed . For these, the number of seeds and contents of any g a l l s present were noted. The contents were categorized according to f l y species (eas i ly d i s t ingu i shed) , larva (dead /a l ive ) , pupa (dead/a l ive ) , adult (dead/a l ive ) , pupal case, or no v i s i b l e remains. U. a f f i n i s g a l l s are hard and woody and those of U. quadri fasc iata are th in and papery. . The e f fects of f e r t i l i z a t i o n and watering on l a r v a l mortal i ty and development were evaluated by comparing the contents of g a l l s in developed buds from treated and contro l plants in 1980. Dif fuse knapweed buds were d i ssected during the period March 3 1 - A p r i l 28, 1981, and spotted knapweed buds were dissected during the period March 2 1 - A p r i l 8, 1981. In the di f fuse knapweed c o l l e c t e d from Robertson's in 1980, 4.1% of the g a l l s were attacked by the mite, Pynotes sp. (cf . H a r r i s , 1980a). It i s not known what the other sources of l a r v a l and pupal mortal i ty were. ADDITIONAL STATISTICAL METHODS A set of Tables (4 .2 , 4.6, 4.9-4.13 below) are used to present the main e f fec t s of two way analyses of variance of data from the treatments in 1979. S t a t i s t i c a l tests of in terac t ion terms are not presented i f the terms are not s i g n i f i c a n t at a=0.05. 1 35 RESULTS RESPONSE OF PLANTS DIFFUSE KNAPWEED The resul t s of analys i s of variance on d i f fuse knapweed c h a r a c t e r i s t i c s in 1979 show that both treatments s i g n i f i c a n t l y increased the number of buds per plant (Tables 4 .2 -4 .4 ) . The stronger response of bud numbers per plant to Table 4.2 - E f f ec t of f e r t i l i z a t i o n and watering on the t o t a l number of d i f fuse knapweed buds and the number of developed buds, 1 979 Character Treatment Si te Control Low High Buds Water NC 27.30 39.20 37. 25 a R 27. 1 3 36.57 45. 36 b F e r t . NC 23.67 34.47 45.62 c R 24.43 37.81 46.71 d Developed Water NC 10.88 15.10 1 5. 30 e Buds R 10.70 1 6.83 1 9.24 f F e r t . NC 8.90 1 4.57 1 7.82 g R 10.68 1 6.66 1 9.39 h S i t e s : NC - Ned's Creek, R - Robertson's. Let ters ind icate s t a t i s t i c a l tests of main e f f ec t s : a. F=5.18, df=2,171, p=0.006; b. F=6.42, df=2,l69, p=0.002; c. F=15.32, df=2 ,171 , p<0.00l; d. F=9.19, df=2,169, p<0.001; e. F=4.51, df=2,171, p=0.0l2; f. F=6.93, df=2,169, p=0.00l; g. F=11.63, df=2 ,171 , p<0.00l; h. F=5.46, df=2,169, p=0.005. nitrogen add i t i on compared to the effect of watering may be p a r t l y due to the timing of the f e r t i l i z e r app l i ca t ions (after b o l t i n g had begun). T a b l e 4.3 - E f f e c t of f e r t i l i z a t i o n and w a t e r i n g on d i f f u s e knapweed c h a r a c t e r i s t i c s and I n s e c t a t t a c k , Ned's Creek 1 9 7 9 F e r t i l i z a t i o n L e v e l None Low - H i g h W a t e r i n g L e v e l None Low High None Low H i g h None Low H i g h H e i g h t 19+1 2 0 ± 1 2111 2011 2311 2211 2011 24+1 2111 B u d s / P l a n t 16±2 22+3 32+6 3215 3714 34+4 3416 5817 45+7 Chewed/Plant 0.2±0 1 0.5±0 2 0.4+0.2 0.610.3 1.410.4 1.410.5 0.5+0.3 3.1+0 9 0.810 3 Chewed/Buds 0.01+0 01 0.03+0 01 0.0110.01 0.0210.01 0.0310.01 0.04+0.01 0.0110.01 0.0510 01 0.0210 01 Dev./PI ant 5.4±0 8 9+1 12+2 14±3 14±2 16±2 14+3 22+3 18+3 Dev./Buds 0.33+0 03 0.4210 04 0.41+0.05 0.3710.03 0.3610.04 0.4310.03 0.3510.04 0.3610 03 0.3910 03 UA/Plant 3.2+0 8 4±1 1113 712 7+1 1012 1418 1514 9+2 UA/Dev. 0.62+0 08 0.48+0 05 0.8610.07 0.5910.06 0.4910.05 0.6810.06 1.0710.09 0.7210 05 0.55+0 05 UA/Attacked 1.5±0 1 1.26+0 06 1.67+0.08 1.6310.09 1.4210.08 1.7210.07 2.210.1 1.6510 07 1 .44+0 06 Prop A t t UA 0.41±0 05 0.38+0 04 0.5110.03 0.3610.03 0.3410.03 0.40+0.03 0.48+0.03 0.4310 02 0.3810 03 UO/Plant 0.7+0 3 1 .610 5 2.910.9 312 2.310.7 2.710.8 312 412 2.810 8 UQ/Dev. 0.13+0 06 0.18+0 05 0.2310.04 0.2610.06 0.1710.08 0.1810.03 0.20+0.04 0.19+0 03 0. 1710 03 UO/Attacked 1.9+0 5 1 .7+0 2 1.610.2 2.2+0.3 1.610.2 1.610.1 1.710.2 1 .7+0 2 1 .410 1 Prop A t t UQ 0.07+0 03 0.1110 02 0.1510.02 0.1210.02 0.1110.02 0.1110.02 0.1210.02 0.1110 02 0.12+0 02 S e e d s / P l a n t 25±4 54110 76117 68+14 80118 89119 55120 110120 80115 Seeds/Oev. 4.9±0 4 6.3+0 3 6.110.3 5.5+0.3 6.010.3 5.9+0.3 4.310.3 5. 1+0 3 4.810 3 Seeds/Prod 7.2+0 4 7.5+0 3 7.710.3 6.7+0.3 8.110.3 7.310.3 6.410.3 7.0+0 2 7.0+0 3 Prop A b o r t e d 0.07±0 01 0.1210. 02 0.1010.02 0. 1110.02 0.0910.01 0.1010.02 0.0910.02 0.0810 02 0.0810 01 T a b l e 4.4 - E f f e c t of f e r t i l i z a t i o n and w a t e r i n g on d i f f u s e knapweed c h a r a c t e r i s t i c s and i n s e c t a t t a c k , Robertson's 1979 F e r t i l i z a t i o n Level None Low H i g h W a t e r i n g L e v e l None Low High None Low H i g h None ' Low H i g h H e i g h t 18+1 18.0+0. .8 2212 21 + 1 2111 2312 17.810. .9 2312 2411 B u d s / P l a n t 25±5 19±2 29+4 3215 3719 46+6 25+3 54111 62111 Chewed/Plant 2.210 .7 1 . 1+0 .3 2.910, .9 5+3 611 1113 4.610. .7 1313 2115 Chewed/Buds 0.0810 .02 0.06±0. .02 0.1010 .02 0.1210.03 0.17+0.03 0.20+0.03 0.2010. .03 0.2310.03 0.2810. 04 Dev./Plant . 11+2 9.010, .8 12+2 1212 18+5 2013 9+ 1 23+5 2615 Dev./Buds 0.5210. .04 0.4910. .03 0.4210. .02 0.3710.03 0.52+0.02 0.4810.03 0.3910. .03 0.4710.04 0.4510. 03 UA/Plant 6+ 1 711 10+3 813 26114 1714 612 18+6 24+5 UA/Dev. 0.57±0. .06 0.73+0. .07 0.8010. .07 0.6810.07 1.4510.08 0.8410.05 0.6410. .07 0.7610.05 0.90+0. 05 UA/Attacked 1.65±0. .09 1.5310. . 10 1.6310. .09 1.7010.10 2.22+0.08 1.6210.06 1.63+0. . 10 1.7310.06 1.7210. 06 Prop A t t UA 0.35+0. .03 0.4710. .04 0.49+0. .03 0.40+0.03 0.65+0.03 0.51+0.03 0.3910. .04 0.4410.02 0.5210. 02 UQ/Plant 1.4+0. 5 1.0+0. 3 1.210. .6 0.510.3 1.010.3 2.510.7 1.2+0.4 412 5+ 1 UQ/Dev. 0.12+0. 03 0.1110. .03 0.10+0. .03 0.0410.02 0.0510.02 0.1210.03 0.1210. ,03 0.15+0.03 0.1810. 02 UQ/At tac k e d 1.5+0. 2 1.3+0. 2 1.5+0. .3 1.4+0.2 1.4+0.1 1.5+0.2 1.410. 2 2.010.2 1.2910. 08 Prop A t t UQ 0.08±0. 02 0.0810. 02 0.07+0. ,02 0.0310.01 0.0410.01 0.0810.02 0.0910. 02 0.0810.01 0.1410. 02 S e e d s / P l a n t 51+9 36+4 4117 4919 65+17 88+22 41110 97122 1 11125 Seeds/Dev. 4.6+0. 3 4.0+0. 3 3.510. 2 4.210.3 3.510.2 4.410.2 4.5+0.3 4.210.2 4.210. 2 Seeds/Prod 6.2+0. 3 5.6+0. 3 5.410. 3 5.710.3 5.610.2 6.2+0.2 6.710. ,4 5.610.2 5.510. 2 Prop A b o r t e d 0.0710.02 0.08+0.02 0.1010.02 0.0910.02 0.0610.01 0.0810.02 0.1010.02 0.0810.02 0.0710.01 1 38 Based on a comparison of the means for the main effects at the two di f fuse knapweed s i t e s in 1979 (Table 4 .2) , there d id not appear to be s i g n i f i c a n t d i f ferences in the responses to the treatments which could be a t t r i b u t e d to d i f ferences in the nutr ients ava i l ab l e at the two s i t e s . I conclude that the s i g n i f i c a n t d i f ferences between s i t e s (Table 4 .1) , in p a r t i c u l a r in n i t r a t e concentrat ion, were unimportant r e l a t i v e to the experimental treatments. The f e r t i l i z e d and watered d i f fuse knapweed in 1980 d i f f e r e d s i g n i f i c a n t l y from the c o n t r o l in the number of buds per plant (log transformed data, t=3.04, df=68, p=0.003; Table 4 .5) . The proport ion of buds developed per plant was much lower on the f e r t i l i z e d and watered plants than on the contro l (Table 4.5) . The di f ference was p r i m a r i l y due to se l ec t ive feeding by grasshoppers on f e r t i l i z e d plants (discussed below). F e r t i l i z e r "burn" was apparent at the higher rate of f e r t i l i z a t i o n in 1980, despite the higher rate of watering. SPOTTED KNAPWEED The f e r t i l i z a t i o n and watering in 1979 had r e l a t i v e l y l i t t l e effect on spotted knapweed. The effects on both t o t a l buds and developed buds were in the d i r e c t i o n expected (Table 4.6) , however t o t a l U. a f f i n i s g a l l s , to ta l U. quadr i fasc ia ta g a l l s , and t o t a l seeds were not affected by f e r t i l i z a t i o n or watering at Chase (Table 4 .7) . F e r t i l i z a t i o n and watering in 1980 produced spotted knapweed with many more buds and developed buds per plant than contro l plants (Table 4 .8) . There was no evidence of f e r t i l i z e r 1 39 Table 4.5 - Ef fec t of f e r t i l i z a t i o n and watering on d i f fuse knapweed c h a r a c t e r i s t i c s and insect attack, Robertson's 1980 Character * Control Exper imental Number of Plants 50 20 Height 3 0 . 0 ± 0 . 7 28. 1 ± 1 . 3 Buds/Plant 72±6 124±21 Chewed/Plant 6 . 6 ± 1 . 2 2 9 . 3 ± 6 . 1 Chewed/Buds 0 . 0 9 ± 0 . 0 1 0 . 2 3 ± 0 . 0 3 Developed/Plant 3 0 . 7 ± 2 . 4 3 5 . 9 ± 8 . 1 Developed/Buds 0 . 4 4 ± 0 . 0 2 0 . 2 8 ± 0 . 0 3 UA/Plant 37±4 37±14 UA/Developed 1 . 2 0 ± 0 . 0 4 1 . 0 2 ± 0 . 0 5 UA/Attacked 2 . 1 8 ± 0 . 0 5 2. 1 9 ± 0 . 0 7 Prop Att UA 0 . 5 5 ± 0 . 0 1 0 . 4 7 ± 0 . 0 2 UQ/Plant 8 . 3 ± 1 .3 5 . 0 ± 2 . 8 UQ/Developed 0 . 2 7 ± 0 . 0 2 0 . 1 4 ± 0 . 0 2 UQ/Attacked 1 . 8 4 ± 0 . 0 7 1 . 7 7 ± 0 . 1 3 Prop Att UQ 0 . 1 5 ± 0 . 0 1 0 . 0 8 ± 0 . 0 1 Seeds/Plant 106±11 106±32 Seeds/Developed 3 . 4 7 ± 0 . 1 0 2 . 9 7 ± 0 . 1 6 Seeds/Produc ing 6 . 0 3 ± 0 . 1 2 5 . 9 8 ± 0 . 2 4 Prop Aborted 0 . 2 0 ± 0 . 0 1 0 . 1 3 ± 0 . 0 1 * Deta i l ed descr ipt ions of these characters are given in Table 2.5. "burn" at t h i s s i t e . RESPONSE TO PLANTS GALL FLIES DIFFUSE KNAPWEED Only l i m i t e d numbers of f l i e s were observed in 1979. If the numbers of f l i e s on plants in each of the nine treatment plots are combined, the expected r e l a t i o n s h i p with the number of buds or developed buds is a p o s i t i v e l inear one. This was observed at both d i f fuse knapweed s i t e s for both 1 40 Table 4.6 - Ef fec t of f e r t i l i z a t i o n and watering on the t o t a l number of spotted knapweed buds and the number of developed buds, 1979 Character Treatment S i te Contro l 'Low High Buds Water C 7.42 7.95 8.45 a F e r t . C 5.78 8.25 9.78 b Developed Water C 2.95 2.70 3.13 c Buds F e r t . C 2.58 2.92 3.28 d S i t e : C - Chase. Let ters indicate s t a t i s t i c a l tests of main e f fec t s : a. F=0.14, df=2,171, p=0.867; b. F=14.38, df=2,171, p<0.00l; c. F=0.35, df=2,171, p=0.704; d. F=1.09, df=2,171, p=0.337. species of f l i e s (r>0.70, df=7, p<0.05 in a l l cases except for U. quadr i fasc iata vs . t o t a l buds at Robertson's , r = 0.60, df = 7, p>0.05). The counts of f l i e s on plants in 1979 indicate that the r e l a t i v e abundance of f l i e s was about twice as great at Ned's Creek (370 vs. 170 at Robertson's ) . In 1980, both species of f l y were observed more frequently on the contro l plants than on the f e r t i l i z e d and watered p lants . This was true based on number of p lants and on number of buds (x 2 ^7.3, df=1, p<0.0l for a l l comparisons). The average of 0.27 f l i e s per bud on contro l p lants was almost three times as great as the average of 0.098 for f e r t i l i z e d p l a n t s . This trend holds over the ent i re range of buds per plant (Figure 4.3) . This was in the opposite d i r e c t i o n of the response expected, e spec ia l ly considering the observations in 1979. Within the group of f e r t i l i z e d p lants , a p o s i t i v e c o r r e l a t i o n was observed between buds per plant and f l i e s per p l a n t . There was no o v e r a l l Table 4.7 - E f f e c t o f f e r t i l i z a t i o n and w a t e r i n g on s p o t t e d knapweed c h a r a c t e r i s t i c s and I n s e c t a t t a c k . Chase 1979 F e r t i l i z a t i o n Level None Low H i g h Watering L e v e l None Low High None Low H i g h None Low H i g h Height 2 1 + 1 22+ 1 1911 2 1 + 1 1811 20+2 1 9 + 1 2212 2111 Buds/Plant 6.1±0. .7 6.4+0.9 4.9+0. .5 811 7.4+0, .8 10+2 8.3+0. .8 1 0 + 1 1112 Chewed/Plant O.3±0, . 1 1.410 .3 0.510, . 1 0.910. .3 2.310. .8 1.810, .8 2.7+0. .8 3.5+0. .8 3.110, .8 Chewed/Buds 0.03±0. . 0 1 0.24+0, .06 0.1010, .03 0.12+0. .04 0.2810. .07 0.1310, .04 0.2610. .06 0 . 3 1 + 0 . .06 0.2410. .05 Dev./Plant 3.2+0. .4 2.810, .05 1.910.3 3.2+0. .6 2.010. .5 3.610. .7 2.4+0.3 3.410. .5 4.011. .0 Dev./Buds 0.54±0. .04 0.4510 .05 0.3710, .04 0.40+0. .04 0.2910, .04 0.3810 .04 0.3210. .04 0.3710. . 0 5 0.3610. .05 UA/Plant 3.0+0. .7 3.5+0. .8 3.211. .0 612 311 3.7+0. .9 1.7+0.4 5+2 7+2 UA/Dev. 1.0±0. .2 1.3+0. .2 1.810. .3 1.9+0. .3 1.610. ,3 1.110. .2 0.810. .2 1.710. .3 1.810. 2 UA/At t a c k e d 2.3+0. .2 2.1+0. .2 2.310. .4 3.0+0. ,3 3.010. .4 2.210. .2 1.810. .2 3.210. .5 2.810. 2 Prop A t t UA 0.44+0. .07 0.60+0. .07 0.7710. .07 0.6310. .06 0.5610. .08 0.4910. .06 0.4210. .07 0.55+0. .06 0.6210. 06 UO/Plant 0.2+0. . 1 0.0510. .05 0.3+0. . 1 0.610. .3 0.110. . 1 0.8+0. .4 0.3+0. . 1 0.5+0. 3 0.1510. 08 UQ/Oev. 0.05±0. ,04 0.02+0. .02 0.1710. .09 0.1810. .07 0.0610. 06 0.210. . 1 0 . 1 1 + 0 . .07 0.1510. 08 0.0410. ,02 UQ/Attacked 1.5±0. .5 1 .0 1.510. 3 1.610. .3 2.0 2.5+0. .6 1.7+0. 3 2.310. ,5 1 .0 Prop A t t UQ 0.0310. ,02 0.02+0. 02 0.1110. . 0 5 0.1110. ,04 0.03+0. .03 0.09+0. ,03 0.07+0. 04 0.0710. 03 0.04+0. ,02 Seeds/Plant 2715 30+7 1313 1816 9+3 30110 1213 20+4 30110 Seeds/Dev. 9.010. 4 10.8+0. 7 7.2+ 1 . 0 5.810. 4 4.810. 8 8.5+0. 9 5.2+0. 8 6.5+0. 5 7.6+0. 9 Seeds/Prod 9.710. 3 1 1.810. 6 8.711. 0 8.010. 8 6.910. 9 1 2.310. 7 8.110. 8 9.810. 8 1 2.410. 9 Prop A b o r t e d 0.0910.03 0.09+0.04 0 . 1 1+0.04 0.0410.01 0.0810.02 0.1310.03 0.07+0.02 0.07+0.02 0.1010.03 1 42 Figure 4.3. Counts of t o t a l g a l l f l i e s observed per plant for treated and contro l d i f fuse knapweed plants at Robertson's in 1980. 70. o • o • I o ° °o a o fertilized . control 40 80 120 160 200 240 280 320 360 400 Buds per plant 1 44 Table 4.8 - Effect of f e r t i l i z a t i o n and watering on spotted knapweed c h a r a c t e r i s t i c s and insect at tack, Chase 1980 Character * Control Exper imental Number of Plants 50 30 Height 26 .4±1 .1 2 6 . 0 ± 1 . 5 Buds/Plant 4 . 5 ± 0 . 3 1 6 . 9 ± 2 . 7 Chewed/Plant 0 . 5 0 ± 0 . 13 5 . 3 0 ± 0 . 7 4 Chewed/Buds 0 . 0 8 ± 0 . 0 2 0 . 3 2 ± 0 . 0 3 Dev . /P lant 2 . 6 4 ± 0 . 1 8 7 . 5 0 ± 0 . 9 0 Dev./Buds 0 . 5 8 ± 0 . 0 3 0 . 4 2 ± 0 . 0 3 UA/Plant 3 . 2 ± 0 . 5 7 . 3 ± 1 . 6 UA/Developed 1 . 2 2 ± 0 . 1 1 1 . 0 3 ± 0 . 0 9 UA/Attacked 1 , 8 7 ± 0 . 1 3 1 . 9 1 ± 0 . 1 0 Prop Att UA 0 . 6 5 ± 0 . 0 4 0 . 5 4 ± 0 . 0 3 UQ/Plant 0 . 8 4 ± 0 . 2 6 0 . 7 0 ± 0 . 3 7 UQ/Developed 0 . 3 2 ± 0 . 0 8 0 . 1 0 ± 0 . 0 3 UQ/Attacked 2 . 1 0 ± 0 . 3 4 2 . 1 0 ± 0 . 3 5 Prop Att UQ 0 . 1 5 ± 0 . 0 3 0 . 0 5 ± 0 . 0 1 Seeds/Plant 40±4 150±26 Seeds/Dev. 1 5 . 4 ± 0 . 8 2 1 . 2 ± 0 . 6 Seeds/Prod. 1 7 . 2 ± 0 . 7 2 4 . 4 ± 0 . 2 Prop. Aborted 0 . 0 3 ± 0 . 0 1 0 . 0 5 ± 0 . 0 2 * De ta i l ed descr ipt ion of these characters are given in Table 2.5. d i f f erence between the two f l y species in the r a t i o of f l i e s observed on f e r t i l i z e d plants to f l i e s observed on contro l p lants (x 2=0.79, df=1, p=0.374). SPOTTED KNAPWEED In contrast to the response on f e r t i l i z e d d i f fuse knapweed, g a l l f l i e s responded to f e r t i l i z e d spotted knapweed in 1980 as expected. U. a f f i n i s occurred on f e r t i l i z e d and c o n t r o l plants in approximately the same proport ion as the number of buds and developed buds per p lant . Of the t o t a l 1 45 number of U . a f f i n i s a d u l t s , 67% were observed on f e r t i l i z e d plants compared with 69% of the t o t a l buds (x 2 =0 .23 , df=1, p=0.632) and 63% of the developed buds (x 2 =3 .37 , df=1, p=0.066). The proportion of U . quadr i fasc i a ta of the t o t a l f l i e s was lower on f e r t i l i z e d plants (F i sher ' s exact t e s t , p<0.05) . OTHER HERBIVORES The responses of various other herbivores also indicate that plants had been a l t ered in b i o l o g i c a l l y s i g n i f i c a n t ways. D i f f erent herbivores responded to the changes at the d i f f eren t s i t e s . At Ned's Creek in 1979 and at Robertson's in both 1979 and 1980, grasshopper chewing damage was evident on the f e r t i l i z e d and watered di f fuse knapweed plants which were c o l l e c t e d at the end of the summer. The estimates of the damage caused (Tables 4 . 