.-. THE EFFECT OF DIETHYLCARBAMAZINE AND ' LEVAMISOLE ON THE DEVELOPMENT OF DIROFILARIA IMMITIS TN AEDES TR|SERlATUS Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY EIDI KASKA ' 977 .6 4_0. VON‘ an? o '55 01“ ' .a. ‘40' _ ' I” ' ‘ I “a. . ’ . a” l - .1. . o-fl. . . I '> n D 1 Jet 4-dg-I . I u‘. . JV.” . _ I m AIN‘- onfl ' -o M' 4 . ~r- an. I . '-W"‘ ”D I. «‘4‘ .v»- » -v~¢ ' . 1-.- 0 ¢ ' . ' ‘ ' ' ,f‘. - ' C '. d.’ a . .u fi.v.>-<’—’-‘~-’.’p “‘~ _.‘,. ' - . . a -- ~ . mm wvdwo ---l° J . ‘ - '9‘ w . ~ - . 00-4 "n“. .- ._-' v, ’fl‘al' . ‘ " a > - ,. .,.'. g - ,- ,'. .. Ql' . . - - , . . -ktr,vronn." ':.A, ‘ . , . . I . . _ ‘r‘ ~~~ nap . I -"~ ' a”, ‘ - : . . I o-lr- ' f rota . ‘ ‘ « ‘ o v: e r V' . l o u T . a . 01"- ‘ . ‘ .I , . . ‘ uf‘ ' ‘ A 3 . . . . ' ' ' ‘ ‘ ' ' (lasa- o‘:r . ‘ r o c ,.....»O- u- . - . -_. t rot (nu-1H r '0 0 v p . ' 0 f1’; 4"- T , o .. . . ‘. ' o -" or ‘ e _ - - -_-._..-. “5—...”‘Hl‘oanMW' . THE EFFECT OF D THE DEVEL gm investigatioj :55, diethylcarbamz aiming stages 0f Q Mosquito m 3.1 tag. immitis micrc Stead blood meal cor teals were given mosq Echique. Dissectic ‘F'Ps‘oximately two wee EveTOpmental stages ABSTRACT THE EFFECT OF DIETHYLCARBAMAZINE AND LEVAMISOLE ON THE DEVELOPMENT OF DIROFILAELA IMMITIS IN AEDES TRISERIATUS By Heidi Kaska An investigation was conducted on the effect of two drugs, diethylcarbamazine (DEC) and levamisole. on the de- 've10ping stages of Dirofilaria immitis (dog heartworm) in 'the mosquito Agggg triseriatus. Mosquitoes were infected ‘with Q, immitis microfilariae and one week later given a second blood meal containing the test drugs. Both blood .meals were given mosquitoes using an artificial feeding technique. Dissections of infected mosquitoes took place approximately two weeks after infection and numbers and deve10pmental stages of larvae were noted. Analyses of data indicated that both drugs produced statistically significant results, with certain treatment groups showing as much as a 44 percent reduction in the number of larvae develOping. A closer examination of the data revealed a bimodal distribution of larval numbers in dissections performed over time and fewer numbers of larvae in those mosquitoes in which 2, immitis develOpment was at 'mid-stage. Definite modes of action for these drugs were not established during this research but the results sug- gested that DEC may slow down larval develOpment and Heidi Kaska levamisole may kill or inhibit early develOpmental stages of the larvae. THE EFFECT OF THE DEVE . Mi 1“ partial De 1 THE EFFECT OF DIETHYLCARBAMAZINE AND LEVAMISOLE ON THE DEVELOPMENT OF DIROFILARIA IMMITIS IN AEDES TRISERIATUS By Heidi Kaska A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Entomology 197? To Ed ii I am deeply 8- '.eson, for his 133' Etf throughout my 23.} ways during m; ersity and was a1T isaperson for whc perscnal admiratior films for his he iiing many of the fife a busy schedul imld also like 1: Tiraduate Committ “ Donald C. ares 42f; contribute d ‘t 0 aEEEStions . ACKNOWLEDGEMENTS I am deeply grateful to my advisor, Dr. Harold D. Newson, for his patience, kindness, and constant encourage- ‘ment throughout my research. Dr. Newson supported me in many ways during my graduate studies at Michigan State Uni- versity and was always available with help and advice. He is a person for whom I have the highest professional and personal admiration. I am also grateful to Dr. Jeffrey F. Williams for his helpful comments on my work and for pro- viding many of the supplies needed for my research. Des- pite a busy schedule, he always took time to assist me. I would also like to thank sincerely the other members of 'my Graduate Committee for their guidance and friendship: Drs. Donald C. Cress, Gary R. H00per, and Richard W. Merritt. Each contributed to my work by his beneficial criticisms and suggestions. Many others provided considerable assistance during my research and deserve special credit: Ms. Gail Peel and Ms. Jean Beemsterboer, laboratory technicians at the Small Animal Clinic at Michigan State University. who cheerfully responded to my frequent requests for blood samples; Dr. Charles Cress of Michigan State University and Mr. Michael Bower of the Statistical Institute at Texas A&M University iii (5.: provided stati flexes MM UTIiVG £53; Dr. Woodbrid sElied the Ohio trig of the Unive; Eamd Alabama st: fdhints on mosqu: Edniversity of 1‘ Eu membrane in t} E: of these persc I would eSpec j 333E my stay at M iirewarding 9Xper “1131511988 to teac Tiling mosquitoes research, and for h reeded mosquitoes; :elpful suggestions Mina and for hi 515% w StUdent at Mi. 3: and fol, the ‘ Ma dagme . who provided statistical assistance; Dr. Norman 0. Dronen of Texas A&M University who generously provided computer time: Dr. Woodbridge Foster of the Ohio State University who supplied the Ohio strain of Agggs triseriatus; Dr. George B. Craig of the University of Notre Dame who provided the Wal- ton and Alabama strains of A. triseriatus and gave me help- ful hints on mosquito rearing; and Dr. Paul R. Grimstad of the University of Notre Dame who suggested using the Natura- Lamb membrane in the artificial feeding of mosquitoes. To each of these persons I extend sincere thanks. I would especially like to thank my friends for their making my stay at Michigan State University such a pleasant and rewarding experience: Mr. Mori Zaim for his unfailing willingness to teach me the many things I did not know about handling mosquitoes, for his thoughtful comments about my research, and for his generously providing me at times with needed mosquitoes: and Dr. Henry B. Lewandowski for his many helpful suggestions. for his teaching me much about mosquito rearing. and for his making me feel so welcome as a new gra- duate student at Michigan State University. Most of all, I would like to thank my husband Ed for all the love and support he has given me over these past two years. and for the sacrifices he has made helping me obtain this degree. iv 2550? TABLES . . ° ISIOF FIGURES . . 3‘0? APPENDICES . TEQDUCTION . . . IBETURE REVIEW Distribution of Taxonomy and Bi Other Hosts . . Diagnosis and '1 Pharmacology oi acology 0! Artificial Feec‘ FBI-103$ AND MATERIAI Proposed Resea: osqulto Rearir TABLE OF CONTENTS LIST OF TABLES O O O O O O O O 0 LIST OF FIGURES . . . . . . . . LIST OF APPENDICES . . . . . . . INTRODUCTION 0 O O O O O O O O 0 LITERATURE REVIEW Distribution of the Disease Taxonomy and Biology of the Other Hosts . . . . . . . . Diagnosis and Treatment . . Pharmacology of Diethylcarbamazine Pharmacology of Levamisole Artificial Feeding Technique METHODS AND MATERIALS Proposed Research . . . . . Mosquito Rearingt. . Parasite Aedes aegypt11(ROCK strain) . . Aedes triseriatus (Michigan strain) Aedes triserzLatus (Alabama strain) Aedes triser::atus (Walton strain) Aedes triserr atus (Ohio strain) Artificial Feed1ng Apparatus Membrane Preparation and Artificial Feed Setup . . . . . . Sorting the Mosquitoes . . Blood Source and Drug Concentrations Drug Concentrations and Preparation . Dissection Techniques . . . Experimental Procedures . . RESULTS Background Information Experiments Experiment 1 . . . . . Experiment 2 . . . . . Experiment 3.. . . Experiments 4 and 5 . Experiments 6 and 7 . neooooflooooo ouggoomoooooooo Page viii xi xiii T1 Experim EXperim Summary Experi Drug-Feeding EXperim Summary Emlysis of Phase 0 Phase AA AAA AAA goowoxamteawmo—I TABLE OF CONTENTS (continued) Experiment 8 . . . . . . . . . . . . Experiment 9 . . Summary of Background Information Experiments . . . . . . . . . . . . Drug-Feeding Experiments . . . . . . . . Experiments 10 through 16 . . . . . Summary of Drug-Feeding Experiments Analysis of Experimental Results . . . Phase One Analysis . . . . . . . . (1) Analysis of variance . . (2) Orthogonal analysis . . . (3) Non— —orthogonal analysis . Phase Two Analysis . . . . . . . (4) Frequency distributions . (5) Correlation coefficients and means I O C O O O O O C O (6) "DevelOpmental variances" (7) Power curves . . . . . . . (8) Bimodality in mosquito dissections . . . . . . . . . DISCUSSION Results of Background Information Experiments . . . . . . . . . . . . . . Feeding Rates . . . . . . . . . . Mosquito Mortalit after Infection Effect of Microfi aremia Level on Mosquito Survival . . . Susceptibility of Mosquito Strains D. immitis . . . . . . Results of Drug-Feeding Experiments . . Preliminary Considerations . . . Experimental Results . . . Results of Statistical Analyses . Hypothesis A--Real asynchrony Hypothesis B--Apparent asynchr Explanation of Experimental Results . . . . . Alternate Modes of Drug Action . . . to on SUMMARY AND CONCLUSIONS Overall Statements . . . . Implications of Experimental Results . . Suggestions for Further Study . . . . . . vi y Page 53 55 57 58 58 64 64 64 69 69 72 72 73 79 84 86 86 87 88 88 89 92 92 9# 102 10h 106 107 EETEDIX A . . - EEZJDIX B . . . EPE‘JDIX C . . . LEEDIX D . . LIST OF REFERENC] TABLE OF CONTENTS (continued) Page APPENDIX A . . . . . . . . . . . . . . . . . . . . . . . 109 APPENDIX B . . . . . . . . . . . . . . . . . . . . . . . 131 APPENDIX C . . . . . . . . . . . . . . . . . . . . . . . 149 APPENDIX D . . . . . . . . . . . . . . . . . . . . . . . 152 LIST OF REFERENCES . . . . . . . . . . . . . . . . . . . 154 vii nob-U l. Feeding rates 1‘ h. x-- in Experiment 1 Mortality rates fed upon dog bl concentrations Feeding rates 0 and}. triseria perimentj . ' Wage number -' lmmitis larv‘ straTn) ' War Of late mm? larvae in strain) from Ex Bram) from E mfiier of late “\lg larvae in Palm) frOm E “Tiber of '- late $§ 1 39 in rain) rom Ex LIST OF TABLES Table Page 1. Feeding rates for A. aegypti (ROCK strain) in Experiment 1 o o o o o o o o o o o o o o o o o 0 1+5 2. Mortality rates of A. aegypti (ROCK strain) fed upon dog blood containing three different concentrations of Q. immitis microfilariae . . . . 46 3. Feeding rates of A. aegypti (ROCK strain) and A. triseriatus (Michigan strain) in Ex- periment 3 I I I I I I I I I I I I I I I I I I I I #8 4. Average number of late second and third stage 2. immitis larvae per A. triseriatus (Michigan Straln) I I I I I I I I I I I I I I I I I I I I I I 50 5. Number of late second and third stage 2. im- mitis larvae in A. triseri t (Michigan strain) from Experiment 4 . . . . , . . . . . . . . 52 6. Number of late second.and.third st e D. im- mitig larvae in A. triseriatus (Mic igan strain) from Experiment 5 . . . . . . . . . . . . . 52 7. Number of late second and third stage D. Am: miti larvae in A. triseriatu (Michigan stra ) from Experiment 5 . . . . . . . . . . . . . 54 8. Number of late second and third stage 2. Am: 't' larvae in A. triseriatus (Michigan stra1n) from EXperiment 7 . . . . . . . . . . . . . 54 9. Average number of late second and third stage 2. ' it's larvae per A. tzégezigtgs (Alabama strain from Experiment . . . . . . . . 55 10. Average number of late second and third stage D. immitis larvae per A. t eria s in EXPeriment 9 I I I I I I I I I I I I I I I I I I 56 11. Number of late second and third stage 2. im- mitis larvae in A. triseriatus (Alabama Strain) from Experiment 10 o o o o o o o o o o o o 59 viii I \_ - . . \l‘ :- . . ‘gr -. - , . a.) - .3519 ;2 Number of la“- miti larvae Brain) from Number of lai‘ m_it_i_§ larvae strain) from :4 Number of 1a1 mitis larvae strain) from Number of lat m larvae strain) from Nlpnber of lat m larvae strain) from ”Ember 01' lat Eli—94 larvae Strain) from smary chart third stage D 16' grouped a LIST OF TABLES (continued) Table Page 12. Number of late second and third stage 9. Am: m'ti larvae in A..triseriatus (Alabama strain) from Exper1ment II . . . . . . . . . . . 59 13. Number of late second and third stage 2. im— mitis larvae in A. triseriatus (Alabama stra1n) from Experiment 12 . . . . . . . . . . . . 6O 14. Number of late second and third stage D. im- mitis larvae in A. triseriatus (Alabama strain) from Experiment 13 . . . . . . . . . . . . 6O 15. Number of late second and third stage 2. im- mitis larvae in A. triseriatus (Alabama strain) from Experiment 14 . . . . . . . . . . . . 61 16. Number of late second and third stage D. im- mitis larvae in A. triseriatgg (Alabama strain) from Experiment 15 . . . . . . . . . . . . 61 17. Number of late second and third stage D. im- mitis larvae in A. triseriatus (Alabama strain) from Experiment 16 . . . . . . . . . . . . 62 18. Summary chart of number of late second and third stage 2, ' itis in A, trisefigtus Alabama stra1n rom xperumen s rough l6, grouped according to dissection day . . . . . . 62 19. Summary chart of means of late second and third stage D. immiti in A. tri eriatus (Alabama strain) from experiments in which dissections were performed on day 16 after infection, plus overall means for each treatment grouP . . . . . . . . . . . . . . . . . . 63 20. Mathematical model for analysis of variance of a randomized block design experiment with replicates in subclasses . . . . . . . . . . 21. Final analysis of variance of the results from Experiments 10, 11, 12, 14, and 15 . . . . . . 68 22. Orthogonal analysis (seven contrasts) per- formed on the results of Experiments 10, 11! 12! 11+! and 15 o o o o o o o o o o o o o 0 ix LI: latle u ’1 5" '3:— t Results of nor contrasts) pe: periments 10. Number of obs: group; ercen‘ were 0 and 1' development 0: Summary of av. and third sta, in total body dissections o Alabama stra correlation c from individu; comparisons o; almes (ADV) w Winning (EV. ML: develop] 3111a strain) tal Varirmces Constant Valu f” thOse tre greater than ETGSBion from LIST OF TABLES (continued) Table Page 23. Results of non-orthogonal analysis (eight contrasts) performed on the results of Ex- periments 10, ll, 12, 14, and 15 . . . . . . . . . 72 24. Number of observations for each treatment group; ercentage of observations which were 0%pand 100% developed; and total % develOpment of that treatment group . . , , . , , 75 25. Summary of average number of late second and third stage D. immiti larvae found in total body, head7thorax, and abdomen dissections of infected A. triseriatus (Alabama strain) mosquitoes along with correlation coefficients (r) calculated from individual data . . . . , , , , , , , . . . . 77 26. Comparisons of actual developmental vari- ances (ADV) with estimated variances at the beginning (EVB) and middle (EVM) of D. Em: 'mitlg development in A. triseriatus (Ala- bama strain) and with estimated developmen- tal variances (EDV) . . . . . . . . . . . , , , , , 80 27. Constant values for the power curve (y==axb) for those treatment groups with r-values greater than the r-values for linear re- gression from Table 25 . . . . . . . . . . . . . . 82 28. .Mean number and percentage of 2. immitig larvae in abdominal and head/thoracic re— gions of A. triseriatus for Experiments 10, llilzilniandlsoooooooooooooooolOJ. II {151‘ e I L. Apparatus for art quitoes . . , , . r o 9 Artificial feedin mosquitoes in thi 3‘ Needing cup for U mg apparatus shc 0v91‘2111 means frc ”hand 15 of lat .- 1mm1t1s larvae a strain) for add controls x. Overall means f r< y'fim} 15 0f la' ' l S arva. 5m S‘fl‘aini for soups (fl) and 01 Observations fro] “up Cont Percent r01 .2’ 3-:— as LIST OF FIGURES Figure Page 1. Apparatus for artificial feeding of mos- quitoes . . . . . . . . . . . . . . . . . . . . . . 21 2. Artificial feeding apparatus used to feed mosquitoes in this research . . . . . . . . . . . . 29 3. Feeding cup for use in the artificial feed- ing apparatus shown in Figure 2 . . . . . . . 4. Overall means from Experiments 10, 11, 12, 14, and 15 of late second and third stage 2. immitis larvae in A. triseriatus (Ala- bama strain) for DEC treatment groups (%) and controls . . . . . . . . . . . . . . . . . . . 65 5. Overall means from Experiments 10, 11, 12, 14, and 15 of late second and third stage D. ' ' ° arvae in A, tri§¥r1g%u§ (Ala- 'Bama 3 rain for levamiso e rea ent I I groups (fl) and controls . , , , 66 6. Observations from Experiment 11, Treatment Group Control 2, distributed according to percentage 2. immitis develOpment in A. 3;;- seriatus (Alabama strain) . . . . . . . . . . . . . 74 7. Experimental data from Experiment 11, Con— _0 51 trol Group 2 fitted to the equation y= 1.69x ' with O-values from the data substituted with OIl-values o o o o o o o o o I o o o o o o o o o o 83 8. Average number of D. immitis larvae in head/ thorax, abdomen, and total body of infected A. triseriatus (Michigan strain) dissected overtlm'e..................... 85 9. Average number of Q. immitis infective larvae ;per individual mosquito . , . . , . , , , . . . . . 91 xi t Figure '1 .L 12. Hypothetical larvae with: in which a l is showing e ment of the Hypothetical larvae withi in which a I in the thora Hypothetical and third 31' seriatus g1. mental rate: HYPOthetica' and third 3 ser'atus sh me n treatment g laMal deve Hmothetice andfihird e ir‘fi 81 gr: hum er 11p 1 Velleiiegtw LIST OF FIGURES (continued) Figure Page 10. Hypothetical distribution of D. immitis larvae within a total mosquito popuIatlon in which a large portion of the pOpulation is showing either early or late develop- mentof'thelar’vae................93 ll. Hypothetical distribution of Q. immitis larvae within a total mosquito population in which a portion of the larvae present in the thorax is missing from observations . . . . 95 12. Hypothetical distribution of late second and third stage D. immitis larvae in A. t i- geriatus given that DEC slows down develOp- mental rates of the larvae . . . . . . . . . . . . 97 13. Hypothetical distribution of late second and third stage D. immitis larvae in A. tri- seriatus showing how early dissection af- fects the number of larvae from each DEC treatment group, given that DEC slows down laflaldevelopment oooooooooooooooo 98 14. Hypothetical distribution of late second and third stage D. ° m1ti§ larvae in A. _21- er tu showing how late dissection affects gfie num%er of larvae from each DEC treatment group. given that DEC slows down larval de- velopment . . . . . . . . . . . . . . . . . . . ... 99 xii :;;-:r.iix i'lu Number of la immitis larv ms Michige Number of la immitis lam mange Number of 1; gig lam Lu§ Michig; Number 0f 12 'tis lar- fififi-fiChig: LIST OF APPENDICES Appendix Page A—l. Number of late second and third stage D. immitis larvae in individual A. triseria- tus (Michigan strain) from Experiment 4 . . . . . 110 A-2. Number of late second and third stage D. immitis larvae in individual A. triseria- tug (Michigan strain) from Experiment 5 . . . . . lll .A-3. Number of late second and third stage D. Ammitis larvae in individual A. triseria- tus Michigan strain) from Experiment 6 . . . . . 113 A-4. Number of late second and third stage D. 1mm1%1§ larvae in individual A. triseria- tus Michigan strain) from Experiment 7 . . . . . 115 A-5. Number of late second and third stage D. immiti larvae in individual A. t iseria- tus (KEabama strain) from Experiment IO . . . . . 117 A-6. Number of late second and third stage D. imm1ti§ larvae in individual A. triseria- Ag§_ Alabama strain) from Experiment 11 . . . . . 118 A97. Number of late second and third stage D. immiti larvae in individual A. triseria— tus (Alabama strain) from Experiment 12 . . . . . 122 A-8. Number of late second and third stage l_3_. ' it' larvae in individual A. trisegia- tug (Alabama strain) from Experiment 13 . . . . . 124 A-9. Number of late second and third stage D. ' iti larvae in individual A. triseria- ggg (Alabama strain) from Experiment 14 . . . . . 126 A-lO. Number of late second and third stage D. immitis larvae in individual A. triseria- tgg (Alabama strain) from Experiment 15 . . . . . 128 xiii LI 5'. £;:er.dix 9.1 I : a a) . Number of ' itis l. tus (Alab; Number of ”t' 1: strain r! meal (Con 8 and is , dissectio: Number of m itis 1 only the from EXpe according Number of lmmit' 1 strain) 1‘ ing to 10 ”“1981‘ of lmmtis 1 m f mg to lo Amnber of lmmitis 1 .Str‘ain f mg to 1c Number of it's 1 $%tyi ing to It Ember 01 JJumjftis gums } ng t0 lc LIST OF APPENDICES (continued) Appendix A-ll. B-2. B-3. B-4. B-5. B-6. 3-8. Number of late second and third stage D. i i is larvae in individual A. triseria- tus (Alabama strain) from Experiment 16 . . Number of late second and third stage D. immiti? larvae in A. triseriatus (Alabama strain receiving only the infective blood meal (Control 2). Data is from Experiment 8 and is grouped according to location of diBSeCtion I I I I I I I I I I I I I I I I Number of late second and third stage D. immiilg larvae in A. triseriatus receiving only the infective blood meal. Data is from Experiment 9 and larvae are grouped according to location of dissection . . . . Number of late second and third stage D. " it' larvae in A. triseriatus (Alabama strain) from Experiment 10, grouped accord- ing to location of dissection . . . . . . . Number of late second and third stage D. ' itis larvae in A. triseriatus (Alabama strain) from Experiment 11. grouped accord- ing to location of dissection . . . . . . . Number of late second and third stage D. immitis larvae in A. triseriatus (Alabama strain) from Experiment 12, grouped accord- ing to location of dissection . . . . . . . Number of late second and third stage 2. immiti? larvae in A. triseriatus (Alabama strain from Experiment 13. grouped accord- ing to location of dissection . . . . . . . Number of late second and third stage D. immitis larvae in A. tr'ser atus (Alabama strain) from Experiment 1 , grouped accord- ing to location of dissection . . . . . . . Number of late second and third stage 2. immitis larvae in A. triseriatus (Alabama strain) from Experiment 15. grouped accord- ing to location of dissection . . . . . . . xiv Page I 130 o 131 . 132 . 134 . 136 . 140 . 142 . 144 . 146 LIST OF upendix 5-9. Number of 1at1 )1. immiti larva strain from. ing to locat1 Analysis of v Experiments 1 all the data ‘- . Orthogonal ar formed on the 11! 129 11+, 2 eluding zeros Calculations sis performe. 10: 11, 12’ LIST OF APPENDICES (continued) Appendix Page B-9. C-1. D-l. Number of late second and third stage D. immiti larvae in A. triseriatus (Alabama strain) from Experiment 16, grouped accord- ing to location of dissection . . . . . . . . . . 148 Analysis of variance of the results from Experiments 10, ll, 12, 14, and 15 u31ng all the data (including zeros) . . . . . . . . . 150 Orthogonal analysis (seven contrasts) per- formed on the results of Experiments 10, ll, 12, 14, and 15 using all the data (in— cluding zeros) . . . . . . . . . . . . . . . . . 151 Calculations for the non-orthogonal analy- sis performed on the results of Experiments 10, ll, 12. 14, and 15. shown in Table 23 . . . . 152 Dirofilari ofthe domestic netiate host. fond in the ri ing. producing ease. If left ie't-ilitation o: 35 10g heartwo lactic drugs 't 533' ’50 Preven Although effect of thee INTRODUCTION Dirofilaria immitis (Leidy) is a filarial parasite of the domestic dog for which the mosquito serves as inter- mediate host. In its mature state. the parasite can be found in the right ventricle and pulmonary artery of the dog. producing a condition known as canine heartworm dis- ease. If left untreated. this disease may result in severe debilitation or death of the animal. Because of this threat of dog heartworm disease, many dog owners administer prophy- lactic drugs to their dogs daily during the "mosquito sea- son" to prevent development of the parasite in their dogs. Although considerable research has been done on the effect of these drugs on the development of the heartworm in the dog. almost nothing is known about their possible effects on developing microfilariae in an already-infected ‘mosquito. The developmental period for Q. immitis in the mosquito is approximately two weeks in temperate climates and during this interval the mosquito may take another blood meal, conceivably one which could contain small amounts of these prophylactic drugs. If certain compounds 'were found to retard or prevent development of Q. immifiis in.the mosquito, the secondary effect of reducing the in- cidence of the parasite in amosquito pOpulation may be :fcmsiderable attic Purposes {the drug on is an importam ‘lo answer affect of prop guito populati :2“ two dmgs' the developme] mus 0n msquitoes we no test drug mined for Both the ini tantaining I artif is '1: tus thesis of considerable worth. Thus in choosing a drug for pr0phy- lactic purposes in the dog, one might find that the effect of the drug on developing microfilariae in the mosquito would be an important consideration. To answer some of these questions about the potential effect of pr0phylactic drugs on parasite levels within a mos- quito population, an investigation was made into the effect of two drugs, diethylcarbamazine (DEC) and levamisole, on the development of D. 1mmit1s in the mosquito Agggg :21- seriatus. One week after infection with Q. immitis, test 'mosquitoes were given a blood meal containing one of the two test drugs. These mosquitoes were later dissected and examined for numbers and development of Q. immitis larvae. Both the initial infective blood meal and the later drug- containing blood meal were provided to the mosquitoes using, an artificial feeding device. The results are reported in this thesis. Q. immitis talent in regic Far East. the ] ”bets of mosqui' Pical climates ease. however, Eparted to be 331th America the more temp In the t Eminent alc "Nito, 1969a isnse, Lind labile, Alab it)” r"‘vport Populations diSCusSe C1 b‘ reported fr f-. (mallenst e .1 3‘ 'I LITERATURE REVIEW Distribution of the Disease 49. immitis has a worldwide distribution but is most pre- valent in regions with more tropical climates such as the Far East, the Pacific, and equatorial Africa, where high num- bers of mosquitoes persist yearround. Regions with less tro- pical climates are by no means free from dog heartworm dis- ease, however. and infection rates among dogs have also been reported to be high in certain portions of South America, North America, Australia, and northern Africa, as well as in the more temperate parts of EurOpe (Soulsby, 1965). In the united States the disease is. expectedly, most prevalent along the southern Atlantic and Gulf coasts (Otto, 1969a), where mosquito populations are particularly dense. Lindsey (1961) reported that among pound dogs in Mobile, Alabama, 42 percent had circulating microfilariae. Other reports indicating that up to 63 percent of local dog populations had circulating Q. immitis microfilariae were discussed by Otto (1972). Numerous cases have also been reported from the Middle Atlantic and New England states (wallenstein & Tibola. 