@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Land and Food Systems, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Lyons, Donald Barry"@en ; dcterms:issued "2010-10-07T19:53:32Z"@en, "1988"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Models of phenology of the pine false webworm (PFW), Acantholyda erythrocephala and one of its parasitoids were developed from relationships between PFW spatial distribution and microweather. Development of subterranean stages of PFW was simulated from rate-summation models developed from nonlinear regression equations describing the relationship between temperature and rate of development of post-diapause prepupae and pupae. Defoliation caused by PFW increased the soil's exposure to solar radiation resulting in higher soil temperatures and a corresponding reduction in development time of subterranean stages. Predictability was enhanced slightly when the distribution of insects and temperature of the soil were incorporated into the model. Increasing the time increment used in the model from 1 to 4 h did not adversely affect its resolution. Mating and oviposition of PFW occur within a few hours of emerging from the soil and the majority of PFW eggs were mature and ready for deposition at female emergence. Potential fecundity of PFW was accurately predicted from adult wet and dry weights. The oviposition pattern of PFW was also described by a model based on temperature-dependent oviposition and ageing rate functions. The effect of larval web construction on the development of arboreal stages was investigated. When exposed to sunlight, the web traps heat and raises the body temperature of its inhabitants. A model was developed and used to examine the significance of the web microclimate for development of larvae. Relationships between web temperatures, canopy temperatures and standard meteorological methods were developed to permit using data from standard weather stations to drive the model. Larval development increased by 1.4 to 2.8 d when estimated web temperatures were incorporated into the model, while development was retarded by 2.6 to 4.0 d when canopy temperatures were used instead of meteorological screen temperatures. Two ichneumonid parasitoids, Sinophorus megalodontis and an undescribed species of Olesicampe were reared from eonymphs of PFW. Morphological methods for distinguishing the immature stages of the parasitoids were developed. A predictive model for subterranean development and adult longevity of Olesicampe sp. was used to describe and to compare phenological observations from emergence traps, Malaise traps and dissections of host larvae. The effectiveness of the parasitoids as natural control agents is discussed in relation to host synchrony, encapsulation, and multi- and superparasitism."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/29016?expand=metadata"@en ; skos:note "PHENOLOGY AND BIOMETEOROLOGY OF PINE FALSE WEBWORM (HYMENOPTERA: PAMPHILIIDAE) AND ITS PARASITOIDS IN SOUTHERN ONTARIO By DONALD BARRY LYONS B . S c , C a r l e t o n U n i v e r s i t y , 1973 M.Sc, C a r l e t o n U n i v e r s i t y , 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of P l a n t Science) We accept t h i s t h e s i s as conforming to the r e q u i r e d standard THE UNIVERSITY OF BRITISH March 1988 © Donald Barry Lyons, COLUMBIA 1988 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 P l a n t Science The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 12 A p r i l 1988 DE-6G/81) ABSTRACT Models of phenology of the pine f a l s e webworm (PFW), Acantholyda e r y t h r o c e p h a l a and one of i t s p a r a s i t o i d s were developed from r e l a t i o n s h i p s between PFW s p a t i a l d i s t r i b u t i o n and microweather. Development of subterranean stages of PFW was simulated from rate-summation models developed from n o n l i n e a r r e g r e s s i o n equations d e s c r i b i n g the r e l a t i o n s h i p between temperature and rate of development of post-diapause prepupae and pupae. D e f o l i a t i o n caused by PFW i n c r e a s e d the s o i l ' s exposure to s o l a r r a d i a t i o n r e s u l t i n g i n higher s o i l temperatures and a corresponding r e d u c t i o n i n development time of subterranean stages. P r e d i c t a b i l i t y was enhanced s l i g h t l y when the d i s t r i b u t i o n of i n s e c t s and temperature of the s o i l were i n c o r p o r a t e d i n t o the model. I n c r e a s i n g the time increment used i n the model from 1 to 4 h d i d not a d v e r s e l y a f f e c t i t s r e s o l u t i o n . Mating and o v i p o s i t i o n of PFW occur w i t h i n a few hours of emerging from the s o i l and the m a j o r i t y of PFW eggs were mature and ready f o r d e p o s i t i o n at female emergence. P o t e n t i a l f e c u n d i t y of PFW was a c c u r a t e l y p r e d i c t e d from a d u l t wet and dry weights. The o v i p o s i t i o n p a t t e r n of PFW was a l s o d e s c r i b e d by a model based on temperature-dependent o v i p o s i t i o n and ageing rate f u n c t i o n s . i i The e f f e c t of l a r v a l web c o n s t r u c t i o n on the development of a r b o r e a l stages was i n v e s t i g a t e d . When exposed to s u n l i g h t , the web t r a p s heat and r a i s e s the body temperature of i t s i n h a b i t a n t s . A model was developed and used to examine the s i g n i f i c a n c e of the web m i c r o c l i m a t e f o r development of l a r v a e . R e l a t i o n s h i p s between web temperatures, canopy temperatures and standard m e t e o r o l o g i c a l methods were developed to permit u s i n g data from standard weather s t a t i o n s to d r i v e the model. L a r v a l development i n c r e a s e d by 1.4 to 2.8 d when estimated web temperatures were i n c o r p o r a t e d i n t o the model, while development was r e t a r d e d by 2.6 to 4.0 d when canopy temperatures were used i n s t e a d of m e t e o r o l o g i c a l screen temperatures. Two ichneumonid p a r a s i t o i d s , Sinophorus megalodontis and an u n d e s c r i b e d s p e c i e s of Olesicampe were reared from eonymphs of PFW. M o r p h o l o g i c a l methods f o r d i s t i n g u i s h i n g the immature stages of the p a r a s i t o i d s were developed. A p r e d i c t i v e model f o r subterranean development and a d u l t l o n g e v i t y of Olesicampe sp. was used to d e s c r i b e and to compare p h e n o l o g i c a l o b s e r v a t i o n s from emergence t r a p s , Malaise t r a p s and d i s s e c t i o n s of host l a r v a e . The e f f e c t i v e n e s s of the p a r a s i t o i d s as n a t u r a l c o n t r o l agents i s d i s c u s s e d i n r e l a t i o n to host synchrony, e n c a p s u l a t i o n , and m u l t i - and s u p e r p a r a s i t i s m . i i i TABLE OF CONTENTS s Abstract i i L i s t of tables v L i s t of figures v i i Acknowledgements x Chapter 1. General introduction 1 1.1 Biology of pine f a l s e webworm 1 1.2 Phenology and biometeorology 3 .1.3 Study areas • • • • 6 1.4 Meteorological equipment 7 Chapter 2. Subterranean development and adult emergence of pine f a l s e webworm 9 2.1 Introduction 9 2.2 Materials and methods 10 2.3 Results 23 2.4 Discussion 41 Chapter 3. Oviposition-fecundity models for pine false webworm 55 3.1 Introduction 55 3.2 Materials and methods 56 3. 3 Results 61 3.4 Discussion 80 Chapter 4. Arboreal development of pine f a l s e webworm ...... 84 4.1 Introduction 84 4.2 Materials and methods 85 4.3 Results 91 4.4 Discussion 117 Chapter 5. Biology and phenology of Sinophorus megalodontis and Olesicampe sp. (Hymenoptera: Ichneumonidae), parasitoids of pine f a l s e webworm 126 5.1 Introduction 126 5.2 Materials and methods • 127 5 . 3 Results 132 5.4 Discussion 156 Chapter 6. General conclusion 162 Lit e r a t u r e c i t e d 168 0 i v LIST OF TABLES Table 2.1. Number ,of emerging adults of PFW captured i n emergence traps i n 1983-1986 31 Table 2.2. Mean development times (days) and rates for post-diapause prepupal development of PFW reared at constant temperatures 31 Table 2.3. Mean development times (days) and rates for pupae of PFW reared at constant temperatures 32 Table 2.4. Mean development times (days) and rates for pronymph to adult development of PFW reared at constant temperatures 34 Table 2.5. Estimated parameters and goodness-of-fit s t a t i s t i c s of models for development of subterranean stages of PFW .... 40 Table 2.6. Estimated parameters and goodness-of-fit s t a t i s t i c s of development v a r i a b i l i t y functions for subterranean stages of PFW 40 Table 2.7. Deviations (dev.) i n days of observed adult emergence from simulated emergence and the deviation as a proportion (pro.) of t o t a l simulation time for adults emerging from s o i l boxes 46 Table 2.8. Deviations (dev.) i n days of observed adult emergence from simulated emergence and the deviation as a proportion (pro.) of t o t a l simulation time 46 Table 3.1. Mean wet and dry body weights, and number of mature oocytes of dissected females of PFW trapped from three plots i n 1983 and 1986 65 Table 3.2. Predicted potential fecundities from wet- and dry-weight regressions for females of PFW trapped i n LA plantation 1986 65 Table 3.3. Regression s t a t i s t i c s for relationships between number of mature oocytes and l i v e weight for PFW 70 Table 3.4. Mean longevities and fecundities of PFW females reared at constant temperatures 70 Table 4.1. Representative temperatures i n webs of PFW and associated meteorological variables 100 Table 4.2. Mean development times (days) and rates for eggs of PFW reared at constant temperatures 103 Table 4.3. Mean development times (days) and rates for larvae of v PFW reared at constant temperatures 103 Table 4.4. Mean dry weights (±SE) of larvae of PFW reared at constant temperatures : 110 Table 4.5. Deviations (dev.) i n days of observed egg hatch from simulated egg hatch and the deviation as a proportion (pro.) of t o t a l simulation time for the 1984 single-cohort simulation and the 1985 multiple-cohort simulation pooled for a l l three days 110 Table 4.6. Deviations (dev.) i n days of observed l a r v a l drop from simulated l a r v a l drop and the deviation as a proportion (pro.) of t o t a l simulation time 115 Table 5.1. Mean number of parasitoids per host (x) and percent p a r a s i t i z e d of PFW larvae c o l l e c t e d by branch sampling from v e r t i c a l s t r a t a at Anten M i l l s (AMI) i n 1983 and Lakehurst (H) i n 1986 145 Table 5.2. Mean number of parasitoids per host (x) and percent p a r a s i t i z e d of PFW larvae c o l l e c t e d by branch sampling from ca r d i n a l d i r e c t i o n s at Anten M i l l s (AMI) i n 1983 145 Table 5.3. Mean number of parasitoids per host (x) and percent p a r a s i t i z e d of PFW larvae c o l l e c t e d from drop traps at Anten M i l l s (AMI) i n 1983 147 v i LIST OF FIGURES Fig. 2.1. Cumulative emergence patterns of adults of PFW i n 1983 and 1986 24 Fi g . 2.2. Cumulative v e r t i c a l d i s t r i b u t i o n of larvae of PFW i n s o i l . Curve i s regression (Equation 2.1) of pooled cumulative frequencies as a function of s o i l depth. Symbols are observed cumulative frequencies by sex, lo c a t i o n and pooled 27 Fig. 2.3. (A) S o i l temperature p r o f i l e for 2 A p r i l 1986 i n a moderately- to heavily-defoliated zone at LA plantation. Sol i d l i n e s are estimated temperatures at indicated depths (cm) and dotted l i n e s are recorded average s o i l temperatures. (B) Cumulative degree-days at the LA plantation. S o l i d l i n e s are predicted temperatures and dotted l i n e s are observed temperatures. Lines occur i n order of depth i n s o i l 29 Fig. 2.4. Development rates of male ( s o l i d l i n e s ) and female (dashed l i n e s ) (A) post-diapause prepupae, (B) pupae and (C) pronymphs to adults of PFW at constant temperatures. C i r c l e s and t r i a n g l e s are mean observed rates for males and females, respectively. Curves are regressions from Equations 2.11a and b. Both weighted (dotted l i n e ) and unweighted (dashed l i n e ) regressions are shown for female prepupal development 36 Fi g . 2.5. Developmental v a r i a b i l i t y of male ( s o l i d l i n e s ) and female (dashed l i n e s ) (A) post-diapause prepupae, (B) pupae and (C) pronymphs to adults of PFW at constant temperatures. C i r c l e s and t r i a n g l e s are cumulative frequencies of observed i n d i v i d u a l rates/predicted rates from development rate regressions (Equation 2.11) for males and females, respectively. Curves are regressions from Equations 2.12a (B,C) and b (A) 38 Fi g . 2.6. Simulated emergence of PFW males ( s o l i d l i n e s ) and females (dashed l i n e s ) from s o i l boxes i n (A) heavily-defoliated and (B) l i g h t l y - d e f o l i a t e d zones. Symbols are observed emergence of males (tria n g l e s ) and females ( c i r c l e s ) 42 Fig. 2.7. Simulated ( l i n e s ) and observed ( c i r c l e s ) emergence of adults of PFW from LA plantation i n 1986: (A) males and (B) females from moderate- to heavily-defoliated zone and (C) males and (D) females from l i g h t l y d e f o l i a t e d zone. Consecutive l i n e s of the same type represent pupal eclosion, adult.. eclosion and adult emergence, respectively. Different l i n e types depict simulations using one depth ( s o l i d ) , multiple depths (short dash) and multiple depths with four hour increments (long dash) 44 Fig. 3.1. Cumulative ovi p o s i t i o n frequency by PFW on high and low red pine branches at LA plantation i n 1986 62 v i i F i g . 3.2. Regressions for square root of fecundities of-.group I-III females of PFW as a function of square root of the i r l i v e weight * 67 Fi g . 3.3. D i e l pattern of egg laying by PFW females at 23.4 and 14.9°C. Light and dark bands under histograms are photophases and scotophases, respectively. V e r t i c a l l i n e s are standard errors of means 71 Fi g . 3.4. Mean oviposition rate (±SE) to 50% fecundity for PFW as a function of temperature and (B) cumulative eggs/female at each constant temperature as a function of normalized time. Lines are regressions 73 Fi g . 3.5. Mean ageing rate (±SE) of PFW females as a function of temperature and (B) cumulative eggs/female at each constant temperature as a function of normalized time. Lines are regressions 75 Fi g . 3.6. (A) Predicted number of eggs (dotted l i n e ) compared with observed number of eggs ( s o l i d l i n e ) deposited by PFW i n outdoor insectary, (B) simulated (dotted l i n e ) compared with observed ( s o l i d l i n e ) number of ovipositing females and (C) simulated ovposition pattern ( s o l i d l i n e ) for PFW compared with observed o v i p o s i t i o n pattern (dotted l i n e ) of natural population and pattern of female emergence (dashed l i n e ) at Lakehurst i n 1986 78 Fig. 4.1. Frequency d i s t r i b u t i o n s of l a r v a l head capsule widths of PFW c o l l e c t e d from AM i n 1983 and 1984, and from LA i n 1986. Modes corresponding to the l a r v a l i n s t a r s are indicated by Roman numerals 92 Fi g . 4.2. Mean stage (±SE) of development of PFW from (A) three v e r t i c a l s t r a t a at the AM plantation i n 1983 and (B) three d e f o l i a t i o n zones i n the LA plantation i n 1986 94 Fi g . 4.3. Cumulative drop of ultimate instars of PFW i n the sample plots i n (A) 1983 at the Anten M i l l s plantation (plots AMI and AM2) and (B) 1986 i n H and M zones of the Lakehurst plantation 97 Fi g . 4.4. Frequency d i s t r i b u t i o n s of head-capsule widths of larvae of PFW reared at the seven constant temperatures and the alte r n a t i n g temperature regime 105 Fig. 4.5. Predicted ( l i n e s ) and observed (symbols) egg hatch of PFW for (A) single-cohort simulation and multiple-cohort simulations using (B) observed hourly a i r temperatures, (C) estimated hourly a i r temperatures and (D) observed canopy temperatures. S o l i d l i n e i n (A) i s simulation r e s u l t s using observed hourly a i r temperature and dashed l i n e i s res u l t s using estimated hourly temperatures from maximum and minimum temperatures. S o l i d l i n e s i n (B-D) are simulation r e s u l t s for v i i i i n d i v i d u a l cohorts and the dashed l i n e s are results for -population . 108 F i g . 4.6. Predicted (circles=males, triangles=females) and observed ( l i n e s ) l a r v a l drop of PFW from the H- and M-defoliated zone of LA plantation i n 1986; (A) no microclimatic corrections, (B) canopy temperature correction ( s o l i d line=males, dashed line=females), (C) l i n e a r ( s o l i d line=males, dashed line=females) and nonlinear (dotted line=males, dotted-dashed line=females) web temperature correction, and (D) canopy and web corrections ( s o l i d line=males, dashed line=females) 113 F i g . 5.1. Immature stages of PFW parasitoids: (A) larvae of &. raegalodontis emerging from the egg; (B) egg of Olesicampe sp. ; (C) f i r s t - i n s t a r larvae of Olesicampe sp.; (D) cocoons of £. megalodontis (right) and Olesicampe sp. ( l e f t ) 134 F i g . 5.2. Emergence (A, B) and Malaise (C, D) trap captures of parasitoids of PFW from the LA plantation i n 1986 137 Fi g . 5.3. Mean numbers (±SE) of p a r a s i t o i d eggs (A, B) and encapsulated larvae (C, D) per PFW l a r v a l i n s t a r c o l l e c t e d by branch sampling at Anten M i l l s i n 1983 (A, C) and Lakehurst i n 1986 (B, D) 140 Fi g . 5.4. Mean numbers (±SE) of t o t a l immature parasitoids, p a r a s i t o i d eggs, and encapsulated p a r a s i t o i d larvae per PFW larvae c o l l e c t e d by branch sampling at Anten M i l l s i n 1983 (A, C) and Lakehurst i n 1986 (B, D). The numbers over the symbols are number of host larvae 143 Fi g . 5.5. Post-diapause temperature-dependent development rates for (A) within host and (B) within cocoon development of Olesicampe sp. C i r c l e s are i n d i v i d u a l observed rates and l i n e s are regressions curves (Equation 2.11b) 150 Fi g . 5.6. (A) Temperature-dependent ageing rate of Olesicampe adult females as a funtion of temperature and (B) cumulative d i s t r i b u t i o n function of normalized death times. C i r c l e s are (A) observed mean ageing rates (±se) and (B) cumulative mortality at each normalized time. Regression l i n e s are indicated 152 Fig. 5.7. Simulated development of subterranean stages of Olesicampe sp. from moderately- (A) and l i g h t l y - d e f o l i a t e d (B) plo t s of LA plantation i n 1986. S o l i d l i n e s (A, B) represent r e s u l t s using temperatures from one depth while the dashed l i n e s represent summed development of insects at several depths i n s o i l . F i r s t curves are emergence of p a r a s i t o i d larvae from hosts and second curves are emergence of adults from s o i l . C i r c l e s are observed emergence of adults from s o i l . (C) Simulated r e l a t i v e abundance ( s o l i d l i n e ) of females and Malaise trap captures scaled to size (dotted l i n e ) 154 i x ACKNOWLEDGEMENTS I would sincerely l i k e to thank Dr. W.G. Wellington for his supervision during the course of these investigations. I'am also indebted to Drs. B.D. Frazer, C.L. Gass, and J.H. Myers for he l p f u l and informative discussions during my stay at U.B.C. The present work could not have been completed without the assistance and support of the s t a f f of the Great Lakes Forestry Centre, Canadian Forestry Service. Dr. C.R. Sullivan provided the impetus and Drs. T.J. Lysyk, V.G. Nealis, J. Regniere, and D.R. Wallace enhanced my understanding of insect ecology. Expert technical assistance throughout the work was provided by S. Fera. In addition, seasonal assistance was provided by M. Bishop, H. Devon, M. Finn, M. Holmes, C. Kirchmeir, R. Wagner, N. Watson, and R. Whitcroft. Staff of the Forest Insect and Disease Survey helped me locate populations of the pine fal s e webworm. I would also l i k e to thank Mrs. B. McMaster and Mr. G. Dunsmore for allowing me to conduct research on th e i r properties and the Ministry of Natural Resources, Midhurst D i s t r i c t , for permission to conduct work on Crown land. The following i n d i v i d u a l s graciously examined and i d e n t i f i e d p a r a s i t o i d specimens: G.E. Bisdee and M. Sanborne (Biosystematics Research Centre, Ottawa, Ontario), M. F i t t o n ( B r i t i s h Museum of Natural History, London), R. Hinz (Einbeck, Germany), H. Townes (American Entomological In s t i t u t e , G a i n e s v i l l e , F l o r i d a ) . I would e s p e c i a l l y l i k e to thank my wife Martha for her support and understanding. x 1 Chapter 1 General Introduction 1.1 Biology of Pine False Webworm eryth.ro- 26. T. antennis setaceis, corpore caeruleo, capite cephala. rubro. Uddm. dis s . 89. Habitat i n Pino. Mas totus ater. ore t i b i i s q u e a n t i c i s l u t e i s . Linnaei (1758) The pine f a l s e webworm (PFW), Acantholyda erythrocephala (L.) (Hymenoptera: Pamphiliidae), i s endemic to the Palearctic Region, occurring i n Great B r i t a i n , central and northern Europe to Lapland, and east to the Caucasus and western Siberia, Korea and Japan (Middlekauff 1958, Charles and Chevin 1977). This species was f i r s t reported i n North America i n 1925 from Pennsylvania (Wells 1926), and has since been reported i n New Jersey (Soraci 1938), New York (Middlekauff 1938), Connecticut (Plumb 1945) and the Lake States (Wilson 1977). Larval PFW were f i r s t reported i n Ontario i n 1961 i n Scarboro Township (Eidt and McPhee 1963). PFW now occurs throughout most of southern Ontario and i n the Lake-of-the-Woods region i n northwestern Ontario (Syme 1981). The host plants of PFW are nine species of Pinus. Pinus resinosa A i t . (red pine) and P. strobus L. (white pine) are the most favored and P. nigra Arnold (Austrian pine) the least favored i n North America (Middlekauff 1958). Pinus s y l v e s t r i s L. (Scots pine) i s preferred to P. nigra i n mixed stands i n .Europe (Jahn 1967, Schmutzenhofer 1975). Adult males precede females, i n spring, from earthen c e l l s i n the s o i l (Schwerdtfeger 1944). Females mate soon after emergence. U n f e r t i l i z e d females can deposit viable eggs that develop into males. Eggs are l a i d i n contiguous rows on the flattened surface of one year-old pine needles (Griswold 1939). A s l i t i s cut i n the needle by the ovipositor into which a crease of the chorion i s inserted (Middlekauff 1958). Eggs obtain moisture from the vascular system of the plant (Schwerdtfeger 1941). Newly hatched larvae descend needles to the twigs where they spin s i l k around the base of the needles and feed gregariously on the sides of needles above the f a s c i c l e . While maturing, larvae enlarge t h e i r web and feed on the basal ends of needles that they p u l l towards themselves (Griswold 1939). Accumulation of s i l k , frass, exuviae and uneaten needles forms a web enclosing the larvae. One to 35 (average 6 to 10) larvae . inhabit s i l k e n tubes within t h i s web (Schwerdtfeger 1944). Male larvae pass through f i v e instars and females have six during the course of t h e i r arboreal development (Schwerdtfeger 1941). Larvae f a l l to the ground and burrow into the s o i l upon completion of feeding i n early summer. These larvae or eonymphs, mold earthen c e l l s and enter summer diapause. Most eonymphs emerge from diapause and transform i n autumn to 3 -=pronymphs which have c h a r a c t e r i s t i c pupal eyes v i s i b l e through the l a r v a l skin (Schwerdtfeger 1941). A portion of eonymphs may remain i n diapause for several years (Schwerdtfeger 1944). PFW i s e s s e n t i a l l y univoltine (Griswold 1939), pronymphs and less commonly eonymphs i n prolonged diapause are the overwintering stages. Pupation occurs i n spring when s o i l temperatures increase. 1.2 Phenology and Biometeorology Phenology i s the study of the e f f e c t s of climate and season on natural events (Stedinger e_t a l . 1985). Insect phenology concerns the physiological processes governing development, dormancy, reproduction and ageing (Tauber and Tauber 1973). The temporal d i s t r i b u t i o n of the insect's stages throughout the season determines the degree of synchrony of various stages with food resources and mortality agents such as parasitoids, predators and pathogens (Willmer 1982). Although other variables, such as n u t r i t i o n , moisture and photoperiod can affect development, the primary dri v i n g force i s temperature (Curry e_t a l . 1978, Baker 1980). Oviposition, fecundity, longevity (Hogg and Gutierrez 1980, Mason and Mack 1984) and processes at the population l e v e l (e.g., mating, migration) are also influenced by temperature. A phenological model i s a description of the temporal progression of events i n the l i f e of insects as a function of environmental conditions. Phenological models based only on temperature provide 4 s i g n i f i c a n t predictive power. An understanding of phenology can provide information about species d i s t r i b u t i o n (Tauber and Tauber 1973) and the more complex processes involved i n population dynamics, for example the benefits that occur when reproduction and development are synchronized with favorable conditions. A b i l i t y to predict phenological events for pests allows precise timing of the application of control agents, be they chemical or b i o l o g i c a l . This i s e s p e c i a l l y c r i t i c a l when the duration of susceptible stages i s b r i e f (de Reede and de Wilde 1986). Timing of surveys to determine abundance also requires a knowledge of when the developmental stages are present. Economic losses can be forecast i f the onset of the destructive stage can be predicted (Dennis e_t a l . 1986). Process-oriented models are e s s e n t i a l l y hypotheses describing the predicted behavior of a system. Tests of these hypotheses include the degree to which the models mimic natural events under f i e l d conditions. A phenology model i s not only a t o o l to predict seasonal a c t i v i t y of an insect population, but also provides a framework within which questions can be addressed about underlying mechanisms a f f e c t i n g that population and gaps i n knowledge can be i d e n t i f i e d (Regniere 1982). The major impediment to understanding the e f f e c t s of weather and temperature on insects i s the d i f f i c u l t y i n r e l a t i n g 5 macroel-imatic variables to the microclimate that impinges on the insect's habitat (Wellington 1957, Gilbert et a l . 