3 - 4 . 5 , 4.9) are s l i g h t underestimates, because the chewed branches were no longer v i s i b l e . At Ned's Creek, the damage was r e l a t i v e l y l i g h t ; o v e r a l l , 2 . 4 ± 0 . 3 % of the buds were chewed. There was no l inear e f fect of the treatments on the proport ion of buds chewed. At Robertson's , the grasshopper attack was greater (overal l mean 16±1%) and the proport ion of buds chewed increased s i g n i f i c a n t l y with watering and f e r t i l i z a t i o n . The same increased attack with treatment was evident for Robertson's in 1980 (Table 4 . 5 ) . At Chase in 1979, the proport ion of spotted knapweed buds chewed by grasshoppers increased with f e r t i l i z a t i o n and watering, though the e f fect of watering was not l inear (Table 4 . 9 ) . In 1980, the proport ion of buds or branches chewed on the f e r t i l i z e d plants was much higher than the contro l (Table 4 . 8 ) . 1 46 Table 4.9 - Ef fect of f e r t i l i z a t i o n and water ing on the proportion of knapweed buds chewed, 1 979 Spec ies Treatment Si te Contro l Low High Di f fuse Water NC 0.013 0.035 0.024 a Knapweed R 0.131 0. 155 0. 196 b F e r t . NC 0.016 0.030 0.026 c R 0.081 0. 163 0.237 d Spotted Water C 0. 1 37 0.275 0. 1 57 e Knapweed F e r t . C 0. 1 25 0. 175 0.269 f S i t e s : NC - Ned's Creek, R - Robertson's , C - Chase. Let ters indicate s t a t i s t i c a l tests of main e f f ec t s : a. F=4.98, df=2,171, p=0.008; b. F=3.82, df=2,170, p=0.024; c . F=2.10, df=2,171, p=0.126; d . F=22.01, df=2,l70, p<0.00l; e. F=6.92, df=2,171, p=0.00l; f. F=6.66, df=2,171, p=0.002. Unident i f i ed spi t t lebugs (Homoptera: Cercopidae) a l so responded to the changes caused by the added nitrogen in spotted knapweed in 1979. Only 8.9% of the t o t a l number of sp i t t l ebugs (N=45) were observed on u n f e r t i l i z e d plants compared to the expected one t h i r d (x2 = 12.1, df =1, p<0.001). The cercopids a lso attacked f e r t i l i z e d plants more heavi ly than would be expected on the basis of bud numbers (x 2=5.65, df=1, p=0.0l7) . CHANGE IN INTERACTION ATTACK LEVELS 1979 DIFFUSE KNAPWEED The proport ion of buds aborted was not s i g n i f i c a n t l y affected by the increased resources ava i lab le to the plants at e i ther of the two d i f fuse knapweed s i tes in 1979 (Table 4.10). The outcome of g a l l f l y attack could also r e f l e c t increased 1 47 Table 4.10 - Ef fec t of f e r t i l i z a t i o n and watering on the proport ion of knapweed buds aborted, 1 979 Species Treatment S i te Control Low High Di f fuse Water NC 0.091 0.097 0.093 a Knapweed R 0.086 0.072 0.083 b F e r t . NC 0.096 0. 100 0.085 c R 0.084 0.076 0.080 d Spotted Water C 0. 065 0.079 0.113 e Knapweed F e r t . C 0.095 0.083 0.080 f S i t e s : NC - Ned's Creek, R - Robertson's, C - Chase. Let ters indicate s t a t i s t i c a l tests of main e f fec ts : a. F=0.09, df=2 ,171, p=0.9l8; b. F=0.58, df=2,170, p=0.561; c. F=0.60, df=2 ,171, p=0.548; d. F=0.21, df=2,170, p=0.8l4; e. F=2.44, df=2 ,171, p=0.090; f. F=0.26, df=2 ,171, p=0.770. plant resources i f the proport ion of buds aborted was unchanged and the number of g a l l s per developed bud increased. At Ned's Creek, the number of U. a f f i n i s g a l l s per developed bud in contro l plants was not s i g n i f i c a n t l y d i f f erent from the number in a l l treated plants combined (t=1.03, df=120, p=0.305), however the e f fec t s of the treatments were quite complex. With no f e r t i l i z a t i o n , the number of U. a f f i n i s ga l l s per developed bud increased with watering. At the high f e r t i l i z a t i o n l e v e l , the number of U. a f f i n i s g a l l s per developed bud dropped with increased watering (hence the s ign i f i cance of the in t erac t ion term in the ANOVA; Table 4 .11) . The changes in the number of U. a f f i n i s ga l l s per developed bud with the treatments were not consistent between d i f fuse knapweed s i t e s . At Robertson's , the number of U. a f f i n i s g a l l s per developed bud in a l l treated plants combined was 1 48 Table 4.11 - Effect of f e r t i l i z a t i o n and watering on the number of U. a f f i n i s g a l l s per developed bud, 1 979 Species Treatment S i te Control Low High Di f fuse Water NC 0.789 0. 598 0.677 a Knapweed R 0.632 0. 703 0.861 b F e r t . NC 0.678 0.590 0.747 c R 0.697 0.784 0.778 d Spotted Water C 1 .26 1 . 54 1 . 52 e Knapweed F e r t . C 1 .27 1 .50 1.51 f S i t e s : NC - Ned's Creek, R - Robertson's , C - Chase. Let ters indicate s t a t i s t i c a l tests of main e f f ec t s : a. F=7.63, df=2,2350, p<0.00l; b. F=9.99, df=2,2369, p<0.00l; c. F=5.62, df=2,2350, p=0.004; d. F=1.46, df=2,2369, p=0.232; e. F=1.11, df=2,490, p=0.330; f. F=0.84, df=2,490, p=0.431. The interact ions of the main ef fects at Ned's Creek and Chase were s t a t i s t i c a l l y s i g n i f i c a n t (F=11.98, df=4,2350, p<0.001; F=4.68, df=4,490, p<0.00l, r e s p e c t i v e l y ) , but in nei ther case was the in teract ion consistent among r e p l i c a t e s . s i g n i f i c a n t l y higher than in contro l plants (t=3.34, df=275, P<0.001). There was an increase in U. a f f i n i s g a l l s per developed bud with watering (Table 4.11). The increase was consistent among rep l i ca te s and among f e r t i l i z a t i o n l e v e l s . This was also the only treatment-s i te-year combination which had a s i gn i f i can t increase in the proport ion of buds developed (Table 4.12). There was no treatment ef fect on the number of U. quadr i fasc ia ta g a l l s per developed bud that was consis tent among rep l i ca tes at e i ther d i f fuse knapweed s i t e (Table 4.13). (The effect of nitrogen at Robertson's was not cons is tent among r e p l i c a t e s . ) 1 49 Table 4.12 - E f f ec t of f e r t i l i z a t i o n and watering on i the proport ion of knapweed buds developed, 1979 * Species Treatment S i t e Control Low High Diffuse Water NC 0.379 0.403 0.436 a Knapweed R 0.476 0.557 0.552 b F e r t . NC 0.396 0.421 0.401 c R 0.517 0.517 0.549 d Spotted Water C 0.486 0.530 0.447 e Knapweed F e r t . C 0. 520 0.450 0.493 f * - Proportions are adjusted by subtracting chewed buds from the t o t a l number of buds per p lant . S i t e s : NC - Ned's Creek, R - Robertson's, C - Chase. Let ters indicate s t a t i s t i c a l tests of main e f fec t s : a. F=1.81, df=2,171, p=0.167; b. F=5.04, df=2,170, p=0.008; c. F=0.38, df=2,171, p=0.684; d. F=0.84, df=2,170, p=0.433; e. F=1.57, d f = 2 , l 7 l , p=0.211; f. F=2.11, df=2,171, p=0.l24. The i n t e r a c t i o n of the main effects at Robertson's was s t a t i s t i c a l l y s i g n i f i c a n t (F=4.80, df=4,l70, p<0.00l), but was not consistent among r e p l i c a t e s . The main effect of watering was cons is tent among r e p l i c a t e s . The number of g a l l s per bud was not s i g n i f i c a n t l y reduced by any of the treatments in 1979, despite large increases in the number of buds per p l a n t . SPOTTED KNAPWEED Neither of the two predicted responses of insect attack to improved plant qua l i ty were detected with the treatments of spotted knapweed in 1979 (bud abort ion , Table 4.10; U. a f f i n i s g a l l s per developed bud, Table 4.11; U. q u a d r i f a s c i a t a g a l l s per developed bud, Table 4.13). ATTACK LEVELS 1980 DIFFUSE KNAPWEED The proportion of buds aborted was s i g n i f i c a n t l y lower on f e r t i l i z e d and watered plants 1 50 Table 4.13 - E f f e c t of f e r t i l i z a t i o n and watering on the number of U . quadr i fasc iata g a l l s per developed bud, 1979 Spec ies Treatment S i te Control Low High Dif fuse Water NC 0.214 0.181 0. 188 a Knapweed R 0.094 0. 139 0. 1 46 b F e r t . NC 0. 193 0.200 0. 185 c R 0.110 0.087 0. 172 d Spotted Water C 0.113 0.079 0. 1 33 e Knapweed F e r t . C 0.067 0. 168 0.093 f S i t e s : NC - Ned's Creek, R - Robertson's, C - Chase. Let ters ind ica te s t a t i s t i c a l tests of main e f f ec t s : a. F=0.49, df=2,2350, p=0.614; b. F=1.43, df=2,2369, p=0.239; c . F=0.11, df=2,2350, p=0.892; d. F=6.87, df=2,2369, p=0.00l; e. F=0.27, df=2,490, p=0.763; f. F=1.87, df=2,490, p=0.!56. than on c o n t r o l plants (Table 4.5) apparently supporting the p r e d i c t i o n of reduced bud abortion in treated p lant s , however the reduction i s only marginally s i g n i f i c a n t i f the effect of chewed buds is removed (t=1.70 , df = 68, p=0.094). There was no increase in U . a f f i n i s g a l l s per developed bud corresponding to the lower proportion of buds aborted; instead, the number of g a l l s per developed bud dropped s i g n i f i c a n t l y for both species . The proportion of developed buds attacked was lower for both U . aff i n i s (x 2 = 12.72, df=1, p<0.00l) and U . quadr i fasc i a t a ( x 2 = 22 .88, df=1, p<0.001) and the numbers of g a l l s in attacked buds were not s i g n i f i c a n t l y d i f f erent for e i ther f l y species (Table 4.5) . The lower proportions of buds aborted and attacked on treated plants are consistent with the lower counts of f l i e s per plant r e l a t i v e to c o n t r o l s . The 151 proport ional reduction in g a l l s per bud was much less than the reduction in the number of adult f l i e s . After adjust ing for the e f fect of grasshopper damage, the proportion of buds developed was s i g n i f i c a n t l y lower on treated d i f fuse knapweed. This drop p a r a l l e l s the drop in the proportion of buds attacked. An increase in both of these var iables was observed with the treatments at Robertson's in 1979. SPOTTED KNAPWEED Bud abort ion was not s i g n i f i c a n t l y reduced on the f e r t i l i z e d and watered spotted knapweed plants in 1980 (Table 4.8) nor d id the number of g a l l s per developed bud increase. A lower proportion of buds on f e r t i l i z e d plants was attacked by U. a f f i n i s than on the contro l s (Table 4.8; x 2=4.47, df=1, p=0.035). The same r e l a t i o n s h i p holds for U. quadr i fa sc ia ta (Table 4.8; x 2 =11.5, df=1, p=0.00l). The number of g a l l s in attacked buds d id not d i f f e r s i g n i f i c a n t l y between treated and untreated p l a n t s . The r e l a t i v e l y smaller proportion of U. quadr i fasc ia ta g a l l s in f e r t i l i z e d plants was consistent with the r e l a t i v e l y fewer adults of th i s species observed on these p lants . At Chase in the same year, there was a s l i g h t decl ine in g a l l s per developed bud despite approximately equal numbers of adul ts per developed bud. This reduction was due to a drop in the proport ion attacked (Table 4.8) which suggests that a f e r t i l i z e r e f fect on the timing of bud growth may be respons ib le . 1 52 LARVAL SURVIVAL AND DEVELOPMENT DIFFUSE KNAPWEED F e r t i l i z a t i o n and watering of d i f fuse knapweed in 1980 increased morta l i ty and the rate of development for U. a f f i n i s larvae (Table 4.14). Table 4.14 - Contents of Urophora g a l l s from contro l and treated di f fuse knapweed p lant s , Robertson's 1980 PERCENT IN CATEGORY * G a l l Contents Control a f f i n i s Treated U. quadr i fasc ia ta Control Treated Live Larvae 51 .9 44.0 1 .2 0.0 Dead Larvae 18.2 22.9 9.8 28.3 Proceeding in Development 29.9 33.0 89.0 71.7 (Successful) ** (71.5) (67.5) (57.5) (62.0) N 715 698 82 99 Columns may not add to 100% because of rounding. Proport ion of larvae proceeding in development that survived to time of d i s s e c t i o n . Larvae of th i s species in treated plants were more l i k e l y to die than larvae in contro l plants (x 2=7.96, df=1, p=0.005). A higher proport ion of U. a f f i n i s larvae in treated plants proceeded in development to the next generation ( i . e . d id not enter diapause; excluding dead larvae , x 2=4.57, df=1, p=0.033). I detected no s i g n i f i c a n t d i f ference in the morta l i ty of U. a f f i n i s larvae proceeding in development between treated and untreated plants (x 2=1.04, df=1, p=0.308). The counts of l i v e U. quadr i fasc ia ta larvae were too low to obtain r e l i a b l e estimates of l a r v a l morta l i ty or of the 1 53 proportion proceeding in development, though the data suggest higher morta l i ty in f e r t i l i z e d plants (Table 4.14). The success of larvae proceeding in development was not s i g n i f i c a n t l y af fected by f e r t i l i z a t i o n (x 2=0.30, df=1, p=0.584). SPOTTED KNAPWEED Larvae of U. af f i ni s were s i g n i f i c a n t l y af fected by f e r t i l i z a t i o n and watering of spotted knapweed in 1980. The proport ion of l i v e larvae in a l l g a l l s was s imi lar in treated and contro l plants (62.8%, N=218 vs . 62.1%, N=161 on the c o n t r o l ) , but l a r v a l morta l i ty was reduced (from 33.5% to 16.5%) and an increased proport ion proceeded in development (from 4.4% to 21.1%). If the larvae which proceeded in development are excluded, 35% of larvae died in contro l plants compared with 21% in the f e r t i l i z e d plants (x 2=8.12, df=1, p=0.004). U. quadr i fasc ia ta emerged from a l l 21 g a l l s in f e r t i l i z e d spotted knapweed plants as compared with adults emerging from only 71% of 42 g a l l s in c o n t r o l p lants (F i sher ' s exact tes t , p<0.005). 1 54 DISCUSSION RESPONSE OF PLANTS Both treatments increased the number of buds per p l a n t , with the exception of the watering treatment at Chase in 1979. S i m i l a r l y , Schirman (1981) observed both s o i l moisture and s i t e p r o d u c t i v i t y ef fects on bud product ion. The weaker response of d i f fuse knapweed to the treatments in 1980 was the opposite of what was expected, given the higher a p p l i c a t i o n rates . This observation suggests that the plants were c lose to the ir phys io log i ca l l i m i t for nitrogen absorption in 1979 and that the addit ion of more f e r t i l i z e r a c t u a l l y had a de le ter ious effect on plant growth. This e f fect occurred despite the higher watering l eve l s and the higher r a i n f a l l in 1980, which would have tended to a l l e v i a t e the negative e f fec t of a too heavy f e r t i l i z e r a p p l i c a t i o n . The contrast between 1979 and 1980 was much c loser to the expected di f ference on spotted knapweed. The same ef fects were observed in both years and the response of bud numbers and developed buds to the treatment was stronger in 1980. This suggests that spotted knapweed can u t i l i z e higher l eve l s of appl i ed nitrogen than dif fuse knapweed. RESPONSE TO PLANTS GALL FLIES The data for 1979 indicate that there was no 1 55 important deviat ion from the d i s t r i b u t i o n of f l i e s expected on the basis of bud numbers at any of the three s i t e s . S i m i l a r l y , the response of the f l i e s to f e r t i l i z e d and watered spotted knapweed in 1980 matched the change in bud numbers as expected. The data from Robertson's in 1980 were an exception to th i s pat tern . The g a l l f l i e s were observed r e l a t i v e l y less often on the treated p lant s , despite higher numbers of buds per p lant . This reduced f l y density l ed to a lower proportion of buds containing g a l l s for both spec ies . The higher r e l a t i v e density of grasshoppers on treated p lants could have caused the lower f l y density i f f l i e s were d i s turbed by the grasshoppers and l e f t the knapweed p lants . This explanation is consistent with the observations from a l l other s i t e s and years. OTHER HERBIVORES Vince and V a l i e l a (1981) observed a s imi lar invertebrate response to the ones I obtained when they f e r t i l i z e d marsh p l o t s . Their treatment increased the standing crop of insect herbivores , p r i m a r i l y homopterans and grasshoppers, the same groups that responded to improved knapweed q u a l i t y . The response of grasshoppers to f e r t i l i z e d