', 9 f} LIEfiAflY ’ Michigan Mate University This is to certify that the thesis entitled THE BIOLOGY OF PATASSON SP. NEAR SORDIDATUS (HYMENOPTERA: MYMARIDAE) AND ITS IMPACT ON THE CARROT WEEVIL presented by ROBERT D. COLLINS has been accepted towards fulfillment of the requirements for M- S . degree in ENII‘QMQLOGY W¢d~¢fi Major prayssor DateD e 18 191 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU RETURNING MATERIALS: Piace in book drop to LJBRARJES remove this checkout fror up your record. FINES win be charged if book is returned after the date stamped below. THE BIOLOGY OF PATASSON SP. NEAR SORDlDATUS (HYMENOPTERA: M'YMARIDAE) AND IT lM THE CARROT WEEVlL by Robert D. Collins A THESIS Submitted to Michigan State University . in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE . I982 ABSTRACT THE BIOLOGY OF PATASSON SP. NEAR SORDIDATUS (HYMENOPTERA: MYMARIDAE) AND ITS IMRACI ON THE CARROT WEEVIL by Robert D. Collins Laboratory and field studies were conducted to examine the biology of Patasson sp. near sordidatus, a mymarid egg parasitoid of the carrot weevil, Listronotus orggonensis (LeConte), and to evaluate the impact of the parasitoid on host populations. Patasson sp. near sordidatus adults were short-lived, and both longevity and developmental times were influenced by temperature and superparasitism. Emergence periodicity was affected by both temperature and photoperiod. Courtship and mating behaviors were described. Carrot weevil and parasitoid populations were spatially but not entirely temporally synchronous. Nominal parasitism rates (the overall percent of eggs parasitized) were found to overstate the true effect of the parasitoid on the weevil population due to the observed density-dependent weevil mortality in the absence of parasitism. The parasitoid was found to have a significant though minor role in carrot weevil mortality rates in the crop site . ACKNOWLEDGEMENTS For their invaluable assistance in the preparation of this thesis, I would like to thmk the members of my graduate committee, Dr. George W. Bird, Dr. Stuart H. Gage, Dr. Dem L. Haynes, md particularly Dr. Ed Grafius, my major professor. I am also indebted to Dr. James Bath md the Department of Entomology for the research opportunities and facilities provided. 11' 2. TABLE OF CONTENTS Introduction . . ...................... 1 Adult Longevity ...................... 4 2.l Methods ....................... 5 2. LI Effect of Temperature ............. 5 2.1.2 Adult Feeding . . . . . . ........... 6 2.l.3. Adult Size and Superparasitism .......... 6 2.2Results.......... .......... ....8 2.2.l Superparasitism ......... . ...... 8 2.2.2 Sexw. . .......... 11 2.2.3 Survivorship .................. 11 2.2.4 Adult Feeding ................. 11 2-2-5 Temperature .................. 13 2.3 Discussion ..... . ................. 17 Fecundity and Oviposition .................. 19 3.l Methods ....................... 19 3.l.l Fecundity ................... 19 3.l.2 Oviposition . . ................ 22 3.2 Results . ....................... 23 3.2.l Fecundity ................... 23 3.2.2 Oviposition .................. 29 3.3 Discussion ....................... 29 iii TABLE OF CONTENTS, continued 4. Development . ....................... 33 4.! Methods .......... . ............. 34 4.2 Results . . . ..................... 35 4.2. l Temperature .................. 36 4.2.2 Sex . . . . . ................. 38 4.2.3 Superparasitism. ................ 38 4.3 Discussion . ...................... 38 Emergence Periodicity. . . . . . . ............. 41 5.lMethods........... ............ 41 S.l.l Role of Temperature md Photoperiod ....... 41 5.l.2 Diel Emergence Pattern . ............ 42 5.l.3 Role of Exogenous Cues ............. 43 5.2 Results . . . ..................... 44 5.2.I Role of Temperature md Photoperiod ....... 44 5.2.2 Diel Emergence Pattern ............. 46 5.2.3 Role of Exogenous Cues ............. 48 5.3 Discussion . ...................... 51 Courtship md Mating .................... 53 6.l Methods ..... . . ................. 53 6.l.l Courtship md Mating Behavior .......... 53 6. l .2 Mating Frequency ................ 54 6.2Results........h ................ 54 6.2.l Courtship and Mating Behavior .......... 54 6.2.2 Mating Frequency ................ 57 6.3 Discussion ....................... 59 iv TABLE OF CONTENTS, continued Host-Parasitoid Synchrony . . ................ 60 7.IMetnods........ .............. ..61 7.I.I Temporal Synchrony. .............. 61 7.l.2 Spatial Synchrony ................ 62 7.2 Results . . . . .................... 63 7.2.l Temporal Synchrony ............... 63 7.2.2 Spatial Synchrony ................ 64 7.3 Discussion . ...... . ............... 65 Host md Parasitoid Resource Utilization .......... . . 67 8.lMethods.............. .......... 69 8.l.l Field Observations . . . ............ 69 8.l.2 Survival of Carrot Weevil Larvae ......... 72 8.2 Results ........... . . . .......... 73 8.2.l Carrot Weevil Resource Utilization ........ 73 8.2.2 Parasitoid Resource Utilization ........ . . 82 8.3 Discussion..... ..... ....... 9O 8.3.l Carrot Weevil Resource Utilization ........ 90 8.3.2 Parasitoid Resource Utilization .......... 93 Differential Pesticide Effects. ................ 98 9.l Methods. . . . ....... . ............ 99 9.l.| Study Area ................... 99 9. l.2 Pesticide Applications .............. 99 9.l.3 Adult Carrot Weevil Trapping ........... 101 9.l.4 Carrot Plant Sampling .............. 101 TABLE OF CONTENTS, CONTINUED 9.2 Results ....... . . . .............. 104 9.2.l Carrot Weevil Infestation .......... . . 104 9.2.2 Parasitism . .................. 107 9.3 Discussion ....................... 108 ID. Summary and Conclusions ....... . ....... . . . 110 Literature Cited . . ............ . . . . . . . . 118 vi Figure I. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Finge 8. Figure 9. Figure l0. Figure ll. LIST OF FIGURES Survivorship of adult Patasson n. sp. at 26° C. . Effect of temperature on weighted mean langevityofadult Patassonn.sp. . . . . . . . Effect of mating status and number of parasitoids per host egg on the temporal gigstributian of oviposition by Patasson n. sp. at C.OOOOOOOOOOOO_OOOOOO Effect of temperature on the mean duration of the life cycle of Patasson n. sp. . . . . . . . Temporal distribution of emergence from host eggs by Patasson n.sp. in: a) constant light, constmt temperature; b) cyclic photoperiod, constant temperature; and c) constant light, cyclictemperature. . . . . . . . . . . . . Effect of temperature on the temporal distribution of emergence from host eggs by PatassmnOSPO o o o o o o o o o o o a a 0 Cumulative emergence from host eggs by adult male and female Patasson n.s “RC during the first twohaursofphotap fieat 23 C. . . . . . . Effect of phatoperiod an the temporal distribution of emergence from carrot weevil eggs by adult E_____atasson n. sp. at 23° C. . . . . Position of Patasson n. sp. during copulation. . Mean percent of carrot weevils surviving to late instar larvae or pupae after transfer of eggs to uninfested carrot plants. . . . . . . Temporal distribution of parasitism by Patasson n.97s9p. and inviability of carrot weevil eggs in vii O O O O O O 12 . . 16 ...... 25 ...... 40 . . . . . . 45 . . 47 . ..... 49 . ..... 50 ...... 58 . . . 81 ...... 83 LIST OF FIGURES, continued Figure l2. Figure l3. Figure I4. Figure l5. Field density of carrot weevil eggs md Patasson n. sp. immatures within hast eggs in Clinton Co., Michigan, in I979. . . . . . . ....... 85 Effect of per plant host egg density on: a) mean number of Patasson n. sp. developing per carrot plant, and b5 mean number of carrot weevil eggs parasitized per plant.. . . . . . ...... 89 Relationship between nominal and effective rates of parasitism of carrot weevil eggs on carrots by Patasson n. sp. for a given distributionofhasteggsperplant. . . . . . . . . . . . 95 Mem percent of carrot weevil eggs parasitized by Patasson n. sp. in: 0) plots treated at plmting with a systemic insecticide (aldicarb); b) plots treated with fol iar applications of four insecticides (azinphmmethyl, fenvalerate, diazinan, md axamyl); and untreated controls, ClintanCquichigcn, I980. . . . . . . . . . . . . . . 109 viii Table I. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. Table l0. Table II. Table l2. LIST OF TABLES Composition of diet fed to adult Patasson n. sp. Mean head capsule width of adult Patasson n. sp. as affected by number of parasitoids per hast egg md sex. . . Mean longevity of adult Patasson n. sp. at 26° C as affected by sex and nu r of parasitoids per h“? ”g. 0 O 0 Effect of diet ano mem longevity of adult E___atassm n. sp. at 26°C. . . Mem longevity of adult Patasson n. sp. as affected by number of parasitoid? per host egg md temperature. Number of offspring produced by female Patasson n. sp.as asinfluenced by age and mating sustat offemale. . . . . ......... Developmental time of Patasson n. sp. as influenced by temperature, sex, md number per w ”9. O I O O O O O O O Mem per plant density of carrot weevil larvae by larval instar md sampling date, I979 ..... Number of carrot weevil eggs per clutch and per infested carrot plant and the number of clutches per infested plant as affected by sampling date. Spatial distribution of carrot plmts and carrot weevil eggs among 30 x 30 cm grids, I979. . . . Spatial distribution of carrot weevil eggs parasitized by Patasson n. sp. among 30 x 30 cm grids, I979. O O ........ O O O O 0 Nominal and effective rates of parasitism by Patasson n. sp. in I979 ............ ix 74 79 96 LIST OF TABLES, continued Table I3. Table I4. Application dates and rates applied for five chemical insecticides in field #3 at the Hammond Farm, Clinton Co., Michigan, in I980. ...... 100 Percent of carrot plants with carrot weevil eggs or larvae as affected by insecticide treatment andsamplingdate. 105 I. Introduction The carrot weevil, Listronotus orggonensis (LeConte) was first reported as a pest of economically importmt crops in I902 (Chittenden I909). Its presence in Michigcn was reported as early as I9l5 (Henderson I940), and it has since been observed to be widely distributed at low population densities in many vegetable growing areas within the state (Otto I978). The primary crops attacked are parsley, carrots and celery, the latter two being of particular importance in Michigm. In Michigan, the carrot weevil is a periodic pest. Reports of damage have been sporadic, md economically significant infestation levels have generally been localized. However, severe damage in specific fields has occurred. Prevention of economic damage by carrot weevils is primaily accomplished by crop rotation. Dispersal of this insect occurs as a result of walking. Flight quears to play no significant role in inter-field movement. This facilitates control by crop rotation and usually results in a confinement of infestations to field margins which are adjacent to fields infested during the previous growing season. Control of carrot weevil populations by the use of chemical insecticides is limited. There are no compounds currently registered in Michigan specifically for the control of the carrot weevil in crops other than celery. In addition, the incidental effectiveness of insecticides used for the control of other pest species is questionable. The egg and larval stages of the carrot weevil are spent in a relatively protected environment inside the host plant, and pupation occurs below the soil surface. Therefore, only adult weevils experience direct contact with insecticides applied as foliar sprays, and the effectiveness of these chemicals appears to be limited. Significmt effects of natural enemies on carrot weevil populations have not been reported. Several associations between carrot weevils and parasitoids or parasites have been observed (Chittenden I924, Whitcomb I965, Ryser I975), but in each instmce only a negligible impact on the weevil population was noted. However, a mymarid egg parasitoid was discovered attacking carrot weevil eggs in significant numbers during the I979 growing season in Clinton County, Michigm. This parasitoid has been identified as Patasson so. near sordidatus (E. Grissell, USDA Systematic Entomology Laboratory, personal communication), and appears to represent a new species. It will be referred to hereafter as Patasson n. sp. A survey conducted in I980 revealed the presence of this parasitoid at several other locations in Michigan, including vegetable growing areas near Hudsonville md Grant (Ottawa and Newaygo Counties, respectively). Simonet (Ohio Agricultural Research md Development Center, personal communication) has since observed the presence of this parasitoid at Willard, Ohio (Huron Co.). This parasitoid may represent a significant factor affecting carrot weevil populations in Michigm and could contribute to the periodic and localized nature of weevil infestations. In I979, a series of studies was initiated to examine the biology of Patasson n. sp., and to evaluate the impact of this parasitoid on carrot weevil populations. The objectives of laboratory studies of the biology and life cycle of Patasson n. sp. were to: I) examine the effects of temperature, sex, and the number of parasitoids developing per host egg on adult longevity; 2) determine the relationship between adult size and the number developing per host egg; 3) examine parasitoid fecundity as influenced by the number developing per host egg and mating status; 4) determine the temporal pattern of oviposition; 5) estimate developmental rates and the effect of number per host egg, temperature, and sex; 6) estimate the sex ratio by examining adults reared from field-collected host eggs; 7) investigate the effect of temperature md photoperiod on the periodicity of emergence by adults from host eggs and the role of endogenous md exogenous mechmisms; and 8) observe courtship md mating behavior. These studies were designed to provide basic information concerning the biology of the parasitoid in order to more completely evaluate its role as m agent of biological control in field situations. Additional laboratory and field studies were undertaken to investigate relationships between the carrot weevil and Patasson n. sp. The objectives of these studies were to determine the degree of spatial and temporal synchrony between host md parasitoid populations, to examine resource exploitation strategies by both the carrot weevil md by Patasson n. sp., md to evaluate the sigtificmce of interactions between the two sets of strategies. Carrot weevil oviposition strategies will determine the may of resources (host eggs) available to the parasitoid, and the way in which the parasitoid exploits the available host eggs will influence the efficiency with which the weevil utilizes its resources. A final study was conducted to evaluate the relative impact of chemical insecticide applications on host md parasitoid populations. Relevant literature is briefly summarized in the appropriate sections of the following report. A more complete review of the literature concerning the carrot weevil, the genus Patasson, and the closely related mymarid genus m is given by Collins md Grafius (I982). 2. Adult Longevity The adults of species closely related to Patasson n. sp. are short lived. Estimates ,of mean adult longevity under a variety of conditions have ranged from less than two days for Angphes {REESE (Anderson and Paschke I970a) to I2 days for an unidentified Patasson species (Fisher et al. I96 I). Longevity has been found to be strongly influenced by temperature (Anderson and Paschke I970a, Stoner and Surber I97 I). Differences based on the sex of the individual have also been reported (Aeschlimann I977, Ahmad I979). The influence of superparasitism on developmental times and on the size of omits has been discussed (Anderson and Paschke I969). However, the effect of superparasitism on adult longevity as a measure of overall vigor has not been examined. Tumbull and Chant (I96I) suggested that the availability of food for adult parasitoids may play (:1 important role in their success as biological control agents. Several investigators have used various dilutions of honey in distilled water as a source of food for Id>oratory colonies of mymarid parasitoids (Balduf I928, Kevm I946, Ahmad I979, Vidmo et al. I979). This was presumably done to increase the longevity and general vigor of the adults. However, no quantitative information is available concerning the effect of adult feeding on aspects of the life cycle. The objective of this study was to examine the effect of various factors on the adult longevity of Patasson n. sp. The effects of temperature, sex, superparasitism and adult feeding were investigated. The relationship between superparasitism and adult size was also studied. This information will facilitate an understanding of parasitoid-host interactions in the field. 2.I Methods Carrot plants were collected from the border rows of an untreated area of field #3 at the Hammond farm near East Lansing, Michigan, during July, I980. Carrot weevil eggs were extracted from the plants and placed individually in inverted 65 ml clear plastic containers on a 5.5 cm disc of Whatman #3 qualitative filter paper. Eggs were kept in an environmental chamber at 26° C with a light-dark cycle of l6:8 (photophase beginning at 0600 h). Distilled water was applied to the filter paper when necessary to maintain a high relative humidity. The eggs were monitored twice a day, at 0800 h and at 2000 h (s l h) did the emergence of adult parasitoids recorded. 2.l.I Effect of Temperature When the first parasitoid had emerged from the egg within a container, the container was rmdomly assigned to an environmental chamber set at I7, 20, 23, 26, or 29° C with a photoperiod of I6:8. The adults emerging from at least 50 parasitized carrot weevil eggs were monitored at each temperature. A larger number of host eggs were monitored at 26° C. Limitations on the number of available environmental chambers preclUded the examination of all temperatures simultaneously. Adult longevity was observed by monitoring the status of parasitoids within each container every I21l h at 0700 h and I900 h. The longevity of each parasitoid was estimated by assuming that both emergence and death occurred at the midpoints of the observed l2: I h intervals. The parasitoids emerging from a single host egg were kept in the original rearing container throughout the period of observation to avoid potential injury resulting from their transfer to individual containers. In some instances, individual parasitoids emerged from the same egg during different monitoring periods. When two or more adults of the some sex emerged from an egg during different periods md also died during different periods, the longevity of the specific individuals was ambiguous. This will not affect treatment means but will influence associated variances. The set of assumptions which yielded the greatest contribution to the treatment variance was selected in all such situations. The actual variance associated with any mean would thus be equal to or less than the reported value. These situations occurred infrequently and affected the variance estimates only slightly. 2.I.2 Adult Feeding A honey and beer based diet (Table I) was made available to the adult parasitoids at all temperatures. The diet was introduced onto the filter paper using a hypodermic syringe inserted through the soft plastic base of the container. Parasitoids were fed when the initial emergence was observed within each container and every morning thereafter as long as at least one adult remained alive. Approximately 0.05 ml of diet per container was used at each I feeding. To determine the effect of the artificial diet on adult longevity, a control group was given a diet of distilled water in the manner described above. The control group was maintained at 26° C with a photoperiod of l6:8. 