1960; Rothstein et al.. 1961; Tritch et a1.. 1973). where mosquito populations are also high but Inore seasonal, and heartworm disease has also been reported in Ontario, Canada (Otto. 1969a). 3 The Preval. if: along the firm mn_c0aste ages in areas anthem “lme isolated 01‘ 1‘ tough cases seas of the fitto. 1969a: the lississij resulted fro Essever, rec :;:e heartw< states. Al' dense of th Zaubach (18 in-state dq The prevalence of the disease inland is usually lower than along the coast. and high rates of infection reported from non-coastal regions (e.g., 30 to 40 percent infection rates in areas of northern and southern Illinois and in southern Minnesota; Otto, 1969a) have usually occurred in isolated or localized areas of high mosquito density. Al- though cases of heartworm can be found in virtually all areas of the United States east of the Mississippi River (Otto. 1969a), in the past cases in those states west of the Mississippi River have been infrequent and have usually resulted from infected dogs being brought into those areas. However, recent reports indicate that the incidence of ca- nine heartworm disease is on the rise in the arid western states. A118 and Greve (1974) reported a 6.5 percent inci- dence of the disease in Iowa and, in Oklahoma, Kocan and Laubach (1976) reported adult worms in 7.3 percent of the in-state dogs sampled. Tagonomy and Biology 0; the Parasite The filarial nematode Dirofilaria immitis was first described by Gruby and Delfond in 1843 and named by Leidy in 1856. Fulleborn described the life cycle of this para- site in 1908 and identified the mosquito as the intermediate host, necessary for completion of the develOpmental cycle of the nematode (Bradley. 1971). Other workers described de- 've10ping microfilariae in dog and cat fleas thus giving sup- ;port to the theory that possibly several haemophagous mopods were r (Breinlt :itls ’- ::' “10% heartworm ;arasite of (10% as W atzut multiple flog and cat i" :1. but not ‘l‘he adu aisle and a and finite 1: rd the adL :arcus, wi fithin the filth has ’fi‘ - .. why, "4‘ J 0 W5 :iseharg if, \ ‘7351 I arthropods were responsible for the transmission of Q. immitis (Breinl, 1920). This hypothesis of multiple vectors of dog heartworm continued to receive acceptance until 1956 when Newton and Wright (1956) described another filarial parasite of dogs in the United States belonging to the ge— nus Qipetalonema. This discovery clarified the confusion about multiple vectors of 2. immitis and established that dog and cat fleas were apparently vectors of Dipetg1gnema spp. but not of Q. immitis (Bradley, 1971). The adult worms of Q. immitis, found in the right ven- tricle and adjacent blood vessels of the heart, are slender and white in color. The adult females measure 25 to 30 cm and the adult males measure 12 to 18 cm. The female is vivi- parous, with fertilized eggs developing into active embryos within the uterus of the female. The vitelline membrane, ‘which has the appearance of a sheath. is stretched over the embryo within the uterus, but is shed before the embryos are discharged by the female into the host's bloodstream (Soulsby, 1965). Microfilariae are colorless and transparent, and range in size from 307 to 322 lam with a width of 6.7 to 7.1/4m (Newton.& Wright, 1956). (Dipetalonemg microfilariae are annaller in size.) [2. immitis microfilariae lack a sheath, :as was mentioned above. and unlike Dipetalonema spp. in which the tail is hooked, Q. immitis microfilariae possess straight tails (Newton & Wright, 1956). 12. immitis micro- :filariae possess no intestine or esophagus, but when the drofilariae are E the interior of the 3:: staining are : :retory cell, varit Mich are used . :c‘uiorofilariae ( Although 12. 1 Food more or less are 24-hour per :icrofilariae beir. mam period (01 guts that the pat :olocale. E.g., 219141?) observed a at 1630 hours and 3951), in France 112000 hours and as made by Hawki Nlated to oxyser. During the C Herefilariae Ci] $01)) reported M: . :n m the Six term-10h the)I 3qu flI‘St 24 ou- 2.1 a ' L mo of the N" ‘microfilariae are stained a column of nuclei is visible in the interior of the microfilaria. Other features visible upon staining are a nerve ring, an excretory pore, an ex- cretory cell, various genital cells, and an anal pore, all of which are used in differentiation of the various species of microfilariae (Taylor, 1960a). Althougth. immitis microfilariae are present in the blood more or less continuously, their numbers fluctuate over a 24-hour period with five to fifty times as many ‘microfilariae being found at the maximum period as at the ‘minimum period (Otto, 1969b). Soulsby (1965) discussed re- ports that the pattern of this periodicity varies according to locale. E.g., in the United States, Schnelle and Young (1944) observed a maximum number of circulating microfilariae at 1630 hours and a minimum at 1100 hours. Euzeby and Laine (1951), in France, found a maximum number of microfilariae at 2000 hours and a minimum at 0800 hours. The suggestion was made by Hawking (1956, 1967) that this periodicity is related to oxygen tension. During the course of a blood meal, the mosquito ingests microfilariae circulating in the blood of the dog. Taylor (1960b) reported that in Agggg aegypti these microfilariae remain in the stomach of the mosquito for the first 24 hours, after which they migrate to the Malpighian tubules. Other authors have reported that this movement can occur within the first 24 hours. Further development occurs at the dis- tal end of the Malpighian tubules and for the first week it larvae can first two days {and circula4 Lately 250 to seer, the lam stage larva I first moult mi the secc 7'3?) and in‘l 3'? infectio] 5.? 2‘3 fan W1 after the s tutu1‘38 in‘l 081.133 thrt “cad, Wile 5:439 laru the larvae can be found inside the cells of the tubules after which they develop within the lumen of the tubules. For the first two days larvae appear very much like microfilariae found circulating in the blood of the dog and measure approxi- 'mately 250 to 300/am by 8 to lO/Mm. By the fourth day, how- ever, the larvae shorten and thicken to form a "sausage" stage larva measuring 220 to 240/4m by 20 to 25 flm. The first moult occurs at about the eighth day of development and the second stage larvae possess certain elementary excre- tory and intestinal cells. By approximately the tenth day of infection, the third stage larva develops, measuring 500 by ZO/am with the principle organs already formed. Shortly after the second moult the larvae work their way from the tubules into the body cavity of the mosquito. Movement then occurs through the thorax and into the cephalic spaces of the head. When found in the head, this third or "infective" stage larva may measure 900/qm.(Taylor. l960a.b; Soulsby, 1965). The infective larvae can remain in the head for several days although many of them may be lost from the head (Ho et al.. 1974). (It has been suggested that this loss may occur during the act of obtaining a sucrose meal.) The mosquito may react to the developing larvae by allowing complete de- velopment of the parasites, succumbing to the parasites due 'to injury inflicted during the migration of the larvae into the Malpighian tubules or into the body cavity, or, as re- ;ported by Brug (1932) and Kartman (1956), reacting geologically gage of develop velopment withou Infective l the dog) while tozaotually "ir. ’ the mouthpe all droplet oi the dog. The 15 the dog and intc me made by the 1353a;NcGreevy e Nmoults in t] isvelops in the times (Kline & Nierany occur :1965) at wh 1”" Followi l’fiavfi physiologically by encapsulating the larvae during an early stage of develOpment (Soulsby, 1965). Lack of larval de- velopment without encapsulation can also occur. Infective larvae are passed to the definitive host (the dog) while the mosquito feeds. The infective larva is not actually "injected" into the animal but, rather, escapes from the mouthparts of the mosquito and may be found in a small droplet of liquid which is deposited on the skin of the dog. The larva then works its way through the skin of the dog and into the bloodstream, generally through the punc- ture made by the mosquito at the site of feeding (Kartman, 1953agMCGreevy et al., 1974). The infective larva undergoes two moults in the dog (Orihel, 1961) during which time it develops in the submuscular membranes and subcutaneous tissues (Kume & Itagaki, 1955). Migration to the heart generally occurs two to three months after infection (Souls- by, 1965) at which time the worms vary in size from 3.2 to 11 cm. Following arrival in the right ventricle of the heart. the worms reach maturity after which the females re- lease microfilariae into the blood. Circulating microfilariae are usually not found for at least eight months after the initial inoculation. Although the longevity of the adult worms varies, some adult females reportedly have lived and produced microfilariae for more than five years (Otto, 1969a). Over 60 species of mosquitoes have been shown in the laboratory to be capable of supporting the developing stages of _I_)_. immitis, though there is considerable variation among EstEPtible mo ismlepment Of rah support ’ a "fat may occur 3 g immitis is W 1121.1 genera. more limitec seals) and wh: the develOpmel 21‘ larger mos susceptible mosquito species in the length of time needed for deve10pment of the parasite, the number of larvae which each can support, and the degree of development of the parasite that may occur. At identical temperatures the development of Q. immitis occurs somewhat more rapidly in mosquitoes of the AnOQheles genus than in mosquitoes of either Agggg or 931g; genera. Species of smaller mosquitoes which possess a more limited stomach capacity (and ingest smaller blood :meals) and whose Malpighian tubules are small, may support the deve10pment of fewer numbers of larvae than the species of larger mosquitoes (Soulsby, 1965). Other Haste Although canine heartworm disease is primarily a dis- ease of dogs, 9. immitis has been shown to occur in other carnivores such as foxes, wolves, coyotes, and cats (Coffin, 1944; Erickson, 1944; Otto, 1972; Sharp, 1974), and, more rarely, in non-carnivores such as beavers (Foil & Orihel, 1975) and marine animals such as seals. (Ninety adult worms were found in the heart of a male harbor seal in California; Medway & Wieland, 1975.) Although infection also may occur in humans, it is generally felt that complete development of the nematode cannot occur in man (Center for Disease Control, 1974). In all, more than 40 human cases of dirofilariasis have been reported (Otto, 1972), and presence of the worm is nearly always discovered inadvertantly. such as on a routine chest x-ray, where immature worms are detected in the lung 01‘ pulmor lesion (Navarrete. smh‘man cases : 2119118974) rep my symptoms d Ziezotherapy clea As one would ricrofilariae we] sored to reach 1 til“. in a human . i'aapatient wit Front infection Dog heartW imination of. : zoiified Knott' :ter microfil ar mix-rays may Neckson, 1969a .zlariae provid to diseaSe (m; 58"} ‘ 9“ Conga 31 m lI‘Eatment the ' meetion . < d. . a prop] afferent Stag. 10 the lung or pulmonary artery in the form of a small "coin" lesion (Navarrete, 1972; Robinson et al., 1974). Although such human cases are usually asymptomatic, Feldman and jHolden (1974) reported a case of a female patient with pul- monary symptoms diagnosed as having presumptive D. immitis. Chemotherapy cleared up the symptoms. As one would expect in the cases described, circulating Inicrofilariae were not demonstrated as the worms never ap- peared to reach sexual maturity. One recent case of infec- tion in a human with circulating microfilariae was reported in a patient with lupus erythematosus, however, but the ap- parent infection cleared up without treatment (Green, 1975). Diaggosis and Treatment Dog heartworm disease is most frequently detected by examination of a blood sample from the dog using either the modified Knott's or a filtration technique to determine whe- ther microfilariae are present. Additionally, heartsounds and x-rays may also be used in the diagnosis of the disease (Jackson, 1969a,b) because the number of circulating micro- filariae provides no real indication of the severity of the disease (number of adult worms in the heart and sub- sequent congestion thereof). Treatment of active cases involves first eliminating the infection and then preventing reinfection by administra— tion of a prophylactic drug. Different drugs are used for different stages of the disease as a drug used to treat one rive heartworm sodium which 0: :icrofilariae : tree weeks (0' gist. is recomme Moss treated :3” 1110-. 1967 P1“OPhYIaxi if :5; frOm two 3.130 season" a {it}. further I‘e 5791‘3' Six month FM- Admin :f circulating I isrgic or tox misole (levam. ‘sourrently be: :icrofilaricide i‘d in prophylas if‘mver a m + , .he dog d 11 stage of the disease may not be effective against another stage. Many drugs used to treat an active infection are often too toxic to be used routinely for prophylaxis. Ac- tive heartworm infections are treated with Thiacetarsamide sodium which causes the adult female worms to release their microfilariae all at once and then slowly die in two to three weeks (Otto, 1969b). Dithiazanine iodide or Stibo- phen is recommended to destroy the circulating microfilariae in dogs treated in the above manner (Otto, 1969b: Merck and Co., Inc., 1967). Prophylaxis currently is limited to daily administration of DEC from two months prior to two months after the "mos- quito season" at a dosage of 1.5 mg/lb (Jackson, 19690), with further recommendations that Thiacetarsamide be given every six months to kill any adult worms that may have de— veloped. Administration of DEC to a dog with high numbers of circulating microfilariae can produce severe, presumably allergic or toxic, reactions (Desowitz et al., 1975). Tet- ramisole (levamisole), a new broad-spectrum anthelmintic, is currently being tested for its potential use both as a :microfilaricide in the treatment of active heartworm disease and in prophylaxis (Tulloch et al., 1970) of the disease. However, discussion continues as to the proper dosage levels for the dog and the side effects and contraindications. Diethylcar ‘my-N-methylpi MN? by Hewi seed for use a rilized in a 5 :itrate tablet trade name Caz-j item) and has Noses of M rd other film Fawood, 1969; 1mg and 13°“ (1972) latively n°n_ t that Oral admi “reached a to hours afte :leared from 1 gm in the I tree times t] atesting lab. PM“ 01’ two ‘3‘! u- - 159d by ratc 12 nggmacology of Diethylcarbgmazine Diethylcarbamazine, the generic name for l-diethylcar- bamy-44methylpiperazine, was first reported as a filaricide in 1947 by Hewitt et a1. (1947). Although DEC was first pre- pared for use as a hydrochloride. it is now'most commonly utilized in a 50 percent biologically active dihydrogen citrate tablet form. This compound is marketed under the trade name Caricide (for use in dogs) and Hetrazan (for use in man) and has been used in the treatment of developing stages of flgchgrer1g b crofti, A. pacifica, Bzggiggma1ayi. and other filariases for the past twenty years (James & Harwood, 1969; Bryan & Southgate, 1976). Long and short term studies discussed by Burkhart and Alford (1972) on DEC's pharmacology have shown that it is re- latively non-toxic to vertebrates. Plasma assays showed that oral administration of the drug was rapidly absorbed and reached a peak concentration in the plasma approximately two hours after dosage, after which the drug was rapidly cleared from the blood. No detectable trace of the drug was found in the plasma after 24 hours. Daily administration of three times the recommended dose (9.9 mg/kg) to test dogs by a testing laboratory of the American Cyanamid Company for a period of two years produced no evidence of toxicity as judged by rates of survival. food intake, hematology, clini- cal chemistry. organ weights, and gross and microscopic pathology. Studies through F1 and F2 generations showed that DEC had no significant long term effects on reproduction. In studies 12.53am (1955a) aitroken down aim ring intac inhe degradat 2331' occurred, are excreted a 32* Mltls s 11°: SUCh upt finding medium ill m s the (1955b) stir-e agains t Stal' (1950) d showed no ill 6 31“. ”0100.1 dre ml. with t} Q W 29:”: BOlutiOn < t?) p1‘QbOScis 2 Such resu; :t'aéham’ 1955‘ '9 J"'f‘milizim 0 o ‘ .1"y~ .-'~€ their Su' \4 Min Of ml 1:382“... "““en‘t . 1'", a1. ( 13 In studies with cotton rats, hooded rats, and monkeys, Bangham (1955a) showed that DEC was metabolized very rapidly and broken down into four metabolites, each with the pipera- zine ring intact. Partial demethylation was an early step in the degradation, after which further stripping of the side chain occurred, until finally piperazine and methylpiperazine were excreted as the end products. Although microfilariae of Q, immitis showed uptake of the drug after one hour in :xitgg. such uptake was only one-third the amount of the sur- rounding medium (Bangham, 1955b). in gitgg studies with DEC and its metabolites by Bangham (1955b) showed that all of these compounds were in- active against Litomosoides carinii microfilariae. Hawking et a1. (1950) demonstrated that microfilariae of Q. repens showed no ill effects after having been placed in citrated heart blood drawn one-half to two hours after dosing the animals with the drug. In the same study. AnOQheles maculi- pennis atroparvus infected with Q. gepens fed on a one per- cent solution of DEC in glucose had well-developed larvae in the proboscis after 14 days. Such results led Hawking et al. (1950) and later authors (Bangham, 1955b) to speculate that DEC acts as a filaricide by immobilizing the microfilariae in the bloodstream or modi- fying their surfaces for later removal by phagocytosis. Exa— mination of microfilariae and host tissues before and after treatment with DEC in cases of onchocerciasis have led Gibson et al. (1976) to speculate from observable changes in lfi 'microfilarial cuticle that DEC alters the muc0polysaccharide or collagen of the cuticle unmasking the worm to the verte- brate host and allowing the body to recognize it as foreign. Pharmacology of Levamisole The anthelmintic activity of tetramisole hydrochloride (glrz,3.4,6-tetrahydro-6-phenyl imadazole (2,1-b) imazole hydrochloride) was first reported by Thienpoint et al. (1966). Investigators concluded that it was the levorotatory isomer of this drug which was biologically active and it is this isomer (lrtetramisole or levamisole) which is now in use in the United States as a general anthelmintic for sheep, cattle, and swine. The hydrochloride form is generally given in oral administrations of the drug whereas levamisole phosphate solu- tion is marketed as an injectable anthelmintic (Bradley, 1976). Although levamisole's primary use is in treatment of the many internal parasites of livestock. recent studies have shown that it may also be effective in treating certain types of cancer in man and animals and in certain.immune deficiency or autoimmune diseases such as rheumatoid arthritis (Tripodi et al., 1973: Schuermans. 1975: Grob. 1976). Its effect in these diseases is apparently to stimulate the immune system. Preliminary investigations have indicated that levamisole shows promise as a.Q. immitis microfilaricide and adulticide. Mills and Amie (1975) stated that daily doses of 10 mg/kg of levamisole cleared microfilariae from the bloodstream after 7 to 11 days of treatment. when the levamisole treatment gs given tC cement. iced circul aim worms :91200 mg 0 Other 5 effective ag Jar and Na sale (1 to 5 amicrof as of leva gig of DEC Station in 31ie(197u) : 3:10 Ina/kg : liemfilariat EilCtion in & Patiel Microfilm g3“ to trea. meats (J01 against micr. tirfain nema- IE pigment < 532nm 1a] limit 15 was given to dogs three weeks after Thiacetarsamide sodium treatment. Tulloch and Anderson (1971, 1972) reported re- duced circulating microfilariae levels and elimination of adult worms in infected dogs each dosed with a total of 1100 to 1200 mg over approximately six weeks. Other studies have indicated that levamisole is also effective against develOping stages of other filarial worms. Zaman and Natarajan (1973) showed that oral doses of levami- sole (1 to 50 mg/kg) eliminated circulating Breinlia g2;- gang; microfilariae in slow lorises after only one to four days of levamisole treatment, whereas treatment with 50 mg/kg of DEC over a similar time period resulted in only reduction in the numbers of circulating microfilariae. Duke (1974) found that levamisole injected intramuscularly at 10 mg/kg for 15 days in a chimpanzee greatly reduced microfilariae counts and Zaman and Lal (1973) found marked reduction in microfilariae numbers in two Wuchereria hénr crofti patients treated with levamisole. Similar reductions in.microfilarial counts were obtained when levamisole was used to treat Brugia malayi in.man (O'Holohan & Zaman, 1974) and cats (Joon-Wah et al.. 1974; Rogers & Denham, 1976). ..§n.!it£g studies have shown that levamisole is effective against microfilariae and third stage infective larvae of certain.nematodes. Rogers and Denham (1976) showed that a one percent concentration of levamisole can kill third stage infective larvae of Brugia pahangi in 21322 within five minutes and that a 0.0# percent solution can kill the 16 infective larvae within 50 to 60 minutes. These authors also reported that 0.5 to 1.0 percent solutions of levamisole can kill 2. pahangi microfilariae in yitgg within 15 minutes whereas 0.04 percent and even 0.01 to 0.03 percent concentra- tions can kill the microfilariae within 24 hours. Unpub— lished data reported by the same authors (Rogers & Denham, 1976) showed that significantly higher concentrations of DEC are required to affect E. pahangi microfilariae. More than 0.2 percent DEC was required to kill all g. nahaggi micro- filariae in 24 hours in 211:9, whereas 0.04 percent levami— sole killed microfilariae in the same amount of time. A 0.7 percent levamisole concentration killed microfilariae in six hours whereas 0.3 percent DEC was needed to produce similar results. (Hawking et al. (1950) reported that a 4‘mg/ml con— centration of 50 percent biologically active DEC was inef- fective against microfilariae of Q. repens in_xitgg.) Studies on the effect of levamisole on Brugia pahangi in Agggg aegypti indicated that when the drug was given to mosquitoes in sugar solution, concentrations as low as 0.08 percent resulted in stunted third stage or sausage stage larvae; greater concentrations reduced the number of larvae per infected mosquito (Rogers & Denham, 1976). Similarly, Gerberg et al. (1972) found that a 0.1 percent solution of levamisole administered in sucrose solution to A. aegypti mosquitoes infected with Q. immitis eliminated or prevented complete development of the nematode. Such studies show that levamisole is apparently directly effective against 3.