knapweed suggests that one of the fac tors l i m i t i n g the attack by grasshoppers may be the low nitrogen content of knapweed. McNei l l and Southwood (1978) argue that low nitrogen may act as a plant defence. Popova (1964; c i t e d in Watson, 1972) and Fletcher (1961; c i t e d in Watson, 1972) both observed low prote in l eve l s in knapweed. Bo l t ing p lants had less than 8.3% true 1 56 prote in (by dry weight) at the beginning of the summer. This dec l ined to close to zero by the end of the summer. These values (cf . Strong et a_l. , 1984) indicate that the low nitrogen content of knapweed may indeed be a b a r r i e r to feeding by d e f o l i a t o r s . The g a l l f l y larvae are probably feeding on the t i s sues in bo l t ing plants which have the highest nitrogen concentrations (McNeil l and Southwood, 1978). Because i t was concentrated on treated p lant s , the feeding by grasshoppers tended to equalize the number of developed buds between treated and contro l plants (cf . Parker, 1984). S i m i l a r l y , Onuf e_t a l . (1977) observed that a marked increase in the nitrogen content of mangroves led to a loss of biomass to herbivores four times greater than on u n f e r t i l i z e d t ree s . As a r e s u l t , the biomass of f r u i t and observed production of leaves were not s i g n i f i c a n t l y d i f f e r e n t between groups. The response of other herbivores in the Urophora- Centaurea system i l l u s t r a t e s the "leakiness" of e c o l o g i c a l system boundaries. For every set of organisms and i n t e r a c t i o n s , there are forces outside of the system which influence the v a r i a b l e s included in the system. The experiments described in t h i s Chapter indicate that the response of other herbivores to v a r i a t i o n in knapweed plant q u a l i t y may s u b s t a n t i a l l y a l t e r the g a l l fly-knapweed system. CHANGE IN INTERACTION EFFECT OF BUD ABORTION The improved resource status of the 1 57 plants d id not s i g n i f i c a n t l y a l t e r the proportion of buds aborted, except to reduce i t s l i g h t l y at Robertson's in 1980. These observations suggest that the experimental treatments d id not a l t e r knapweed's propensity to abort buds under heavy insect a t tack . Much larger increases in s o i l moisture than used in t h i s Chapter might reduce the propensity for knapweed to abort buds (Chapter V ) . A reduced propensity to abort buds could be consistent with the resu l t s obtained i f the o v e r a l l attack on treated plants was more intense. This would be re f l ec ted in more g a l l s per developed bud. Cons i s t ent ly higher numbers of g a l l s per developed bud were only observed in one instance: watering treatments at Robertson's in 1979. This increase was c o r r e l a t e d with an increase in the proport ion of buds developed. (A dec l ine in the proport ion of buds developed was corre la ted with a drop in the proport ion of buds attacked at Robertson's in 1980). The proport ion of buds developed may be an indicator to a q u a l i t a t i v e l y higher "carrying capacity" of buds. Because the changes in the number of g a l l s per developed bud are due p r i m a r i l y to changes in the proportion of buds attacked, a simpler mechanism consis tent with the effect of bud abort ion discussed in Chapter III i s poss ib le : the higher proportion of buds developed may lead to an elevated encounter rate with buds su i table for o v i p o s i t i o n . If knapweed plants do not change the proportion of buds aborted because of insect a t tack , the proportion of buds l e f t 1 58 undeveloped in the normal sequence of development may a lso be constant. Roze (1981) concluded that weather d id not regulate the proportion of undeveloped buds per p lant . S i m i l a r l y , Aker (1982) suggests that Yucca whipplei T o r r . responds to moisture changes by adjust ing inf lorescence s ize rather than the proportion of buds developed. LARVAL SURVIVAL AND DEVELOPMENT The greater m o r t a l i t y of U. a f f i n i s and U. quadr i fasc ia ta larvae in f e r t i l i z e d d i f fuse knapweed in 1980 may be due to the des truc t ive e f fec t of grasshopper feeding on the p lants ' vascular systems or to the effect of f e r t i l i z e r "burn". In e i ther case, the increased morta l i ty in treated plants would be due to a reduction in the quantity or qua l i ty of plant resources a v a i l a b l e to the l arvae . Two other explanations are poss ib le for the increased morta l i ty : heavier attack by natural enemies or increased i n t r a s p e c i f i c competition for space. Neither explanation i s l i k e l y . The g a l l f l i e s have no known paras i tes in North America (Myers and H a r r i s , 1980). I n t r a s p e c i f i c competit ion for space in treated plants would be lower than in c o n t r o l p l a n t s ; the number of g a l l s per developed bud was lower in treated plants for both species of g a l l f l y (Table 4 .5) . In contrast to the effect of~treat ing d i f fuse knapweed, U. a f f i n i s l a r v a l morta l i ty was lower in f e r t i l i z e d spotted knapweed. The di f ference between the two plant species suggests that spotted knapweed a l loca te s increased resources to ind iv idua l buds. This i s consistent with the increased number 1 59 of seeds per bud in f e r t i l i z e d spotted knapweed (Table 4 .8) . F e r t i l i z e d d i f fuse knapweed had fewer seeds per bud, despite fewer g a l l s per developed bud than control p lants . L a r v a l development rates of both insect species responded p o s i t i v e l y to the combination of f e r t i l i z e r and a d d i t i o n a l water supplied to spotted knapweed. U. a f f i n i s l a r v a l development was also faster in treated d i f fuse knapweed. (No d i f ference in U. quadr i fasc i a ta l a r v a l development was detected.) This suggests that the flow of nutrients to i n d i v i d u a l buds was s i g n i f i c a n t l y a l t e r e d by the treatments. The resu l t of the faster l a r v a l development would be to increase the s ize of the second generation r e l a t i v e to the f i r s t generation. These data suggest that the p h y s i o l o g i c a l condit ion of the plant and i t s ef fect on l a r v a l development may modify the process of diapause i n i t i a t i o n (cf . Brown et a l . , 1979). If greater resource a v a i l a b i l i t y during l a r v a l development is corre la t ed with a greater p r o b a b i l i t y of f ind ing sui table ov ipos i t i on s i t e s in the l a t t e r part of the season, th i s f l e x i b i l i t y in diapause i n i t i a t i o n may be se lected for (Tauber et a l . , 1986). EFFECT OF PLANT QUALITY ON POPULATION DYNAMICS The impact of a s h i f t in plant qua l i ty (e .g . nitrogen or water a v a i l a b i l i t y ) on the populat ion dynamics of insects depends on how the s h i f t a l t e r s the net outcome of a ser ies of processes. The f i r s t process i s the p lant ' s a l l o c a t i o n of a d d i t i o n a l resources. Bigger and more vigorous plants may not be better 160 food. Such plants may have improved defences against herbivore attack (Fraenkel , 1959; Rhoades, 1983). The second process is the insec t ' s detect ion of the q u a l i t a t i v e change in the plant (e .g . Van Emden, 1972). The t h i r d process i s the insect ' s exp lo i ta t ion of the add i t i ona l food resources . This process depends on the insec t ' s funct ional and numerical responses (Solomon, 1949) with the i r associated time lags . F i n a l l y , there are the processes underlying the insec t ' s populat ion dynamics, s u r v i v a l , development, reproduction and d i s p e r s a l , each of which may be a l t ered by plant q u a l i t y changes (e .g . McNei l l and Southwood, 1978; but see Auerbach and Strong, 1981). The nature of any change in the population dynamics r e l i e s on the change in the magnitude and timing of the preceeding processes . The condit ions under which an insect outbreak w i l l occur following an increase in plant q u a l i t y are probably rare in natural systems. In general , such increases are probably absorbed by a c o l l e c t i o n of herbivores , g iv ing the appearance of regulat ion of primary produc t iv i ty (Mattson and Addy, 1975). Cases where food q u a l i t y is implicated in insect outbreaks (e .g . Kimmins, 1971; White, 1976) seem to require a re servo i r of plant biomass on which one or more generations of the insect can sustain high net rates of reproduct ion. FERTILIZATION AS A MANAGEMENT TOOL It i s un l ike ly that f e r t i l i z a t i o n w i l l improve the a b i l i t y of forage plants to r e s i s t knapweed invas ion . In the absence of 161 graz ing , Popova (i960; c i t ed in Watson, 1972) found that horse manure addi t ions increased the percent cover by knapweed from 5 6 . 4 ± 2 . 3 % to 7 4 . 1 ± 4 . 4 % at the expenses of grasses and forbs . Myers and Berube (1983) f e r t i l i z e d a l i g h t l y grazed range with ammonium n i t r a t e and found no effect on knapweed or grass biomass, in part because c a t t l e on that range fed p r e f e r e n t i a l l y on treated p l o t s . F e r t i l i z a t i o n may prove useful through a synerg i s t i c effect with h e r b i c i d e . Sheley et JELL. (1984) discovered that a f a l l a p p l i c a t i o n of pic loram with f e r t i l i z e r produced the highest grass y i e l d s and the best contro l of spotted knapweed. The mechanism they proposed was increased competition from the grasses af ter the res idua l effect of the herbic ide had worn o f f . SUMMARY This Chapter describes the effect of f e r t i l i z a t i o n and watering treatments on knapweed, the response of the adult f l i e s to treated p l a n t s , and the consequences of f l y attack on the p l a n t s . Knapweed plants were s i g n i f i c a n t l y affected by both f e r t i l i z a t i o n and watering, though the d e t a i l s of the changes d i f f e r e d between years , between s i t e s , and between plant species . The most consistent response to the treatments was an increase in numbers of buds per p lant . The higher l e v e l of f e r t i l i z a t i o n and watering at Robertson's in 1980 over 1979 led to f e r t i l i z e r "burn"; the same increase at Chase gave a much greater response in 1980. 162 In general , adult f l i e s were observed on treated plants in d i rec t proport ion to the increase in the number of buds per p lant . The one exception was the reduced number of adults recorded on treated d i f fuse knapweed plants in 1980. The reduction appeared to be due to the increased density of grasshoppers on treated p lant s . Grasshoppers, sp i t t l ebugs , and cows responded p o s i t i v e l y to improved plant q u a l i t y . I examined changes to three factors which l i m i t g a l l f ly populat ions . The g a l l f l i e s are l i m i t e d by the density of ov ipos i t i on s i t e s since the experimental increases in developed buds per plant led to a proport iona l increase in g a l l s per p lant . In general , plants d id not s i g n i f i c a n t l y a l t e r the proportion of attacked buds that were aborted. In the two cases where the proport ion of buds developed changed, the numbers of g a l l s per developed bud sh i f t ed in the same d i r e c t i o n . Surviva l of larvae in f e r t i l i z e d d i f fuse knapweed plants was reduced r e l a t i v e to the c o n t r o l . S u r v i v a l of U. a f f i n i s larvae in treated spotted knapweed plants was improved over the contro l p lants . There was no evidence of poss ib le outbreaks by the g a l l f l i e s in response to the experimental treatments . F e r t i l i z a t i o n w i l l be of l i m i t e d use in the management of the knapweed problem. 1 63 V. POPULATION LIMITATION OF TWO INTRODUCED INSECTS: PROCESSES WITHIN AND BETWEEN YEARS The choice of s p a t i a l and temporal scale in studies of populat ion dynamics may dramatical ly a l t e r the observed behaviour of the system. Large f luctuat ions and "extinctions" at a l o c a l l e v e l may appear smoothly continuous when they are aggregated (e .g . Huffaker, 1958; Nicholson and Ba i l ey , 1935). This Chapter examines the ef fect of a change of temporal scale on populat ion l i m i t i n g processes for two introduced insec t s . The ga l l - forming f l i e s , Urophora a f f i n i s F r f l d . and U. q u a d r i f a s c i a t a (Meig.) (Diptera: T e p h r i t i d a e ) , lay eggs in the immature flower buds of d i f fuse knapweed, Centaurea d i f f u s a Lam., and spotted knapweed, C. maculosa Lam. (Asteraceae) . Previous Chapters focussed on processes l i m i t i n g the g a l l f l y populations which acted within a s ingle season. This Chapter evaluates the same processes as they act between one or more seasons. The conclusions from the rest of th i s thes is are contrasted with the resul ts of comparisons among years. Three years of f i e l d observations are combined with data c o l l e c t e d by e a r l i e r workers on th i s system extending back to the o r i g i n a l in troduct ion of the g a l l f l i e s (e .g. Berube, 1980; H a r r i s , 1980a,b; Myers and H a r r i s , 1980; Roze, 1981). Chapters I through IV concluded that two factors were important in l i m i t i n g the populations of g a l l f l i e s . The f i r s t , a v a i l a b i l i t y of ov ipos i t ion s i t e s , was affected by the resource status of the plant (Chapter IV) , and by non-random attack by 1 64 the f l i e s in space and time (Chapters I , I I , and I I I ) . Both of these components of o v i p o s i t i o n s i t e a v a i l a b i l i t y should be evident in the comparisons among years . Changes in p r e c i p i t a t i o n from year to year should be corre la ted with changes in the number of buds per p lant . The h i s t o r i c a l d i s t r i b u t i o n of g a l l s should r e f l e c t non-random f ly attack. In a d d i t i o n , the density of mature plants was constant for the analys i s of processes within a s ing le season; th i s component of ov ipos i t i on s i t e a v a i l a b i l i t y w i l l vary among years. G a l l formation decreases seed production by knapweed p lants . If seed production is re la ted to mature plant dens i ty , the g a l l f l i e s w i l l l i m i t the a v a i l a b i l i t y of o v i p o s i t i o n s i t es two or more years l a t e r . The second factor l i m i t i n g the g a l l f l y populations i s the abortion of buds that are heavi ly attacked by insects (Chapters I , I I , and I I I ) . While the propensity for a plant to abort buds may vary from year to year, there should be a c o r r e l a t i o n between the proportion of buds aborted and the number of ga l l s per developed bud. Bud abort ion was one way that U. a f f i n i s reduced the a v a i l a b i l i t y of su i tab le ov ipos i t i on .sites for U. quadr i fasc iata (Chapter I I ) . Among year comparisons l i k e these are not contro l l ed experiments and thus cannot be conclus ive demonstrations of a p a r t i c u l a r process. They can, however, act as supporting evidence and do give an apprec ia t ion of the r e l a t i v e importance of d i f f e r e n t processes as they in teract in the undisturbed system. 1 65 This Chapter (1) evaluates the effect on g a l l f l y populations of dif ferences among years in o v i p o s i t i o n s i t e a v a i l a b i l i t y , in p a r t i c u l a r due to p r e c i p i t a t i o n , non-random at tack , and changes in seed product ion, and (2) discusses the l i m i t a t i o n of the number of g a l l s per developed bud, e s p e c i a l l y as a resu l t of bud abortion and the in terac t ion between the two g a l l f l y species . 