2.l.3 Adult Size and Superparasitism Between one and six parasitoids were observed to emerge from a single Table 1. Composition of diet fed to adult Patasson n. sp. Honey 200 ml Old Milwaukee beer 306 ml Ascorbic acid 1o 9 L-cyatene hydrachloride 2 g Distilled water 560 ml carrot weevil egg. The relationship between adult size and the number of parasitoids developing per host egg was examined. Several measures of body size were evaluated in a preliminary study, including thorax length, overall body length, antennal length, and head capsule width. Only head capsule width, which is unaffected by conditions encountered subsequent to emergence, was found to be consistently related to the number of parasitoids per egg. Head capsule widths were measured for parasitoids kept at 260 C subsequent to their death. Measurements were made with an ocular micrometer at 50 x. Each micrometer division was ca. 0.0092 mm. 2.2 Results 2.2. I Superparasitism Adult size (as measured by head capsule width) was inversely related to the number of parasitoids developing per host egg (Table 2). Females were slightly, but not significantly, larger than males within each parasitoid per host egg category (t-tests, p>.05). The adult longevity of Patasson n. sp. was also found to be related to the degree of superparasitism. The longevity of all parasitoids (males and females) kept at 26° C decreased as the number of parasitoids developing in the host egg increased (Table 3). The most pronounced differences were observed between adults reared from eggs in which one, two, or three parasitoids emerged. This relationship is similar to that found between size and the degree of superparasitism. Tlnerefore, longevity may be related to size. The correlation coefficient between mean adult longevity and mean head capsule width was 0.980 and is significant (t-test, p<.0|). Table 2. Mean head capsule width of adult Patasson n. sp. as affected by 'fiumber of parasitoids per host egg and sex. Mean head capsule width (mm) No. of parasitoids a per egg Males Females Total One Mean 0.244 0.256 0.252 SD 0.018 0.016 0.018 N 34 52 86 Two Mean 0.212 0.220 0.217 SD 0.017 0.017 0.017 N 63 103 166 Three Mean 0.195 0.200 0.198 SD 0.018 0.019 0.019 N 51 84 134 Four Mean 0.180 0.185 0.182 SD 0.025 0.020 0.023 N 18 15 33 Five Mean 0.169 0.170 0.169 SD 0.012 0.020 0.012 N 8 2 l0 a. Means followed by the same letter are not significantly different at p-0.05, Neuman-Kuel multiple range test. Student- 10 Table 3. Mean longevity of adult Patasson n. sp. at 26 C as affected by sex and number of parasitoids per host egg. Mean adult longevity (h) No. of parasitoids a per egg Males Females Total One Mean 100.42 83.17 90.00 SD 40.28 29.55 35.04 N 38 58 96 Two Mean 66.25 69.69 68.35 SD 27.06 25.11 25.80 N 73 121 194 Three Mean 59.24 60.68 60.14 SD 21.21 21.30 21.24 N 63 105 168 Four Mean 49.60 58.15 53.57 SD 23.94 22.44 23.16 N 15 13 28 Five Mean 52.80 --- 52.80 SD 23.40 --- 23.40 N 5 --- 5 a. Means followed by the same letter are not significantly different at p-0.05, Student- Neuman-Kuel multiple range test. 11 2.2.2 Sex The sex of the adult appears to have only a small effect on parasitoid longevity at 26° C (Table 3). A significant difference was found between the mean longevity of males and females only for adults emerging from host eggs in which they had developed singly (t-test, p<.05). For this group, the males lived an average of 20.7% (l7.25 h) longer than the females. The females lived slightly (though not significantly) longer in all other categories. These data suggest that sex based differences in longevity play a relatively minor role in the dynamics of parasitoid populations in the field. 2.2.3 Survivorship Survivorship of adult parasitoids kept at 26° C was relatively high during the first two days following emergence (Figure I). After two days, 83.3% of all adults were still alive. Mortality increased over the next two days and only l7.796 were still alive after four days. A few adults remained alive up to 7.5 days after emergence. Adult longevity has been shown to be related to the number of parasitoids developing per egg, and to a lesser degree to the sex of the individual. Thus, the survivorship curve of a particular field population would be determined in part by the distribution of individuals by parasitoid per egg category and by sex. The parasitoids examined in this study were reared from field collected eggs, and are thus assumed to be representative of the composition of the field population. 2.2.4 Adult Feeding The average lifespan of adults reared from eggs in which one, two, three, 12 .0 am no .mn r: aamummmu Dance mo mannuo>e>uam .H ouoaem ONIAIAHRS lNBOUSd 13 or four parasitoids developed increased 57.3, 74.4, I l4.8, and 80.0% respectively when the honey and beer diet was made available (Table 4). Mean longevity was significantly greater in each parasitoid per egg category for parasitoids fed the artificial diet (SNK tests, p<.05). Insufficient data were available for comparisons involving adults from host eggs in which five parasitoids developed. The relationship of feeding to other factors such as fecundity and the source of nutrition in the field is not known. However, feeding by the adults appaently plays an important role in the longevity of adults, and may consequently affect other aspects of the biology of this parasitoid. 2.2.5 Temperature Significant differences have been demonstrated between parasitoid per host egg categories. Therefore, longevity comparisons between temperatures must be done independently for each parasitoid per egg category. These comparisons generally indicate an inverse relationship between temperature and longevity (Table 5). Comparisons between the means for all adults at each temperature are not initially feasible because each temperature group consists of a different distribution among parasitoid per egg categories. However, overall comparisons can be made by weighting the mean longevity for each category by the fraction of parasitoids at all temperatures in the category. The means connputed in this manner demonstrate the decreased longevity which occurred at increased ten'peratures for the parasitoids examined (Figure 2). Within the range of observed temperatures, longevity more than doubled as a result of a I0o C reduction in temperature. 14 Table 4. Effect of diet on mean longevity of adult gatagggn n. sp. at 26 C. Mean longevity (h) (males + females)° No. of Adults Adults fed parasitoids distilled artificial per egg water honey diet One Mean 57.20 90.00 SD 22.21 35.04 N 30 96 Two Mean 39.22 68.35 SD 14.74 25.80 N 56 194 Three Mean 28.00 60.14 SD 12.88 21.24 N 12 168 Four Mean 28.50 53.57 SD 10.99 23.16 N 8 28 In each row, means followed by the same letter are not significantly different at p-0.05, Student-Neuman-Kuel multiple range test. 15 Table 5. Mean longevity of adult Egtggagn n. sp. as affected by number of parasitoids per host egg and temperature. Mean adult longevity (hla Number of parasitoids per host egg Temperature 1 2 3 4 5 6 17 C Mean 230.00 165.60 128.67 109.28 a 122.40 --- SD 87.96 38.88 52.56 61.68 10.08 N 12 30 45 28 5 --- 20 C Mean 139.46 119.18 103.67 --- 103.20 SD 39.36 29.40 47.16 --- 60.96 --- N 37 44 36 --- 5 --- 23 C Mean 156.00 94.87 74.00 72.37 b 44.00 54.00 SD 36.00 41.04 31.56 39.36 17.40 34.92 N 3 32 54 32 15 12 26 C Meam 90.00 68.35 60.14 53.57 be 52.80 --- SD 35.04 25.80 21.24 23.16 23.40 --- N 96 194 168 28 5 --- 29 C Mean 48.00 60.00 43.27 34.72 c --- --- SD --- 17.76 22.08 22.68 --- --- N 1 40 66 28 --- --- a. In each column, means followed by the same letter are not significantly different at p-0.05, Student-Neuman-Kuel multiple range test. 16 180- A 8 V >. h- 120- >' u: (D 2: O _| . Z 80- I( m _ 2 C3 u: E q, 40- III 3 0 I I I I I 17 20- 23 26 29 TEMPERATURE Pa) Figure 2. Effect of temperature on weighted mean longevity of adult Egtassgn_n. sp. (log Y - 10.99 - 2.08 log.x, a - 0.981). t 17 2.3 Discussion The results of this study show that Patasson n. sp. is relatively short lived. At 26° C, nearly two-thirds of the adults perished before reaching 3.5 days of age. Adult feeding was found to influence adult longevity. The availability of an artificial diet increased mean longevity by more than 50% in all parasitoid per egg categories. In contrast, differences based on sex were relatively slight. The degree of superparasitism appeared to have a major impact on adult longevity. Adults emerging from host eggs in which fewer parasitoids developed were larger and lived longer. This is apparently due to competition among the larvae for the fixed quantity of available resources. The greater the number of lava within a host egg, the more intense is this competition. The nature of the connpetition between parasitoid larvae within the host egg is not known. Interference competition may occur among some species. This is suggested for ;A_I;I_agrus 9.191993 where only one adult successfully develops regardless of the initial number of parasitoid eggs within the host egg (MacGill I934). However, mortality in most species does not increase appreciably until a high level of superparasitism is reached (Vidano et al. I979, Statterthwait I93l, Anderson and Paschke I969). A typical result of lower levels of superparasitism is the production of smaller, less vigorous offspring (Jackson I96I, Chamberlin I924, Anderson and Paschke I969, Statterthwait I93I). The relationship between mortality prior to emergence as adults and superparasitism for Patasson n. sp. is not known. However, adults emerging from host eggs in which a larger number of parasitoids have developed were smaller and had a shorter lifespan. This suggests that 18 some form of exploitation competition may have occurred. In some species this competition may be reduced by the oviposition behavior of the female, which lays more eggs in larger host eggs than in smaller ones (Statterthwait l93l). However, this strategy may not be feasible when host eggs are scarce (Jackson I96l). . ' Temperature strongly influenced adult longevity. The relationships which were demonstrated may be useful in evaluating the effects of temperature variation on adult longevity in the field. However, estimates based on observations made under constant conditions may tend to overestimate the magnitude of such effects. In the field, the parasitoid may adjust the microenvironmental conditions in which it finds itself by changing locations, or by other adaptive behaviors. This will serve to dampen the amplitude of variations in mocroenvironmental conditions and thus mitigate their effects on the parasitoid population. l9 3. Fecundity and Oviposition F ecundity is generally low among Patasson and Angphes species. Estimates of the mean number of progeny per female for various species have ranged from l7 to 36 (Williams et al. l95l, Anderson and Paschke I970a, Stoner and Surber l97l, Aeschlimann I977, Ahmad I979). Oviposition usually begins soon after the emergence of the adult female (MacCill I934, Williams et al. l95l, Jackson I96I, Anderson and Paschke I969, Maltby et al. l97l). Most parasitoid oviposition occurs within the first few days in the life of the adult female (Kevan I946, Williams et al. I9SI). Patasson and Anophes are arrhenotokous; virgin females produce only male offspring by haploid parthenogenesis. Mated females produce progeny of both sexes. Anderson and Paschke (I968) stated that the female-male sex ratio among Patasson and Anghes species is usually near 3:l. Tine sex ratio of field populations is dependent on both the percent of female offspring produced by mated females and the percent of virgin females. Several aspects of Patasson n. sp. oviposition and fecundity were examined in this study. The effects of mating status and superparasitism were estimated. The sex ratio among the progeny of mated females was recorded, and the initiation of oviposition by newly emerged females was investigated. 3.l Methods 3.l.l Fecundity Fecundity was examined by exposing carrot weevil eggs to adult female parasitoids and then rearing the eggs to determine the number of eggs parasitized and the nunnber of progeny produced. Host eggs less than 24 h old 20 were obtained from a laboratory culture of carrot weevils known to be free of parasitism. The female parasitoids used in earlier trials were reared from host eggs collected from an untreated area of a carrot field at the Hammond Farm. Those used in later trials were obtained from weevil eggs which were parasitized in the laboratory. All of the females used in this study had emerged within six hours of trial initiation. Each female parasitoid was kept in an inverted 65 ml clear plastic container with a moistened disc of filter paper. An artificial diet was introduced daily following the procedure described in Section 2.I.2. The females were kept in an environmental chamber at 23° C with a light-dark cycle of l6:8. Photophase initiation was at 0600 h. Each trial involved the introduction of a group of cdffbt weevil eggs into a container with a female parasitoid. The eggs were introduced on a moistened wedge of cardboard with a central depression to prevent them from rolling off. Each host egg group was left in the container with the female for 24: l h, then removed and replaced by a new group of eggs. The host eggs were known to be 0-24: I h old at the time of their introduction, and 24-48s2 h old at the time of their removal from the parasitoid container. A new group of eggs was presented to each female every 24: l h throughout her lifetime. Female parasitoids were exposed to 20-50 host eggs during each 24sl h trial period. The number used was determined by the availability of host eggs and by parasitoid age. Preliminary results indicated that oviposition is relatively greater during the first few days following the emergence of the females from the host egg, and declines rapidly thereafter. Therefore, more eggs were used in trials involving younger females. Only three of 4I5 trials resulted in the 21 parasitization of more than 90% of the available eggs, and in no instance were all eggs parasitized. No individual female parasitized more than 39.796 of the eggs available to her during her life. It was thus assumed that host egg availability did not limit parasitoid oviposition. The effect of mating status on fecundity was examined by recording the oviposition of females in three categories: I) virgin females, 2) females allowed to mate once within two hours after their emergence and then denied further access to males, and 3) females kept in containers in which one or two males were introduced within two hours of female emergence and allowed to remain throughout the life of the female. Females in the third category were thus not precluded from mating more than once. The females used in all three categories emerged singly from their host eggs. This prevented rrnating except where desired. The number of paasitoids which developed per host egg was shown to affect both adult longevity and size (Section 2). These factors may in turn affect fecundity. Therefore, the fecundity of a fourth group of parasitoids was monitored by selecting females from host eggs in which two parasitoids had developed. These females were placed in individual containers. One or two males were then introduced into each container within two hours of the female's emergence and allowed to remain throughout the life of the female. The terrparal distribution of parasitoid oviposition was determined by calculating the mean percent of total offspring produced during each day following emergence from the host egg. The progeny produced by each female received equal weighting regardless of the absolute level of production. If it is assumed that development and survival from oviposition to adult emergence is 22 constant, then the mean percent of progeny produced during each daily time interval will equal the mean percent of oviposition. Mated females were not segregated according to whether they were known to mate only once or were allowed continual access to males, as this distinction had no apparent relevance to temporal oviposition patterns. All of the eggs exposed to Patasson n. sp. females that failed to produce either parasitoid adults or carrot weevil larvae were examined under a dissecting microscope at 25x. Any evidence of parasitoid pupae or adults that had died before emerging were credited to the fecundity totals of the appropriate female. However, parasitoid eggs and larvae are small and difficult to see. It is probable that parasitoids which died at an early stage in their development were overlooked. 3.l.2 Oviposition Host eggs were exposed to newly emerged parasitoid females to determine the age at which oviposition begins. Parasitoids that emerged between 0600 and 0800 h were selected from a laboratory culture. Male-female pairs were placed in clear plastic containers at 0800 h t l5 min on the day of their emergence. Five carrot weevil eggs less than 24 h old were introduced into each container as described above. At I000 h 2 l5 min these eggs were removed and replaced by a new group of five eggs. This was repeated at l200 h and at I400 h 1 l5 min. Host eggs introduced at I400 h were removed at 0800 h 1 l5 min the following morning. The host eggs were then individually placed in clear plastic containers and reared at 23° C and the number of parasitoids emerging was recorded. 23 3.2 Results 3.2.l Fecundity The mean number of progeny among all females that developed singly within the host egg was 24.7 (s=32.3). However, one half of these 36 females produced no offspring. The mean for females that produced at least one offspring was 49.4 (s=29.3). The number per female ranged from four to 90. One of the females that was allowed continuous access to males produced only males among her I7 offspring. Therefore, it was assumed that this female was not fertilized despite the presence of a companion mole, and the female was considered to be a virgin in the analysis which follows. Mating status appeaed to have no influence on the fecundity of females that had at least one offspring. The mean numbers of progeny produced by virgin females, females that mated once, and those allowed continual access to males were 4I.I, 52.6, and 60.3, respectively (not significantly different, ANOVA, p>.05). Mating status also did not appear to affect the likelihood that a female would produce offspring. Five of l2 virgin females and three of II once mated females had no progeny, while I0 of I3 females allowed continual access to males had no offspring. Superpaasitism The mean number of offspring among females emerged from host eggs in which two paasitoids developed was 22.I (s=22.6). Twelve of the I8 females examined produced at least one offspring. The number per female ranged from four to 64. The mean for reproducing females was 33.2 (s=l9.7) offspring (not 24 significantly different from the mean for females in all mating status categories which emerged singly from the host egg, t-test, p>.05). Although not demonstrated statistically, a relationship between fecundity and superparasitism may still exist since females that developed singly produced an average of 49. I% more offspring than females from host eggs with two paasitoids. All of the females that developed gregaiously and produced offspring were paired with one or two males throughout their adult lives. However, two of these females produced only male offspring. It was assumed that these females (which produced four and I2 offspring) did not successfully mate. Since no significant relationship was demonstrated between fecundity and mating status, all females emerging from two-parasitoid host eggs were considered together, regardless of the sex of their progeny. The fecundity estimates which have been presented may represent an underestimate of the number of eggs laid by parasitoid females if superpaasitism contributed to an increased mortality rate among the immatures. Aeschlimann (I977) reported that mortality among Patasson lameerei was low during development. However, mortality due to laval competition has been observed for several related species (MacGill I934, Vidano et al. I979, Statterthwait l93l, Anderson and Paschke I969). Among the 7477 carrot weevil eggs monitored, 9.7% were inviable and no evidence of paasitoid or weevil development was found. Of those eggs exposed to parasitoid females that produced at least one offspring, I0.9% were inviable, while only 7.6% inviability was encountered for eggs exposed to females that had no offspring. These values ae significantly different (X2 test, p<.0l) and suggest that activity by ovipositing parasitoids may have played a 25 greater role in host egg mortality than was shown by the emergence of adult parasitoids. Competition among parasitoid larvae may have contributed to this additional mortality. Temporal Distribution of Oviposition The cumulative temporal distributions of oviposition by all virgin females, all mated females from host eggs in which two parasitoids developed, and all mated females that developed singly are presented in Figure 3. Each of these distributions is significantly different (X2 test, p<.0l). Most oviposition occurred during the first four days following emergence at 23° C. Oviposition by mated females from superparasitized host eggs occurred rapidly, and 84.9% of all oviposition by these females was completed within three days after their emergence. Significant oviposition (l6.4%) by mated females that developed singly in the host egg occurred during the first day of adult life. Oviposition proceeded at a relatively constant rate through the fifth day following emergence, by which time,86.8% of total oviposition had been completed. No oviposition was observed for virgin females during the first day. Beginning on the second day, oviposition proceeded rapidly and was 94.2% completed by the end of the fourth day. Sex Ratio The percent of females among the offspring of once-mated females and females allowed continued access to males was 77.7 and 7 I.2, respectively (Table 6). These values were calculated by summing the number of progeny of each sex produced by all females in the appropriate mating status category. The 26 ‘fl—WO—C “T - ‘ z. 3 l ....o... mm FEMALES/ONE PER EGG i ........ VIRGIN muss/ONE ma sac PERCENT OF TOTAL OVIPOSITION ...— HATED muss/Two PER sac I T I I I l 3 4 5 C 7 3 AGE (days) Figure 3. Effect of mating status and number of parasitoids per host egg on the temporal distribution of oviposition by Patasggg n. sp. at 23° C. 27 eaawm a. zaaaan Om Ommaonwsm whoaaaaa UK maamwo monoaaoa :. an“ an womwcooaoa aw one man Bowman anmnca Om moamwa. ZcBUmH Om Ommownmam anomaaaa fine now won Om moaupo banana Aaawav zmnwom onmnca H M u e .m a q a eOan wnamwa banana apnea oaama eonmw was am mp do am pm a n any mmamwma mm hm mm am we a a a and zmwoa pm my ya we pm a a N we a mmamwm mm.a am.» ma.n a~.u um.a am.u a.a a.a qq.q enamwo banana two: noonwaaaw a madman no amwma aonmw Nma Haw um qe NM A N II mau mnamwom Han ku hm m& up a a II haw zmwaa um we we we pp p N II Haw a unamwo m~.u aw.w mq.