1976, Wellington and Trimble 1984). Although microclimatic temperatures have been measured, few studies have done so for a long enough period that meaningful comparisons may be made with meteorological standards (Baker 1981). Two possible approaches to r e l a t e Stevenson screen data and microclimate are: make measurements of the microclimate under various macroclimatic conditions and determine the thermal requirements for development of the insect i n the laboratory and use the results to compare predicted development from Stevenson screen data with actual development of the insect i n the f i e l d (Morris and Fulton 1970). The differences between observed and predicted events are then attributed to differences between macroclimate and microclimate. The f i r s t approach was used with European pine shoot moth, Rhyacionia buoliana (Green 1968). Temperatures of the insects' microhabitats were recorded and used to construct models which relate infested bud temperatures to Stevenson screen temperatures. Wellington (1954) pointed out the value of shelter builders for studying the ef f e c t s of weather on insect populations. Larvae of PFW l i v e inside shelters i n the canopy and eggs occur on the surface of the fo l i a g e . Eonymphs, pronymphs and pupae l i v e i n the s o i l , thus, PFW inhabits two d i f f e r e n t habitats. The adults, at d i f f e r e n t times i n t h e i r l i v e s move between both habitats. Its shelter-building behavior, abundance i n southern 6 Ontario-,<-and i t s variety of habitats make PFW an excellent species for the study of the e f f e c t s of microclimate on phenology. The central theme of t h i s thesis i s to determine whether empirical relationships between the thermal conditions i n the microhabitat of the insect and conventional meteorological measurements enhance the p r e d i c t a b i l i t y of phenology models. The present study involves the development of phenology models for PFW. Individual Chapters deal with the development of models for successive phenophases of PFW. Chapter 2 covers the period from spring thaw u n t i l adult emergence. Chapter 3 examines oviposition and fecundity i n an attempt to predict the temporal d i s t r i b u t i o n of egg laying. Chapter 4 investigates the development of the resulting arboreal stages of PFW and the role that microclimate plays i n t h i s development. In Chapter 5, the biology and phenology of two parasitoids of PFW are explored to understand the e f f e c t s that i n t e r s p e c i f i c synchrony have on population dynamics. Factors inducing and maintaining diapause states were not evaluated i n t h i s investigation. 1.3 Study Areas F i e l d investigations were conducted i n red pine plantations i n south central Ontario. In 1983-1985, studies were conducted i n two s i t e s i n Simcoe County. One plantation, planted i n 1969, was 1.2 km North of Anten M i l l s and another was 3.8 km East of 7 Craighurst<--. Two one-hectare plots (100 X 100 m) i n the Anten M i l l s plantation (AMI and AM2) and one plot i n the Craighurst plantation (CR) were used. Each plot was marked off i n a grid of 10-m i n t e r v a l s . During 1985-86 the .study was conducted i n a plantation (planted i n 1970-72) 3.5 km NNW of Lakehurst (LA) i n Peterborough County. S o i l s i n a l l locations were loamy sands to sandy loams. The s o i l i n the LA plantation was underlaid with calcareous rocks. 1.4 Meteorological Equipment During 1983-85, standard aspirated Stevenson Screens (1.3 m above ground) [Atmospheric and Environment Service (AES), Environment Canada] were i n s t a l l e d i n clearings at the AM and CR plots, each with a maximum and minimum thermometer and a standard r a i n gauge (AES). Daily maximum and minimum temperatures and r a i n f a l l were recorded at about 0900 hours. An e l e c t r o n i c weather station and ele c t r o n i c data loggers (Campbell S c i e n t i f i c Inc. (CS), Logan, Utah) were used i n conjunction with the Stevenson Screen at the AM loc a t i o n i n 1984-85. The el e c t r o n i c s t a t i o n consisted of a tri p o d tower (CS, Model CM10K) to which meteorological sensors were fixed. A combined temperature and humidity probe (CS, Model 207) was mounted on the tower i n a ven t i l a t e d screen 1.5 m above the ground. Global solar radiation was measured using a pyranometer (Li-Cor Inc., Lincoln Nebraska, Model LI-200S) with cosine correction. P r e c i p i t a t i o n was measured with a Sierra tipping 8 bucket r a i n gauge (CS, Model RG2501). These sensors were monitored by a CR-7 (CS) data logger programmed to sample the sensors every 5 seconds and output the averages, maxima, minima or t o t a l s on the hour. At the LA plantation, only the e l e c t r o n i c weather station was used i n 1986. A second CR-7 data logger was added i n 1985-86 allowing comparisons to be made between d i f f e r e n t s i t e s within the plantations. Copper-constantan thermocouples were used to measure the thermal microclimate of the insects' habitats within the plantations. The gauge of the wire depended on the application. Wire diameter used was smaller when di r e c t solar radiation was greatest. Thermocouples were calibrated i n an i c e bath with a mercury-in-glass thermometer. 9 Chapter 2 Subterranean development and adult emergence of pine f a l s e webworm. 2.1 Introduction PFW overwinters i n the s o i l as'pronymphs inside c e l l s formed i n l a t e summer by body movements of eonymphs. As the s o i l warms i n spring, overwintered pronymphs transform to pupae i n the c e l l s and pupae i n turn emerge as adults. The duration of the pupal stage i s reported to l a s t from 15-18 days (Prozorov 1925, Middlekauff 1958) and i s influenced by temperature (Schwerdtfeger 1941). Peak emergence of males from the s o i l precedes (protandry) that of females (Schwerdtfeger 1944). The date of adult emergence depends on l o c a l weather. In France, adults appear i n A p r i l and occasionally i n the l a s t days of March (Charles and Chevin 1977). In western Siberia, adults are f i r s t observed at the end of May (Prozorov 1925). Adult emergence, i n New York, occurs from mid-April u n t i l mid-May (Middlekauff 1958), si m i l a r to the emergence period i n Germany (Schwerdtfeger 1941) and Ontario (Syme 1981). Development of subterranean insects i s regulated by the thermal environment of the s o i l (Morse e_t a l . 1985). S o i l temperatures have temporal and s p a t i a l v a r i a t i o n . It i s e s s e n t i a l to know the depth i n the s o i l at which an insect dwells as temperature varies with s o i l depth (Logan e_t a l . 1979, 10 Willmer 1982). Horizontal d i s t r i b u t i o n of s o i l insects also influences developmental processes ( G r i f f i t h s 1959) through changes i n exposure to solar r a d i a t i o n (Wallace and S u l l i v a n 1963). Emergence patterns of insects may also be affected by var i a t i o n s i n temperature (Tatar 1984). While functional advantages of protandrous emergence have been suggested, l i t t l e information exists as to how the pattern i s achieved (Lederhouse et a l . 1982). Protandry may result from sex-related differences i n developmental rates, s p a t i a l d i s t r i b u t i o n or other factors. The objectives of t h i s phase of the study were to examine emergence patterns of PFW i n the f i e l d ; determine the s p a t i a l d i s t r i b u t i o n of the subterranean stages of PFW; determine the v a r i a b i l i t y of the s o i l thermal microclimate and r e l a t e i t to the d i s t r i b u t i o n of PFW i n the s o i l ; determine the r e l a t i o n s h i p between temperature and development of the subterranean stages of PFW and examine the processes and s t a b i l i t y of the emergence pattern using computer simulations. 2.2 Materials and Methods Adult Emergence. Adult emergence patterns were determined using emergence traps. Traps were made from aluminum window screening shaped into cones wired at the bottom to s t e e l hoops (0.25 m2 basal area). Emerging adults climbed through a small hole at the top of the cone and were captured i n c o l l e c t i n g 11 b o t t l e s containing 70% ethanol. Emergence was monitored during 1983-85 with 50 traps placed i n each of the three sample grids (Section 1.3), according to computer-generated random coordinates. During 1986, 33 emergence traps were placed i n each of three PFW-defoliation zones within the LA plantation. D e f o l i a t i o n zones were subjectively i d e n t i f i e d as heavy (H), moderate (M) and l i g h t (L). Traps were placed at 5-m i n t e r v a l s i n three rows 5 m apart. Each trap was held down with three wire tent pegs. S o i l and l i t t e r was forced up against the trap to prevent escape of insects. Adults were removed from c o l l e c t i n g b o t t l e s and from the i n t e r i o r of traps each day at about 1000 hours. V e r t i c a l D i s t r i b u t i o n of PFW i n S o i l . The v e r t i c a l d i s t r i b u t i o n of PFW eonymphs i n the s o i l at the LA plantation was determined i n August 1984. S o i l l i t t e r was scraped away and a core 15 cm deep was extracted from the s o i l using a core sampler. The distance of each eonymph from the s o i l surface was measured to the nearest cm. Eonymphs were preserved i n 70% ethanol for l a t e r sex determination based on head capsule s i z e . Head capsules were measured using a c a l i b r a t e d ocular micrometer mounted on a stereo dissecting microscope. S o i l d i s t r i b u t i o n of PFW i n the AM plantation was compared with the d i s t r i b u t i o n i n the LA plantation i n the following manner. Branches with larvae were c o l l e c t e d at the Lakehurst 12 plantation i n 1985 and kept i n w a t e r - f i l l e d jars inside 40 by 40 by 40 cm screen cages on the s o i l surface at one of the Anten M i l l s plots (AMI). Tops of the cages were covered with chicken wire and the sides were buried i n the ground to prevent l a r v a l escape. After larvae completed development, they dropped to the ground and burrowed into the s o i l . S o i l i n the enclosures was dug up i n October and the v e r t i c a l d i s t r i b u t i o n of larvae was determined. Pronymphs with larger head capsules were females, while smaller i n d i v i d u a l s were males (Schwerdtfeger 1941). The mean v e r t i c a l d i s t r i b u t i o n s of each sex i n the s o i l were compared. The v e r t i c a l d i s t r i b u t i o n of larvae i n the s o i l was described by: V(z) = (1 - a)qa w (2.1) where a = (ZL - Z)/(ZL - zrj) V(z)=cumulative proportion of i n d i v i d u a l s and z=soil depth. Depth was normalized between lower (ZL) and upper (zrj) l i m i t s . Parameters q and w were estimated by nonlinear regression. S o i l Temperature. S o i l temperature was measured using 24-AWG copper-constantan thermocouple wire inserted through lengths of Tygon tubing. Thermocouple junctions were sealed to the end of tubes with epoxy r e s i n . Square holes were dug i n the s o i l to a depth of 20 cm. Thermocouples were placed on one side of each hole at 5 and 10 cm depths. Soil.and humus were replaced and the thermocouples were wired as d i f f e r e n t i a l pairs 13 to data loggers. Thermocouples were i n s t a l l e d i n autumn to allow s o i l to s e t t l e before temperature recordings were made the following spring. Ten pairs of thermocouples were placed i n a moderate to heavy de f o l i a t e d zone and ten pairs were i n s t a l l e d i n a l i g h t l y d e f o l i a t e d zone at the LA plantation. Data loggers were used to monitor thermocouples every 5 s and calculate hourly average temperatures. Temperature recording began i n winter p r i o r to the disappearance of the snow cover while the ground was s t i l l frozen. S o i l temperatures at multiple depths were estimated from temperatures at the 5 and 10 cm depths incorporating temporal and s p a t i a l v a r i a b i l i t y , a correction for asymmetry of the wave and a correction for phase lag r e s u l t i n g from the slower heating of the s o i l with increased depth. At each f i e l d location, the hourly average temperatures recorded by i n d i v i d u a l thermocouples were averaged at each depth ( i . e . , 5 and 10 cm). These average temperatures were used to estimate temperatures at other depths within the v e r t i c a l d i s t r i b u t i o n of PFW i n the s o i l . The hourly temperature d i f f e r e n t i a l (D z) between the temperature at a depth of 10 cm and any other depth z was calculated (Regniere e_t a l . 1981) using: D 2 \" 14 D e l t a z Cos{180[l + (t + 5)/16]} when H t i 10 D e l t a z Cos[180(t - l l ) / 8 ] (2.2) when 10 < t < 20 D e l t a z Cos{180[l + (t - 19)/16]} when 20 ^ t ^ 24 where D e l t a z = (R z - R v)/2, t i s time of day, Ry i s the d a i l y temperature range at 10 cm, and R z i s the d a i l y temperature range at depth z. R z was calculated as a function of s o i l depth using: R z = R y/exp[k(z - y)] (2.3) where y i s 10 cm and k i s the r e c i p r o c a l of the damping depth and i s calculated d a i l y (Rosenberg 1974) from: ln(R v/R x) k = ± (2.4) x - y where R x i s the temperature range at 5 cm and x i s 5 cm. The preceding algorithm assumes the temperature wave at the unknown depth i s symmetrically d i s t r i b u t e d around the reference temperature wave. However, temperature waves at the two observed depths were seldom symmetrical. The difference between the temperature maxima (Tmaxx - Tmaxy) was usually greater than the difference between minima (Tmin x - Tminy). If t h i s r e l a t i o n s h i p holds true for a l l depths, the proportion (PD) of the temperature range at the unknown depth that occurs above the reference wave i s computed d a i l y from: (Tmaxx - Tmaxy) PD = . 1 : (2.5) (Tmin x - Tminy) + (Tmaxx - Tmaxy) and a corrected temperature difference ( D c o r r ) i s determined from: D c o r r = D z + D e l t a z - PD (R z - R y) (2.6) The temperature at any depth for any hour of the day i s estimated from: T 2 ( t ) = T y ( t ) - D c o r r (2.7) The phase lag (ty - t z ) i n temperature that occurs with increasing depth for the predicted and reference temperatures was determined from: t y - t z = ((z-y)/2)(p/(7ra)) 0- 5 (2.8) where p i s the period of the wave i n seconds ( i . e . , 8.64 X 10 4 s) and the thermal d i f f u s i v i t y a = 7r/k2p (Rosenberg 1974). Temperature-dependent Development. Duration of the prepupal period ( i . e . , time for termination of diapause and morphogenesis to pupa) was determined using pronymphs c o l l e c t e d i n l a t e October and early November. Pronymphs were placed i n damp vermiculite i n p l a s t i c boxes and gradually acclimatized to a -0.5°C storage temperature. Individuals were removed from storage a f t e r about three months and placed i n 1.9-mL s h e l l v i a l s stoppered with foam plugs. The v i a l s were placed on t h e i r sides i n p l a s t i c boxes and covered with moist paper towels. Boxes were put into paper bags and placed i n one of a series of constant-temperature chambers set at 1.7, 5.8, 7.5, 10.7, 15.0, 19.1, 22.4, and 23.0°C. Another group of spring c o l l e c t e d pronymphs was reared at 4.1°C. Preliminary investigations indicated that pupation did not occur at temperatures greater than 23.0°C. Pronymphs were examined twice d a i l y u n t i l pupal 16 eclosion. Pupae emerging from pronymphs described above were used to determine the duration of pupal development at constant temperatures. Pupae emerging from 15°C were d i s t r i b u t e d among 11 constant temperature chambers ranging from 2.3 to 31.7°C, while pupae emerging at other temperatures were l e f t i n the chambers i n which they eclosed. One group of insects was alternated d a i l y between 7.4 and 15°C chambers u n t i l adult eclosion to compare t h e i r developmental periods with pupae kept throughout the stage at 7.4 and 15°C. Pupae were examined twice d a i l y u n t i l adult emergence. Sex of pupae was determined either by examination of emerged adults or from the shape of terminal abdominal segments of pupae (Kolomietz 1967). Adults produced at 7.4°C were weighed to determine the r e l a t i o n s h i p between size and developmental rate. Total duration of subterranean development ( i . e . , pronymph to adult) was determined i n a si m i l a r manner. Laboratory overwintered pronymphs were placed i n s h e l l v i a l s on t h e i r sides i n transparent p l a s t i c boxes. V i a l s were covered with damp vermiculite and boxes were placed at 12 constant temperatures ranging from 1.8 to 31.0°C. Boxes at 1.8°C were moved to 15°C af t e r 90 days to speed up development. Boxes were examined twice d a i l y and newly emerged adults were removed. Data from indivi d u a l s moved between low and high 1? temperatures were used to estimate time required to complete development at the low temperature using a i n t e r p o l a t i o n technique. B r i e f l y , the method used was: t L = t L / [1 - ( t H / t H ) ] (2.10) where i s estimated time an i n d i v i d u a l requires to complete development at low temperature, t-^ i s actual time spent at low temperature, i s average time required to complete development at high temperature (determined from i n d i v i d u a l s reared at high temperature for entire stage) and t H i s time that indi v i d u a l s a c t u a l l y spend at high temperature (Regniere 1987). Individual development times were converted to development rates (1/time) and mean development rates (b m) were computed for each experimental temperature (T). Matched-asymptote equations (Logan e_t a l . 1976) were used to describe the re l a t i o n s h i p between mean developmental rates and temperature. Parameters were estimated using program PAR of BMDP (Dixon 1983). Equation 2.11a i s an exponential function coupled with a decay function, while Equation 2.11b i s a l o g i s t i c function combined with a decay function. The choice of equation was determined graphically by the shape of the re l a t i o n s h i p near the optimal temperature (Regniere e_t a l . 1981). The equations are: b m(T) = ? 1 {exp(P 2r) - exp[P 2 - (1 - r)/P 3]} (2.11a) and bm = ?1 {[1 + exp(P 2 - P 3 r ) ] _ 1 - exp[(r - 1)/P 4]} (2.11b) where T = (T - T b) / ( T m - T b) T m and T]-, are maximum and minimum temperatures above and below 18 which development did not occur, and P 1 - P 4 are parameters estimated by nonlinear regression. V a r i a b i l i t y i n development rates for a l l temperatures can be mathematically described i f the d i s t r i b u t i o n of development rates can be normalized into standard units. The method developed by Regniere (1984) was used to describe developmental v a r i a b i l i t y of a l l subterranean stages of PFW. This requires s e l e c t i n g an appropriate normalizing constant, i n t h i s case, the estimated mean development rate calculated using Equation 2.11. Individual rates at each constant temperature were divided by the temperature-dependent solution of Equation 2.11. Normalized data from each temperature were pooled and a cumulative d i s t r i b u t i o n was constructed. A l o g i s t i c function: Y = [1 + exp{-K(X-C)}]~ 1 / (? (2.12a) or a Poisson function: Y = t l - expC-KX)] 1 7^ (2.12b) was f i t t e d to the cumulative frequency d i s t r i b u t i o n using nonlinear regression. Parameters K, Q and C i n Equation 2.12a describe the slope, skew and p o s i t i o n of the sigmoid curve. Parameters K and Q behave s i m i l a r l y i n Equation 2.12b, a function better suited for describing sigmoid d i s t r i b u t i o n s with extreme po s i t i v e skews. The inverted forms of these equations are X = C - l n ( Y \" Q - 1)/K (2.13a) and. : X = - l n ( l - Y Q ) / K (2.13b) 19 for the l o g i s t i c and Poisson curves, respectively. These describe the v a r i a b i l i t y factor X of the development rate for the Yth percentile of the population. Pronymphs coll e c t e d i n the f a l l and overwintered i n the laboratory were reared outdoors at the LA plantation inside 37 by 19 by 14.5 cm high p l a s t i c boxes p a r t i a l l y f i l l e d with s o i l . This emulated subterranean development i n natural conditions but eliminated s p a t i a l v a r i a b i l i t y . A 29 by 15.5 cm hole was cut i n the top of each box and covered with fine-mesh p l a s t i c screening. Drain holes were d r i l l e d i n the bottoms of the boxes. Thermocouples were glued to the center of the bottoms of the boxes. F i f t y s h e l l v i a l s each containing one l a r v a were taped on t h e i r sides to the bottoms of six boxes and covered with s o i l to a depth of 11 cm. Three boxes were placed i n a heavily d e f o l i a t e d zone and three were placed i n a l i g h t l y d e f o l i a t e d zone. Each box was buried u n t i l the s o i l l e v e l i n the box was even with the surrounding s o i l . Thermocouples were connected to data loggers programmed to record hourly average temperatures. Boxes were examined d a i l y for adult emergence. A rate-summation model to predict the emergence of the adults from the s o i l boxes was constructed based on the FORTRAN program provided by Regniere (1984). Phenology Model Construction. The modeling approach used i s based on the assumption that development i s an additive process (Logan e_t a l . 1976). For each d i s c r e t e time step (At), the 20 proportion of development completed was multi p l i e d by the time step. For a l l subterranean stages of PFW, At = 0.0417 d. To incorporate developmental v a r i a b i l i t y , the hourly development rate was multiplied by factor Xy (from Equation 2.13) for each percentile (Y) of the population and development was accumulated u n t i l development reached unity (Regniere 1984). Thus, development of a percentile of the population (Py(t)) at time t was approximated by the difference equation P Y ( t ) = P Y ( t - At) + b m(T) At Xy (2.14) The proportion of the population that had completed development at time t was then: 100 P(t) = [ ^ P Y ( t ) ] / 100 (2.15) Y = 1 Development of PFW i n s o i l c e l l s following diapause termination consisted of prepupal and pupal phenophases. Subterranean development of PFW culminated i n the emergence of adults from the s o i l . The model assumes that a l l overwintered pronymphs are at the same stage of development when s o i l temperatures rose above freezing. Development of each sex was computed independently and simultaneously since male and female subterranean stages of PFW have d i f f e r e n t temperature-dependent development rate and v a r i a b i l i t y functions. The number of developing cohorts (c) of PFW simulated during each stage of the multiple-cohort subterranean phenology model was determined by the number of days i t took for the population to complete the previous phenophase (Wagner ejt a l . 1985). The l i m i t of a class i n t e r v a l for the d i s t r i b u t i o n of cohorts was set at 1000 hours since emergence traps were examined at t h i s time. The f i r s t cohort (c = 1) began development at the end of the day that the f i r s t i n d i v i d u a l s had completed the previous phenophase. The l a s t cohort (c = n) began development when the l a s t of the population had completed the previous phenophase. Each d a i l y cohort represented the proportion of the population beginning the phenophase (Pb c) by time t d which was incremented i n d a i l y steps. P ( t d ) was derived from Equation 2.15, and: Pb c(At d) = P ( t d ) - P ( t d - A t d ) (2.16) n and The proportion of the cohort ( P f c ) f i n i s h i n g the phenophase by time t was also computed from the rate-summation model (Equation 2.14). Each cohort simulation resulted i n a separate cumulative frequency d i s t r i b u t i o n of the proportion completing the phenophase. The cumulative proportion of the population (Pp(t)) that had completed development was the sum of the cohort d i s t r i b u t i o n s weighted by the proportion of the population i n the cohort. When Pp(t) = 1.0 the population has completed the phenophase. The model was written i n FORTRAN. Input for the model consisted of hourly average s o i l temperatures from the heavy to moderate-defoliation zone and from the l i g h t - d e f o l i a t i o n zone of the LA plantation i n 1986. S o i l temperature f i l e s started 22 March when the ground was s t i l l frozen. To determine i f v e r t i c a l d i s t r i b u t i o n of the s o i l c e l l s influenced emergence pattern of PFW, the model was modified to incorporate t h i s element of s p a t i a l v a r i a b i l i t y . The proportion [p(Az)] of insects i n 1.5 cm increments ( i . e . , Az) of the s o i l from the surface to the bottom of the lowest i n t e r v a l containing larvae was determined from the depth-dependent solution to Equation 2.1 where: For simulations involving microclimatic s t r a t i f i c a t i o n of development ( i . e . , depth i n s o i l ) , s o i l temperature was calculated at the midpoint of each v e r t i c a l stratum. Hourly temperature at six depths was used as the input for the model. The rate-summation model was then calculated for each stratum (s) and developmental status of the population (P(t,z)) was determined from: s = 1 where Ps i s the proportion of PFW i n the stratum that had completed development. p(Az) = V(z) - V(z - Az) (2.18) n (2.19) 2.3 Results 23 Adult Emergence. Peak female emergence occurred aft e r peak of male emergence i n most situations (Fig. 2.1). Onset of emergence was similar from 1983-1985, although, the pattern was obscured by low adult numbers during 1984 and 1985. Onset of male emergence i n 1984-85 i n the three plots ranged from 12 May (day 133) to 17 May (day 137), while the onset of female emergence ranged from 12 May to 19 May (day 140). Duration of emergence was 4 to 26 days i n the same years. The emergence pattern was s i m i l a r i n 1986, but occurred e a r l i e r as a res u l t of d e f o l i a t i o n of the pine canopy and a warmer spring ( Fig. 2.1). Observed onset of emergence i n the heavy d e f o l i a t i o n zone preceded the onset of emergence i n the moderate d e f o l i a t i o n zone by two days, which i n turn preceded emergence from the l i g h t d e f o l i a t i o n zone by seven days. Throughout the LA plantation the emergence period ranged from 27 A p r i l to 26 May. Female percentage of emerging adults from a l l locations (Table 2.1) varied from 16.7 to 56.9. V e r t i c a l D i s t r i b u t i o n of PFW i n S o i l . Differences between the mean depth of male and female eonymphs were always le s s than the v e r t i c a l sampling i n t e r v a l (1 cm) and not s t a t i s t i c a l l y s i g n i f i c a n t , therefore frequencies of PFW of both sexes at each v e r t i c a l depth increment were pooled. Si m i l a r l y , the difference between mean depths of eonymphs at each lo c a t i o n was l e s s than the sampling i n t e r v a l , so the data were again pooled. Since the 24 Fig. 2 .1 . Cumulative emergence patterns of adults of PFW i n 1983 and 1986. 25 120 130 140 150 160 170 2 LU > < 115 120 125 130 135 140 145 150 DAY OF YEAR 26 upper l i m i t of the d i s t r i b u t i o n was the s o i l surface, ZJJ = 0. The lower l i m i t was the maximum depth of the sample that contained larvae (9 cm) (Fig. 2.2). For these data, q = 3.484 and w = 2.705 (Equation 2.1) provided an equation which had excellent f i t (R 2 = 0.999). S o i l Temperature. Maximum observed s o i l temperatures never exceeded 17°C pr i o r to the completion of adult emergence at a l l locations. A simulated s o i l temperature p r o f i l e i s shown i n Fi g . 