1 66 MATERIALS AND METHODS WEATHER To test whether the resource status of plants is changed by p r e c i p i t a t i o n , weather data were obtained from Environment Canada. The r a i n f a l l measurements at the Kamloops Airpor t weather s tat ion used in th i s Chapter were consistent with the patterns recorded at Chase, B . C . , at the opposite end of the South Thompson River v a l l e y , and a lso agreed with the days r a i n f a l l was recorded at the study s i t e s . PLANT COLLECTIONS IN 1979 S imi lar experimental p lots were es tabl i shed at each of the three study s i tes in 1979 to examine the ef fect of changes in plant q u a l i t y on the g a l l f l y population dynamics (Chapter I V ) . The p lots described here were the contro l p lots for these experiments. At each s i t e , three one m2 p lo t s were located in a s t ra ight l i n e , separated by 3 m in an area with a v i s u a l l y uniform density of knapweed. A t o t a l of twenty plants were followed at each s i t e , f ive each from the two end p lo t s and ten from the centra l p l o t . Each plant was se lected as the nearest bo l t ing plant to an a r b i t r a r i l y chosen point within the p l o t . Five of the points in each plot were taken as the center of the plot and the four points b i sec t ing the s t ra ight l i n e s between the center and the four corners of the p l o t . In the c e n t r a l p l o t , four of the a d d i t i o n a l points were taken as the points b i sec t ing the s tra ight l ine s between the se lected plants near adjacent p lot d iagonals . The tenth plant was chosen by b l i n d l y f l i p p i n g a 1 67 p e n c i l into the p l o t . If the nearest plant to the p e n c i l t i p was not previously selected and was within the plot boundaries, i t was chosen. Selected plants were rare ly nearest neighbours. P lots at a l l s i tes were establ ished on May 28, 1979. Di f fuse knapweed plants were c o l l e c t e d on August 22 at Ned's Creek and on August 25 and 26 at Robertson's. Spotted knapweed at Chase was co l l ec ted at three day in terva l s from August 8 to August 20. Four plants were c o l l e c t e d on each day. PLANT COLLECTIONS IN 1980 In 1980, s imi lar experimental p lo t s were establ ished on June 1 at Robertson's and on June 2 at Chase. At each s i t e , a rectangular g r i d of f i f t y (5x10) points was placed over a 3 m x 6.5 m port ion of the f i e l d with a v i s u a l l y uniform density of knapweed. The plant which had begun to bolt nearest each point on the g r i d was staked. The points on the g r i d were far enough apart so that staked plants were r a r e l y nearest neighbours. A l l d i f fuse knapweed plants were c o l l e c t e d on August 23, 1980. Spotted knapweed plants were c o l l e c t e d just before the f i r s t seed head shed i t s d r i e d f l o r e t s . This c o l l e c t i o n extended over the period August 1 to August 24 as seed heads matured. Ear ly in the summer of 1980, the owner sprayed the Ned's Creek s i t e with the herbic ide picloram (4-amino-3,5,6-t r i c h l o r o p i c o l i n i c a c i d ) . Forty f ive plants which survived the herbic ide treatment were c o l l e c t e d from Ned's Creek on September 12. 1 68 PLANT COLLECTIONS IN 1981 On October 9, 1981, spotted knapweed plants were c o l l e c t e d at Chase (N=50) and di f fuse knapweed plants were c o l l e c t e d at Robertson's (N=20). These plants were the nearest p lants to randomly selected points . The late c o l l e c t i o n date means that some seed may have been los t from spotted knapweed heads. PLANT DISSECTIONS Plants were c o l l e c t e d i n d i v i d u a l l y by c l i p p i n g them off at ground l e v e l . They were then stored in folded and stapled paper bags at room temperature u n t i l d i s s e c t i o n . When they were d i s sec ted , the height of each plant was measured and the s i z e , developmental s ta tus , and locat ion of each bud were recorded. Buds large enough to contain e i ther g a l l s or seeds were i n d i v i d u a l l y d i s sec ted . For these, the number of seeds and contents of any g a l l s present were noted. U. a f f i n i s g a l l s are hard and woody and those of U. quadri fasc iata are th in and papery. 169 RESULTS AMONG YEAR DIFFERENCES NED'S CREEK The number of buds on d i f fuse knapweed plants at Ned's Creek increased from 1979 to 1980 (Table 5.1) . The proportion of buds that matured increased from Table 5.1 - Diffuse knapweed c h a r a c t e r i s t i c s and Urophora attack l e v e l s , Ned's Creek 1979- 1980 Character * 1 979 1 980 Number of Plants 20 45 Buds/Plant 1 6 . 2 ± 1 . 9 5 0 . 6 ± 5 . 8 Chewed/Plant 0 . 5 5 ± 0 . 2 0 1 . 0 8 ± 0 . 30 Chewed/Buds 0 . 0 3 ± 0 . 0 1 0 . 0 2 ± 0 . 0 1 Dev. /Plant 5 . 1 ± 0 .8 2 6 . 4 ± 2 . 5 Dev./Buds 0 . 3 3 ± 0 . 0 3 0.5710.02 UA/Plant 3 . 2 ± 0 . 8 1 5 . 9 ± 3 . 4 UA/Dev. 0 . 6 2 ± 0 . 0 8 0 . 6 1 ± 0 . 0 3 UA/Attacked 1 . 5 0 ± 0 . 10 1 . 8 4 ± 0 . 0 7 Prop Att UA 0 . 4 1 ± 0 . 0 5 0 . 3 3 ± 0 . 0 2 . UQ/Plant 0 . 7 ± 0 . 3 3 . 3 ± 1 . 5 UQ/Dev. 0 . 1 3 ± 0 . 0 6 0 . 1 3 ± 0 . 0 1 UQ/Attacked 1 . 8 6 ± 0 . 4 6 1 . 4 7 ± 0 . 0 8 Prop Att UQ 0 . 0 7 ± 0 . 0 3 0 . 0 9 ± 0 . 0 1 Seeds/Plant 25±4 161+26 Seeds/Dev. 4 . 8 8 ± 0 . 4 1 6 . 2 5 ± 0 . 1 3 Seeds/Produc ing 7 . 2 2 ± 0 . 3 5 7 . 2 3 ± 0 . 1 3 Prop. Aborted 0 . 0 7 ± 0 . 0 2 0 . 2 1 ± 0 . 0 2 * Deta i led descr ipt ions of these characters are given in Table 2.5. 1979 to 1980. The number of g a l l s per developed bud was not s i g n i f i c a n t l y d i f f eren t between years for both species of f l y , however the proportion of buds aborted increased t h r e e f o l d . 1 70 More of the developed buds produced seeds in 1980 than in 1979; the number of seeds in buds which contained seeds d id not change s i g n i f i c a n t l y while the number of seeds in a l l developed buds increased. This was not consistent with the r e l a t i v e l y constant numbers of g a l l s per developed bud, however i t i s not known to what extent the herb ic ide and the resu l t ing dras t i c reduction in plant density a l t e r e d f l y attack or seed product ion. ROBERTSON'S The number of buds per d i f fuse knapweed plant at Robertson's increased between 1979 and 1980 and between 1980 and 1981 (Table 5 .2) . The proport ion of buds that matured -dropped from 1979 to 1980 (t=2.10, df=28, p=0.045) and then d id not change s i g n i f i c a n t l y between 1980 and 1981. Attack by U. a f f i n i s increased over the three years, both in terms of g a l l s per attacked bud and the proportion of buds attacked. The proport ion of buds aborted complemented the proportion of buds developed; i t increased between 1979 and 1980, but d id not change s i g n i f i c a n t l y between 1980 and 1981. Attack by U. quadri fasc i a ta increased from 1979 to 1980, but dec l ined from 1980 to 1981. The number of seeds in a l l developed buds was negatively re la ted to U. a f f i n i s attack, yet i t was only between 1980 and 1981 that the number of seeds in buds containing seeds decl ined s i g n i f i c a n t l y . CHASE The height of spotted knapweed plants at Chase increased from 1979 to 1981, but the number of buds only increased between 1980 and 1981 (Table 5.3) . The proportion of buds that 171 Table 5.2 - Dif fuse knapweed c h a r a c t e r i s t i c s and Urophora attack l e v e l s , Robertson's 1979-1981 Character * 1979 1980 1981 Number of Plants 20 50 20 Height (cm) 1 8±1 30±1 39±2 Buds/Plant 25±5 72±6 132±20 Chewed/Plant 1 . 8 ± 0 .7 6 . 6 ± 1 . 2 1 3 . 3 ± 2 . 6 Chewed/Buds 0 . 0 6 ± 0 .02 0 . 0 9 ± 0 . 0 1 0 . 0 9 ± 0 . 0 1 Dev. /Plant 1 1 . 1 ± 1 .5 3 0 . 7 ± 2 . 4 5 5 . 6 ± 7 . 2 Dev./Buds 0 . 5 2 ± 0 .04 0 . 4 4 ± 0 . 0 2 0 . 4 5 ± 0 . 0 3 UA/Plant 6 .4±1 .0 3 6 . 9 ± 4 . 0 116+17 UA/Dev. 0 . 5 7 ± 0 .06 1 . 2 0 ± 0 . 0 4 2 . 1 0 ± 0 . 0 5 UA/Attacked 1 . 6 5 ± 0 .09 2. 1 8 ± 0 . 0 5 2 . 7 3 ± 0 . 0 5 Prop Att UA 0 . 3 5 ± 0 .03 0 . 5 5 ± 0 . 0 1 0 . 7 7 ± 0 . 0 1 UQ/Plant 1 . 4 ± 0 .5 8 . 3 ± 1 . 3 1 1 . 6 ± 2 . 7 UQ/Dev. 0 . 1 2 ± 0 .03 0 . 2 7 ± 0 . 0 2 0 . 2 1 ± 0 . 0 2 UQ/Attacked 1 . 5 0 ± 0 .17 1 . 8 4 ± 0 . 0 7 1 . 6 3 ± 0 . 0 8 Prop Att UQ 0 . 0 8 ± 0 .02 0. 1 5 ± 0 . 0 1 0 . 1 3 ± 0 . 0 1 Seeds/Plant 51 ±9 106±11 93±22 Seeds/Dev. 4 . 5 5 ± 0 .27 3 . 4 7 ± 0 . 10 1 . 6 8 ± 0 . 0 8 Seeds/Produc ing 6 . 2 0 ± 0 .26 6 . 0 3 ± 0 . 1 2 4 . 2 6 ± 0 . 1 3 Prop. Aborted 0 . 0 7 ± 0 .01 0 . 2 0 ± 0 . 0 1 0 . 2 0 ± 0 . 0 2 * Detai led descr ipt ions of these characters are given in Table 2.5. developed did not change s i g n i f i c a n t l y . Attack by U. a f f i n i s did not change s i g n i f i c a n t l y between 1979 and 1980, except for an increase in the proportion of developed buds at tacked. From 1980 to 1981, a l l measures of U. a f f i n i s attack increased dramat ica l ly ; almost a l l of the developed buds contained g a l l s . The proportion of buds aborted d id not behave as pred ic ted; the proportion dec l ined from 1979 to 1980, despite nons igni f icant changes in U. a f f i n i s g a l l d e n s i t i e s . The proport ion of buds aborted in 1981 was approximately the same as in 1979 even 1 72 Table 5.3 - Spotted knapweed c h a r a c t e r i s t i c s and Urophora attack l e v e l s , Chase 1979-1981 Character * 1 979 1980 1981 Number Plants 20 50 50 Height (cm) 2 1 . 4 ± 1 .2 26.4±1. 1 3 2 . 5 ± 0 . 9 Buds/Plant 6 . 1 0 ± 0 .73 5 . 0 2 ± 0 . 30 8 . 2 2 ± 0 . 58 Chewed/Plant 0 . 2 5 ± 0 .10 0 . 5 0 ± 0 . 13 1 . 6 0 ± 0 . 26 Chewed/Buds 0 . 0 3 ± 0 .01 0 . 0 8 ± 0 . 02 0 . 1 7 ± 0 . 02 Dev . /P lant 3 . 1 5 ± 0 .36 2 . 6 4 ± 0 . 18 4 . 2 6 ± 0 . 27 Dev./Buds 0 . 5 4 ± 0 .04 0 . 5 8 ± 0 . 03 0 . 5 5 ± 0 . 03 UA/Plant 3 . 0 ± 0 .7 3 . 2 ± 0 . 5 1 6 . 7 ± 1 . 5 UA/Dev. 1 . 0 0 ± 0 .18 1 . 2 2 ± 0 . 1 1 3 . 9 6 ± 0 . 1 7 UA/Attacked 2 . 2 7 ± 0 .23 1 . 8 7 ± 0 . 1 3 4 . 2 6 ± 0 . 1 6 Prop Att UA 0 . 4 4 ± 0 .07 0 . 6 5 ± 0 . 04 0 . 9 3 ± 0 . 02 UQ/Plant 0 . 1 5 ± 0 . 1 1 0 . 8 4 ± 0 . 26 1 . 1 0 ± 0 . 30 UQ/Dev. 0 . 0 5 ± 0 .04 0 . 3 2 ± 0 . 08 0 . 2 6 ± 0 . 06 UQ/Attacked 1 . 5 0 ± 0 .50 2 . 1 0 ± 0 . 34 1 . 9 0 ± 0 . 26 Prop Att UQ 0 . 0 3 ± 0 .02 0 . 1 5 ± 0 . 03 0 . 1 4 ± 0 . 02 Seeds/Plant 2 6 . 6 ± 4 .6 3 9 . 9 ± 3 . 5 1 9 . 7 ± 2 . 6 Seeds/Dev. 9 . 0 ± 0 .4 1 5 . 4 ± 0 . 8 4 . 7 ± 0 . 3 Seeds/Produc ing 9 . 7 ± 0 .3 1 7 . 2 ± 0 . 7 5 . 7 ± 0 . 3 Prop. Aborted 0 . 0 9 ± 0 .03 0 . 0 3 ± 0 . 01 0 . 0 8 ± 0 . 02 * Deta i l ed d e s c r i p t i o n s of these characters are given in Table 2.5. though attack by both g a l l f l i e s was much higher in 19 Attack by U. quadr i fasc iata increased sharply from 1979 to 1980, p r i m a r i l y through an increased number of buds attacked, but was not s i g n i f i c a n t l y changed between 1980 and 1981. The number of seeds per developed bud also increased between 1979 and 1980, but in 1981 dropped to 30% of i t s value in 1980. The contrast in seeds per developed bud between 1980 and 1981 may be p a r t l y due to a l a t e r c o l l e c t i o n date in 1981. 1 73 EFFECT OF RAINFALL Based on a simple ex trapo la t ion of numbers of buds at Robertson's in 1979 and 1980 and the observed to ta l r a i n f a l l for the three months June to August, the pred ic t ion of t o t a l buds per plant for 1981 is 76.2 buds per p l a n t . This is s i g n i f i c a n t l y lower than the observed value of 132±20 for 1981 (Table 5 .2) . The d i f ference between the observed and predicted values indicates that t o t a l p r e c i p i t a t i o n over the three months is not an adequate pred ic tor of t o t a l number of buds produced per p lant . If only p r e c i p i t a t i o n between the time of f i r s t bud and f i r s t flower is considered ( i . e . June 10 to Ju ly 27), a d i rec t r e la t i onsh ip is observed between the number of buds per plant and the r a i n f a l l for those three years (Figure 5 .1) . Using log transformed data, the r e l a t i o n s h i p i s : ln(BUDS+l) = 0 . 0 2 5 6 ( ± 0 . 0 0 3 2 ) R A I N (mm) + 2 . 8 3 ( ± 0 . 1 3 ) , F=66.0, df=1,88, r=0.65, p<0.00l. The number of buds per plant at Ned's Creek a lso increased from 1979 to 1980 with increased p r e c i p i t a t i o n (Figure 5.1) . Other poss ib le sources of var ia t i on in the number of buds per plant include the s ize of rosette reserves and the density of b o l t i n g p l a n t s . Bud production by spotted knapweed plants appeared not to have the same response to changes in p r e c i p i t a t i o n ; the number of buds per plant dropped between 1979 and 1980 despite increased r a i n f a l l (Table 5 .3) . In contras t , plant height changes were consistent with changes in p r e c i p i t a t i o n . 1 74 Figure 5.1. E f f ec t of t o t a l r a i n f a l l during the per iod June 10 - Ju ly 27 on the f i n a l number of buds per d i f fuse knapweed p l a n t . The predicted value for Robertson's for 1981 i s based on s tra ight l ine extrapolat ion from the values for 1979 and 1980. The regression l i n e shown is based on the observed values for Robertson's 1979 to 1981. Deta i l s of the regression are given in the text . 1 76 DISCUSSION BUD DENSITY In the short term, the density of buds ava i l ab l e to the g a l l f l i e s i s a l tered by the resource status of the plants and by the timing of bud i n i t i a t i o n . The c o r r e l a t i o n of r a i n f a l l with buds per plant (Figure 5.1) supports the conclusion of Chapter IV: bud numbers are d i r e c t l y af fected by the water a v a i l a b l e to the bo l t ing p lant . This implies that bud dens i t i e s w i l l vary among years depending on the p r e c i p i t a t i o n in those years . Schirman's (1981 ) data a lso support the same conc lus ion . A subs tant ia l f rac t ion of su i table buds are unattacked in any given year because of the timing of f l y attack r e l a t i v e to bud i n i t i a t i o n (Chapter I I ) , because of low f ly dens i t i e s (Chapters II and I I I ) , and because of the cumulative ef fect of aborted buds on g a l l production (Chapter I I I ) . Appendix VA describes an analys i s of the g a l l d i s t r i b u t i o n s to estimate the proport ion of buds that were a c t u a l l y ava i l ab l e to the o v i p o s i t i n g g a l l f l i e s . The proportion of developed buds a v a i l a b l e at Ned's Creek in 1980 was estimated to be 0.72. At Chase one year l a t e r , the estimated proportion was 0.96. The smaller proport ion on di f fuse knapweed may r e f l e c t the more extended per iod of bud i n i t i a t i o n in that species . In the long term, the density of su i table ov ipos i t ion s i t e s depends on the density of mature plants and hence on seed production and the surv iva l of immature p l a n t s . The g a l l f l i e s have a major impact on seed production within and among plants 1 77 (Chapter I ) . The data from the among year comparisons at Robertson's and Chase indicate that U. a f f i n i s g a l l density i s negatively re lated to the number of seeds per developed bud. It is not poss ib le to evaluate the reduction in seed output d i r e c t l y using a l l the h i s t o r i c a l data since only bud dens i t ies were recorded from 1973 to 1978; seed production was not measured. At the two s i t e s where the best time ser ies are ava i lab le (Ned's Creek and Chase), the changes in g a l l and bud density per unit area are s imi lar (Figures 5.2, 5 .3) . The f l i e s increased r a p i d l y , peaked, and d e c l i n e d . Bud dens i t i e s decl ined fol lowing the peak in f l y d e n s i t i e s . Bud dens i t i e s have remained low; dens i t i e s of buds in 1985 were v i r t u a l l y i d e n t i c a l to dens i t i e s in 1979 (Myers, pers . comm.). The reduction in seed production w i l l be even more extreme than the drop in dens i t i e s of buds because the number of seeds per bud w i l l a lso dec l ine due to g a l l formation within developed buds. Assuming that unattacked d i f fuse knapweed produces 12.5 seeds per developed bud and that unattacked spotted knapweed produces 26.6 seeds per developed bud (Watson, 1972), the estimated reduction in numbers of seeds produced per plant as a resu l t of f l y attack was 89% for d i f fuse knapweed at Robertson's in 1981 and 84% for spotted knapweed at Chase in the same year. Many questions remain regarding annual changes in bud density inc lud ing: How important i s seed input r e l a t i v e to competition among plants? How important are plant s ize and density p r i o r to b o l t i n g r e l a t i v e to s o i l moisture leve ls during 1 78 Figure 5.2. Dens i t ies of d i f fuse knapweed buds and t o t a l Urophora g a l l s per m2 at Ned's Creek 1972-1979. Density data are from Harr i s (1980a). Also shown is the r e l a t i v e density of g a l l s per developed bud (1972-1980). Relat ive density data are from H a r r i s (unpublished data) and data in th i s t h e s i s . V e r t i c a l l ines give ± one standard e r r o r . No standard errors were given for absolute dens i t i e s of g a l l s 1972-1974 in the o r i g i n a l reference. Standard errors for the r e l a t i v e dens i t i e s of g a l l s are c lose to the s ize of the symbol for 1972-1974. 