< aq.a ma.a a.a a.a II up.» a. coma Hounoaaan nan Ommawnwom Om aware mnamwma (awn: manhood omooww mnoa are same one. a. boom enunaooon nan Ommuunwao Om naunnaa: mmauwoa (awn: manhood mnoa 50am ammo macs area: one On etc vanaawanaa aaanmam. 28 contribution made by each female to the total sex ratio was thus weighted by the number of offspring she produced. Females within each mating category were not segregated according to superparasitism. The percent of female offspring among all mated females was 74.0. This corresponds with sex ratio estimates for closely related species (Anderson and Paschke I968). Female progeny were produced by once-mated females for six days after mating had taken place. Total reproduction by both groups was negligible beyond this time (l.0% of the total). This indicates that female paasitoids that mate shortly after emergence ae able to store sperm and oviposit fertilized eggs throughout most of their reproductive lives. Since the energy associated with additional mating efforts would not be compensated by any substantial increase in female progeny, it would appea unlikely that females would ordinarily mate more than once. Nimber of Eggs Paasitized The mean number of eggs parasitized per female was 33.3 (s=2l.5, n=30). This is more than twice the average number of eggs paasitized by Angphes M females (Anderson and Paschke I970a). There were no significant differences among mating status categories in the mean number of host eggs paasitized by females that parasitized one or more eggs (SNK test, p>.05). Thus, the mating status of female parasitoids does not appea to affect either the number of offspring or the number of host eggs paasitized. Similarly, no significant difference was shown between the mean number of eggs parasitized by females emerging singly from their host eggs and by females emerging from 29 superpaasitized eggs (t-test, p>.05). The mean number of parasitoids per parasitized host egg was nearly equal among all adult categories. 3.2.2 Oviposition Patasson n. sp. oviposition may begin within a few hours after the emergence of the female. All females examined had emerged between 0600 and 0800 h. One female began to oviposit between I000 and I200 h on the morning during which she emerged. Another female began to oviposit between I200 and I400 h. The intial oviposition of three females took place between I400 h and 0800 h an the following day. No oviposition occurred between 0800 and I000 h, when host eggs were first made available. However, only five of the SI females tested paasitized one or more host eggs. Thus, the central tendency of oviposition initiation can not be estimated. The low rate of parasitism in this experiment may be due to the low number of host eggs used in each trial. When 20.50 host eggs were used in the procedures described ealier, 55.6% (of the females produced at least one offspring. 3.3 Discussion The fecundity observed for mated females was greater than for virgins. However, this difference was not statistically significant. This may be attributable to an inherently high degree of variation associated with individual parasitoid fecundity. Stoner and Surber (I97l) obtained a similar result for Anophes ovijentatus. 30 Although initial oviposition was delayed, the percent of virgin females that eventually produced offspring did not appear to be different than the percent among mated females. This contradicts the findings of Williams et al. (I95I) who observed that only one of 20 virgin Patasson [15.192 females examined produced offspring. However, Stoner and Surber (I97 I) reported no disinclination to oviposit by virgin Anghes ovijentatus females. Superpaasitism was also found to have no statistically significant effect on fecundity even though the mean number of offspring produced by females which emerged singly from the host egg was greater than the mean for females which developed in superpaasitized eggs. A relationship between superpaasitism and fecundity was observed by Jackson “96” who attributed the reduced fecundity to the smaller size of females emerging from superpaasitized eggs. Most of the oviposition by both mated and virgin females occurred within 3-4 days of their emergence at 23° C. Mean adult longevity at this temperature was found to be 3.82 days, and survivorship was relatively high during the first few days following emergence, declining rapidly thereafter (Section 2). This suggests that most paasitoid females survive the critical period during which reproductive output is at a maximum, and a large portion of the female population will contribute to the next generation of paasitoids. This is in agreement with the findings of Anderson and Paschke (I970a) who observed that the number of paasitized eggs per female does not increase appreciably with the longevity of the female once they are more than a few days old. The initiation of oviposition by virgin females appears to be delayed for about a day. The benefits of this strategy would presumably be an increased 31 likelihood of mating and thus producing female offspring, and hence an increased contribution to subsequent generations per unit of reproductive effort. The opportunity costs associated with the delay would be related to the foregone reproductive output of males and the probability of death during the first day following emergence. By the third day after emergence, the currnulotive oviposition of virgin and mated females are approximately equal. The costs associated with longer delays ae mpaently prohibitive. The delayed oviposition strategy of virgins may also act as a regulating mechanism tending to equilibrate the population sex ratio. Paasitism of carrot weevil eggs has been observed to occur continuously from early .lune until the middle of August (Section 8). Parasitoid development was shown to require an average of II.9‘days at 2362, and mated females complete about 50% of their total oviposition within the first three days after emerging as adults. The average generation time is thus approximately I5 days at 23° C. This suggests that the parasitoid may complete up to five generations during the carrot weevil's oviposition period at this temperature. The percent of female offspring among once-mated females was 77.7. These females were observed mating within a few hours after their emergence, and the matings were successful since female offspring were produced by each of the adults observed. It is therefore assumed that this is the maximum possible percent of female offspring which will be produced by a field population in which all females have mated. Assuming that total fecundity by virgin and mated females is not significantly different, the percent of virgin females in a field population (V) may be estimated by: 32 (E0 3.I) V = I00 - ((f/77.7) - I00) where f represents the observed percent of females among parasitoids emerging from field-collected host eggs. Assuming that the percent of virgin females is related to population density, a relationship could also be derived between the observed sex ratio and field population density. This relationship is based on the assumption that the sex ratio among mated females is uninfluenced by field conditions. However, Flanders (I947) noted that among many parasitic hymenoptera, mated females tend to produce more female offspring at higher host densities. Jackson (I96I) observed that the rate of oviposition by mated Caaphractus M females influenced the sex ratio. The effect of environmental factors on the sex ratio among the offspring of mated Patasson n. sp. females is not known. Therefore, the application of the relationship described by Equation 3.I to field populations may only be useful for relative corrpaisons of populations assumed to be subjected to simila external influences. 33 4. Development Clausen (I940) observed that most mymaids ae solitary and only one immature parasitoid usually develops per host egg. However, gregarious development occurs in several Patasson and Angphes species, including Patasson sp. near sordidatus. No polyembryony has been observed among these genera. ' The developmental process and immature stages of several species of Patasson, Anghes, and related genera have been described (Balduf I928, Clark l93l, Statterthwait I93I, Jackson I96I, Anderson and Paschke I969). Development from the egg to the adult stage usually takes I0-l5 days. However, development under certain conditions may take considerably longer (Kevan I946, Bilboni I970). Several factors may influence developmental rates. The effects of temperature on development have been reported by a number of observers, including Fisher et al. (l96l), Anderson and Paschke (I970a), and Stoner and Surber (I97I). Developmental thresholds have been reported for Patasson lameerei by Leibee (I979) and for Anophes fl_av_i£_e_s_ by Anderson and Paschke (I969). Developmental time has been found to be influenced by sex (Leibee I979, Anderson and Paschke I969), host species (Statterthwait l93l), and relative humidity (Anderson and Paschke I968). Statterthwait (I93l) reported no correlation between the nun'ber of Patasson calendrae developing in the host egg and developmental time. However, such a relationship has been observed for some species (Jackson l96l). Vaiatiion in developmental time has also been found for different local populations of the some species, and at different times of the year among the same population. 34 The time required for the development of Patasson n. sp. from the egg to the adult stage is reported here. The effects of temperature, superparasitism, and sex ae examined. 4.I Methods Patasson n. sp. developmental times were estimated at I7, 23, and 29° C. Groups of 20-50 carrot weevil eggs were exposed to individual female parasitoids for 241 I h at 23° C. The same egg groups were also used to examine fecundity (Section 3). Each host egg group was then allocated among the three temperature regimes in one of three ways: I) divided equally Into three subgroups with each subgroup reared at a different temperature; 2) 2/3 reaed at 23° c and l/3 at 29° c; or 3) all reared at 23° c. The allocation among temperature regimes for a paticula egg group was determined by the availability of space within the environmental chambes used. The eggs in each subgroup were placed together in a 65 ml clear plastic container with a moistened 5.5 cm disc of Whatman #3 qualitative filter paper. Distilled water was added when necessary to maintain a high relative humidity, and all hatched carat weevil Iavae were removed daily. The eggs in each subgroup were kept in a common container for 4, 6, and 8 days at 29, 23, and I7° C, respectively. Preliminay observations indicated that the expiration of these intervals precedes the earliest possible parasitoid emergence. After the initial intervals, each egg was placed in a sepaate container. Eggs were then monitored for paasitoid emergence a minimum of six times per day at 0600, 0800, l000, I200, I400, and 2200 h. Some of these eggs were monitored more frequently to precisely determine emergence times in 35 conjunction with the procedures described in Section 5. The time interval during which each parasitoid emerged was thus determined and developmental times were calculated. 4.2 Results Patasson n. sp. development occurs entirely within the host egg. The only preliminary external evidence of paasitism is the anpearance of lesions on the host egg surface, which appaently result from penetration by the paasitoid ovipositor. Simila observations have been reported for Gonipterus scutellatus eggs paasitized by Patasson m (Clark l93l) and for Aga_bu_§ bipustulatus L. (Coleoptera: Dytiscidae) eggs parasitized by ggraphractus M (Jackson I96I). Paasitized carrot weevil eggs exposed to Patasson n. sp. females in the laboratory frequently exhibit many such lesions. Five or six lesions may appear on an egg from which only one paasitoid emerges. This suggests that many of the lesions represent unsuccessful attempts at host egg penetration, or that the mortality of immature stage paasitoids is high. The role of superpaasitism and paasitoid mortality was discussed in Section 2. The appearance of these lesions is highly vaiable and cannot be used as a consistent ealy indication of parasitism. The first reliable external evidence is the appearance of the red compound eyes of the pupae which as usually visible through the host egg chorion. Pigmentation of the compound eyes as an indicator of pupal stage development has also been noted for several related species (Niemczyk and Flessel I970, Statterthwait I93I, Kevan I946, Jackson I96I). Also visible at about the same time are three ocelli, dorsally, and the apically sclerotized mandibles, ventrally. The first appearance of compound 36 eyes among egg groups took place an average of ”.3, 8.l and 5.7 days after parasitoid oviposition at I7, 23, and 29° C, respectively. These data give approximations for the length of time required for larval development. One or twa days prior to the emergence of the adult parasitoids, the host egg becomes cornsiderably dakened. Black, pre-emergent adults may be visible at this time. The developmental process is terminated with the emergence of the adult from the host egg. 4.2. I Temperature The mean developmental times of Patasson n. sp. at I7, 23 and 29° C were I7.55, Il.93, and 8.68 days, respectively, demonstrating a significant inverse relationship between developmental time and tennpeature (SNK test, p<.0l) (Table 7). These data are based on the midpoints of the known intervals in which paasitoid oviposition and emergence occurred. Development took I02% longer at no C than at 29° C. The coefficient of variability at each temperature was relatively low, ranging from I0.I% at 29° C to I4.3% at 23° C. Successful development occurred at each of the temperatures examined. The percent of all eggs exposed to female parasitoids from which adult parasitoids were reared was 47.75, 4l.44, and 39.05 at I7, 23, and 29° c, respectively (not significantly different, X2 test, p>.05). These data are based only on egg groups which were allocated equally among the three temperature regimes and in which there was at least one parasitized egg. This implies that the temperature extremes examined do not approach the upper and lower developmental thresholds for this paasitoid. 37 Table 7. Developmental time of gatggggn n. sp. as influenced by temperature, sex, and number per host egg. Mean developmental time (h) Temperature ( C) 17 23 29 a No. per egg One Mean 17.90 12.41 8.85 SD 2.24 1.97 0.82 N 34 458 68 Two Mean 16.88 a 11.39 8.56 SD 1.27 1.17 0.90 N 18 391 58 Sex0 Male Mean 16.91 b 12.07 8.41 SD 1.97 1.82 0.88 N 29 340 69 Female Mean 18.36 a 11.84 9.00 SD 1.79 1.63 0.77 N 23 524 60 Totalb Mean 17.55 a 11.93 8.68 SD 2.01 1.71 0.88 N 52 866 129 a. In the same row» means followed by the same letter are not significantly different at p-0.01, Student-Neuman-Kuel multiple range test. In the same column, means followed by the same letter are not significantly different at p80.05, t test. b. Means followed by the same letter are not significantly different at p-0.0S, Student- Neuman-Kuel multiple range test. 38 4.2.2 Sex Mean developmental times of females were found to be significantly greater than means for males at both l7 and 29° C, but not at 23° C (Table 7). Although significant in two of three instances, the relative differences in developmental times were small: 8.6, I.9, and 7.l% at I7, 23, and 29° C, respectively. Sex-based differences in developmental times may thus play only a minor role in field population dynamics. 4.2.3 Superpaasitism Mean developmental times for paasitoids which emerged from host eggs in which one or two parasitoids developed were significantly different only for paasitoids reaed at 23° C (Table 7).~ At this tempeature, paasitoids developing singly took an average of L0 day longer to develop than those which developed in two-paasitoid eggs. Paasitoids which developed singly also required a somewhat longer developmental time at l7 and 290 C, but the differences were not significant. 4.3 Discussion Patasson n. sp. development from the egg to the adult stage took an average of I I.9 days at 23° C. Development was shown to be very sensitive to temperature. The high degree of variability under constant temperature conditions reported by Statterthwait (I93l) and by Jackson (l96l) was not encountered for this paasitoid. The completion of development was found to‘occur more quickly among parasitoids which developed in superpaasitized eggs at 23° C. Adult longevity 39 has also been demonstrated to be shortened among paasitoids from superpaasitized eggs (Section 2). A field population will generally be composed of individuals which emerged from host eggs in which vaious numbers of parasitoids developed. Thus, the vaiation in longevity and developmental times attributable to supepaasitism will tend to spread out the emergence and oviposition period of parasitoids originally oviposited on the same day. This may reduce the risks to the paasitoid population associated with a short-term reduction in host egg availability or temporally unsuitdnle environmental conditions. Estimates of developmental times, combined with adult longevity estimates (Section 2), indicate that most of the life of the parasitoid is spent inside the carrot weevil egg as an egg, larva, or pupa. Developmental time as a percent of total longevity is relatively constant. An average of 7|.8, 73.4, and 77.4% of the paasitoid's life occurred inside the host egg at I7, 23, and 29° C, respectively, even though the overall length of the life cycle more than doubled as the tennpeature was decreased from 29 to no C (Figure 4). Therefore, most of the life cycle occurs in a relatively insulated environment: inside the host egg, which is in turn inside the plant petiole. This may provide protection from external mortality factors such as low humidity, extreme temperatures, predation, hypepaasitism, and perhaps most significantly from the effects of chemical pesticides. A substantial portion of the total parasitoid population may survive pesticide applications while still inside the host egg even though the adults may suffer high mortality. 40 ”T rmnxnmms 20-1 Hill 15-- (I - ' 7’ . " / . , I / I V ” 10" ’1’ ’1’ ’1' /’ /’ I I r’ 1’ I, ’ 1 1 ’1 f ,4 V 1 1 I I ’ I / V ’I I 1’ ” I, 1’ ’ I I I ’ A t I ’ I 1 1’ 1’ 1’ I’ 1’ V I, ’1 ” ’I 4 ’ 1 V I 1’ I ’ ‘ I ’ ’ ’ ’ b ' 1’ ’ I 1’ d 1’ 1” 1I " 1’ /’ I I I ' ” ’I’ ’I ”’ ’4 1’ 1’ ’ ' ’ ” , 1 ’ 1 o _ A i 23 °c 29 °c TEMPERATURE Figure 4. Effect of temperature on the mean duration of.the life cycle of Patassgn n. sp. 41 5. Emergence Periodicity The diel pattern of emergence among Patasson and Anaghes species has received little attention. Emergence usually occurs during the morning. Anderson and Paschke 0970:) suggested that early morning emergence of M“; 3.011% was related to the high relative humidity which occurs during this time. However, in the laboratory they observed that increases in relative humidity did not always induce emergence, and suggested that a biological rhythm may be involved. Aeschlimann (I977) observed that A. M males tend to emerge earlier thm females. The effect of temperature and superparasitism on emergence patterns has not been reported. Three experiments were conducted to investigate the temporal pattern of adult parasitoid emergence. The first of these experiments was designed to test for the presence of diel emergence periodicity and to examine the potential roles of temperature and photoperiod. This study provided some evidence of the existence of such a periodicity. A second experiment was designed to more precisely define the dimensions of this rhythm. The final procedure represents at attempt to evaluate the relative roles of endogenous and exogenous factors in the observed periodicity. 5.! Methods 5.l.l The Role of Temperature and Photoperiod Carrot plmts were collected from m untreated portion of field #2 at the Hammond F arm on 6 md 8 August I980. These plants were brought to the laboratory and carrot weevil eggs were extracted. These eggs were randomly 42 assigned to one of three environmental chambers employing different combinations of temperature md photoperiod: l) constant temperature of 23° C md constant light; 2) constant ten'perature of 23° C with a photoperiod of l6:8, photophose beginning at 0600 h; and 3) temperature regime of 26° C for l6 h (0600 to 2200 h) and 20° C for 8 h, with constant light. Two hundred potentially parasitized host eggs were assigned to each of these temperature-photoperiod combinations. Each egg was placed individually in a 65 ml clear plastic container with a moistened 5.5 cm disc of Whatman #3 qualitative filter paper. All eggs were monitored once every eight hours at 0600, I400, and 2200 h. Parasitoid adult emergence was recorded for each eight-hour interval. Only the first parasitoid emerging from each egg was recorded. 