2.3A. Cumulative degree-days (> 0°C) at each observed and predicted depth i n the s o i l are shown i n Fig. 2.3B. Temperature-dependent Development. No pupa emerged at temperatures greater than 23.0°C. Mean prepupal development period was longer for females than for males at a l l but the lowest temperature (Table 2.2). Lack of an observable difference at 1.7°C r e f l e c t e d the high v a r i a b i l i t y at t h i s temperature. Sim i l a r l y , female pupal development took longer than male development at a l l but the lowest temperature (Table 2.3). Pupae from prepupae reared at 15°C and subsequently kept at another temperature during pupal development had a consistently shorter developmental period than i n d i v i d u a l s that experienced a lower temperature as prepupae (Table 2.3). For pupae reared at 7.4°C, there was no c o r r e l a t i o n between development times and female l i v e weights (r=-0.005, df=55, P=0.968). There was a s i g n i f i c a n t negative c o r r e l a t i o n (r=-0.377, df=50, P=0.006) between these variables for males. 27 F i g . 2.2. Cumulative v e r t i c a l d i s t r i b u t i o n of larvae of PFW i n s o i l . Curve i s regression (Equation 2.1) of pooled cumulative frequencies as a function of s o i l depth. Symbols are observed cumulative frequencies by sex, l o c a t i o n and pooled. 28 SOIL DEPTH (cm) 29 F i g . 2 .3 . (A) S o i l temperature p r o f i l e for 2 A p r i l 1986 i n a moderately- to heavily-defoliated zone at LA plantation. S o l i d l i n e s are estimated temperatures at indicated depths (cm) and dotted l i n e s are recorded average s o i l temperatures. (B) Cumulative degree-days (>0°C) at LA plantation. S o l i d l i n e s are predicted temperatures and dotted l i n e s are observed temperatures. Lines occur i n order of depth i n s o i l . Table 2 .1 . Number of emerging adults of PFW captured i n emergence traps i n 1983-1986. Year Plot Number of adults % Female 1983 AMI 1882 45. 6 a AM2 537 44. 5 a CR 355 50.7 1984 AMI 153 22. 9 a AM2 98 32. 7 a CR 65 55.4 1985 AMI 26 53.8 AM2 18 16. 7 a CR 21 28.6 a 1986 LA(H) 832 52.0 LA(M) 2026 40. l a LA(L) 262 56. 9 a a s i g n i f i c a n t l y d i f f e r e n t from 50% (x 2; P < 0.05) Table 2.2. Mean development times (days) and rates for post-diapause prepupal development of PFW reared at constant temperatures. Temp. (°c5 Sex n Mean time (d) SE Mean rate (l/d) SE 1.7 M 40 27.9 2.2 0. 053 0. .007 F 64 27.8 1.6 0. 047 0. .004 5.8 M 16 9.2 0.8 0. 121 0. .011 F 20 10.9 0.9 0. 106 0'. 010 7.5 M 51 7.5 0.5 0. 172 0, .015 F 66 8.9 0.4 0. 136 0 .008 10.7 M 34 5.2 0.3 0. 222 0 .015 F 47 5.7 0.3 0. 203 0 .012 15.0 M 202 3.4 0.1 0. 387 0 .017 F 283 4.0 0.1 0. 295 0 .008 19. 1 M 21 2.1 0.2 0. 637 0 .090 F 20 3.1 0.2 0. 353 0 .023 22.4 M 7 1.5 0.3 0. 786 0 . 107 23. 0 a F 4 3.3 0.6 0. 343 0 .066 M 3 1.7 0.2 0. 611 0 .055 a no female pupae eclosed at t h i s temperature. Table 2.3. Mean development times (days) and rates for pupae of PFW reared at constant temperatures. Temp. (\"CO Mean time Mean rate Sex n (d) SE (1/d) SE 2.3 M 19 98.0 1.4 o.oioj? 0.010? 0.020?° 0.019to° 0.000 F 47 97.2 0.8 0.000 .4.1 M 9 50.6 0.3 0.000 F 10 52.4 0.4 0.000 4. l a M 2 55.5 0.5 0.018 0.000 F 4 58.8 0.4 0.017, 0.028£° 0.027to° 0.000 5.8 M 16 36.1 0.2 0.000 5.8 a F 26 37.4 0.2 0.000 M 6 37.6 0.4 0.027 0.000 F 25 39.0 0.4 0.026. 0.037?° 0.0Z6*0 0.000 7.4 M 54 26.9 0.1 0.000 F 67 27.8 0.1 0.000 7.4 a M 16 27.8 0.3 0.036 0.001 F 40 28.6 0.2 0.035. 0.063? 0.062;b 0.000 11.0 M 39 15.8 0.1 0.000 F 52 16.1 0.1 0.000 11.0 a M 35 16.0 0.1 0.063 0.001 F 53 16.4 0.1 0.061, 0.095? 0.092? 0.101? 0.097^ 0.000 15.0 M 31 10.5 0.1 0.001 F 46 10.9 0.1 0.001 15.4 M 37 9.9 0.1 0.001 F 42 10.4 0.1 0.001 18.6 M 23 7.8 0.1 0.129? 0.120^° 0.001 F 27 8.4 0.1 0.002 18. 6 a M 34 7.6 0.1 0.132 0.002 F 58 7.9 0.1 0.127 0.001 23.1 M 4 6.3 0.3 0.161 0.007 F 4 6.6 0.2 0.151 0.003 23. l a M 34 6.0 0.1 0.168? 0.160?° 0.171? 0.162;b 0.002 F 27 6.3 0.0 0.002 26. 5 a M 20 5.9 0.1 0.003 F 9 6.2 0.2 0.006 a pupae reared at these temperatures were from prepupae reared at 15°C. •k mean developmental rates used to construct temperature-dependent rate model. ° development rate s i g n i f i c a n t l y d i f f e r e n t ( t - t e s t ; P<0.05) than rate for pupae with d i f f e r e n t prepupal thermal hi s t o r y reared at comparable temperature. 33 Adults did not emerge from the vermiculite boxes at temperatures greater than 20.5°C. Adult males emerged p r i o r to females at a l l but the highest temperatures (Table 2.4). M o r t a l i t y i n vermiculite boxes was high at a l l temperatures. Adult emergence was highest at intermediate temperatures. Insects died before reaching the pupal stage at the highest temperatures. At a l l temperatures, a portion of the insects died as adults before emerging from v i a l s . To test the i n t e r p o l a t i o n technique, estimated times for pupal development at 7.4°C were calculated from the insects that were moved d a i l y between 7.4 and 15.0°C. These times were compared with actual development times for pupae reared throughout the stage at 7.4°C (Table 2.3). Since the amount of time i n d i v i d u a l s spent at each temperature was variable, depending on when they pupated or eclosed as adults, these i n d i v i d u a l times were used instead of means i n the computations. Estimated times for males (n=15) and females (n=22) were 27.0 and 26.5 days, respectively. Although the estimate for females was le s s than the estimate for males, mean estimated rates were not s i g n i f i c a n t l y d i f f e r e n t from actual rates (males; t=-0.31, p=0.76: females; t=-1.99, p=0.06). Development periods for t o t a l subterranean development at 1.8°C were also estimated using t h i s technique (Table 2.4). T;b for a l l subterranean stages was set at 0°C since Table 2.4. Mean development times (days) and rates fo r pronymph to adult development of PFW reared at constant temperatures. Temp. (°C) Sex n Mean time (d) SE Mean rate (1/d) SE 1.8 a M 14 138.1 7.5 0.008 0. 000 F 25 146.2 5.8 0.007 0. 000 5.8 M 11 54.1 1.8 0.019 0. 001 F 24 58.1 0.9 0.017 0. 000 7.4 M 10 40.4 1.4 0.025 0. 001 F 20 43.3 1.0 0.023 0. 001 11.0 M 11 22.4 0.8 0.045 0. 002 F 16 26.7 0.8 0.038 0. 001 12.5 M 18 22.9 0.6 0.044 0. 001 F 25 23.6 0.4 0.043 0. 001 15.0 M 13 15.8 0.4 0.064 0. 001 F 28 16.9 0.3 0.060 0. 001 15.5 M 8 14.9 0.4 0.069 0. 004 F 16 15.4 0.5 0.066 0. 002 18.6 M 2 11.5 0.0 0.087 0. 000 F 10 11.0 0.1 0.092 0. 003 20.5 M 2 12. 3 0.8 0.082 0. 005 F 15 11.9 0.4 0.086 0. 003 estimated development times by i n t e r p o l a t i o n technique. 35 development progressed at the lowest temperatures tested. T m was set at temperatures just greater than the highest temperatures at which development was observed. The three-parameter equation (Equation 2.11a) was f i t t e d to prepupal and pronymph to adult subterranean development data (Fig. 2.4A, C), while pupal development (Fig. 2.4B) was better described using the four-parameter model (Equation 2 . l i b ) . The i n i t i a l model for female prepupal development did not adequately r e f l e c t the observed data at low temperatures (Fig. 2.4A), temperatures c r i t i c a l for development i n the f i e l d . Since sample sizes used to develop the function were disproportionate between groups, a weighted regression based on sample size was substituted. Parameter estimates for the equations are l i s t e d i n Table 2.5. Although the R 2-values for the weighted models suggested a poorer f i t than the unweighted models (Table 2.5), deviations between predicted and observed development rates for the former were smaller at low temperatures. Pupal and pronymph to adult developmental v a r i a b i l i t y were described using Equation 2.12a (Fig. 2.5B, C), while the p o s i t i v e skewed d i s t r i b u t i o n of prepupal v a r i a b i l i t y was described by Equation 2.12b (Fig. 2.5A). Parameter estimates for these equations are i n Table 2.6. Only the middle 90% of the v a r i a b i l i t y functions for prepupal development was used i n the simulations. This eliminated i n d i v i d u a l s with extreme developmental rates u n r e a l i s t i c a l l y suggested by Equation 2.12b. 36 F i g . 2.4. Development rates of male ( s o l i d l i n e s ) and female (dashed l i n e s ) (A) post-diapause prepupae, (B) pupae and (C) pronymphs to adults of PFW at constant temperatures. C i r c l e s and t r i a n g l e s are mean observed rates for males and females, respectively. Curves are regressions from Equations 2.11a and b. Both weighted (dotted l i n e ) and unweighted (dashed l i n e ) regressions are shown for female prepupal development. 37 0 5 10 15 20 25 0 5 10 15 20 25 30 0.10 - i 0 5 10 15 20 T E M P E R A T U R E (°C) 38 Pig. 2.5. Developmental v a r i a b i l i t y of male ( s o l i d l i n e s ) and female (dashed l i n e s ) (A) post-diapause prepupae, (B) pupae and (C) pronymphs to adults of PFW at constant temperatures. C i r c l e s and t r i a n g l e s are cumulative frequencies of observed i n d i v i d u a l rates/predicted rates from development rate regressions (Equation 2.11) for males and females, respectively. Curves are regressions from Equations 2.12a (B, C) and b (A). 39 Table 2.5. Estimated parameters and goodness-of-fit s t a t i s t i c s of models for development of subterranean stages of PFW. Parameters Stage Sex Pi ?2 ?3 P4 Tm Tb Pv 2 Prepupae M 0. ,057 3 .071 0. 041 - 24. 0 0.0 0. 996 M a 0. ,060 2 .984 0. 038 - 24. 0 0.0 0. ,995 F 0. ,066 2 .264 0. 061 - 24. 0 0.0 0. .969 pa 0. .051 3 . 184 0. 134 - 24. 0 0.0 0. .895 Pupae M 0. .209 2 .981 5. 470 0, .031 29. 0 0.0 0. .999 F 0, , 199 2 .927 5. 369 o, .031 29. 0 0.0 0, .998 Pronymph M 0. .008 3 .374 0. 105 - 24. 0 0.0 0, .993 to Adult F 0. .007 3 .481 0. 092 - 24. 0 0.0 0 .996 parameters estimated for regressions weighted by sample siz e . Table 2.6. Estimated parameters and goodness-of-fit s t a t i s t i c s for development v a r i a b i l i t y functions for subterranean stages of PFW. Parameters Stage Sex K C Q Pv 2 Prepupae M 2. .398 - 0. .274 0. ,995 F 2. .603 - 0. .317 . 0. . 999 Pupae M 34. .849 1. .009 0. .684 0. . 999 F 38. .254 1. .022 0, . 564 0. .999 Pronymph M 14. .281 0, . 983 1, .905 0, .998 to Adult F 18. .037 1, .006 2 . 148 0, .999 41 Adult emergence was 43% of the i n i t i a l 300 pronymphs placed i n the s o i l boxes. The emergence pattern of these adults was used to validate the subterranean development model under variable-temperature regimes measured i n the boxes from the two d e f o l i a t i o n zones. The model predicted emergence as much as 12 days too l a t e i n the heavily d e f o l i a t e d zone (Fig. 2.6A), while predictions (Table 2.7) were i n closer agreement to observed emergence i n the l i g h t l y d e f o l i a t e d zone (Fig. 2.6B). Phenology Model Validation. For both d e f o l i a t i o n zones and both sexes (Fig. 2.7), predicted adult emergence from single and mul t i p l e - s t r a t a models were only s l i g h t l y d i f f e r e n t . Although the r e s u l t s for the 10, 50, and 90% percent emergence classes were variable (Table 2.8), i n general the incorporation of multiple s t r a t a into the model s l i g h t l y enhanced accuracy. Increasing the time increment of the multiple-strata model from 1 to 4 h ( i . e . , At=0.167 d) d i d not adversely af f e c t r e s o l u t i o n of the model. 2.4 Discussion Wellington (1957) suggested that comparisons between two extreme habitats provided a more r e l i a b l e method of assessing the e f f e c t s of weather on insect populations than long-term studies i n one habitat. In the present instance, degree of exposure to sunlight i n the habitat was c l e a r l y an important variable that had to be taken into account. D e f o l i a t i o n opens 42 Fi g . 2.6. Simulated emergence of PFW males ( s o l i d l i n e s ) and females (dashed l i n e s ) from s o i l boxes i n (A) heavil y - d e f o l i a t e d and (B) l i g h t l y - d e f o l i a t e d zones. Symbols are observed emergence of males (tria n g l e s ) and females ( c i r c l e s ) . 44 F i g . 2 . 7 . Simulated ( l i n e s ) and observed ( c i r c l e s ) emergence of adults of PFW from LA plantation i n 1986: (A) males and (B) females from moderate- to heavi l y - d e f o l i a t e d zone and (C) males and (D) females from l i g h t l y - d e f o l i a t e d zone. Consecutive l i n e s of the same type represent pupal eclosion and adult emergence, respectively. Different l i n e types depict simulations using one depth ( s o l i d ) , multiple depths (short dash), and multiple depths with four-hour increments (long dash). 45 NOIJLdOdOUd Table 2 . 7 . Deviations (dev.) i n days of observed adult emergence from simulated emergence and the deviation as a proportion (pro.) of t o t a l simulation time for adults emerging from s o i l boxes. Percent emergence 10% 50% 90% Plot Sex dev. pro. dev. pro. dev. pro. L M 1.5 0.04 0.8 0.02 4.4 0.10 F 1.5 0.04 2.6 0.06 5.5 0.12 H M 7.5 0.22 8.7 0.23 11.8 0.26 F 9.6 0.27 10.4 0.27 11.7 0.26 Table 2.8. Deviations (dev.) i n days of observed adult emergence from simulated emergence and the deviation as a proportion (pro.) of t o t a l simulation time. Percent emergence 10% 50% 90% Model Plot Sex dev. pro. dev. pro. dev. pro. one depth B M 0.4 0. 01 0.0 0.00 0.2 0.00 F 1.7 0. 04 1.6 0.03 2.4 0.04 F M 3.4 0. .08 0.9 0.02 1.7 0.03 F 2.5 0. 06 1.8 0.04 6.0 0.11 s i x depths B M 0.2 0. .00 0.3 0.01 0.5 0.01 F 1.6 0. .04 2.0 0.04 1.4 0.03 F M 1. 1 0. .03 0.5 0.01 1.0 0.02 F 0.9 0. .02 1. 1 0.02 • 3.9 0.08 four-hour B M 0.6 0. .01 0.3 0.01 0.5 0.01 increments F 2.0 0, .04 1.9 0.04 1.3 0.02 F M 2.6 0, .06 0.5 0.01 1.1 0.02 F 2.0 0, .05 1.3 0.03 4.1 0.08 4? the canopy and changes the pattern of l i g h t penetration to the understory (Schowalter e_t a l . 1986). D e f o l i a t i o n by the pine moth, Dendrolimus p i n i . increased the magnitude and range of d i u r n a l s o i l temperatures (Ierusalimov 1973), as occurred i n the three d e f o l i a t i o n zones examined i n the present study. D e f o l i a t i o n by PFW affected s o i l temperatures and consequently the temporal pattern of emergence of the adults, but did not a l t e r the sexual sequence of emergence. Protracted emergence of Hyphantria cunea could be p a r t i a l l y a ttributed to variable degrees of solar r a d i a t i o n on overwintering s i t e s i n the s o i l (Morris and Bennett (1967). Temperature d i f f e r e n t i a l s between shaded and exposed s i t e s resulted i n a 2-3 week difference i n the emergence of N. swainei (Tripp 1965). Observed difference i n emergence of adults of PFW from the habitat extremes examined i n the present i n v e s t i g a t i o n was approximately one week. The variable sex r a t i o of PFW observed i n t h i s i n v e s t i g a t i o n suggested d i f f e r e n t i a l mortality between the sexes or d i f f e r e n t i a l sex determination mechanisms. Female PFW generally predominate i n le s s heavily-infested habitats (Jahn 1967). Where the insects were more numerous, the sex r a t i o was even or favored males, i n agreement with t h i s study. For the three d e f o l i a t i o n zones examined i n the LA plantation, males predominated where the insects were most abundant (M-zone) and females were more numerous where the the sawflies were least abundant (L-zone). Where densities were intermediate (H-zone) the sex r a t i o was 1:1. These differences may have resulted from 48 the f a i l u r e of some females to mate p r i o r to ovipositing at high d e n s i t i e s . Eonymphs were found below the humus layer at depths of 5 to 11 cm i n Europe (Schwerdtfeger 1941, Jahn 1967). Here, eonymphs were found from just below the s o i l surface to a depth of 9.cm. The mean depth was less than the 5 cm minimum suggested i n e a r l i e r studies. V e r t i c a l d i s t r i b u t i o n s of the s o i l c e l l s of PFW were not s i g n i f i c a n t l y d i f f e r e n t at the two locations sampled. Eonymphs of Acantholyda p o s t i c a l i s . i n some c l i m a t i c zones, overwinter deeper i n the s o i l i n response to reduced snow cover and s o i l desiccation (Kolomietz 1967). Since the two plantations were only separated by a distance of 112 km, c l i m a t i c e f f e c t s on overwintering depths between the two s i t e s were minimal. Since red pines are planted only on loamy sands and sandy loam s o i l s (Bassett 1984), the e f f e c t s of s o i l type would be minimal. Although the v e r t i c a l d i s t r i b u t i o n of females was consistently deeper than the v e r t i c a l d i s t r i b u t i o n of males, the sampling i n t e r v a l employed i n the study was not adequate to detect the differences. Increasing the resolution of the sampling i n t e r v a l , however, would not enhance the u t i l i t y . o f the s o i l d i s t r i b u t i o n function since s o i l temperatures do not vary greatly over the short distances involved. S o i l temperatures i n spring have been described as a monotonically increasing function of time (Logan et a l . 1979). While not s t r i c t l y true, since s o i l temperatures fluctuate 49 d i u r n a l l y , the net e f f e c t over a period of days or weeks i s a r i s i n g temperature. Although degree-days are not used here, rate summations are analogous to degree-day summations and the l a t t e r are a more standardized means of comparing heat accumulations. S o i l i n the l i g h t l y - d e f o l i a t e d zone accumulated heat at a slower rate than did s o i l i n the moderately- to heav i l y - d e f o l i a t e d zone. Accumulations of heat slowed as depth increased, even within the top 10 cm of the mineral s o i l . PFW inhabiting s o i l c e l l s at d i f f e r e n t v e r t i c a l depths experience d i f f e r e n t microclimates. S o i l temperatures predicted from a i r temperatures would be the most p r a c t i c a l approach for pest management purposes since soil-temperatures records are rar e l y available. Although some progress has been made i n t h i s d i r e c t i o n (Morse e_t al. 1985), accurate predictions for forested s o i l s are d i f f i c u l t (Novak and Black 1985). The model of s o i l temperature presented here imparts the needed resolution for use i n process-oriented phenological models and could be e a s i l y modified to predict hourly s o i l temperatures from maximum and minimum s o i l temperatures. The microclimate of s o i l - i n h a b i t i n g insects can be considerably d i f f e r e n t than that experienced by dwellers above ground. Several authors have used s o i l temperatures to predict phenological events for subterranean insects (Logan e_t al- 1979, Regniere e_t al- 1981, Morse e_t al- 1985). Phenological 50 predictions based on a i r temperature are not appropriate for insects i n the s o i l ( C o l l i e r and Finch 1985). S o i l and a i r temperature inputs provided similar r e s u l t s for simulation models of the black cutworm, Agrotis i p s i l o n (Kaster and Showers 1984). Bare and grass-covered s o i l s accumulated heat faster than a conventional meteorological screen (Baker 1981). In the present investigation, however, s o i l temperatures under the plantation canopy never approximated the extremes of the a i r temperatures. The 5-cm depth at which s o i l temperatures were recorded was near the median depth of the insects i n the s o i l . I f the thermocouples were at another depth the multi-stratum model might have provided much better predictions than the single-stratum model. The c u r v i l i n e a r nature of the temperature versus development rate functions was evident for a l l stages of PFW. Rates departed from l i n e a r i t y at both the highest and lowest temperatures. Prepupal development and pupal eclosion began soon aft e r s o i l temperatures rose above freezing. S i m i l a r l y pupal development also progressed when the s o i l temperature had only moderately warmed. The determination of development rates at low temperatures was c r u c i a l for the construction of pre d i c t i v e phenological models for PFW. Maximum s o i l temperatures observed during subterranean development of PFW never approached the temperature maxima for development. The i n h i b i t i o n of development at high temperatures observed i n the laboratory would r a r e l y occur i n the f i e l d . 51 Development rates of male prepupae and pupae were greater than those of females. The accelerated rate of male development explains the protandry observed i n the f i e l d . Differences i n development were unrelated to size differences within a sex. Therefore, size differences between the sexes i s not related to differences i n development rates (Lederhouse et a l . 1982). High mortality of insects i n the vermiculite boxes was s i m i l a r to the mortality of subterranean stages of PFW i n the laboratory experiments. The insects are extremely susceptible to desiccation and i n f e c t i o n once removed from t h e i r s o i l c e l l s and placed i n a r t i f i c i a l environments. Exarate pupae are also very f r a g i l e and e a s i l y damaged. The i n t e r p o l a t i o n technique was a useful method for determining development rates at low temperatures and c r i t i c a l i n predicting the phenology because development begins early i n the spring when temperatures are very low. Alternating temperatures reduced the developmental period i n calendar time and hence the exposure of insects to pathogens and desiccating conditions. The technique i s v a l i d i f developmental v a r i a b i l i t y i s not affected by temperature and developmental rates of i n d i v i d u a l s are constant during a stage. The construction of v a r i a b i l i t y functions by pooling standardized rates from a l l temperatures suggested that the f i r s t assumption was v a l i d . Exposure of insects to high temperatures early i n a stage may 52 retard subsequent development (Turnock et aJL. 1986). This did not occur with the pupae of PFW that were moved d a i l y between low and intermediate temperatures. However, incubation of the prepupa at moderate temperatures did result i n retardation of pupal development when reared at lower temperatures. Diapause i s the most s i g n i f i c a n t synchronizing element i n insect l i f e cycles (Tauber and Tauber 1981) e s p e c i a l l y for regulating voltinism. But the termination of diapause and subsequent morphogenesis of prepupae of PFW was more variable than pupal development. For PFW and some other insects (e.g., Hyphantria cunea) (Morris and Fulton 1970), prepupal development imparts a high degree of v a r i a b i l i t y to the phenology. This v a r i a b i l i t y , although dramatic i n physiological time, i s n e g l i g i b l e i n calendar time since spring s o i l temperatures are constantly increasing, which i n turn increases developmental rates and obscures the differences accumulated at low temperatures. The single-stage ( i . e . , pronymph to adult) model provided an adequate prediction of adult emergence of PFW from the s o i l boxes for only one of the d e f o l i a t i o n zones. Temperatures recorded i n the boxes i n the he a v i l y - d e f o l i a t e d zone exceeded the upper threshold for pronymph to adult development l a t e i n the development period. As has been demonstrated, subterranean development i s a combination of several temperature-dependent processes, each with i t s own maximum and minimum for 53 development. This model treated subterranean development as one process. The temperatures were often less than the optimal for pupal development, but greater than the maximum for prepupal development, and thus predicted slow or no development when rates would a c t u a l l y be near the maximum. Temperatures experienced by the insects i n the a r t i f i c i a l environments of the boxes were often higher than those that would be normally encountered by the subterranean stages under natural conditions. The s l i g h t l y better f i t achieved by incorporating v e r t i c a l d i s t r i b u t i o n into the simulation model did not j u s t i f y the increased computational complexity and time required. Increasing the time increment decreased the computational time without a f f e c t i n g the resolution of the simulation. The protracted emergence pattern of PFW adults resulted from: inherent v a r i a b i l i t y i n development rates of prepupae and pupae, and microclimatic differences r e s u l t i n g from the v e r t i c a l and horizontal d i s t r i b u t i o n of insects i n the s o i l . The v e r t i c a l d i s t r i b u t i o n determined the temperatures of the insects' habitat v i a the diu r n a l fluctuations and seasonal increases i n s o i l temperature during spring. D i f f e r e n t i a l shading of the s o i l surface, due i n part to degree of d e f o l i a t i o n , had a profound e f f e c t on s o i l temperatures and hence emergence patterns. The s t a b i l i t y of the prot.androus emergence pattern resulted 54 from the d i f f e r e n t i a l rates of prepupal and pupal development between the sexes and to a lesser degree from t h e i r d i f f e r e n t i a l v e r t i c a l d i s t r i b u t i o n . PFW has become more common i n recent years i n Ontario (Syme 1981). Although not widespread, PFW has been l o c a l l y abundant i n pine plantations. Even i n low numbers, PFW i s e s p e c i a l l y destructive i n Christmas tree plantations, where the webs and d e f o l i a t i o n reduce the trees' market value. The c h a r a c t e r i s t i c a l l y small tree size and lack of crown closure i n Christmas tree plantations would resu l t i n a very d i f f e r e n t s o i l microclimate than those found i n other kinds of plantations. Since s o i l temperatures are employed i n the model, i t should therefore provide adequate predictions of adult emergence i n the Christmas tree areas. Although there i s currently no i n s e c t i c i d e registered for use against t h i s species (Syme 1981), the present model w i l l be useful for timing control operations. 55 Chapter 3 Oviposition-fecundity models for pine f a l s e webworm. 