180 Figure 5.3. Dens i t ies of spotted knapweed buds and t o t a l Urophora g a l l s per m2 at Chase 1971-1979. Density data are from H a r r i s (1980a). Also shown is the r e l a t i v e densi ty of g a l l s per developed bud (1971-1981). Re lat ive density data are from Harr i s (unpublished data) and data in th i s thes i s . V e r t i c a l l i n e s give ± one standard e r r o r . No standard errors were given for absolute dens i t i e s of g a l l s 1971-1974 in the o r i g i n a l reference . Density (per m2) x 100 Galls / developed bud T 8 T 182 bud i n i t i a t i o n ? Is there a d i f ference in the density of g a l l s and seeds produced from small densely-packed plants compared to larger and more widely-spaced plants? A l l of these questions have immediate management impl i ca t ions . Roze (1981) has provided prel iminary evidence on some of these issues , but her work must be extended and re f ined . EFFECT OF BUD DENSITY ON GALL DENSITY The reproductive p o t e n t i a l of the Urophora adults i s s u f f i c i e n t to take advantage of large increases in bud dens i t i e s . At Ned's Creek from 1979 to 1980, there was a threefo ld increase in the number of buds per plant and the number of ga l l s per developed bud remained constant . From 1979 to 1981 at Robertson's, there was a f i v e f o l d increase in the number of buds per plant and the number of U. a f f i n i s g a l l s per developed bud increased over 260%. The f e r t i l i z a t i o n and watering treatment at Chase in 1980 increased the number of buds per plant almost fourfo ld and the density of U. a f f i n i s g a l l s per bud dropped only s l i g h t l y (Chapter IV) , though t h i s observation may include an effect of f l i e s moving onto treated p l a n t s . Despite the ir large reproductive p o t e n t i a l , the g a l l f l i e s are l i m i t e d by the density of su i table o v i p o s i t i o n s i t e s (Chapters I and IV) . The effect of th i s r e l a t i o n s h i p over time is c l e a r l y seen in Figures 5.2 and 5.3; as bud dens i t i e s dec l ined at Ned's Creek and Chase af ter 1976, g a l l dens i t i e s per unit area a lso dropped. Thus the f luctuat ions in insect density w i l l be a function of resource a v a i l a b i l i t y . Dempster and 1 83 P o l l a r d (1981) argue that t h i s i s a widespread causal l ink for insec t s . GALL DISTRIBUTIONS Table 5.4 demonstrates that the number of g a l l s per developed bud d id not increase monotonically at the two s i t e s where the best time ser ies are a v a i l a b l e . This l i m i t a t i o n occurred while a s i g n i f i c a n t f rac t ion of su i table buds d id not contain any g a l l s (Appendix VA). Obviously other factors aside from the absolute number of suitable buds are l i m i t i n g g a l l f l y populat ions . Among s i t e d i f ferences a f fec t the mean number of g a l l s per developed bud. The number of U. a f f i n i s g a l l s per developed bud at Robertson's in 1981 was higher than in any year at Ned's Creek (Table 5.4) . The d i f f erences between these s i t e s are presumably a function of the encounter rate with sui table buds ( in turn a function of buds per plant) and of the "carrying capacity" of i n d i v i d u a l buds (Chapter IV) . Story and Nowierski (1984) followed the increase of U. a f f i n i s at f ive spotted knapweed s i t e s in Montana over four years . At the conclusion of t h e i r study, the percent of developed buds attacked ranged from 63-99% over the f ive s i t e s and the mean number of g a l l s per developed bud ranged from 2.1 to 9.3. Again, among s i t e d i f ferences s i g n i f i c a n t l y affected the outcome of f l y a t tack . The two plant species d i f f e r in the number of g a l l s each bud can support (Myers and H a r r i s , 1980). The higher "carrying capacity" of spotted knapweed buds is re f l ec ted in the higher 184 Table 5.4 - Urophora g a l l s per developed bud at the three study s i t e s , 1973-1981 * Year Ned's Creek Robertson's Chase Urophora af f i n i s 1 973 0 . 1 4 ± 0 . 0 1 ** 0.4310.01 1 974 0 . 2 3 ± 0 . 0 1 0.3010.01 1975 0 . 7 9 ± 0 . 0 4 1.3410.05 1 976 1 . 2 4 ± 0 . 0 3 3.1810.09 1 977 0 . 8 4 ± 0 . 0 4 0 . 1 0 ± 0 . 01 4.8910. 14 1 978 1 . 1 0 ± 0 . 0 4 2.5210.05 1 979 0 . 6 2 ± 0 . 0 8 0.5710. 06 1 .83 + 0. 10 1980 *** 0 . 6 1 ± 0 . 0 3 1.2010. 04 1.2210.12 1 981 2.10+0. 05 3.96 + 0. 17 Urophora quadr i fasc iata 1 973 0 . 0 6 ± 0 . 0 1 1 974 0 . 2 9 ± 0 . 0 1 1 975 0 . 7 8 ± 0 . 0 5 1976 0.4210.03 0.1710.03 1 977 0 . 4 0 ± 0 . 0 3 0.2410. 01 0.1110.02 1 978 0 . 0 5 ± 0 . 0 1 0.8110.03 1 979 0 . 1 3 ± 0 . 0 2 0.1210. 03 0.1210.03 1 980 0 . 1 3 ± 0 . 0 1 0.2710. 02 0.3210.08 1 981 0.2110. 02 0.2610.06 Both Urophora species combined 1 973 0 . 2 0 ± 0 . 0 1 0.4310.01 1 974 0 . 5 2 ± 0 . 0 2 0.3010.01 1 975 1 . 5 8 ± 0 . 0 6 1.3410.05 1 976 1 . 6 5 ± 0 . 0 4 3.3510.09 1 977 1 . 2 4 ± 0 . 0 4 0.3410. 02 4.9910.13 1 978 1 . 1 5 ± 0 . 0 4 3.3310.06 1 979 0 . 7 5 ± 0 . 0 6 0.6910. 07 1.9510.10 1 980 0.7410.04 1.47+0. 04 1.55 + 0. 14 1 981 2.3010. 05 4.2210.17 * Data for 1973-1978 from Harr i s (unpublished data) . ** MeaniS.E. *** Ned's Creek was sprayed with herbic ide in th i s year. average number of g a l l s per bud observed at the populat ion peak (Table 5 .4) . Harr i s (1980a) suggests that the d i f ference between the two knapweeds is due to a d i f ference in the area of the i r flower receptacles , the s i t e s of g a l l formation. 185 The two contrasts just discussed explain part of the observed range in the mean number of g a l l s per developed bud, but they do not account for e i ther the l i m i t a t i o n of g a l l s per developed bud or the dec l ines fol lowing peak g a l l s per developed bud observed at Ned's Creek and Chase. The temporal patterns indicate that the numbers of g a l l s per developed bud were prevented from cont inuing the increase observed during establishment of the g a l l f l i e s and that there are factors act ing over a longer time span to reduce the g a l l s per developed bud. Chapters I , I I , and III showed that the d i s t r i b u t i o n of attack by the g a l l f l i e s was not randomly d i s t r i b u t e d in space or time. One of the outcomes of non-random attack, bud abort ion , may have prevented the increase in g a l l s per developed bud during 1975 to 1978. While Roze's (1981) data on "superparasit ized" buds are not d i r e c t l y comparable to the data on aborted buds presented in th i s thes i s , her data can probably be used to indicate the q u a l i t a t i v e changes in bud abortion over time. She reported that the proportion of "superparasit ized" buds at Ned's Creek was c lose to 0% in 1975 and then jumped to a value of about 10% in 1976 which then d id not change not iceably from 1976 to 1978. At Chase, the proportion of "superparasit ized" buds tracked the numbers of g a l l s per developed bud from 1975 to 1978, peaking at about 10% of the t o t a l number of buds in 1977. Her data are consistent with the hypothesis that bud abort ion prevented an increase in the number of g a l l s per developed bud. 186 The pattern of the proport ion of buds aborted from 1979 to 1981 at Robertson's (Table 5.2) and Chase (Table 5.3) may be explained by a dec l in ing propensity to abort buds over the three years combined with the observed changes in f l y a t tack . A dec l in ing propensity to abort buds is consistent with the improved plant q u a l i t y (measured by plant height and number of buds) which in turn was corre la ted with the heavier p r e c i p i t a t i o n from 1979 to 1981. Chapter IV concluded that the propensity to abort buds was not a l t ered by changes in the resource status of the p l a n t s . The magnitude of the d i f ferences in p r e c i p i t a t i o n between years r e l a t i v e to the treatments in Chapter IV may account for the contrast ing r e s u l t s . The high level ' of watering at Ned's Creek in 1979 added 6.9 mm to the natural p r e c i p i t a t i o n compared to a d i f ference of over 60 mm between 1979 and 1981 (Figure 5 .1) . Many factors could act over several years to cause the decl ine in the number of g a l l s per developed bud, however the only one for which there i s any evidence is the drop in bud density (Figures 5.2 and 5.3) . If the reduced bud densi ty led to a reduced encounter rate with su i table buds, fewer g a l l s per bud would r e s u l t . This hypothesis pred ic t s that large changes in the density of sui table buds w i l l a f fect the number of g a l l s per developed bud. The evidence from Chapter IV does not support th i s hypothesis, however the density of unsuitable ( i . e . aborted or undeveloped) buds was a lso increased in these experiments. The two exceptions to the general pat tern , the data from Robertson's 187 in 1979 and 1980, are consistent with the hypothesis that the c r i t i c a l factor i s the r a t i o between the numbers of su i table and unsuitable buds. The context of ov ipos i t ion and feeding s i t es may have a strong inf luence on the a b i l i t y of insects to locate them (reviewed by K a i r e v a , 1983). A comparison of natural and experimental changes in plant density (MS in prep.) w i l l help resolve th i s i ssue . INTERACTION BETWEEN GALL FLY SPECIES In the among year comparisons, the number of U. a f f i n i s and U. quadr i fasc ia ta g a l l s per plant changed together; both are corre la ted with the number of buds per p l a n t . S imi lar corre la t ions were observed in Chapter I for adult f l i e s and in Chapter IV for g a l l s in f e r t i l i z e d and watered p l a n t s . The d i s t r i b u t i o n of g a l l s of the two species within buds shows a qui te d i f f e r e n t pat tern . Consistent and s i g n i f i c a n t d i f ferences in the number of g a l l s per developed bud are evident between the two f l y species (Table 5.4). In every s i t e -year combination except 1974 at Ned's Creek and 1977 at Robertson's , U. quadr i fasc ia ta had fewer g a l l s per developed bud than U. a f f i n i s . Berube (1980) argues that U. a f f i n i s has a negative effect on the number of U. quadr i fasc iata g a l l s per bud. The d i s t r i b u t i o n of g a l l s may r e f l e c t a negative in terac t ion between the Urophora species in two ways: (1) the numbers of g a l l s per developed bud with the other species may be lower than without, or (2) the proport ion of buds containing g a l l s of both species 188 may be smaller than that predicted from independent attack by the two species . Myers and Harr i s (1980) have addressed the f i r s t c r i t e r i o n and demonstrated that the numbers of g a l l s per developed bud do behave as i f there i s a negative i n t e r a c t i o n . The second c r i t e r i o n was tested by comparing the number of unattacked buds predicted by independent attack with the observed zero c l a s s . In sixteen out of seventeen d i s t r i b u t i o n s from the three s i t e s , the number of buds pred ic ted to be unattacked by e i ther species was greater than the observed zero c la s s , the d i r e c t i o n expected on the basis of a negative in teract ion (signs tes t , p=0.00l). For ten of these d i s t r i b u t i o n s , the d i f ference was s i g n i f i c a n t (x 2 t e s t , a=0.05). For the one exceptional d i s t r i b u t i o n (Chase in 1978) for which the comparison suggests a pos i t ive i n t e r a c t i o n , the d i f ference was s i g n i f i c a n t (x 2 t e s t , a=0.05). If the zero c lasses ca lcu lated from the truncated negative binomial (Appendix VA) are compared with the zero c lasses predic ted by independent attack, seventeen out of seventeen d i s t r i b u t i o n s deviate in the d i r e c t i o n of a negative in terac t ion (signs t e s t , p<0.00l). The presence of the two f ly species are not independent of one another. The ef fect i s in the d i r e c t i o n predic ted i f a negative interact ion were occurr ing . There are , of course, several other possible explanations for th i s lack of independence besides competit ion. In a s imi lar a n a l y s i s , McEvoy (1984) concluded that U. a f f i n i s and U. quadr i fasc ia ta g a l l s were independently and randomly d i s t r i b u t e d in a spotted knapweed populat ion in Oregon. 189 His a b i l i t y to detect a non-random pattern was l i m i t e d , however, because in the sample he co l l ec t ed only two buds contained two U. a f f i n i s g a l l s and only 11 out of 869 buds contained two U. quadr i fasc i a ta g a l l s . No buds contained more than two g a l l s of e i ther species . There i s evidence for a negative influence of U. a f f i n i s on U. q u a d r i f a s c i a t a from the annual changes in the d i s t r i b u t i o n s of g a l l s . The proportion of U. aff i n i s g a l l s in the t o t a l number of g a l l s increased over time at Ned's Creek (Spearman r=0 .7 l , df=5, 0.05<p<0.1) and at Robertson's (Figure 5.4) . Combined with the evidence from Chapter I I , Berube's (1980) c la im for a negative in terac t ion is supported. In the face of th i s de leter ious influence from U. a f f i n i s , how does U. quadr i fasc ia ta pers i s t? Figure 5.4 shows that U. quadr i fasc i a ta const i tutes approximately 5-20% of the t o t a l g a l l s formed at the release s i t e s , even af ter the two species have been i n t e r a c t i n g for several years. B irch (1979) gives two reasons why competit ive exclusion might not occur: (1) species dens i t i e s are kept below the l e v e l necessary for competitive e f f ec t s , or (2) refuges in time or space. The dens i t i e s at the release s i t e s are high enough for competitive e f fects to be observed (Chapter I I ) , though the competition is not symmetrical (Lawton and H a s s e l l , 1981). The di f ference in bud s ize preferences (Berube and H a r r i s , 1978) w i l l only lead to a temporal refuge at the beginning of the season i f buds are large enough for U. quadr i fa sc ia ta ov ipos i t ion pr ior to heavy attack by U. a f f i n i s . There i s a set of unattacked or only l i g h t l y 1 90 Figure 5.4. Changes in the proport ion of U. a f f i n i s g a l l s of a l l Urophora g a l l s over time at the three study s i t e s . 1973 1974 1975 1976 1977 1978 1979 1980 Year 1981 1 92 attacked buds that were i n i t i a t e d af ter the f i r s t generation that could act as a refuge la te in the summer for the r e l a t i v e l y larger second generation of U. q u a d r i f a s c i a t a (Appendix U A ; Roze, 1981). Both of these poss ib le refuges are contingent on the year to year v a r i a t i o n in the numbers of g a l l f l i e s and the r e l a t i v e timing of attack on t h e i r host p l a n t s . The among year comparisons show that between some years the numbers of ga l l s per developed bud for the two species move together and between other years they move in opposite d i r e c t i o n s (Table 5.2, 5.3) . Thus other factors af fect the U. quadr i fasc iata dens i t ies besides the density of U. a f f i n i s . A refuge in space for U. quadri fasc i a ta may ex i s t by v ir tue of superior reproductive and d i s p e r s a l a b i l i t i e s . U. quadr i fasc ia ta may be more fecund than U. a f f i n i s . At Robertson's in 1980, the r a t i o of observed g a l l s to observed insects was much higher for U. quadr i fasc i a ta (4.99 vs . 