- I._-’ I 17 nematodes rather than having to be first activated by meta- bolic processes in the dosed animal. Clearance tests in pigs have shown no traces (at 0.05 PPM sensitivities) of levamisole in the blood of an animal dosed with 8 mg/kg after 24 hours (Johnson et al., 1972). The most accepted explanation of levamisole's acti- vity is that levamisole probably acts by inhibiting the succinic dehydrogenase pathways of the nematode (Van den Bossche & Janssen, 1967; PharmIndex, 1973) resulting in helminth paralysis and dislodgement. Such paralysis pre- ceded by hyperactive coiling of Breinlia sergenti adults was reported by Natarajan et al. (197%) and has also been associated with increased muscular tone of the worms. Coles and Jenkins in personal communication to Rogers and Denham (1976) reported that ig,xitgg 0.01 percent concen- trations of levamisole caused reversible paralysis of the adult Nippostrongylus brasiliensis (an internal parasite) when the worms were left in the drug. The authors postu- lated that this recovered motility resulted from recovery of activity of a neuroreceptor in the worm. (The drug at high doses stimulates the receptor for a time but a point is reached where the receptor can no longer be stimulated and the worm recovers its motility.) Although levamisole appears to be as effective as DEC against develOping stages of Q, immitis and other filarial worms, controversy exists regarding its mammalian toxicity. Single oral dosages as low as 5 mg/lb have resulted in 18 vomiting, diarrhea, lack of appetite, and "distress" in treated dogs (Jackson, 1972), although these symptoms may appear only after the first dosage. Administration of the g and ; isomers at levels of 15, 20, and 25 mg/kg/day to cats infected with develOping stages of Brugia pahangi also produced salivation and "distress" for several hours after dosing (Rogers & Denham, 1976). Alford in Jackson (1972) reported that daily administration of the d; isomers of tetramisole (which is slightly less than 50 percent bio- logically active) to dogs at levels of 30 mg/kg, 15 mg/kg, and 7.5 mg/kg resulted in four out of six female dogs de- veloping hemolytic anemia after 13 weeks. No changes were seen in similarly treated male dogs, however, and attempts by Alford to duplicate his results were unsuccessful. Artificial Feeding Technique Current artificial feeding techniques involve the use of a membrane to cover the blood source and an apparatus to hold the blood and membrane and possibly regulate such fac- tors as temperature of the blood, etc. The blood used in the artificial feeding of mosquitoes is usually taken from the preferred host of that particular mosquito species (e.g.. in the case of A. aegypti, human or mammalian blood; in the case of Qulgx_territans, amphibian blood). Fresh, whole blood is usually used and is always either heparinized or defibrinated. However, hemolysed or frozen blood, erythro— cyte extract (Rutledge et al.. 1964), or outdated human “I ‘b \1' 19 blood (as described by Tarshis, 1959) can be prepared and used with success. Types of membranes used in such artificial feeding techniques have ranged from fresh ones. such as rat skin (Woke, 1937), rabbit skin (Yoeli, 1938), and chicken skin (Bishop & Gilchrist, l9h#), to artificial or prepared mem- branes, such as the Baudruche membrane (a bovine intesti- nal preparation) used by Greenberg (1949) and the Trojan brand NaturaLamb condom membrane used by Dr. Paul Grimstad of the University of Notre Dame (personal communication). Currently. most workers find these latter two membranes most satisfactory because of their ease of preparation and storage. Kartman (1953b)and.later authors such as Wade (1975) demonstrated that mosquitoes can successfully be infected with filarial worms using the artificial feeding technique and that anticoagulants apparently have no observable effect of the development of such larvae. The advantages to in- fecting mosquitoes artificially with such inoculants as filarial worms are not only reduced cost and experiment simplification. but also. as pointed out by Weiner and Brad- ley (1970), the titer of the infected blood can be regulated by dilution with "normal" blood. Microfilariae from such nematodes as Q. immitis are seemingly quite hardy. Bemrick et a1. (1965) showed that blood containing microfilariae could be stored for up to four months at -68°C and then thawed and fed artificially a mosquitoes . Lifective stage is transported ‘ taming microfi'. Sizcessfully in element of th Most of th gas has involv :ezbrane and th etal. (1974) 1 :tier suitable 73-5 tubes, plac :19 cage, Refi Eé’llation of E 7138(if and ‘3'}. ion-colloidal 3 the blood. “ls-filoped by R1 (:2 .1, . ‘1: water ‘to 3‘» 20 to mosquitoes. Development of the microfilariae to the infective stage in such mosquitoes was near normal levels. Ponnudurai et al. (1971) reported that filarial worms can be transported between laboratories by mailing blood con- taining microfilariae at room temperature and later can be successfully introduced to susceptible mosquitoes with de- velopment of the worms going to completion. Most of the variability in artificial feeding techni- ques has involved the feeding apparatus used to house the membrane and the blood. A simple system described by Wills et a1. (1974) involves placing the blood in test tubes or other suitable containers capped with membrane and inverting the tubes, placing them either over the mosquito cage or in the cage. Refinement of the technique has concentrated upon regulation of such variables as the homogeneity of the mix— ture (if and when the blood were mixed with non-soluble or non-colloidal substances) (Behin, 1967) and the temperature of the blood. A frequently cited system now in use is that developed by Rutledge et a1. (1964) involving circulating warm water to maintain the blood or other material at a con- stant temperature. This apparatus, shown in Figure 1, can be autoclaved and is especially designed for the feeding of infectious agents to mosquitoes. Other authors such as Wade (1975) have modified such feeding apparatuses to include a method for stirring the blood mixture. Figure l . Apparatus for (Adapted from 1964.) 21 . no ‘ .' o‘. ' ‘ _ ‘ -’~ .1";.\°.:J.;;‘.;..'..;‘.\.'..'.-..'.-.. -‘ ‘- . In. .I. ' .l a :5 'uo : 0.1:. ' . I u H . n. ‘ ' . ' a '. .’ ‘ ..~.. ~ . ' o ‘ l.. '{g.. . o o - . '0?'.:,.:'~.' 2:; s... : i'.'#. c . '-'-.‘:‘-“:v_-&__~5.'{-'.‘it‘-?"_ .' " artificial feeding of mosquitoes. a figure by Rutledge et al., The Procedl } an initial : rtion on “10qu :.*_s m the mo ad (2) perform issigned to (191? lemisole on ‘3 Ho et al. from the head C in insure that Salts here, all tentwere perfc .ae constraini METHODS AND MATERIALS Proposed Resggggh The procedure followed during this research involved (1) an initial period of gathering certain background infor- mation on mosquito feeding behavior, development of Q. im- mitis in the mosquito strain chosen, and research techniques, and (2) performing several replicates of a single eXperiment designed to determine the effect of diethylcarbamazine and levamisole on developing Q. immitis in the mosquito. Ho et al. (1974) reported loss of infective larvae from the head or mouthparts of infected mosquitoes over time. To insure that such losses would not affect experimental re- sults here, all dissections from a single replicate experi- ‘ment were performed over a 24-hour time period. (Given this time constraint, no more than approximately 80 mosquitoes could be dissected in one day.) Investigations have previous- ly been conducted on the effect of DEC and levamisole on microfilariae, infective larvae. or adults ig_1it§g. Experi- ments involving the effects of these two drugs on deve10ping ,2. immitis generally have involved administration of the drug to the mosquitoes in sucrose or other sugar solutions (Hawking et al., 1950; Gerberg et al.. 1972). The use of mosquitoes in screening anti-filarial and anti-malarial drugs, 22 mastering 1‘ robe finite eff 31.7;Keegan’ 1 grater. in thi 331112098 in a taining the d zit-3 drug in sue the blood inges goes directly 1 1c the divertic ifficulties 01 :dn-infected d: 1 because of “alien mosquitoe Edth infected figs using a. incial feed" TWO spec "ring thGSe {’54 3‘ isms ti - de “me UniVer t Ram h 1%. “'an ‘ J e t 1332‘: 23 administering the compounds in such a manner, has proved to be quite effective and relatively inexpensive (Terzian, 1947; Keegan, 1968; Gerberg, 1971; Gerberg et al., 1972). However, in this research the drugs were given to infected mosquitoes in a second blood meal to better simulate their obtaining the drugs under natural conditions. (Provision of the drug in such a manner may be vitally important in that the blood ingested by the mosquito during the blood meal goes directly to the midgut whereas sucrose solution goes to the diverticulum or crop.) Additionally, because of the difficulties of obtaining and maintaining infected dogs and non-infected drug-dosed dogs for the mosquitoes to feed upon and because of the variability of feeding rates reported when.mosquitoes feed directly upon dogs, mosquitoes were both infected with 2. immitis microfilariae and given the drugs using a modification of the previously described ar- tificial feeding apparatus. Mosquito Rearing Two species and five strains of mosquitoes were used during these experiments. Initially, the ROCK strain of Agggg aegypti, maintained in our laboratory at Michigan State University, was chosen for use in this research as ,5. aegypti had been shown to successfully support the de- velopment of Q. immitis (Taylor, 1960b; Gerberg et al.. 1971; McGreevy et al., 1974). However, early experiments indicated that Q. immitis would not develop in this particular strain mosquito“ (Th :irely surprising attained in the tssmetimes 105 filarial worms f0 Anative stI‘ :hetic‘nigan stra d obtained from State University search. EXperi ;.lgmgt_ig develo salary of this Mi reef in the lat triseriatus cc Items from the :f?{otre Dame. we: 33- Woody Foster, PFOV‘ided by Dr. fiszePtibility tc Evilopment in a] Es ‘, 31m on or each he techniclliees I i5,- ‘v , The Colony c 24 of mosquito. (This problem with the ROCK strain was not en- tirely surprising, as strains of mosquitoes which have been maintained in the laboratory for some time have been known to sometimes lose their ability to support deve10pment of filarial worms for unknown reasons; Taylor, 1960b.) A native strain of A. triseriatus (hereafter called the Michigan strain) initially chosen because of expediency and obtained from treeholes in wooded areas on the Michigan State University campus, was then considered for use in this research. Experiments indicated that this strain supported ,9. immitis deve10pment. However, when my particular sub- colony of this Michigan strain failed to mate and maintain itself in the laboratory, one strain from an established ,A. trisgriatus colony at the Ohio State University and two strains from the Vector Biology Laboratory at the University of Notre Dame were obtained. The Ohio strain (provided by Dr. Woody Foster, OSU) and the Walton and Alabama strains (provided by Dr. George Craig, UND) were all tested for susceptibility to Q. immitis. Although Q. immitis completed development in all three of these strains, the Alabama strain was chosen for use in the remainder of this research. A des— cription of each mosquito strain and its rearing and mainte- nance techniques appears below. .Aedes aegypti (ROC§_strgigl The colony of g. aegypti mosquitoes maintained at the Pesticide Research Center laboratory at Michigan State Uni- versity was originally obtained from Dr. George Craig of the 25 University of Notre Dame. Dr. Craig's laboratory initially received this strain from Dr. D. W. Jenkins of the Rocke- feller Institute in 1959. This strain of A. aegypti is con- sidered the best laboratory strain available because of its fecundity and vigor. These mosquitoes are relatively large and uniform and are used in numerous laboratories as the "typical" strain of mosquito (University of Notre Dame, 1974). Eggs of this strain were obtained from existing colonies at MSU and after embryonation were hatched in distilled water and transferred to enamel rearing pans (10"x l6"x 2%"), 350 mosquitoes per pan. Diet during this larval stage consisted of Tetramin, a commercially available fish food. Rearing pans were kept in an insectary maintained at 78:20F and 80:10% relative humidity. Pupation began approximately seven days after hatching. Adults emerged two days after pupation and were thereafter kept in 12"x 12"x 12" mosquito cages and maintained on 10 percent sucrose solution absorbed onto cotton. Two days after the infective blood meal, the experi- mental group of mosquitoes was provided with oviposition ma- terials (filter paper placed in a beaker containing approxi- mately one inch of water). (Although not all authors allow blood-fed mosquitoes to oviposit it was felt that nearly natural conditions should be maintained during this research. Additionally, as mosquitoes were expected to receive a second blood meal, it was felt that allowing them to oviposit after the first blood meal might increase the number which fed during the second blood meal.) Mosquitoes kept separately 26 for stock purposes were allowed to feed every two weeks on an immobilized guinea pig and eggs were later collected. Aedes triseriatus Michi an strain This strain of A. triseriatus was initially obtained from larvae collected from treeholes in wooded areas on the campus of Michigan State University. A successful colony was maintained in the insectary at the Pesticide Research Cen- ter by Mr. Mori Zaim but my attempt to maintain a subcolony failed because the adults would not mate. Attempts to en- courage mating behavior by lowering the temperature in the insectary to 72°F, by reducing the malesfemale ratio from 1:1 to 1:2, by increasing the size of the cage in which the mosquitoes were being maintained from 12"x 12"x 12" to 24"x 24"x 24", and by repositioning the lighting in the in- sectary proved unsuccessful. General maintenance techniques and blood meals were provided in a manner similar to that described above, for A. aegypti, although larval rearing pans contained 250 larvae each and water used for hatching was first deoxygenated with nitrogen gas. Aedes triseriatus Alabama strain Eggs from this strain of mosquito were obtained from Dr. George Craig of the University of Notre Dame in September 1976. Dr. Craig received this strain from H. Schoof of the Calhoun Technical Development Laboratory, U. S. Public Health Service, Savannah, Georgia, in July 1969, but the strain was reported to be initially collected from Alabama in the 19308 and had been maintained in the laboratory ever since that 27 time (University of Notre Dame, 1974; personal communication with Dr. Craig, 1976). This strain of mosquito is noted for its high fecundity and is considered to be an excellent laboratory mosquito (University of Notre Dame, 1974). Rearing procedures for this strain of A. trigeriatus dif— ferred somewhat from those for the two previously discussed strains. After embryonation, eggs of this strain were hatched in distilled water primed with a small amount of Difco nu- trient broth powder to reduce the oxygen content of the water. Shortly after hatching, mosquitoes were transferred to enamel rearing pans, approximately 250 mosquito larvae per pan, and were fed Tetramin fish food during the course of their de- ve10pment. Rearing pans were kept in the insectary, maintain- ed at 72i2°F and 80:10% relative humidity. Pupation began about 11 days after hatching and the pupal period lasted approximately two days. The adults used for stock purposes were placed in a 24"x 24"x 24" cage whereas mosquitoes which were used for experimental purposes were placed in 18"x 18"x 18" cages. Approximately three days after the experimental group of mosquitoes received their infective blood meal, oviposi- tion materials were placed in the cage and eggs were collect- ed.v As genetic selection of experimental mosquitoes was to be avoided, these eggs were later discarded. Aedes triseriatus (Walton strain) Eggs of this strain were also obtained from Dr. Craig of the University of Notre Dame and were collected by R. Beach at the Izaac Walton Preserve, St. Joseph Co., Indiana in f 33.9 1969- lished by fo the dating git]. This a. the harm: ieies trise This 11 Trio State at the med versity ap {pears anal 28 June 1969. The University of Notre Dame colony was estab- lished by forced c0pulation for two generations after which time mating occurred naturally (University of Notre Dame, 1974). This strain of mosquito was reared and maintained in the manner described above for the Alabama strain. Aedgggtriseriatus (Ohio strain) This mosquito, obtained from Dr. Woody Foster at the Ohio State University in September 1976, was established at the medical entomology laboratory of the Ohio State Uni- versity approximately 15 months prior to our obtaining it (personal communication with Dr. Foster, 1976). This Ohio strain was reared and maintained during the course of this research in the manner described for the A. triseriatug Alabama strain. Artificial Feeding Apparatus The apparatus used to provide mosquitoes with blood meals during the course of this research heated stationary sources of water which were in direct contact with the blood with small electric light bulbs (7-watt "night lights") (Figure 2). Such lights, wired in parallel, were enclosed inqu ml beakers sealed against a board with latex caulking to prevent electric shock. The temperature of the water and the blood was monitored with a small aquarium thermometer and the temperature of the water was regulated by manually turning the lights on and off. (A variable voltage regulator or "light dimmer" could also be used to control the 29 8d rr 8 S eag Phhr r _ flan utte age b1 ciit .Jno c e k wwe im .1 bol .6 m 8n 8 rs tu md ..o niee tt 1 a . erm iaflo ch mdc dle dt .t 9.5 1 elte enml 1.1 06y e.lah eoeu El lhb Ffwt Fch. Artificial feeding apparatus used to feed mos- quitoes in this research. Figure 2 . rater temIl givides e rlt'mut tt transports Membrz The a used durir lood meal tested . ( fed femal; Pirated 1‘; about 50 1 then seen: and eight feeding a for holdi Vith the m1 Pie lipped t C 30 water temperature.) The cost of such an apparatus which provides eight feeding stations is no more than seven dollars, without the dimmer, and it is lightweight and can be easily transported. Membrane Prgparation and Artificial Feeding Setup The artificial feeding apparatus shown in Figure 2 was used during this research both to infect mosquitoes with Q. immitis microfilariae and to provide them with a second blood meal containing various dosages of the two drugs tested. On the day of the first blood meal, previously un- fed female mosquitoes, approximately one week old, were as- pirated from their cage and placed in 24 pint jars, with about 50 female mosquitoes per jar. Mosquito netting was then secured over the tOp of these jars with rubber bands and eight jars at a time were placed in position under the feeding apparatus containing the blood. The containers for holding the blood consisted of eight small plastic cups with the bottoms cut out (open cylinders) (Figure 3). A small piece of Saran Wrap was loosely stretched across the lipped top of the cup and secured with rubber bands. Approxi- ‘mately 3‘ml of blood was poured into this section of the cup. A.membrane was then tightly stretched over this blood source and also secured with rubber bands. The cup was then in- verted and water was poured into the upper section. The heating apparatus, consisting of the light bulbs and glass beaker shield section, was then placed in the water and the 31 Thermometer Saran wrap .Membrane Figure 3. Feeding cup for use in the artificial feeding apparatus shown in Figure 2. 32 temperature was monitored with the aid of a small aquarium thermometer. Throughout the feeding period, the blood was maintained at approximately 102°F (dog body temperature) by turning the light bulbs on and off. Mosquitoes generally landed upon the membrane surface within the first five minutes after placement of the entire apparatus on the netting-covered jars containing the mos- quitoes. Feeding soon ensued and within 15 minutes most of the mosquitoes which were to feed engorged and drOpped to the bottom of the jar. Jars were removed from under the feeding apparatus after 20 minutes and the procedure was repeated. During the early course of this research, fresh bovine small intestine obtained from a local slaughterhouse, was used as a membrane source. The intestine was cut into 2"x 2" portions, placed in a pan containing saturated sodium chloride solution and soaked for approximately five minutes to help loosen the inner mucosa and the remainder of the gut contents. The inner mucosa was then removed by gently scraping it with the blunt edge of a scissors blade. The re- maining outer membrane was then placed in a saturated sodium chloride solution and soaked for another five minutes. Any remaining inner mucosa was gently rubbed away with the fin— ger and the membrane left was then rinsed several times in distilled water. The remaining outer membrane was tan colored, opaque, quite elastic and flexible, and was very easy to work with. Remaining portions of unprepared gut me then 1 also found the, and 1 Later, zez'rrane s . 33w gut men has invol they were p three or fc 7.25 . 33 were then frozen for later preparation and use. It was also found that the membranes could be prepared ahead of time, and frozen for later use. Later, Trojan brand NaturaLamb condoms were used as membranes. These condoms are commercially prepared from sheep gut and are quite flexible and much thinner than the cow gut membrane described above. Preparation of these mem- branasinvolved carefully washing off the lubricant in which they were packaged. Each membrane was then divided into three or four sections, ready for use in the feeding appara- tus. Sorting_the Mosgpitogg During the early eXperiments, blood-fed mosquitoes were sorted after being anesthetized with carbon dioxide gas generated from dry ice. Engorged mosquitoes were carefully separated from unengorged ones and placed in a new holding cage. This technique was extremely time- consuming and it was impossible to insure that no mosqui— toes would escape or become damaged when sorted in such a manner, so the following sorting technique was soon adopted. After being offered the first blood meal, all mosquitoes from the feeding jars were placed in a 12"x 12"x 12" mos- quito cage and unengorged mosquitoes or incompletely en- gorged mosquitoes were then aspirated from that cage and destroyed. The cagewas carefully checked visually several times for the presence of any non—fed mosquitoes. 