5.l.2 Diel Emergence Pattern The results of the preceding study indicated that at a constant temperature of 23° C and a photoperiod of l6:8, most parasitoid emergence occurs during the first eight hours of light. The objective of the procedure described here was to more precisely define the parameters of this pattern. . Parasitized carrot weevil eggs were obtained by exposing groups of 20-50 eggs to individual parasitoid females. Egg groups were then allocated among three environmental chambers md reared at I7, 23, md 29° C. All eggs were subjected to a l6:8 photoperiod, with photophose beginning at 0600 h. The some egg groups were also used to examine parasitoid fecundity md developmental times (Sections 3 and 4). As the developing parasitoids approached the pupal stage (as indicated by the appearance of red compound eyes), the eggs were placed in individual 43 containers and monitored daily at two-hour intervals during the first eight hours of the photophose (0600, 0800, IOOO, l200, and MOO h) and again at 2200 h. Adult emergence was recorded (bring each of these intervals. Preliminary results indicated that a large portion of parasitoid emergence occurred (bring the first two hours of the photophose. The emergence pattern was further examined by continuously monitoring some of the eggs during this two-hour interval. Emergence times were recorded to the nearest minute from 0600 to 0800 h for a randomly selected subset of eggs reared at 23° C. 5.l.3 Role of Exogenous Cues It was hypothesized that acblt emergence was cued by photophose. This hypothesis was examined by recording the emergence of parasitoids which were reared under a l6:8 LD cycle, did then exposed to a single extended scotophase just prior to their emergence. If emergence was controlled primarily by an endogenous mechanism, then the scotophase extension should not significantly alter the emergence pattern following entrainment to the l6:8 LD cycle. However, if emergence was primarily controlled by exogenous cues (photophose), then the emergence peak should shift immediately to accommodate the altered photophose. A subset of the parasitized eggs used in the procedure described daove was selected for this experiment. Only eggs reared at 23° C were used. On each day that a trial was initiated, eggs were selected from which the emergence of adult parasitoids was believed to be imminent. These expectations were based on preliminary results of the developmental study which used the some set of parasitized carrot weevil eggs (Section 4). Eggs which had been exposed to the 44 some female parasitoid during the some 24 h interval were randomly divided into two groups. One group continued to be reared at 23° C with a photoperiod of l6:8, photophose beginning at 0600 h. The second group was placed in a different environmental chamber at 23° C in which the scotophase was extended by two hours, so that it would last until 0800 h. Eggs were exposed to the prolonged l0 h scotophase only once. Developing parasitoids which did not emerge during the photophose which immediately followed the extended scotophase were returned to the original l6:8 light-dark cycle beginning at 2200 h. Subsequent to the day of the scotophase extension, all eggs in both groups were monitored until no further parasitoid emergence occurred. Emergence for both groups was monitored at two-hour intervals from 0600 to MOO, md again at 2200 h. 5.2 Results Prior to emergence, fully formed adults were often seen moving about within the host egg. A circular exit hole was chewed in the host egg chorion. Once this hole was completed, emergence usually occurred within a few minutes. Parasitoids which developed in the some host egg usually emerged within a few moments of one mother. Occasionally the emergence of the lost adult from m egg was several hours later thm the emergence of the first adult. The newly emerged adults generally remained stationary within l-2 cm of the vacated host egg md vigorously preened-themselves for up to 30 min. 5.2.I Role of Temperature md Photoperiod No diel pattern was exhibited by the emergence of parasitoids from eggs reared under constant light md temperature conditions (Figure S-a, X2 test, p> 45 ‘5 EMERGENCE as EMERGENCE S S | I 95 EMERGENCE 8 8 l l 3 l 3 l flux. OQEV. A300 . czar. ””8 3.8 88 :8 3.8 38 88 :8. 23m .225. ...! la. I._ . HRXV b m Illlrld m ”moo .... Humane m. L noon ..o 36 3.3.6 .... anaconuw unannwccnwoa on cannonsom nnoa rape noon 6% convene: a. no. u: my nonnaman whore. oosonuon "cavonmncnmu UV oknwwo crunowonwoo. oocanman anaconmncnou moo av ooaunmon mesn. nkowwo noawonmacno. 46 .05). A pronounced pattern was found among parasitoids reared under cyclic regimes of either temperature or photoperiod (X2 tests, p<.05). Over half of the emergence from parasitized eggs kept in constant light and subjected to a cyclic temperature regime occurred during the eight-hour period at 20° C.(Figure S-c). During the first half of the l6-hour period at 26° C, 35.3% of the observed emergence occurred, and only 7. l% occurred during the second half of this period. These results suggest that the lower temperature was more conducive to parasitoid emergence, which would correspond to early morning in the field. Alternatively, parasitoid emergence could be associated with changes in temperature, rather than dasolute levels, since both the eight- hour low temperature period did the first half of the IG— hour high temperature period were initiated by 6° C temperature changes. Parasitoid emergence from eggs reared at a constant temperature and a light-dark cycle of l6:8 occurred predominantly during the first eight hours of the photophose (Figure S-b). Only l8.5% occurred during the remaining l6 h. These data again suggest that parasitoid emergence may occur in the field during the morning, when temperatures are lower. 5.2.2 Diel Emergence Pattern A distinct diel pattern was exhibited by the emergence of parasitoids reared at I7, 23, and 29° C with a photoperiod of l6:8. Peak emergence occurred during the first two hours of the photophose, and diminished rapidly thereafter (Figure 6). The percent of total emergence which occurred during the first eight hours of the photophose was 97.2, 8l.6, and 79.4 at l7, 23, and 29° c, respectively. This is in close agreement with the pattern described previously 47 $ 11°C 1 ‘ 8 ‘ n %Emetgence 3 2:. t 96 Emergence 3 i H a l 604 0 29°C 2 o 404 3’ a "7' In 20‘ .. ‘11—» c F osoo “00 Time Figure 6. Effect of temperature on the temporal distribution of emergence from host eggs by Patasson n. sp. for the emergence of parasitoids from field-collected eggs which were reared under the same temperature md photoperiod (Figure S-b). Although similar in general form, the relative temporal distributions at each temperature were significantly different from one mother (X2 tests, p<.05). Emergence was less concentrated in the first two hours of light at higher temperatures. The percent of emergence occurring during this interval was 66.0, 55.4, md 32.9 at l7, 23, and 29° C, respectively. These data suggest that the effect of photoperiod on emergence periodicity is mediated by temperature. Superparasitism did not affect the diel emergence pattern. The temporal emergence distribution for parasitoids reared at 23° C which emerged singly was not sig'iificmtly different that the distribution for those which emerged from host eggs in which two parasitoids developed lxz test; p>.05). The diel emergence patterns of males and females reared at 23° C were also not significantly different (X2 test, p>.05). However, among parasitoids which emerged from host eggs monitored continuously from 0600 to 0800 h, moles appeared to emerge earlier than the females (Figure 7). One hour after photophose initiation, the number of males reached 59 compared to only 2i females (X2 test, p<.0l). After two hours of light, the number of emerged females was greater that the number of males. 5.2.3 Role of Exogenous Cues The temporal emergence pattern among parasitoids reared at .23° C appears to be controlled by exogenous environmental cues. When the scotophase was extended by two hours, the emergence mode was shifted by two hours (Figure 8-b). The distribution of emergence for parasitoids exposed to the 49 NUMBER EMERGED TlME Figure 7. Cumulative emergence from host eggs by adult male and female Patasson n. sp. during the first two hours of photophase at 23“_C. 50 .nhoc ucosvoansm Ham :0 am can .cofimcmuxo on» «0 mac «nu Au co mzoum Houucou a Ca ammo scum can «name acoovomnaw ago :0 Ac can «codacouxo 0:» «a hop any an :0 cofincuuxo unannououm onu ou coaomxu ammo Baum «codacouxa unannououa n a a on nodum hop and :0 Ac mama GA No pawuomouogm o yucca cannon ammo poo: Ado scam mcwmuoEo Honssz .0 0mm pa .mn .: cowaouom Jason an ammo Hw>ou3 youuoo EOHM oucomuufio mo cOausnauuowo HouomEou on» :0 GOHuumouonn mo uuommm .m ousmwm l 89.. as: 83 hi i ——U ° Li .8 N .o F l .3 w m. _ l m 82 05:. 83 8a 83 38 t j I.“ u c U :3 N [ .0 I2 9 m 9 rll Ia m. u a .. v 1 r. n— o 51 photophose extension was significantly different than the distribution for the . control group (Figure 8-c) (X2 test, p<.0l). Saunders (I976) observed that the endogenous nature of a circadim rhythm associated with a particular activity is revealed upon transferring an insect from cyclic to canstmt conditions if the activity continues to occur at cpproximately the same time as before. The absence of a residual emergence peak during the interval 0600 to 0800 (which was normally the first two hours of the photophose) suggests that the timing of emergence was immediately shifted to conform to the new environmental conditions. No evidence was found to suggest that m endogenous mechmism caused the emergence rhythm to 'free run' in spite of altered exogenous cues. Upon returning to the 'normal' l6:8 photoperiod subsequent to the scotophase extension, the emergence peak shifted back to the original time interval. The temporal distributions of parasitoids exposed to the scotophase extension and of parasitoids in the control group were not sigtificantly different (X2 test, p>.05; Figure 8—d 8: e). 5.3 Discussion Temperature 01d photoperiod were shown to influence the emergence periodicity of Patasson n. sp. When parasitized eggs were kept in constant light with a cyclic temperature regime consisting of eight hours at 20° C and I6 hours at 26° C, most emergence occurred at the lower temperature. Most emergence from eggs kept at a constant temperature occurred during the first two hours of the photophose when exposed to a l6:8 photoperiod. This is in agreement with observations made for several related species (Aeschlimmn I977, Anderson and 52 Paschke l970b). The association between the initiation of the photophose and peak emergence was greater at lower temperatures. V No relationship was found between superparasitism and emergence periodicity. Sex appeared to have no major effect on the overall emergence pattern, but males tended to appear earlier than females during the first two hours of light. Earlier emergence by males may increase the likelihood that the males will be able to locate conspecific females. immediately following emergence, both males and females remain relatively stationary for up to 30 min near the vacated host egg md preen themselves. Emerging in advance of the females, the males will be able to complete this activity before the females and begin searching for the stationary females. Anderson md Paschke (l970b) suggested the possible involvement of a biological rhythm in the emergence of Anghes M35. MHovirever, the emergence of Patasson n. sp. appears to be the result of responses to exogenous environmental cues. No endogenous rhythm was demonstrated. 53 6. Courtship and Mating The mating behavior of several Patasson and Angphes species has been described. Mating usually occurs soon after adult emergence, especially among gregarious species in which sibling matings are common. Males of several species were observed to mate more than once. Anderson and Paschke (I969) reported that the average mating frequency among male m Mwas 3.5 times. They suggested that since this corresponds with - the sex ratio normally observed for this species, the females were likely to mate only once. Several studies were conducted in which the courtship aid mating behavior ' of Patasson n. sp. was examined. The first of these studies had two objectives: I) to observe md describe courtship md mating behavior, and 2) to determine how soon mating ca'i occur after omit emergence. The second study was designed to determine the frequency of mating among male md female parasitoids. 6.l Methods 6.I.I Courtship atd Mating Behavior Newly emerged male and female parasitoid pairs were introduced into inverted 65 ml clear plastic containers and their courtship and mating behavior observed. All of the parasitoids used had emerged singly from laboratory-reared host eggs a'ld were thus known to be virgin at the start of the observations. The emergence time of each parasitoid was known so that age at copulation could be recorded. The courtship and mating behavior of each male-female pair was observed with a 4x hand lens. Components of parasitoid courtship and mating behavior 54 were identified in preliminary observations. The sequence and duration of components were verbally described and recorded on a cassette recorder for each male-female pair observed. Data were later transcribed from the recordings, md the duration of specific behavioral components estimated with a stop watch. 6.l.2 Mating Frequency Virgin male aid female parasitoids were introduced by pairs into clear plastic containers md observed until copulation occurred. All of the parasitoids used were less that l2 h old when copulation first took place. After mating, each parasitoid was transferred to a separate container. Twenty-four :2 h later a virgin of the opposite sex was introduced into each of the 20 containers with the previously mated parasitoids. Each of the new virgins had emerged within l2 h of their trmsfer. Each new pair was observed for 30 min or until mating had taken place. If mating was observed both parasitoids were discarded. If no mating occurred, the introduced virgin was removed md a new virgin was introduced 24 t 2 h later. 6.2 Results 6.2.I Courtship and Mating Behavior Mating by Patasson n. sp. occurs soon after the adults emerge from pupation. Courtship and mating behavior was observed for both males and females within three minutes of their emergence. Sixteen of 26 virgin females examined mated within one hour of emergence, md another eight mated within two hours. 55 Little overt courtship behavior was observed for Patasson n. sp. females. The female usually remained stationary until the male attempted to mount her, at which time she elevated her dadomen. Five components of male parasitoid courtship md mating were indentified: I) general excitation, 2) wing fanning display, 3) antennation of the female abdomen, 4) mounting, and 5) copulation. Each of these components is described below in the order in which they normally occurred. General Excitation General excitation of the male usually occurred as soon as a virgin female was introduced into a container with a virgin male. This behavior was mmifested by a general increase in movement md by brief, discontinuous bouts of wing fmning. The wings of Patasson n. sp. are normally held horizontally over the abdomen. During wing fanning the wings were elevated to a vertical position and rapidly vibrated in short bursts lasting ca. ”3 s each. Wing fanning bouts during the period of general excitation usually involved l-4 bursts. The occurrence of this behavior was erratic. It sometimes began within l- 2 s of the introduction of the female, but at other times did not occur at all. The chrotion of this behavior was also quite variable, and it was difficult to determine when it begm and ended. Wing aning Display Male courtship display was characterized by a relatively prolonged bout of continuous wing fanning, involving up to 37 cycles of wing elevation md 56 vibration (i=l6.l, s 52:20). This behavior was observed only when the male was within IO—IS mm of the female. The average duration of this display was 6.4 s (s §=I.I). No cmsistent orientation to the female appeared to be necessary for this display. Antennation The male continued the wing fmning display as he moved into a position behind the female. Wing fmning ceased when the attennae of the male made contact with the posterior end of the female abdomen. The dadomen of a receptive female was usually elevated at this time. The male then begm to attennate the female's abdomen. Both male and female parasitoids remained stationary (bring this phase. Mounting The male then attempted to mount the female. Receptive females usually remained stationary (I'ld mounting was accomplished quickly. Unreceptive females attempted to walk away. The male generally followed, often haiging on to the retreating female. The male continued to antennate the female's abdomen while mounting. Antennation and mounting occurred so rapidly that estimates for each individual component could not be made. The average duration of the two activities combined was 2.9 s (s i=0’6) Copulation Copulation occurred with the female in a normal standing position, with her dadomen raised. The male leaned back, resting on his outstretched wings and 57 inserted the aedeagus at the mterior end of the female abdomen (Figure 9). The male often stroked the female's abdomen with his prothoracic legs during copulation. The average duration of copulation was 56.7 s (s §=I l.5). Following copulation, both the male md female usually remained within lO-IS mm of one mother md preened themselves for up to 30 min. The average length of the male courtship sequence, from the beginning of the wing fanning display to the initiation of copulation was 8.8 s (s J-‘=l.2). Successful mating sequences always involved the wing fmning display, antenna- tion md mounting components, in that order. This sequence was interrupted only by the escape of the female by walking or flying away. If the female mmaged to get more than lO-IS mm away from the male, the male would revert to the general excitation stage or would cease to exhibit courtship behavior. 6.2.2 Mating Frequency Male parasitoids readily mated more than once. All of the ten males examined mated with virgin females during each of the first two days following their emergence. Four of five males mated a third time within one hour of previous copulation on the second day. None of the ten females were observed to mate a second time under the laboratory conditions. Although multiple mating by females under field condi- tions is not entirely ruled out, it appears to be unlikely. It has been demonstrated that females are capable of producing female offspring for up to six days after mating, and total parasitoid oviposition beyond this age is greatly reduced (Section 3). Therefore, it would appear that no selective advantage would accrue to females that mated a second time to offset the associated time, energy, and risk. 58 .EOauoHamoo moduap .ma .: _dddddqdm mo cofiuwaom .o unamwm 59 6.3 Discussion Patasson n. sp. readily mated within a few minutes after emergence from the host egg. Matings were frequently observed between siblings which emerged from the some host egg. Mathews (I975) stated that such matings a'e common among parasitic Hymenoptera. Similar mating behavior has been observed for other Patasson md Angphes species (Mossop I929, Statterthwait l93l, Williams et al. I95l, Aeschlimmn I977). A delay in mating following emergence has been reported only for Anaghgs @293 (Anderson md Paschke I968). Doweil and Horn (I975) noted that early, predispersal mating between siblings may tend to lower the genetic variablility within a parasitoid population. A relationship between genetic variability md success as a biological control agent has been suggested (Tumbull and Chmt l96I, Simmmds I963). Force (I967) has also recognized variability as a beneficial attribute of m effective natural enemy. However, F lmders (I947) noted that gregarious parasitoid species generally mate earlier, which results in a more stable sex ratio. This has been supported by observations which suggest that parthenogenic reproduction by virgin females is infrequent among field populations (Kevan I946, Williams et al. l95l). The advmtages associated with a stable sex ratio apparently compensate for the reduced genetic variability among many arrhenotokous species. The importmce of a well-balanced sex ratio is apparently reflected in the observed delay in the initiation of oviposition by virgin Patasson n. sp. females (Section 3). 