3.1 Introduction Temperature during the oviposition period has a profound e f f e c t on o v i p o s i t i o n rates of most species examined. The pattern of oviposition influences the age d i s t r i b u t i o n of subsequent stages and therefore must be included i n phenological models. Oviposition pattern i s determined by the emergence pattern, s u r v i v a l and behavior of female sawflies and modified by weather (Rumphorst and Goosen 1960). F l i g h t period of PFW i s reported to extend from 21 days (Rumphorst and Goosen 1960) to a month (Jahn 1967). Adults are d i u r n a l and peak f l i g h t a c t i v i t y occurs between 1100 and 1500 hours on sunny, calm days with only minimal a c t i v i t y on cool, rainy or windy days (Rumphorst and Goosen 1960). L i t t l e i s known about the fecundity of PFW. Three females prevented from ovipositing and dissected contained 27, 37 and 41 eggs (Schwerdtfeger 1941). Several immature eggs were also found. Reported fecundities for PFW vary from 16 (Middlekauff 1958) to 35 (Rumphorst and Goosen 1960) eggs per female. Oviposition and fecundity r e s u l t from a complex i n t e r a c t i o n of several processes a l l affected by temperature. The objective of t h i s i n v e s t i g a t i o n was to examine these processes i n the f i e l d and i n the laboratory, to determine the e f f e c t s of 56 temperature on the processes and to develop a model of reproduction. 3.2 Materials and Methods Ovipo s i t i o n Pattern i n F i e l d . Terminals of twenty branches were examined d a i l y at about 1000 hours for newly deposited eggs during 1986. Ten branches at a height of 3.5 m and another ten branches at 1.5 m were selected around a small opening i n the plantation. Needles with new eggs were marked at each observation and the number recorded. Pairs of newly emerged adults from emergence traps were placed on red pine fo l i a g e i n lantern-chimney rearing containers. The lantern chimneys rested on the l i d s of 225-mL ointment j a r s f i l l e d with water. Cut ends of branches were inserted i n t o the jars through holes i n the l i d s . Tops of the chimneys were covered with muslin. An absorbant-cotton wick provided drinking water and honey was smeared on the muslin top. The rearing containers were housed i n a screened insectary situated i n a large clearing i n the LA plantation. A t o t a l of 90 pairs of adults was set up i n the rearing containers on 4, 5, 6 and 8 May 1986. Containers were examined d a i l y i n the morning, and eggs and dead females were removed and recorded. Temperature i n the insectary was recorded from thermocouples hanging from the c e i l i n g . Rearing containers were shaded by white sheets. 57 P o t e n t i a l Fecundity. Females col l e c t e d from the emergence traps i n the three plots i n 1983 (Section 2.2) were used to determine the rela t i o n s h i p between potential fecundity and body weight. Females were b r i e f l y dipped i n acetone and weighed. Females were then dissected and the number of mature oocytes i n the ovarioles was counted. Dissected females and oocytes were placed for a minimum of 24 h i n an oven at 70°C and weighed again. The number of eggs dissected from females was regressed against wet and dry weights. Square-root transformations of number of eggs, and wet and dry weights s t a b i l i z e d the variance and y i e l d e d l i n e a r relationships. Regression l i n e s were compared by analysis of covariance using the SAS GLM procedure (SAS I n s t i t u t e 1985). Potential fecundity of females c o l l e c t e d from emergence traps i n 1986 was estimated from t h e i r weights using the regressions and compared with observed p o t e n t i a l fecundity from dissections. Oogenesis. Samples of 50 newly-emerged females was weighed and randomly assigned to one of three treatments. One group (I) was confined with males on red pine branches i n lantern-chimney rearing containers and another group (II) was confined with males without fo l i a g e . Containers were placed i n a controlled-environment chamber at 23.4°C with a 15:9 (L:D) photoperiod. The number of residual oocytes was determined by d i s s e c t i o n and the number of eggs deposited by.group-I i n d i v i d u a l s was counted d a i l y . After three days, group-II 58 i n d i v i d u a l s were k i l l e d and -preserved i n 70% ethanol. The l a s t group (III) was k i l l e d and preserved i n ethanol. Ov i p o s i t i o n at Constant Temperatures. Adults, c o l l e c t e d from the s o i l at the LA plantation, were shipped on ice to the laboratory where they were used for oviposition studies. Females were weighed and confined with males on red pine branch t i p s i n chimney rearing containers. Containers were placed i n four constant-temperature chambers (14.9, 18.1, 23.4 and 26.6°C) with a 15:9 (L:D) photoperiod. Branches were examined d a i l y at the end of the scotophase (0800 hours) and a l l eggs were removed and counted. Death dates of females were recorded and cadavers were preserved i n 70% ethanol for l a t e r d i s s e c t i o n to determine residual oocyte content. To determine the pattern of egg laying for a group of females incubated at 14.9 (n=24) and 23.4°C (n=17), eggs were removed at the end of the scotophase and at 5-h i n t e r v a l s during the photophase, culminating at the start of the next scotophase, for the f i r s t seven days of t h e i r oviposition periods. Mean time to 50% fecundity of each i n d i v i d u a l was calculated at each temperature. Times were transformed to rates (1/time) and means calculated for each temperature. Mean rates, weighted by sample siz e , were regressed as a function of temperature. To describe o v i p o s i t i o n pattern around the means, a two-parameter Vfeibull function (Equation 3.1) was f i t t e d to the cumulative 59 d i s t r i b u t i o n of eggs per females as a function of normalized time ( o v i p o s i t i o n times/mean oviposition times). F(x) = 1 - exp(-x/r?)0) ( 3 . 1 ) where F(x) = the proportion of eggs deposited by normalized time x, and 7? and 0 are parameters estimated by nonlinear regression (SAS I n s t i t u t e 1 9 8 5 ) . Mean ageing rates (averages of reciprocals of i n d i v i d u a l l o n g e v i t i e s ) , weighted by i n i t i a l number of females, were regressed against temperature. Cumulative d i s t r i b u t i o n of female lo n g e v i t i e s was described as a function of normalized time (longevity/mean longevity). Reproductive p o t e n t i a l was calculated as the sum of age s p e c i f i c fecundities (eggs/female/day) for each temperature. Simulation of Oviposition-Longevity . A model to describe the egg-laying pattern was constructed from ovi p o s i t i o n and ageing rate functions. The model employed two physiological time scales, one for the rate females aged and the other for o v i p o s i t i o n a l rate. Temperature-dependent ageing rate was calculated for each hour of the day and the physiological age of the cohort mean was accumulated (physiological age = E [ageing rate per day * 1 / 2 4 ] ) . The cumulative proportion of the cohort that had died by normalized time x was solved from the cumulative d i s t r i b u t i o n function (Equation 3 . 1 ) . When the p h y s i o l o g i c a l age of the cohort mean equaled one, the proportion 60 of the cohort that had died-was 0.50. Proportion of the cohort a l i v e at each time i n t e r v a l i s 1 - F(x). The number of insects a l i v e at time t i s the product of the proportion a l i v e and the i n i t i a l number of insects i n the cohort. Chronological time of events was determined by keeping track of the hour of the day that was input into the model. Si m i l a r l y , temperature-dependent hourly oviposition rate was computed and the physiological age of the cohort mean was accumulated. The cumulative proportion of eggs deposited by the cohort was estimated from the d i s t r i b u t i o n of cumulative eggs/female function. The cumulative number of eggs deposited by an average i n d i v i d u a l was the product of the cumulative proportion and reproductive potential. For each cohort, the number of eggs deposited d a i l y by an average female was calculated at 1000 hours. Total egg production during the d a i l y i n t e r v a l was determined from the number of eggs per i n d i v i d u a l times the number of surviving females. The model was used to predict number of eggs deposited, pattern of egg deposition and r e l a t i v e abundance of females i n the screened insectary. Four cohorts, one for each s t a r t i n g date, were simulated. Temperatures recorded by thermocouples i n the insectary were used as model input. The cumulative proportional oviposition of natural populations was also simulated. This simulation was i n i t i a l i z e d 61 with the observed emergence ;times of the females i n the LA plantation i n 1986 (Fig. 2.IB). The number of cohorts ovi p o s i t i n g equaled the number of days females emerged, while the proportion emerging each day represented the proportion of the population i n each cohort. A i r temperatures recorded from the meteorological screen were used as input for t h i s component model. Simulation r e s u l t s were compared with egg depositions on the high and low branches i n the LA plantation i n 1986. 3.3 Results Oviposition Pattern i n the F i e l d . I n i t i a l cumulative deposition of eggs occurred sooner i n the lower branches than i n the high branches (Fig. 3.1). Egg accumulation was about the same i n both s t r a t a towards the l a t e r half of the ovi p o s i t i o n period. No eggs were deposited on 3 May (day 124). The maximum temperature on that day was 7.4°C, while the maximum temperature the previous and following days were 10.2 and 20.9°C, respectively. P o t e n t i a l Fecundity. The relationships between the number of mature oocytes dissected from newly emerged females and t h e i r wet or dry weights were highly s i g n i f i c a n t (P<0.0001) for females from the three plots trapped i n 1983. Since the slopes (wet weight, F=l.20, df=2,1206, P=0.30; dry weight, F-0.71, df=2,1195, P=0.49) and the distances (wet weight, F=1.82, 62 Fig . 3.1. Cumulative oviposition frequency by PFW on high and low red pine branches at LA plantation i n 1986. 115 120 125 130 135 DAY OF YEAR 64 df=2,1208, P=0.16; dry weight, F=0.39, df=2,1197, P=0.68) between the regression l i n e s were homogenous for both r e l a t i o n s h i p s , the data were pooled to y i e l d the following: oocytes = -1.11 + 0.77 wet weight (r 2=0.88) oocytes = -0.13 + 1.09 dry weight (r 2=0.91) Mean number of oocytes, and mean wet and dry weights were s i g n i f i c a n t l y d i f f e r e n t (Table 3.1) between the three d e f o l i a t i o n zones trapped i n 1986, indicating that increased d e f o l i a t i o n resulted i n a reduction i n female weight and oocytes/female. To te s t the models a b i l i t y to describe fecundity from body weight, p o t e n t i a l fecundities of the females emerging from the three d e f o l i a t i o n zones of the LA plantation i n 1986 were estimated from t h e i r wet and dry weights using the appropriate pooled regression. Estimated pote n t i a l fecundities approximated observed pote n t i a l fecundities (Table 3.2). Not only were the models able to predict p o t e n t i a l fecundity of pooled samples for the three LA zones, the models were rigorous enough to predict the p o t e n t i a l fecundities of females from each d e f o l i a t i o n zone (Table 3.2). Mean deviations of the predicted from observed were smallest for the heaVily-defoliated zone. The dry-weight model consistently yielded smaller deviations from observed than the wet-weight model, although the difference was only s i g n i f i c a n t for the l i g h t l y - d e f o l i a t e d zone ( t - t e s t , P<0.05). Table 3.1. Mean wet and dry body weights, and number of mature oocytes of dissected females of PFW trapped i n 1983 and 1986. Mean (SE) wet Mean (SE) dry Mean (SE) no. weight weight of Plot n (nig) (mg) oocytes 1983 AMI 822 67. .lc(0. 62) 23.5b(0.26) 26.9b(0.31) AM 2 221 v 84. .6a(0. .94) 30.8a(0.39) 35.3a(0.48) CR 170 70. ,4b(l. .12> 24.5b(0.47) 28.0b(0.60) 1986 H 135 51. .0o(l. .31) 16.2c(0.54) 18.7c(0.77) M 170 61. .6b(l. .07) 21.2b(0.49) 25.9b(0.68) L 120 76 .5a(0 .86) 29.4a(0.39) 35.0a(0.55) Means within columns and within years followed by same l e t t e r were not s i g n i f i c a n t l y d i f f e r e n t (Duncan's [1955] multiple range test, P > 0.05). Table 3.2. Predicted potential fecundities from wet- and dry-weight regressions for females of PFW trapped i n LA plantation i n 1986. Predicted Mean (SE) deviation mean (SE) from observed Plot no. Of oocytes no. of oocytes Wet weight .5 (0.38) H 19. . 1 (0.61) 0 M 24. 0 (0.50) -1 .8 (0.35) L 31. ,2 (0.41) -3 .7 (0.37) Pooled 24. .5 (0.38) -1 .6 (0.22) Drv weight .4 (0.36) H 18, .3 (0.63) -0 M . 24. 2 (0.57) -1 .7 (0.29) L 33. .7 (0.46) -1 .3 (0.35) Pooled 25. .0 (0.44) -1 .2 (0.19) 66 Oogenesis. Mean numbers of oocytes dissected from females of the three treatment groups and t h e i r mean l i v e weights are. shown i n Table 3.3. For group I females, the number of oocytes i s the t o t a l of the deposited eggs plus the residual number of mature eggs present i n t h e i r abdomens at death. Of the 50 group II i n d i v i d u a l s , only 28 survived to the t h i r d day following introduction into the rearing containers. For each treatment, the square roots of the number of eggs were regressed against the square roots of the l i v e weights (Table 3.3) and the regression l i n e s (Fig. 3.2) were compared using analysis of covariance. Increases i n square root of number of eggs resulted i n comparable increases i n square roots of wet and dry weights (F=1.64, df=2,118, P=0.20), but the distances between the rela t i o n s h i p s were d i f f e r e n t (F=44.66, df=2,120, P=0.0001). Group I and II females contained s i g n i f i c a n t l y more mature oocytes than did group III females (Table 3.3). I n i t i a l mean l i v e weight of group I and III females were not s i g n i f i c a n t l y d i f f e r e n t , but i n i t i a l weights of surviving group II females were s i g n i f i c a n t l y larger. Oviposition at Constant Temperatures. Females f a i l i n g to oviposit were eliminated from further analysis. Oviposition at a l l temperatures began the day females were introduced into the rearing containers. Hence, the preoviposition period of PFW was l e s s than one day at these temperatures and i s thus smaller than the observation i n t e r v a l employed. At each temperature, egg laying was characterized by an i n i t i a l burst of depositions 67 Pig. 3.2. Regressions for square root of fecundities of groups I-III females of PFW as a function of square root of t h e i r l i v e weight. 68 10 n followed by a sharp decline i n the egg-laying rate. Fecundities at the four experimental temperatures were not s i g n i f i c a n t l y d i f f e r e n t , but longevity was inversely proportional to temperature (Table 3.4). Maximum fecundity was 73 eggs by one female at 18.1°C. Ovipositions at 14.9 and 23.4°C occurred predominately during photophase although occasionally eggs were observed at the end of the scotophase (Fig. 3.3). Ovipositions were concentrated at the beginning of the photophase at 23.4°C. Oviposition rate increased with temperature (Fig. 3.4A) as follows: y = -0.230 + 0.034x where y = rate to 50% fecundity ( d a y s - 1 ) and x = temperature (°C) (r 2=0.918; F=22.28; df=l,3; P=0.042). Estimated parameters for the cumulative d i s t r i b u t i o n ( F i g . 3.4B) of o v i p o s i t i o n times (Equation 3.1) were r?=1.7488 and 0=0.9228 (R2=0.983). Temperature-dependent ageing rate function (Fig. 3.5A) for female PFW was y = -0.101 + 0.014x where y = mean ageing rate ( d a y s - 1 ) and x = temperature (°C) (r 2=0.962; F=50.57; df=l,3; P=0.019). Cumulative mortality (Fig. 3.5B) as a function of normalized time was y = 0.092 + 0.435x where y = cumulative mortality and x = normalized time (r 2=0.964; F=1712.06; df=l,65; P<0.0001). Observed reproductive potentials at 14.9, 18.1, 23.4 and 70 Table 3 . 3 . Regression s t a t i s t i c s for relationships between number of mature oocytes and l i v e weight for PFW. Mean (SE) Group n no. oocytes Mean (SE) l i v e weight (mg) Regression Slope Intercept I a 46 36. . 3a (2. .26) 60. l a (2. ,55) 0. 95 -1. ,36 0. 70 II 28 39. ,4a (1. .68) 72. 9b (2. .96) 0. 74 -0. .06 0. 88 III 50 26, ,8b (1 .27) 63. , 3a (2. .37) 0. .82 -1, .34 0. 85 a Mean number of oocytes for group I individ u a l s i s sum of deposited and retained eggs. Means followed by same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (Duncan's [1955] multiple range test; P > 0.05). Table 3 . 4 . Mean longevities and fecundities of PFW females reared at constant temperatures. Mean (SE) Mean (SE) Temperature longevity fecundity (°C) n (days) (eggs/female) 14.9 24 27.5a (3.17) 32.9a (3.81) 18.1 20 14.3b (2.11) 33.7a (4.49) 23.4 75 9.2bc (0.68) 34.6a (1.94) 26.6 11 7.1c (1.33) 42.0a (3.46) Means followed by same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (Duncan's [1955] multiple range test; P > 0.05). 71 F i g . 3.3. D i e l pattern of egg laying by PFW females at 23.4 and 14.9°C. Light and dark bands under histograms are photophases and scotophases, respectively. V e r t i c a l l i n e s are standard errors of means. T I M E ( D A Y S ) 7 3 F i g . 3.4. (A) Mean oviposition rate (±SE) to 50% fecundity for PFW as a function of temperature and (B) cumulative eggs per female at each constant temperature as a function of normalized time. Lines are regressions. 74 6 8 10 12 14 16 18 20 22 24 26 28 TEMPERATURE (°C) UJ < UJ u_ o O UJ UJ > I— < o 0 1 2 3 4 5 6 7 8 9 10 NORMALIZED TIME (TIME/MEAN TIME) 75 F i g . 3.5. (A) Mean ageing rate (±SE) of PFW females as a function of temperature and (B) cumulative proportion of dead females as a function of normalized time at each constant temperature. Lines are regressions. 76 6 8 10 12 14 16 18 20 22 24 26 28 TEMPERATURE (*C) 0.0 H —i 1 , , , , 0.0 0.5 1.0 1.5 2.0 2.5 3.0 NORMALIZED TIME (TIME/MEAN TIME) 26.6 °C were 43.2, 43r3, 44.3 and 51.1 eggs/female, respectively. Since fecundities at these temperatures were not s i g n i f i c a n t l y d i f f e r e n t , i t was assumed that these values were not d i f f e r e n t . For modeling purposes a pooled mean of 44.6 eggs/female, based on i n i t i a l number of females, was used. Simulation of Oviposition-Longevity. The d i s t r i b u t i o n pattern of egg deposition predicted by the model was very s i m i l a r to the observed pattern (Fig. 3.6A). Simi l a r l y , the pattern of female a v a i l a b i l i t y for oviposition (Fig. 3.6B) was s i m i l a r for observed and predicted, except the rate of removal of females from the predicted population was more gradual than the rate of removal of the observed population. Both observed and simulated populations were a l l dead at about the same time. Sum of fecundities for a l l females predicted by the model was 2465 eggs, while the observed number of deposited eggs was 2015. This translates to a predicted fecundity of 27 eggs/female and a observed fecundity of 22 eggs/female The difference between observed and expected fecundity r e f l e c t s the more gradual disappearance of the simulated population with a greater proportion of females available for oviposition towards the end of the o v i p o s i t i o n period. For simulations of egg deposition i n the f i e l d ( F ig. 3.6C), low temperatures at the beginning of the oviposition period resulted i n l i t t l e or no predicted o v i p o s i t i o n while observed ovipositions were taking place i n the f i e l d . The rate of 78 Fig . 3.6. (A) Predicted number of eggs (dotted l i n e ) compared with observed number of eggs ( s o l i d l i n e ) deposited by PFW i n outdoor insectary, (B) simulated (dotted l i n e ) compared with observed ( s o l i d l i n e ) number of ovipositing females and (C) simulated o v i p o s i t i o n pattern ( s o l i d l i n e ) for PFW compared with observed o v i p o s i t i o n pattern (dotted l i n e ) of natural populations and pattern of female emergence (dashed l i n e ) at Lakehurst i n 1986. -2 CD 80 predicted ovipositions increased dramatically when the temperature sharply increased, but the majority of natural egg depositions had occurred by t h i s time. The oviposition pattern i n the f i e l d was very similar to the pattern of emerging females (Fig . 3.6C). 3.4 Discussion Egg deposition i n the f i e l d stopped when maximum a i r temperatures were 7.4°C. This was close to the threshold predicted by the l i n e a r regression for oviposition rate. Maximum a i r temperatures of 10.2'C resulted i n ovipositions. The a b i l i t y of females to oviposit at such low temperatures i s probably enhanced by t h e i r dark color. Females began ovipos i t i n g as they moved up the trees r e s u l t i n g i n lower branches accumulating eggs sooner than upper branches. The low ambient temperatures during the beginning of the ovi p o s i t i o n period accentuated t h i s trend. Female body weight i s an unreliable estimate of p o t e n t i a l fecundity i n some sawflies due to varying degrees of fat u t i l i z a t i o n by maturing females (Heron 1966). However, both wet and dry weights were accurate predictors of p o t e n t i a l fecundity for PFW. Dry weights produced closer agreement between observed and expected p o t e n t i a l fecundities, but the increased resolution probably would not j u s t i f y the extra e f f o r t required i n determining dry weights. As demonstrated i n the previous chapter, the degree of d e f o l i a t i o n of the host trees had a profound e f f e c t on the phenology of emerging adults of PFW v i a a l t e r a t i o n of s o i l microclimate. In addition, associated with t h i s depletion of the foliage resource was a reduction i n the mean weight of emerging females which resulted i n a reduced p o t e n t i a l fecundity. Consequently, the e a r l i e s t females to emerge were those with the lowest potential fecundities. Since the fecundity models retained t h e i r predictive a b i l i t i e s for females from a l l d e f o l i a t i o n zones, i t was clear that stunted females from heavily d e f o l i a t e d zones were unable to shunt other reserves to egg production following starvation stress of the larvae. A few females at emergence contained no mature eggs. Sawflies vary i n t h e i r reproductive state at adult emergence. Although most females are ready to oviposit, the proportion of mature to immature oocytes present i n t h e i r abdomens varies from species to species. For some species the number of oocytes does not change after emergence (Wilkinson e_t a l . 1966, Lyons 1970), and thus dissections of newly emerged females provided a r e l i a b l e estimate of fecundity. Eggs of other species (Prebble 1941, Lyons 1964, Heron 1966, Ives fit a l -1968) undergo some postemergence maturation. A few immature oocytes were always observed during dissections of newly emerged females of PFW. Evidence presented here suggests that females underwent post-emergence maturation of these immature eggs. Females that were allowed to oviposit, produced more mature eggs than did females s a c r i f i c e d at emergence. Poor s u r v i v a l of females i n h i b i t e d determination of whether or not females kept 82 a l i v e and prevented^fTom ovipositing mature eggs. The larger s i z e of surviving 'females of t h i s group biased the r e s u l t s so that an estimate of egg maturation for t h i s group was u n r e l i a b l e . At a l l experimental temperatures, PFW began ovipositing within the f i r s t 24 h after emergence. Thus, PFW possesses a n e g l i g i b l e preoviposition period. Females are diurnal and even at constant temperatures i n the laboratory r e s t r i c t ovipositions to daylight hours. Fecundity of PFW was not dependent on temperature over the range of temperatures examined. This does not hold true for a l l insects (Hogg and Gutierrez 1980, Mason and Mack 1984). Conversely, ageing and oviposition rates of PFW were s i g n i f i c a n t l y temperature dependent as was the resultant pattern of oviposition. Oviposition and ageing are temperature-dependent phenomenon and can be treated using rate-summation techniques (Regniere 1984). Models incorporating these functions for PFW reared i n an outdoor insectary provided reasonable estimates for absolute fecundity, o v i p o s i t i o n pattern and female abundance. For these models to be incorporated into population dynamics models a functional r e l a t i o n s h i p between degree of d e f o l i a t i o n and fecundity needs to be developed. Only i n t r i n s i c s u r v i v a l (Curry and Feldman 1987), i n the absence of other mortality agents, i s •u t i l i z e d i n .the present model. Models l i k e t h i s one that keep mortality and oviposition as separate processes can be used to 83 t e s t other population mortality factors (Curry and Feldman 1987). Attempts to model the r e l a t i v e oviposition pattern of natural populations resulted i n poor estimates of cumulative emergence. During the beginning of the oviposition period i n the f i e l d , temperatures were below 10.2°C. These temperatures were below the lowest experimental temperature. The model predicted l i t t l e or no oviposition while observed females were layin g eggs. Oviposition rates as a function of these low temperatures may be nonlinear and higher than extrapolated rates. Conversely, females which have dark bodies may increase t h e i r temperature behaviorally. For p r a c t i c a l purposes, accurate estimates of cumulative oviposition to predict phenology can be obtained from cumulative adult emergence. PFW deposits a low number of large high quality eggs. The majority of eggs i s mature and ready to be oviposited when the female ecloses and emerges from the s o i l . Even at low temperatures, most eggs are deposited shortly after emergence. The black bodies of the females probably allow them to be active at low temperatures. This rapid deposition of eggs means that i f females succumb to a mortality agent, the eggs are already deposited. As a r e s u l t , s u r v i v a l of females has l i t t l e e f f e c t of generation s u r v i v a l . D e f o l i a t i o n a f f e c t s fecundity by reducing egg numbers and thus acts as a feedback system.in regulating population numbers. 84 Chapter 4 Arboreal development of pine false webworm. 4.1 Introduction Eggs of PFW are deposited on pine needles of the previous-year's fo l i a g e . Eggs hatch, under f i e l d conditions, a f t e r f i v e to eight days (Prozorov 1925) usually during late May (Middlekauff 1958) or after as long as three weeks of incubation (Griswold 1939). Upon hatching the larvae form conspicuous webs i n which they feed and develop. The ultimate-instar larvae drop to the ground from mid- to la t e June (Middlekauff 1958). These observations are of l i t t l e value for predicting seasonal timing of PFW since development i s a temperature-dependent process. Local weather during the development of these arboreal stages may have an ef f e c t on these processes. To accurately predict development, heat accumulations must be known i n the microhabitats where PFW develop. Meteorologists use conventional methods for measuring weather to permit comparisons between s i t e s . Temperatures are recorded from instruments placed i n standardized v e n t i l a t e d screens (e.g. Stevenson screen). Any insect habitat i s made up of a mosaic of microclimates fluctuating dynamically over time and space (Baker 1980), consequently, extreme differences may exist between these meteorological standards and temperatures i n insect habitats (Wellington 1950, Smith 1954, Morris and Fulton 1970). 