0n th ties were ”titer. place: engorged a' :ri'oed and sized ice < in. Those :crd blood 2rd were re im. Que: res were ( Blood T does 1 at the Col: “100d we filariae (1 3:01 Cont; Tied Husk: 'Q98 and I u. g “daf‘rl “3 S bl< 34 On the day of the second blood meal, infected mosqui— toes were first transferred from their cage into jars and then placed under the feeding apparatus. Mosquitoes that engorged at this time were separated in the manner des- cribed and placed into small, especially prepared, pint- sized ice cream cartons, no more than 25 mosquitoes per car- ton. Those mosquitoes which did not engorge during this se— cond blood meal were used as one of the two control groups and were replaced into their original cage for later dissec— tion. Questionably engorged or partially engorged mosqui- toes were destroyed. Blood Soppce and_Drug,Concentratipns Blood used during these experiments was obtained from dogs kept for blood donation or research purposes at the College of Veterinary Medicine at MSU. Two types of blood were used; blood containing circulating micro- filariae (from a Basset hound named Scotch) and normal blood containing no microfilariae (from a mixed-breed dog named Husky). Both male dogs were kept indoors at all times and neither dog received medication of any kind. Husky's blood was used both to "dilute" Scotch's blood, which contained a very high microfilariae count (600 to over 1000 microfilariae per 20/04 depending on the time of day when it was drawn), and to provide mosquitoes with a second blood meal, at which time it was mixed with small amounts of the test drugs. Dilution of Scotch's blood as done bec :ces fed UPC Biliary e t a 34. Addi :irculating rte-i feedi gested by fan that ac 21:4 concen sei during 3n the dog was '- s11 anticc TE 1C’thlitc 4 of t} Siding Clips 7‘73 distr'n 35 was done because of reported of high mortality among mosqui- toes fed upon dogs with a high microfilaremia (Travis, 1947; Duxbury et al., 1961; Weiner & Bradley, 1970; Kuntz & Dobson, 1974). Additionally, microfilariae tend to settle when not circulating (that is, they settle on the membrane of the in- verted feeding apparatus) and the number of microfilariae ingested by artificially fed mosquitoes might be much higher than that acquired during natural feeding. For this reason, a 1:4 concentration of Scotch's blood to Husky's blood was used during the course of these experiments. On the day of the infective blood meal, blood from each dog was collected in 5 ml vacuum tubes containing EDTA as an anticoagulant. Such blood was immediately brought to the mosquito laboratory and combined. Approximately 2.5 to 3.0 ml of this blood.mixture was poured into each‘of eight feeding cups, the blood being continually shaken to insure even distribution of microfilariae throughout the medium. These prepared cups were then capped with membrane and pre- sented to female mosquitoes in the manner described previously. For the second blood meal, only Husky's blood was used and it was drawn in the manner described above, immediate- ly brought to the laboratory, and mixed with various concen- trations of the test drugs. At the time of this second blood meal, 2.5 ml of blood (final volume) was placed in each of the eight feeding cups. Two of them contained small amounts of 0.9 percent saline (mixed with the blood) as a control, L; and each of the remaining cups contained a 10'3, 10' , or "3'5 percer 33.112310?) Of Little 3 levamisc after norms that the cc :earc}: adeq A mom 3 consider ford (1972) 11% citrate (rich would Eight, Le ‘39 monito nation pea :f the Oral it a}. (194 Use Of DEC 11' the Same Til dOse . five-n to a "33 K's/1 C 73s Plasma. ROSEFE e I “its .ed t 36 10-5 percent concentration of either DEC or levamisole. Pre- paration of these drug concentrations is described below. Drag Concentrations and Preparation Little information was available on the amount of DEC or levamisole that would be present in the blood of a dog after normal prophylactic treatment. However, it is felt that the concentration of the drugs used during this re- search adequately represents the entire range of probability. A normal dose of DEC used for prophylaxis of Q. immitis is considered to be 3.3 mg/kg (drug base). Burkhart and Al- ford (1972) reported about dosing dogs with 45.4 mg/lb of the citrate form of DEC (50 percent biologically active) which would be equivalent to 50 mg/kg (base drug) of dog weight. Levels of DEC in the plasmas of such dosed dogs were monitored after dosing and after two hours the concen- tration peaked at 5 flg/ml plasma (5 mg/l), or one-tenth of the oral dose given. Similarly. the results of Harned et al. (1948) showed that one—fifth to one-tenth of an oral dose of DEC administered intravenously produced approximate- ly the same results (as judged by dog symptomology) as the oral dose. Therefore if a normal dose of 3.3 mg/kg were given to a dog, after two hours one would expect to find 0.33 mg/l of the drug (3.3 x 10'5 percent concentration) in the plasma. Rogers and Denham (1976) in dosing cats with levamisole indicated that an oral dose of 10,000 mg/kg of levamisole mid be necessar the drug in the t mm therefore 12 ing in the blood zhtadose of 1 acphylaxis asatir extrapolating f N meet to find 2. Sezause Gerberg e :rcentration of fifl.hmitis in fimyngof the “105 all at once 11365318/1 of 1 515'“ Percent of 53.3 percent (con. Ta Thus for t Percent Concentr- ste in these The BOUI‘Ce 733:: «one "as Car ‘r—o . w on “100 mg of Eii‘rod out 9 . T} '7pel‘ce n't Sal; +. LL. 9 drug has :\._ E .11 CH 37 would be necessary to produce a one percent concentration of the drug in the blood. A 1 mg/kg oral dose of levamisole would therefore produce a 10'“ percent concentration of the drug in the blood. Tulloch and Anderson (1972) suggested that a dose of l mg/lb, administered orally, is adequate for prophylaxis against 2. immitis in the dog. At this dosage, extrapolating from Rogers and Denham's findings. one would expect to find 2.2 x 10.5 percent levamisole in the blood. Because Gerberg et al. (1972) showed that a 10"3 percent concentration of levamisole affected the developing stages of Q. immitis in mosquitoes, and, because even if a maximum of S‘mg/kg of the drugs were to enter the bloodstream of a dog all at once the concentration found there would be only 65 mg/l of blood or 6.5 x 10"3 percent (given that eight percent of an average-sized dog's body is blood), a 10"3 percent concentration as an upper limit is not illogi- cal. Thus for the reasons presented, 10'3, 10'“, and 10"5 percent concentrations of each of these drugs were chosen for use in these experiments. The source of DEC for preparation of the various concen- trations was Caricide@)tablets, each containing 400 mg of DEC citrate. Each pill actually weighed 0.9h8 grams and to obtain 100 mg of the base drug, O.#7u grams of the pill were weighed out. This amount of DEC was placed in 1000 ml of 0.9 percent saline to produce a 10'2 percent concentration of the drug base and sonicated for five minutes to insure even distribution of the drug. (Saline was used as the diluent f 01‘ Wild lyse 1‘ he ml of th iiluted into OJL ail per :ffhese thI igblood. the blood we provided by :1 of blood The te milable i STCCK, as t Iiiespread gécording 1 Eterial CC 3L2190 1 W100 - fefflilatl‘aio. Es PlaCed a‘»*2 A per 38 diluent for these concentrations because sterile water alone would lyse red blood cells when added to the dog blood.) One ml of the 10"2 percent concentration was then serially diluted into 9 ml of sterile saline twice to produce 10'3 and 10'“ percent concentrations. One-quarter ml of each of these three concentrations was placed into 2.25 ml of dog blood. Thus the final concentrations of the drug in a, and 10'5 percent. A control was the blood were 10'3, 10' provided by placing 0.25 ml of 0.9 percent saline in 2.25 ml of blood. The tetramisole used during this research was only available in the liquid form, used for treatment of live- stock, as this particular drug is not yet authorized for widespread use in dogs by the Food and Drug Administration. According to the package instructions, one mg of the liquid material contained the equivalent of 182 mg of levamisole HCl, a 90 percent active compound (Jackson, 1972), or 164 mg of 100 percent active compound. Six-tenths ml of this formulation, containing approximately 100 mg of levamisole, was placed in 1000 m1 of sterile saline solution to produce a 10"2 percent concentration of the drug. Serial dilutions were made to obtain 10"3 and 10'“ percent concentrations, and 0.25 ml of each of these concentrations was placed into 2.25 ml of blood to obtain final concentrations of 10-3, 10'“, and 10"5 percent levamisole. One—quarter ml of 0.9 percent saline was placed into 2.25 ml of blood as a control. Through0 the drugs wer tion of the (1 Send in the luring each : were thoroug ;‘.pettes wer are step of eere made . were kept s‘ Proximately Dissec Quitoes we] Tlito. On 1118 315%: 31“ 39 Throughout these experiments. new concentrations of the drugs were prepared every two weeks to preclude deteriora— tion of the compounds (although this phenomenon had not been found in the literature). Pipettes were used only once during each step of the drug concentration preparations and were thoroughly washed before reuse at a later time. These pipettes were marked and the same pipette was used at the same step of preparation each time the drug concentrations were made. Prepared mixtures were shielded from light and were kept stored, when not in use, in a refrigerator at ap- proximately 40°F. Qissection Techniques Dissections of eXperimentally-treated infected mos- quitoes were made to determine whether the drugs had any effect on the development of Q. immitis larvae in the mos- quito. 0n the day of dissection, cartons containing the mosquitoes which had ingested the various drug concentra- tions were drawn at random and then completely dissected before the drug and dosage the mosquitoes had received was seen. At the time of dissection the mosquito to be examined was killed by knocking it several times against the side of the carton. (Freezing the mosquitoes or treating them in any other similar manner reduced the motility of the larvae found inside. which in turn made the larvae more difficult to identify and count.) The dead mosquito was placed on a clean slide under a dissecting microscope and legs and wings were removed. In the mam g. of the mOSCl‘ slide in a drop ezglf‘j anterior teased open and :aitder of the es the head, : apart. The he; phargmgeal are found larvae i so particular 3353’ infect iw quito and Con my Carefull eoiplete, eac 380799, and fly I? f Ound i l 40 In the manner described by Jones (1967), the entire gut of the mosquito was removed and placed on a separate slide in a drop of distilled water. The terminal and now- empty anterior abdominal segments of the mosquito were teased cpen and any remaining contents expelled. The re- mainder of the abdomen was broken away from the thorax, as was the head, and the thorax was opened and muscles teased apart. The head was teased cpen and the buccal cavity and pharyngeal area very carefully examined. (Burton (1963) found larvae in such unusual places as palpi and antennae, so particular care was taken in examining the entire head.) Many infective larvae were found in the labium of the mos- quito and could be overlooked, so the labium was always very carefully slit longitudinally. After dissection was complete, each slide was examined under the compound micro- sCOpe and the results recorded. Because so many larvae were found in the cervical region and in the anterior part of the thorax, it was difficult to ascertain whether these larvae were in the head or the thorax so counts from these two areas were added together. No larval stain was used during the microscOpic exami- nation because so many mosquitoes had to be examined during one day. Methylene blue, a quick stain, was not used be- cause in some cases its use reduced the motility of the Q. immitis larvae and made them more difficult to locate. The criteria used for determining the various developmental stages of the larvae was their size. Late second stage 41 larvae (post "sausage" stage) and early and late third stage larvae were added together in counts, though notation was made as to the approximate stage of each larva identified (Iyengar, 1957; Taylor, 1960b). Experimental Procedures Nine trials or experiments were conducted to gather background information for this research. Tasks included refining experimental procedures and techniques, determining the best mixture of Scotch's blood and Husky's blood to use in infecting mosquitoes, examining development of Q, immitis in the mosquito, and determining the best post-infection day for dissection of the mosquitoes. Although the procedures used in Experiments 1 through 9 varied somewhat from ex- periment to experiment, they were generally similar. Pre— viously unfed female mosquitoes of the same age group were offered a dilution of microfilaria-infected blood in the ar- tificial feeding device. (After Experiment 7. the Natura- Lamb membrane was used instead of the cow gut membrane in this device.) To reduce mosquito mortality. mosquitoes were not provided with this first infective blood meal until they were approximately one week old (Duxbury et al., 1961; Weiner & Bradley, 1970; Intermill, 1973). Following the 20 minute feeding period, engorged mosquitoes were sorted either by anesthetization (Experiments 1 and 2) or by aspiration (Experiments 3 through 9). Infected mosquitoes were kept in the insectary at 78:20F and 80:10% relative humidity, except where otherwi :csquitoes in fected mosqui Fesults secti :csquitoes re ere utilizec rlth Q. ELL ‘I-Ekl’ld, an :eived a sec Seven e Z’eito. For ale mO'Squ'n am dilut: tificial fe Eter fee d1 1.730 12"x 1 relative hl. 42 where otherwise mentioned. A second blood meal was provided mosquitoes in Experiments A, 5, and 6, and dissections of in- fected mosquitoes occurred at intervals specified in the Results section of this thesis. In those experiments where mosquitoes received a second blood meal, two control groups were utilized: one in which the mosquitoes had been infected with Q. immitis but had not received a second blood meal of any kind, and a second group in which infected mosquitoes re— ceived a second blood meal containing no drug. Seven experiments were conducted to determine the effect of DEC and levamisole on developing Q. 'mmitis in the mos- quito. For Experiments 10 through 16 previously unfed fe- male mosquitoes, approximately one week old, were offered a 1:4 dilution of microfilariae-containing blood in an ar- tificial feeding apparatus, using the NaturaLamb membrane. After feeding, engorged mosquitoes were sorted by aspiration into 12"x 12"x 12" cages and maintained at 72:2°F and 80:10% relative humidity. One week later a second blood meal con- taining the test drugs was offered to the infected mosqui- toes in the same manner as the infective blood meal. Two control groups were used as in Experiments 4, 5, and 6. Dissection of the mosquitoes from Experiments 10, 11, 12, 1“, and 15 took place on post-infection day 16. Dissections for Experiment 13 mosquitoes took place early, on day'lh after infection, and late dissections for Experiment 16 mos- quitoes were performed on day 17. Occasionally not all mos- quitoes could be dissected on the appointed day. In such 43 cases, the extra mosquitoes belonged to Control group 2 and were usually dissected during the first 12 hours of the following day. (The results of these late Control group 2 dissections were listed separately from the dissections per— formed on time.) Back leeriment l Scotch's bl 1:5. and 1:11 CO cffered to m zine what effect adamsequent n: 339 Shown in Ta‘t Percent for 337 After feedj separate Cages e RESULTS Background Information Experiments Experiment 1 Scotch's blood was mixed with Husky's blood in 1:1, 1:5, and 1:11 concentrations. These concentrations were offered to Aedes aegypti (ROCK strain) mosquitoes to deter- mine what effect, if any, each might have on feeding rates and subsequent mosquito mortality. Mosquito feeding rates are shown in Table 1. The overall feeding rate was 78.2 percent for 337 females fed and 94 females not fed. After feeding, engorged mosquitoes were placed in three separate cages according to the concentration of blood they had been fed. Examinations of these cages for the presence of dead mosquitoes were made over the next several days and the results appear in Table 2. Mortality among infected mos- quitoes was low after the first week of infection (1 to 2 dead per day) although mortality during the first six days after infection ranged from 68.5 to 82.6 percent. Dissections were performed of dead mosquitoes during the first week after infection and of the surviving mosqui- toes (I to 2 per day) almost daily up to 19 days after infec- tion. Microfilariae or early first stage larvae were seen distributed throughout the Malpighian tubules within 24 44 45 Table l. Feeding rates for A. aegypti (ROCK strain) in Experiment 1. Conc. Engorged Unfed % Females Jar No. Received Females Females Fed 1 1:1 32 6 84.2 2 1:1 30 8 78.9 3 1:1 20 3 86.9 Total 1:1 115 18 86.5 5 1:5 31 16 65.9 6 1:5 26 12 68.4 7 1:5 40 1 97.6 8 1:5 30 10 75.0 9 185 3 5 37-5 Total 1:5 130 44 74.7 10 1:11 4 12 25.0 11 1:11 36 8 81.8 12 1:11 23 6 79.3 13 1:11 22 4 84.6 14 1:11 7 2 77.8 Total 1:11 92 32 74.2 46 Table 2. Mortality rates of A. aegypti (ROCK strain) fed upon dog blood containing three different concentra- tions of Q. immitis microfilariae. Blood Conc. Day Post- No. Mosqs. Cumulative Cage No. Received Infection Living % Mortality 1 1:1 0 115 0 l 67 41.7 2 40 65.2 3 36 68.7 4 34 70.4 5 33 71.3 6 31 73.0 2 1:5 0 130 0 1 102 21.5 2 75 42.3 3 61 53.1 4 52 60.0 5 44 66.2 6 41 68.5 3 1:11 0 92 0 l 61 33.7 2 36 60.9 3 26 71.7 t: 22 76.1 5 20 78.3 6 16 82.6 4 :u . 4- a flu Ah .1 al.: a I] I cl 2 €me . r r CH :5 .J «.3— Ju ‘1 d 47 hours after the infective blood meal. By the fourth day, some larvae, shortened in length and enlarged in width, could be seen in the distal ends of the tubules. Later dissections showed that development past the "sausage" stage never did occur and most of the larvae were encapsulated by the mos- quito in the earliest stages of development. In six mosqui- toes dissected 19 days after the initial infective meal four had no larvae visible at all and two contained several encap- sulated larvae each. Experiment 2 .A. aegypti (ROCK strain) mosquitoes were allowed to feed on a 1:2 concentration of blood containing microfilariae, and feeding rates and Q. immitis development in the mosquito were examined. The overall feeding rate for this experiment was 52.6 percent (103 females fed and 93 females unfed). Seventy—two hours after the blood meal, 48 of the 103 en- gorged mosquitoes had died. Results of dissections performed on the 13th and 17th days after the infective blood meal were similar to those in Experiment 1: the nematode apparently did not develop in this strain of mosquito. Experiment 3 Both A. aegypti (ROCK strain) and A. triseriatus (Michi- gan strain) mosquitoes were allowed to feed on a 1:4 dilution of Q, immitis-containing blood and both feeding rates and nematode development were observed. Results are reported in Table 3. 48 Table 3. Feeding rates of A. aegypti (ROCK strain) and A. triseriatus (Michigan strain) in Experiment 3. Mosquito Engorged Unfed % Females Jar No. Species Females Females Fed 1 A. ae ti 8 7 53.3 2 10 11 47.6 3 18 20 47.4 4 9 12 42.9 5 12 8 60.0 6 20 6 23.1 7 7 24 22.6 8 11 24 31.4 9 9 5 64.3 10 21 52 28.8 11 17 15 53.1 12 5 14 26.3 13 17 22 43.6 l4 14 10 58.3 15 23 18 56.1 16 29 10 74.4 17 15 23 39-5 Total 295 281 51.2 18 A. triseriatus 13 21 39.4 19 12 11 52.2 20 15 21 41.7 21 7 12 36.8 22 24 8 75.0 23 9 7 56-3 Total 80 80 50.0 49 Mortality of engorged mosquitoes was extremely low. (The aspiration method of sorting engorged mosquitoes was used here.) Three days after infection only 10 out of 296 infected A. aegypti and four out of 80 infected A. triseria- Egg mosquitoes had died. By the 19th day a total of 19 A. triseriatus mosquitoes had died. A. aegypti were dissected on days 3, 4, and 6 after infection. As in Experiments 1 and 2, Q. immitis had not developed beyond the "sausage" stage. Encapsulation of early first stage larvae was again seen in many of the mos- quitoes dissected. A. triseriatus dissections, performed on days 3, 4, 6, 8, 10 through 14, and 17 post infection, revealed defin— ite progressive deve10pment of the nematode. Sausage stage larvae were seen on the third day and subsequent dissections showed continued larval development so that by the eighth day many late second stage larvae could be seen in the Mal— pighian tubules. By the tenth day, early third stage larvae were present in the tubules, and by day 11 many third stage larvae were observed in the abdominal hemocoele and thorax. By day 12, many infective larvae were found in the head and proboscis regions and any larvae still remaining in the Mal- pighian tubules were in at least the late second or early third stages of deve10pment. Dissections done after the 12th day showed a progressive decrease in the number of larvae found in the Malpighian tubules. A summary of A. triseriatus dissections appears in Table 4. 50 Table 4. Average number of late second and third stage D. immitis larvae per A. triseriatus (Michigan strain). Day Post— No. Mosqs. No. Larvae No. Larvae Total No. Infection Dissected in Abdomen in Head/Thorax Larvae 8 2 21.0 0 21.0 10 4 19.3 0 19.3 11 2 .5 0.5 9.0 12 6 3.8 16.2 20.0 13 4 3.3 7.0 10.3 14 4 2.0 9.0 11.0 17 4 1.0 6.3 7.3 The maximum number of infective stage larvae was found in the head region on post-infection day 12; therefore, it was decided that the best day for dissection, under the ex- perimental conditions used here, was the 12th day after ini— tial infection of mosquitoes. It was also decided that A. apgyppi would no longer be used in this research. Experiments 4 and_5 The purpose of these two experiments was to evaluate laboratory procedures used to provide the second blood meal before research proceeded on the major objective of the study. Both groups of A. triseriatus (Michigan strain) mosquitoes used in these experiments were hatched and reared together. Experimental procedures and techniques used were identical, with both groups of mosquitoes being allowed to feed on a 1:4 concentration of Scotch's blood. Feeding rates at the first blood meal for each experiment appeared to be about average (that is, about 50 percent fed); however, v-r 113‘ n r n: A‘ 51 the number of mosquitoes available for attempted feeding was low. Only 53 of approximately 100 mosquitoes from the Experiment 4 group and 35 of approximately 75 mosquitoes from the Experiment 5 group engorged. Six days later, a second blood meal containing various concentrations of the experimental drugs was made available to the groups of infected mosquitoes. Feeding rates this time were quite low: of the 38 original Experiment 4 mos- quitoes which were still alive at the time of this second feeding, only 9 took a blood meal; and of the 30 remaining Experiment 5 mosquitoes, only 5 fed this second time. Dissections from both Experiments 4 and 5 were begun on the 12th day after infection and the number of late second and third stage larvae recorded for each of the test and control groups. Results are shown in Tables 5 and 6. (See Appendix A for data on individual observations.) Merments 6 and 7 The purpose of these two experiments was to evaluate the experimental procedures involved in providing mosquitoes with a second drug—containing blood meal and gather data on infection levels within individual mosquitoes. The results of these experiments are discussed together because mosquitoes from each of these two groups were hatched on the same day and reared together. Infection of Experiment 6 and 7 mos- quitoes on 1:4 concentrations of Scotch's blood took place five days apart, although otherwise identical experimental procedures were used for both experiments. 52 Table 5. Number of late second and third stage D. immitis larvae in A. triseriatus (Michigan strain from Experiment 4. Treatment Day Post- No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10'3 — — - — DEC 10-4 12 2 5.0 2.0 DEC 10'5 12 1 30.0 - Lev 10'3 - - _ _ Lev 10'1+ 12 l 3.0 - Lev 10’5 12 1 6.0 — Control 1* 12 3 16.7 12.8 Control 2* 12 6 27.8 5.4 Control 2* 13 12 19.4 2.8 Table 6. Number of late second and third stage 2. immitis larvae in.A. triseriatus (Michigan strain from Experiment 5. Treatment Day Post- No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10"3 - - — - DEC 10'“ - - - — DEC 10‘5 12 1 19.0 - Lev 10"3 12 1 8.0 - Lev 10'“ - - - - Lev 10"5 - - - - (Control 1* 12 1 15.0 - Control 2* 13 16 14.1 1.4 *In these and following tables, Control 1 refers to the con- trol group containing infected mosquitoes taking a second blood meal containing saline rather than a drug and Control 2 refers to the control group containing infected mosquitoes not taking a second blood meal at all. 53 Only 34 mosquitoes from Experiment 7 took an infective meal and for this reason these mosquitoes were not given a second blood meal. Dissections of the Experiment 6 mosquitoes, which did take a second blood meal, were begun on day 12 af- ter infection and the results are shown in Table 7. Experi- ment 7 mosquitoes were dissected 14 days after infection and the results appear in Table 8. Additional data on the num- bers of second and third stage larvae from individual mosqui- toes can be found in Appendix A. Experyp’ ent 8 In this experiment, evaluations were made of feeding rates and developmental success of Q. 