1.91 for U. a f f i n i s ; x 2=427, df=1, p<0.00l) . The rate of increase in the f i r s t years of establishment was higher for U. quadr i fasc ia ta than for U. a f f i n i s ( H a r r i s , 1980a). Roze's (1981) data from a var ie ty of d i f fuse knapweed s i t e s in 1977 indicated that U. quadr i fasc i a ta was present at more s i tes than U. a f f i n i s (six to one), suggesting that of the two g a l l f l y species U. quadr i fasc iata spreads more r a p i d l y . The r e l a t i v e importance of these factors has not been evaluated. In combination, they appear to be s u f f i c i e n t to ensure that U. quadri fasc ia ta remains a s i g n i f i c a n t component of the insect complex that i s developing on the knapweeds in North 1 93 Amer i c a . SUMMARY This Chapter focusses on an analys is of the changes among years at the release s i t e s . I considered changes over the three years of f i e l d work and h i s t o r i c a l data extending back to the in troduct ion of the g a l l f l i e s . The density of ov ipos i t i on s i t e s appears to have depended on r a i n f a l l , the a v a i l a b i l i t y of su i tab le buds in time, and, in the longer term, on the reduction in seeds by the g a l l f l i e s . The f l i e s were l imi t ed by the density of a v a i l a b l e and sui table buds. The number of g a l l s per developed bud was d i f ferent among the release s i t e s and between knapweed species . Bud abortion changed with insect attack as expected and the propensity for plants to abort buds may have changed from year to year. The decl ine in the density of su i tab le buds may account for the drop in g a l l s per developed bud fo l lowing peak g a l l dens i t i e s . There i s further evidence from g a l l d i s t r i b u t i o n s that U. a f f i n i s had a negative influence on JJ. quadr i fasc i a t a , though U. a f f i n i s has not completely excluded U. q u a d r i f a s c i a t a . 194 APPENDIX VA. ESTIMATION OF BUD AVAILABILITY There are two basic elements of aggregation in the Centaurea- Urophora system. Because of the refuge in time for knapweed buds (Chapter I I ) , a v a r i a b l e number of buds w i l l never be a v a i l a b l e for at tack. When these buds are included in computation of means, var iances , and variance-mean r a t i o s , misleading and spurious re su l t s may be obtained. The resul ts would indicate clumping (because of the large zero c lass) even i f the attack on ava i lab l e buds was per fec t l y random and independent. The second element of aggregation is the d i s t r i b u t i o n of g a l l s between a v a i l a b l e buds. METHODS An estimate of the c o n t r i b u t i o n of the two elements may be obtained by the use of e i ther the truncated Poisson (Cohen, 1960) or the truncated negative binomial (Sampford, 1954). Since the truncated Poisson is a l i m i t i n g case of the truncated negative binomial (when k=inf in i ty ) and the l a t t er can accommodate cases where clumping is due to both elements, the truncated negative binomial seems to be better suited to the estimation problem. A FORTRAN program which computed both the truncated Poisson and truncated negative binomial parameters for the ava i lab le d i s t r i b u t i o n s was wr i t t en . The truncated Poisson f i t t i n g was based on Cohen (1960) and the truncated negative binomial on the maximum l i k e l i h o o d method of Sampford (1954) using an i n i t i a l estimate suggested by Brass (1958). An example of an observed 195 d i s t r i b u t i o n and the two f i t t e d d i s t r i b u t i o n s are given in Figure 5.5. The number of buds "unseen" in th i s analys i s is simply the d i f f erence between the number of buds that are unattacked in a given sample and the number of buds that are estimated to be unattacked as a resul t of the s t a t i s t i c a l propert ies of the d i s t r i b u t i o n of attack. The proportion "unseen" by o v i p o s i t i n g females i s the number of buds "unseen" div ided by the t o t a l number of buds in the sample. The assumption in th i s ana lys i s i s that the d i s t r i b u t i o n of attack is well described by the truncated negative binomial d i s t r i b u t i o n or the truncated Poisson d i s t r i b u t i o n . RESULTS The entr ies in the fourth column of Tables 5.5, 5.6, and 5.7 indicate that the estimated truncated Poisson d i s t r i b u t i o n d i f f e r s s i g n i f i c a n t l y from the observed d i s t r i b u t i o n in a number of cases (x 2 tes t , a=0.05). This r e f l e c t s the clumped d i s t r i b u t i o n of g a l l s within the ava i lab le buds. Thus the a d d i t i o n a l parameter ava i lab le for f i t t i n g the truncated negative binomial is useful for descr ib ing the observed d i s t r i b u t i o n s accurate ly . The two d i s t r i b u t i o n s tend to give quite s i m i l a r estimates of the proportions of buds unavai lable to the g a l l f l i e s . At Ned's Creek, the estimated proportion of buds "unseen" by the g a l l f l i e s was very large at low f ly dens i t i e s (0.70 in 1973; Table 5.5) . As r e l a t i v e f l y density increased, the proport ion dropped. It appears that i t had s t a b i l i z e d at a 1 96 gure 5.5. Observed d i s t r i b u t i o n of U. a f f i n i s g a l l s in developed buds at Ned's Creek in 1979 and f i t t e d , d i s t r i b u t i o n s . The large proport ion of buds "unseen" i s evident . In th is case, both truncated d i s t r i b u t i o n s give a good f i t to the observed d i s t r i b u t i o n (see Table 5.5) . 197 400 | | observed 300. truncated Poisson distribution >» o c a> 3 _ CT 0) 200. I s * r * ' I * * r * * [ * * r * ' • * * W i l l i • • • • • • • • • i i • • • • • • • • i 1 1 1 1 1 truncated negative binomial distribution 100. I * * r * I * * I * * I * • L • I * * l • • I • • r * ' I * * 2 3 4 Galls per bud 1 98 Table 5.5 - Proportion of d i f fuse knapweed buds unattacked and estimated proport ions of buds unavailable to o v i p o s i t i n g g a l l f l i e s , Ned's Creek 1973-1980 Truncated S i g n i f . Truncated Suitable Proportion Poisson D i f f . Negbinomial Estimate Year Unattacked Prop. Unavl . ? Prop. Unavl . ? UROPHORA AFFINIS 1 973 0.88410.004 0.57 Yes 0.77 Yes 1 974 0.81410.004 0.49 Yes 0.67 Yes 1 975 0.43910.024 -0.09 No 0.16 Yes 1 976 0.36710.012 0.19 Yes 0.16 Yes 1 977 0.54810.016 0.40 No 0.42 Yes 1978 0.48410.016 0.38 Yes 0.37 Yes 1 979 0.60110.019 0.36 No 0.44 Yes 1980 0.66710.014 0.55 Yes 0.42 Yes UROPHORA QUADRIFASCIATA 1 973 0.95210.003 0.90 No 0.92 Yes 1974 0.81210.004 0.69 Yes 0.71 Yes 1 975 0.54910.024 0.36 No 0.38 Yes 1976 0.79410.010 0.74 Yes 0.69 Yes 1 977 0.81210.012 0.77 No 0.76 Yes 1 978 0.95910.006 0.87 No 0.93 Yes 1 979 0.87210.013 0.77 No 0.81 Yes 1 980 0.91210.008 0.84 No 0.85 Yes BOTH SPECIES COMBINED 1 973 0.83510.005 0.52 Yes 0.70 Yes 1974 0.65110.005 0.39 Yes 0.45 Yes 1 975 0.22010.020 0.03 No 0.03 Yes. 1 976 0.22710.01 1 0.07 Yes -0.04 Yes 1 977 0.39310.016 0.25 No 0.25 Yes 1 978 0.46010.016 0.35 Yes 0.34 Yes 1 979 0.50910.019 0.26 No 0.33 Yes 1980 0.58810.014 0.44 Yes 0.28 Yes r e l a t i v e l y constant value of about 0.30. The high density and low proport ions unattacked in 1975 and 1976, poss ibly due to delayed plant and insect development in those years, reduced the estimated proportion "unseen" almost to zero. The samples from Robertson's gave a s teadi ly decreasing proport ion unattacked and a corresponding decl ine in the 1 99 estimated proportion "unseen" (Table 5 .6) . The changes in the Table 5.6 - Proportion of d i f fuse knapweed buds unattacked and estimated proportion of buds unavai lable to ov ipos i t ing g a l l f l i e s , Robertson's 1977-1981 Truncated S i g n i f . Truncated . Sui table Proportion Poisson D i f f . Negbinomial Estimate Year Unattacked Prop. Unavl . ? Prop. Unavl . ? UROPHORA AFFINIS 1 977 0 . 9 0 6 ± 0 . 0 0 6 0. 50 Yes 0. 76 No 1978 no data ava i lab le 1 979 0 . 6 5 3 ± 0 . 0 3 2 0. 48 Nd 0. 57 No 1980 0 . 4 4 7 ± 0 . 0 1 3 0. 34 Yes 0. 29 Yes 1981 0 . 2 3 2 ± 0 . 0 1 3 0. 16 No 0. 1 7 Yes UROPHORA QUADRIFASCIATA 1 977 0 . 8 5 1 ± 0 . 0 0 7 0.77 Yes 0 .76 Yes 1 978 no data ava i lab le 1 979 0 . 9 1 9 ± 0 . 0 1 8 0.86 NO 0 .88 No 1980 0 . 8 5 3 ± 0 . 0 0 9 0.80 NO 0 .81 Yes 1 981 0 . 8 7 2 ± 0 . 0 1 0 0.81 NO 0 .81 Yes BOTH SPECIES COMBINED 1977 0 . 7 7 0 ± 0 . 0 0 9 0.60 Yes 0 .60 Yes 1978 no data ava i lab le 1 979 0 . 5 9 5 ± 0 . 0 3 3 0.42 No 0 .52 Yes 1 980 0 . 3 5 0 ± 0 . 0 1 2 0.24 Yes 0 . 1 9 Yes 1 981 0 . 1 B 7 ± 0 . 0 1 2 0. 12 Yes 0 .12 Yes proport ions at Robertson's occurred more slowly than at Ned's Creek and lagged behind the population increase at Ned's Creek by four to f ive years. The estimated proportion of buds "unseen" at Chase decreased almost monotonically during the sampled years u n t i l i t reached a r e l a t i v e l y constant value of 0.05 in 1976 (Table 5 .7) . The proportion of buds unattacked reached a minimum in 1977, 200 Table 5.7 - Proport ion of spotted knapweed buds unattacked and estimated proport ion of buds unavailable to o v i p o s i t i n g g a l l f l i e s , Chase 1973-1981 Truncated S i g n i f . Truncated Suitable Proportion Poisson D i f f . Negbinomial Estimate Year Unattacked Prop. Unavl . ? Prop. Unavl . ? UROPHORA AFFINIS 1973 0 . 6 8 5 ± 0 . 0 0 8 0.35 Yes 0.49 Yes 1974 0 . 7 8 8 ± 0 . 0 0 5 0.59 Yes 0.63 Yes 1 975 0 . 3 1 0 ± 0 . 0 1 9 0.12 Yes 0.16 Yes 1 976 0 . 1 1 7 ± 0 . 0 1 3 0.09 Yes 0.07 Yes 1 977 0 . 0 7 7 ± 0 . 0 1 2 0.07 Yes 0.06 Yes 1978 0 . 1 8 0 ± 0 . 0 0 9 0.13 Yes 0.08 Yes 1979 0 . 2 6 3 ± 0 . 0 2 5 0.17 Yes 0.10 Yes 1980 0 . 3 4 6 ± 0 . 0 4 2 0.14 No 0.10 Yes 1981 0 . 0 7 1 ± 0 . 0 1 8 0.06 Yes 0.05 Yes UROPHORA QUADRIFASCIATA 1976 0 . 9 2 2 ± 0 . 0 1 1 0.91 No 0.90 Yes 1977 0 . 9 4 5 ± 0 . 0 1 0 0.93 No 0.92 Yes 1 978 0 . 6 1 2 ± 0 . 0 1 2 0.52 Yes 0.50 Yes 1979 0 . 9 1 8 ± 0 . 0 1 5 0.85 No 0.88 Yes 1980 0 . 8 4 6 ± 0 . 0 3 2 0.81 No 0.78 Yes 1981 0 . 8 6 3 ± 0 . 0 2 4 0.82 No 0.73 Yes BOTH SPECIES COMBINED 1976 0 . 0 9 4 ± 0 . 0 1 2 0.07 Yes 0.05 Yes 1 977 0 . 0 5 8 ± 0 . 0 1 0 0.05 Yes 0.04 Yes 1978 0 . 1 3 1 ± 0 . 0 0 8 0.11 Yes 0.07 Yes 1 979 0 . 2 2 9 ± 0 . 0 2 4 0.14 Yes 0.06 Yes 1980 0 . 2 9 0 ± 0 . 0 4 0 0.16 No 0.09 Yes 1 981 0 . 0 5 7 ± 0 . 0 1 6 0.04 Yes 0.04 Yes increased u n t i l 1980, and then dec l ined again in 1981. A comparison of Tables 5.5-5.7 with Table 5.4 indicates that the estimated proport ion of buds "unseen" changes roughly inversely to the density of g a l l s per bud. This suggests that there is no f ixed proport ion of buds that are unavailable to the g a l l f l i e s between years , and hence there is no constant proport ion of buds in a seed refuge. An increase in f l y density 201 w i l l not lead to a proport ional reduction in the estimated proportion "unseen", however, because of the observed di f ferences in the timing of attack (Chapter II) and bud abort ion (Chapter I I I ) . The density manipulation experiment described in Chapter II showed that a roughly three fo ld di f ference in adult density changed the proport ion of buds unattacked by only 8%. 202 CONCLUDING DISCUSSION The Introduct ion to th i s thes is i d e n t i f i e d four i n t e r l i n k e d c r i t e r i a for evolut ionary success. The fourth c r i t e r i o n of evolut ionary success, response to v a r i a t i o n , is a key indicator of how successful a species w i l l be in the future . This thesis has focussed on three l eve l s of var ia t ion in the host plants of the g a l l f l i e s : the bud, the p lant , and the population (cf . Morrison, 1984). The processes ac t ing .a t each of these leve ls w i l l be summarized in t u r n . BUDS The flower bud in knapweed plants is the basic unit for both g a l l and seed product ion . As ov ipos i t ion s i t e s , buds are an ephemeral and var iab le resource. The a b i l i t y of buds to support g a l l formation var ies considerably depending on s i ze , l oca t ion within p l a n t s , and time of i n i t i a t i o n . In the face of t h i s v a r i a t i o n , the g a l l f l i e s are se lec t ive about which buds they ov ipos i t i n . The d i rec t consequences of ov ipos i t i on s e l ec t ion are g a l l formation and a reduction in seed product ion. Bud abort ion resu l t s from common se lec t ion c r i t e r i a among insects and high r e l a t i v e dens i t ies of g a l l f l i e s . This outcome of f l y attack el iminates both g a l l and seed product ion. F l i e s do not appear to detect aborted buds without probing and rare ly ov ipos i t in them. PLANTS In the absence of vegetative reproduction, the i n d i v i d u a l plant is the demographic unit for the plant populat ion . The 203 unique pattern of resource a l l o c a t i o n for each plant means that every plant may also be viewed as a d i s cre t e and changing population of buds (cf . White, 1979). Increased resources in the form of nitrogen f e r t i l i z e r and water led to increased bud i n i t i a t i o n . G a l l f l i e s responded to the changes in the populations of buds on the p lant s . In general , g a l l f l i e s were observed on plants in d i r e c t proport ion to the number of buds per p lant . However, g a l l s are unevenly d i s t r i b u t e d among buds on a given p lant . The Urophora species act in a s i m i l a r fashion to " p a r t i a l predators" ( H a r v e l l , 1984), never completely e l iminat ing seed product ion. The way plants a l l o c a t e buds, among branching categories and in time, a l t e r e d the outcome of f l y at tack. Plants aborted buds in response to heavy insect a t tack , but there was minimal compensation for the los t seed product ion. The proportion of buds aborted was not s i g n i f i c a n t l y af fected by f e r t i l i z a t i o n or watering treatments. A higher proport ion of buds maturing appeared to increase the number of g a l l s per developed bud. POPULATIONS Populations give the greatest scope for long-term changes in the plants and are the l e v e l at which year-to-year var ia t i on is c r i t i c a l . Both changing resource a v a i l a b i l i t y between years, for example due to p r e c i p i t a t i o n , and changing plant densi ty , sh i f t the density of o v i p o s i t i o n s i t e s in space and time. 204 Es tab l i shed populations of g a l l f l i e s tracked changes in bud density very w e l l , however the resul t s of Chapter II indicated that a s i g n i f i c a n t f rac t ion of the bud population was unattacked because of a refuge in time. Another a d d i t i o n a l proport ion was unattacked because of the s t a t i s t i c a l propert ies of the g a l l f l i e s ' attack, that i s , a l i m i t a t i o n on the ir searching a b i l i t y . Their a b i l i t y to discover su i table o v i p o s i t i o n s i t e s was further reduced by the cumulative e f fects of bud a b o r t i o n . Bud abortion may be the most important factor preventing an increase in the number of g a l l s per bud. The e f fects of U. a f f i n i s at tack, bud abortion and g a l l formation, reduced g a l l formation by U. quadr i fa sc ia ta . The populat ion trends c l e a r l y indicate that population l i m i t a t i o n occurred at the two o r i g i n a l release s i t e s . Once the g a l l f l i e s were well e s tabl i shed , the numbers of g a l l s per bud changed only within a small range. Further increase was prevented in part by bud abort ion . The drop in seed production was c o r r e l a t e d with a much lower density of o v i