85 V e r t i c a l strata of plant canopies often have d i f f e r e n t microclimates (Strong e_t a l . 1984), which may result i n differences i n development of the arboreal stages of PFW. Also, the webs of PFW may act l i k e heat sinks (Wellington 1950, Shepherd 1958), thereby affecting insect phenology. The purpose of t h i s study was to determine the development of PFW r e l a t i v e to the s p a t i a l d i s t r i b u t i o n of arboreal stages; the relationships between the temperatures i n the microhabitats of PFW and meteorological standards; the role of temperature on behavior of PFW larvae; and the ef f e c t of temperature on egg and l a r v a l growth and development. The ef f e c t s of temperatures i n the arboreal habitats on PFW development was determined through the use of phenological simulations. 4.2 Materials and Methods Sp a t i a l D i s t r i b u t i o n and Development of Arboreal Stages. Larval development i n the f i e l d was assessed using a regular sampling program. A l l 2781 trees within plot AMI were numbered with aluminum tags nailed to t h e i r boles and semiweekly, 50 trees were selected at random from t h i s plot. Red pines have a uniform growth form and flush a new v e r t i c a l whorl of branches from the bole each year, allowing easy d i v i s i o n of the canopy into three v e r t i c a l strata. The leader and f i r s t whorl of branches comprised the uppermost stratum. The next eight whorls 86 were divided equally into-high and low strata. The upper stratum and one branch randomly chosen by whorl number and cardinal d i r e c t i o n from each of the other s t r a t a were sampled. One branch was selected at random according to cardinal d i r e c t i o n from the fourth whorl of each tree i n 1984. Branches were examined and a l l stages of PFW were removed and preserved i n 70% ethanol. Due to the collapse of the PFW population at Anten M i l l s i n 1985, a l l branch sampling during 1986 was conducted i n 3 plots at the LA plantation. The largest plot (900 trees) was established i n a heavily-defoliated zone, while two smaller plots (100 trees) were established i n moderately- and l i g h t l y - d e f o l i a t e d zones. Branches, excluding leaders, were sampled semiweekly and the insects were picked off and preserved i n alcohol. Larval instars were determined using head capsule measurements. Comparisons were made between development of the arboreal stages i n d i f f e r e n t s p a t i a l dimensions. Mean development stage i n each st r a t a on each day was computed (egg = 0, l a r v a l i n s t a r s = 1 to 6). Larval Drop. The period of l a r v a l drop was determined using inverted adult emergence traps (Section 2.2) suspended from three wooden stakes. Glass c o l l e c t i n g bottles were fixed to the funnel stems at the bottom of the cones v i a t i g h t - f i t t i n g holes punched i n the p l a s t i c l i d s of the bottles. 87 F i f t y traps were placed -in plots AMI and AM2 i n 1983-85 as described for the emergence traps. During 1986, 25 traps were placed i n each of the heavily-defoliated and moderately-defoliated zones at the LA plantation. Traps were arranged at 5-m in t e r v a l s i n 2 rows 5 m apart. Traps were checked each morning and a l l PFW larvae were removed and preserved i n 70% ethanol. Sex of the ultimate-instar larvae was determined from head capsule measurements. Micrometeorology. Maximum and minimum d a i l y a i r temperatures were recorded from thermometers housed i n a conventional AES Stevenson screen i n 1983-85. Maximum and minimum d a i l y temperatures, as well as hourly temperatures, were recorded from a temperature probe i n a CS meteorological screen i n 1984-86. Linear regression was used to determine the re l a t i o n s h i p between d a i l y maximums and minimums recorded i n the AES screen and d a i l y maximums and minimums recorded i n the CS screen i n 1984-85. Temperatures recorded i n the CS screen during 1986 were corrected using t h i s l i n e a r regression. Temperatures i n the plantation canopy were recorded using three thermocouples (36-AWG) positioned v e r t i c a l l y , i n branch whorl numbers 3, 5.5, 8 from the top of the trees, at two locations i n plot AMI during 1985. Thermocouples were suspended under t i n f o i l pie plates, the upper surfaces of which were painted white to increase the emissivity and the undersides were painted black to i n h i b i t r e f l e c t i o n . Extension wires near the 88 junctions were painted white or covered with aluminum f o i l to s h i e l d them from solar radiation. The v e r t i c a l d i s t r i b u t i o n of thermocouples was dependent on tree height. Canopy temperature d i f f e r e n t i a l s (AT C) were determined by 1) converting CS screen temperatures to AES standard temperatures using the regression determined e a r l i e r and 2) subtracting these converted temperatures from the canopy temperatures (T c) for each height i n the canopy. Average canopy temperature d i f f e r e n t i a l s for the period of egg and l a r v a l development were then determined for each hour of the day. Canopy temperatures for 1985 were estimated from a i r temperature pooled for a l l heights using l i n e a r regression. D i f f e r e n t i a l pairs of thermocouples (36-AWG) were used to compare temperatures inside and outside the webs of PFW. Outside thermocouples were positioned adjacent to the webs, shielded by pie plates positioned to minimize the interception or r e f l e c t i o n of solar radiation to the web. Web temperature d i f f e r e n t i a l s (AT W) were computed from: AT W = T w - T c (4.1) where T w = web temperature (°C). The e f f e c t of d i f f e r e n t weather conditions on web temperatures, PFW l a r v a l body temperatures and a i r temperatures was determined by p e r i o d i c a l l y recording temperatures using thermocouple probes attached to a portable potentiometer. The body-temperature probe was made by ins e r t i n g a 40-AWG 89 thermocouple junction into a hypodermic needle. Thermocouple leads ran through a porcelain tube. The probe was inserted into the anus of the larvae or into the thorax. Web surface temperatures were recorded with an Instatherm in f r a r e d thermometer (Model 14-220D-15, Barnes Engineering Co., Stamford, Conn.). Egg Development at Constant Temperatures. The ef f e c t of constant temperatures on egg development was determined using eggs obtained from mating pairs of PFW confined on red pine branches i n lantern globe rearing containers (Section 3.2) at 23°C. Branches with eggs l a i d within 24 h were removed and placed i n constant-temperature chambers (3.9, 5.8, 7.0, 10.7, 14.9, 18.4, 23.8, 26.7 and 30.3°C) with 15:9 L:D photoperiods. Branches at the lowest three temperatures were moved to 14.9°C a f t e r 50 days of incubation to allow the eggs to complete t h e i r development and to prevent excessive mortality at these temperatures. Eggs at the f i v e highest temperatures were examined every 12 h u n t i l l a r v a l eclosion was completed. Eggs at lower temperatures were examined d a i l y . Larval Development at Constant Temperatures. Larval development rates were determined at several constant temperatures. Branches with eggs were c o l l e c t e d from the LA plantation and stored at 4°C u n t i l use. Twigs were cut to f i t rearing containers. Cut ends of the twigs were inserted through holes i n the tops of 450-mL ointment j a r s . The tops of the jars 90 were made by r i v e t i n g two l i d s together end to end. The upper l i d held the lantern globe enclosing the fo l i a g e . Branches were placed at 23 °C and examined d a i l y for l a r v a l e closion. Newly hatched larvae were held i n the rearing containers d i s t r i b u t e d among seven constant-temperature chambers (7.3, 10.4, 14.9, 18.1, 24.0, 26.7 and 29.0°C) and a fluctuating-temperature chamber set to alternate between 27 and 8°C ( i . e . , square wave, average 20.3°C) with the day-night l i g h t cycle. Larvae were examined d a i l y u n t i l they dropped when they were preserved i n 70% ethanol for sex determination by head capsule measurement. Dry weight of ultimate-instar larvae was determined. Egg Development i n the F i e l d . Since eggs i n the f i e l d are of variable and unknown age, eggs of known age were used to va l i d a t e models of egg development under f i e l d conditions. Eggs were obtained from mated females placed i n p l a s t i c bags enclosing red pine branches. Females were removed after 24 h; eggs were marked and examined d a i l y u n t i l hatch. Females were bagged at two heights i n the canopy i n 1984 and thermocouples (24-AWG) were fastened to the underside of the branches to record hourly average temperatures near the eggs. S i m i l a r l y i n 1985, females were allowed to oviposit over a three-day period and eggs that were deposited each day were observed throughout t h e i r development. Temperatures i n 1985 were measured using pie-plate-shielded thermocouples placed at the same height i n 91 the canopy. Behavior of Larvae at High Temperature. Individual branches containing l a t e - i n s t a r larvae were placed i n front of a General E l e c t r i c sunlamp (275 W) i n the laboratory. A thermocouple attached to a potentiometer was placed adjacent to a larva i n the web. The a c t i v i t y of the larva was noted when the l i g h t was turned on and the temperature i n the web rose. 4.3 Results S p a t i a l D i s t r i b u t i o n and Development of Arboreal Stages. The modes of the the polymodal frequency d i s t r i b u t i o n s of l a r v a l head capsule measurements (Fig. 4.1) representing l a r v a l instars occurred at about the same head capsule widths i n a l l years. The size l i m i t s (mm) of the six l a r v a l instars were: I < 1.005; 1.006 < II < 1.255; 1.256 < III < 1.555; 1.556 < IV < 1.955; 1.956 < V < 2.355; and VI > 2.356. A l l insects were i n the egg stage on 3 and 7 June (days 154 and 158) (Fig. 4.2A). Hatch began i n the leader and high s t r a t a by 10 June (day 161) and i n the low stratum by 14 June (day 165). As the season progressed, the insects high i n the tree were more advanced than lower insects. Although the insects i n the high stratum were s l i g h t l y advanced compared to those i n the low stratum, the differences were not as pronounced as the differences between the leader and upper strata. There was no s i g n i f i c a n t difference i n mean development of insects among the 92 Fig. 4.1. Frequency d i s t r i b u t i o n s of l a r v a l head capsule widths of PFW c o l l e c t e d from AM i n 1983 and 1984, and from LA i n 1986. Modes corresponding to the l a r v a l i n s t a r s are indicated by Roman numerals. 2500n 1875 1250-1983 625-IV VI >l~nlTlTTrl Ijllll-rTTT-rTTa-. 125T 100 1984 < > DC 75-< III 50 IV 25- VI r f T l 1600 -i 1200 1986 800-400-III IV VI JllR 0.70 0.90 1.10 1.30 1.50 1.70 1.90 2.10 2.30 2.50 2.70 HEAD CAPSULE WIDTH (mm) 94 Fi g . 4.2. Mean stage (±SE) of development of PFW from (A) three v e r t i c a l s trata at the AMI plantation i n 1963 and (B) three d e f o l i a t i o n zones i n the LA plantation i n 1986. 95 96 d i f f e r e n t cardinal points of the tree i n 1984. Development i n the heavily- and moderately-defoliated zones was :not s i g n i f i c a n t l y d i f f e r e n t i n 1986 (Fig. 4.2B), however, development was s l i g h t l y slower i n the l i g h t l y - d e f o l i a t e d zone. Larval Drop. Frequency d i s t r i b u t i o n s of head capsule measurements of larvae dropping from trees had two well-defined modes. The pattern of l a r v a l drop of the sexes was determined (Fig. 4.3), assuming that the larger i n d i v i d u a l s were females and the smaller larvae were males. The median time of male drop preceded the drop of females by 1 to 2 days i n both 1983 and 1986. There were only s l i g h t differences i n the timing of l a r v a l drop between plots i n both years. Micrometeorology. The re l a t i o n s h i p between the temperatures recorded i n the AES Stevenson screen and the CS screen temperatures was y = 0.554 + 0.928 x (F=9467.8; df=1,206; P<0.0001; r 2=0.979) where y = AES screen temperatures (°C) and x = CS screen temperatures (°C). The slope was s i g n i f i c a n t l y d i f f e r e n t from 1 (t=5.090; df=206; P=0.0001) and the intercept was not s i g n i f i c a n t l y d i f f e r e n t from zero (t=-1.159; df=206; P=0.2478), i n d i c a t i n g a s l i g h t temperature elevation i n the CS screen at high temperatures. A l l temperatures recorded i n the CS screen were corrected using the regression equation. The r e l a t i o n s h i p between the canopy temperature ( T c ) , pooled for a l l heights, and the standardized a i r temperature i n the 97 Fig . 4.3. Cumulative drop of ultimate instars of PFW i n sample plots i n (A) 1983 at the Anten M i l l s plantation (plots AMI and , AM2) and i n (B) 1986 i n the H and M zones of the Lakehurst plantation. 1.0 0.8 -0.6 -1983 7< 7 7 0. o cc Q Ui > < _J Z) 5 =} o 0.4-0.2-/ II: It-0.0 183 187 1.0n 0.8-0.6-0.4-0.2-1986 MALE AM1 FEMALE AM1 MALE AM2 FEMALE AM2 191 195 199 203 — i 207 • i • i .' i i i i ' 7 .»•/ .'/ MALE H .'/ FEMALE H 7 MALE M FEMALE M 0.0 162 166 1 4 -170 — i — 174 178 182 —i 186 DAY OF YEAR 99 meteorological screen (T a) was: T c = - 0.244 + 0.967Ta (4.2) (F=522482.1; df=l,8218; P<0.0001; r 2=0.985). Maximum d i f f e r e n t i a l recorded between the web and outside a i r temperatures i n 1984 was 13.0°C, but the mean difference (±SE) was only 0.7±0.02°C. The maximum temperature recorded i n the canopy was 30.2°C, while web temperatures peaked at 42.5°C. Mean web temperature difference (AT W) was described by: AT w ( t ) a{b + s i n [2?r(t - c)/24]} when 0700 ^ t ^ 2200 hours (4.3) -0.215 when 2200 < t < 0700 hours where t = time of day, a = amplitude (1.833), b = period (0.269), and c = phase (8.419) of the wave (R2=0.971). Estimated web temperature was also described as a l i n e a r function of canopy temperature using: T w = -1.789 + 1.147TC (4.4) (F=192789.1; df=1,7549; P<0.0001; r 2=0.962). The slope and intercept of the regression l i n e were s i g n i f i c a n t l y d i f f e r e n t from 1 (t=56.19; df=7549; P=0.0001) and 0 (t=-37.45; df=7549; P=0.0001), respectively. Table 4.1 l i s t s some representative temperatures associated with webs of PFW. Temperature v a r i a t i o n at each reading resulted from the small time constants of the fine thermocouple probes. Web temperature on a clear day [18 June (day 169)] was as much as 7°C i n excess of ambient a i r temperatures recorded Table 4 . 1 . Representative temperatures in webs of PFW and associated meteorological var iab les Ta S W Tc Tw Ti Ts Day of Time of year day (°C) (WirT2) Cms\"1) C O C C ) C C ) C C ) Weather Web 169 1117 17 . 8 0. 823 3.12 15 .8-16 .6 24 .0-24 .2 18.7 c lear o ld 1118 16 . 1-17 . 1 22 .0-22 .2 21 . 5-21 .7 c lear new 1 126 16 .3-16 .7 17 .9-18. . 3 c lear shaded/new 1 130 16 . 3-16 . 5 22 . 8-23 .6 20 .3-20 . 6 19.2 c lear new 1550 19 . 5 0.739 3.58 19 . 8-20 . 5 22 .0-24, , 1 23 .4-23, . 5 c lear new 1600 18 .2-18 . 3 19 .0-19, .7 18 .8-19, .0 cloud new 1627 18 , .8 0.606 3.63 19 .0-19. .7 20 .1-20, .6 21 .9-22, . 1 c lear new 170 1407 23 . 5 0 . 586 1 .72 22 .6-22. .8 24 .4-25, . 5 c lear new 1429 21 .6-22. . 1 25 .0-26. .7 23 .5-23. , 8 cloud new 175 1356 12 . ,6 0 . 340 3.72 1 1 .3-11. .9 1 1 .2-12. .4 12 . 8 overcast new 1415 12 . 1 0 . 328 3.35 1 1 .7-12. 5 12 .4-13. 0 12 . 8 overcast new 178 1 150 18 . 2 0 . 139 2.06 18 .0-18. 1 19 .2-19. 9 19 .3-19. 5 18.7 overcas t / r a in new Ta average a i r temperature for hour in which the web temperature was recorded S average solar rad ia t ion for hour in which the web temperature was recorded W average wind v e l o c i t y during hour in which the web temperature was recorded Tc a i r temperature from thermocouple placed adjacent to web Tw web temperature from thermocouple inserted into web Ti body temperature of PFW larva in web from thermocouple-syringe probe inser ted into anus Ts web surface temperature from infrared thermometer 101 outside the web. Temperatures recorded i n the older,-aisually uninhabited webs were consistently higher than the temperatures recorded i n newer webs. Older webs were darker than newer webs. Newer portions of webs usually contained green, fr e s h l y cut needles and l i v i n g needles. Temperatures of the larvae approximated the temperatures i n the webs under clear sunny skies and were i n excess of ambient a i r temperatures. A r t i f i c i a l shading of the web or the passing of a cloud i n conjunction with the high wind v e l o c i t i e s on 18 June (day 169) r a p i d l y cooled the web to near ambient a i r temperatures. Web temperatures remained above web temperatures on 19 June (day 170), with the lower wind v e l o c i t y , even when thick clouds were present. Web temperatures recorded from the shaded side of the web were closer to ambient temperatures than temperatures on the side exposed to d i r e c t solar radiation. Web temperatures and PFW body temperatures were only s l i g h t l y elevated above ambient a i r temperatures under overcast skies or wet conditions. The angle of incidence of the solar r a d i a t i o n profoundly influenced web temperatures. Under a constant radiant load, webs with t h e i r long axis t i l t e d from an acute to an obtuse angle to the sun increased i n temperature by as much as-4°C. High Temperature Behavior. As web temperatures reached 33-35°C, under the radiant heat load produced by the U.V. lamp, the larvae on the radiated side of the web moved out of t h e i r s i l k e n tubes. Larvae i n i t i a l l y fastened s i l k strands to nearby 102 needles, but assumed resting positions among the strands with t h e i r heads oriented towards the l i g h t . Larvae i n tubes leading away from the lamp moved towards the shaded side of the web, but as temperatures i n the front of the web increased, these larvae also vacated the webs. Some larvae dropped from the web on s i l k e n threads i f the temperature rose above 33°C, while others writhed e r r a t i c a l l y and died. Constant-temperature Development. The mean times for egg hatch at each experimental temperature and the corresponding development rates are given i n Table 4.2. A l l eggs incubated at 30.3°C f a i l e d to hatch. Since the s u r v i v a l of eggs at 26.7°C was noticeably reduced, the value of the temperature maximum (T m) was estimated to be 30.0°C. Development rates of eggs at 3.9, 5.8, and 7.0°C were estimated by the i n t e r p o l a t i o n technique (Section 2.3). Since mean rates of egg development after transfer of eggs to 14.9°C were a l l s i g n i f i c a n t l y d i f f e r e n t than the mean development rate of eggs reared throughout the stage at 14.9°C ( t - t e s t ; P < 0.0001), a l l eggs had undergone development at the lower temperatures p r i o r to transfer. The value of the temperature minimum (T b) was set at 3.0°C since the difference at 3.9°C was small. Eggs remaining at low temperatures had developed at 3.9 and 5.8°C and although head capsules were evident, the eggs f a i l e d to hatch. Larvae developed at a l l experimental temperatures (Table 4.3). However, newly eclosed larvae f a i l e d to es t a b l i s h on a l l Table 4.2. Mean development times (days) and rates f-or eggs of PFW reared at constant temperatures. Temp. (°C) Mean time Mean rate n (d) SE (l/d) SE 3.9 18 16. 0.4 0.053 0. ,000 282. 5.8 0.004 0. 000 5.8 49 12. 0.2 0.067 0. ,000 126. 1.6 0.008 0. .000 7.0 13 6. :$ 0.4 0.173 0. .004 73. 5.0 0.015 0. .001 10.7 265 35. .9 0.2 0.028 0, .000 14.9 254 20, .4 0.1 0.049 0, .000 18.4 239 14, . 1 0.1 0.071 0 .000 23.8 258 8 .5 0.0 0.118 0 .001 26.7 109 7 .9 0.0 0. 127 0 .001 30.3 No development mean times to completion of development a f t e r transfer to 14.9°C. estimated mean development times using i n t e r p o l a t i o n technique. Table 4.3. Mean development times (days) and rates for larvae of PFW reared at constant temperatures. Temp. (°C) Sex n Mean time (d) SE Mean rate (l/d) SE 7.3 M 4 73.3 1. 8 0. 014 0. 001 F 2 75.5 6. 5 0. 013 0. 001 10.4 M 9 60. 3 1. 6 0. 017 0. 000 F 9 66.0 1. 7 0. 015 0. 000 14. 9 M 17 36.7 1. 5 0. 028 0. 001 F 36 40.2 0. 9 0. 025 0. 001 18. 1 M 42 22.2 0. 3 0. 045 0. 001 20. 3 a F 52 25.6 0. 4 0. 039 0. ,001 M 7 17.6 0. 9 0. 058 0. ,003 F 8 20.0 0. 5 0. 050 0. .001 24.0 M 20 15.8 1. 0 0. 066 0. .002 F 26 17.0 0. 3 0. 059 0. .001 26.7 M 24 - 14.0 0. 2 0. 072 0, .001 F 47 16.4 0. 3 0. 062 0. .001 29.0 M 31 13. 5 0. 4 0. ,076 . 0, .002 F 39 15.5 0. 3 0. .065 0 .001 average of alternating temperatures. 104 but two branches at 7.3°C. Consequently, the value -£or u t i l i z e d for modeling purposes was set at 7°C. The value of T m was estimated from preliminary experiments to be about 33.0°C. Development rates as functions of temperature for the egg and l a r v a l stages were developed as described i n Section 2.2. Parameter estimates for Equation 2.11a for egg development were P1=0.008, P2=4.393 and P3=0.131 (R2=0.996). Functions (Equation 2.11b) for l a r v a l development were developed separately for each sex with parameter estimates of Pi=0.094, P2=2.051, P3=4.446 and P4=0.048 (R2=0.995) for males and P1=0.084, P2=1.977, P3=4.191 and P4=0.053 (R2=0.992) for females. Estimated parameters for developmental v a r i a b i l i t y functions for eggs and larvae were obtained as described i n Section 2.3 using Equation 2.12a. Estimated parameters for eggs were K=16.257, C=0.994 and Q=0.796 (R2=0.997). Parameters for male larvae were K=18.257, C=1.043 and Q=2.112 (R2=0.999) and for female larvae were K=16.540, C=0.969 and Q=1.378 (R2=0.999). The means of the d i s t r i b u t i o n of head capsule widths of the l a s t - i n s t a r larvae at the lowest temperatures were smaller than at intermediate temperatures (Fig. 4.4). Head capsule sizes at the a l ternating temperature (average 20.3°C) more cl o s e l y resembled head capsule sizes of larvae reared at the higher of the two temperatures. Mean dry weights of larvae were larger at higher temperatures with a maximum at 24.0°C (Table 4.4). A s i g n i f i c a n t l y lower dry weight of females was evident at the 105 Fig. 4 . 4 . Frequency d i s t r i b u t i o n s of head-capsule widths of larvae of PFW reared at seven constant temperatures and an alternating temperature regime. 106 ill < > DC < _i u_ O DC LU 03 Z 20 15 10 5 0 15 10 5 0 10-5 0 20 15 10 5-0 10\" 5 0 5 0 5 0 5 0 29.0°C 26.7°C JZL 24.0°C i s . r c J Q 14.9\"C 10.4\"C J Z L n 7.3°C 8.0/28.0°C 1.90 2.10 2.30 2.50 2.70 HEAD CAPSULE WIDTH (mm) 107 highest temperature. Mean dry weights of the larvae—reared at the alternating temperature were less (not s i g n i f i c a n t ) than would be anticipated for larvae reared at the average of the fluc t u a t i n g temperatures. Va l i d a t i o n of Egg Phenology Model. A computer simulation based on the described development rate and v a r i a b i l i t y functions was used to compare observed egg development i n the f i e l d i n 1984 with predicted egg development. A FORTRAN algorithm modified from the program developed by Regniere (1984) was used. The simulation began on 1 June (day 152) when eggs were deposited on branches by the bagged females. Two independent sets of temperature data were used as input for the model: average hourly a i r temperatures recorded from the CS temperature probe and hourly temperatures computed from maximum and minimum temperatures recorded i n the Stevenson screen, using half-cosine functions (Regniere 1982). Predicted emergence was within three days of observed emergence (Fig. 4.5A) for both simulations. The simulation using recorded hourly temperatures was i n closer agreement with observed r e s u l t s than the simulation using estimated hourly temperatures, but the difference i n deviations of observed from simulated hatch was small (Table 4.5). The deviations between observed and expected for both simulations were greatest at the time of ' 50% emergence for the three egg hatch percentages compared. However, the deviations were less than 16% of the t o t a l 108 F i g . 4.5. Predicted ( l i n e s ) and observed (symbols) egg hatch of PFW ( s o l i d l i n e s ) for (A) single-cohort simulation and multiple-cohort simulations using (B) observed hourly a i r temperatures, (C) estimated hourly a i r temperatures and (D) observed canopy temperatures. S o l i d l i n e i n (A) i s simulation r e s u l t s using observed hourly temperatures and dashed l i n e i s r e s u l t s using estimated hourly temperatures from d a i l y maximum and minimum temperatures. S o l i d l i n e s i n (B-D) are simulation r e s u l t s for i n d i v i d u a l cohorts and the dashed l i n e s are r e s u l t s for population. 110 Table 4.4. Mean3, dry weights (±SE) of larvae of PFW reared at constant temperatures. Males Females Temp. cc5 n Wt. (mg) n Wt. (mg) 7.3 1 8. 3d 1 35.6cd 10.4 9 18.2+0.75c 9 31.4±1.92d 14.9 17 20.l±2.21bc 36 36.3±1.14cd 18.1. 42 24.7±0.64ab 52 49.0±1.19ab 20. 313 7 20.8±0.50bc 8 ' 41.2±1.43bcd 24.0 20 27.2±0.83a 26 52.0±1.78a 26.7 24 23.7±0.74abc 47 45.8±1.22abc 29.0 31 23.0±0.72abc 39 40.4±1.38bcd a means i n the same column followed by the same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (Duncan's [1955] multiple range test, P < 0.05). average of fluc t u a t i n g temperature. Table 4.5. Deviations (dev.) i n days of observed egg hatch from simulated egg hatch and the deviation as a proportion (pro.) of t o t a l simulation time for the 1984 single-cohort simulation and the 1985 multiple-cohort simulation pooled for a l l three days. Percent emergence 10% 50% 90% Year Model input dev. pro. dev. pro. dev. pro. 1984 Obs. A i r Temp. 0. .4 0. .03 2. . 1 0. 14 1. .0 0. 06 Est. A i r Temp. 0. . 5 0. .04 2. . 5 0. 16 1. .2 0. ,07 1985 Obs. A i r Temp. -3. . 9 0. . 19 -2. .6 0,11 -3. . 5 0. .14 Est. A i r Temp. -1. . 9 0. .09 -1 . 