'mmitig for the Alabama strain of A. triseriatpg. The Alabama strain was infected with a 1:4 concentration of Scotch's to Husky's blood and kept at 75:20F. An estimated 75 to 80 percent of the fe- males fed during the 20 minute feeding period. Mortality of infected mosquitoes over the next 15 days was relative- ly low. Dissections of infected mosquitoes, performed on days 3, 5, 7, 9, 12, and 13 after infection, indicated the same, progressive development of Q. immitis that was described in Experiment 3. Dissections performed on days 15 and 16 indicated that on these days most larvae were in the head/ proboscis region and, after the 13th day, few larvae were still in the Malpighian tubules. Results of these dissec- tions are shown in Table 9. A more detailed account of the number of larvae observed in individual dissections ap- pears in Appendix B. l..:l 1 54 Table 7. Number of late second and third stage D. immitis larvae in A. triseriatus (Michigan strain) from Experiment 6. Treatment Day Post— No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10‘3 12 2 12.5 4.5 DEC 10’” 12 2 3.5 2.5 DEC 10'5 12 3 10.7 3.8 Lev 10"3 12 4 4.8 1.6 Lev 10’” 12 8 9.3 2.3 Lev 10'5 12 1 9.0 _ Control 1 12 7 10.3 1.5 Control 2 13 17 15.8 3.1 Control 2 14 27 10.9 1.1 Control 2 15 13 10.7 1.8 Table 8. Number of late second and third stage D. immitis larvae in A. triseriatus (Michigan strain) from Experiment 7. Standard Error Treatment Group (%) Mean No. Larvae Day Post- Infection No. Mosqs. Dissected DEC 10‘3 - - - - DEC 10'“ - - - - DEC 10"5 — — - - Lev 10"3 - - - - ‘Lev 10'“ - - - - Lev 10"5 - - - - Control 1 — - - - Control 2 14 20 55 Table 9. Average number of late second and third stage 9. immitis larvae per A. triseriatus (Alabama strain) from Experiment 8. Day Post- No. Mosqs. No. Larvae No. Larvae Total No. Infection Dissected in Abdomen in Head/Thorax Larvae 12 14 2.1 0 2.1 13 6 703 202 9'5 15 12 1.0 3.7 4.7 16 10 1.8 2.9 4.7 Experiment 9 Four strains of A. triseriatus (Walton, Alabama, Ohio, and Michigan) were compared for feeding rates and suscepti- bility to Q. immitis development. Mosquitoes of each strain were fed a 1:4 concentration of infected blood and maintain- ed in the insectary at 72:20F. Dissections were performed on days 2, 5, 12, 13, 15, and 16 after infection and no real variation in progressive rates or levels of nematode development was apparent although two Michigan and one Walton strain mosquitoes contained some encapsulated first stage larvae. Third stage, pre-infective larvae, were observed in the Malpighian tubules of all mos— quitoes dissected on the 12th day after infection and by the 13th day, infective larvae were found in the heads of the Alabama, Ohio, and Walton strains. A progressive shift in the numbers of larvae from the Malpighian tubules to the head region was seen in the later dissections. (Because of the few number of infected Michigan strain mosquitoes, this .Cu 56 strain was not dissected until day 15.) A summary of the dissection results is shown in Table 10. A more detailed account of these dissections appears in Appendix B. The temperature in the insectary during this experiment (72:20F) was lower than it had been in previous experiments (78:20F) and development of the nematode occurred at a slower rate. Consequently, the Optimum day for dissection of infected mosquitoes maintained at the lower temperature was no longer day 12 after infection. Examination of dis- section data indicated that post-infection day 15 or 16 was best for dissection because peak numbers of third stage larvae were found in the head regions on these days. Table 10. Average number of late second and third stage 2. immitis larvae per A. triseriatus in Experi- ment 9. Mosq. Day Post— No. Mosqs. No. Larvae No. Larvae Total Strain Infection Dissected in Abdomen in Head/Th. LarvaeH Alabama 12 4 21.3 0 21.3 13 5 6.8 6.8 13.6 15 10 2.7 11.3 14.0 16 19 0.8 10.9 11.7 Walton 12 4 14.5 0.5 15.0 13 5 9.6 0 9.6 Ohio 12 5 11.6 0 11.6 13 4 7.8 0 7.8 Michigan 15 8 2.1 12.3 14.4 Q! ; vi H“ .« AU .N L .1: 57 Observations on the survival rates and feeding acti- vities of the Alabama, Walton, and Ohio strains of A. pp;- seriatus indicated little or no differences. Dissections of the spermathecae from week—01d females of these strains indicated that successful mating had occurred and hatch rates of eggs of each strain were very similar. However, the Ala- bama strain was chosen for the remainder of this project be- cause I had on hand many more embryonated eggs of this strain than of the Ohio and Walton strains. Summary of Background Information Experiments The results of the first nine experiments dictated the direction of the remainder of this research. The ROCK strain of A. aegypti was eliminated when it was determined that it did not support Q. immitis development. The Michigan strain of A. triseriatus allowed complete development of this nema- tode but was not used because the females in my particular subcolony were not inseminated. Differences among the Ohio, Walton, and Alabama strains of A. triseriatus appeared to be slight, but the Alabama strain was chosen over the others for the remainder of this research because more eggs were on hand. A 1:4 concentration of Scotch's infected blood to Husky's "normal" blood was used because this concentration resulted in the deve10pment of high numbers of Q. immitis larvae (8 or more per mosquito) with little mosquito mortal- ity. The amount of blood needed when using this mixture (4 ml of Scotch's blood and 20 m1 of Husky's blood) also was the maximum that could be safely drawn from each dog for the experiments conducted each week. 58 The temperature in the insectary was lowered to 7212°F to accommodate the experiments of some other graduate stu- dents and this extended the developmental time of Q. immitis in the test mosquitoes. As a result, dissections during the drug feeding experiments were made on the 16th rather than the 12th day post—infection. The NaturaLamb membrane was used instead of the laboratory-prepared cow gut membrane because it was easier to prepare and its use resulted in increased mosquito feeding rates. Finally, aspiration was chosen as a sorting technique as it was less time-consuming than the carbon dioxide anesthetization procedure and it produced lower mosquito mortality. Dppg-Feeding Expgriments Experiments 10 tpxopgh l6 Dissection results for each experiment are shown in Tables 11 through 17. Details of individual dissections appear in Appendix A. Summ of -Feedin E eriments A summary of Tables 11 through 17, showing the means for each treatment group from the various trials. is pre- sented in Table 18. (The results of late Control group 2 dissections were not used to prepare Table 18.) An esti- mated mean of 6.18 for the missing Experiment 10 observa- tion in Table 18 was calculated using the formula T + b B - S b - 1 t - 1 59 Table 11. Number of late second and third stage D. immitis larvae in A. triseriatus (Alabama strain) from Experiment 10. Treatment Day Post- No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10‘3 16 3 9.33 0.67 DEC 10'“ 16 3 2.33 1.33 DEC 10'5 - - - - Lev 10"3 16 11 6.45 1.34 Lev 10-4 16 6 10.67 5.43 Lev 10"5 16 6 12.83 1.96 Control 1 16 15 9.27 1.76 Control 2 16 10 11.60 1.63 Control 2 17 10 11.60 1.67 Control 2 18 10 12.33 4.64 Table 12. Number of late second and third stage D. immitis larvae in A. triseriatus (Alabama strain)_fram__ Experiment 11. Treatment Day Post- No. Mosqs. Mean No. Standard Group (fl) Infection Dissected Larvae Error DEC 10‘3 16 12 8.25 1.50 DEC 10'” 16 18 7.67 1.08 DEC 10'5 16 9 6.56 1.99 Lev 10‘3 16 4 6.50 1.94 -Lev 10'” 16 7 9.86 2.39 Lev 10"5 17 9 9.33 1.93 Control 1 16 30 7.77 1.19 Control 2 17 63 8.00 0.75 6O Table 13. Number of late second and third stage D. immitis larvae in A. triseriatus (Alabama strain) from Experiment 12. Treatment Day Post- No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10‘3 16 6 10.33 1.76 DEC 10'“ 16 6 6.17 1.28 DEC 10'5 16 11 4.91 0.96 Lev 10'3 16 5 7.60 2.01 Lev 10'4 16 8 7.63 2.63 Lev 10‘5 16 6 14.83 4.22 Control 1 16 14 12.79 2.06 Control 2 16 18 10.11 1.72 Table 14. Number of late second and third stage 2. immitis larvae in A. triseriatus (Alabama strain) from Experiment 13. Treatment Day Post- No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10"3 14 10 7.20 1.43 DEC 10‘“ 14 11 7.91 0.72 DEC 10'5 14 14 8.21 0.96 Lev 10"3 14 6 7.50 1.64 Lev 10'“ 14 4 7.50 1.04 Lev 10'5 14 10 8.90 1.16 Control 1 14 17 6.88 0.98 Control 2 14 21 9.05 1.04 61 Table 15. Number of late second and third stage 9. immitis larvae in A. triseriatus (Alabama strain) from Experiment 14. Treatment Day Post— No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10'3 16 5 9.80 2.63 DEC 10-4 16 7 11.14 1.53 DEC 10‘5 16 10 6.90 1.70 Lev 10"3 16 7 7.43 1.91 Lev 10'“ 16 6 8.33 1.33 Lev 10‘5 16 5 8.60 3.56 Control 1 l6 17 10.82 1.15 Control 2 l6 14 12.79 2.16 Table 16. Number of late second and third stage D. immitis larvae in A. triseriatus (Alabama strain) from Experiment 15. Treatment Day Post- No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10‘3 16 19 6.32 1.24 use 10'“ 16 12 6.83 1. 50 DEC 10’5 16 13 5.31 0.88 Lev 10"3 16 16 4.19 0.78 Lev 10- 16 13 6.08 1.41 Lev 10'5 16 11 7.00 1.17 (Control 1 16 28 7.54 1.49 Control 2 16 9 8.89 2.37 62 Table 17. Number of late second and third stage D. immitis larvae in A. triseriatus (Alabama strain) from Experiment 16. Treatment Day Post— No. Mosqs. Mean No. Standard Group (%) Infection Dissected Larvae Error DEC 10‘3 17 5 6.40 2.52 DEC 10’“ 17 4 2.75 1.03 DEC 10’5 17 6 2.83 1.08 Lev 10“3 17 8 2.38 0.96 Lev 10-4 17 8 4.00 0.80 Lev 10'5 17 6 3.67 0.95 Control 1 17 8 2.88 0.81 Control 2 l7 6 6.50 0.76 Table 18. Summary chart of number of late second and third stage D. immitis in A. triseriatus (Alabama strain) from Experiments 10 through 16, grouped according to dissection day. Treatment Exper. Exper. Exper. Elle-gen]. Exper. Exper. Exper. Group (%) 10 11 13 1 DEC 10'3 9.33 8.25 10.33 9.80 6.32 7.20 6.40 DEC 10-4 2.33 7.67 6.17 11.14 6.83 7.91 2.75 DEC 10‘5 - 6.56 4.91 6.90 5.31 8.21 2.83 Lev 10'3 6.45 6.50 7.60 7.43 4.19 7.50 2.38 Lev 10’” 10.67 9.86 7.63 8.33 6.08 7.50 4.00 Lev 10"5 12.83 9.33 14.83 8.60 7.00 8.90 3.67 Control 1 9.27 7.77 12.79 10.82 7.54 6.88 2.88 Control 2 11.60 8.00 10.11 12.79 8.89 9.05 6.50 63 where T = the sum of all other weekly values for the DEC 10'5 percent treatment group, B = the sum of all other values for the various treatment groups of Experiment 10, t = the total number of treatment groups, including control groups (8), b = the total number of trials or experiments (5), and S = the sum of all values shown for all experiments and all treatment groups (Snedecor, 1956). Treatment means from only Experiments 10, ll, 12, 14, and 15 and total means for each treatment group from those five trials are shown in a second summary table, Table 19. (The overall mean for the 10'5 percent concentration of DEC was calculated using the estimated mean of 6.18 for the missing value from Experiment 10.) Table 19. Summary chart of means of late second and third stage D. immitis in A. triseriatus (Alabama strain) from experiments in which dissections were perform- ed on day 16 after infection, plus overall means for each treatment group. Treatment Exper. Exper. Exper. Exper. Exper. Overall Group (%) 10 ll 12 1 15 Means DEC 10'3 9.33 8.25 10.33 9.80 6.32 8.81 DEC 10‘“ 2.33 7.67 6.17 11.14 6.83 6.83 DEC 10'5 (6.18) 6.56 4.91 6.90 5.31 (5.97) Lev 10"3 6.45 6.50 7.60 7.43 4.19 6.43 Lev 10’” 10.67 9.86 7.63 8.33 6.08 8.51 Lev 10’5 12.83 9.33 14.83 8.60 7.00 10.52 Control 1 9.27 7.77 12.79 10.82 7.54 9.64 Control 2 11.60 8.00 10.11 12.79 8.89 10.28 y——— 64 Graphs of the overall treatment group means for diethyl— carbamazine and levamisole shown in Table 19 are presented in Figures 4 and 5. Analysis of Experimental Results The results of the drug-feeding experiments were ana- lyzed in two phases, each dealing with a major question about this research and each involving several separate analyti- cal steps. To determine whether any drug treatment group showed a significant difference in Q. immitis numbers which developed, an analysis of variance, an orthogonal analysis, and a non-orthogonal analysis were performed. To determine whether individual stages of Q. immipig larvae were affected by diethylcarbamazine or levamisole and whether the larvae in all mosquitoes of a test group were synchronized (at the same stage of development), frequency distributions, corre- lation coefficients, "deve10pmental variances", power curves, and deve10pmental cycles were examined. Each of these ana- lyses is presented below. Phase one Analysis (1) Analysis of variance. The research performed for this thesis was classified as a randomized block design ex- periment with replications in subclasses (Dr. Charles Cress, Michigan State University, personal communication). Because the number of observations in the subclasses was dispr0por— tionate and there was a missing observation from Experiment 10, the computer was used to perform the analysis of variance 12 11 10 9 8 3 7 E ’4 6 6 5 Z a 3 2 1 Figure 65 o 0 DEC 10"3 DEC 10—4 DEC 10"5 Control 1 Control 2 4. Overall means from Experiments 10, ll, 12, 14, and 15 of late second and third stage D. immitis lar- vae in A. triseriatus (Alabama strain) for DEC treatment groups (7) and controls. 12 11 10 9 8 3% 7 g. 6 5 2: L. 3 2 1 Figure 66 Lev 10'3 Lev 107E Lev 10"5 Control 1 Control 2 5. Overall means from Experiments 10, ll, 12, 14, and 15 of late second and third stage D. immitis lar- vae in A. triseriatus (Alabama strain) for levami- sole treatment groups (fi) and controls. 67 and orthogonal contrasts. The program package used was SAS (Statistical Analysis System) by Barr, Goodnight, Sall and Helwig of the SAS Institute in Raleigh, North Carolina, and the computer work was done at the Texas A&M University Com— putation Center. To arrive at the analysis of variance (ANOVA) apprOpri- ate for my experimental design, shown in Table 20, two ini- tial analyses of variance were performed using a general linear models procedure. The first ANOVA adjusted variation in treatment groups for weekly variation and adjusted inter- action variation for both weekly and treatment group varia- tion. The second ANOVA adjusted weekly variation for varia- tion in treatment groups. These two analyses of variance Table 20. Mathematical model for analysis of variance of a randomized block design experiment with repli- cates in subclas es (where 0' is estimated variance, K1, ¢» , K2 are constants, and W, T, WT, and E refer respectively to weeks, treat- ment, weeks*treatment, and error). (Graybill, 1976) Degrees of Source of Variation Freedom Estimated Mean Square Weeks (adjusted for 2 d 2 d 2 treatment) 4 d E + K1 WT + K2 w Treatment (adjusted 2 a 2 2 for weeks) 7 6E + K1 WT H“T WeekstTreatment (ad- 2 2 justed for weeks 27 6E + KldWT and treatment, both) Error 433 6E2 "'1 68 produced the apprOpriate ANOVA in which weeks were adjusted for treatment groups, treatment groups were adjusted for weeks, and the interaction between weeks and treatment groups was adjusted for both weeks and treatment groups. The final product is shown in Table 21. (See Appendix C.) Table 21. Final analysis of variance of the results from Experiments 10, ll, 12, 14, and 15. Mean Level of Source of Variation df Square F-value ‘ Significance Weeks (adjusted for Treatment (adjusted for weeks) 7 118.8106 4.1025 0.01, df7'27 WeekseTreatment adjusted for both weeks and treat- 27 28'9603 0'7953 NS ment) Error 433 36.4152 - TOTAL 471 - - F-values for weeks and for treatment were computed using MSW*T rather than MSW§T + MSE as a divisor, to be conserva- tive. F—values for both the weekly experiments and the treat- ment groups indicate significance at the 0.01 level. Signi- ficance for weeks can be overlooked, however, as significant variance among weeks was expected due to weekly variation in the microfilaremia count of Scotch's blood and other un— controllable experimental conditions. 69 (2) Orthogona;_analysis. To pinpoint the exact source of variation among the treatment groups, seven different orthogonal contrasts were performed comparing: (1) control group one to control group two; (2) all drug treatment groups together to both control groups combined; (3) all three concentrations of DEC together to all three concentra- tions of levamisole together; (4) the 10.3 percent and 10—4 percent concentrations of both drugs to the 10"5 percent concentrations of both drugs; (5) the 10"3 percent concen- trations of both drugs to only the 10')+ percent concentra- tixe of both drugs; (6) the 10"3 percent and 10-4 percent concentrations of DEC plus the 10'5 percent concentration of levamisole to the 10'5 percent concentration of DEC and the 10'3 percent and 10—4 percent concentrations of levami- sole; and (7) the 10"3 percent concentration of DEC and the 4 10' percent concentration of levamisole to the 10')+ per- cent concentration of DEC and the 10’3 percent concentra- tion of levamisole. The results of this orthogonal analy- sis are shown in Table 22. The only contrasts which showed significance were con— trasts 2 and 6 (both of which were significant at the 0.01 level), indicating that the results of the drug treatments differed significantly from those of both control groups, and that DEC and levamisole at the concentrations tested be- haved quite dissimilarly. (See Appendix C.) (3) Non-orthogonal analysis. To determine a more exact source for the treatment variation indicated by the 70 Table 22. Orthogonal analysis (seven contrasts) performed on the results of Experiments 10, ll, 12, 14, and 15. Source of Variation df sgizge F-value Sigfiigicggce “2:212:88th . 1.7.7... - Contrast l 1 2.0512 0.0708 NS Contrast 2 1 286.5938 9.8961 0.01, df1,27 Contrast 3 1 34.1332 1.1786 NS Contrast 4 1 8.6453 0.2985 NS Contrast 5 1 21.5740 0.7450 NS Contrast 6 1 387.0572 13.3651 0.01, df1,27 Contrast 7 1 92.0246 3.1776 NS WeekseTreatment 5.22.2282: £22223 27 - 1 ment) Error 433 36-4152 - TOTAL 471 .. .. 71 results of the analysis of variance and orthogonal contrasts, eight non-orthogonal contrasts were performed, manually, using the formula (W) (Zea-wwe , 13 where (referring back to Table 18) i and j refer to columns and rows, reSpectively, §ij = the mean value for each cell, n.lj = the number of observations represented by each cell, cij = a weighted value assigned to each contrast, and MSTW = 28.96, the mean square for the interaction of weeks and treatment groups (from the analysis of variance). t2 values were arrived at using the mean values for each treatment group from Experiments 10, 11, 12, 14, and 15 only, with 6.18 used as an estimated mean for the missing value in Ex— periment 10. A summary of these non-orthogonal contrasts appears in Table 23. The t2 results indicated that, when conservative levels of significance at df7’27 of 8.08 at the 0.05 level and 12.35 at the 0.01 level were used, only those contrasts comparing DEC to Control group 1 and the 10'3 percent con- centration of levamisole to Control group 1 were significant. (Such calculations used an n value of l for the missing Ex- periment 10 cell. If the harmonic mean (5.57) had been used instead, the t2 values for the DEC vs Control 1 con- trast and the DEC 10"5 percent vs Control 1 contrast would have been 10.09 and 14.07, respectively, indicating Ll.‘ 72 Table 23. Results of non-orthogonal analysis (eight con— trasts) performed on the results of Experiments 10, ll, 12, 14, and 15. Contrasts t2 Level of Significance DEC vs Control 1 8.55 0.05, df 7,27 Lev vs Control 1 2.25 NS DEC 10‘32 vs Control 1 0.54 NS DEC lO—u% vs Control 1 5.51 NS DEC 10-5% vs Control 1 7.05 NS Lev 10-3% vs Control 1 8.75 0.05, df7'27 Lev 10'u% vs Control 1 1.16 NS Lev 10-5% vs Control 1 0.67 NS 2 significance at the 0.05 and 0.01 levels.) The t compu- tations for each of these contrasts appears in Appendix D. ‘EpagegTwo Analysis (4) Frequency distpibutipns. To determine whether or not the distribution of larvae within treatment groups fit the normal curve, frequency distributions were compiled based upon percentage development within that particular mos- quito dissection. For example, if the results of three mos- quito dissections revealed (0,2), (6,0), and (7,8) larvae per section of each mosquito, where x of (x,y) represents the num- ber of larvae found in the abdominal region and y of (x,y) re- presents the number of larvae found in the head/thorax, then percentage development for each mosquito observation would be 73 100(1 - ), or 100, O, and 53 Percent, respectively. __£__ x + y A frequency distribution of the treatment group from Ex— periments 10, ll, 12, 14, and 15 with the largest number of observations (n = 63) is shown in Figure 6. As can be seen from Figure 6, over 50 percent of the observations fall into the 0 and 100 percent categories in- dicating a wide divergence in larval development within the same mosquito p0pulation. Similar findings are seen when one examines all treatment groups from the experimental trials, although, of course, such divergence can only be evident when most larvae are neither at the end or beginning of their developmental cycle (neither at the 0 percent or 100 percent end of the scale). The results from other treatment groups are shown in Table 24. (5) Correlation coegficiepts and means. Because fre- quency distributions are limited in that they are based upon percentages rather than raw data, correlation coefficients (r), which use raw rather than converted data, were com- puted by the formula 3 1 1 r = 11%- ’ where 8xy = n—l zxiyi " 323131 ' x y sx - :23x12;ié::ij)é/g % and s = 2yiz-(Zyi 2% é . y n - 1 An r value to 0 to —l.0 indicates possible asynchrony of larval development and an r value of 0 to +1.0 indicates 74 A.covoommfio opwsUmoe Some scum ow>hma Mo honesc Havop map psomongon amp Mono mafiehom newness Hmsofl>flocfi onav .Asflmnpm mean umHoo mflPHESH .m owmpcoopog op mcfionooom povspHLPmfle .N Homecoo macho pcospwona .HH enoaauomxm scum mcowvm>nomno .0 oesmfim pcosmoao>om owmvnooeom 00H mm-oo mm-om om-oe mouoo mm-om mane: mmuon mNION mH-OH muH o mN OH m o N HH 6 HH WH N e mN m mN N A mH N o N N H NH 5 0H 0 NH m m :H a 2 OH 5 3 6H 3 a e H o m 6H N s m mm.N n AOOHV m N m m. OW.©H H Ammlcmv N U HH oo.HH u Ammuoev m:.m n Amm-omv H OH m AN ow.o u AmNqu - u Ame-oev OH r m oo.HH n Aomuomv om.eH u Amo-oov N NH % 3H om.:H u Ameuoov om.m u Ammuomv H .w mH HH.e u Ammsoev mN.N u Ameuoov N :H m a; n STONV one n 18-63 3 .N H om.OH u AmHnHv mm.m u HmN-ONv N 6H . m 6N.m u “OOHIOAV om.0H n AmH-OHv m o No.N u Aeoaomv - u “oqu NH N AH.o u AmNuov AH.m u gov ”meomopmo nomo CM newness pom msmoz om .cx H Illtll‘ ‘\ 4.! Ni Nlrfl firfit 75 a: mm mm NH N N 0N AH e Hopes AHH.NNV Ae.eNV Ao.va ANH.:NV Ao.HNV ANH.NNV AN.ANV AN.NNV mono pm a m NN HH NH NH NH NH NH .o>oomno .oz mH psoaaaoaxm NN NH on a: NN NH NH 2N a Hopes AeH.omv Ao.NNV Ao.ONv Amm.mmv Ao.HNV AOH.ONV Ao.mmv ANN.o:V oeco em a 3H NH m N m 0H m m .m>uomeo .02 :H psoeanomxm NN me an HN Ne NN NN Ne a Hmeoe AHH.NHV AeH.:HV Amm.mmv ANN.NNV Hoe.ONv ANN.NHV AoN.NHV Amm.ov mono em a NH :H N N N HH N N .m>noono .oz NH pcoeroaxm HN NH 3: NH om m e N a Hopes ANN.NNV Am.oNV HHH.NNV Ao.HNv Ao.NNv Ao.NNV Ao.va Ao.Nev ooze pm a mm on m n d 0 ma NH .m>nomno .02 HH pcoeapomxm NN NN NH NH ca - o NN a Heeoe Ao.ONN AN.oNV Ao.mmv Ao.omv ANN.NHV - Ac.OOHv Ao.mmv more we e OH NH N N HH - m. N .m>eoono .62 OH Pcoeflpomxm N H N-OH -OH N-OH N-OH -OH NIOH Aav ozone Hoopcoo Hoeecoo eoH >oH >oH can own one remapoooe .A «Aaw+xw\xuvnHvoOH an NocHanoPov .Qsohw PQNEPNoHP Peep mo Paosmoao>oc % Havop New “ROOH u a New R0 u x_Aa.xv oposzv Nonoao>oo ROOH one R0 who; roan; mcoHpm>uompo mo ommvsoopoa “azopw p:NEPNoup some pom mc6flpm>pomno mo nopewz .em oHnNe 76 in-phase larval development. The correlation coefficient is most reliable as an indicator of synchronization when the sample is large and the observations are taken at neither the beginning of larval movement out of the abdomen nor after most larvae have disappeared from the mosquito mouth- parts. A summary of weekly means and computed r values from Experiments 10, 11, 12, 14, and 15 is shown in Table 25. As can be seen in Table 25, 27 out of the 39 treatment groups had negative r values. The average r value for all treatment groups was -0.16. This indicated a slight negative corre- lation between the number of Q. immitis larvae in the abdom- inal and head/thoracic regions of mosquitoes. The only r value which was significant at the 0.01 level was that for Experiment 11, Control Group 2 (-0.32 at df61), but this 'may be due to the low number of observations within the treatment groups. (6),"Develppmenta;:variances". Because the correlation coefficient (r) demonstrates correlation between head/thor- acic and abdominal larval numbers but does not adequately differentiate between extremeg in degree of larval develop- 'ment, a comparison wasmade of variance computed for the data obtained at the time of dissection and predicted vari- ances at the beginning, end, and middle of larval development. (E.g., the correlation coefficient r = -l.0 for paired num- bers (8,1) and (1,8), where x and y of (x,y) represent numbers of Q. immitis larvae in the abdomen and head/thorax 77 Table 25. Summary of average number of late second and third stage 2. immitis larvae found in total body, head/ thorax, and abdomen dissections of infected A. tri- seriatus (Alabama strain) mosquitoes along with correlation coefficients (r) calculated from in- dividual data. (See data sheets in Appendix B.) Treatment Exper. Exper. Exper. E er. Exper. Group (%) 10 ll 12 14 15 DEC 10"3 9.33 8.25 10.33 9.80 6.32 Head/Thorax 6.67 7.58 2.67 7.40 5.21 Abdomen 2.67 0.67 7.67 2.40 1.11 Corr. coef. -0.94 +0.42 +0.03 -0.65 —0.03 DEC 10-4 2.33 7.67 6.17 11.14 6.83 Head/Thorax 2.33 7.39 0.67 9.71 5.50 Abdomen 0 0.28 5.50 1.43 1.33 Corr. coef. - +0.33 -0.62 -0.10 -0.24 DEC 10'5 - 6.56 4.91 6.90 5.31 Head/Thorax — 6.33 1.73 5.90 5.00 Abdomen - 0.22 3.18 1.00 0.31 Corr. coef. - +0.81 +0.01 +0.68 -0.62 Lev 10"3 6.45 6.50 7.60 7. 