p o s i t i o n s i t e s . This in turn reduced g a l l dens i t ies to the ir present l e v e l s . It i s not known how the number of g a l l s per bud w i l l change with bud dens i ty . Several processes have been explored in th i s thes i s , each with t h e i r own s p a t i a l and temporal sca le . In switching to a broader s p a t i a l scale and a longer temporal sca le , some processes disappear and others become more prominent. The ef fect of o v i p o s i t i o n se lect ion is combined with the effect of the r e l a t i v e t iming of bud i n i t i a t i o n and g a l l f l y emergence in 205 generating non-random g a l l d i s t r i b u t i o n s . At the other extreme, the r e l a t i v e rates of d i s p e r s a l of the insects and the plants w i l l a f fect the extent to which the plant populat ion is reduced by insect attack over several years , yet d i s p e r s a l w i l l not be apparent in the day-to-day changes occurr ing on a s ingle p lant . For p r a c t i c a l reasons, i t is necessary to l i m i t a study such as th i s one to c e r t a i n s p a t i a l and temporal scales . (Because of the economic importance of the weeds, the scales may be greater than i f the study had no management impl i ca t ions . ) An add i t iona l constra int is imposed by the requirement that the processes be regular and cons i s tent . The l a t t e r requirement ar i ses from the s c i e n t i s t ' s des ire for r e l i a b l e knowledge, knowledge that can be used to make p r e d i c t i o n s . Processes which are sporadic or inconsistent in t h e i r e f fec ts may be informative, but through descr ip t ion rather than systematic experimentation. For every eco log i ca l system, there are processes which are not a part of the defined system, but which have an impact on i t s behaviour. Expanding the system d e f i n i t i o n , perhaps by increasing the s p a t i a l or temporal scale of a n a l y s i s , does not avoid the problem. Perhaps the c l eares t example of such a process was the herbic ide treatment at Ned's Creek in 1980. It had a d r a s t i c ef fect on plant density and had the po tent ia l to r a d i c a l l y change the in terac t ion between the g a l l f l i e s and the ir hosts . (Remarkably, i t d id not .) Yet the human decis ion and the act ion that followed were outside the system I was studying and could not have been predic ted from the processes 206 occurr ing within and among years. Our a b i l i t y to evaluate evolutionary success from the response of population dynamics to environmental v a r i a t i o n i s neces sar i ly l i m i t e d , but as we struggle to understand the evolut ionary process and how i t interacts with our human goals , information l i k e th i s may provide essent ia l c lues . 207 LITERATURE CITED Aker, C . L . 1982. Regulation of flower, f r u i t , and seed production by a monocarpic p e r e n n i a l , Yucca whipple i . Journal of Ecology 70:357-372. Andres, L . A . 1977. The b i o l o g i c a l c o n t r o l of weeds, pp. 153-176 in J . D . Fryer and S. Matsunaka, e d i t o r s . Integrated Control of Weeds. Un ivers i ty of Tokyo Press , Tokyo. Andres, L . A . , and R.D. Goeden. 1971. The b i o l o g i c a l contro l of weeds by introduced natural enemies, pp.143-164 in C . B . Huffaker, e d i t o r . B i o l o g i c a l C o n t r o l . Plenum Press, New York. Andrewartha, H . G . and L . C . B i r c h . 1954. The D i s t r i b u t i o n and Abundance of Animals. U n i v e r s i t y of Chicago Press, Chicago. Auerbach, M . J . , and D.R. Strong. 1981. N u t r i t i o n a l ecology of He l i con ia herbivores: experiments with plant f e r t i l i z a t i o n and a l t e r n a t i v e hosts . E c o l o g i c a l Monographs 51:63-83. Bach, C . E . 1981. Host plant growth form and d i v e r s i t y : e f fects on abundance and feeding preference of a s p e c i a l i s t herbivore , Acalymma v i t t a t a (Coleoptera: Chrysomelidae). Oecologia 50:370-375. Bardner, R . , and K . E . F l e t c h e r . 1974. Insect in fes tat ions and the i r e f fects on the growth and y i e l d of f i e l d crops: A review. B u l l e t i n of Entomological Research 64:141-160. Berube, D . E . 1978a. L a r v a l descr ip t ions and biology of T e p h r i t i s  d i l a c e r a t a [Diptera: T e p h r i t i d a e ] , a candidate for the b iocontro l of Sonchus arvensis in Canada. Entomophaga 23: 69-82. Berube, D . E . 1978b. The basis for host plant s p e c i f i c i t y in T e p h r i t i s d i l a c e r a t a and T. formosa (Diptera: T e p h r i t i d a e ) . Entomophaga 23:331-337. 208 Berube, D . E . 1980. In terspec i f i c competition between Urophora  af f i n i s and U. quadri fasc iata (Diptera: Tephri t idae) for o v i p o s i t i o n a l s i tes on d i f fuse knapweed (Centaurea  d i f f u s a : Compositae). Z e i t s c h r i f t e fur angewandte Entomologie 90:299-306. Berube, D . E . and P. H a r r i s . 1978. Bud s ize se lec t ion by o v i p o s i t i n g U. a f f i n i s and U. quadri fasc ia ta in B . C . pp. 210-217 in Proceedings of the Knapweed Symposium (Oct. 6-7, 1977: Kamloops, B . C . ) . Berube, D . E . , and J . H . Myers. MS. Reproductive i s o l a t i o n between Urophora aff in i s and U. quadr i fasc ia ta (Diptera: T e p h r i t i d a e ) . B i r c h , L . C . 1979. The effect of species of animals which share common resources on one another's d i s t r i b u t i o n and abundance. F o r t s c h r i f t e Zoologie 25:197-221. Borg ia , G. 1981. Mate se lect ion in the f l y Scatophaga s t e r c o r a r i a : female choice in a male-contro l led system. Animal Behaviour 29:71-80. Brass , W. 1958. S impl i f i ed methods for f i t t i n g the truncated negative binomial d i s t r i b u t i o n . Biometrika 45:59-68. Brown, G . C . , A . A . Berryman, and T . P . Bogyo. 1979. Densi ty-dependent induction of diapause in the codl ing moth, Laspeyresia pomonella (Lepidoptera: Ole threut idae ) . Canadian Entomologist 111:431-433. Brown, J . H . 1984. On the re la t i onsh ip between abundance and d i s t r i b u t i o n of species. American N a t u r a l i s t 124:255-279. Brues, C T . 1920. Select ion of food plants by insects with s p e c i a l reference to Lepidoptera. American N a t u r a l i s t 54: 313-332. Buchner, A . , and H. Sturm. 1971. F e r t i l i z e r A p p l i c a t i o n in Intensive Agr icu l ture (3rd e d . ) . DLG - V e r l a g , Frankfur t /Main , Germany. 209 Burk, T . 1981. S ignal ing and sex in aca lyptrate f l i e s . F l o r i d a Entomologist 64:30-43. Calcote , V . R . 1975. Pecan weevi l : feeding and i n i t i a l ov ipos i t i on as re la ted to nut development. Journal of Economic Entomology 68:4-6. Campanella, P . J . 1975. The evolut ion of mating systems in temperate zone dragons (Odonata: Anisoptera) . I I . L i b e l l u l a luctuosa (Burmeister) . Behaviour 54:278-310. Chapin, F . S . , I I I , and M. S lack . 1979. Ef fec t of d e f o l i a t i o n upon root growth, phosphate absorpt ion , and resp ira t ion in n u t r i e n t - l i m i t e d tundra graminoids. Oecologia 42:67-79. Chew, F . S . 1975. Coevolution of p i e r i d b u t t e r f l i e s and the i r cruc i ferous foodplants . I . The r e l a t i v e qua l i ty of ava i l ab l e resources. Oecologia 20:117-127. Chew, F . S . 1977. Coevolution of p i e r i d b u t t e r f l i e s and the i r cruc i ferous foodplants . I I . The d i s t r i b u t i o n of eggs on p o t e n t i a l food p l a n t s . Evo lu t ion 31:568-579. Cohen, A . C . 1960. Est imating the parameter in a condi t iona l Poisson d i s t r i b u t i o n . Biometrics 16:203-211. Courtney, S .P. 1981. Coevolution of p i e r i d b u t t e r f l i e s and t h e i r cruc i ferous foodplants . I I I . Anthocaris cardamines L . s u r v i v a l , development and o v i p o s i t i o n on d i f ferent hostp lants . Oecologia 51:91-96. Cox, W . J . , and H.M. Reisnauer. 1973. Growth and ion uptake by wheat supplied nitrogen as n i t r a t e , or ammonium, or both. Plant and S o i l 38:363-380. Crawley, M . J . 1983. Herbivory: The Dynamics of Animal-Plant Interact ions . Blackwell S c i e n t i f i c P u l i c a t i o n s , Oxford. Decau, J . , and B. P u j o l . 1973. Comparative effects of i r r i g a t i o n and nitrogen f e r t i l i z e r on the q u a l i t a t i v e and quant i ta t ive production of d i f f e r e n t maize c u l t i v a r s . Annals of Agronomy 24:359-373. 210 Dempster, J . P . , and E . P o l l a r d . 1981. Fluctuat ions in resource a v a i l a b i l i t y and insect populat ions . Oecologia 50:412-416. Deth ier , V . G . 1941. Chemical factors determining the choice of food plants by P a p i l i o larvae . American N a t u r a l i s t 75:61-73. Dieleman, F . L . 1969. Ef fec t s of g a l l midge in fe s ta t ion on plant growth and growth regulat ing substances. Entomologia experimentalis et appl ica ta 12:745-749. Dixon, A . F . G . 1971a. The role of aphids in wood formation. I . The effect of the sycamore aphid, Drepanosiphum  platanoides (Schr.) (Aphididae) on the growth of sycamore, Acer pseudoplatanus ( L . ) . Journal of appl ied Ecology 8: 165-179. Dixon, A . F . G . 1971b. The role of aphids in wood formation. I I . The effect of the lime aphid, E u c a l l i p t e r u s t i l i a (L . ) (Aphididae) on the growth of l ime, T i l i a v u l g a r i s Hayne. Journal of appl ied Ecology 8:393-399. Dixon, A . F . G . 1976. Timing of egg hatch and v i a b i l i t y of the Sycamore aphid, Drepanosiphum platanoides ( S c h r . ) , at bud burst of Sycamore, Acer pseudoplatanus L . Journal of Animal Ecology 45:493-603. van Emden, H . F . 1972. Aphids as phytochemists. pp. 25-43 in J . Harborne ( e d i t o r ) . Phytochemical Ecology. Academic Press , N.Y. Faeth, S . H . , E . F . Connor, and D. Simberloff . 1981. Ear ly leaf absc i s s ion: a neglected source of morta l i ty for f o l i v o r e s . American Natura l i s t 117:409-415. Feeny, P. 1970. Seasonal changes in oak leaf tannins and nutr ients as a cause of spring feeding by winter moth c a t e r p i l l a r s . Ecology 51:565-581. Finney, D . J . , and G . C . V a r l e y . 1955. An example of the truncated Poisson d i s t r i b u t i o n . Biometrics 11:387-394. 21 1 Fourcroy, M . , and C. Braun. 1967. Observations sur la ga l l e de 1'Aulax qlechomas L . sur Glechoma hederocea L . Second p a r t i e . H i s t o l o g i e et ro le physiologue de la coque s c l e r i f i e e . M a r c e l l i a 34:3-30. Fraenke l , G. 1959. The raison d'etre of secondary plant substances. Science 129:146-170. Frankton, C , and G . A . M u l l i g a n . 1977. Weeds of Canada. Canada Department of A g r i c u l t u r e , Ottawa. Gigon, A . , and I . H . Ror i son . 1972. The response of some e c o l o g i c a l l y d i s t i n c t plant species to n i t ra te and ammonium n i trogen . Journal of Ecology 60:93-102. Green, T .W. , and I . G . Palmblad. 1975. Effects of insect seed predators on Astragalus c i b a r i u s and Astragalus utahensis (Leguminosae). Ecology 56:1435-1440. Gromko, M . H . , and D.W. Py le . 1978. Sperm competit ion, male f i tne s s , and repeated mating by female Drosophila  melanogaster. Evo lut ion 32:588-593. Hare, J . D . , and D . J . Futuyma. 1978. Di f ferent effects of v a r i a t i o n in Xanthium strumarium L . (Compositae) on two insect seed predators . Oecologia 37: 109-120. Harper, J . L . 1977. Populat ion Biology of P lants . Academic Press , New York, New York. H a r r i s , P. 1972. Insects in the population dynamics of p l a n t s . Royal Entomological Society Symposium. H a r r i s , P. 1980a. Establishment of Urophora a f f i n i s F r f l d . , and U. quadr i fasc i a ta (Meig.) (Diptera: Tephrit idae) in Canada for the b i o l o g i c a l c o n t r o l of d i f fuse and spotted knapweed. Z e i t s c h r i f t e fur angewandte Entomologie 89:504-514. H a r r i s , P. 1980b. Ef fec t s of Urophora a f f i n i s F r f l d . , and U. quadr i fasc ia ta (MeigTl (Diptera: Tephrit idae) on Centaurea d i f fusa Lam., and C. maculosa Lam. (Compositae). Z e i t s c h r i f t e fur angewandte Entomologie 90:190-201• 212 H a r r i s , P . , and R. Cranston. 1979. An economic evaluat ion of contro l methods for d i f fuse and spotted knapweed in western Canada. Canadian Journal of Plant Science 59:375-382. H a r v e l l , D. 1984. Why nudibranchs are p a r t i a l predators: i n t r a c o l o n i a l v a r i a t i o n in bryozoan p a l a t a b i 1 i t y . Ecology 65:716-724. H a s s e l l , M.P. 1978. The Dynamics of Arthropod Predator-Prey Systems. Princeton Univers i ty Press . Pr inceton , N . J . Hendrix, S.D. 1979. Compensatory reproduction in a b i ennn ia l herb fol lowing insect d e f l o r a t i o n . Oecologia 42:107-118. Hirose , Y . , Y. Suzuki , M. Takagi , K. Hiehata, M. Yamasaki, H. Kimoto, M. Yamanaka, M. Iga, K. Yamaguchi. 1980. Population dynamics of the c i t r u s swa l lowta i l , P a p i l i o  xuthus Linne (Lepidotera: P a p i l i o n i d a e ) : mechanisms s t a b i l i z i n g i t s numbers. Researches in Population Ecology 21 :260-285. H O r i , K. 1974. Plant growth-promoting factor in the s a l i v a r y gland of the bug, Lygus d i spons i . Journal of Insect Physiology 20:1623-1627. Howard, R.D. 1978a. The influence of male-defended o v i p o s i t i o n s i tes on ear ly embryo morta l i ty in b u l l f r o g s . Ecology 59: 789-798. Howard, R.D. 1978b. The evolut ion of mating s trateg ies in b u l l f r o g s , Rana catesbeiana. Evolut ion 32:850-871. Hubbard, W.A. 1971. B i o l o g i c a l weed c o n t r o l . Annual Report Research S ta t ion . Canada Department of A g r i c u l t u r e . Kamloops, B r i t i s h Columbia. Huber, F . 1974. Neural in tegrat ion (central nervous system), pp. 3-100 in M. Rockstein, e d i t o r . The Physiology of Insecta , IV. 2nd e d i t i o n . Academic Press, New York. 213 Huffaker, C . B . 1958. Experimental studies on predat ion: d i spers ion fac tors and predator-prey o s c i l l a t i o n s . H i l g a r d i a 27:343-383. Jankiewicz, L . S . , H. F l i c h , and H. Antoszewski. 1970. Pre l iminary studies on the trans locat ion of C - l a b e l l e d ass imi la tes and PO towards the g a l l evoked by Cynips (D ip lo l eps i s ) quercus fo l i i L . on oak leaves. M a r c e l l i a 36: 163-172. Jones, R . E . , N. G i l b e r t , M. Guppy, and V. Nea l i s . 1980. Long distance movement in P i e r i s rapae. Journal of Animal Ecology 49:629-642. Kaireva , P. 1983. Influence of vegetation texture on herbivore populat ions: resource concentration and herbivore movement, pp. 259-289 in R . F . Denno and M.S. McClure, e d i t o r s . V a r i a b l e Plants and Herbivores in Natural and Managed Systems. Academic Press, New York. Kimmins, J . P . 1971. V a r i a t i o n s in the f o l i a r amino ac id composition of f lowering balsam f i r [Abies balsamea(L.) M i l l . ] and white spruce [Picea glauca(Moenchl Voss] in r e l a t i o n to outbreaks of the spruce budworm [Choristoneura  fumiferana(Clem)]. Canadian Journal of Zoology 49: 1005-1011. Kingso lver , J . G . 1979. Thermal and hydric aspects of environmental heterogeneity in the pitcher plant mosquito. E c o l o g i c a l Monographs 49:357-376. Lawton, J . H . , and M.P. H a s s e l l . 1981. Asymmetrical competit ion in insec t s . Nature 289:793-795. Lev in , S . A . , and J . D . Udovic . 1977. A mathematical model of coevolving populat ions . American Natura l i s t 111:657-675. Louda, S.M. 1982. L i m i t a t i o n of the recruitment of the shrub Haplopappus squarrosus (Asteraceae) by flower- and seed-feeding in sec t s . Journal of Ecology 70:43-53. 214 Louda, S.M. 1984. Herbivore effect on s ta ture , f r u i t i n g , and leaf dynamics of a native c r u c i f e r . Ecology 65:1379-1386. McEvoy, P .B . 1984. Recent advances in b i o l o g i c a l c o n t r o l of spotted knapweed by insects , pp. 57-65 in Proceedings of the Knapweed Symposium. B u l l e t i n 1315. Plant and S o i l Science Department and Cooperative Extension Serv ice , Montana State U n i v e r s i t y . Bozeman, Montana. M c N e i l l , S . , and T . R . E . Southwood. 