60 7. Host-Parasitoid Synchrony Low parasitism rates have been observed among several Patasson and Angghes species during the early part of the host‘s oviposition period (Statterthwait I93I, Anderson md Paschke I968, Bilboni I964, Dysart I97I, Ellis I973, Aeschlimann I977). This could result either from difficulty in surviving periods when primary host eggs are unavailable, or from poorly synchronized host md parasitoid life histories. The degree of synchrony may vary between locations (Mailloux md Pilon I970) or yearly at the same location (Gage md Haynes I975). Carrot weevil adults overwinter in areas adjacent to previously infested fields (Chmdler I926, Wright and Decker I957). Overwintering may also occur on the crop site. However, when rotatim of host md non-host crop plmts occurs, the locations of carrot weevil egg populations will change yearly. A concomitmt movement by the parasitoid is thus required. Parasitoid dispersal md host-searching abilities have been the subject of many investigations, especially as they relate to the introduction of exotic parasitoid species (Juillet I960, Hendricks I967, Anderson and Paschke I970), Maltby et al. l97I, Biever I972). Environmental factors which may influence dispersal include temperature, relative humidity, wind velocity and direction, and crop density. The objective of this study was to examine the degree of spatial and temporal overlap between Patasson n. sp. and the carrot weevil. 62 parasitism begins following the intiation of carrot weevil oviposition. .Weeds were rmdamly collected and brought to the laboratory for examination. Carrot weevil eggs were extracted md reared to determine the presence md extent of parasitism. All weed species present were included in samples collected on 9 May I980. Carrot weevil eggs were found in only one weed species (Me; sp.). Subsequent samples taken on I8 May md 4 June included only this weed species. Parasitoid Overwintering Overwintering (or aestivation) by several Patasson and Anghes species takes place in the form of advmced larvae inside the eggs of the primary host (Streams md Fuester I966, Aeschlimmn I977). The role of carrot weevil eggs in the overwintering of Patasson n. sp. is not known. Carrot plmts were collected in field #2 at the Hammond Farm in an area which had high population densities of both weevils md parasitoids earlier in the season. All plants in a co. 2 m2 area were sampled on 26 September I979, about one month after oviposition by carrot weevils had ceased (Section 8). These plmts were taken to the laboratory and examined for the presence of Patasson n. sp. in carrot weevil eggs. 7. I .2 Spatial Synchrony Carrot plants were collected from the ten western-most seven-row beds of field #2 at the Hammond Farm on I I July and I August I979, and examined for both carrot weevil infestation md parasitized weevil eggs. Thirty plants were collected from each of the ten western-most beds of the field on each date. 64 sarmles of this weed species collected on I8 May, but none were found on 4 June. These data indicate that carrot weevil oviposition began at least as early as 9 May in I980. Carrot weevil larvae were reared from all nine of the eggs collected from weed samples; none were found to be parasitized. Parasitism of carrot weevil eggs may not yet have been initiated, or a low parasitism rate coupled with the small size of the sample examined may have prevented detection of sparsely distributed parasitized eggs. No carrot weevil eggs were found among the weeds collected on 4 June. However, four eggs were collected from carrot plmts sampled on the previous day from m adjacent carrot bed (Section 9). One of these four eggs was parasitized by Patasson n. sp. Thus, evidence of parasitism appeared as early as 3 June during the spring of I980. Parasitoid Overwintering No carrot weevil eggs (parasitized or otherwise) were found in 64 carrot plants collected on 26 September I979, although there was abundant evidence of earlier weevil activity among the plants collected. Carrot weevil eggs thus appear to have no role in Patasson n. sp. overwintering. However, the sample size was relatively small and the possibility that the parasitoid overwinters in carrot weevil eggs at very low population densities cannot be entirely eliminated. 7.2.2 Spatial Synchrony The total percent of plants infested in the western most I0 beds was 39.3 and 40.3 on II July and I August, respectively. Weevil activity was more 65 concentrated along the field margin throughout the growing season. Parasitism rates appeared to be uncha'tged over space, even though the absolute number of available hast eggs decreased toward the field interior. The number of parasitized eggs in each bed on both sampling dates was not simificmtly different than the number which would be expected if a constant parasitism rate was assumed throughout the ID beds (x2 test, p>.05). These data indicate that the parasitoid and host populations were spatially synchronous within the study field. The contribution of the parasitoid to the natural control of the weevil appears to be uniform over the field, and is apparently not affected by' host egg density or distmce from the overwintering sites. 7.3 Discussion Carrot weevil oviposition was observed to begin early in May I980 in the weeds adjacent to a previously infested carrot field. Oviposition on carrots was not observed until early June. Although Patasson n. sp. adults were active in mid-May, parasitism of carrot weevil eggs was not observed until early June. However, the number of weevil eggs examined early in the season was small, so that the relationship between the initiation of weevil oviposition and parasitism could not be determined with certainty. The coincidence of initial weevil oviposition md parasitism will affect parasitism rates early in the season. This is especially importmt in celery production where seedlings are transplmted to the field in April in Michigan. If substmtial parasitism does not occur until later in the weevil oviposition period it may be too late to significmtly influence the level of crop damage. 66 Environmental conditions may differentially affect the emergence from overwintering of a parasitoid and its host (Gage md Haynes I975). Thus, the terrporol synchrony of early season weevil and parasitoid activities warrmts further study under different environmental conditions. Achlt Patasson n. sp. were active in the field several months after the end of the carrot weevil oviposition period. No evidence was found of parasitoid overwintering in carrot weevil eggs. Since overwintering of adults among Anghes md Patasson species has not been reported, an alternative overwinter ing host is suggested for this parasite id. The mechmism by which Patasson n. sp. overwinters will play a critical role in determining the size of the parasitoid population at the onset of carrot weevil oviposition. The low parasitism rates that have been observed early in the weevil's oviposition period suggest that overwintering may represent a weak link in the parasitoid's life history. Spring dispersal of overwintering carrot weevil adults mpears to occur prior to the initiation of oviposition. When host and non-host crops are rotated, there is little overlap in the year-to-year weevil distributions except in areas adjacent to infested fields. However, parasitism rates were found to be uninfluenced by distance from the field margin during I979. This suggests that the dispersal abilities of the parasitoid are sufficient to accommodate the yearly shift in spatial distribution by the weevil. 67 8. Host md Parasitoid Resource Utilization Carrot weevil oviposition patterns reflect the way in which the weevil exploits available host plants. Host egg availability will in turn influence host finding and exploitation by Patasson n. sp. The objective of this study was to examine the relationship between weevil and parasitoid resource utilization strategies. . Suitable host plants serve as food md oviposition sites for adult carrot weevils, and subsequently as a food source for their larvae. The availability of suitable host plants may represent a major limiting factor on carrot weevil populations in non-agricultural settings. Oviposition strategies determine the efficiency with which the weevil utilizes' this resource in producing viable offspring. The mmner in which this major resource is exploited is thus of particular interest. Host plants must reach a certain size or developmental threshold before they are used as oviposition sites by the carrot weevil (Boyce I927, Wright I953, Stevenson I976b). Oviposition ends in late summer or early fall, apparently in response to shortened photoperiod and reduced temperature rather thm specific plmt conditions (Whitcomb I965, Stevenson I976b, Simonet md Davenport I98I). Therefore, this resource may be defined as all suitable host plmts which have attained the required size threshold during the weevil's oviposition period. Carrot weevil egg clutch size is generally low. Observed mem clutch sizes have rmged from L3 to 5.0 (Harris I926, Boyce I927, Pepper md Hagrnmn I938, Wright md Decker I958, Martel et al. I976). The number of clutches per plmt is also generally low (Whitcomb I965). This results in a low egg density per plmt. Carrot weevil larval densities per plmt are also usually low. In naturally 68 occurring field populations it is rare to find more than a few late instar larvae per plant. Chmdler (I926) reported as many as 2-3 larvae per carrot plmt. A similar observation was made by Boyce (I927) on celery. Several observers have reported instances of higher numbers of larvae per plant. Ryser (I975) reported as my as seven larvae per carrot plmt- Wright (I953) found up to l5, and Pepper (I942) reported as many as 22 larvae per plmt. However, these observations are difficult to assess, as mem values are not reported. The number of various instar larvae also are not given. Carrot weevil oviposition patterns were observed during the I979 growing season in a commercial carrot field. Spatial and temporal relationships were monitored throughout the oviposition period. Egg clutch size and per plmt densities were recorded and their relationship to various measures of plmt status evaluated. Larval populations were also observed. Optimal carrot weevil oviposition strategies will maximize the number of surviving offspring per unit of reproductive effort. The carrot weevil is m endemic species. It has presumably evolved its oviposition strategies within a resource array unaffected by man. These strategies may be suboptimal within a I commercial agricultural setting. Conditions significantly different than those in which the strategies originally evolved may be encountered. Therefore, m experiment was conducted to evaluate the efficiency of naturally occurring oviposition strategies within the altered resource context represented by a commercial carrot field. The relationship between the survival of larvae and artificially imposed oviposition strategies in the form of various egg densities per plant was examined. Carrot weevil eggs represent a fundamental resource to the parasitoid 69 population. Weevil oviposition strategies will determine the spatial and temporal availability of host eggs. The strategies employed by the parasitoid in exploiting this resource will determine the effect of the parasitoid on the weevil population. Patasson n. sp. resource utilization was examined by observing the relationship between parasitism md host egg clutch size and per plant egg densities. The effect of plmt status on parasitism rates, md the spatial distribution of parasitoids were investigated. 8.I Methods 8.I.I Field Observations Carrot weevil egg md larval distribution patterns md parasitism were monitored in m untreated portion of a commercial carrot field during the I979 growing season. Carrot plants were periodically collected in m untreated am of field #2 at the Hammond Farm near East Lmsing in Clinton County, Michigm. let Sampling Procedure lets were initially collected on I3 June from the western-most seven- row carrot bed to provide a preliminary estimate of the extent of the carrot weevil distribution. Seven plmts were randomly collected every five meters. Subsequent plant samples were collected from the three western-most carrot beds. All plmts within a I-2 at2 area in each bed were collected per sampling period. A mem of HS plants was collected in each bed on each sampling date. The general sampling locations on each sampling date were 70 randomly selected. Sorrples were taken from immediately adjacent areas in the three beds. Within each bed, a board was placed perpendicular to the direction of the carrot rows. Retractable measuring tapes were attached to each end of the board, separated by a known distmce. As each plant was collected, its location was recorded by noting its distance from the attachment point at each end of the board. The triangular measurements for each plant were later converted to grid coordinates so the precise location of each plant could be determined with respect to all other plmts collected within the bed. The approximate area sampled in each bed was estimated by determining the minimum md maximum values for both the x and y coordinates among the plants within each bed. Carrot Weevil Egg md Larval Data All plants collected were brought into the laboratory where they were carefully examined for the presence of carrot weevil eggs md larvae. The data collected for each plmt included: I) the number of adult weevils (a rare occurrence), 2) the number of feeding or oviposition punctures, 3) the number of hatched and unhatched eggs in each egg clutch encountered, and 4) the number of each instar larvae found md their location (in the root or foliage). Plant Data The total and root fresh weights were recorded for all of the plants sampled. Additional plant dimensions were recorded for a randomly selected subsample of approximately l096. These data included: I) the dry weight of the foliage and the root, 2) the length of the taproot md the maximum foliage 71 height, 3) the maximum diameter of the taproot, and 4) the area of the leaves and stems. Parasitoid Data The presence of Patasson n. sp. was not known prior to the 28 June san'pling date. Thus, no parasitoid data were available from plants collected on the first three sampling dates. Beginning with plants collected on 28 June, carrot weevil eggs were reared in the laboratory to determine parasitism rates. A portion of the eggs extracted from plmts collected on 28 June were allocated to several clear plastic containers. An average of 37.5 eggs were placed in each container, without regard to their egg clutch or plant source. These eggs were reared at room temperature. The relative humidity was maintained at a high level by periodically adding distilled water to a filter paper disc located in the bottom of each inverted container. The total number of parasitoid adults emerging from these eggs was recorded. The eggs from each clutch extracted from plmts collected on 9 July were reared in a separate container. The source plmt for each clutch was known so that parasitism rates could be compared to the available plant dimensions. All containers were mmitored daily, and distilled water added when required to keep the filter paper moist. Monitoring of each egg group (clutch) continued until the status of all eggs was determined (i.e., until a carrot weevil larva hatched, a parasitoid emerged, or the eggs were clearly inviable). After extraction from the plmt, approximately three weeks were required to determine the ultimate disposition of all eggs. The sex, number, md emergence date for all parasitoids from the eggs in each clutch were recorded. It was also noted whether the 72 cavity from which the eggs were extracted was sealed with a fecal covering. The eggs extracted from plmts collected on all three subsequent sampling dates were reared individually. This allowed a determination of the number of parasitoids emerging from each host egg. 8.I.2 Survival of Carrot Weevil Larvae Mechanisms which may account for the low density of late instar larvae which has been observed in the field were examined experimentally by mmually transferring various numbers of eggs to plants in m uninfested carrot field, md recording the number of late instar carrot weevil larvae and pupae which successfully developed. Carrot weevil eggs less thm three days old were obtained from a laboratory culture. These eggs were mmually trmsferred to carrot plants within a pesticide-free carrot field located at the Michigm State University Experimental Muck Farm, Bath, Michigan. The eggs were transferred using a fine bristle paint brush dipped in carrot baby food. The eggs were placed on the carrot plmts near the plant base at the junction of two or more stems. The carrot baby food facilitated the adherence of the eggs to the plmts. Different per plmt egg densities (l,2,4,6,8,l0,l5, and 20) were rmdamly assigned to every third row. Two guard rows thus separated the carrots in each treatment (egg density). From five to seven replications (plmts) of each density were tested, with all plants within a row receiving the same number of eggs. This procedure was repeated four times at one-week intervals. The survival of weevil eggs was determined by collecting the plants 73 approximately three weeks after eggs had been placed on them. The plmts were brought to the laboratory and examined for the presence of carrot weevil larvae. The numbers of each instar were recorded. Total and root fresh weights were recorded for all plants. The soil surrounding the sampled plants was also examined. All soil within I0—l5 cm laterally of the plants was collected to a depth of ca. I5 cm. Pupae found in this soil could not be assigned to specific plants, but were allocated proportionally to the plants in the treatment. The plants in all guard rows and those adjacent to the treated plants were also collected and examined for the presence of carrot weevil larvae. This was done to determine the extent of my larval migration between plants or rows. 8.2 Results 8.2.I Carat Weevil Resource Utilization Larval Density Per plmt larval density remained low throughout the growing season; on average of l.46 larvae per infested plant was recorded (Table 8). Only 3.78% of the plmts had four or more larvae. Similar distributions were found for each larval instar. Larvae mpeared to feed first on the stems in the crown area of the plmt. Most first and second instars were found in the foliage (90.2 and 5l.4%, respectively). After this food source was exhausted, the larvae moved below ground to feed on the root. Most third and fourth instars were found in the root of the plant (66.4 and 88.8%, respectively). Whitcomb (I965) observed that on smaller carrot plants with 2-4 leaves, the larvae usually go directly to the root 74 Table 8. Mean per plant density of carrot weevil larvae by larval instar and sampling date, 1979. Mean number of larvae per plant --------------- Sampling date------------------ 13 21 28 9 21 1 13 Larval instar Jun Jun Jun Jul Jul Aug Aug Total 1st instars Mean 1.05 1.35 1.28 1.26 1.33 1.29 1.00 1.26 SD 0.22 0.81 0.62 0.60 0.51 0.49 0.00 0.62 N 20 34 46 38 6 7 2 153 2nd instars Mean 1.33 1.31 1.00 1.38 1.22 1.00 1.27 1.25 SD 0.58 0.85 0.00 0.83 0.76 0.00 0.47 0.70 N 3 13 11 32 36 6 11 112 3rd instars Mean --- 105a 106a 1007 1027 1019 1006 1019 SD --- 0.71 0.00 0.27 0.59 0.48 0.00 0.49 N --- 2 3 14 41 31 9 100 4th instars Mean --- --- --- 1.00 1.26 1.18 1.00 1.19 SD --- --- --- 0.00 0.60 0.39 0.00 0.48 N --- --- --- 5 38 22 10 75 Total Mean 1.09 1.38 1.30 1.53 1.77 1.44 1.17 1.46 SD 0.29 0.82 0.66 0.97 1.06 0.66 0.46 0.85 N 23 48 56 73 86 54 30 370 75 upon hatching md frequently kill the plant. To the extent that smaller plmts were killed in this manner md thus excluded from the samples collected, the earlier data may underestimate the proportion of larvae located in the carrot roots. Fourth instars were not observed until 9 July, md substantial numbers of fourth instars (5% of all larvae) were not observed until 2l July. This suggests that only one corrplete generation was produced in carrots in I979. Otto (I978) observed two generations on celery in Michigan in I977. However, transplmting of celery seedlings into the field usually begins early in April. Therefore, celery plmts developed sufficiently to serve as carrot weevil oviposition sites are availdale in the field rmch_earlier thm carrots. Also, the spring of I977 was exceptionally warm and adult activity was first noted several weeks earlier than in I979. Carrot weevil oviposition ceases in mid-August. Thus, weevils produced during the current season on carrots would have little time for their own oviposition. It is assumed, therefore, that most of the oviposition which occurs during a particular growing season is done by adults produced during the preceding year. Egg Clutch Size Mem carrot weevil egg clutch size was I.99 eggs during the I979 growing season (Table 9). An egg clutch was defined as all eggs, hatched or unhatched, located in the same oviposition cavity. This mem is intermediate to the values reported by Wright md Decker (I958) and by Martel et al. (I976). Clutches ranged from one to seven eggs, but only 9.4% of all clutches) had four or more 76 Table 9. Number of carrot weevil eggs per clutch and per infested carrot plant and the number of clutches per infested plant as affected by sampling date. No. of No. of No. of clutches/ eggs / Sampling eggs la infested infested date clutch plant“ plant° 13 Jun Mean 1.49 d 1.29 b 1.93 b SD 0.69 0.64 1.28 N 280 217 217 21 Jun Mean 2.00 b 1.53 a 3.07 a SD 1.02 1.02 2.39 N 336 219 219 28 Jun Mean 1.82 c 1.43 a 2.61 ab SD 0.95 0.70 1.75 N 228 159 159 9 Jul Mean 2.24 a 1.53 a 3.42 a SD 1.17 0.89 2.48 N 220 144 144 21 Jul Mean 2.20 a 1.65 a 3.63 a SD 1.15 1.05 2.78 N 196 119 119 1 Aug Mean 2.17 ab 1.39 ab 3.02 ab SD 1.10 0.75 2.07 N 132 95 95 13 Aug Mean 2.40 a 1.33 ab 3.20 ab SD 1.09 0.68 2.20 N 80 60 60 Total Mean 1.99 1.45 2.87 SD 1.15 0.72 2.20 N 1472 1013 1013 a. In the same row, means followed by the same letter are not significantly different at p-.05, Student-Neuman-Kuel multiple range test. 77 eggs. Mem clutch size on each sampling date ranged from I.82 to 2.40 eggs per cavity. Mem clutch size was significmtly lower early in the season, but no differences were found among samples collected from 9 July through l3 August (SNK test, p>.05). This is the reverse of the trend observed by Niemczyk md Flessel (I970) for Hyper-g M wherein clutch size decreased as the season progressed. Clutches Per Plant The mean number of clutches per infested plmt was I.45 over the entire period of observation (Table 9). This is somewhat lower than the value reported by Whitcomb (I965). The mem number of clutches per plmt was relatively constmt (bring the weevil oviposition period, although the initial mem was significmtly lower thm the mems on several subsequent sampling dates (SNK test, p<.05). Eggs Per Plant The combination of low mem clutch size and the low number of clutches per infested plmt resulted in a low number of eggs per infested plant. The mean number of eggs per infested plmt (including hatched md unhatched eggs) was 2.87 for the entire observation period (Table 9). As my as l8 eggs were found on a single plant, but 89.4% of the infested plmts had five or fewer eggs. Mems for each sampling date ranged from I.93 to 3.63 eggs per plmt. Although simificmt differences were found, no temporal trend was apparent. 