2 0:05 -0. ,8 0. .03 Obs. Canopy Temp. -0. .8 0, .03 0, .4 0.02 2. .5 0. .08 I l l simulation time. As a test of modeling multiple cohorts, the development of the eggs deposited 14 to 16 June (days 165 to 167) i n the f i e l d during 1985 by bagged females was simulated. The results were compared with actual egg development over the same period of time deposited by bagged females. The proportion of eggs deposited on each day ( i . e . , 0.555, 0.287, and 0.158, respectively) was used as the proportion i n each cohort starting development. Three sets of temperature data were used to drive the model: 1) hourly temperature averages (Fig. 4.5B); 2) estimated hourly temperatures (Fig. 4.5C); and 3) average hourly temperatures recorded from thermocouples at the same height i n the canopy as the eggs (Fig. 4.5D). Estimated egg hatch for each of the three d a i l y cohorts were pooled and compared with the pooled observed egg hatch (Table 4.5). Simulations driven by both estimated a i r temperature and canopy temperatures produced smaller deviations from observed than did the simulation based on observed a i r temperature. The canopy temperature simulation better predicted the onset of egg hatch when compared with the estimated a i r temperature simulation, but deviated more towards the termination of the egg hatch period. Arboreal Development Simulations. A model was used to simulate the combined development of eggs and larvae. The simulation was i n i t i a l i z e d with the observed oviposition pattern of wild females at the LA plantation i n 1986 (Fig. 3.1). The 112 observed proportion of eggs l a i d each day represented the proportion of insects i n the d a i l y cohorts. Eggs began development during a 14-day period, consequently, 14 cohorts were simulated simultaneously and the completion of development of the larvae was proportioned accordingly. Model input was the corrected a i r temperature recorded i n the meteorological screen at the LA plantation during 1986. After a preliminary simulation with no corrections for microclimatic differences, the model was modified to incorporate: i ) a canopy temperature correction function from the pooled canopy temperatures i n 1985 (Equation 4.2), i i ) the l i n e a r web correction function (Equation 4.4), i i i ) the non-linear web correction function (Equation 4.3) and i v ) a combination of i and i i i . In addition, only the fastest developing 90 percent of the population was simulated to determine i t s e f f e c t on the d i s t r i b u t i o n of dropping times. This was accomplished by d i v i d i n g the upper 90 percent of the v a r i a b i l i t y function into percentile categories. The output of the model was predicted drop of l a s t - i n s t a r male and female larvae from the trees. The predicted drop was compared to observed captures i n the traps, pooled for H- and M-defoliation zones, from the LA plantation i n 1986. Results of the simulation (Fig. 4.6A) without correction for microclimatic differences provided an adequate prediction of the onset of l a r v a l drop (Table 4.6), but was less accurate i n predicting the termination of the arboreal stage. Introducing 113 Fig. 4.6. Observed (circles=males, triangles=females) and predicted ( l i n e s ) l a r v a l drop of male and female PFW from the St-and M-defoliated zones of LA plantation i n 1986; (A) CS screen temperatures, (B) canopy temperature correction, ( s o l i d line=males,.dashed line=female) (G) l i n e a r ( s o l i d line=males, dashed line=females) and nonlinear (dotted line=males, dotted-dashed line=females) web temperature corrections and (D) web and pooled canopy corrections ( s o l i d line=males, dashed line=females). 114 115 Table 4.6. Deviations (dev.) i n days of observed l a r v a l drop from simulated drop and the deviation as a proportion (pro.) of simulation time. Percent drop 10% 50% 90% Microclimatic correction Sex dev. pro. dev. pro. dev. pro. none M 2. .5 0. .05 2. 9 0. ,05 8. 4 0. 13 F 2. .6 0. .05 5. .6 0. , 10 9. 4 0. 14 T c (pooled) M 5. .4 0. . 10 6. .9 0. , 12 11. 1 0. 17 T w (Equation 4.4) F 6. .6 0. . 12 9. .5 0. , 15 12. 0 0. 17 M 1. .2 0. .02 1. .7 0. ,03 6. 3 0. 10 T w (Equation 4.3) F 1, .7 0. .03 4. .3 0. 08 8. 0 0. 12 M 0. .8 0. .02 1 . , 3 0. .02 5. 6 0. 09 F 1. .4 0. .03 3. .8 0. .07 7. 5 0. 12 T c and T w M 3. .8 0. .07 4. ,8 0. .09 9. 2 0. 14 F 4. .4 0. .08 6. .9 0. . 12 9. 7 0. 15 T c and T w M 3. .7 0. .07 4. . 1 0. ,07 5. 3 0. 09 (90% v a r i a b i l i t y ) F 3. . 9 0, .07 6. .2 0. . 10 7. 8 0. 12 ) 116 the canopy-temperature correction function into-^the model (Fig. 4.6B) resulted i n larger deviations between observed and v predicted l a r v a l drop. Inclusion of either web-temperature correction functions into the model (Fig. 4.6C) resulted i n predictions with s l i g h t l y smaller deviations from observed than the basic model. Using nonlinear correction function for web temperature predicted s l i g h t l y faster development than did the l i n e a r function, and was i n closer agreement with predicted drop. When both microclimatic correction functions were employed (Fig. 4.6D), the r e s u l t s were more accurate than when only the canopy function was used, but were not an improvement over basic or web-temperature corrected models. The number of days between male and female drop was much greater for a l l simulated development (4.3-5.1 d at 50% drop) than for observed development (2.1 d at 50% drop). In addition, deviations between predicted and observed increased from the onset to the termination of l a r v a l drop for a l l simulations. The difference i n deviations between the onset and termination of l a r v a l drop was reduced by using 90 percent of the developmental v a r i a b i l i t y . Inclusion of estimated web temperatures into the model reduced development time from 1.2 to 2.8 d over corrected screen temperatures. Including estimated canopy temperature into the model, however, retarded development by 2.6 to 4.0 d. Simulation r e s u l t s combining both canopy and web correction functions suggested that higher temperatures are required to 117 reduce the deviations between simulated and observed l a r v a l drop. 4.4 Discussion The r e s u l t s corroborated the observation that PFW males and females have f i v e and six l a r v a l i n s t a r s , respectively (Schwerdtfeger 1941). The class i n t e r v a l s for instars occurred at about the same head capsule widths for the three years for which frequency d i s t r i b u t i o n s were compared. However, the head-capsule sizes for North American populations were considerably larger than head capsules of a European population (Schwerdtfeger 1941). The width of ultimate l a r v a l head capsules i n the laboratory study also increased with temperature to a maximum, after which i t was inversely correlated with increases i n temperature. Low and high temperature e f f e c t s on head capsule size have been reported for other insects (Uvarov 1931, Guppy 1969). Differences i n head capsule sizes observed i n the constant temperature rearings suggest that the European population either experienced d i f f e r e n t temperatures or are ge n e t i c a l l y d i f f e r e n t . The weight of l a s t - i n s t a r larvae also varied with temperature. Thus, the temperature experienced by the larvae affected t h e i r a b i l i t y to assimilate food and t h e i r a b i l i t y to accumulate resources for fecundity. The growth rate of PFW, l i k e Colias larvae (Sherman and Watt 1973), i s probably greatest at or near i t s optimal range of temperatures for feeding. 118 O s t r i n i a n u b i l a l i s had s i g n i f i c a n t l y larger heads and greater weight when reared i n variable temperatures than larvae reared under comparable constant temperatures (Beck 1982). This was not the case with PFW. However, the maximum and minimum temperatures used i n the variable regime were outside the range for e f f i c i e n t weight gain. Head capsule sizes of f i e l d c o l l e c t e d larvae were comparable to larvae reared i n the intermediate constant temperatures i n the laboratory. Larvae c o l l e c t e d from webs from the four cardinal quadrants of trees developed at the same rate. Larvae developed slower i n the least d e f o l i a t e d trees of the same plantation. The difference i n development of the larvae i n the d e f o l i a t i o n zones probably resulted from differences i n adult emergence and degree of exposure of the webs to solar radiation. The l i g h t l y d e f o l i a t e d zone also produced the heaviest larvae so that the slower development was not because of inadequate n u t r i t i o n . Slight differences i n development were observed i n di f f e r e n t canopy str a t a . The fastest development occurred near the top of trees. The differences i n development rate were related to the degree of exposure of the webs to solar radiation. While sampling programs to determine PFW developmental status can ignore cardinal d i r e c t i o n , they must incorporate v e r t i c a l s t r a t i f i c a t i o n . Plant canopies experience a d i f f e r e n t microclimate than the standards of conventional meteorology. The red pine canopies i n 119 which PFW reside were generally cooler than clearings used for meteorological screens. Temperatures of canopies underwent a diu r n a l temperature pattern with the deviations dependent upon time of day and height i n the canopy. Several functions have been proposed for shelters b u i l t by phytophagous insects. These include protection from enemies (Morris 1976), desiccation (Henson 1958, Willmer 1980, Strong fit al. 1984) and phototoxicants (Berenbaum 1978). However, the sa l i e n t feature of these shelters i s that the i n t e r n a l temperature d i f f e r s from the temperature of the surrounding a i r (Wellington 1950, Barbosa fit al. 1983). Insect shelters have been likened to greenhouses that mitigate spring temperatures i n the Temperate Zone (Wellington and Trimble 1984). Solar r a d i a t i o n has a profound e f f e c t on elevating temperatures of exposed plant structures, feeding insects and the inside of these i n s e c t - b u i l t shelters (Wellington 1950, Wellington 1954). The r e l a t i o n s h i p between the external and i n t e r n a l climates of these structures i s complex. The webs of PFW exhibit a diurnal pattern of heating and cooling related to the diurnal i n t e n s i t y of solar r a d i a t i o n (Henson and Shepherd 1952, Shepherd 1958). As with other shelter-building insects (Henson 1958), the magnitude of the pattern was modified by the degree of cloud cover, ra i n , wind and the color, orientation, and size of the web. Elevations of web temperatures may be due i n part to re f l e c t e d and transmitted r a d i a t i o n i n addition to global solar r a d i a t i o n . 120 As with other shelter-building insects (Wellington 1954, Henson 1958, Green 1968), PFW larvae vacate the web to avoid c r i t i c a l high temperatures. Temperatures i n webs of PFW occasionally r i s e above tolerable l e v e l s for the larvae. The larvae then respond by moving away from these overheated areas. When they leave the web, the larvae also leave i t s boundary layer and thus improve t h e i r convective heat exchange. F i n a l l y , the larvae reduce the surface area exposed to the radiation by positioning themselves with t h e i r heads towards the radiation source. The shelters b u i l t by some early-emerging insects are aids to e x p l o i t i n g the cold temperatures of early spring (Wellington 1950). Yet when PFW hatches from the egg, eastern tent c a t e r p i l l a r (Malacosoma americanum) i n the same plantation had completed the tent-building phase. Other non-webbing pine-feeding sawflies i n the same plantation (Neodiprion nanulus nanulus. Diprion s i m i l i s and D. frutetorum) were i n more advanced stages of development than PFW. Neodiprion n. nanulus males were dropping and forming cocoons several days i n advance of the PFW l a r v a l drop. Although the web-building habit of PFW accelerates development other adaptations do not allow PFW to take advantage of early spring temperatures i n i t s present d i s t r i b u t i o n i n North America. The web of PFW i s multifunctional. In addition to 121 thermoregulation, the silkspinning habit of the-.larvae serves to f a c i l i t a t e mobility (Kolomietz 1967) and attachment to the food plant (Strong e_t a l . 1984). The larvae move along the substrate ( i . e . , a needle, a branch or any other surface) on t h e i r backs by moving t h e i r heads from side to side and attaching a s i l k ladder under which they t r a v e l . This behavior was observed on numerous occasions i n the present study i n a l l but the ultimate i n s t a r . Some pine-feeding sawflies feed from the d i s t a l t i p s of the needles towards the base (Ghent 1958). However, PFW larvae are unable to grasp t h e i r food and b i t e off the needles at the base and feed as the needles are drawn into the web. Disturbed PFW larvae ret r a c t into the web away from the food and need a method of securing the needles. Examination of a number of webs containing in t a c t needles revealed that the larvae had guyed the needles with strands of s i l k p r i o r to feeding on them. Severed needles were fixed i n p o s i t i o n with s i l k strands. These s i l k guys also could serve to slow the movement of parasitoids around the web. Male PFW larvae begin development at the same time as females, but because they pass through one less instar, they precede the females from the trees. Thus, sampling after the onset of l a r v a l drop w i l l be biased towards the larger females. The described model treats l a r v a l development as one stage. Differences i n the development of i n d i v i d u a l instars may have affected model predictions. In some insects, d i f f e r e n t i n s t a r s 122 have d i f f e r e n t developmental rate l i m i t i n g parameters (Beck 1982). The c r y p t i c habits of PFW made i t d i f f i c u l t to observe the rates of development of the di f f e r e n t i n s t a r s . Nonlinear equations employed i n t h i s study provided r e a l i s t i c descriptions of the development rates of eggs and larvae of PFW. Unlike the subterranean stages of PFW, the arboreal stages experience both high and low temperature extremes. Insect embryogenesis can occur at temperatures below which eggs can hatch (Morris and Fulton 1970, Beck 1983), and t h i s happened with PFW. Interpolation techniques allowed t h i s low temperature development to be incorporated into the development rate equations. Observed canopy temperatures provided an accurate estimate of egg development when used as input to the model. This indicates that egg development on the surface of pine needles was not s i g n i f i c a n t l y influenced by radiant heating or cooling. Temperatures of conifer f o l i a g e have been reported to be as much as 5.6°C warmer than the surrounding a i r when exposed to solar r a d i a t i o n and as much as 3°C below ambient temperatures on clear nights (Wellington 1954). Temperature elevations of as much as 6.3°C i n the mines of needle mining insects have also been observed (Henson and Shepherd 1952). In addition, differences i n egg development of the moth, Orgyia pseudotsugata. between thinned and unthinned Douglas f i r canopies were attributed to differences i n solar r a d i a t i o n (Wickman and Torgersen 1987). 123 Although the s l i g h t difference between observed and predicted PFW egg hatch thus may have resulted from compensatory diurnal heating and nocturnal cooling of the foliage, i t may also have been a re s u l t of an averaging of temperatures from convection within the canopy. For example, use of infrared thermometry showed that needle temperatures did not d i f f e r greatly from a i r temperatures i n well-ventilated Douglas f i r canopies (Tan e_t a l . 1978). The most accurate prediction for onset and median egg hatch was provided using ambient temperatures recorded i n the plantation canopy as model input. Estimated a i r temperatures from maximum and minimum thermometers provided estimates of egg hatch i n 1984 and 1985 as accurate as uncorrected observed a i r temperatures recorded i n the CS screen. Employing uncorrected observed a i r temperature predicted egg development too early. Overheating of the CS screen would explain t h i s discrepancy. V e r t i c a l differences i n canopy temperatures were not responsible for the observed faster development of PFW i n the upper canopy. Furthermore, the oviposition pattern as a function of tree height did not result i n the accelerated development i n the upper canopy since female PFW begin ov i p o s i t i o n i n the lower canopy. Therefore, the faster development i n the upper canopy probably resulted from the increased exposure of these webs to solar radiation. 124 Behavior of monarch b u t t e r f l y c a t e r p i l l a r s .reduced the l a r v a l period by 10 to 50% by af f e c t i n g the:thermal environment (Rawlins and Lederhouse 1981). This reduction decreased the period larvae were exposed to predators, parasitoids and pathogens. Elevations of web temperature i n the present study reduced the developmental period i n calendar time by only 1.2 to 2.8 days over screen temperatures. Conversely, canopy temperatures increased the developmental period when compared with screen temperatures. Simulations suggested that observed larvae experienced more thermal units than was predicted by the model. For example, the extreme d e f o l i a t i o n encountered i n the Lakehurst plantation i n 1986 was very d i f f e r e n t from d e f o l i a t i o n i n the plantation on which the empirical microclimatic re l a t i o n s h i p s were based. Severe d e f o l i a t i o n at Lakehurst opened the canopy to such an extent that PFW webs were exposed to intense solar r a d i a t i o n for longer periods. The climate experienced by PFW arises from a series of compensatory processes i n t h e i r microhabitats. PFW webs constructed from frass, exuviae, and needle fragments act l i k e heat sinks i n d i r e c t sunlight and raise the body temperatures of the inhabitants, while red pine canopies lower the a i r temperatures outside the webs below screen temperatures. The microclimate i s further modified by the insects' own a c t i v i t i e s . Empirical relationships designed to describe these processes i n pre d i c t i v e models for PFW w i l l not be very accurate i f the degree of d e f o l i a t i o n i n the stand i s not taken into account. 125 The duration of simulated l a r v a l drop was considerably longer than observed. This discrepancy appeared at the end of the l a r v a l development, suggesting that problems may s t i l l exist i n the v a r i a b i l i t y functions developed from constant-temperature laboratory rearings. Simulations using only the fastest 90 percent of the population p a r t i a l l y a l l e v i a t e d t h i s problem. Because f o l i a g e q u a l i t y i n the laboratory deteriorated as the larvae grew, the development of the slower indiv i d u a l s may have been further retarded by q u a l i t a t i v e changes i n t h e i r food. This factor may also explain the d i s p a r i t y between predicted and observed l a r v a l drops for males and females, since females took longer to develop than males i n the laboratory. Larvae of PFW, unlike the eggs and subterranean stages, can a c t i v e l y respond to changes i n thermal conditions. Larvae can move to avoid c r i t i c a l high temperatures and t h e i r behavior also may improve t h e i r developmental rate. In t h i s type of simulation, the thermal maximum for development i s a sensitive parameter (Gold e_t a l - 1987). The estimated thermal maximum for l a r v a l PFW i s very near the high temperature at which active avoidance i s e l i c i t e d , so that errors i n estimating t h i s parameter could have serious e f f e c t s on model predictions. Since the current version of the model uses mean temperature deviations, not extremes, thermal maxima would ra r e l y be encountered i n the simulations. 126 Chapter 5 -Biology and phenology of Sinophorus megalodontis and Olesicampe sp. (Hymenoptera: Ichneumonidae), parasitoids of pine false webworm. 5.1 Introduction A p a r a s i t o i d must coincide temporally and s p a t i a l l y with i t s host to cause s i g n i f i c a n t mortality of the host population ( G r i f f i t h s 1969, Huffaker e_t a l . 1977). Asynchrony between host and par a s i t o i d has been suggested as the cause of f a i l u r e of many b i o l o g i c a l control programs (Hoy 1976, Huffaker e_t a l . 1977, Ehler and Andres 1984), as has releasing adult parasitoids when suitable hosts were unavailable ( G r i f f i t h s and Lyons.1968, Hoy 1976). I n a b i l i t y of parasitoids to exploit hosts i n a l l host habitats can also result i n f a i l u r e s ( M i l l e r 1983). Several insect parasitoids attack PFW i n Europe (Schwerdtfeger 1941, 1944, Rumphorst and Goosen 1961, Kolomietz 1967, Schmutzenhofer 1975). The only record of parasitism of PFW i n North America i s Barron's (1981) description of a new species of Ctenopelma (Ichneumonidae) ovipositing i n eggs of PFW. Two species of ichneumonids were reared from the sawfly host during .the present investigation. Sinophorus megalodontis Sanborne belongs to a species group that attacks web-spinning sawflies (Sanborne 1984). Another member of the species group, S_. crassifemur (Thomson), attacks PFW larvae and related species i n the Palea r c t i c region (Schwerdtfeger 1944). Only the 127 b i o l o g i e s o£-rS_. validus and S_. t u r i o n i s (S_. rufifemur) have been studied i n any d e t a i l (Timberlake 1912, T o t h i l l 1922, J u i l l e t 1959, Morris 1976). Another ichneumonid reared from PFW i s an undescribed species (H. Townes pers. comm.) of the poorly known genus Olesicampe ( B i l l a n y ejb al. 1985). Some species of t h i s very large cosmopolitan genus are responsible for s i g n i f i c a n t mortality of sawfly populations (Price and Tripp 1972, B i l l a n y et al. 1985) and some have been successfully introduced as b i o l o g i c a l control agents against sawflies (Turnock and Muldrew 1971, Quednau and Lim 1983, Drooz e_t al. 1985). The objective of t h i s investigation was to study the l i f e h i s t o r y and the temporal and s p a t i a l a c t i v i t y patterns r e l a t i v e to the host of these ichneumonids, and to estimate parameters for developing temperature-dependent development rate and ageing rate functions to be incorporated into a predictive phenology model. Methods for d i f f e r e n t i a t i n g the immature stages of the two species were investigated. Three methods of determining the phenology of the adult parasitoids were compared. 5.2 Materials and Methods Description of Immature Stages. Egg morphology was determined by dissecting adult females of the two parasitoids species. Eggs dissected from host larvae were compared with these eggs. The morphology of the two species of parasitoids 128 was determined-by examining larvae that had p a r t i a l l y emerged from the egg and by associating S_. megalodontis larvae with empty chorions. Parasitoid eggs and l a r v a l head capsules were measured with an ocular micrometer mounted on a stereo microscope. Adult Emergence and F l i g h t Period. The temporal pattern of p a r a s i t o i d emergence from the s o i l was determined using traps (Section 2.2). Traps were examined d a i l y during the parasite emergence period i n late May and early June. Adult parasitoid a c t i v i t y was monitored with Malaise traps (D.A. Focks Ltd., G a i n e s v i l l e F l o r i d a ) . Traps were positioned at the edge of a clearing adjacent to plot AMI i n 1983 and i n the heavily- and moderately-defoliated zones within the LA plantation i n 1986. Adults of Olesicampe sp. and £. megalodontis were removed d a i l y at 1000 hours and preserved i n 70% ethanol for l a t e r i d e n t i f i c a t i o n . Temporal and S p a t i a l D i s t r i b u t i o n of Parasitism. To determine the host stage attacked and the s p a t i a l d i s t r i b u t i o n of parasitism, a l l the host larvae c o l l e c t e d by branch sampling and l a r v a l drop trapping at the AMI plantation i n 1983 and by branch sampling at the H plot i n 1986 (Section 3.2) were dissected and the incidence of p a r a s i t o i d eggs and larvae was noted. Total number of £•!. megalodontis was determined from the sum of the unhatched eggs plus larvae or hatched eggs, whichever 129 was greater. The number of Olesicampe sp. was estimated as the sum of the eggs iand larvae, since hatched eggs were undetectable. To determine the s p a t i a l d i s t r i b u t i o n of parasitism, host larvae were pooled for a l l sample days from the three v e r t i c a l strata i n the plantation canopy sampled i n 1983. To determine the eff e c t of time of host drop on incidence of parasitism, samples of dropping larvae were dissected and pooled at three-day i n t e r v a l s for the f i r s t 12 days of the drop period while the l a s t eight days made up the f i n a l sample. Subterranean Development and Adult Longevity of Olesicampe sp. Post-diapause development rates and adult longevity of Olesicampe sp. at constant temperatures were determined using p a r a s i t i z e d PFW larvae c o l l e c t e d i n late autumn from the s o i l of the LA plantation and para s i t o i d adults reared from these larvae. Parasitized larvae are distinguishable from unparasitized hosts because they r e t a i n eonymphal c h a r a c t e r i s t i c s and are distinguished from eonymphs i n prolonged diapause by th e i r yellow coloration. Parasitized larvae were stored as described previously for unparasitized pronymphs (Section 2.2). Upon removal from cold storage the larvae were placed i n 1.9-mL s h e l l v i a l s and incubated at nine constant temperatures (1.8, 4.3, 7.1, 10.6, 15.6, 18.6, 23.3, 27.1 and 29.9°C). Temperature-dependent development rate and v a r i a b i l i t y functions for Olesicampe sp. were determined using nonlinear 130 regression of in d i v i d u a l development rates (Regniere 1984). Freshly emerged adults of Olesicampe sp. were placed i n d i v i d u a l l y i n rearing v i a l s and randomly assigned to eight constant-temperature chambers (4.1, 5.8, 11.0, 15.0, 19.0, 23.0, 27.0 and 31.0°C) to determine the effect of temperature on longevity. V i a l s were made by gluing together the l i d s of two 52-mL p l a s t i c snap-on cap v i a l s . The bottom of one v i a l was removed and replaced with muslin. A hole was d r i l l e d through the l i d s into which an absorbent cotton wick was inserted, running from the upper insect v i a l to the lower w a t e r - f i l l e d v i a l . Honey was smeared on the muslin. Parasitoids were examined d a i l y and time of death recorded. Mean ageing rates ( i . e . , mean of reciprocals of longevity) of p a r a s i t o i d females at each constant temperature was determined and regressed as a li n e a r function of temperature (Section 3.2). Normalized longevity times were obtained by di v i d i n g i n d i v i d u a l longevities by mean longevity for each temperature. The cumulative d i s t r i b u t i o n of normalized times pooled for a l l temperatures was f i t t e d to a two parameter Weibull function (Section 3.2). A model of subterranean development and longevity of Olesicampe sp. was developed (Section 2.2 and 3.2). Overwintered larvae of Olesicampe sp. were assumed to be at an equal stage of development as spring s o i l temperatures began to increase. For modeling purposes, subterranean development of Olesicampe sp., i n spring, consisted of: 1) post-diapause development of the larvae of Olesicampe sp. within the host and 2) emergence of the parasitoid larvae from the host, cocoon formation and adult emergence. Two types of s o i l temperature data were used i n the subterranean development model: average hourly s o i l temperatures recorded at a 5-cm depth at the LA plantation i n 1986 and s o i l temperatures at s i x depths, estimated from s o i l temperatures recorded at two depths (5 and 10 cm) from the same locations (Section 2.2). It was assumed that the v e r t i c a l d i s t r i b u t i o n of p a r a s i t i z e d hosts i n the s o i l was similar to the d i s t r i b u t i o n of unparasitized hosts. Insects i n each 1.5 cm increment of s o i l developed at a rate determined by the estimated temperature for that stratum. For every time step i n the simulation the rate-summation model was simultaneously solved for a l l six s p a t i a l l y d i s t r i b u t e d groups. Then, t o t a l development was apportioned for the entire population. Temperature data from the moderately- and l i g h t l y - d e f o l i a t e d zones of the plantation were employed i n the simulations. The model incremented time i n i n t e r v a l s of 1 or 4 h. Average hourly a i r temperatures were used to estimate longevity i n the model. The number of cohorts simulated i n the longevity model was determined by the number of days that adults emerged from the s o i l . For each time step, mean ageing rate of 132 the cohort was determined and the f r a c t i o n a l age of the cohort was accumulated. The cumulative proportion of the cohort dying was calculated from the Weibull function. The proportion surviving was one minus the proportion dying. The proportion of the population surviving was the product of the proportion of the cohort surviving and the proportion of the population i n the cohort. Observed emergence was used to i n i t i a l i z e the longevity component of the model. Simulated adult longevity was compared to Malaise trap captures of adult parasitoids scaled to match the proportion of l i v i n g adults. 