1978. The ro le of nitrogen in the development of insec t /p lant r e l a t i o n s h i p s , pp. 77-98 in J . B . Harborne, e d i t o r . Biochemical Aspects of Plant and Animal Coevolut ion. Academic Press , London. Maddox, D.M. 1979. The knapweeds: the i r economics and b i o l o g i c a l contro l in the western states , U . S . A . Rangelands J_:139-141. Mani, M.S. 1964. Ecology of Plant G a l l s . Dr. W. Junk, The Hague. Mattson, W . J . , J r . 1980. Herbivory in r e l a t i o n to plant nitrogen content. Annual Review of Ecology and Systematics 11:119-161. Mattson, W . J . , and N.D. Addy. 1975. Phytophagous insects as regulators of forest primary product ion. Science 190:515-522. van der Meijden, E . 1971. Senecio and T y r i a in a Dutch dune area. A study on an in terac t ion between a monophagous consumer and i t s host p lant , pp. 390-404 in P . J . den Boer and G.R. Gradwel l , e d i t o r s . Dynamics of Populat ions: Proceedings of the Advanced Study I n s t i t u t e on Dynamics of Numbers in Populat ions . Oosterbeek, the Netherlands. 7-18 September, 1970. Centre for A g r i c u l t u r a l Publ i sh ing and Documentation, Wageningen. M i l l e r , E . R . , and H. Dingle . 1982. The ef fect of host plant phenology on reproduction of the milkweed bug, Oncopeltus  fasc ia tus , in t r o p i c a l F l o r i d a . Oecologia 52:97-103. 215 Mispagel , M . E . 1978. The ecology and bioenergetics of a c r i d i d grasshopper, Bootett ix punctatus, on creosotebush, Larrea  t r identata , in the northern Mojave desert . Ecology 59:779-788. Monro, J . 1967. The exp lo i ta t ion and conservation of resources by populations of insects . Journal of Animal Ecology 36: 531-547. Mooney, H . A . , K . S . Wi l l i ams , D . E . L i n c o l n , and P.R. E h r l i c h . 1981. Temporal and s p a t i a l v a r i a b i l i t y in the in terac t ion between the checkerspot b u t t e r f l y , Euphydryas chalcedona and i t s p r i n c i p a l food source, the C a l i f o r n i a n shrub, Diplacus aurant iacus . Oecologia 50:195-198. Moore, L . R . 1978a. Seed predation in the legume C r o t a l a r i a . I . Intens i ty and v a r i a b i l i t y of seed predation in native and introduced populations of C. p a l l i d a A i t . Oecologia 34: 185-202. Moore, L . R . 1978b. Seed predation in the legume Crota lar i a . I I . Corre la tes of in terp lant v a r i a b i l i t y in predation i n t e n s i t y . Oecologia 34:203-223. Moore, R . J . , and C . Frankton. 1974. The T h i s t l e s of Canada. Monograph 10. Research Branch, Canada Department of A g r i c u l t u r e . Ottawa, Ontar io . M o r r i s , R . F . 1963. Pred ic t i ve population equations based on key f a c t o r s . Memoirs of the Entomolgical Society of Canada 32: 16-2 1 . Morr i son , P .D. 1984. Evaluat ing the effect iveness of Urophora  a f f i n i s as a c o n t r o l agent for di f fuse and spotted knapweed, pp. 53-57 in Proceedings of the Knapweed Symposium. B u l l e t i n 1315. Plant and S o i l Science Department and Cooperative Extension Serv ice , Montana State U n i v e r s i t y . Bozeman, Montana. Morse, D . H . , and R . S . F r i t z . 1982. Experimental and observat ional studies of patch choice at d i f f eren t scales by the crab spider Misumena v a t i a . Ecology 63:172-182. 216 Myers, J . H . 1981. Interact ions between western tent c a t e r p i l l a r s and wild rose: a test of some general plant herbivore hypotheses. Journal of Animal Ecology 50:11-25. Myers, J . H . 1985. Ef fect of p h y s i o l o g i c a l condi t ion of the host plant on the o v i p o s i t i o n a l choice of the cabbage white b u t t e r f l y , P i e r i s rapae. Journal of Animal Ecology 54:193-204. Myers, J . H . , and D . E . Berube. 1983. Di f fuse knapweed invasion into rangeland in the dry i n t e r i o r of B r i t i s h Columbia. Canadian Journal of Plant Science 63:981-987. Myers, J . H . , and P. H a r r i s . 1980. D i s t r i b u t i o n of Urophora g a l l s in flower heads of d i f fuse and spotted knapweed in B r i t i s h Columbia . Journal of Appl ied Ecology 17:359-367. Myers, J . H . , and B . J . Post. 1981. Plant nitrogen and f luctuat ions of insect populat ions: a test with the cinnabar moth - tansy ragwort system. Oecologia A8:151-1 56. Nicholson, A . J . , and V . A . B a i l e y . 1935. The balance of animal populations - Part I . Proceedings of the Zoolog ica l Society , London. 1935:551-598. Onuf, C P . , J . M . T e a l , and I . V a l i e l a . 1977. Interact ions of nutr i ent s , plant growth, and herbivory in a mangrove ecosystem. Ecology 58:514-526. Osborne, D . J . 1972. Mutual regulat ion of growth and development in plants and insects , pp. 33-42 in H . F . van Emden, Insect /Plant Re la t ionsh ips , Symposia of Royal Entomological Society (London). No.6. B lackwe l l , Oxford. Parker, G.A. 1970. Sperm competition and i t s evolut ionary consequences in the insec ts . B i o l o g i c a l Review 45:525-567. Parker, G .A . 1974. Courtship persistence and female-guarding as male time-investment s t ra teg i e s . Behaviour 48:157-184. 217 Parker, M.A. 1984. Local food deplet ion and the foraging behavior of a s p e c i a l i s t grasshopper, Hesperotett ix  v i r i d i s . Ecology 65:824-835. Pinthus, M . J . , and E . M i l l e t . 1978. Interact ions among number of s p i k e l e t s , number of grains , and grain weight in the spikes of wheat (Tri t icum aestivum L . ) . Annals of Botany 42:839-848. Port , G . R . , and J . R . Thompson. 1980. Outbreaks of insect herbivores on plants along motorways in the United Kingdom. Journal of Applied Ecology 17:649-656. P r i c e , P .W. , C . E . Bouton, P. Gross, B .A. McPheron, J . N . Thompson, and A . E . Weis. 1980. Interact ion among three trophic l e v e l s : influence of plants on in teract ions between insect herbivores and natural enemies. Annual Review of Ecology and Systematics 11:41-65. Prokopy, R . J . 1981. E p i d e i c t i c pheromones that influence spacing patterns of phytophagous insects , pp. 181-213 in D.A. Norlund, R . L . Jones, and W.J . Lewis ( e d i t o r s ) . Semiochemicals: Their Role in Pest C o n t r o l . Wiley, N .Y. Rausher, M.D. 1979a. Egg recogni t ion: i t s advantage to a b u t t e r f l y . Animal Behaviour 27:1034-1040. Rausher, M.D. 1979b. Larva l habitat s u i t a b i l i t y and ov ipos i t i on preference in three re lated b u t t e r f l i e s . Ecology 60:503-511. Rausher, M . D . , and D.R. Papaj. 1983. Demographic consequences of d i s c r i m i n a t i o n among conspecif ic host plants by Battus  phi lenor b u t t e r f l i e s . Ecology 64:1402-1410. Rhoades, D . F . 1983. Herbivore population dynamics and plant chemistry, pp. 155-220 in R . F . Denno and M.S. McClure, e d i t o r s . Var iab le Plants and Herbivores in Natural and Managed Systems. Academic Press, New York. Richards , L . J . , and J . H . Myers. 1980. Maternal influences on s ize and emergence time of the cinnabar moth. Canadian Journal of Zoology 58:1452-1457. 218 Roland, J . 1986. Success and f a i l u r e of Cyzenis a lbicans in c o n t r o l l i n g i t s host the winter moth. PhD. Thes i s . Department of Zoology, U n i v e r s i t y of B r i t i s h Columbia. Vancouver, B . C . 213pp. Roze, L . D . 1981. The b i o l o g i c a l c o n t r o l of Centaurea d i f fusa Lam. and C. maculosa Lam. by Urophora a f f i n i s Frauenfeld and U. quadri fasc ia ta Meigen (Diptera: T e p h r i t i d a e ) . PhD. Thes i s , Dept. of Plant Science, U n i v e r s i t y of B r i t i s h Columbia, Vancouver, B . C . Roze, L . D . , and B.D. Frazer . 1978. B i o l o g i c a l contro l of di f fuse and spotted knapweed by Urophora a f f i n i s and U. quadri fasc ia ta in B r i t i s h Columbia, pp. 239-254 in Proceedings of the Knapweed Symposium (Oct. 6-7, 1977: Kamloops, B . C . ) . Sampford, M.R. 1954. The truncated negative binomial d i s t r i b u t i o n . Biometrika 42:58-69. SAS I n s t i t u t e . 1982. SAS User's Guide: S t a t i s t i c s . 1982 E d i t i o n . SAS Ins t i tu te Inc. Cary, North C a r o l i n a . Schaffer , W.M., and M . L . Rosenzweig. 1978. Homage to the red queen. I . Coevolution of predators and the i r v ic t ims . Theore t i ca l Population Biology 14:135-157. Schirman, R. 1981. Seed production and spr ing seedling establishment of d i f fuse and spotted knapweed. Journal of Range Management 34:45-47. Schlager, G. 1960. Sperm precedence in the f e r t i l i z a t i o n of eggs in Tr ibo l ium castaneum. Annals of the Entomological Society of America 53 : 557-560. Schroeder, L . A . , and M. Maimer. 1980. Dry matter, energy, and nitrogen conversion by Lepidoptera and Hymenoptera larvae fed leaves of black cherry . Oecologia 45:63-71. S c r i b e r , J . M . , and F . Slansky, J r . 1981. The n u t r i t i o n a l ecology of immature insec ts . Annual Review of Entomology 26:183-211. 219 Sheley, R . L . , R . H . C a l l i h a n , and C . H . Huston. 1984. Improvement of spotted knapweed-infested pastures with picloram and f e r t i l i z e r , pp. 21-22 in Proceedings of the Knapweed Symposium. B u l l e t i n 1315. Plant and S o i l Science Department and Cooperative Extension Serv ice , Montana State U n i v e r s i t y . Bozeman, Montana. S l a t k i n , M . , and J . Maynard Smith. 1979. Models of coevo lut ion . Quarter ly Review of Biology 54:233-263. Smith, D . C . , and R . J . Prokopy. 1980. Mating behaviour of Rhagoletis pomonella (Diptera: T e p h r i t i d a e ) . V I . S i t e of ear ly season encounters. Canadian Entomologist 112:585-590. Smith, R . L . 1979. Repeated copulat ion and sperm precedence: patern i ty assurance for a male brooding water bug. Science 205:1029-1031. Sokal , R . R . , and F . J . Rohlf . 1969. Biometry: The P r i n c i p l e s and Prac t i ce of S t a t i s t i c s in B i o l o g i c a l Research. W.H. Freeman, San Franc i sco . Solomon, B . P . 1981. Response of a hos t - spec i f i c herbivore to resource densi ty , r e l a t i v e abundance, and phenology. Ecology 62:1205-1214. Solomon, M . E . 1949. The natural contro l of animal populat ions . Journal of Animal Ecology 18:1-35. Stapanian, M.A. 1982. A model for f r u i t i n g d i sp lay : seed d i s p e r s a l by birds for mulberry trees . Ecology 63:1432-1 443. Stephenson, A . G . 1980. F r u i t set , herbivory , f r u i t reduct ion , and the f r u i t i n g strategy of Catalpa speciosa (Bignoniaceae). Ecology 61:57-64. Stephenson, A . G . 1981. Flower and f r u i t abort ion: proximate causes and ultimate funct ions . Annual Review of Ecology and Systematics 12:253-279. 220 Story, J . M . 1978a. Biology of spotted knapweed in Montana, pp. 31-38 in Proceedings of the Knapweed Symposium (Oct. 6-7, 1977: Kamloops, B . C . ) . Story, J . M . 1978b. Ef fec t of U. a f f i n i s on spotted knapweed seed product ion, pp. 208-209 in Proceedings of the Knapweed Symposium (Oct. 6-7, 1977: Kamloops, B . C . ) . Story , J . M . , and N . L . Anderson. 1978. Release and establishment of Urophora a f f i n i s (Diptera: Tephri t idae) on spotted knapweed in western Montana. Environmental Entomology 1_: 445-448. Story, J . M . , and R.M. Nowierski . 1984. Status of b iocontro l knapweed research in Montana, pp. 65-70 in Proceedings of the Knapweed Symposium. B u l l e t i n 1315. Plant and S o i l Science Department and Cooperative Extension Service , Montana State U n i v e r s i t y . Bozeman, Montana. Strong, D . R . , J . H . Lawton, and T . R . E . Southwood. 1984. Insects on Plants : Community Patterns and Mechanisms. Harvard Univers i ty Press , Cambridge, Massachusetts. Sutton, R.D. 1984. The e f fect of host plant flowering on the d i s t r i b u t i o n and growth of hawthorn p s y l l i d s (Homoptera: P s y l l o i d e a ) . Journal of Animal Ecology 53:37-50. Tauber, M . J . , C . A . Tauber, and S. Masaki . 1986. Seasonal Adaptations of Insects . Oxford Univers i ty Press, New York. T a y l o r , F . 1981. Ecology and evo lut ion of phys io log ica l time in insec t s . American N a t u r a l i s t 117:1-23. Thompson, J . N . , and P.W. P r i c e . 1977. Plant p l a s t i c i t y , phenology, and herbivore d i s p e r s i o n : wild parsnip and the parsnip webworm. Ecology 58:1112-1119. Udovic, D . , and C . L . Aker. 1981. F r u i t abortion and the regulat ion of f r u i t numbers in Yucca whipple i . Oecologia 49:245-248. 221 Uvarov, B .P . 1931. Insects and c l imate . Transactions of the Entomological Society of London 79:1-247. V a r l e y , G . C . 1947. The natural contro l of population balance in the knapweed g a l l - f l y (Urophora jaceana). Journal of Animal Ecology 16:139-187. V a r l e y , G . C , and G.R. Gradwell . 1960. Key factors in populat ion s tudies . Journal of Animal Ecology 29:399-401 . V ince , S.W., and I . V a l i e l a . 1981. An experimental study of the structure of herbivorous insect communities in a sa l t marsh. Ecology 62:1662-1678. Watson, A . K . 1972. The biology and contro l of Centaurea d i f f u s a Lam., and C. maculosa Lam. MSc. Thes is , Dept. of Plant Science, Univers i ty of B r i t i s h Columbia, Vancouver, B . C . 101 pp. Watson, A . K . , and A . J . Renney. 1974. The biology of Canadian weeds. 6. Centaurea d i f fusa and C. maculosa. Canadian Journal of Plant Science 54:687-701. White, J . 1979. The plant as a metapopulation. Annual Review of Ecology and Systematics 10:109-145. White, T . C . R . 1976. Weather, food, and plagues of l ocus t s . Oecologia 22:119-134. Whitham, T . G . 1978. Habitat se l ec t ion by Pemphigus aphids in response to resource l i m i t a t i o n and competi t ion. Ecology 59:1164-1176. Whitham, T . G . 1980. The theory of habitat s e l e c t i o n : examined and extended using Pemphigus aphids. American N a t u r a l i s t 115:449-466. Wiklund, C. 1975. The evolutionary r e l a t i o n s h i p between adult ov ipos i t ion preferences and l a r v a l host plant range in P a p i l i o machaon L . Oecologia 18:185-197. 222 Wil l iams , R . J . , K. Broersma, and A . L . van Ryswyk. 1979. The ef fects of nitrogen f e r t i l i z a t i o n on water use by crested wheatgrass. Journal of Range Management 32:98-100. Wilson, F . 1960. A Review of the B i o l o g i c a l Control of Insects and Weeds in A u s t r a l i a and Aus tra l ian New Guinea. Technical Communication No. 1, Commonwealth Ins t i tu te of B i o l o g i c a l C o n t r o l , Ottawa. Wilson, F . 1964. The b i o l o g i c a l contro l of weeds. Annual Review of Entomology 9:225-244. Zwolfer, H. 1969. Urophora s iruna-seva (Hg.) (Diptera: Trypet idae) ; a p o t e n t i a l insect for the b i o l o g i c a l contro l of Centaurea s o l s t i t i a l i s L . in C a l i f o r n i a , pp. 105-155 in Technical B u l l e t i n No. 11, Commonwealth Ins t i tute of B i o l o g i c a l C o n t r o l , Commonwealth A g r i c u l t u r a l Bureaux, Farnham Royal , England. Zwolfer, H. 1970. Invest igat ions on the host s p e c i f i c i t y of Urophora a f f i n i s F r f l d . (Diptera: Trypet idae) . Progress Report of the Commonwealth Ins t i tu te of B i o l o g i c a l C o n t r o l , no. 25, 28 pp. Zwolfer, H. 1974. Das T r e f p u n k t - P r i n z i p als Komrnunikationsstrategie und Isolation-mechanismus bei Bohrfl iegen (Diptera: Trype t idae ) . Entomologia Germanica 1:11 -20. Zwolfer, H. 1978. An ana lys i s of the insect complexes associated with the heads of European Centaurea maculosa populat ions , pp. 139-163 in Proceedings of the Knapweed Symposium (Oct. 6-7, 1977: Kamloops, B . C . ) . Zwolfer, H. 1979. Strateg ies and counter s trategies in insect population systems competing for space and food in flower heads and plant g a l l s . F o r t s c h r i f t e Zoologie 25:331-335. 223 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0097476/manifest

Comment

Related Items