78 Egg Density and Plant Status No consistent relationships were found between carrot weevil egg density md plant growth status on my sampling date, as measured by the correlation coefficient between the number of eggs on a plmt and maximum foliage length, root length, leaf area, foliage dry weight, md root dry weight. Apparently, carrot weevil oviposition was not affected by the size of the host plmt once the plmts had reached the necessary size threshold for attack. Among plmts sampled on I August, significmt negative correlations were found between egg density md several measures of plant size. However, this occurred near the end of the carrot weevil oviposition period, and may reflect plant growth inhibition resulting from earlier weevil infestation. Spatial Distribution The spatial distribution of carrot weevil eggs within each bed at each sampling was measured by arbitrarily dividing the area sampled into 30 x 30 cm grids md recording the number of hatched and unhatched eggs within each grid. The mem number of eggs per grid md the associated variance was calculated by considering only grids which contained at least one carrot plant (Table I0). The distribution of carrot weevil eggs appeared to be aggregated within grids (as measured by the magnitude of the variances relative to the associated mems). Since egg density per plant was law, this distribution may be the result of limited movements by female weevils rather thm the spatial aggregation of several females. The distribution of plants within grids was also somewhat aggregated. A significant correlation was shown between the number of carrot weevil eggs and 79 .Auaau he so >9 ~a.&nm an .s an couauwocw ma.anm ua mucouflmwcmwm .u eUHHm 5000 Ed QUGMHQ HO H0955: 0:“ 0G0 >ufiacop mmu Hw>oo3 uouuou ucaam pom onu aoo3uon ucowuammuou sawuaaouuou .n .CMHm Sumo cw ammo Hw>oo3 #OHHQU HO HOQE§G can OHGMHQ HO H0939: $003903 HQOHOMNHOOU COfiUOHOHHOD e0 men.l emu. no am.m mh.m me ae.- om.~ and ma NHH.I oo Ham. mo m¢.m oo.m we co.mm em.v mad a «an.l oo «mm. v0 ah.m Ne.¢ Hm mo.em no.0 Han Hm 5mm. so «mm. mm mm.m eh.v em m~.mm ea.h How a was. so mem. me do.o ma.m Ho m¢.m~ No.m can on maa.l ss men. we ma.¢~ mo.m no no.5v $0.0 can am «a Hz 2 .Ha> com: 2 .Ho> cam: ouac a mcwamEam uouaufiUAMMoou ouaaam uouuou ammo Hfi>ou3 uouuou ceauoaouuoo .mhmu .o0wum EU on x mm moose ammo Hw>oo3 uouuou can aucoam uouuoo mo cofiuanduuoao Aaauamm .aa canes 80 the number of plmts per grid as expected, since the mean number of eggs per plmt was fairly constant (Table 9). No density-dependent associations were found between the mem number of weevil eggs per plant md plant density. ' Carrot Weevil Egg Density md Survival In the artificial infestation experiment, few carrot weevils successfully developed to reach the third or fourth instar or the pupal stage regardless of the initial egg density. No relationship was found between the number of carrot weevil eggs originally placed on a carrot plant md the subsequent number of larvae ’md pupae recovered (ANOVA of the raw data and data trmsformed by X’a/Xm, p>.05). This indicates m inverse relationship between larval survival rate and initial egg density per plmt (Figure I0), and suggests the operation of a density-dependent mortality mechmism in the field. A similar relationship was noted byEl-Dessouke md Stein (I970) who observed that the survival rate of the weevil 3.922 hispidulus was reduced by intraspecific competition at high larval densities. The destructim of unhatched eggs has been observed to occur in the laboratory as the result of chewing by newly hatched carrot weevil larvae on unhatched eggs. Thus, cannibalism could contribute to the density dependent mortality which was observed. Although not statistically significant, the average number of larvae and pupae developing per plmt increased as initial egg density increased. Thus, optimal oviposition under conditions of scarce resources (e.g., suitable host plmts) may shift to a higher per plmt density than those which were observed in commercial field situations. 81 20-4 0 ..l .< E: o >' - a: l5 2) 00 'l- E (3 101 P a: U] a. 0 :2 if. s-I __. 2 O o I r i ‘1 I I l l 2 4 6 8 10 is 20 NUMBER OF EGGS Figure 10. Mean percent of carrot weevils surviving to late instar larvae or pupae after transfer of eggs to uninfested carrot plants (1og.Y - —1.47 - 0.53 log.x, R = 0.862). 82 Significant differences were demonstrated between the number surviving per plant within each egg density class among eggs transferred on different sampling dates (ANOVA, data transformed by Xem p<.05). This suggests that survival conditions may change during the growing season in response to plmt conditions, weather conditions, or other environmental factors. However, the eggs were trmsferred under unnatural conditions, so the observed temporal differences may have little relevmce to weevil-environmental relationships under more natural conditons. No consistent correlation was found between the number of larvae surviving per plmt md several measures of plant size within each sampling date. No evidence of carrot weevil larval infestation was found in my of the guard row plants. Pepper (I942) and Whitcomb (I965) have reported that the larvae are capable of moving between plant roots. However, such movement appears to be a rare occurrence and may not be a significant factor in larval survival unless it is important early in the season when plmts are much smaller. 8.2.2 Parasitoid Resource Utilization Parasitism Rates The percent of carrot weevil eggs parasitized during the I979 growing season was substmtial, reaching a peak of 49.32 on I August (Figure II). The average percent of eggs parasitized during the period of observation (28 June to l3 August) was 22.8. These data suggest that parasitism by Patasson n. sp. may have significmtly affected the carrot weevil population. In addition, carrot weevil larvae hatched from only 55.77% of all eggs examined, and an additional 2l.40% were inviable. Part of the observed inviability may have resulted from undetected parasitism. PERCENT 83 10¢ DMUJBEE Li _.__, v 73H CNHKH“WEWILIUEWAE so- 25‘ EMRMHHIZED C I— ] I I assure “UL 2140:. 1AUG 13 AUG DATE Figure 11. Temporal distribution of parasitism by Patasson n. sp. and inviability of carrot weevil eggs in 84 Increased mortality among developing parasitoids has been observed to result from superparasitism (Anderson md Paschke I969, Jackson l96l, Miller I966, MacGilI I934). Multiple parasitism may also diminish parasitoid survival. Anderson md Paschke (I969) observed that as the number of Athes m females allowed access to a given number of host eggs increased, the number of parasitoid eggs per host egg increased md the survival rate decreased. Early instar parasitoid larvae are difficult to detect. Thus, some of the inviability among carrot weevil eggs may have been the result of parasitoid larvae which themselves failed to develop as a result of high egg density per host. Paasi'taid md Host Egg Density The density of unhatched carrot weevil eggs declined from 86.3 per m2 on 2| June to 20.0 per m2 on I3 August (Figure l2). This decline was mparently unrelated to parasitism by Patasson n. sp. since most or all of the oviposition was attributable to overwintered females. Parasitoid density, as measured by the number of immatures reared from host eggs, increased to 40.5 per m2 by I August, md then‘declined. On I August the number of immature parasitoids per m2 exceeded the number of host eggs. The temporal pattern of parasitism by Patasson n. sp. was characterized by a low parasitism rate during late June md early July when host egg densities were highest. Similar trends have been observed for other Patasson md Angphes species (Statterthwait l93l, Bilboni I964, Anderson and Paschke I968, Dysart l97l, Ellis I973, Aeschlimmn I977). This pattern could be the result of high parasitoid mortality due to either inherent vulnerability to overwintering conditions or a scarcity of overwintering host eggs. Alternatively, the low early season parasitism rates may be the result of inadequate host-finding 85 ”1 "-“O-w-IDHMTmfilPMUSPKHDS A 60“ on . E ‘. O 5 ”a: >. I: G) :2 “I a 30-1 e’ \ :s-t \. O 5 l l 1 l l 7 21“ ”JUN SJUL 21M IAUG 13AUG DATE Figure 12. Field density of carrot weevil eggs and Patasson n. sp. immatures within host eggs in Clinton Co., Michigan in 1979. 86 abilities which are surmounted later in the season only by virtue of greatly increased parasitoid density. Sex Ratio Sixty-seven percent of the parasitoids reared from field-collected carrot weevil eggs during I979 were females, somewhat lower than the percent of female offspring produced in the laboratory by mated females (Section 3). This suggests that a significmt fraction of the ovipositing female parasitoids had not mated, which may inhibit rapid population growth. Based an equation 3.I (Sectim 3), l3.l% of the females parasitizing the sampled eggs were virgins. This estimate assumes that the sex ratio among offspring produced by mated females in the field was the same as that observed in the laboratory. The occurrence of virgins among this parasitoid population appears to be higher thm has been observed for other species. Kevm (I946) noted that reproduction by virgin Patasson n_i_t_en_s females was rare. Anderson md Paschke (I968) found evidence of ovipositim by virgin Anghes m females in only one of 87 parasitized eggs examined. Although not statistically different (X2 test, p>.05), the percent offemales increased from 59.7 among adults reared from eggs collected on 9 July, to 74.5 on l3 August. Thus, the estimated frequency of virgin females decreased from 22.9% to 4.l% (based an equation 3. l, Section 3). The sex ratio of parasitoids emerging from host eggs collected in the field on a given date reflects the mating status of the parental adults at some earlier date. Assuming that the parasitoid life cycle requires about two weeks in the field (Section 4), the sex ratio reported for parasitoids emerging from eggs 87 collected on l3 August would be a reflection of the mating status of parasitoid adults in the field during the preceeding two weeks. Therefore, the maximum percent of mated females in the field mpeared to have occurred around the time of maximum parasitoid density. This suggests a relationship between population density md the frequency of mating. This relationship could be useful in the derivation of population density estimates. Parasitism and Access to Host Egg Cavities The fecal plug with which the female carrot weevil covers the oviposition cavity had no effect on parasitism. Among eggs collected from 9 July to I3 August, parasitism of eggs from cavities which were covered with a fecal plug was 40.I%, while parasitism of eggs from unsealed cavities was 40.9%. The fecal plug thus appears to have no inhibitory effect on the parasitoid, nor does it appear to enhmce hast location. Parasitism and Host Egg Clutch Size Parasitoid host finding was not influenced by host egg clutch size. No significmt differences were found in the percent of clutches parasitized within clutch size categories (X2 test, p>.05). Thus, the probability that a clutch will be located by a parasitoid female is not affected by clutch size. Parasitoid oviposition behavior was also uninfluenced by clutch size. No differences in the observed degree of superparasitism were attributable to clutch size (X2 test, p>.05). Similarly, the percent of eggs parasitized within a clutch was unrelated to clutch size. Thus, once a clutch was located by a female parasitoid, her treatment of individual host eggs was unaffected by the number of eggs available. 88 The number of parasitoids which emerged per host egg remained nearly constmt, increasing from 2.l8 among eggs collected on 2l July to 2.33 on l3 August (not significant, X2 test, p>.05). Thus, the degree of superparasitism was relatively constant, even though relative parasitoid and host egg densities varied greatly (Figure I2). Parasitism and Per Plant Egg Density The number of available host eggs per carrot plant appeared to have no effect on parasitism. The percent of plants with one or more parasitized eggs was not related to host egg density per plmt (X2 test, p>.05). Similarly, egg density per plmt had no significant effect on the number of parasitoids which emerged per parasitized egg or the percent of available eggs parasitized among parasitized plants. Plant size also had no apparent effect on parasitoid host finding or oviposition behavior. No significant correlations were found between total plant fresh weight and the number or percent of parasitized host eggs (F tests, p>.05). These data indicate that the probability that a plant with host eggs will be found was not affected by egg density per plant, and that once a plant with host eggs was located, parasitoid oviposition behavior was not influenced by the number of eggs available. Therefore, the absolute number of parasitoids and parasitized eggs per plant was determined by the number of eggs available (Figure I3). Spatial Distribution The spatial distribution of parasitoids was examined by recording the number of adults that emerged from carrot weevil eggs collected within 89 s-I 00 O 8 .- Ill 8 s 7" '2 {D <( 18- I! '< a. s— as 2 4- 5 t: 00 3-‘ .( II '( a zq IL 0 o' 1'1 :2 l I I l I l I I 1 3 3 4 5 8 7 3 NUMBER OF EGGS PER PLANT a-e Figure 13. Effect of per plant host egg density on a) mean number of Patasson n. sp. (a) (y a 1.64 + 0.73 X, R . 0.718), and b) mean number of carrot weevil eggs parasitized per platt (O) (Y I 0.92 + 0.32 X, R - 0.685). 9O abitrarily designated 30 x 30 cm grids. The distribution of paasitoids appeared to be aggregated when all grids were considered, as well as when grids with one or more host eggs were cmsidered (Table II). However, this is primarily the result of the aggregated distribution of host eggs, as shown by the highly significmt correlations between the number of paasitized eggs md the number of available host eggs per grid. The distribution of parasitized eggs was more random at the end of the carrot weevil oviposition period when host egg density was low md approached a rmdom distribution itself. Parasitism rates were generally not dependent on host egg density (Table II). A significant correlation between the percent of eggs parasitized and egg density was found-on I August. However, this relationship explained only I0.2% of the observed variation in parasitism rates. 8.3 Discussion 8.3.l Carat Weevil Resource Utilization Carrot weevil oviposition was characterized by small clutch size md low egg density per plant. let status appeared to have no influence on oviposition. Egg populations were spatially aggregated, but this aggregation was not related to host plmt density. Density-dependent mortality (presumed to result primaily from cmnibalism) was shown to occur when different carrot weevil egg densities were experimentally imposed on carrot plmts. ln naturally occurring field populations, much of this density-dependent mortality may be avoided by low egg densities per plant. This pattern of oviposition may represent m adaptive strategy which originally evolved to efficiently exploit the weevil's naturally 91 .uoou m .~&.unm ua os can ma.alm an s an couauficcw oucaudmficmwm .o .Ufinm soap ca ammo poo: canauqa>a mo HunEac on» can ammo cuufiuaaouam mo acauuum on» cou3uon ucu«0wmwuoo sawuaaouuou .n .Uqu 50am ca ammo coufiuaoauam mo Hanson any can ammo poo; manaAfia>o mo Manes: 05¢ cua3uun ucuauwmmaou ceauoauuuoo .a moa.l s mem. av aoo.s mam.u mo «he.a mum.a 0:4 ma o Sam. so new. we ohm.¢ acm.a .mo mmm.m N¢Q.H mod A ans. oo one. am hav.e Nev.a co ama.m NhH.H Han AN oNH.I so amm. em cmh.~ «no.5 ac emm.a amh.& Hob a am Am 2 .uo> coax z .ua> coo: dump a a mafiamsam uoucaaofimmoou ammo one: :ueS npwuo madam Add c0auaduuuou caum Ham ammo puuwuaaauam mo Hones: coo: .mhma .ocwuu Eu amxan mcoaa .as .: cOaaauom an couwuqoouam ammo Hfi>au3 uouuau mo :ofiusnwuuaflp Hafiuamm .HA canoe 92 occurring host plmts. Wild host plants have a much smaller taproot, and unlike commercial plmt varieties, may be unable to support a large number of larvae. The observed field oviposition pattern may thus avoid much of the wasted reproductive energy expenditure which would be associated with greater egg densities per plmt. Low carat weevil egg densities may not always occur. Celery collected nea Hudsonville, Michigan, in July of I980 had m average of ”J eggs per plmt (s=l3.9, n=l4). Clutch sizes appeared to be compaable to those found on carrots. The greater egg density may thus be attributed to a lager number of clutches per plmt. The higher egg density on celery may be the result of differences between the two plmts. These differences could be based on qualitative suitability, or on qumtitative differences such as root size md extent of foliage. However, quantitative differences among carrot plmts were found to have no effect on egg densities. Alternatively, the observed egg density differences may be attributable to differences in the density of ovipositing female weevils. Celery is trmsplanted into the field in April in Michigm. Thus, suitable host plmts are usually available in the field when spring weevil activity begins. This may tend to limit dispersion, resulting in increased density. This appears to be the more probable explanation as there is no evidence that celery is utilized differently than carrots by simila densities of adult weevils. No evidence is available concerning the survival of larvae on celery. However, it is assumed that the density-dependent mortality manifested on carrots would function on celery as well. This implies that reproductive efficiency would decrease as the population density increased. 93 Carat weevil oviposition behavior has presumably evolved to efficiently utilize more spatially diffuse resources than ae encountered in a monoculture of commercial carats or celery. The likelihood of dense aggregations of adults following spring dispersion would appea to be diminished in a resource aray unaltered by man. 8.3.2 Paasitoid Resource Utilization No relationship was found between the parasitism rate md clutch size or host egg density per plant. Paasitism was also unaffected by plant size. The spatial distribution of paasitized carrot weevil eggs was aggregated, and appeared to be determined by the host egg distribution. Turnbull md Chant (l96l) observed that the relationship between the pest species md the affected crop greatly influences the effectiveness of a parasitoid species in preventing economic damage. A single late insta carat weevil lava is generally sufficient to render a carat plmt unmaketable. It has been demonstrated that only a few carat weevil lavae will develop per plmt regadless of the initial number of eggs. Therefore, virtually all of the carrot weevil eggs on a plmt must be paasitized to avoid economic injury. An effective parasitism rate may be defined as the percent of plants in which all carat weevil eggs ae paasitized, thereby preventing laval damage. The probability that a particula egg will be parasitized is unaffected by egg density per plmt, and is independent of the status of other eggs in the clutch. Thus, the effective paasitism rate (EPR) for a given nominal parasitism rate (i.e. the overall percent of eggs paasitized), and a given distribution of plants among weevil egg density categories cm be estimated by: 94 (EG8.I) EPR=§|(PioNPRi)/:§'Pi I: = where Pi is the number of plants in egg density category i, NPR is the nominal paasitism rate, md k is the number of density categories. The relationship between the nominal md effective paasitism rates cm be calculated for my carat weevil egg density distribution. Using the distribution for the entire period of observation during I979, the relationship would be as depicted as in Figure l4. At lower parasitoid population densities, the parasitoid is relatively ineffective at preventing damage to the carrot plants. At higher densities the effective rate will more closely approximate the nominal rate. During the entire I979 observation period, 22.8% of all carrot weevil eggs were paasitized. The effective paasitism rate during this period was only 8.5%. Estimates of effective paasitism rates were made based an observed carat weevil egg distributions for each sampling period (Table I2). These estimates conform closely to the observed percent of plants in which all available host eggs were paasitized. . The effective paasitism rate has been defined in terms of the percent of plants in which all eggs were parasitized, but it also applies to the survival of carat weevil lavae. As only one or occasionally two lavae will generally complete their development on a single plant, the effect of parasitism on the next generation of weevils will also be related to the number of plants in which all available eggs have been parasitized. Carrot weevil oviposition patterns are characterized by low per plant egg densities, which may represent m adaptation for optimizing the number of offspring per unit of reproductive effort. However, as paasitoid oviposition appeas to be distributed independently of host egg 95 100- m .. .I- 8° < a: 2 2’. I-_- so- a: < a: < a In .. 