43 4.19 Head/Thorax 1.91 3.25 1.80 5. 57 3.94 Abdomen 4.55 3.25 5.80 1.86 0.25 Corr. coef. -0.17 -0.21 -0.32 +0. 02 -0.14 Lev 10'“ 10.67 9.86 7.63 8.33 6.08 Head/Thorax 9. 33 8.57 3.00 4.67 4.92 Abdomen 1. 33 1.29 4.63 3.67 1.15 COI‘I‘. coef. ‘0. 35 ‘0025 ‘0052 '0068 -0015 Lev 10’5 12.83 9.33 14.83 8.60 7.00 Head/Thorax 10. 50 5.22 6.83 6.00 4.55 Abdomen 2. 33 4.11 8.00 2.60 2.45 Corr. coef. -0. 48 -0.14 -0.50 +0.12 -0.07 Control 1 9.27 7.77 12. 79 10.82 7.54 Head/Thorax 6.93 6.33 7.07 9.41 4.93 Abdomen 2.33 1.43 5. 71 1.41 2.61 Corr. coef. +0.18 -0.17 -0. 34 -0.32 +0.08 Control 2 11.60 8.00 10.11 12.79 8.89 Head/Thorax 4.70 3.14 3. 83 9. 79 5.00 Abdomen 6.90 4.86 6. 28 3. 00 3.89 Corr. coef. -0.16 -0.32 +0.13 -0. 45 -0.21 78 of individual mosquitoes. The first two pair of numbers, however, show a more "synchronous" development of larvae within that treatment group of two mosquitoes than the last two pair.) Although a comparison of "developmental variances" represents no valid statistical test, it is discussed here for illustrative purposes. To compare existing larval synchrony levels within treatment groups to 100 percent synchrony, four types of "variance" were computed from v1, wl, v2, w2, . . . vi, wi, where i is the number of observations in that treatment group: (a) "actual developmental variance" (ADV) or variance (82) where vi = 51 and WJ.- = y1 ; (b) "estimated variance" or variances (82) if all larvae were at the beginning of their cycle (EVB), where vi = -(x1 + yi) and W1 = 0 ; (c) "estimated variance" or variance (s2) if all larvae were at the mid of half-way point of development (EVM), where vi = -%(x1 + yi) and wi = #(x.l + yi) 3 (d) "estimated developmental variance" (EDV) or vari- ance (s2) if all larvae at the present developmental level were 100 percent synchronized, where EDV = H (EVB — EVM) + EVM .An.index ADV/EDV was computed to compare existing deve10pmen- 'ta1 variance in a treatment group with variance expected if all observations within that treatment group were 100 percent synchronized at the existing stage of development. 79 Above computations performed for the results of Experi- ments 10, ll, 12, 14, and 15 are shown in Table 26. Although most treatment groups had ADV/EDV ratios over 1.0, ratios under 1.20 (or more conservatively, 1.25) were not considered significant. Seventeen of 39 treatment groups had ADV/EDV ratios greater than or equal to 1.20, ranging as high as 1.83, and the treatment group with the highest number of ob- servations (n = 63) (Experiment 11, Control Group 2) had an ADV/EDV ratio of 1.53. (7) Powepycurvep. Data was fit to a variety of curves (power, exponential, logarithmic) to determine whether lar- val numbers fluctuated with stage of development. The three previous analyses of experimental data suggested development of Q. immitis larvae within treatment groups was not syn- chronous, but provided little information about whether similar numbers of larvae were found in late, early, and middle stages of development. (Frequency distributions show- ed fluctuations in larval numbers as development progressed (Figure 6) but were based upon percentages rather than raw data and were affected by the magnitude of the numbers. Cor- _ relation coefficients indicated negative correlations in ab- dominal and head/thoracic larval numbers but fit the data to only a linear equation.) Tryouts showed that the data fit the exponential curve (y = aebx) poorly, fit a linear regression and logarithmic curve equally well, and, in many cases, fit the power curve (y = axb) best. (Closeness of r or r2 (coefficient of 80 AN.H NN.H NH.H NH.H No.H NH.H NN.H oN.H >nman< . 2N.NN NN.NN oe.Ne NN.NH NN.N NN.HH NN.NN mn.o~ >nm u NN.NN NN.NN NN.NN NN.NH NN.N ea.o HN.NH NN.NH s>m m2; HN.oe Ho.ee NN.NN NN.HN NN.N NH.NH eo.mN Ho.eN m>m 11 NN.NN NN.NN NN.NN mo.mN me.m NN.NH NH.NN oa.eN >n< NN.H NH.H ee.H om.H HH.H No.6 eo.H N:.H enmxsne . NN.NN NN.NN HN.NN NN.HN NN.NN NH. N no.0 NN.NN >om a NH.NN HN.NN NN.NN NN.HN ee.oN en. H NN.N NN.NN s>m nte NN.NN NN.HN No.6e NN.NN NN.NN NH.NN NN.He NH.Ne m>meo11 NN.NN oa.me NN.NN NN.NN NN.NN HN.HN me.He NN.NN >n¢ AH.H ee.H NN.H AN.H HN.H NH.H NH.H mo.H >nmxsn< . NN.Ne HN.NN NN.NN NN.HN NN.NN H .N NN.NH Nmemm >em a NN.NN NN.NN NN.NN an.NN NN.0N o .N NN.NH N .nm s>m nae NN.NN No.2. NN.NOH NoN NN.NN NH.HH HNJH HNNN N>ma1 NN.NN NN.NN Ne.eNH NH.o NN.NN NN.HH Ne.NH NN.NN >n<4 NN.H ON.H 0N.H NH.H NN.H No.6 No.6 No.6 >nm\sq« . om.NN H:.Nm He.Nm NN.NN NN.NH NN.NN NN.NN NN.NN >nm a No. N HN.NN HN.NN NN.NN NN.NH eN.NH NN.NN NN.NN s>m mar No. m NN.NN NN.NN $.13 om.NH moNN NN.NN 3.6m Emil oH.He NN.NN NN.HN NH.NN NN.NH NN.NN NN.NN oo.NN >n¢ NH.H 6H.H No.H NH.H NN.H - oo.H HN.H >nN\>n< . NN.Ne NN.NN HN.ON NN.NoH NN.AH . Na.m NN.NN >om r 8.3 NH.NN NN.NN 3.2. NN.NH .. HTN SNN Em m0 No.5: NN.NN NN.NN NN.HHH AN.ON . ea.n AN.NN m>meoai mm.m¢ NN.NN No.eN NN.HNH NN.HN . sm.m oo.Nm >94 N H , -oH e-oH :OH -OH 113 -OH ARV ozone Hoopeoo Hooeooo >oH eoH eoH one can one peoapoooe .mafiflu 30:38.; Hmvsoemoaoeroc oopofipmo new: 98 Aswan: oaonmgw mspoauom -H e .< 8H eooeooHoeoo mHeHaaN em.eo xs>mv oHoo a.ooo Am>mv woeeoH on one we mooSNHpmNr oopmfipmo new; 255 moocmrnmer Horne oao>ov H950.» Ho mcomwumaaoo .mm 35.3. 81 determination) to 1.0 determined "goodness of fit" to the curves.) Constants for the power curve equations which best fit treatment groups are shown in Table 27. Regression coef- ficients a and b were computed from the formulas _ 2(1n xi)(1n yi) - (gin xi)(Zln yi)/n b — and 2(ln xi)2 - (Zln xi)2/n 1 . l . Because zeros could not be used in the power curve compu- tations, 0.1 values were substituted for 0 values in the data placed into the above equations producing minor but inconsequential flattening of the curves. Power curve r values can be compared with correlation coefficients in Table 25. Table 27 shows that exponent b for most of the power curves was a negative number and that the curves produced "hug" the x and y axes. (The closer the b value is to zero, the flatter the curve.) Such curves confirm the frequency distribution results that the total number of larvae is greater when deve10pment is in either its earliest or latest stages than mid-stage; however, the power curve r values showed significance at the 0.01 level in only two cases (Experiment 11, Control Group 2, and EXperi- :ment 15, DEC 10'5 percent). Data from Experiment 11, Control Group 2 fitted to its power curve is shown in Figure 7. 82 Table 27. Constant values for the power curve (y = axb) for those treatment groups with r-values greater than the r-values for linear regression from Table 25. Treatment 2 Correlation Experiment Group (%) a b r r Coefficient 10 Lev 10‘“ 1.11 -0.66 0.25 0.50 -0.35 10 Control 1 0.81 0.30 0.08 0.28 +0.18 11 Control 1 0.54 —o.26 0.03 0.17 -0.17 11 Control 2 1.69 -0.51 0.21 0.46 -O.32 12 DEC 10"3 6.91 0.05 0.03 0.18 +0.03 12 DEC 10‘“ 1.39 -0.77 0.39 0.63 —0.62 12 DEC 10'5 1.65 —0.23 0.06 0.25 +0.01 12 Lev 10'” 0.79 -0.97 0.49 0.70 -0.52 12 Lev 10"5 2.96 -o.75 0.51 0.71 —0.50 14 DEC 10'3 1.75 -o.72 0.54 0.73 -0.65 14 Lev 10'3 0.21 0.22 0.01 0.09 +0.02 14 Lev 10‘5 0.12 0.57 0.08 0.27 +0.12 14 Control 2 1.75 -0.61 0.29 0.54 —0.45 15 DEC 10'3 0.40 -0.28 0.04 0.19 -0.03 15 DEC 10‘” 0.48 -0.41 0.12 0.35 -0.24 15 DEC 10‘5 0.32 —0.68 0.90 0.95 -0.62 15 Lev 10‘” 0.56 -0.31 0.09 0.30 -0.15 15 Control 2 2.23 -o.35 0.09 0.30 —o.21 83 30 28 24 22 ' 20 18 16 14 ° ° 12 No. of Larvae in Head/Thorax 10 12 14 16 18 20 of larvae in Abdomen CD9 Figure 7. Experimental data from Experiment 11, Con- -0 51 trol Group 2 fitted to the equation = 1.69x ' with O-va ues from the data substitu ed with 0.1- values. 84 (8)Bimodal;ty in mosquito dissections. The only mos- quito dissections performed over time were those done to de- termine the optimum time for mosquito dissection. Tables 4, 9, and 10 show that a dip in total larval numbers oc- curred lasting one to two days after which larval numbers increased and thereafter tapered off slowly. Accompanying this dr0p in total larval numbers was a drOp-off in abdo- minal larval numbers and a rise in head/thoracic larval numbers. All fluctuations occurred about the time when third stage larvae migrate from the Malpighian tubules in- to the thorax and head region and resulting graphs showed a distinctly bimodal distribution of larvae over time. A graph of data from Table 4 is shown in Figure 8. 85 Total no. larvae ----- No. larvae in abdomen ~m-~No. larvae in head/ thorax Larvae No. 8 9 1o 11 12 13 14 15 16 17 Days Figure 8. Average number of D. m larvae in head/ thorax, abdomen, and total body of infected A. tri er'atus (Michigan strain) dissected over t1fie. (Data taken from Table 4.) DISCUSSION Results of Baqxground Infppmation Experiments Feedinngates The results of the membrane feeding experiments in— dicated that satisfactory numbers of mosquitoes could be in- fected with Q. immitis using the artificial feeding device. Overall feeding rates for A. aegypti mosquitoes fed upon the gut membrane in Experiments 1 and 3 were 78.2 and 51.2 per- cent, respectively; feeding rates for A. triseriatus mos- quitoes ranged from an overall rate of 50 percent in Ex- periment 3 before the introduction of the NaturaLamb mem- brane to approximately 75 to 80 percent after the introduc- tion of the membrane in Experiment 8. Mosguito Mortality'aftgpplnfecpipp High mortality rates dropped after switching from the anesthetization method of sorting engorged mosquitoes to the aspiration method. The cumulative mortality rate at three days post-infection for Experiment 1 A. xggyppxgmos- quitoes sorted by anesthetization after receiving a 1:5 concentration of Scotch's blood, was 42.3 percent. The cumulative mortality rate for A. aggyppx infected in Ex- periment with a similar (1:4) concentration of blood but sorted by aspiration was only 3 percent three days after 86 87 the blood meal. As is shown in Table 2, mortality during the first day after infection in those mosquitoes sorted by anesthetization ranged from 21.5 to 41.7 percent for three different concentrations of blood, whereas after sorting by aspiration was introduced in Experiment 3, mor- tality rates during the first day after the blood meal were almost negligible. Effect of Migrofilaxemia Level on Mosqpito Surviyal Before it was realized that the ROCK strain of A. a_gyp— 3; would not support 2. immitis development, a test of three concentrations of blood containing microfilariae was con- ducted over time to determine if, as reported by Duxbury et a1. (1961) and as discussed by Travis (1947) and Weiner and Bradley (1970), high microfilaremia counts are frequent- ly associated with high mortality rates of infected mosqui- toes. Cumulative mortality rates among the three groups receiving the different concentrations of blood were very similar (73.0, 68.5, and 82.6 percent for 1:1. 1:5, and 1:11 concentrations, respectively, at six days after in- fection), but as the microfilariae were not developing in this strain of mosquito, these results are entirely incon- clusive. Susceptibility of Mosquito Stpaips tgAD. immitis Reports of varying Q. immitis susceptibilities in mos- quitoes from different geographical sites are discussed by Intermill (1973) and other authors. Four different strains of A. triseriatus (Alabama, Walton, Ohio, and Michigan) 88 were tested in Experiment 9 for Q. immitis development. De- ve10pment in the four strains was very similar but the Ala- bama and Michigan strains showed deve10pment of most larvae (see Table 10). (These results, however, are based on very few mosquitoes and should not be considered conclusive.) Results of Drug-Feeding Experiments .Epgliminary_00nsiderations Before proceeding with interpretation of the experimen- tal results, thought should be given to some potential drug-filariae cause-effect relationships. For example, as the research was conducted here, if a particular drug or substance were to kill developing Q. immitis larvae, fewer numbers of larvae might be found in that treatment group receiving the highest concentration of that drug. If the drug were to slow down the rate of larval development, more larvae might be found in the Malpighian tubules than in the head/thorax (compared to the control groups), or higher to- tal numbers of larvae because fewer had been lost from the head. Finally, if the drug were to act by speeding up the developmental rate of Q. immitis, a greater proportion of larvae might be found in the head than in the abdominal region, if no larvae had escaped from the proboscis or in any other way been "lost" from the mosquito. To complicate interpretation and analysis, a drug might also have a com- pounding effect; that is, affect both the developmental rate and numbers of developing larvae, or affect just one 89 developmental stage. Although a drug might show greatest effect at its highest concentration, a leveling-off or even inverse effect might also be seen. Experimental Results The Phase One analyses performed on the results of Experiments 10, ll, 12, 14, and 15 indicated statistically significant differences in the effects of DEC compared to levamisole at the three concentrations tested as well as in certain concentrations of these drugs as compared to the control groups. Although the results of the non—orthogonal contrasts performed indicated that the only individual con- centration which differed significantly from the control groups was the 10'3 percent concentration of levamisole, the DEC 10-5 percent concentration would have shown signi- ficance if the harmonic mean of 5.57 had been used as part of the calculations instead of 1. Figures 4 and 5 show that the levamisole treatment groups results are ones which might be expected if levami- sole were to cause inhibition or death of Q. immitis larvae at its highest concentration. The results of the three DEC concentrations, however, indicating fewer numbers of larvae found at the lowest concentration of the drug, are surpris- ing. .As DEC has been considered relatively ineffective ip yixpp against 2. immitis microfilariae, it might be expected to produce no effect or a very minimal effect on developing larvae. Levamisole might be eXpected to have statistically significant effects on developing larvae, 90 especially at higher concentrations, because it is active against filariae Ag yippp and in mosquito test systems. Additionally, as levamisole itself is a more potent drug than identical concentrations of DEC, as demonstrated by comparison studies involving dosed animals and ip_yxppp stu- dies, one might expect that 1evamisole would show signifi- cance at drug concentrations lower than DEC. (NOTE: During this research it was realized that such a pattern for DEC was deve10ping. To eliminate such possible factors as impr0per1y prepared DEC concentrations, pipette contamination, or researcher fatigue during dissect- ing, certain preventative measures, previously discussed in this thesis, such as preparing new solutions of DEC (and levamisole) every two weeks, using new pipettes, and dis- secting mosquitoes randomly, were taken. The DEC results did not change with the implementation of these measures.) All of these expectations are based upon the assump- tion that the Q. immitis developmental pattern for the ex- perimental trials followed that found by other authors. As is shown in Figure 9, H0 et a1. (1974) found that over time the total number of third stage larvae within an infected mosquito p0pulation gradually increased and then dropped off, accompanied by a drop in infective larval numbers found in the head. The numbers of infective larvae found in the abdomen increased after the second stage larval moult and then dr0pped substantially after larvae (supposedly) moved into the thorax and head regions. Although the developmental 91 25 . Total no. larvae ----- No. larvae in abdomen —u—~—No. larvae in thorax wmwmeo. larvae in head 20 Larvae No. Figure 9. Average number of Q. ' itis infective larvae per individual mosquito. (Adapted from a figure on the mean number of Q. ‘ itis in infected A. togoi by Ho et al., 1974.) 92 rate of Q. immitis may vary in different species of mosquitoes, and the ambient temperature at which they are maintained, the usual developmental pattern is approximately that shown in Figure 9. Any deviation from this pattern in laboratory tests would have a bearing on the interpretation of any re- sults obtained. Results of Statistical Analyses The end product of the dissections over time in this research (Figure 8) was a distinctly bimodal distribution of Q. immitis larvae, quite different from that shown in Figure 9. However, it does confirm the results of the other Phase Two analyses (4) through (7): total larval numbers in both individual mosquitoes and mosquito populations were lower mid—cycle than early or late in the developmental cy- cle and that "asynchrony" (real or apparent) in larval de- ve10pment may be a contributing factor. fiypgthesis A--Rea;_asynchrony. As Figure 10 shows, asynchrony in larval deve10pment can occur when one portion of mosquitoes show early larval development and the other portion show late development. Such a pattern explains both demonstrated bimodality in larval numbers over time and the reduction in larval numbers in mosquito popula- Ations dissected at mid—cycle. Although each group of de- ve10ping larvae (early and late) would themselves show distributions like that in Figure 9, lags in time needed for deve10pment and rapid loss of infective larvae within the first day after appearance in the head (35.1 percent Larvae No. 93 Total no. larvae ~_"**No. larvae in head Figure 10. Time Hypothetical distribution of Q. immitis larvae within a total mosquito pOpulation in which a large portion of the population is showing either early or late development of the larvae. 9# loss over a two-day period; Ho et al., 1974), would produce a dip in total larval numbers at the apparent mid-cycle point. fiypothesis B--Apparent asynchrony. Although Hypothesis A does offer an eXplanation for much of the data obtained during this research, early and late larval deve10pment in an apparently uniform mosquito population is improbable. An- other explanation for bimodality, and one which would also explain the reduction in total larval numbers within indivi- dual mosquitoes (rather than mosquito populations) at the mid-cycle point of development (Figure 9), is that asynchrony is apparent rather than real. According to this hypothesis, a two-peak distribution (like that in Figure 8) exists but represents larvae which had migrated to the thorax but were not discovered during dissection. (See Figure 11.) Infec- tive larvae look similar to thoracic muscles and differen— tial staining was not done during dissections so it is pos- sible that a certain segment of the larval pOpulation was not counted. According to this hypothesis, the drOp in the total larval numbers would be most dramatic if and when in- fective larvae made a massive migration into the thorax. Frequency distributions and deve10pmental variances computed from such data would be identical to that obtained if Hypo— thesis A were true. Explanation of experimental results. Whether or not either or both of the above—discussed hypotheses are true, a dip in the number of Q. immitis larvae over time did occur here and numbers of larvae at mid-stage development 95 ----- Actual total no. larvae Total no. larvae found ----------- “No. larvae in head Larvae No. Time Figure 11. Hypothetical distribution of Q. immitis larvae within a total mosquito pOpulation in which a portion of larvae present 1n the thorax is missing from observations. 96 were lower than at early or late-stage deve10pment. The inverse relationship between DEC concentrations and numbers of larvae might therefore be explained if DEC affects the rate of larval development, slowing them down most at the highest concentration of the drug° Such a slowdown would be proportional to the amount of drug given and could re- sult in a distribution of larvae like that shown in Figure 12 in which the least concentrated dose group falls into the "valley" of the graph and the higher concentrations groups (and lesser developed larvae) time-wise precede the smallest concentration group. If DEC does act in the above-described manner, then graphs of dissections performed early or late in the larval developmental cycle should be slightly different from that shown in Figure 12. Evidence supporting the above "slow- down" hypothesis comes from Experiments 4, 6, 13, and 16. Experiment 13 results represent early dissection of mosqui- toes (day 14 after infection in mosquitoes incubated at 72:20F) and show the DEC 10"3 percent concentration with lower numbers of larvae than the two other DEC concentrations (Figure 13). Experiments 4, 6, and 16 represent late'mos- quito dissections and show DEC concentration 10'“ percent ‘with the lowest number of larvae of the three test groups (Figure 14). (Experiments 4 and 6 are considered late dis- sections because, even though dissections were done on post- infection day 12, the mosquitoes from those trials were in- crubated at a higher temperature and almost 80 percent of the 97 DEC 10-4% DEC 10-5% DEC 10' % Control 1 Control 2 CZ 00 Nl-‘UNH II II II II II Larvae No. Time Figure 12. Hypothetical distribution of late second and third stage D. immitis larvae in A, triseria- tus given that DEC slows down developmental rates of the larvae. (The information here comes from Tables 11, 12, 13, 15, and 16.) 98 DEC 10‘3% DEC 10'5% DEC 10’ % Control 1 Control 2 CO NHUNH II II II II II Cl Larvae H No. Time Figure 13. Hypothetical distribution of late second and third stage D. immitis larvae in A. triseria- tus showing how early dissection affects the number of larvae from each DEC treatment group, given that DEC slows down larval deve10pment. (The information here comes from Table 14.) 99 1 = DEC 10:2% 2 = DEC 1.0—5o 3 = DEC 10 % Cl = Control 1 CZ = Control 2 Q) E m .4 6 z: Time Figure 14. Hypothetical distribution of late second and third stage D. immitis larvae in A. triseria- tus showing how late dissection affects the number of larvae from each DEC treatment group, iven that DEC slows down larval deve10pment. The information presented here comes from Tables 5, 7, and 17.) ' 100 larvae counted then were found in the head/thorax, indi- cating that development was well along the way.) At this point one might feel that some indication of which asynchrony hypothesis is responsible for the above DEC pattern could be found by examining the proportion of head/thorax larvae to total body counts for each of the experimental test groups. (C.f. Figures 10 and 11.) If hypothesis A were correct, the pr0portion of head/tho- racic larvae should be higher in mosquitoes dissected just prior to the "dip" and lower in those mosquitoes dissected just after the dip, compared to similar proportions for earlier-dissected mosquitoes. Numbers of larvae should drop in both abdominal and head/thoracic regions over this period of time. If hypothesis B were correct, the propor- tion of head/thoracic larvae should be higher compared to that of earlier dissections. Although numbers of larvae found in the Malpighian tubules might drop, the number of larvae in the head/thoracic region should increase over time. The results of such examinations, however, are gen- erally inconclusive. (See Table 28.) The results of Experiments ll, 12, 14, and 15 support Hypothesis A, in- dicating that the percentage of head/thoracic larvae to ~total body larvae for the DEC 10'5 percent group was gen- erally lower than for the 10'3 or lO'L+ percent treatment groups and that the number of larvae in the head/thoracic and abdominal regions were reduced in the 10"5 percent DEC concentration. However, hypothesis B is also supported by lOl 33 00.0 00 00.0 00 00.0 00 00.3 00 00.0 xaposa\x 00 00.0 00 00.0 00 00.0 00 30.0 03 00.3 2050000 0 0000200 00 00.0 00 03.0 03 00.0 00 03.0 00 00.0 xmnoga\x 00 00.3 00 03.0 00 00.0 00 00.0 00 00.0 20s000< 0 0000:00 00 03.0 00 00.0 30 00.0 33 00.3 00 00.0 xwuo;e\0 00 00.3 00 00.0 03 00.0 00 00.0 00 00.00 cwflowmm m... 00 00.0 33 00.0 00 00.3 00 00.0 00 00.0 xauoze\: 00 00.3 00 00.3 00 00.0 00 00.0 00 00.0 cmflommm 0 0:..