5.3 Results Description of Immature Parasitoids. Eggs of S_. megalodontis were brown, smooth and broader at one end than the other (Fig. 5.1A). Mean length of eggs dissected from host larvae was 0.968 mm (S.D.=0.048, n=30) and mean width at the broadest point was 0.238 mm (S.D.=0.020). Eggs of S_. megalodontis were v i s i b l e through the integuments of the host larvae. Eggs of Olesicampe sp. (Fig. 5.IB) were clear, curved and broader at one end than at the other. Mean length of 30 eggs of t h i s species was 0.566 mm (S.D.=0.041) and mean width was 0.167 mm (S.D.=0.028). Chorions from hatched &. megalodontis were d i s c e r n i b l e . Empty chorions of Olesicampe sp. were never encountered. F i r s t - i n s t a r larvae of both parasitoid species were of the ichneumonid mandibulate-caudate type (Quednau and Lim 1983). F i g . 5.1. Immature stages of PFW parasitoids: (A) la r v a of S_ megalodontis emerging from the egg; (B) egg of Olesicampe sp (C} f i r s t - i n s t a r l a r v a of Olesicampe sp. ; CD) cocoons of S_. megalodontis (right) and Olesicampe sp. ( l e f t ) . 135 Head capsules of S_. megalodontis-.:(Fig. 5.1A) were generally larger than those of Olesicampe sp. (Fig. 5.1C), but with some overlap i n both height and length. However, the diagonal length of the larvae from the t i p of the mouthparts to the dorsal end of the head capsule was consistently longer i n the former species. The head capsule of S_. megalodontis was darker, i t s deeper shade of yellow probably res u l t i n g from a greater degree of s c l e r o t i z a t i o n . In addition, the length of the t a i l i n S. megalodontis was about as long as the body, while the length of the t a i l i n Olesicampe sp. was about half the body length. The cocoon of S_. megalodontis was longer, darker, smoother and more tubular than the Olesicampe sp. cocoon (Fig. 5.ID). Adult Emergence and F l i g h t Period. Parasitoids c o l l e c t e d from the emergence traps were pooled for heavily- and moderately-defoliated zones. Only six individuals, three of each species, were captured i n the l i g h t l y - d e f o l i a t e d zone and were subsequently disregarded. The proportion of females of £. megalodontis c o l l e c t e d was not s i g n i f i c a n t l y d i f f e r e n t from 50% (n=42, %female=52.4, x2=0.10, P>0.05 ), however, c o l l e c t i o n s of Olesicampe sp. contained s i g n i f i c a n t l y more males (n=56, %female=35.7, x2=4.57, P<0.05). Adults of both species were captured i n emergence traps beginning 23 May (day of year 143) (Figs. 5.2A, B). Emergence periods for Olesicampe sp. and S_. megalodontis lasted 16 and 17 days, respectively. Mean (SE) 'emergence of male £• megalodontis preceded emergence of females by 3.1 (1.03) days; males of Olesicampe sp. preceded females by 136 1.6 (0.81) days. There were no s i g n i f i c a n t differences i n time of emergence of comparable sexes for the two parasitoid species (males; t=1.77, df=40, p=0.084: females; t=0.09, df=54, P=0.930). Peak adult emergence coincided with about average in s t a r 1.5 and lasted u n t i l about average instar 3.8 i n 1986 (Fig. 4.2B). Parasitoids from the Malaise traps, i n the moderately d e f o l i a t e d - and heavily-defoliated zones, were pooled. The onset of the observed f l i g h t periods of the two species (Figs. 5.2C, D) coincided with the onset of emergence from the s o i l . Malaise traps c o l l e c t e d s i g n i f i c a n t l y more males than females of both p a r a s i t o i d species (Olesicampe sp.; n=637, %female=30.6, X2=95.78, P<0.05: S_. megalodontis: n=538, %female=16.9, x2=235.57, P<0.05). Although percentages of females i n the emergence traps were greater than i n the Malaise traps for both species, the difference was only s i g n i f i c a n t for S_. megalodontis (Olesicampe sp. x2=0.626, P=0.429; S_. megalodontis x2=31.24, P<0.001). The f l i g h t period for both species lasted 28 days u n t i l 18 June (day 169). The mean (SE) f l i g h t period of male S_. megalodontis preceded the mean f l i g h t period of females by 2.7 (0.55) days; males of Olesicampe sp. preceded females by 2.0 (0.30) days. The f l i g h t periods of the two parasitoids were also s i g n i f i c a n t l y d i f f e r e n t (males; t=9.72, df=887, P<0.001: females; t=5.33, df=284, P<0.001). The depression i n trap c o l l e c t i o n s i n the middle of the f l i g h t period occurred during a period of low temperature and r a i n f a l l . F i g . 5.2. Emergence (A, B) and Malaise (C, D) trap catches parasitoids of PFW from the LA plantation i n 1986. A. 8-6-r-_ l O a < O CC 80-LU m 5 Z 60 -40-I \" I\" 20-142 146 i FEMALE OLESICAMPE i MALE OLESICAMPE i—r ~ i — i — i — i — i — i i FEMALE OLESICAMPE > MALE OLESICAMPE 150 154 158 162 166 170 Jp, FEMALE SINOPHORUS MALE SINOPHORUS 1 1 1 1 1 1 i FEMALE SINOPHORUS i MALE SINOPHORUS 4M LplnlflO | 142 146 150 154 158 162 166 170 DAY OF YEAR CD 139 Coll e c t i o n s from the Malaise trap i n 1983 yielded a maximum of f i v e adults of one sex and species per day. Females of Olesicampe sp. were caught from 15 to 29 June, while the males' f l i g h t extended from 13 June to 6 July. For both sexes of S_. megalodontis. the f l i g h t period lasted from 18 June to 6 July. F l i g h t period of adults was coincident with average host instars 0.9 to 4.8 i n 1983 (Fig. 4.2A). Temporal and S p a t i a l D i s t r i b u t i o n of Parasitism. Unhatched eggs of &. megalodontis were found i n a l l ins t a r s of PFW (Fig. 5.3A) i n 1983 and 1986. The majority of eggs was found i n the fourth i n s t a r i n 1983 and i n the f i r s t i n s t a r i n 1986. S i m i l a r l y , the eggs of Olesicampe sp. were d i s t r i b u t e d among a l l host i n s t a r s i n 1983 (Fig. 5.3B), but were lim i t e d to host i n s t a r s I-III i n 1986. Larvae of both pa r a s i t o i d species were found only i n i n s t a r s II-VI i n 1983, but i n a l l host instars i n 1986. There were encapsulated larvae of S_. megalodontis i n a l l host i n s t a r s , but they occurred mainly i n host ins t a r s IV-VI (Fig. 5.3C). Encapsulated larvae of Olesicampe sp. were found only i n the l a s t three host i n s t a r s (Fig. 5.3D). Eggs of S_. megalodontis were also found i n ultimate host ins t a r s V and VI and eggs of Olesicampe sp. were found i n ultimate host instar V c o l l e c t e d i n drop traps i n 1983. A similar comparison was not avail a b l e for 1986. The t o t a l number of p a r a s i t i z e d PFW larvae increased 140 F i g . 5.3. Mean numbers (±SE) of parasitoid. eggs (A, B) and encapsulated larvae (C, D) per PFW l a r v a l i n s t a r c o l l e c t e d by branch sampling at Anten M i l l s i n 1983 (A, C) and Lakehurst i n 1986 (B, D). 0.10 0.08-0.04-0.02-0.00 0.025-1 0.020-0.01S-0.010-0.005-0.000-0.20 0.16-0.12 0.08-0.04 0.025-0.020-0.015-0.010-0.00S-II III IV V VI INSTAR 142 dramatically (Fig. 5.4) throughout the host larval-. megalodontis was not s i g n i f i c a n t l y d i f f e r e n t i n the upper two st r a t a . There were no s i g n i f i c a n t differences i n percent parasitism for pooled samples c o l l e c t e d from two heights i n 1986 (Table 5.1). Host larvae were concentrated i n the high s t r a t a i n 1983, but the number of hosts co l l e c t e d from the two s t r a t a were approximately the same i n 1986. The greatest concentration of parasitism by £• megalodontis was i n the southern and western quadrants of the trees (Table 5.2). The eastern quadrant had the lowest incidence of parasitism, while the northern aspect was intermediate. The percentage of hosts p a r a s i t i z e d , however, was only s i g n i f i c a n t l y lower i n the eastern quadrant. The incidence of Olesicampe parasitism was s i g n i f i c a n t l y greater i n the western quadrant 143 Fig. 5.4. Mean numbers (±SE) of t o t a l immature parasitoids, p a r a s i t o i d eggs, and encapsulated pa r a s i t o i d larvae per PFW larvae c o l l e c t e d by branch sampling at Anten M i l l s i n 1983 (A, C) and Lakehurst i n 1986 (B, D;. The numbers over the symbols are number of host larvae. Table 5.1. Mean number of parasitoids per host (x) and percent p a r a s i t i z e d of PFW larvae c o l l e c t e d by branch sampling from v e r t i c a l s t r a t a at Anten M i l l s (AMI) i n 1983 and Lakehurst (H) i n 1986. Year Height NO. hosts SE <* i a SE G 2 P S. megalodontis 1983 Leader 2870 0. 037 0. 004 3. 2a 0. 3 92. ,031 <0. 005 High 9655 0. 051 0. 003 4. l a 0. 2 Low 3750 0. 128 0. ,010 7. 8b 0. ,4 1986 High 2129 0. , 116 0. 009 9. , 8a 0. ,7 3. .578 >0. ,05 Low 2014 0. .089 0. .007 8. . l a 0. .6 Olesicampe sp 1983 Leader 2870 0. .009 0, .002 0. . 8a 0, .2 49 .844 <0 .005 High 9655 0, .019 0. .001 1, .8b 0 . 1 LOW 3750 0, .033 0, .003 3 . l c 0 . 3 1986 High 2129 0 .029 0 .004 2 . 9a 0 .4 0 . 164 >0 .05 LOW 2014 0 .031 0 .004 3 . l a 0 .4 Percentages within the same year and within a species followed by the same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (G 2, P>0.05) Table 5.2. Mean number of parasitoids per host (x) and percent p a r a s i t i z e d of PFW larvae c o l l e c t e d by branch sampling from cardinal d i r e c t i o n s at Anten M i l l s (AMI) i n 1983. NO. Aspect hosts X SE b a SE G 2 P S. megalodontis North 3506 0. 068 0. 005 5. ,4a 0.4 74.266 <0.005 East 4187 0. 040 0. 004 2. ,9b 0.3 South 3713 0. ,095 0. ,007 6. . 8a 0.4 West 1999 0. , 108 0. ,013 6. . l a 0.5 Olesicampe sp. North 3506 0. .025 0. .003 2, . 5ab 0.3 10.969 <0.05 East 4187 0. .017 0. .002 1, ,6b 0.2 South 3713 0. .022 0, .002 2 . lab 0.2 West 1999 0. .032 0, .005 2 .7a 0.4 Percentages within a species followed by the same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (G 2, P>0.05) 146 compared with the eastern quadrant. Fewer host larvae were . c o l l e c t e d from the western aspect of the host trees. The period of PFW l a r v a l drop lasted for 20 days (Section 4.3). The mean number of parasitoids (Table 5.3) of each species per host, from both sexes of the host larvae, increased over time, so that later-dropping larvae exhibited a greater incidence of parasitism. For the enti r e l a r v a l drop period, the proportion of p a r a s i t i z e d larvae was si m i l a r between the sexes. The proportion (0.184, SE=0.013, n=894) of male PFW larvae p a r a s i t i z e d by S_. megalodontis was not s i g n i f i c a n t l y d i f f e r e n t than the proportion (0.163, SE=0.019, n=367) of females p a r a s i t i z e d (G2=0.797, df=l, P>0.05). Similarly, the proportion of males (0.060, SE=0.008, n=894) and females (0.065, SE=0.013, n=367) p a r a s i t i z e d by Olesicampe sp. d i d not s i g n i f i c a n t l y d i f f e r (G2=0.143, df=l, P>0.05). The effectiveness of both p a r a s i t o i d species was lim i t e d by superparasitism, multiparasitism and encapsulation by the host. Although both parasitoids species superparasitized hosts, the occurrence of superparasitism was greatest for S_. megalodontis. The proportion (SE) of pa r a s i t i z e d S_. megalodontis. co l l e c t e d by branch sampling i n 1983, branch sampling i n 1986, and l a r v a l drop trapping i n 1983, that were superparasitized were 0.185 (0.013), 0.107 (0.015), and 0.247 (0.024), respectively, while the proportion of Olesicampe sp. that were superparasitized were 0.017 (0.004), 0.000 (0.000), and 0.006 (0.005), respectively. Table 5.3. Mean number of parasitoids per host (x) and percent p a r a s i t i z e d of PFW larvae c o l l e c t e d from drop traps at Anten M i l l s (AMI) i n 1983. NO. Dates hosts x SE , a SE G 2 P S. megalodontis - male 184-186 293 0.105 0. 019 9. 9a 1. 7 63. 984 <0. 05 187-189 423 0.251 0. 035 16. 5ab 1. 8 190-192 102 0.363 0. 079 26. 5bc 4. 4 193-195 56 0.741 0. 163 48. 2cd 6. 7 196-203 20 1.050 0. 276 60. Od 11. 0 S. megalodontis - female 184-186 62 0.097 0. 044 8. l a 3. 5 38. 479 <0. 05 187-189 187 0.160 0. 040 10. 6a 2. 3 190-192 45 0.089 0. 043 8. 9a 4. 7 193-195 41 0.732 0. 201 41. 4b 7. 7 196-203 32 0.567 0. 149 43. 8b 8. 8 Olesicampe sp. - male 184-186 293 0.027 0. ,010 2. 7a 1. ,0 12. 326 <0. 05 187-189 423 0.069 0. ,013 6, ,6a 1. ,2 190-192 102 0.088 0. ,028 8, , 8ab 2. ,8 193-195 56 0.107 0. ,042 10. ,7ab 4. , 1 196-203 20 0.120 0, .066 15, .Ob 8. .0 Olesicampe sp. - female 184-186 62 0.032 0 .023 3 . 2a 2 .2 8. .032 >0. ,05 187-189 187 0.048 0 .016 4 . 8a 1 .6 190-192 45 0.067 0 .038 6 .7a 3 .7 193-195 41 0.195 0 .072 16 .7a 5 .8 196-203 32 0.094 0 .052 9 .4a 5 .2 a Percentages within a species and sex and followed by the same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (G 2, P>0.05) 148 Host larvae contained up to nine immature S_. megalodontis. The greatest number of Olesicampe sp. found i n one host was three. The proportion (SE) of p a r a s i t i z e d PFW larvae from the same c o l l e c t i o n s that were multiparasitized were 0.165 (0.012), 0.125 (0.016), and 0.172 (0.022), respectively. Subterranean Development and Adult Longevity of Qi epi 27.1°C) and low (1.8°C) temperatures. Larvae were capable of cocoon formation at 5.8°C and successfully emerged after transfer to 15.6°C. For these 149 parasitoids, development periods were estimated using the i n t e r p o l a t i o n technique (Section 2.2). Larvae allowed to spin cocoons at 15°C and moved to 27.1°C f a i l e d to emerge. Equation 2. l i b adequately described both within host (Fig.. 5.5A) and within cocoon (Fig. 5.5B) development rates of Olesicampe sp. Estimated parameters were Pi=0.327, P2=2.546, P3=5.568, P4=0.017, Tm=31.0, and Tb=1.0 (R2=0.848) for within host development and Px=0.085, P2=2.285, P3=4.411, P4=0.015, Tm=25.0 and T]-,=5.0 (R 2 =0.848) for within cocoon development. V a r i a b i l i t y functions for development of these stages were also developed as described previously (Section 2.2). Parameters for Equation 2.12a were K=6.373, C=0.753, and Q=0.365 (R2=0.997) for within host development and K=44.928, 0=1.017, Q=2.511 (R2=0.999) for within cocoon development. The temperature-dependent ageing rate function (Fig. 5.6A) for adults of Olesicampe sp. was: y = - 0.034 + 0.011 x ( r 2 = 0.784) where y = ageing rate and x = temperature C O (F=21.8; df=l,7; P=0.003). The parameter estimates for the Weibull function describing the cumulative d i s t r i b u t i o n of normalized times to death (Fig. 5.6B) were T?=0.997 and 0=1.471 (R2=0.991). Predicted emergence from the moderately-defoliated zone preceded the emergence from the l i g h t l y d e f o l i a t e d zone by seven and eight days for larvae and adults, respectively (Figs. 5.7A, B). Deviations of predicted emergence from observed emergence, 150 Fig . 5.5. Post-diapause temperature-dependent development rates for (A) within host and (B) within cocoon development of Olesicampe sp. C i r c l e s are i n d i v i d u a l observed rates and l i n e s are regressions curves (Equation 2.11b). TEMPERATURE (°C) 152 Fig . 5.6. (A) Temperature-dependent ageing rate of Olesicampe adult females as a function of temperature and (B) cumulative d i s t r i b u t i o n function of normalized death times. C i r c l e s are (A) observed mean ageing rates (±SE) and (B) cumulative mortality at each normalized time. Regression l i n e s are indicated. T T 1 1 1 1 1 1 1 1 1 1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 TEMPERATURE C C ) NORMALIZED TIME (TIME/MEAN TIME) 154 Fig . 5.7. Simulated development of subterranean stages of Olesicampe sp. from moderately- (A) and l i g h t l y - d e f o l i a t e d (B) plots of LA plantation i n 1986. S o l i d l i n e s (A, B) represent res u l t s using temperatures from one depth while the dashed l i n e s represent summed development of insects at several depths i n s o i l . F i r s t curves are emergence of p a r a s i t o i d larvae from hosts and second curves are emergence of adults from s o i l . C i r c l e s are observed emergence of adults from s o i l . (C) Simulated r e l a t i v e abundance ( s o l i d l i n e ) of females and Malaise trap captures scaled to size (dotted l i n e ) . 155 1 0 0 115 130 145 160 175 DAY OF YEAR 156 i n the moderately-defoliated zone was 3.2 days for 50% of the population i n the single-stratum and 2.5 days i n the multiple-stratum simulations. Increasing the time increment of the multiple-stratum model from 1 to 4 h had no e f f e c t on the deviations. Since emergence of p a r a s i t o i d larvae from the hosts occurs i n s o i l and parasitoids were rare i n the s o i l , t h i s event was not observed d i r e c t l y . Low numbers of emerging adults i n the l i g h t l y d e f o l i a t e d zone also prevented comparisons between observed and predicted emergence. The longevity models predictions (Fig. 5.7C) c l o s e l y approximated the observed f l i g h t period of adults of Olesicampe sp. However, weather affected f l i g h t behavior, r e s u l t i n g i n peaks and troughs of trap captures throughout the adult period, obscuring the pattern of longevity. 5 . 4 Discussion Sinophorus megalodontis and Olesicampe sp. are s o l i t a r y endoparasitoids of PFW and, l i k e t h e i r hosts, are univoltine. Unlike Olesicampe lop h r y i . which overwinters i n i t s host, Neodiprion swainei. as an egg or f i r s t i n s t a r larva (McLeod 1975), Olesicampe sp. overwinter i n the host larvae as l a t e - i n s t a r larvae. The p a r a s i t o i d larvae prevent the hosts from transforming into pronymphs. Larval development i s completed i n the spring when larvae emerge from host cadavers. A l l that remains of the host i s the empty c u t i c l e . The p a r a s i t o i d larvae spins a cocoon i n the earthen c e l l , pupates 15? and emerges as an adult. Sinophorus megalodontis emerges from the host i n f a l l and overwinters i n i t s own cocoon within the host's earthen c e l l . This overwintering mode was also reported i n S_. validus (Morris 1976). Parasitoid adults, l i k e t h e i r sawfly hosts, emerge protandrously from the s o i l as seen i n both emergence and Malaise traps. The emergence and f l i g h t periods of the two ichneumonid species were contemporaneous, despite the d i f f e r e n t strategies for subterranean development. The f l i g h t period of S_. megalodontis adults was somewhat longer than that of Olesicampe sp. The higher proportion of males i n Malaise traps, compared with emergence traps, r e f l e c t e d sex-related differences i n f l i g h t behavior. Males spent t h e i r time searching near the ground for females, while females searched the trees. Malaise traps positioned near the ground intercepted a greater number of f l y i n g males. A si m i l a r bias was reported i n f l i g h t trap captures of cocoon parasitoids of N. swainei (Price 1971). There was considerable year-to-year v a r i a t i o n i n the date of adult emergence, as there was with host development. The onset of emergence i n 1986 of both pa r a s i t o i d species was synchronized with the host's early i n s t a r s and with l a t e r i n s t a r s i n 1983. Dissections of parasitoids from host larvae confirmed that yearly vari a t i o n s i n host-parasitoid temporal coincidence occur, r e s u l t i n g i n differences i n host stage attacked. 158 The females of both species oviposit into a l l instars of host larvae. Parasitoid eggs hatch soon after deposition, as evidenced by the presence of pa r a s i t o i d larvae i n f i r s t - i n s t a r host larvae. Larvae remain as f i r s t i n s t a r s u n t i l the l a s t i n s t a r of the host. Adults mated and began ovipositing soon af t e r emergence as indicated by the coincidence of the onset of emergence and the occurrence of pa r a s i t i z e d host larvae. Pa r a s i t o i d attacks are d i s t r i b u t e d equally between the sexes of the host. Thus, the parasitoids do not affect the sex ra t i o s observed i n PFW populations. The synchrony of Olesicampe sp. with s p e c i f i c host stages i s not as consistent as i n other species of Olesicampe attacking early i n s t a r s of t h e i r hosts (Ives e_t a l . 1968, McLeod 1975, Thompson and Kulman 1976, B i l l a n y and Brown 1980). Similarly, the stage attacked by j&. megalodontis i s variable, unlike S_. validus (Morris 1976) and i t s s i b l i n g species, &. crassifemur. which also attacks Acantholyda spp. (Kolomietz 1967). The majority of species of Sinophorus that attack webspinning insects have long ovipositors (Sanborne 1984). Sinophorus megalodontis and i t s P a l e a r c t i c counterpart £. crassifemur have long ovipositors and are well adapted for attacking l a t e i n s t a r s i n webs. Sinophorus crassifemur has been observed attacking t h i r d and fourth in s t a r s of the congeneric species A. p o s t i c a l i s (Kolomietz 1967). Silkspinning behavior of the sawfly and long ovipositors of parasitoids are probably r e v o l u t i o n a r y responses to s i m i l a r insects. Olesicampe sp. does not have a comparably 159 extensive ovipositor and i s probably better suited for attacking e a r l i e r i n s t a r s that are not well concealed i n t h e i r webs. Females of Olesicampe sp. are apparently unable to discriminate between previously p a r a s i t i z e d hosts ( i . e . , superparasitism) and hosts attacked by S_. megalodontis ( i . e . , multiparasitism). Encapsulation of the par a s i t o i d larvae by hemolymph inclusions of the host was common. Encapsulation, regardless of the degree of host synchrony, was mainly r e s t r i c t e d to la t e i n s t a r s of the host. Parasitized hosts may develop more slowly than unparasitized insects ( M i l l e r 1983). Larvae of the European pine sawfly, N. s e r t i f e r . developed slower when pa r a s i t i z e d by Lophyroplectus luteator ( G r i f f i t h s 1975). Sinophorous validus had l i t t l e e f f e c t on the development rate of i t s host H. cunea (Morris and Bennett 1967). The increasingly greater incidence of parasitism i n l a t e r dropping host larvae i n 1983 suggested that development of PFW larvae may be retarded by p a r a s i t i z a t i o n . This observation i s confounded by the coincidence of attacking parasitoids and late i n s t a r s of PFW i n that year and the p o s s i b i l i t y that e a r l i e r developing larvae escaped parasitism. The high proportion of l a t e r dropping larvae that were parasitized, compared with the proportion p a r a s i t i z e d observed during branch sampling, supports the developmental retardation supposition. 160 The d i s t r i b u t i o n of i n d i v i d u a l parasitoid species within forest canopies d i f f e r s with respect to microclimate (Weseloh 1976). PFW larvae were seldom seen on branches outside of webs, unless they had completely defoliated the branch or u n t i l they were ready to drop. Both parasitoid species were most abundant i n the high stratum and least abundant i n the leader and f i r s t whorl of branches and intermediate i n the lower stratum. The d i s t r i b u t i o n of parasitoids i s influenced by the d i s t r i b u t i o n of the host larvae. In the heavily d e f o l i a t e d plantation sampled i n 1986, the host larvae were equally d i s t r i b u t e d i n the high and low st r a t a and the parasitoids were also equally d i s t r i b u t e d among the hosts. Parasitized hosts were f a i r l y evenly d i s t r i b u t e d around the sides of the tree. Larvae of Olesicampe sp. were not capable of cocoon-forming behavior at the two lowest temperatures used i n the constant temperature rearings. In variable temperatures i n the f i e l d , larvae probably survive these temperatures and form cocoons during periods of higher s o i l temperature. Observed p a r a s i t o i d emergence was within 3 to 5 days of predicted emergence. Incorporating v e r t i c a l d i s t r i b u t i o n of temperatures i n the s o i l into the model reduced the deviations between observed and predicted emergence by about one day. Reducing the number of time steps u t i l i z e d i n the model had no ef f e c t on the deviations between observed and expected events. 161 The longevity component of the par a s i t o i d model provided close agreement between the observed f l i g h t period of females of Olesicampe sp. and the simulated longevity. This occurred even when other sources of mortality were not included i n the model and the model was based on par a s i t o i d longevity i n the laboratory i n the absence of hosts. The transcontinental d i s t r i b u t i o n of S_. megalodontis (Sanborne 1984) suggests that t h i s species i s endemic to North America. Unidentified species of Olesicampe and Sinophorus have been reported from species of Cephalcia i n Canada (Eidt 1969). It i s l i k e l y that the species of Olesicampe described here originated i n the New World. Even though members of these genera attack PFW i n the Old World, these Nearctic species are poorly adapted to t h e i r recently introduced host. The evidence presented here suggests that these parasitoids, as a resu l t of lack synchronization with the host, poor searching a b i l i t y , and encapsulation are not having much impact on the host populations. Techniques developed here w i l l be useful for evaluating pote n t i a l candidates for future b i o l o g i c a l control of PFW. The developed models w i l l a s s i s t i n synchronizing releases of potential b i o l o g i c a l agents. 