2 4o '— o “I u. u. I“ 20— - l l l I 1 ° 20 so so so 100 NOMINAL PARASITISM RATE Figure 14. Relationship between nominal and effective rates of parasitism of carrot weevil eggs on carrots by Patasson n. sp. for a given distribution of host eggs per plant. (Nominal rate a percent of eggs parasitized: effective rate a percent of plants in which all eggs are parasitized). 96 Table 12. Nominal and effective rates of parasitism by Patasson n. sp. in 1979. Effective parasitism rate Sampling Naminal a . date para. rt. Estimated Observed 28 Jun 9.35 3.34 --- 9 Jul 19.00 5.07 --e 21 Jul 39.41 13.76 14.92 1 Aug 49.32 21.92 18.37 13 Aug 29.63 8.38 14.71 a. Percent of carrot weevil eggs parasitized. b. Percent of infested carrot plants in which all carrot weevil eggs are parasitized. 97 density, lower egg densities per plant will increase the inhibitory effect of the paasitoid an the weevil population. Several additional factors may influence the effectiveness of a paasitoid A as m agent of biological control. The destruction of paasitized host eggs by newly hatched carat weevil lavae would reduce the parasitoid population growth rate. Barton and Stehr (I970) observed that newly hatched _Q_u_l£r_n_9_ melangms larvae do not interfere with eggs paasitized by Angghes Mg. However, newly hatched carat weevil lavae were observed to occasionally destroy other carrot weevil eggs in the laboratory. Presumably, paasitized eggs could also be destroyed. The frequency at which this phenomenon occurs is not known. Simmonds (I948) observed that the value of a paasitoid as a biological control agent was also influenced by the relative developmental times of the host md the paasitoid. At 23° C, the mem developmental time of Patasson n. sp. was found to be “.9 days (Section II). Simonet and Davenport (l98l) reported that carrot weevil eggs hatched in m average of 7.l days at 23.9° C. Using these data in the equation given by Simmonds (I948), the effective paasitism rate computed for the entire observation period (8.5%) .would be further reduced to only 5.I%. This reduction is lagely the result of the longer time spent inside the host egg by the developing parasitoid. This result suggests that paasitism by Patasson n. sp. represents a relatively minor factor in carrot weevil mortality. 98 9. Differential Pesticide Effects Control of carat weevi I infestations by chemical insecticides has been the focus of numerous investigations (Hagmann I938, Pepper I942, Wright I953, Semel I957, Wright md Decker I957, Whitcomb I965, Matel et al. l975a, Pepper md Ryser I975, Stevenson l976a). Pesticide usage is frequently assumed to inhibit the effect of paasitoid populations on their host species. The relative impact of pesticides on Patasson n. sp. and carat weevil populations must be understood in order to evaluate the role of the paasitoid as m agent of natural control within a commercial agricultural setting. It has been demonstrated that the egg, laval md pupal stages of Patasson n. sp. comprise mproximately three- fourths of its total lifespan (Section 4). Thus, most of the paasitoid's life is spent in a relatively insulated environment inside the host egg, which is in turn inside the plmt petiole. This may render a degree of protection to the paasitoid population from the effects of pesticide applications. Although the acblt population may be severely diminished, a residual population of immatures may remain relatively unaffected. It was also demonstrated that most paasitoid oviposition occurs within the first few days after the adults have emerged (Section 3). Paasitoids which have survived a particula pesticide mplication as immatures may be able to emerge and complete most or all of their potential oviposition during the relatively safe interval between applications. This possibility would be enhanced by the use of pesticides which break down quickly in the environment. The objective of this study is to examine the relative effects of several pesticides on the host md paasitoid populations. 99 9.l Methods 9. M Study Area This study was conducted during the I980 growing season in field #3 at the Hammond Fam in Clinton County, Michigm. This field was plmted in onions in I979, and carats (bring I980. Carrots were planted in four-row beds running north and south. Beds were I.2 m wide and were separated by ca. 40 cm. The study aea consisted of the two beds adjacent to the eastern field margin. The field magin was ca. I2 m wide and consisted of a strip of trees md weeds separating two simila fields. Carats had been grown in the field to the east of this magin (field #2) in I979, md the field was planted in bems in I980. The eastern field was known to have been infested with carat weevils in I979, 'md carat weevil eggs were paasitized by Patasson n. sp. Carats (cv. GT26) were planted in the study field on 3 May, except for the two eastern-most beds. Carrots (cv. Gold King) were planted in the two eastern beds on 9 May. Fertilizer was applied to this field prior to planting at a rate of 400 lb. per acre. Herbicides were regulaly applied to the entire field. 9.I.2 Pesticide Applications Five chemical insecticides were examined. These included one systemic insecticide (aldicab) applied in the row with the seed at planting, and four folia sprays: azinphosmethyl, oxamyl, fenvalerate, and diazinon (Table I3). A randomized block design was used. Each of the three blocks contained eight experimental plots, one for each of the five pesiticides and three control plots. Each plot was l0 m long and two beds wide (ca. 3 m). The entire study aea 100 Table 13. Application dates and rates applied for five chemical insecticides in field number three at the Hammond Farm, Clinton Co., Michigan 1980. Date 3 13 23 30 7 14 21 4 11 18 25 Treatment May Jun Jun Jun Jul Jul Jul Aug Aug Aug Aug Aldicarb (Temik 156) X 1.5 lb ai/A Oxamyl (Vydate L) X X X X X X X X X X 1.0 lb ai/A Fenvalerate (Pydrin 2.4 EC) X X X X X X X X X X 0.1 lb ai/A Diazinon (Diazinon assa) x x x x x x x x x x 0.5 lb ai/A Azinphosmethyl (Guthion 2 SC) X X 1.0 lb ai/A 101 consisted of the southern 240 m of the two eastern beds, divided into 24 contiguous experimental plots. Beginning on I8 June, the northern and southern- most one meter of each experimental plot were kept clea of all vegetation. This was done to prevent pesticide drift or overlap between plots and to inhibit movement of weevils between plots. 9.I.3 Adult Carat Weevil Trapping The initial carat weevil adult population was assessed by trapping adults in the weeds along the eastern field magin. The objective of this procedure was to correlate early spring trap catch data with subsequent field infestation levels. Ja traps baited with carat baby food were used, as described by Ryser (I975) md modified by Grafius md Otto (I979). Two traps were placed in the weeds adjacent to each experimental plot on I7 May, and were checked on 22 May. The number of adult carat weevils captured in each trap was recorded, the adults were released at the point of their capture, md then the traps were removed. This trapping was done at about the time of initial carat plmt emergence, which was first observed in the field on l6 May, md in the experimental plots on 2l May. Therefore, the weevil distribution was assumed to be uninfluenced by the carrot plmts at the time of this sampling. 9.I.4 Carat Plant Sampling Adult carrot weevil activity within the study aea was monitored periodically tl'a'oughout the growing season. This activity was measured indirectly by the examination of host plants. Several plant sampling techniques were used at different times during the growing season, and are outlined below. 102 Destructive Sampling The most reliable method of estimating carrot weevil activity is by collecting plmts in the field md bringing them into the laboratory for closer examination. However, this tecl'nique completely destroys the plants, so that. the lager the sample size, the more the processes being estimated ae disrupted. The destructive sampling technique was employed on three occasions. A preliminay sampling was undertaken on 3 June when the carrot plants were just reaching the three-leaf stage. Approximately five plants were collected at one meter intervals from the center of the second row of the maginal bed in each experimental plot. The fresh weight of each plmt was recorded md the plants were examined for evidence of carat weevil activity. The data collected for each plmt included the number of feeding or oviposition punctures, the number of egg clutches, and the number of larvae. The number of hatched md unhatched eggs was noted for each clutch. The position of each lava was recorded as in the root or in the foliage. The laval insta was determined for each lava by measuring the head capsule width. Only plants with eggs (hatched or unhatched) or larvae were included in the analysis. lets having punctures believed to be made by weevils but which were without eggs or lavae were not considered to be affected by the weevils. These punctures were assumed to be the result of adult feeding which produces no appreciable damage. Unhatched carrot weevil eggs were extracted from the plant petioles md placed individually in 65 ml clear plastic containers. These eggs were reared in an environmental chamber at a constant 26° C with a I6:8 photoperiod (photophose beginning at 0600 h). 103 A second set of samples was taken on the day preceding the initial application of the folia sprays (I2 June). The sampling procedure was the same as that described above, except that five plants were collected from each of the eight rows in the experimental plots. The third set of samples was taken approximately one month after the initial folia spray applications. Since the outside rows were partially damaged by the sprayer tires, samples (seven plmts each) were collected only from the inner six rows of each plot. Patially Destructive Sampling An alternative sampling technique was employed on two occasions (3 July and I August). These samplings were undertaken to provide additional information concerning paasitism rates while causing a minimum of damage to the carat plmts. This procedure involved the visual inspection in the field of all plants within one row of each plot. Whenever a stem was located which appeaed to have m oviposition hole, it was removed from the plant. No more than one stem was taken from any plant. Thus, the number of stems collected within a plot would be an index of the number of plants showing evidence of carat weevil activity. The stems were then taken to the laboratory and carrot weevil eggs were extracted. These eggs were reaed and information recorded on viability md paasitism. Final Carat Weevil Damage Assessment A final assessment of the extent of weevil activity was made at havest (I3 September). This date was about one month past the time at which weevil 104 oviposition ended. Therefore, no information concerning paasitism was obtained. All of the plmts in the middle six meters of each plot were pulled and the roots were inspected in the field for evidence of damage by carat weevil lavae. The number of plmts with and without appaent weevil damage was recorded. A late season encroachment of weeds into some of the experimental plots produced considerable vaiation in the number of plmts in each plot. In addition, nematode (Meloidggme M) damage, and what appeaed to be damage caused by excessive water associated with m unusual amount of fall rain, resulted in a relatively high plmt loss. However, damage from causes believed to be unrelated to carat weevil lavae were not considered in this study. 9.2 Results 9.2.I Carat Weevil Infestation Only 2.l% (3 of I4I) of the carat plants sampled on 3 June had evidence of carrot weevil activity (eggs or larvae). Mem plant fresh weight on this date was 0.I0 g (5:0.06). Individual plant weights ranged from 0.0I to 0.34 g. The carat plmts were appaently too small to support a significmt carrot weevil infestation at this time. The mem percent of plmts sampled on l2 June which were affected by carrot weevils was 6.6 (Table I4). Simificant differences were shown among the treatment mems (ANOVA, p<.005). The at-plmting systemic insecticide (aldicarb) treatment was significantly different from each of the other treatments. No other treatment differences were found. This is consistent with expectations as only aldicab had been applied by this sampling date. Aldicab would thus appear to inhibit weevil activities early in the season. 105 Table 14. Percent of carrot plants with carrot weevil eggs or larvae as affected by insecticide treatment and sampling date. Mean percent of carrot plants with carrot weevil eggs or 1arvae°b Treatment 12 June 11 July 23 Sept. Azinphosmethyl 7.4 a 7.8 b 23.6 b Diazinon 6.6 a 28.3 ab 21.7 b Aldicarb 0.7 b 18.4 b 24.1 b Oxamyl 6.8 a 14.3 b 12.0 b Fenvalerate 3.7 a 16.6 b 13.2 b Controls 9.2 a 41.9 a 41.5 a a. Data were transformed by X'= arcsinVX prior to statistical analysis. The three control plots in.each block were considered as subsamples of a single treatment. The significance of differences in treatment means was tested using the pooled sampling and experimental error mean squares divided by the pooled degrees of freedom. b. In the same column, treatment means followed by the same letter are not significantly different at p-0.05, Student-Neuman-Kuel multiple range test. 106 No relationship was found between the number of adults captured prior to plmting in baited ja traps in the weedy aea adjacent to the experimental plots md the level of infestation recorded within the associated experimental plots an l2 June (F tests of correlation coefficients, p>.05). Relative adult trap catches early in the season thus may not be a useful measure with which to predict the location md severity of subsequent plmt damage within the crop site. The mean percent of plants affected by weevil activity (those with carrot weevil eggs or lavae) in the control treatments among plmts sampled on I l July was significmtly different from the means for experimental plots in which azinphosmethyl, oxamyl, fenvalerate, and aldicab were applied (SNK test, p<.05) (Table I4). Thus, these insecticides generally appeaed to result in the reduction of weevil oviposition among treated plants. I The mean number of plants encountered an 3 July with carat weevil eggs (hatched or unhatched) among the stems selected based on a visual inspection of plmts while still in the field was significantly lower in the azinphosmethyl treatment than in the controls (SNK test, p<.005). However, this sampling failed to detect differences between the control mem and the means in the plots treated with oxamyl, fenvalerate, and aldicab which were shown by the I I July destructive sampling. Therefore, the validity of this method can neither be confirmed nor refuted since the results were intermediate between the results of the I2 June and II July destructive samplings. However, this method has the advmtage of producing less damage to the plmts, md requires only a fraction of the time required for a complete examination of the entire plant. This technique may be appropriate for some applications where large aeas need to be sampled and time is a critical factor. 167 The mem number of infested carrot plmts in the control plots sampled on I August was significmtly different than the mems of plots treated with fenvalerate and oxamyl (SNK test, p<.05). The validity of the patially destructive sampling procedure for estimating damage is questionable on this later sampling date. Only the number of plants encountered which had hatched or unhatched eggs were detected. However, the rate of weevil oviposition is diminished this late in the season. Detection of oviposition is also more difficult on lager plmts. The technique overlooked any larvae located inside the plmts, which could be expected to represent a large portion of the immature weevil population. The percent of plmts damaged by carat weevil lavae at havest differed significmtly depending on treatment (Table I4). The control treatment mem was significmtly different thm all other mems (SNK test, p<.05). None of the meals for plots treated with chemical insecticides were different from one mother. All of the chemicals appeaed to have inhibited destructive weevil activities, md all appeaed to be approximately equal in effectiveness. 9.2.2 Paasitism Paasitism of carat weevil eggs by Patasson n. sp. was first detected on 3 June, and continued as long as suitable host eggs were available.) No significant differences between paasitism rates in different treatments were found on my sampling date (destructive or non-destructive sampling technique) (ANOVA, p) .05). Paasitism rates were thus relatively uniform among treatments, even though carat weevil egg densities were significmtly higher in the control treatments than in some of the plots in which chemical insecticides were applied. 108 9.3 Discussion The insecticides examined have been demonstrated to significmtly reduce adult carat weevil activities which bring about economic injury to carrot plmts. The systemic insecticide was more effective ealy in the season. As the season progressed, the folia treatments were perhaps more effective. No statistically significmt differences were observed between paasitism rates among the vaious treatments on my sampling date. The percent of weevil eggs paasitized on each date is illustrated in Figure I5, wherein data from the four folia treatments ae pooled. These results imply that Patasson n. sp. may be able to function sucessfully in a typical commercial agro-ecosystem in which the use of chemical pesticides is prevalent. The protected environment in which the paasitoid undergoes its development and the relatively short life cycle appea to enable the paasitoid to maintain a uniform impact on the carat weevil population under vaying conditions. 109 so- ’.o I I I I I I I I so- ' ,5, I A“ \.'o I o A 1 Ill 40" J .... E! 4! T“. I: '-. g /’ .. .0 s / a. 30- ’/ l- a :2 ’d//' In (J 1' E 26- / a, .l a" ...o... CONTROL PLOTS -—-a—— FOLIAR INSECTICIDE PLOTS :od ......... SYSTEMIC INSECTICIDE PLOTS c T r I ‘1 I 3JUN IzJUN sum. nJUL IAua DATE Figure 15. Mean percent of carrot weevil eggs parasitized by Patasson n. sp. in a) plots treated at planting with a systemic insecticide (aldicarb), b) plots treated with foliar applications of insecticides (azinphosmethyl, fenvalerate, diazinon, or oxamyl), and c) untreated controls, Clinton Co., Michigan, 1980. 