- 0 00.0 00 00.0 00 00.0 00 00.0 00 00.3 x00000\: 30 30.0 00 00.0 30 00.0 00 00.0 00 00.0 cmflocwm 0-0 > 0 00.0 30 00.0 00 00.0 0 00.0 - - xmnone\r 30 00.0 00 00.0 00 00.0 00 00.0 - - cmwo0m0 0-0 0 o 00 00.0 00 03.0 00 00.0 3 00.0 0 0 000000\m 00 00.0 00 00.0 00 00.0 00 00.0 000 00.0 cmfi00n< 0 can 3| 00 00.0 30 03.0 30 00.0 0 00.0 00 00.0 x00000\m mm Hm.m 00 03.0 cm 00.N mm wm.0 an 00.0 smacvp¢ 0-00 000 vcmosom 8002 Pamonmm S062 0.080.090 00.00: #2000000 5002 0.080.090 0.0de $3 900.06 ma 3.008.0an0000 10H Psmfiummxm NH 90060000009000 .3” 0.088.00on 0..” Paoawnonxm vcmfiwmne .m0 0008 $0.0 .NH .HH .OH 09:05.06me no.0 msfiwflnomfifi. aw. mo 00000990 0.000.093 >000: EH0 0.00.0080va :0 magma ”g .m mo mwwpsoonmm and 000.05: 00me .mm 000.09 102 the data in Tables 4, 9, and 10, indicating that no larvae were found in the head/thorax of dissected mosquitoes in the days just preceding the "dip." The bimodal model was also examined to explain the levamisole results. Although levamisole behaved in a more "expected" manner than DEC and at first glance appears to act by killing or preventing development of certain numbers of larvae present in dosed mosquitoes, one cannot immedi- ately rule out the possibility that levamisole also acts like DEC, by altering the rates of larval development, with- out first examining the experimental results in more detail. Early and late dissections, however, reveal that unlike DEC, levamisole behaved without regard to time; that is, both an early dissection (Experiment 13) and late dissections (Ex- periments 4, 6, and 16) show that the 10'3 percent levami- sole concentration produced fewest larvae. Although it can- not conclusively be stated that levamisole does Egg affect the rate of development of larvae in dosed mosquitoes, these results suggest another mode of action, such as selective killing of those larvae at an early stage of development. Four out of five experiments shown in Table 28 support this hypothesis. At a given dissection day, fewest numbers of .total larvae and abdominal larvae for all the levamisole ”treatment groups were found inthe 19'3qpercent_group. Alternate Modes of Drug Action Although the eXperimental results discussed here point to DEC's causing a slowdown in Q. immitis larval deve10pment 103 within the test mosquitoes, it is possible that the drug acts in another manner. Other possible explanations for this drug's "inverse" activity are that DEC (1) has a dif— ferent pattern of activity at high vs. low concentrations; (2) affects the mosquito physiology at high concentrations in such a way that more 2. immitis larvae develOp; or (3) affects the nematode physiology at high concentrations in such a way that more larvae develop. (E.g., Coles and Jenkins reported in Rogers and Denham (1976) that reversi- ble paralysis of an adult nematode was observed when the worm had prolonged contact with the test drug levamisole.) SUMMARY AND CONCLUSIONS Overall Statements 1. Mosquitoes could be infected with Q. immitis with an artificial feeding apparatus using diluted blood con- taining microfilariae. 2. The ROCK strain of A. aegypti kept at the Pesti- cide Research Center of Michigan State University would not support the deve10pment of Q. immitis. 3. The Alabama, Walton, Ohio, and Michigan strains of A. triseriatus all allowed complete deve10pment of Q. immitis. 4. Mosquito feeding rates increased by using the Natu- raLamb membrane in the artificial feeding apparatus rather than the laboratory-prepared cow gut membrane. 5. Mortality rates of mosquitoes infected with 2. im- mitig remained quite low (one to two percent per day of in- fection) when mosquitoes were sorted by aspiration rather than by carbon dioxide anesthetization. 6. Daily dissections of infected mosquitoes indicated 'that a drop in the number of larvae observed occurred for a one to two day period shortly after the larvae migrated to- ward the head. 7. It was postulated that reasons for the above— described "dip" in larval numbers might have been either 104 105 asynchronization of larval deve10pment in the mosquito popu- lation as a whole or numbers of larvae present in the thorax not discovered during dissection. 8. The results of Experiments 10, ll, 12, 14, and 15 indicated that fewer number of Q. immitis larvae were pre- sent in mosquitoes dosed with the lowest concentration of DEC (10’5 percent). 9. In view of the "dip" in larval numbers mentioned in (7), above, it was postulated that such results could have been caused by DEC's slowing down the rate of larval de- velopment. 10. Such "inverse" DEC results might also be possible, however, if DEC acted differently at high vs. low concen— trations, or acted upon mosquito or nematode physiology, etc., in such a manner as to promote larval deve10pment. 11. The results of Experiments 10, ll, 12, 14, and 15 also indicated that more D. immitis larvae were present at the lowest concentration of levamisole (10’5 percent). 12. Such results might be possible if levamisole acted upon the nematode by killing or preventing deve10pment of the earlier larval stages. 13. No concentration of either drug completely eli- ixninated Q. immitis larvae in dosed mosquitoes. 14. No statistically significant difference was seen in runnbers of 2. immitis larvae which ultimately develOped in 'the once-fed control group (Control 2) and the twice—fed group (Control 1). 106 15. Although the exact causes for reductions in numbers of Q. immitis larvae deve10ping in infected mosquitoes with- in treatment groups could not be conclusively determined here, the results indicated that both drugs in some way do affect either the rate of deve10pment or numbers of deve10p— ing larvae. implications of ExperimentaAAResgltg The results of these experimental trials raised certain questions which can only be answered by additional study and research. For example: (1) Reductions in larval numbers by as much as 44 percent were observed in certain drug treatment groups al- though complete elimination of larvae was never seen. In- gestion of a single dose of a prophylactic drug possibly could have an effect on the ultimate heartworm transmission rate if mosquitoes contained a small number of infective stage larvae and the drug reduced the number to a subtrans- ‘mission or threshold level, or selectively inhibited either the male or female nematode. A drug-induced delay in the development of larvae could also result in the mosquito obtaining a subsequent blood meal before the larvae had de— ’ velOped completely and could be transmitted to a new dog host. (2) Although reports of resistance to either drug (DEC or levamisole) by Q. immitis was not found in the literature, the question still can be raised whether long—term exposure to small amounts of these drugs may ultimately affect the 107 rate of resistance deve10pment. Several authors (Benz, 1973; Colglazier et al., 1973; Theodorides, 1974) have reported that resistance to chemotherapeutic agents by parasitic ne- matodes does occur. Such resistance generally has deve10ped by exposure of the adult stages of the nematode rather than the immature stages, but drug resistance in Q, immitis de- veloping by exposure in the mosquito is not infeasible. Suggestions for Further Study Neither of the above questions can be answered without additional research and study but, before any such studies are undertaken, it should be kept in mind that such questions regarding reduction of transmission rates and development of drug resistance in D. immitis are only valid if it is known that (1) sufficient numbers of dogs will be taking prophy- lactic drugs and will be dosed at a time of day when biting 'mosquitoes will receive a sufficient dose of the drug, and (2) a number of infected mosquitoes will bite a dosed dog before deve10pment of the nematode is complete. Although the following list of suggested tOpics for re- search is far from complete, it does offer a starting points (1) Repeat the research reported in this thesis, this ltime dissecting treated mosquitoes for all treatment groups over'time to determine conclusively if rates of development are being affected by the drugs. (2) Repeat the same research using blood from drug- dosed dogs rather than mixing the drug with the dog's blood. 108 (3) Determine the probability that an infected mosqui- to will bite a drug-dosed dog before ultimate transmission to a susceptible dog takes place. (4) Determine the average nematode load of infected mosquitoes in nature, to ascertain whether ingestion of the drug(s) would cause a sufficient reduction in larval numbers as to directly prevent transmission of heartworm disease. (5) Determine whether ingestion of the drug(s) by the mosquito in any way affects its later host-finding be- havior. APPENDIX A 109 Because it was virtually impossible to tell whether every mosquito which received the infective blood meal ac— tually ingested one or more microfilariae, only those mos— quitoes which were definitely infected (that is, ones in which Q. immitis larvae could be seen) were used in com- puting mean numbers of larvae per mosquito group, or in the statistical analyses discussed in the main portion of of this thesis. Although such "zero" observations occurred infrequently during the course of the dissections, they are nevertheless listed along with all other individual obser- vations in the following data sheets. Notation of the num- ber of “zero" observations occurring per treatment group is made by placing a number indicating the frequency of the zero observations directly in front of a parenthetical "O". (E.g., 2(0) means that two mosquitoes were found in that particular treatment group in which no larvae could be seen.) ma 00.0 0N.MH 00.00 0 000 00 00.00 :H.NN no.0H m 00 €000 mm mm HMPOB 110 00 00 03 00 00 00 mcowpm>hmmno 00:00>00:0 NH NH aoflvommsH upmom man N H Hoppsoo 0009200 Hohpcoo I 03 u 0.0 n 00.0 0.00 0.0 0 0 00 00 0 00 0 00 00 -00 3-00 own one ARV gnome Pcoermhe .3 psmswnonxm soak Acawnpm smwanowsv mspwwummfinp .fl stcw>0ucfl :0 ow>nwa wdeamfl am mwwpm 000:9 and csoomm mead ho nonssz HI¢ xHszmm< 111 0m:£0#£oo 00 00 0 00 00 00 0 00 30 00 00 30 00 00 00 AoVH 0H mcoHPm>hwmno H050H>H0CH 00 00 I NH NH COHPomMzH npmom ham N H HoppCOO HoppCOD IOH >mH 3m 0H IOH 0H >mH IOH 0mm -00 30000 IOH 0mm Afiv Q5000 PCmEPwmhe .n psmngomxm 800% AsHmupm smmHnoHsv mapmHnmman .«_H050H>H0CH s0 00>00H medssd .Q mmwpm 000:0 0cm 0soomm mpmH mo pcnssz NI< NHDzmmm¢ 112 0000 00 - - - - - - - BMQH I .l I I I .l I mm on.“ I I I I I. I I am 00.30 0.00 - - 0.0 0.00 - - m 0H H - - H H - - z mNN 0H - .- m 0H - - H0089 3 v N 00 0 monvw>0mmno 00000>00c0 - - u - GOHPommcH 00 00 00 00 -0000 000 0 0 -00 -00 -00 -00 -00 0-00 000 00000 0000000 0000000 >00 3>00 >00 000 3000 000 000000000 A003GHP£OUV N|< NHszmm< \ 113 005009000 11 0 m 00 00 0 00 00 00 0H 0H NH 00 0 00 0 0 0 0000 00 00 00 0000 3 0 0 0 0 00 0 0 00 0 0 00 00 0 30 00 0 30 00 3 00 0 . M m. .. m M M g. 0 0 30 0 0 0 0 00 0 0 0 0 00 0 00 00mWWWWWMMmm 00 30 00 00 00 00 00 00 00 00 mMWWM0mmm 0 0 0 0 -00 3-00 -00 -00 3-00 -00 000 00000 HOQPQOO HOnPQOU HOHPCOO HOMPCOO >w.H 5.qu >w.H 0mg 0mm ONO PCwEPMmHB .0 9smeH0omxm 8000 0000090 :wwH:Osz ms9mH0omH09 .d_Hw:0H>H0:H :0 00>00H mH9HssH . m mw09m 000:9 000 000000 m9mH 00 009832 mn¢ xHszmmd 114 00 00 00 30 - 30 00 00 00 00 a 00.0 30.0 00.0 03.0 - 00.0 00.0 30.0 00.0 00.3 mm 00.0 00.0 00.00 00.0 - 03.0 00.0 00.0 30.0 00.0 am 00.00 00.00 00.00 00.00 0.0 00.0 00.3 00.00 00.0 00.00 m 00 00 00 0 0 0 3 0 0 0 z 000 300 000 00 0 30 00 00 0 00 00000 0000 00 00 00 0 0 0 00 30 0 0 0 0 0000 0 00 0 HH mCOflPMPHmeO 0 00 00000>0000 00 30 00 00 00 00 00 00 00 00 mMWWM0mmm 0 0 0 0 0-00 3-00 0-00 0-00 3-00 0-00 000 00000 080000 00.00000 00.00080 00.00000 >00 >00 >00 00.0 000 0000 000000000. 00000000000 0-0 x0mzmmm< 115 UmSCHPCOO 00 NH m m 0 NH 0N m0 0 Hm m0 NH ON mGOHvdzmeO Hm HMSfiw>wccH 000000000 :0 u u u n a I n u9mom awn 0 0 -00 -00 -00 -00 -00 M-00 33 00000 00.3000 00.5000 >00 3>00 >00 000 3000 0000 000000000 .n 9cmafihmmxm Scum A00009m ammfizowsv 03900000009 .d adscfl>wcza :0 00>000 0090580 .0 0w09m 00059 000 uncomm 0900 mo nonadz 31¢ NHDzmmmd 116 i' 0r - H0. 00 - - - - - .. - 000 Adm-0H I I. I I. .l I. I mm 00.0 - - - - - - - 00 03.00 - - - - - - - m 00 - - - - - - - z 000 - - - - - - - 00000 onm 00 30 00 00 0 000090>hmmpo m 00000>0000 :009ommCH :0 I I I I I I I I9mom 009 N 0 -00 -00 -00 -00 -00 -00 000 00000 0000000 0000000 >00 3>00 >00 000 3000 000 000000000 AcmSQHPQoov :Id 03szan 117 00 30 30 00 00 00 00 - 00 0 0000 30.3 00.0 00.0 00.0 00.0 03.0 30.0 - 00.0 00.0 00 00.00 00.0 00.0 00.0 00.3 00.00 03.3 - 00.0 00.0 00 00.00 00.00 00.00 00.0 00.00 .00.00 03.0 - 00.0 00.0 M 0 00 00 00 0 0 00 - 0 0 z 000 000 000 000 00 30 00 - 0 00 00000 00 0 0 00 0000 0 0 0000 00 0 0 0 0 00 00 0 0 00 0 0 0m 0 0 30 00 00 00 00 00 00 0 0 3 0 03 00 00 0 00 m 0 0. 0 3 00 00 0 0 0 0 00 0 00 00 m0 0 0 m 0 30 0 00 00 0 0 00 000000>00000 30 m0 m0 0 0 0 0 - 0 00 00000>0000 20.0 om C 00 00 00 00 00 00 00 - 00 00 -00w0 M0m N 0 0 0 -00 3-00 -00 -00 3-00 -00 000 00000 0000000 0000000 0000000 0000000 >00 >00 >00 000 000 0.00 000000000 .00 98080000xm 8000 0800090 0809000080 00 00>000 0090880 .m 0w09m 00009 080 080000 0900 mo 009852 mI¢ XHszmmd 118 005809800 m m 00 m0 0 0 m 3 0000 m0 0 00 m 0 00 00 m 30 0 0 m .0. 3 m 0 m0 0 0000 m m 0 00 m m m 0 30 00 00000 0 00 m0 0 00 0 00 3 0 3 30 onm 0 0 00 00 m 0 m0 3 00 00 mm 3 m 0 0 m0 0 m 0 N 0000 0 0 00 0 00 m m 0 3 00 0 m 3 N0 0 00 m0 00 m 0 00 00 00 00 m 0 000000>00000 3 N 0 0 m 0 3 0 00000>0000 SCH ow C 00 00 00 00 00 00 00 00 -00w0 m0m m 0 -00 -00 -00 -00 -00 -00 000 00000 0000000 0000000 >00 3>00 >00 000 3000 000 000000000 .00 900800mgxm 8000 0000090 080p00¢v 05900000009 .0 00500>0000 80 00>000 0090880 .m.0w090 00009 000 080000 0900 00 009852 ©I< xHszmmd 119 0|||||0w r0 . . . Jr 005809800 00 m m N. m x: 0 m w 0 m N. 00 3 mm 0 mm onm m. w 0 m 0.0 m 0 0m 30 30 0 0 OH +1“ wCOHPMEmeO 30 00 0000000000 COHPO¢.HCH 00 00 00 00 00 00 00 00 -0000 000 0 0 -00 -00 00 . 0000 0000000 0000000 >00 3000 0M00 mmm 3mmm mmm me00000w A005809800v 01¢ NHszmmd 120 fiilfi_ a 0. J 003000200 mm 0 m0 0 m0 0: m 0 m m 00 00 H mcowpmbpmmpo 00 0000000000 00 00 00 000000000 ma ma ma 0H ma upmom ham 0 0 -00 -00 -00 00 00 0000 0000000 0000000 000 0000 000 000 0w00 mmmm hmw00000w Aumscflvcoov 03¢ xHszmmd 121 0 00 0m 00 on on 00 00 0000 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 mm 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00 0.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 m 00 00 0 0 0 0 00 00 z 000 000 00 00 00 00 000 00 00000 00000 N mfiowpmxrhmwflo 0 0000000000 .0 00 0 a a 00 2 2 000000000 0 0 -00 0-00 0-00 - 0-00 0-00 000 00000 0000000 0000000 000 000 000 000 000 000 000000000 AUmSQHPQoov 01¢ NHszmm< 003:00000 122 0 00 om 00 0000 00 00 d m m m 0 0m 0000 00 00 0 0m 0 mm 00 m0 0 0000 0 0 d 0 00V: 0 0 :0 0 0 0 0 00 m 0 00 N0 00 00 m 0 0 m 00 00 00 00 0 0 0 0 m N 0 0 0 m0 0 00 0 000000000000 0 mm 00 m 0 0 0 00 0000000000 COH om C 00 00 00 00 00 00 00 00 -00w0 M0m 0 0 -00 -00 -00 -00 -00 0-00 000 00000 0000000 0000000 000 0000 000 000 0000 000 000000000 .N0 #:08000mxm 800m Am0whpm 080900¢V 05000000000 .0 00:00>00c0 00 00>000 0000820 .Q 00000 00020 000 080000 0000 00 009832 ml¢ xHszmm< 123 Adm-in 00 00 mm 00 00 00 00 00 Agva 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00 00.0 00.0 00.00 00.0 00.0 00.0 m0.m mm.0 00 00.00 00.00 m0.00 00.0 00.0 00.0 00.0 00.00 m 00 00 0 0 m 00 0 0 z 000 000 00 00 mm an mm 00 00000 0000 m 0 m 0 A0 :0 mflowpmimmpo 00 00 0000000000 00 00 00 00 00 00 00 00 mMWWM0mmm m 0 -00 0-00 -00 -00 0-00 m-00 000 00000 0000000 0000000 000 000 000 000 000 000 000000000 A603000Qoov mid NHszmmd 124 000000000 m 0 0 m m m m 0 00 m 00 m 00 m m w 00 00 00 m0 0 0 m0 m m m 0 m 0 00 0 m 0 0 00 m 00 m 00 0 0 0 0 0 03,0. 0 0 0 0 m0 m m0 m u :0 00 m m w m m 0 m 00 m 00 0 00 00 m0 0 0 0 000000000000 0 0 m 0 0 0 N0 00 0000000000 COH ow C 00 00 00 00 00 00 00 00 -00w0 m0m m 0 -00 -00 -00 -00 -00 -00 000 00000 0000000 0000000 000 0000 000 000 0000 000 000000000 .00 0000000000 0000 0000000 00000000 00000000000 .0 0000000000 00 000000 0000880 .m.0m000 00000 000 000000 0000 mo 00980z w|¢ NHszmmd 125 00 00 00 00 mm 00 0 00 0000 00.0 00.0 00.0 00.0 00.0 00.0 00.0 m0.0 mm 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 m 0m 00 00 0 0 00 00 00 z 000 000 00 on 00 000 00 00 00000 0000 00 m 0 0 0000 00 w 00 0 m w 0000 m00000>00mno 00 0 0 0000000000 000000000 00 00 00 00 00 00 00 00 -0000_000 m 0 -00 -00 -00 -00 0-00 -00 000 00000 0000000 0000000 000 0000 000 000 000 000 000000000 00000000000 0-0 mezmmm< 126 Umschsoo IJ :H 0 HH mH 0H m HovH N o m m 0H m :H OH H Hm m o m H mm mH Hovm NH 3 m HH ovm H 0H H o H m mH m a m m m o H mH m NH HH m a 0 HH mH m m :H om NH NH m OH m chHpm>pmmno m NH mH H m :H m om Hmst>HucH SCH om C 0H 0H 0H wH QH wH mH 0H -pwwm mam m H -OH OH IOH -OH -OH -OH ARV axouo Houpcoo Hoppcoo >mH a>mH >mH 0mm :omd 0mm pcmEHmmpa .dH PaweHngxm scum AmHmnHm wempwHHUEH CH mm>th mHPHSEH .m wwwpm upHAH cam Uncomm meH mo umnezz m|< NHszmm< 127 HH HH H: «H mm um Hmvn mH.~ mH.H om.m mm.H Hm.H mo.~ mm mo.m HN.H mm.m um.m mo.m mm.m am mn.~H mm.0H oo.m mm.m m:.u om.m .m 3H NH m w m m z mmH :mH m: on mm a: Hmpoe HovH NH 0H onm :H mGOHHw>uomno HN m HasUH>HocH 0H wH wH 0H mH 0H mmMWMommm m H -OH -OH -OH -OH Hgv msouu Hahvnoo Honvcoo >mH 3>mg >09 0mm HCosvdmne AUmSSHPCoov 01¢ NHszmm¢ 128 vmchpcOo N O H OH AOVH N HOVH : O H N O HOON N m AOVH m N O H a m HH N H O OH O N NH N NH N OH O ON OH N H O O N O O O OH OH H N NH OH H O O O H O H H OH H H O a N NH N O N O N H O O a O H H H m n O O O OH O N m N N m N O N NH O OH N NH OOOHHO>OOOOO OH H O O N O O O HOOOH>HOcH SCH 00 S OH OH OH OH OH OH OH OH -HOWO MOW N H -OH H-OH -OH -OH O-OH -OH HOV OonO Hoppcoo HOHHOOO >OH >OH >OH Own ONO Oma pcmspmmpa .mH pameHummxm scum Amean newanHUcH CH mm>pmH meHaaH .9 depm vanp and uncomm mva Ho nmnabz OHI< NHDzmmmd 129 NN ON on ON OH NH NN ON AOOO NO.N OH.H NH.H HO.H ON.O OO.O OO.H HN.H mm OH.N OO.N N0.0 NO.O NH.O NH.O OH.O o:.O Om OO.O OO.N O.N OO.O OH.H H0.0 OO.O N0.0 m O ON HH OH OH OH NH OH 2 OO HHN NN ON NO OO NO ONH HNPOO AOVO HH O OH mH O N: O 41‘ H H N: O OH H AOOH OH OOOHHO>OOOOO N O OH HOOOH>HOOH CoHHomHCH OH OH OH OH OH OH OH OH upmom ham N H uOH -OH -OH -OH -OH OH HOV OOOOO HONHOOO HoppcOO >OH O>OH >OH Own sown mama pcmepmmpe AUmSEHPGoov OHI¢ NHszmm¢ 130 NH ON ON ON H: OO NO OO HOOO ON.O HO.O OO.O OO.O OO.O OO.H OO.H NO.N mm NO.H OO.N OO.N NN.N NN.N OO.N OO.N OO.O Om O0.0 OO.N NO.O O.O OO.N OO.N ON.N OO.O m O O O O O O O O z OO ON NN NO OH NH HH NO HOHOO AOOO HOVO AOON H O H AOOO N AOON H H N O N N H H HOV: O O O O O H HOVO O O N O O N O H O O O O N N N H OH O N H O N O O O OOOHOO>OOOOO O H N O H N H H HOOOH>HOOH NH NH NH NH NH NH NH NH WMMWMOMMM N H -OH -OH -OH -OH -OH JI‘-OH HOV Osopo HOOHOOO HoppcOO >OH O>OH >OH Omo OOmO Own H:OEHOOOO .OH vamermmxm scum AmHmnvm wemanHccH CH mm>HmH mHHHEeH .Q mmmpm OHHQP wad Uncomw mpmH mo pmnsdz HHI< NHszmm¢ APPENDIX B 131 APPENDIX B-l Number of late second and third stage 9. immitis larvae in g. triseriatus (Alabama strain) receiving only the infective blood meal (Control 2). Data is from Experiment 8 and is grouped according to location of dissection. Day Post- Infection 12 13 15 16 Larvae Location Abd H/T Abd H/T Abd H/T Abd H/T Individual 5 O 3 O 0 2 O 7 Observations 2 O 3 O O l O l 1 O O 12 O 8 O 2 2 O 6 O O 1 O 11 l O 7 O 1 l O 1 1 0 l6 1 O 3 O 2 2 0 O 2 O 1 1 O 4 10 O 2 2 O 3 10 3 2 3 O 3 4 15 O 2 O O 2 3 0 1 O 2 O 2 0 Total in Abd 29 hh 12 18 Total in H/T O 13 4# 29 N 14 6 12 10 i in Abdomen 2.07 7.33 1.00 1.80 i in H/Thorax O 2.17 3.67 2.90 % in Abdomen 100 77 21 38 % in H/Thorax o 23 79 62 COI‘I‘. COGf. "" +002“ +0.67 “0.32 132 OwschcOo OH O N O m o mH N OH O N O OH O HH 0 MH n O OH OH O OH O NH O NN O O N w m OH H m n O O O NN O O O H OH O O OH NN O O OH O OH O NH N O HN O ON H N O O OH HH H O OH O O O OH O ON N O NH O O O O OH ON O O H O O O OH O OH O N H O N O O ON OOOHHO>OOOOO N H O N O HH O O O OH OH H NH O O O O NN HOOOH>HOOH O\O OOO H\O OOO O\O OOO O\O OOO O\O OOO O\O OOO a\: O94 e\: OOO O\O OOO cowwwmmm QOH 0m S OH OH NH OH NH OH OH OH NH -HOWO mam :OOHOOH: OHOO OHOO sopHmz OOHHOO OOOOOHO OOOOOHO OOOOOHO OOOOOHO opmwmmwm .GOHPommmHO Ho COHPOOOH op wcHOnooom amazonm mam mw>th wad a Hamanmnxm sopHIMH «HOD mum xHszmm¢ .Hmme cooHnlm>HPommcH map OHco wCH>Hmoon mapmHHmmeP .¢ :H mm>an mHHHEEH .Q mmwpm OHHSP and vcoomm OPMH Ho nonezz 133 N0.0- . u - O0.0- O0.0+ O0.0- ON.O+ u .mmoo .nnoo OO O O O O OO HO OO O nanosa\x :H O OH OOH OOH OOH NO N OH OO OOH :OaOO94 :H m ON.NH O O O OO.O O0.0H OO.HH O0.0 O «Oposa\x cH H OH.N ON.N OO.HH OO.O OO.OH OO.O ON.N OO.O ON.HN :oOOO94 :H x O O O O O OH OH O O 2 OO O O O N NON OHH OO O‘ a\O :H Haves NH HO OO OO OO OH NN OO OO O94 OH Haves OH H OH O OH 0 mcova>homno O O . HOOOHpHOcH _ .SOHHmooH a\r O94 a\O O94 O\O O94 O\O O94 a\r O94 O\O O94 a\: O94 e\x O94 H\O O94 Om>uaH ll 00 OH OH NH OH NH OH OH OH NH mmwwm mum g3 OH . SHNHPW . . .OO OEO 83a: 83a: mafia? «5934 «532 c.5934 oszvmos AvmficHPCOUV Nlm NHQZHNAd 13# Omfifiwwcoo l\ :- d'r-IOOOOu-H‘NO r-l OQd’OCDBWOdM I-i HQBOOBQOV—IN H H H H H HHNNd’UN‘nOChQ OOOCdenaavnnbOo N WMOSHWNBQH‘O NNHCDOMHHON Hm MOQHQ‘Hé‘Ofi’NO \OHOHOO HOBNN: OHMOMOu—IHMHQ 000 HH“ 0:}: \O mGOHp nw>hompo H636 IH>HOCH O\O O94 O\O O94 O\O O94 B\O O94 O\O O94 O\O O94 e\z O94 O\O O94 a\O O94 :oHpmooH ow>hmq OH NH OH OH OH OH OH OH OH OOHpomacH -pmom Own N HoHOCoo N N Hoppcoo Hohvcoo H Hoppcoo IOH >mH IOH O>0H IOH >oH IOH 0mm -OH Oomn IOH omfi ARV nacho Hamakmwna .GOHpommmHO Mo 20HpmooH on wnHOuooom Ommaonw .OH Pamanognm Bonn AnHwnpm dauan¢v mzpanmman («.2H mw>th mHHHaHH am_mwwpm OAan ch Onoowm mHOH Ho uwnasz mum NHszmmd 135 OH.O- NO.O- OH.O- OH.O+ OO.O- OO.O- NH.O- . OO.O- .mmmwo NO HO OO ON OH NH ON O ON O\O :H O O OO HO ON NO OO OO OOH HN O94 :H O OO.HH OH.N O0.0 OO.N OO.N OO.H O0.0 O NO.N a\r :H x OO.H O0.0 ON.O O0.0 O0.0H O0.0 HO.H OO.N N0.0 O94 :H x O OH OH OH O O HH O O z ManoHSKm NOH HN. mO Om OHH m on o O OH. Hdube H8809; O OO NO OOH OO OO HN N ON :H Hmpoa O NH 93H». m CH ldzmmpc O O HOOO N N uH>HOOHH GOHPwooH O\O O94 O\O O94 O\O O94 O\O O94 a\O O94 a\: O94 a\O O94 axm O94 O\O O94 a\: O94 OO>nOH OH NH OH OH OH OH OH OH OH mmwwmmmmm N N N H -OH O-OH -OH -OH -OH HOV Osage Houpcoo Honvaoo Hopvcoo HoHEoo >mH NEH >mH can one pawn—Pumps A6992...” Psoov mum NHsznHmd 136 OOSCHPCoo OH O O O O HH O O O O N OH OH N O O O OH H O O O O O O N O O N N O H HH O O O O H O N O O O OH O N N O O O O N O OH O NH O O N O O O O N N OH O N O OH O O O O O O NH N O N O H O HH O O O O H O O O O O O O O OH OH O O O O O O H O O H HH O O O O O O O O O O O O O O H OH O N O O O O H O O O O H O O O OH O N O O N N O NH N N O OH O OH O NH O O N O OH O HN O N OH O N O H O:OH..:§.