162 Chapter 6 General Conclusion The a b i l i t y to predict temporal occurrence of stages of PFW has pragmatic implications for control of t h i s pest species. Although there are no i n s e c t i c i d e s registered for control of PFW, Christmas tree growers apply chemicals to eradicate t h i s sawfly i n Scots pine plantations. For e f f e c t i v e control, sprays should be applied when susceptible stages are available. Spray 'windows' are narrow due to the c r y p t i c behavior of the larvae and a knowledge of the sawfly's phenology i s required for e f f e c t i v e control. In addition, b i o l o g i c a l control agents such as entomophagous insects must be released when suitable stages are present. Understanding phenology also imparts an understanding of processes at the population l e v e l . To investigate the phenology of PFW a systems approach was adopted. Process-oriented phenology models were constructed to predict the development of stages of the sawfly. The existence of subterranean and arboreal stages, and the shelter-building habits of the larvae made PFW i n red pine plantations an in t e r e s t i n g system i n which to study the e f f e c t s of microclimate on phenological predictions. Models of phenology of PFW and one of i t s parasitoids were developed from relationships between PFW s p a t i a l d i s t r i b u t i o n and microweather. Protandrous emergence of adults from the s o i l , observed i n 163 natural populations, resulted from d i f f e r e n t i a l rates of development between the sexes for post-diapause pronymphs and pupae. D e f o l i a t i o n caused by PFW increased the s o i l ' s exposure to solar r a d i a t i o n r e s u l t i n g i n higher s o i l temperatures and a corresponding reduction i n development time of prepupae and pupae. Consequently, adult emergence was accelerated by about one week between habitat extremes. PFW occurred at depths of 0-9 cm i n the s o i l . No differences were detected between v e r t i c a l d i s t r i b u t i o n s of the sexes, nor were differences detected between s i t e s . Developmental rate differences and lack of microhabitat differences for the sexes indicated that a l t e r a t i o n of the microclimate by d e f o l i a t i o n would not affect the sequence of adult emergence. Development of subterranean stages of PFW was simulated from rate-summation models developed from nonlinear regression equations describing the re l a t i o n s h i p between temperature and rate of development of subterranean stages. Developmental v a r i a b i l i t y was described by f i t t i n g nonlinear regressions to cumulative development as a function of normalized rates. Adult emergence was s a t i s f a c t o r i l y predicted from s o i l temperatures recorded at 5 cm. A model was developed that accurately estimated s o i l temperatures at multiple depths from s o i l temperatures recorded at two depths. P r e d i c t a b i l i t y was enhanced s l i g h t l y when the d i s t r i b u t i o n of insects and temperature of the s o i l were incorporated into the model. Increasing the time increment used i n the model from 1 to 4 h 164 di d not adversely affect p r e d i c t a b i l i t y . Female PFW mate and begin ovipositing soon after emergence from the s o i l and the majority of PFW eggs were mature and ready for deposition at female emergence. Potential fecundity of PFW at emergence from the ground was accurately predicted from l i n e a r relationships with adult wet and dry weights. However, laboratory experiments indicated that PFW females mature some eggs following emergence from the ground. Increased degrees of d e f o l i a t i o n by the sawfly resulted i n reductions i n size of adults emerging from def o l i a t e d zones. Consequently, the e a r l i e s t emerging females had the lowest p o t e n t i a l fecundity. Fecundity of PFW was independent of temperature over the range of temperatures examined. Oviposition and ageing rates of PFW were temperature dependent. Females of PFW are diurnal and ov i p o s i t i o n occurred during daylight hours. Oviposition pattern of PFW was also described by a model based on temperature-dependent oviposition and ageing rate functions. Although the model was of value for evaluating population processes, o v i p o s i t i o n pattern i n the f i e l d was more accurately represented by the emergence pattern of adult females from the s o i l . The e f f e c t of l a r v a l web construction on the development of arboreal stages was investigated. When exposed to sunlight, the web traps heat and raises the body temperature of i t s 165 inhabitants. Larvae of PFW behaviorally thermoregulate and avoid c r i t i c a l high temperatures by vacating webs. A model was developed and used to. examine the significance of the web microclimate for development of larvae. Relationships between web temperatures, canopy temperatures and standard meteorological methods were developed to permit using data from standard weather stations to drive the model. A small dynamic temperature gradient existed within the v e r t i c a l canopy strata, but the net difference i n heat accumulation between st r a t a was minimal. Observed differences i n development of larvae of PFW i n d i f f e r e n t canopy s t r a t a were therefore attributed to varying degrees of exposure of webs to solar radiation. Egg development was not affected by solar r a d i a t i o n since egg development was accurately predicted using canopy temperatures. Larval development increased by 1.4 to 2.8 d when estimated web temperatures were incorporated into the model, while development was retarded by 2.6 to 4.0 d when canopy temperatures were used instead of meteorological screen temperatures. As a prerequisite to b i o l o g i c a l control of t h i s sawfly i t was deemed necessary to know what parasitoids attack Ontario populations of the sawfly, t h e i r biology and how e f f i c i e n t they were at reducing host populations. Understanding the biology of the e x i s t i n g p a r a s i t o i d fauna required that immature stages of the parasitoids be distinguishable. Coincidence of host and p a r a s i t o i d stages i n time and space are c r i t i c a l for successful population regulation. Thus, an i n v e s t i g a t i o n of the phenology of the parasitoids was undertaken. 166 Two ichneumonid parasitoids, Sinophorus megalodontis and an undescribed species of Olesicampe were reared from eonymphs of PFW. The species are s o l i t a r y endoparasitoids and univoltine. Although both species have adopted d i f f e r e n t strategies for subterranean development, the phenology of adults was contemporaneous. The two species of parasitoids were d i f f e r e n t i a t e d by morphology of eggs, f i r s t - i n s t a r larvae and cocoons. Malaise and adult emergence traps indicated the pattern of adult emergence. Malaise traps c o l l e c t e d many more adults and also provided information on adult f l i g h t period, but were biased towards captures of males. Dissections of host larvae indicated that both parasitoids oviposit i n a l l instars of the host, and temporal and s p a t i a l variations occur i n synchronization with host i n s t a r s . Effectiveness of the parasitoids i n c o n t r o l l i n g host populations was also l i m i t e d by encapsulation, and multi- and superparasitism. There was some in d i c a t i o n that parasitism resulted i n retardation of host development. A pr e d i c t i v e model for subterranean stages and adults of Olesicampe sp. was developed. Predicted emergence was within 3-5 days of observed emergence. Incorporating v e r t i c a l s o i l d i s t r i b u t i o n into the model only s l i g h t l y enhanced the predictions. Reductions i n the number of time steps i n the model did not change predictions. ' 167 Pine f a l s e webworm i s a suitable candidate for b i o l o g i c a l control attempts. Few North American parasitoids attack t h i s introduced species and established parasitoids are i n e f f e c t u a l at reducing PFW populations. The biologies of native parasitoids are known and sampling methodologies for assessing success are developed. Phenological models have been developed to accurately predict development of a l l l i f e stages of PFW during the growing season. Thus, introductions can be made when suitable host stages are available. Rate-summation models provided accurate estimates of phenological events i n the l i f e h i story of PFW. Development as well as ageing and ovip o s i t i o n could be described using temperature-dependent rate regressions. Inherent v a r i a b i l i t y i n development was described as a function of normalized rates. Oviposition and ageing because of t h e i r greater v a r i a b i l i t y were more accurately described as functions of normalized time. For most stages, incorporation of empirical functions r e l a t i n g microclimate to meteorological standards greatly increased computational complexity and only s l i g h t l y enhanced predictions. 168 Literature Cited Baker, C.R.B. 1980. Some problems i n using meteorological data to forecast the timing of insect l i f e cycles. E.P.P:0. B u l l . 10: 83-91. Baker, C.R.B. 1981. Accumulated temperatures i n some pest habitats i n 1976-78. Agric. Meteorol. 23: 245-253. Barbosa, P., M.G. Waldvogel, and N.L. Breisch. 1983. Temperature modification by bags of the bagworm Thyridopteryx ephemeraeformis (Lepidoptera: Psychidae). Can. Entomol. 115: 855-858. Barron, J.R. 1981. The Nearctic species of Ctenopelma (Hymenoptera, Ichneumonidae, Ctenopelmatinae).Naturaliste can. 108: 17-56. Bassett, J.R. 1984. Red pine plantation management i n the Lake States: a review. Univ. Mich. IFSIM Publ. No. 3. 27p. Beck, S.D. 1982. Thermoperiodic induction of l a r v a l diapause i n the European corn borer, O s t r i n i a nubilalus. J. Insect Physiol. 28: 273-277. Beck, S.D. 1983. Insect thermoperiodism. Ann. Rev. Entomol. 28: 91-108. Berenbaum, M. 1978. T o x i c i t y of a furanocoumarin to armyworms: a case of biosynthetic escape from insect herbivores. Science 201: 532-533. Bi l l a n y , D.J. and R.M. Brown. 1980. The web-spinning larch sawfly Cephalcia l a r i c i p h i l a Wachtl. (Hymenoptera: Pamphiliidae) a new pest of Larix i n England and Wales. Forestry 53: 71-80. B i l l a n y , D.J., T.G. Winter, and I.D. Gauld. 1985. Olesicampe monticola (Hedwig) (Hymenoptera: Ichneumonidae) redescribed together with notes on i t s biology as a parasite of Cephalcia l a r i c i p h i l a (Wachtl) (Hymenoptera: Pamphiliidae). B u l l . Entomol. Res. 75: 267-274. Charles P.-J. and H.Chevin. 1977. Note sur l e genre Acantholyda (Hymenopteres-Symphytes-Pamphiliidae), et plus particulierement sur Acantholyda hieroglyphica ( C h r i s t . ) . Rev. For. Franc. 29: 22-26. C o l l i e r , H. and S. Finch. 1985. Accumulated temperatures for predicting the time of emergence i n the spring of the cabbage root f l y , D e l i a radicum (L.) (Diptera: Anthomyiidae). B u l l . Entomol. Res. 75: 395-404 Curry, G.L. and R.M. Feldman. 1987. Mathematical foundations of 169 population dynamics. Texas ASfM University Press, College Station. Curry, G.L., R.M. Feldman, and P.J.H. Sharpe. 1978. Foundations of stochastic development. J. Theor. B i o l . 74: 397-410. Dennis, B., W.P. Kemp, and R.C. Beckwith. 1986. Stochastic model of insect phenology: estimation and t e s t i n g . Environ. Entomol. 15: 540-546. Dixon W.J. (ed.) 1983. BMDP s t a t i s t i c a l software. Univ. C a l i f . Press. Berkeley, C a l i f o r n i a . Drooz, A.T., J.W. Quimby, L.C. Thompson, and H.M. Kulman. 1985. Introduction and establishment of Olesicampe benefactor Hinz (Hymenoptera: Ichneumonidae), a parasite of the la r c h sawfly, Pristiphora e r i c h s o n i i (Hartig) (Hymenoptera: Tenthredinidae}, i n Pennsylvania. Environ. Entomol. 14: 420-423. Ehler, L.E. and L.A. Andres. 1984. B i o l o g i c a l control: exotic natural enemies to control exotic pests, pp. 395-418 In Exotic plant pests and North American agriculture. Eidt, D.C. and J.R. McPhee. 1963. Acantholyda erythrooephala (L.) new i n Canada. Dep. For. Bi-mon. Prog. Rep. 19(4): 2. Ghent, A.W. 1958. Studies on the feeding orientation of the jack pine sawfly, Neodiprion p r a t t i banksianae. Roh. Can. J. Zool. 36: 175-183. G i l b e r t , N., A.P. Gutierrez, B.D. Frazer, and R.E. Jones. 1976. E c o l o g i c a l relationships. W.H. Freeman and Co., San Francisco. Gold, H.J., W.L. Kendall, and P.L. Shaffer. 1987. Nonlinearity and the e f f e c t s of microclimatic v a r i a b i l i t y on the codling moth population (Cydia pomonella'): a s e n s i t i v i t y simulation. Ecol. Modelling 37: 139-154. Green, G.W. 1968. Weather and insects, pp. 81-112. In W.P. Lowry [ed.], Biometeorology. Proc. 28th Ann. B i o l . C o l l . , Oregon State Univ. Press. G r i f f i t h s , K.J. 1959. Observations on the European pine sawfly, Neodiprion s e r t i f e r (Geoff.), and i t s parasites i n southern Ontario. Can. Entomol. 91: 501-512. G r i f f i t h s , K.J. 1969. The importance of coincidence i n the functional and numerical responses of two parasites of the European pine sawfly, Neodiprion s e r t i f e r . Can. Entomol. 101: 673-713. G r i f f i t h s , K.J. 1975. The biology of Lophyroplectus luteator 170 (Hym.: Ichneumonidae) on Neodiprion s e r t i f e r (Hym.: Diprionidae) i n Ontario. Entomophaga 20: 97-103. G r i f f i t h s , K.J. and L.A. Lyons. 1968. The establishment of Lophyroplectus luteator (Hymenoptera: Ichneumonidae) on Neodiprion s e r t i f e r (Hymenoptera: Diprionidae) i n Ontario. Can. Entomol. 100: 1095- 1099. Griswold, C.L. 1939. Introduced pine sawfly. J. Econ. Entomol. 32: 474-478. Guppy, J.C. 1969. Some ef f e c t s of temperature on the immature stages of the armyworm, Pseudaletia unipuncta (Lepidoptera: Noctuidae), under controlled conditions. Can. Entomol. 101: 1320-1327. Henson, W.R. 1958. The ef f e c t of radiatio n on the habitat temperatures of some poplar-inhabiting insects. Can. J. Zool. 36: 463-478. Henson, W.R. and R.F. Shepherd. 1952. The e f f e c t s of radiation on the habitat temperatures of the lodge pole needle miner, Recurvaria m i l l e r i Busck (Gelechiidae: Lepidoptera). Can J. Zool. 30: 144-153. Heron, R.J. 1966. The reproductive capacity of the larc h sawfly and some factors of concern i n i t s measurement. Can. Entomol. 98: 561-578. Hogg, D.B. and A.P. Gutierrez. 1980. A model of the f l i g h t phenology of the beet armyworm (Lepidoptera: Noctuidae) i n central C a l i f o r n i a . 48: 1-35. Hoy, M.A. 1976. Establishment of gypsy moth parasitoids i n North America: an evaluation of possible reasons for establishment or non-establishment, pp. 215-232. In J.F. Anderson and H.K. Kaya teds.], Perspectives i n forest entomology. Academic Press, New York. Huffaker, C.B., R.F. Luck, and P.S. Messenger. 1977. The eco l o g i c a l basis of b i o l o g i c a l control. Proc. XVth Int. Cong. Entomol. 560-586. Ierusalimov, E.N. 1973. Temperature regime and atmospheric humidity i n a stand damaged by the pine moth. Sylvics 6: 46-52. (unedited translation) Ives, W.G.H., W.J. Turnock, C.H. Buckner, R.J. Heron, and J.A. Muldrew. 1968. Larch sawfly population dynamics: techniques. Manitoba Entomol. 2: 5-36. Jahn, E. 1967. On the population outbreak of the pine f a l s e webworm, Acantholyda erythrocephala Chr. i n the Steinfeld, lower Austria, i n the years 1964-1967. Anzeiger fur 171 Schadlingskunde 39: 145-152. (unedited translation) J u i l l e t , J.A. 1959. Morphology of immature stages, l i f e - h i s t o r y , and behavior of three hymenopterous parasites of the European pine shoot moth, Rhyacionia buoliana ( S c h i f f . ) (Lepidoptera: Olethreutidae). Can. Entomol. 91: 709-719. Kaster, L.V. and W.B. Showers. 1984. Modeling black cutworm (Lepidoptera: Noctuidae) f i e l d development i n Iowa. Environ. Entomol. 13: 384-389. Kolomietz, N.G. 1967. Sawfly weaver ( d i s t r i b u t i o n , biology, damage, natural enemies, control). Nauka. Novosibirsk, USSR, (unedited translation) Lederhouse, R.C., M.D. Finke, and J.M. Scriber. 1982. The contributions of l a r v a l growth and pupal duration to protandry i n the black swallowtail b u t t e r f l y , P a p i l i o polygenes. Oecologia 53: 296-300. Linnaei, C. 1758. Systema naturae. 10th ed. Stockholm, Sweden. 826pp. Logan, J.A., R.E. Stinner, and J.S. Bacheler. 1979. A descriptive model for predicting spring emergence of He l i o t h i s zea populations i n North Carolina. Environ. Entomol. 8: 141-146. Logan, J.A., D.J. Wollkind, S.C. Hoyt, and L.K. Tanigoshi. 1976. An an a l y t i c model for description of temperature dependent rate phenomena i n arthropods. Environ. Entomol. 5: 1133-1140. Lyons, L.A. 1964. The European pine sawfly, Neodiprion s e r t i f e r (Geoff.) (Hymenoptera: Diprionidae). A review with emphasis on studies i n Ontario. Proc. Entomol. Soc. Ontario 94: 5-37. Lyons, L.A. 1970. Some population features of reproductive capacity i n Neodiprion swainei (Hymenoptera: Diprionidae). Can. Entomol. 102: 68-84. Mason, L.J. and T.P. Mack. 1984. Influence of temperature on ovip o s i t i o n and adult female longevity for the soybean looper, Pseudoplusia includens (Walker) (Lepidoptera: Noctuidae). Environ. Entomol. 13: 379-383. McLeod, J.M. 1975. Parasitoid evaluation: the monitoring problem. Melsheimer Entomol. Series. No. 321 17: 1-12. Middlekauff, W.W. 1938. Occurrence of a European sawfly A. erythrocephala (L.) i n N. Y. State. J. New York Entomol. SOC. 46: 438. Middlekauff, W.W. 1958. The North American sawflies of the 172 genera Acantholyda. Cephalcia. and Neurotoma (Hymenoptera, Pamphiliidae). Univ. C a l i f . Publ. Entomol. 14: 51-174. M i l l e r , J.C. 1983. Ecological relationships among parasites and the practice of b i o l o g i c a l control. Environ. Entomol. 12: 620-624. Morris, R.F. 1976. Influence of genetic changes and other variables on the encapsulation of parasites by Hyphantria cunea. Can. Entomol. 108: 673-684. Morris, R.F. and C.W. Bennett. 1967. Seasonal population trends and extensive census methods for Hyphantria cunea. Can. J. Zool. 99: 9-17. Morris, R.F., and W.C. Fulton. 1970. Models for the development and s u r v i v a l of Hyphantria cunea i n r e l a t i o n to temperature and humidity. Mem. Entomol. Soc. Can. no. 70, 60 pp. Morse, B.W., F.J. E l l e r , and H.M. Kulman. 1985. Forecasting emergence of adult yellowheaded spruce sawfly (Hymenoptera: Tenthredinidae). Environ. Entomol. 13: 895-897. Novak, M.D. and T.A. Black 1985. Theoretical determination of the surface energy balance and thermal regimes of bare s o i l s . Boundary Layer Meteor. 33: 313-333. Plumb, G.H. 1945. Miscellaneous insect notes. Connecticut State Entomol., 44th Rep., B u l l . Conn. Agric. Expt. Sta. 488: 420-421. Prebble, M.L. 1941. The diapause and related phenomena i n G i l p i n i a polytoma (Hartig). IV. Influence of food and diapause upon reproductive capacity. Can. J. Res. D. 19: 417-436. Price , P.W. 1971. A comparison of four methods for sampling adult populations of cocoon parasitoids (Hymenoptera: Ichneumonidae). Can J. Zool. 49: 513-521. Price, P.W. and H.A. Tripp. 1972. A c t i v i t y patterns of parasitoids on the Swaine jack pine sawfly, Neodiprion swainei (Hymenoptera: Diprionidae), and p a r a s i t o i d impact on the host. Can. Entomol. 104: 1003-1016. Prozorov, S.S. 1925. The pine redheaded sawfly Acantholyda erythrocephala L. Trans. Siberian Agric. Acad. Suppl. Separate, 21 pp. (abstract). Quednau, F.W. and K.P. Lim. 1983. Olesicampe geniculatae. a new pa l a e a r c t i c ichneumonid parasite of Pristiphora geniculata (Hymenoptera: Tenthridinidae). Can. Entomol. 115: 109-113. Rawlins, J.E. and R.C. Lederhouse. 1981. Development influences 173 of thermal behavior on monarch c a t e r p i l l a r s (Danaus plexippus): an adaptation for migration (Lepidoptera: Nymphalidae: Danainae). J. Kans. Entomol. Soc. 54: 387-408. Reede de, R.H. and H. de Wilde. 1986. Phenological models of development i n Pandemis heparana and Adoxophyes orana for timing of a p p l i c a t i o n of insect growth regulators with juvenile hormone a c t i v i t y . Entomol. Exp. Appl. 40: 151-159. Regniere, J. 1982. A process-oriented model of spruce budworm phenology (Lepidoptera: T o r t r i c i d a e ) . Can. Entomol. 114: 811-825. Regniere, J. 1984. A method of describing and using v a r i a b i l i t y i n development rates for the simulation of insect phenology. Can Entomol. 116: 1367-1376. Regniere, J. 1987. Temperature-dependent development of eggs and larvae of Choristoneura fumiferana (Clem.) (Lepidoptera: T o r t r i c i d a e ) and simulation of i t s seasonal history. Can. Entomol. 119: 717-728. Regniere, J., R.L. Rabb, and R.E. Stinner. 1981. P o p i l l i a japonica: simulation of temperature-dependent development of the immatures, and p r e d i c t i o n of adult emergence. Environ. Entomol. 10: 290-296. Rosenberg, N.J. 1974. Microclimate: the b i o l o g i c a l environment. John Wiley c? Sons, Toronto. Rumphorst, H. and H. Goossen. 1960. Contribution to the biology and control of the pine f a l s e webworm (Acantholyda erythrocephala). Anzeiger fur Schadlingskunde 33: 149-154. (unedited t r a n s l a t i o n ) . Sanborne, M. 1984. A r e v i s i o n of the world species of Sinophorus (Ichneumonidae). Mem. Amer. Entomol. Inst. 38, 403p. SAS I n s t i t u t e 1985. SAS user's guide: s t a t i s t i c s . SAS i n s t i t u t e , Cary, N.C. Schmutzenhofer, H. 1975. Sawflies injurious to forests i n Austria. 1. Acantholyda erythrocephala L. Cbl. Ges. Forstwesen 92: 1-8. (unedited translation) Schowalter, T.D., W.W. Hargrove, and D.A. Crossley J r . 1986. Herbivory i n forested ecosystems. Ann. Rev. Entomol. 31: 177-186. Schwerdtfeger, F. 1941. On the biology of the forest pest Acantholyda erythrocephala L. Photographie und Forschung 3: 161-169. (unedited t r a n s l a t i o n ) Schwerdtfeger, F. 1944. Further observations on the biology of the pine f a l s e webworm, Aoantholyda erythrooephala L. Z. Angew. Entomol. 30: 364-371. (unedited translation) Shepherd, R.F. 1958. Factors c o n t r o l l i n g the i n t e r n a l temperatures of spruce budworm larvae, Choristoneura fumiferana (Clem.). Can. J. Zool. 36: 779-786. Sherman, P.W. and W.B. Watt. 1973. The thermal ecology of some Colias b u t t e r f l y larvae. J. Comp. Physiol. 83: 25-40. Soraci, F.A. 1938. D i s t r i b u t i o n of the sawfly Acantholyda erythrooephala i n New Jersey. J. New York Entomol. Soc. 46: 444. Stedinger, J.R., C.A. Shoemaker, and R.F. Tenga. 1985. A stochastic model of insect phenology for a population with s p a t i a l l y variable development rates. Biometrics 41: 691-701. Strong, D.R., J.H. Lawton, and R. Southwood. 1984. Insects on plants. Harvard Univ. Press, Cambridge, Mass. Syme, P.D. 1981. Occurrence of the introduced sawfly Acantholyda erythrooephala (L.) i n Ontario. Can. For. Serv. Res. Notes 1: 4-5. Tatar, M. 1984. Protandry i n physiological time. Oikos 43: 413-414. Tan, C.S., T.A. Black, and J.U. Nnyamah. 1978. A simple d i f f u s i o n model of tr a n s p i r a t i o n applied to a thinned douglas-fir stand. Ecology 59: 1221-1229. Tauber, M.J. and C.A. Tauber 1981. Insect seasonal cycles: genetics and evolution. Ann. Rev. Ecol. Syst. 12: 281-308. Thompson, L.C. and H.M. Kulman. 1976. Parasite complex of the lar c h sawfly i n Minnesota. Environ. Entomol. 5: 1121-1127. Timberlake, P.H. 1912. A study of the biology of Limnerium validum (Cresson). U.S.D.A. Bur. Entomol., Tech. Ser. 19: 71-92. T o t h i l l , J.D. 1922. The natural control of the f a l l webworm (Hyphantria cunea Drury) i n Canada together with an account of i t s several parasites. Can. Dep. Agric. B u l l . No. 3. 107pp. Tripp, H.A. 1965. The development of Neodiprion swainei Middleton (Hymenoptera: Diprionidae) i n the province of Quebec. Can. Entomol. 97: 92-107. Turnock, W.J., R.P. Bodnaryk, and D. Abramson. 1986. Ef f e c t s of temperature on the rate of pupal-adult development of the 175 noctuid moth Mamestra configurata Wlk.: evidence for d i f f e r e n t i a l e f f e c t s on the i n i t i a t i o n of development and subsequent metamorphic development. Oecologia (Berlin) 68: 422-427. Turnock, W.J. and J.A. Muldrew. 1971. Pristiphora e r i c h s o n i i (Hartig), l a r c h sawfly (Hymenoptera: Tenthredinidae). pp. 175-194. In B i o l o g i c a l control programmes against insects and weeds i n Canada 1959-1968. Tech. Comm. 4. Comm. Inst. B i o l . Control. Uvarov, B.P. 1931. Insects and climate. Trans. Entomol. Soc. London 79: 1-247. Wagner, T.L., H. Wu, R.M. Feldman, P.J.H. Sharpe, and R.N. Coulson. 1985. Multiple-cohort approach for simulating development of insect populations under variable temperatures. Ann. Entomol. Soc. Am. 78: 691-704. Wallace D.R. and C.R. Sullivan. 1963. Laboratory and f i e l d investigations of the ef f e c t of temperature on the development of Neodiprion s e r t i f e r (Geoff.) i n the cocoon. Can. Entomol. 95: 1051-1066. Wellington, W.G. 1950. E f f e c t s of rad i a t i o n on the temperatures of insectan habitats. S ci. Agric. 30: 209-234. Wellington, W.G. 1954. Weather and climate i n forest entomology. Recent studies i n bioclimatology. Meteorol. Mono. 2: 11-18. Wellington, W.G. 1957. The synoptic approach to studies of insects and weather. Ann. Rev. Entomol. 2: 143-162. Wellington, W.G. and R.M. Trimble. 1984. Weather, pp. 399-425. In C.B. Huffaker and R.L. Rabb. Ecological Entomology. John Wiley & Sons. New York. Wells, A.B. 1926. Notes on tree and shrub insects i n southwestern Pennsylvania. Entomol. News 37: 254-258. Weseloh, R.H. 1976. Behavior of forest insect parasitoids. pp. 99-110 In J.F. Anderson and H.K. Kaya [eds.]. Perspectives i n forest entomology. Academic Press, New York. Wickman, B.E. and T.R. Torgersen. Phenology of Douglas-fir tussock moth, Orgyia pseudotsugata. egg eclosion and mortality i n a thinned and unthinned stand (lepidoptera: Lymantridae). Pan-Pacific Entomol. 63: 218-223. Wilkinson, R.C., G.C. Becker, and D.M. Benjamin. 1966. The biology of Neodiprion rugifrons (Hymenoptera: Diprionidae), a sawfly i n f e s t i n g jack-pine i n Wisconsin. Ann. Entomol. Soc. Amer. 59: 786-792. 176 Willmer, P.G. 1980. The effects of a flu c t u a t i n g environment on the water r e l a t i o n s of l a r v a l Lepidoptera. Ecol. Entomol. 5: 271- 292. Willmer, P.G. 1982. Microclimate and the environmental physiology of insects. Adv. Insect Physiol. 16: 1-57. Wilson, L.F. 1977. A guide to insect injury of conifers i n the Lake States. U.S.D.A. Agric. Handbook No. 501.' "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0098147"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Plant Science"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Phenology and biometeorology of pine false webworm (Hymenoptera: Pamphiliidae) and its parasitoids in southern Ontario"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/29016"@en .