110 I0. Summary and Conclusions In I979, a new mymaid egg paasitoid of the carat weevil was discovered in a commercial carrot field near East Lmsing, Michigm. Identified as Patasson sp. nea sordidatus, this paasitoid has since been found at several other locations in Michgm, and in Ohio. Studies were conducted to examine the biology md life cycle of the parasitoid md to evaluate its effectiveness as an agent of biological control. The major findings of these investigations are briefly summaized below. Up to six paasitoids were found to emerge per field-collected host egg. - Most emerged from eggs in which two or three paasitoids developed. The degree of superpaasitism was found to affect the size of the adults as well as adult longevity. Adult longevity was also found to be influenced by temperature. Mem longevity increased from 48.8 h at 29° C to I53.5 h at I7° C. However, this relationship may overstate the effect of temperature on longevity in the field to the extent that the parasitoid is able to change the micro-environmental conditions in which it finds itself by moving from one location to mother. More importmt thm mem longevity is the shape of the survivorship curve. Survivorship was found to be high during the first few days following the emergence of the parasitoid from the. host egg. More than 75% of adults kept at 26° C were still alive after three days. After this time mortality increased rapidly. This is importmt because most parasitoid oviposition is accomplished within a few days of adult emergence regadless of how long the female ultimately lives. Paasitoid oviposition begm within 4.6 h after emergence from the host egg at 23°. More than 50% of all oviposition occurred within two days at 23° C, 111 although some ovipositim occurred up to eight days after adult emergence. The mem number of offspring per female paasitoid was 49.4. Fecundity was not affected by the number of paasitoids that developed in the host egg from which the female emerged, or by whether or not the female mated. Virgin females produced all male offspring, while mated females produced 77.7% female offspring at 23° C. This method of reproduction provides a built-in mechanism for maintaining a stable sex ratio. This is appaently importmt to the paasitoid population since there is an additional mechmism by which a stable sex ratio is maintained in the form of a delay in the initiation of ovipositim by unmated females of about a day. In the field, such a delay would presumably increase the probability that the female will mate, md thus increase the likelihood of producing female offspring. Mean parasitoid developmental time was ll.9 days at 23° (2. Since about one half of all oviposition occurs within two days after emergence, one complete generatim would be completed in about I4 days at this temperature. Therefore, the paasitoid may be able to complete approximately five generations during the period of carat weevil oviposition. Developmental time was strongly influenced by temperature. However, while the mem developmental time increased from 9.0 to I7.6 days as temperature decreased from 29 to I7° C, the fraction of the total life cycle spent as immature stages was relatively constmt (7|.8, 73.4, md 77.4% at I7, 23, and 29° C, respectively). Thus, most of the paasitoid's life cycle is spent in a relatively insulated and stable environment inside the host egg, which is in turn inside the plmt petiole. This may provide some degree of protection from external mortality factors such as predation, dessication, and perhaps most 112 importmtly, from the effects of chemical pesticide applications. The impact of several pesticides (fenvalerate, oxamyl, aldicab, diazinan, and azinphosmethyl) on parasitism was examined in I980. Paasitism rates were not significmtly different in experimental plots treated with the pesticides than in untreated control plots. The paasitoid thus appears to be able to function successfully in a typical commercial agro-ecosystem chaacterized by frequent pesticide applications. . I Emergence periodicity was found to be influenced by bath temperature md photoperiod. Most emergence occurred during the first two hours of the photophose from host eggs reaed under a photoperiod of LD I6:8. This emergence peak was more pronounced at lower temperatures. Emergence appeared to be controlled entirely by exogenous environmental cues. During the first two hours of the photophose, males appeaed to emerge in advmce of the females, which may facilitate their location of conspecific females. Several components of male courtship and mating behavior were identified, including a general excitation phase, a wing fmning display, mtennation of the female's abdomen, mounting, md copulation. The females exhibited no overt courtship behavior. Paasitoid mating readily occurred within two hours of emergence from the host egg. Although this may tend to limit genetic vaiability within a field population due to an increased likelihood of sibling matings, this disadvmtage is appaently compensated for by m increased probability of mating, hence on increased stability of the sex ratio. Males readily mated more than once in laboratory conditions, but multiple matings by females were not observed. Since female parasitoids were able to 113 produce female offspring for up to six days after mating,'repeated matings by females appea unnecessay. Carat weevil and Patasson n. sp. field populations appeared to be spatially synchronous. Paasitism rates were relatively uniform on a given sampling date in different aeas of m infested field even though host egg density diminished with distmce from the field magin. This suggests that paasitoid dispersal capabilities ae adequate to keep pace with movements by the carat weevil population. This may be importmt where host and non-host plmts ae rotated annually, causing a shift in the location of host eggs. Overwintered carrot weevil adults first became active in the field in April or May, depending on enviromental conditims. In I980, carat weevil oviposition was first observed on 9 May on weeds in aeas adjacent to a previously infested carat field. Oviposition on carats was not observed until 4 June. In I980, adult paasitoids were active in the field by mid-May but paasitism of carat weevil eggs was not observed until ealy June. Once begun, parasitism continued throughout the period of carrot weevil oviposition, which ended in mid—August. Paasitism rates ealy in the weevil oviposition period were low. Adult paasitoids remained active in the field as late as 8 November in I979. This was more than two months after the end of the carrot weevil oviposition period and suggests the possible involvement of an alternative overwintering host. Carat weevil laval densities were low in a commercial carat field studied in I979. A mem of only I.46 late instars (third and fourth) was observed per infested plant. Possible mechanisms that could bring about this distribution were examined by mmually trmsferring different numbers of carrot weevil eggs 114 to uninfested carat plmts md recording larval survival. The survival rate of these weevils was found to decrease as the initial egg density per plmt increased. This implies the operation of some density-dependent mortality factor in the field (which is unrelated to paasitism by Patasson n. sp.). Cmnibalism has been observed in the laboratory and could account for this density-dependent mortality. In the field, much of this density-dependent mortality is avoided by an oviposition pattern chaacterized by low egg densities per plmt. Mem clutch size observed in I979 was I.99, and the mem number of eggs per carat plmt was 2.87. No density-dependent association was found between host plmt density and carat weevil egg density. Egg density was uninfluenced by plmt size. This pattern of oviposition may represent an adaptive strategy which has evolved to efficiently exploit naturally occurring weed host plmts. Unlike commercial carat vaieties, these wild host plmts have relatively small root systems which ae not sufficient to support a large number of lavae. The observed oviposition pattern thus avoids much of the wasted reproductive energy expenditure which would be associated with a larger number of eggs per plmt. The probability that a female paasitoid would find a host egg clutch or a plmt with host eggs was uninfluenced by clutch size or the number of eggs per plmt. No density-dependent association was found between paasitism rates and host egg density on either a per-plmt or per-unitoaea basis. Therefore, host seaching by Patasson n. sp. appeaed to be rmdom. Among host egg clutches or plants with host eggs in which one or more eggs were paasitized, the percent of eggs paasitized and the number of paasitoids which emerged per host egg were unaffected by either clutch size or 115 egg density per plant. Therefore, there was a linear relationship between the number of available host eggs and both the number of parasitized eggs and the number of paasitoids produced per plmt. Similar relationships were observed on a per-clutch or per-unit-aea basis. In I979, the nominal paasitism rate (i.e., the overall percent of eggs paasitized) reached a peak of 50.2%. The seasonal mean was 22.8%. However, even in the absence of paasitism, only one or occasionally two carat weevil lavae per plmt successfully complete their development. Therefore, to affect the next generation of carat weevils, the paasitoid must paasitize all of the eggs on a plmt. Since one late insta carat weevil is sufficient to render a carat plmt unfit for maket, all of the eggs on a plmt would also have to be parasitized to prevent economic plmt damage. Therefore, an effective paasitism rate was defined as the percent of plmts on which all available carat weevil eggs were paasitized. For the observed distribution of eggs per plmt, the estimated effective paasitism rate for the I979 growing season was only 8.5%. The amount of time spent by the paasitoid in the host egg relative to the time it takes for the host carat weevil lavae to hatch also affects the interpretation of observed nominal paasitism rates. Using the equation given by Simmonds (I948) that accounts for this relationship, the actual effective paasitism rate was only 5.l% in I979. The effect of Patasson n. sp. on carat weevil populations was examined for two consecutive years in commercial carrot fields with moderate weevil infestations. Observed egg densities per plmt were low. Much higher carat weevil egg densities have been observed in Michigan on infested celery plmts (E. 116 Grafius, Michigm State University, personal communication). Since no density- dependent association between host egg density and the rate of paasitism has been observed, the paasitoid would be even less effective as a control agent at higher egg densities per plmt. Although Patasson n. sp. does not appear to be a major mortality factor affecting carrot weevil populations, it may nonetheless play a more significant role in regulating carat weevil populations than has been "suggested. The findings of the studies presented in this report, ae based on observations made on crop sites under conditions of high carrot weevil egg densities. Under these conditions, low paasitism rates were observed to occur ealy in the weevil's oviposition period. A possible explanation to account for these low rates could be that the paasitoid population is limited by the availability of eggs of an alternative overwintering host. However, the carat weevil is widely distributed at very low population densities among weeds adjacent to field magins. Where the carat weevil population density is very low, the limiting factor on the paasitoid population may be the scacity of carrot weevil eggs during .the growing season relative to the availability of overwintering host eggs. Therefore, Patasson n. sp. may have a significmtly greater role in regulating carat weevil populations at low densities, and may contribute to the periodic nature of carat weevil population outbreaks. An alternative explanation for the low parasitism rates which have been observed to occur early in the carat weevil's oviposition period could involve an inherent susceptibility to severe abiotic environmental conditions such as cold temperatures. Overwintering conditions may differentially affect the weevil md paasitoid populations, and these effects may vary from yea to year. Studies 117 examining host-paasitoid relationships under a variety of conditions ae thus warmted. Overwintering appeas to represent a weak link in the life history of Patasson n. sp. To fully evaluate the role of this paasitoid as an agent. of carat - weevil population control, the mode of overwintering must be more completely understood, and is thus a logical focus of future reseach efforts. In addition, 5 studies examining relationships between Patasson n. sp. and the carat weevil at "low population densities among the weevil's wild host plmts could provide information which could ultimately be exploited to enhmce the paasitoid's effectiveness through mmipulation of conditions outside of the crop site. LITERATURE CITED Aeschlimmn, J. P. I977. Notes on Patasson lameerei (Hymenoptera: Mymaidae), an egg paasitoid of _S_i_t_om spp. (Coleoptera: Curculionidae) in the Mediterrmem region. Entomophaga 22:l I I-I4. Ahmad, R. I979. Note on breeding the agentine stem weevil, erodes bonaiensis (Coleoptera: Curculionidae) and its egg parasite Patasson atomaius (Hymenoptera: Mymaidae). Entomophaga 23:I6 l-62. Anderson, R. C. and J. D. Paschke. I968. The biology and ecology of Angphes Mg (Hymenoptera: Mymaidae) an exotic egg paasite of the cereal leaf beetle. Ann. Ent. Soc. Amer. 6|:I-5. . I969. Additional observations on the biology of Anophes Mm (Hymenoptera: Mymaidae), with special reference to the effects of temperature and superpaasitism on development. Ann. Ent. Soc. Amer. 62: l3l6-2l. . I970a. A biological evaluation of five Europem cultures of Angphes 119112231 (Hymenoptera: Mymaidae), an egg paasite of Qngm—a melangggs (Col.: Chrysomelidae). Entomophaga I5:l07-20. . I970b. Factors affecting the post release dispersal of Anaghes M22; (Hymenoptera: Mymaidae) with notes on its past release development, efficiency, and emergence. Ann. Ent. Soc. Amer. 63:820-28. Balduf, W. V. I928. Observations on the buffalo tree hopper, Gigi gu_b_a_l_u_§_ Fabr. (Homoptera: Membracidae), and the bionomics of an egg paasite, Polygema striaticorne Girault (Hymenoptera: Mymaridae). Ann. Ent. Soc. Amer. 2I:4I9-35. 118 119 Baton, L. C. and F. W. Stehr. I970. Normal development of Anaghes flavipes in cereal leaf beetle eggs killed with X-radiation, and potential field use. J. Econ. Ent. 63:I28-39. Bibolini, C. I964. Sulla biologia della Chrysomela americma L. (Coleoptera: Chrysomelidae). Frustula Ent. 6:I-I22 (Abstract). . I970. Nata preliminae sul paassita oofago Patasson chrysomelae Bkdf. (Hym.-Mymaidae) seriazione delle vittime attaccate lungo il litorale Iigure—toscano. Frustula Ent. _: l—8. (Abstract). Biever, K. D. I972. Effect of temperature on the rate of seach by Trichogramma and its potential application in field releases. Env. Ent. I:l94—97. Boyce, A. M. I927. A study of the biology of the pasley stalk-weevil, Listronotus latiusculus Bohemm, (Coleoptera: Curculionidae). J. Econ. Ent. 20:8l4-2 I. Chamberlin, T. R. I924. Studies on the paasites of the alfalfa weevil in Europe. J. Econ. Ent. I7:623-32. Chmdler, S. C. I926. The economic importance of the carat weevil in Illinois. J. Econ. Ent. I9:490-94. Chittenden, F. H. I909. The pasley stalk weevil. USDA Bull. 82:I4-I9. . I924. The parsley stalk weevil, Listronotus latiusculus Bohemm, a potential pest. Bull Brooklyn Ent. Soc. l9:84-86. Clak, A. F. l93l. The paasite control of Gonipterus scutellatus Gyll. New Zealmd J. Sci. and Tech. l3:22-28. Clausen, C. P. I940. Entomophagous Insects. McGraw-Hill Book Co., N.Y. 688 pp- 120 Collins, R. D. and E. Grafius. I982. Literature review of the carat weevil and paasitoids in the mymaid genera Patasson and Anophes. Mich. State Univ. Pest Mmagement Technical Report (in press). Dowell, R. V. and D. J. Horn. I975. Mating behavior of Bathyplectes curculionis (Hymenoptera: lchneummidae) a paasitoid of the alfalfa weevil, EIZE‘EEE Mg (Coleoptera: Curculionidae). Entomophaga 20: 27I-73. . I Dysart, R. J. I97 I. Distribution of Anophes Meg in Europe and sources of its importation into the United States. Entomophaga I6:445-52. El-Dessouke, S. A. and W. Stein. I970. Interspecific competition between lavae ” F of _Si_tofla_ spp. (Coleoptera: Curculionidae). Oecologia 6:I06-08. Ellis, C. R. I973. Paasitism of M M eggs at Guelph, Ontaio, by Patasson Lamond M rugosifrons. J. Econ. Ent. 66:I059-6I. Fisher, T. w., E. l. Schlinger, and R. Van Den Bosch. I96I. Biological notes on five recently imported paasites of the Egyptim alfalfa weevil, m brunneipennis. J. Econ. Ent. 54:l96-97. F lmders, S. E. I947. Elements of host discovery exemplified by paasitic Hymenoptera. Ecology 28: 299-309. Force, D. C. I967. Genetics in colonization of natural enemies for biological I control. Ann. Ent. Soc. Amer. 60:722-29. Gage, 5. H. and D. L. Haynes. I975. Emergence under natural and mmipulated conditions of Tetrastichus ju_lis, an introduced laval paasite of the cereal leaf beetle, with reference to regional population management. Env. Ent. 4:425-34. Grafius, E. and M. A. Otto. I979. Detection and control of carrot weevil. Mich. State Univ. Exten. Bull. Ee890. 2 pp. 121 Harris, H. M. I926. A new carrot pest, with notes on its life history. J. Econ. Ent. I9:494-97. Henderson, L. 5. I940. A revision of the genus Listronotus (Curculionidae- Coleoptero). Univ. Kans. Sci. Bull. 263288-97. Hendricks, D. E. I967. Effect of wind. on dispersal of Trichogramma W ’semifumatum. J. Econ. Ent. 60:I367-73. Jackson, D. J. l96l. Observations on the biology of Caraphractus M Walker (Hymenoptera: Mymaridae), a parasitoid of the eggs of Dytiscidoe (Coleoptera). Parasitology SI: 269-294. JuiIIet, J. A. I960. Some factors influencing the flight activity of hymenopteraus parasites. Cm. J. Zool. 38:I057-6 I. Kevan, D. K. M. I946. The eucalyptus weevil in East Africa. E. African Agric. J. I2:40-44. Leibee, G. L., B. C. Pass, and K. V. Yeargan. I979. Developmental rates of Patasson lameerei (Hymenoptera: Mymaridae) and the effect of host egg age on parasitism. Entomophaga 24:345-55. MacGiIl, E. J. I934. On the biology of mm (L.) HoI., an egg parasite of the Ieafhopper Erythroneura palidifrons Edwards. Parasitology 26:57-63. Mailloux, G. and J. G. Pilon. I970. Patasson 12$. (Girault) (Hymenoptera: Mymaridae) and Bathyplectes curculionis (Thomson) (Hymenoptera: lchneumonidae), two parasites of m M (GyIIenhaI) (Coleoptera: Curculionidae) in Quebec. Cm. J. Zool. 48:607-08. Maltby, H. L., F. W. Stehr, R. C. Anderson, G. A. Mooreheod, L. C. Barton, and J. D. Paschke. I97I. Establishment in the United States of Anophes m an egg parasite of the cereal leaf beetle. J. Econ. Ent. 64:693-97. 122 Martel, P., C. R. Harris, and H. J. Svec. I975a. Toxicological studies an the carrot weevil, Listronotus orggonensis (Coleoptera: Curculionidae). Can. Ent. l07:47 l-7S. . I976. The life history of the carrot weevil, Listronotus orgganensis (Coleoptera: Curculionidae) under controlled conditions. Can. Ent. I08. 93I-34. Mathews, R. W. I975. Courtship in parasitic wasps. In Price, P. W. (Ed.). I975. Evolutionary Strategies of Parasitic Insects and Mites. Plenum Press, New York & London. Miller, M. C. I966. Emergence md mating of Tetrastichus incertus, a parasite of the alfalfa weevil. J. Econ. Ent. 59:I532-33. Mossop, M. C. I929. Notes on a mymarid parasite of the eucalyptus snout- beetle (Gonipterus scutellatus, Gyll.), and its introduction into South Africa. Dept. Agric. Un. S. Africa Sci. BuII. 8|:I-I9. Niemczyk, H. D. and J. K. Flessel. I970. Population dynamics of alfalfa weevil eggs in Ohio. J. Econ. Ent. 63:242-47. Otto, M. A. I978. Carrot weevil, a management view in Michigan. Unpublished paper. 9 pp. Pepper, B. B. I942. The carrot weevil, Listronotus latiusculus (Bahe), in New Jersey and its control. N.J. Agric. Exp. Stn. Bull. 693:I-20. Pepper, B. B. and L. E. Hogmann. I938. The carrot weevil, Listronotus latiusculus (Boh.), a new pest on celery. J. Econ. Ent. 3l:262-66. Ryser, B. W. I975. Investigations regarding the biology and control of the carrot weevil, Listronotus orggonensis (LeConte), in New Jersey. M.S. Thesis, Rutgers Univ. 123 Saunders, D. 5. I976. Insect Clocks. Pergamon Press Ltd., London. 279 pp. Semel, M. I957. Control of carrot weevil attacking parsley. J. Econ. Ent. 50: I83-84. Simmonds, F. J. I948. Some difficulties in determining by mems of field samples the true value of parasitic control. BuII. Ent. Res. 39:435-40. . I963. Genetics md biological control. Cm. Ent. 95:56I-67. Simonet, D. E., and B. L. Davenport. l98l. Temperature requirements for development and oviposition of the carrot weevil. Submitted to Ann. Ent. Soc. Amer. Statterthwait, A. F. l93I. Anaphoidea calendrae Gahan, a mymarid parasite of eggs of weevils of the genus Calendra. J. New York Ent. Soc. 39:I7 I-90. Stevenson, A. B. I976a. Carrot rust fly: chemical control of first generation larvae in orgmic sail in Ontario. J. Econ. Ent. 69:282-84. . I976b. Seasonalhistory of the carrot weevil, Listronotus orggonensis (Coleoptera: Curculionidae) in the Holland Marsh, Ontario. Proc. Ent. Soc. Ontario I07:7 I-78. Stoner, A. and D. E. Surber. l97I. Development of Anaphes oviientatus, an egg parasite of L192; hesperus, in relation to temperature. J. Econ. Ent. 64: l566-67. Streams, F. A. and R. W. Fuester. I966. Patasson lgn_a in overwintering eggs of the alfalfa weevil. J. Econ. Ent. 59: 33I-33. Turnbull, A. L. and D. A. Chant. I96l. The practice md theory of biological control in Canada. Can. J. Zool. 39:697-753. Vidano, C., A. Arzone, and F. Meotto. I979. Multiplicoziane e disseminazione in Italia di Patasson i_1_i_t_§_n_s_ per la difesa dein eucalipti da Gonipterus scutellatus. Annoli deII'Accademio di Agricoltura di Torino I2I:I-I5 (Abstract). 124 Whitcomb, W. D. I965. The carrot weevil in Massachusetts. Biology and connrol. Univ. Moss. Agric. Exp. Stn. Bull. 550. 30 pp. Williams, J. R., L. A. Mautia, and P. R. Hermelin. l95I. The biological control of Gonipterus scutellatus Gyll. (Coleoptera: Curculionidae) in Mauritius. Bull. Ent. Res. 42:23-28. Wright, J. M. I953. An investigation into the biology and control of the carrot weevil, Listronotus organensis (Leconte). Ph.D. Thesis, Univ. of Illinois. Wright, J. M. and G. C. Decker. I957. Insecticidal control of the carrot weevil in canning carrots. J. Econ. Ent. 50:797. . I958. Laboratory studies of the life cycle of the carrot weevil. J. Econ. Ent. 5I:37-39. m.“ A" “"9 umn. VH3 ""6 SHS EH4 vuo mus UHO E" l IIIIWIIIHIIIII , 3129