$O9O N N O N O O O N N O O H O O O H HOOOH>HOcH a\r O94 e\O O94 O\O O94 O\O O94 O\O O94 a\O O94 a\r O94 O\O O94 :Owwmmmm HHOH ow 3 NH OH NH OH OH OH OH OH upmwm MOW N H -OH -OH -OH -OH -OH -OH AOO OsopO Hoppcoo Hoppcoo >OH O>OH >OH Own Oomn Own pcoapmmpa .CoHvoommHO Ho :prmooH op MCHOnooom Ommsoum .HH anaHungm scum acHwan wewpwHhmH mHvHaEH am depm OanP Una Ocoomm waH Ho umnabz dim NHszmm¢ 137 cmschcoo h 0 HH O o O NN H O N NN O O m N O H O o N O O O O mH O O ON O O ON O H O H O OH O HH O O H H O H O OH O O H NH O H o H OGonw>ummno N H N N HOOOH>HO2H a\r O94 O\O O94 a\O O94 O\O O94 O\O O94 a\x O94 a\O O94 a\z O94 cowwmmmm NH OH NH OH OH OH OH OH :OHpommcH upwom Own N H -OH -OH -OH -OH -OH -OH Amy OzouO HO.HPHHO U HO.HPHHO U >$~H 3>0_H >w.H QB 308 US PgGED-NGLFH HOongpnooO Osm NHOzmmm4 138 Omscwwcoo O O H O O H N O O O O OH OH O O H OH N O H O N O O N O O O O O N O N N m M ERGO szmeo H H HOOOH>HOcH :ovaooH O\O O94 a\O O94 O\O O94 O\O O94 a\O O94 O\O O94 a\O O94 O\O O94 OO>OOH COHPoomcH IV nacho N H -OH -OH -OH -OH -OH -OH HOV HO.HPGOO HO.HPGOU >09 .3er >QH DB 308 DB Pflmfiwwha AOmSCHPCooV Oam NHszmm¢ 139 NO.O- NH.O- OH.O- ON.O- HN.O- HO.O+ OO.O+ NO.O+ .Omoo .OOOO HO OH OO OH OO O O O xmgona\r :H O OO NO OO NO OO NO OO NO :OaOO94 :H O O0.0 OO.H HH.O ON.H ON.O NN.O ON.O NO.O OOO0O9\O :H m OH.O OO.O NN.O NO.O ON.O OO.O OO.N OO.N cmeOO94 :H O OO OO O N O O OH NH 2 OOO OO NO O OH N O O a\: :H HOOOO OOH OOH NO OO OH NO OOH HO O94 :H Hapoe H H O N MN 0 O O O NH 0 O onHpN>ummno N OH HOOOH>HOcH O\O O94 e\m O94 O\O O94 H\O O94 O\O O94 O\O O94 e\m O94 9\O O94 cowwmmmw SCH Ow Q NH OH NH OH OH OH OH OH -OOWO mam N H -OH 1‘ -OH -OH -OH O.OH -OH HOV OOOOO Houpsoo Hoppcoo >OH O>OH >OH ONO OOO Omn Psmspmmpe AUmSQH Paoov dim NHsznHmO. Omschcoo 140 O O N O OH OH H OH N H OH N OH N O O N W O N N O O O OH OH OH O N O O O NN N H O HN N O ON OH O N N O H O H O O O O O O O O O O N N O O O O O O O O O O O H O N N O N OH OH O O O O O O O O N O O HH N OH H O NN O O O H O N O O O O H H O O N O O O OH O O O OH H O O OOOHHO>OOO9O O O O NN ON O N O O O O N O O OH O HOOOH>HOcH O\O O94 O\O O94 e\: O94 B\O O94 O\O O94 O\O O94 O\O O94 O\O O94 cowwmmmw HHOH 00 5 OH OH OH OH OH OH OH OH -OOWO mam N H -OH O..OH -OH -OH O..OH OIOH HOV Osogo HOOOOOO HoppcOO >OH >OH >OH OOO OOO ONO Ozmepmmpe .COHHommmHO Mo :oHPNooH op wnHOhooom Ummsonw .NH PGmEHngNm scum Aanan madandv mSPprmman .O QH mm>hmH wHPHEEH 2m mmmpm OMHAP dam Ozoomm mymH Ho amnesz mum NHszmm< 141 OH.O+ OO.O- OO.O- NO.O- NO.O- HO.O+ NO.O- OO.O+ .mmoo .OOOO NO OO OO HO ON OO OO ON OOO0OE\O OH O OO OO OO OO ON OO HH ON OOOOO94 OH O ON.O HN.O OO.O OO.O OO.O OH.O O0.0 NO.N 582?: OH N. O0.0 NO.N OO.O OO.O OO.H ON.H NO.O NO.N OOOOO94 OH.N OH OH O O O HH O O z OHH OO OO NO ON OO OO OO a\O OH Haves OO OO HO ON O OH O OH O94 OH Haves O O O O N O muopr>nmmno O O HOOOH>HOOH O\O OO4 O\O O94 O\O O94 O\O O94 O\O O94 a\O O94 O\O O94 O\O O94 Oowwmmmm GOHPom CH OH OH OH OH OH OH OH OH -vmom an N H -OH O-OH -OH -OH O.OH -OH HOV OOoOO HOOOOOO HoOpOOO >OH >OH >OH OOO OOO OOO pOoOOOOOs AUmSQHPGoov mum xHszmmd 142 OmszHPCoo O OH O O O OH O O O O O O O OH O O O O O O O O O O O O O O O O H O HH O O O OH O O O OH O O O O O O O OH O OH O OH O OH O O O N O OH O O O N N N O O O O O H O OH O O O O O O O O O HH O O O HH O O O OH O O O O O N O O O O O O O O O O O N O OH O O O OH O O O N O OH O HH O O O O O O O O O O O O O O O OH O O O NH O H O OH O OH O OH O O O O O O OOOHOO>OOO9O O O O O O O O N O O O O O NH O OH HOOOH>HOOH O\O O94 e\m O94 O\m O94 O\O O94 e\O O94 e\m O94 N\O O94 e\m O94 Oowwmmmm :oHpommcH 3H 3H OHH OOH OHH OHH 3H 3H IPmom th N H -OH O.OH -OH -OH O-OH -OH HOV OOOOO HOO9OOO HOOPOOO >OH >OH >OH OOO OOO OOO OOOOPOOOO .QOHpommmHO mo CoHPOOOH op mQHuuoooN Ommsopm .OH PQmEHmexm Eopm Achnpm madandv mSPmHnOwHup .O CH mm>NmH Olm xHszmm< mHPHesH .m mmmpm OMHOO wad Onoomm OPOH Mo amnesz 143 OH.O+ .Hmoo .nuoo O O O O O N O O OOOoOa\O OH O OOH OOH OOH OOH OOH OO OOH OOH OOOOO94 OH O O O O O O OH.O O .O OOOoOa\O OH.N O0.0 OO.O OO.O OO.N OO.N N0.0 HO.N. ON.N OOOOO94 OH.N HN NH OH O O OH HH OH z O O O O O N O O a\O OH Haves OOH NHH OO OO OO OHH NO NN O94 OH Haves O OH O O o O monvw>pmmpo O O HOOOH>HOOH SCH woo O\O O94 O\O O94 H\O O94 O\O O94 O\O O94 O\O O94 H\O O94 e\O O94 Ow>nmm ll! SCH om .OH OH OH OH OH OH OH OH upmwm MM N H -OH O-OH -OH -OH O-OH -OH HOV OOOOO HOOOOOO HOOOOOO >oH >OH >OH OOO OOO OOO HOoOpOOOa AcmSSHPSoov Olm NHszmm4 144 Omsszcoo O NN O N O OH O O O HH O OH O OH O O O N O O O O O O O OH O N O HN O OH O H O OH O O O O O O O N O ON O OH O NH O O O O O O O N O OH O H O O O H O O O OH H O O O O O O N O O N O H O O OH O O NH O N O O N O O O O O OH O OH O O O H O OH OH N N O HH O H O O O O O OOOHPO>OOO9O O O N OH O OH N O O O O OH O O O ON HOOOH>HOOH H\O O94 H\m O94 O\O O94 H\O O94 H\O Op4 O\O O94 O\O O94 O\O O94 Oowwmmmm SCH 0m HM OH OH OH OH OH OH OH OH -OOWO mam N H -OH O-OH -OH -OH O..OH -OH HOV OOoOO HoOpOOO HOOPOOO >OH >OH >OH OOO OOO OOO OOOOOOOOO .coHPommmHO ho cowvmooH Op w GHOnooom Ommsopm .OH Pawanmmxm Scum Achnpm OEMQOHOV { msvanmenP .< :H mm>an mHPHafiH em wwwpm OHHOP Una Ozooom mpmH no 909852 mum xHszmm< 145 O0.0- N0.0- NH.O+ O0.0- N0.0+ O0.0+ OH.O- OOOm- .HOOO .OOOO ON OH OO OO ON OH OH ON OOO0OO\O OH O NN NO ON OO ON OO NO ON OOOOO94 OH O OO.O HO.H OO.N NO.O OO.H OO.H OO.H oO.N OOOoOH\O OH N ON.O HO.O OO.O NO.O N0.0 O0.0 HN.O OO.N OmeOO94 OH M OH NH O O N OH N O 2 NO ON OH NN OH OH OH NH O\: OH HOOOO NOH OOH OO ON OO OO OO NO O94 OH Hmpoe O NH NH O mcoHPm>hmmno O OH HOOOH>HOOH H\O O94 H\O O94 O\O O94 H\O O34 H\O O94 O\O O94 O\O O94 H\O O94 Oowmmmmm OH OH OH OH OH OH OH OH mmwmemmm N H -OH O-OH -OH -OH O-OH O.OH OOO OOOOO HOOPOOO HOOOOOO >OH >OH >OH OOO OOO OOO HOOOOOOOH AUmSSHPSoov mum NHszmm¢ 146 Omchvzoo O O O H o H O O OO N O OH O O O O O OH O H O O O O O O O N O M O N O H H O O O O O OH o O o N o O o H H O O O O HH O N O NH O O O O O O O O O N N O O H N O O OH O O O OH O N O O O O O O O O H H H N O O O N O O O OH O H N O O O O O O H O O O O O O O O O O O H O H O OH O O O H H O O O O N O NH O N O N O N O O N O O H O O O O O O O N O H O O O O O O O O O O O O H O O OH O O O N H N O N O N O O O N O O O O O O N N O OH N O O NH OOOHOO>OOO9O OH N H O O O O O O N O O O O O O HOOOH>HOOH O\O O94 O\O O94 O\O O94 B\O O94 H\O O94 H\O O94 a\O OO4 O\O O94 Oowwmmmw SOH ow OH OH OH OH OH OH OH OH #me mum N H -OH O-OH -OH -OH O-OH OIOH OOO OOOOO Honvnoo HoppCoo >mH >mH >mH can 0mm 0mm anapwmna .GOHWomNMHO mo :OmpmooH 0P mcmcuooow Omnsouw .OH vamanmnKm scum Achqu dauandv mzpanoman .< :H mw>hmH mHPHaEH mm depw Oanp and Onocmm meH Ho nopauz mum NHszmg 147 HN.O- OO.O+ NO.O- OH.O- OH.O- NO.O- ON.O- OO.O- .Nooo .OOOO OO OO OO OH O O ON NH OOOOOO\O OH O OO OO OO HO OO OO OO OO OOOOO94 OH O O0.0 HO.N OO.N OH.H ON.O HO.O OO.H HH.H OOOoOaxz OH.N OO.O OO.O OO.O NO.O OO.O OO.O OO.O HN.O 3834 OH O O ON HH OH OH OH NH OH 2 OO ON NN OH O O OH HN H\O OH Haves OO OOH OO OO OO OO OO OO O94 OH Haves O H . N O O N O O O OH O O O O o O mcoprOrnmmpo O O HOOOH>HOOH O\O O94 O\O O94 O\O O94 a\O O94 a\O O94 O\O O94 O\O O94 H\O O94 Oowwnmmm OH OH OH OH OH OH OH OH mmwmwfim N H -OH O-OH -OH -OH O.OH O-OH OOO OOOOO Honpcoo HoppCoo >mH >mH >mH Can can Can Pamfimone AUmHEHpsoov mam NHQZMOEE 148 O0.0- ON.O+ ON.O- O0.0- O0.0- . OH.O- N0.0- .Nmoo .OOOO OO NN OO OO OH O NO NO OOOoOe\O OH O OO ON OO OO OO OOH OH OO OOOOO94 OH O NH.O OO.O OO.H OO.H OO.O O ON.N OO.O OOOOOO\O OH x OO.N ON.N OO.N OO.N OO.N OO.N OO.O OO.N OOOOO94 OH x O O O O O O O O 2 ON O O HH O O O ON H\O OH HOOOH OH OH OH HN OH NH N NH O94 OH Hmpoe O H O O O H O N H O O H N O O O H O N O H O H O N O O O O O O O O O H O O O O m m O O O m O N O O O O O O O O O N O O N O N O H OH H _ O O O N O H N O N O O O O H O O OOOHOO>OOO9O O O O H O N O O O H O N H O H O HOOOH>HOOH O\O O94 Oxm O94 O\O O94 e\: OO4 O\O O94 B\O O94 O\O O94 O\O O94 Oowwmmmm NH NH NH NH NH NH NH NH mmwmemmm N H -OH OH -OH -OH O OH -OH OOO OsoOO HOOOOOO HOOOOOO >OH O>wH >OH OOO OOO OOO OOOOPOOOO 1N .COHPommmHO mo COHpNooH op mcHOnooom OWQSopw .OH meEHnwmxm Eogm Achhpm meaanOV mszHpmmHmp .< CH mm>HNH meHesH .Q mmmpm ONHOH cam Ozoowm mPNH Mo gmnesz mum mezmmm¢ 149 For the reasons presented in Appendix A, the statis- tical analyses discussed in the main body of this thesis were calculated using only non-zero observations. The re- sults of the analysis of variance and orthogonal contrasts using all the data (including zeros) are presented in this appendix. (Results are essentially identical to those dis- cussed in the text but show significance at a lower level.) APPENDIX C 150 APPENDIX C—l Analysis of variance of the results from Experiments 10, ll, 12, 14, and 15 using all the data (including zeros). Source of Level of Variation df Mean Square F-value Significance Weeks (adjusted 4 255.18 5.76 for treatment) 0'01’ df4,27 Treatment (ad- justed for 7 107.77 2.43 0.05, df7 27 weeks) ' Weeks*Treatment (adjusted for both weeks and 27 44°30 1-14 NS treatment) Error 512 39.02 _ TOTAL 550 _ _ 151 APPENDIX C—2 Orthogonal analysis (seven contrasts) performed on the results of Experiments 10, 11, 12, 14, and 15 using all the data (including zeros). Source of Level of Variation df Mean Square F-value Significance Contrast l 1 26.8898 0.61 NS Contrast 2 1 347.2429 7.84 0.01, dfl,27 Contrast 3 1 13.8664 0.31 NS Contrast 4 1 5.5103 0.12 NS Contrast 5 1 36.4316 0.82 NS Contrast 6 1 216.8264 4.89 0.05, df1,27 Contrast 7 1 117.8109 2.66 NS APPENDIX D 152 APPENDIX D—l Calculations for the non-orthogonal analysis performed on the results of Experiments 10, ll, 12, 14, and 15, shown in Table 23. Contrasts Category Weights C1 DEC vs. Control 1 1 l l 0 0 0 -3 0 C2 Lev vs. Control 1 0 0 0 l 1 1 -3 0 C3 DEC 10’3% vs. Control 1 l o o o o o -1 0 C4 DEC 10—4% vs. Control 1 0 1 0 0 0 0 —l 0 C5 DEC lo'5% vs. Control 1 o o 1 o o o -1 0 C6 Lev lo'3% vs. Control 1 o o o 1 o o -l 0 C7 Lev 10'u% vs. Control 1 0 0 0 0 l 0 —1 0 C8 Lev 10‘5% vs. Control 1 o o o o o 1 -1 o Cl (44.03 + 34.14 + 29.86 - 3(48.l9))2 ((1/3 + 1/3 + l/12 + 9/15 + 1/1 + 1/18 + 1/9 + 9/30 + 1/6 + 1/6 + l/ll + 9/14 + 1/5 + 1/7 + 1/10 + 9/17 + 1/19 + 1/12 + 1/13 + 9/28)28.96)'1 = 8.55 CZ ll (32.17 + 42.57 + 52.59 - 3(48.19))2((1/11 + 1/6 + 1/6 + 9/15 + 1/4 + 1/7 + 1/9 + 9/30 + 1/5 + 1/8 + 1/6 + 9/14 + 1/7 + 1/6 + 1/5 + 9/17 + 1/16 + 1/13 + 1/11 + 9/28)28.96)‘1 2.25 C3 (44.03-—48.l9)2 ((1/3 + 1/15 + 1/12 + 1/30 + 1/6 + 1/14 + 1/5 + 1/17 + 1/19 + l/28)28.96)"1 = 0.54 C4 (34.14 - 48.19)2 ((1/3 + 1/18 + 1/6 + 1/7 + 1/12 + 1/15 + 1/30 + 1/14 + 1/17 + 1/28)28.96)'l = 5.51 C5 = (29.86 - 48.19)2 ((1/1 + 1/9 + 1/11 + 1/10 + 1/13 + 1/15 + 1/30 + 1/14 + 1/17 + 1/28)28.96 '1 = 7.05 continued C6 C7 08 (52.1? 1/15 + 1/15 + (52.59 1/15 + 153 APPENDIX D-l (continued) — 48.19)2 ((1/11 + l/4 + 1/5 + 1/7 + 1/16 + 1/30 + 1/14 + 1/17 + 1/28)28.96)'1 = 8.75 — 48.19)2 ((1/6 + 1/7 + 1/8 + 1/6 + 1/13 + 1/30 + 1/14 + 1/17 + l/28)28.96)'1 = 1/16 -48.19)2 ((1/6 + 1/9 + 1/6 + 1/5 + 1/11 + 1/30 + l/l4 + 1/17 + l/28)28.96)"l = 0.67 LIST OF REFERENCES LI ST OF REFERENCES Alls. M. E. and J. H. Greve. 1974. Canine dirofilariasis in Iowa. J. Am. Vet. Med. Assn. 165: 532-33. Bangham, D. R. 1955a. Metabolism and excretion of luC- labelled diethylcarbamazine. Br. J. Pharmacol. 10: 397-405. Bangham, D. R. 1955b. The mode Rf action of diethylcar— bamazine investigated with 1 C-labelled drug. Br. J. Pharmacol. 10: 406-12. Behin, R. 1967. Artificial feeding apparatus for mosqui- toes. Mosq. News 27: 87-90. Bemrick, W. J., B. L. Buchli, and H. J. Griffiths. 1965. Development of Dirofilaria immitis in Anogheles guadri- maculatus after exposure of the microfilariae to a freezing temperature. J. Parasitol. 51: 954-57. Benz, G. W. 1973. Anthelmintic activities of cambenda- sole in calves: effect on nematode morphogenesis from ova to the infective larvae stage. J. Parasitol. 59: 166-68. Bishop, A. and B. M. Gilchrist. 1944. A method for col- lecting sporozoites of Plasmodium gallinaceum by feed- ing infected Aedes aegypti through animal membrane. Nature 153: 713—14. Bradley, R. E. 1971. The etiology of pathologic changes in canine heartworm disease. ;g_Pathology of Parasi- tic Diseases. S. M. Gaafar (ed.). Purdue University Studies, Lafayette. Indiana. 161-71. Bradley, R. E. 1976. Levamisole resinate as a Dirofilaria immitis microfilaricide in dogs. J. Am. Vet. Med. A8811. 169' 311‘150 Breinl. A. 1920. Preliminary note on the development of the larvae of Dirofilaria immitis in dog fleas Ctenoce- phalus felis. Ann. Trop. Med. Parasitol. 14: 389-92. Brug, S. L. 1932. Chitinization of parasites in mosqui- 154 155 Bryan, J. H. and B. A. Southgate. 1976. Some observations on filariasis in Western Samoa after mass administra- tion of diethylcarbamazine. Trans. Roy. Soc. Trop. Med. Hyg. 70: 39-48. Burkhart, R. L. and B. T. Alford. 1972. Pharmacology and mode of action of diethylcarbamazine. ln_Canine Heart- worm Disease: The Current Knowledge. R. E. Bradley (ed.). University of Florida, Gainesville, Florida, 17-30. Burton, G. J. 1963. Wuchereria bancrofti and Dirofilaria immitis larvae in palpi and antennae of mosquito vec- tors. J. Parasitol. 49: 771-72. Center for Disease Control. 1974. Morbidity and Mortality Weekly Report 23: 374-77. Coffin, D. L. 1944. A case of Dirofilaria immitis infec- tion in a captive-bred timber wolf (Canis occidentalis, Richardson). N. Am. Vet. 25: 611-12. Colglasier, M. L., K. C. Kates, and F. D. Enzie. 1974. Cambendazole-resistant Haemonchug contortus strain in sheep; further experimental developments. J. Parasitol. 60: 289-92. Desowitz, R. S., C. C. Draper, and T. M. Phillips. 1975. The nature of the adverse reaction to diethylcarbama- zine treatment in dogs infected with Dirofilaria im- mitis. Trans. Roy. Soc. TrOp. Med. Hyg. 69: 430. Duke, B. 0. L. 1974. The effect of levamisole on Oncho- cerca volvulus. Trans. Roy. Soc. Trop. Med. Hyg. 68: 71. Duxbury, R. E., A. P. Moon, and E. H. Sadun. 1961. Sus- ceptibility and resistance on Anopheles a adrimaculatus 7 u to Dirofilaria immitis. J. Parasltol. 7: -91. Erickson, A. B. 1944. Helminths of Minnesota Canidae in relation to food habits, and a host list and key to the species reported from North America. Am. Midland. Naturalist 32: 358-72. Euzeby, J. and B. Laine. 1951. Sur la periodicite des microfilaires de Dirofilaria immitis. Rev. Ved. vet. Toulouse 102: 231-38. Feldman, R. G. and M. J. Holden. 1974. Meningeal irri- tation, hemOptysis, and eosinophila--a case of human dirofilariasis? J. Am. Med. Assn. 228: 1018-19. 156 Foil, L. and T. C. Orihel. 1975. Dirofilaria immitis (Leidy, 1856) in the beaver Castor canadensis. J. Parasitol. 61: 433. Gerberg, E. J. 1971. Evaluation of antimalarial compounds in mosquito test systems. Trans. Roy. Soc. Trop. Med. Hyg. 65: 358-63. Gerberg, E. J., F. W. Kutz, and G. S. Tulloch. 1972. Pre- liminary evaluation of 1-tetramisole as an antifilarial in a mosquito test system. In Canine Heartworm Disease: The Current Knowledge. R. E. Bradley (ed.). Univer- sity of Florida, Gainesville, Florida, 105-107. Gibson, D. W., D. H. Connor, H. L. Brown, H. Fuglsang, J. Anderson, B. 0. Duke, and A. A. Buck. 1076. On- chocercal dermatitis: ultrastructural studies of microfilariae and host tissues, before and after treatment with diethylcarbamazine (Hetrazan). Am. J. Trop. Med. Hyg. 25: 74-87. Graybill, F. A. 1976. Theory and Applications of the Linear Model. Duxbury Press, North Scituate, Mass. Green, B. M. 1975. Microfilariae in a patient with sys- temic lupus erythematosis: case report. In Proceed- ings of the Heartworm Symposium '74. H. C. Morgan (ed.). School of Veterinary Medicine, Auburn Univer- sity, Auburn, Alabama, 14-15. Greenberg, J. 1949. A method for artificially feeding mosquitoes. Mosq. News 9: 48-50. Grob, D. 1976. Use of drugs in myOpathies. Ann. Rev. Pharmacol. Toxicol. 16: 215-28. Harned, B. K., R. W. Cunningham, S. Halliday, R. E. Vessey, N. N. Yuda, M. 0. Clark, C. H. Hine, R. Cosgrove, Y. SubbaRow. 1948. Studies on the chemotherapy of filariasis IV. Some pharmacological properties of 1-diethylcarbamyl-4-methylpiperazine hydrochloride, Hetrazan. J. Lab. Clin. Med. 33: 216-38. Hawking, F. 1956. The periodicity of microfilariae IV. Stimuli affecting the migration of the microfilariae of Dirofilaria aethiOps, Q. immitis, Q. repens, Dipe- talonema blanci, and Litomosoides carinii. Trans. Roy. Soc. Trop. Med. ParasitolI'4: 397-417. Hawking, F. 1967. The 24-hour periodicity of microfilar- iae: biological mechanisms responsible for its pro- duction and control. Roy. Soc. (London) Proc., Ser. B 169: 59-76. 157 Hawking, F., P. Sewell, and J. P. Thurston. 1950. The mode of action of Hetrazan on filarial worms. Brit. J. Pharmacol. 5: 217-38. Hewitt, R. I., E. White, W. S. Wallace, H. W. Stewart, S. Kushner, Y. SubbaRow. 1947. Experimental chemothera- py of filariasis II. Effect of piperazine deriva- tives against naturally acquired filarial infections in cotton rats and dogs. J. Lab. Clin. Med. 32: 1304- 13. Ho, B. C., M. Singh, and E. H. Yap. 1974. The fate and migratory patterns of the infective larvae of Brugia malayi, Dirofilaria immitis, and Breinli sergenti in Aedes togoi denied access to a host. J. Med. Ent. 11: 622-28. Intermill, R. W. 1973. DevelOpment of Dirofilaria immitis in Aedes triseriatus Say. Mosq. News 33: 176-81 Iyengar, M. 0. T. 1957. Developmental stages of filariae in mosquitoes. So. Pac. Comm. Tech. Pap. 164: 1-11. Jackson, R. F. 1972. Some new chemotherapeutic agents for dirofilariasis--a preliminary report. in Canine Heart- worm Disease: The Current Knowledge. R. E. Bradley (ed.). University of Florida, Gainesville, Florida, 129-136. Jackson, W. F. 1969a. Physical examination of the heart- worm-gnfected patient. J. Am. Vet. Med. Assn. 54: 377-7 . Jackson, W. F. 1969b. Radiographic examination of the heartworm-infected patient. J. Am. Vet. Med. Assn. 543 380—82 0 Jackson, W. F. 19690. Preventative therapy with diethyl- carbamazine. J. Am. Vet. Med. Assn. 154: 396. James, M. T. and R. F. Harwood, eds. 1969. Herms's Medi- cal Entomology. MacMillan Publishing Co., Inc., New York. 484 pp. Johnson, W. P., R. G. Eggert, G. P. Poeschel, and G. T. Wang. 1972. Levamisole as an anthelmintic for swine. J. Am. Vet. Med. Assn. 161: 1221-25. Jones, J. C. 1967. Methods for dissection of mosquitoes. Mosq. News 27: 76-82. 158 Joon-Wah, M., V. Zaman, and S. Sivanandam. 1974. Anti- filarial activity of levamisole hydrochloride against subperiodic Brugia malayi infection of domestic cats. Am. J. Trop. Med. Hyg. 23: 369-74. Kartman, L. 1953a. An observation on the loss of micro- filariae from the mosquito during its infective meal. J. Parasitol. 39: 571-72. Kartman, L. 1953b. Factors influencin infection of the mosquito with Dirofilaria immitis %Leidy, 1856). Expl. Kartman, L. 1956. Notes on the encapsulation of Dirofilar- ia immitis in the mosquito Aedes aegypti. Am. J. Vet. Res. 17: 810-12. Keegan, H. L., C. M. Fitzgerald, T. L. McCrary, and M. D. Doyles. 1968. Studies on dog heartworm in South Texas. Tex. Rpts. Biol. Med. 26: 321-30. Kocan, A. A. and H. E. Laubach. 1976. Dirofilaria im- mitis and Dipetalonema reconditum infections in Okla- homa dogs. J. Am. Vet. Med. Assn. 168: 419-20. Kume, S. and S. Itagaki. 1955. On the life-cycle of Dirofilaria immitis in the dog as a final host. Br. Veto J. 1113 1.6-250 Kutz, F. and R. C. Dobson. 1974. Effects of tempera- ture on the deve10pment of Dirofilaria immitis (Lei- dy) in AnOpheles guadrimaculatus Say and on vector mortality resulting'from this development. Ann. Ent. Soc. Am. 67: 325-31. Lindsey, J. R. 1961. Diagnosis of filarial infections in dogs I. Microfilarial surveys. J. Parasitol. 47: 695-702. McGreevy, P. B., J. H. Theis, M. M. J. Lavoipierre, and J. Clark. 1974. Studies on filariasis III. Dirofilaria immitis: emergence of infective larvae from the mouthparts of Aedes aegypti. J. Helmin. 48: 221-28. Medway, W. and T. C. Wieland. 1975. Dirofilaria immitis infection in a harbor seal. J. Am. Vet. Med. Assn. 167: 549-50. , Merck and Company, Inc. 1967. The Merck Veterinary Index. Merck and Company, Rakway, New Jersey, 731-35. Mills, J. N. and T. C. Amis. 1975. Levamisole as a micro- filaricidal agent in the control of canine dirofilaria- sis. Austral. Vet. J. 51: 310-14. 159 Natarajan, P. N., V. Zaman, and T. S. Yeoh. 1974. In vitro activity of levamisole on the infective larvae, microfilariae, and adult worms of Breinlia sergenti. Internatl. J. Parasitol. 4: 207-10. Navarrete, A. R. 1972. Pulmonary dirofilariasis. Chest 61: 51-55. Newton, W. L. and W. H. Wright. 1956. The occurrence of a dog filariid other than Dirofilaria immitis in the United States. J. Parasitol. 42: 246-58. O'Holohan, D. R. and V. Zaman. 1974. Treatment of Brugia malayi infection with levamisole. J. Trop. Med. Hyg. 77: 113-15 0 Orihel. T. C. 1961. Morphology of the larval stages of Dirofilaria immitis in the dog. J. Parasitol. 47: 251- 20 Otto, G. F. 1969a. Geographical distribution, vectors, and life cycle of Dirofilaria immitis. J. Am. Vet. Med. Assn. 154: 370-73. Otto, G. F. 1969b. Chemotherapeutic agents. J. Am. Vet. Med. Assn. 154: 387-90. Otto, G. F. 1972. Epizootiology of canine heartworm dis— ease. In Canine Heartworm Disease: The Current Know- ledge. R. E. Bradley (ed.). University of Florida, Gainesville, Florida, 1-15. PharmIndex. 1973. Skyline Publishers, Inc., Portland, Oregon, 130. Ponnudurai, T., D. A. Denham, and G. S. Nelson. 1971. The use of a membrane feeding technique for infecting mosquitoes with filarial worms transported between laboratories. J. Helmin. 45: 415-18. Robinson, M. J., M. Viamonte, and M. Viamonte. 1974. Parasites: diagnostic payoff in coin lesions. Modern Medicine 42: 75. Rogers, R. and D. A. Denham. 1976. Studies with Bru ia pahangi 12. The activity of levamisole. J. Helmin. 503 21'280 Rothstein, N., K. E. Kinnamon, M. L. Brown, and R. W. Carithers. 1961. Canine microfilariasis in the east- ern United States. J. Parasitol. 47: 661-65 160 Rutledge, L. C., R. A. Ward, and D. J. Gould. 1964. Studies on the feeding response of mosquitoes to nutri- Zive solutions in a new membrane feeder. Mosq. News 24: O7 -19 o Schnelle, G. B. and R. M. Young. 1944. Clinical studies on microfilarial periodicity in war dogs. Bull. U. S. Army Med. Dept. 80: 52. Schuermans, V. 1975. Levamisole in rheumatoid arthritis. Lancet 1: 111. Sharp, J. J. 1974. Heartworms in a cat. J. Am. Vet. Med. Assn. 164: 519. Snedecor, G. W. 1956. Statistical Methods Applied to Ex- periments in Agriculture and Biology. Iowa State Col- lege, Ames, Iowa. 534 pp. Soulsby, E. J. L. 1965. Textbook of Veterinary Clinical Parasitology. F. A. Davis Co., Philadelphia, Pa., 100-113 0 Tarshis, I. B. 1959. Feeding Aedes aegypti (L.) through animal derived membranes (Diptera: Culicidae). Ann. Ent. Soc. Am. 52: 681-87. . Taylor, A. E. R. 1960a. Studies on the microfilariae of Loa loa, Wuchereria bancrofti, Brugia malayi, Diro- filaria immitis, Q, repens, and Q. aethlops. J. Hel- "lino 3n! 13"260 Taylor, A. E. R. 1960b. The development of Dirofilaria immitis in Aedes aegypt'. J. Helmin. 34: 27-38. Terzian, L. A. 1947. A method for screening antimalarial compounds in the mosquito host. Science 106: 449-50. Theodorides, V. J. 1974. Development of resistance by parasites to anthelmintics. Internatl. Congress Parasitol. Proc. 3: 1390-91. Thienpoint, D., 0. F. J. Vanparijs, A. H. M. Raeymaekers, J. Vandenberk, P. J. A. Demoen, F. T. N. Allewijn, R. P. H. Marsboom, C. J. E. Niemegeers, K. H. L. Schellekens and P. A. J. Janssen. 1966. Tetrami- sole (R8299), a new, potent, broad-spectrum anthel- mintic. Nature 209: 1084-86. Travis, B. V. 1947. Relative efficiency of six Species of mosquitoes from Guam, M. I., as developmental hosts for Dirofilaria immitis. J. Parasitol. 33: 142-45. 161 Tripodi, D., L. C. Parks, and J. Brugmans. 1973. Drug induced restoration of cutaneous delayed hypersensiti- vity in anergic patients with cancer. N. Eng. J. Med. 289: 354. Tritch, V. R., H. A. James, and M. P. Ratner. 1973. Canine filariasis (Dirofilaria immitis and Dipetalonema reconditum): survey in southwestern Connecticut. Am. J. Vet. Res. 34: 1473-75. Tulloch. G. S. and R. A. Anderson. 1971. Acceptability of daily l-tetramisole by pound dogs infected with Diro- filaria immitis. Nature 232: 644. Tulloch. G. S. and R. A. Anderson. 1972. Tetramisole and canine dirofilariasis. ;p_Canine Heartworm Disease: The Current Knowledge. R. E. Bradley (ed.). Univer- sity of Florida, Gainesville, Florida, 101-104. Tulloch. G. S., G. Pacheco, H. W. Casey, and R. A. Anderson. 1970. Efficacy of selected drugs as prophylactic agents against Dirofilaria immitis of dogs. Internatl. Con- gress Parasitol. Proc. 2: 348. University of Notre Dame, Vector Biolo y Laboratory. 1974. Mosquito stock list (mimeographed7. Van den Bossche, H. and P. A. J. Janssen. 1967. Mode of action of levamisole on Ascaris. Life Science 6: 1781-82. Wade, J. 0. 1975. An improved method for the membrane feeding of mosquitoes. Trans. Roy. Soc. Trop. Med. Hyg. 69: 19. Wallenstein, W. L. and B. J. Tibola. 1960. Survey of canine dirofilariasis in a Maryland area: incidence of Dirofilaria immitis and Dipetalonema. J. Am. Vet. Med. Assn. 137: 712-16. Weiner, D. J. and R. E. Bradley. 1970. Ability of some mosquitoes to transmit Dirofilaria immitis in Florida. Mosq. News 30: 406-10. Wills, W., D. F. Carroll, and G. E. Jones. 1974. A sim- ple method for artificially feeding mosquitoes. Mosq. News 34: 119-120. Woke, P. A. 1937. Comparative effects of the blood of different species of vertebrates on egg production of Aedes aegypti Linn. Am. J. Trop. Med. 17: 729-45. 162 Yoeli, M. 1938. Note on the experimental infection of Ano- pheles elutus with Plasmodium faliciparum by feeding ghrgggh a prepared animal membrane. Riv. Malar. 17: 2" o Zaman, V. and M. Lal. 1973. Treatment of Wuchereria ban- crofti with levamisole. Trans. Roy. Soc. Trop. Med. Hyg. 67: 610. Zaman, V. and P. N. Natarajan. 1973. The antifilarial activity of levamisole on Breinlia sergenti compared wgth diethylcarbamazine. J. Trop. Med. Hyg. 76: 126- 2 . "I7'11411471474144“