I‘HES‘S lllllllllllllllllIll”UHllllllllllllllllHlllllllllllllllll 3 1293 01683 0220 This is to certify that the thesis entitled Preferential "ate selection by wales as a modictive isolating ned'mism between the selloutajl species; Ward em (Lepictpbera, Papiliorfidae) presented by rank David m ' has been accepted towards fulfillment of the requirements for Wane in W 5 Major professor Date 8/24/1993 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University PLACE iN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MTE DUE OCT 2 5 2005 E" E is l 'I . - ‘ g. an:- F . DATE DUE m i 3 250:? DATE DUE if use Wages-p.14 PREFERENTIAL MATE SELECTION BY MALES AS A REPRODUCTIVE ISOLATING MECHANISM BETWEEN THE SWALLOWTAIL SPECIES; PAPILIO GLAUCUS AND P. CANADENSIS (LEPIDOPTERA, PAPILIONIDAE) BY Mark David Deering A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ENTOMOLOGY Department of Entomology 1998 ZUBSTKUACI‘ PREFERENTIAL MATE SELECTION BY MALES AS A REPRODUCTIVE ISOLATING MECHANISM BETWEEN THE SWALLOWTAIL SPECIES; PAPILIO GLAUCUS AND P. CANADENSIS (LEPIDOPTERA, PAPILIONIDAE) BY Mark David Deering Papilio glaucus and P. canadensis (Lepidoptera: Papilionidae) in the United States have a narrow hybrid zone between them. Hybrids of both species are viable and host plants are found on either side of the hybrid zone, so interspecific mating would be expected to extend the hybrid zone. Mate preference as a pme-zygotic isolating mechanism may account for the narrowness of the zone. Research conducted in Florida and Michigan attempted to determine the importance of such a mechanism. In Florida, 1997, P. glaucus males chose conspecific, yellow females for copulation 94.8% of the time (n=69), choosing heterospecific mates only 5.2%. In 1998 males chose P. glaucus females in all cases (n=30). In Northern. Michigan P. canadensis ‘males .did not chose conspecific females. Rather, of 493 copulations recorded in 1997 (21 pair combinations) heterospecific P. glaucus females were chosen 82.3%. Mate choice by males could play a large role in gene flow outside the hybrid zone. Dedication This work is dedicated to my Mother and Father, who never gave up on me. HI ACKNOWLEDGMENTS There are many people who helped to make this project possible, and I would like to recognize a few of them. Firstly, my beautiful and supporting wife, who helped me throughout this endeavor. Dr. J. Mark Scriber, my advisor and friend for his many hours of help and dedication to making sure that this project succeeded. I would like to thank the many graduate and undergraduates who helped in the lab raising stock for this project as well as offering moral support. I thank the department of Entomology whose faculty and staff were ever supportive and helpful with problems and solutions. IV TABLE OF CONTENTS VI. LIST OF TABLES VII. LIST OF FIGURES 1. INTRODUCTION 9. MATERIALS AND METHODS STOCK ACQUISITION AND HUSBANDRY SITE CHOICE FOR MALE MATE PREFERENCE STUDIES FEMALE PROCESSING AND TETHERING THE EXPERIMENTS ANALYSIS 21. RESULTS 31. DISCUSSION 40. LITERATURE CITED 24. 24. 26. 27. 27. 29. 30. TABLE TABLE TABLE TABLE TABLE TABLE LIST OF TABLES MARCH\APRIL 1997 FLORIDA DATA TABLE COPULATION NUMBERS AND DIFFERENCES, FLORIDA 1997. STATISTICAL ANALYSIS OF FLORIDA 1997 DIFFERENCE DATA. MARCH\APRIL 1998 FLORIDA DATA TABLE. COPULATION NUMBERS AND DIFFERENCES. FLORIDA 1998. STATISTICAL ANALYSIS OF FLORIDA 1998 DIFFERENCE DATA. 1997 MICHIGAN DATA TABLE. COPULATION NUMBERS AND DIFFERENCES. MICHIGAN 1997. STATISTICAL ANALYSIS OF MICHIGAN 1997 DIFFERENCE DATA. VI Page 2. 6. l4. 17. 22. 25. 28. FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE LIST OF FIGURES THE HYBRID ZONE IN MICHIGAN EXPECTED PATTERNS OF MATE CHOICE IN TWO MATE RECOGNITION SYSTEMS WING LENGTH AND PUPAL WEIGHT FOR THE TWO PAPILIO SPECIES STUDIED TETHERING ARRANGEMENT FOR FEMALE BUTTERFLIES IN MATE CHOICE EXPERIMENTS 1997 FLORIDA TOTALS 1998 FLORIDA TOTALS 1997 MICHIGAN TOTALS VII Introduction It is known that a hybrid zone exists between the two closely related swallowtail species Papilio glaucus and P. canadensis (Rothschild & Jordan 1906). Rothschild and Jordan (1906) first described P. canadensis as a sub-species of P. glaucus, and it has since been designated as a species (Hagen et a1. 1991). The hybrid zone, which corresponds closely with the boreal/temperate plant tension zone (Hagen 1990, Scriber & Gage 1995), has been delineated well by researchers using adult wing morphology of P. glaucus and P. canadensis (Luebke et a1. 1988), electrophoresis (Hagen & Scriber 1989) as well as other diagnostic traits (Scriber 1996, Hagen et al. 1991). The hybrid zone is narrow across most areas of contact between the two species, though it broadens across the Appalachian range in New York, due to the geographic overlap of different habitats and temperature zones caused by the altitude differences throughout that area (Scriber 1996). Gene flow across the Wisconsin and Michigan areas of the zone appears to be especially limited (Luebke et al 1988, figure 1). This may be due in part to the lack of suitable forest edge and hedgerow habitat in this area, but also may be due to other biotic and abiotic isolating mechanisms. The narrowness of the zone through this area is not unexpected as it corresponds to the Figure 1: The Hybrid Zone in Michigan. transition zone between the plant ecotones (Remington 1968). Other Papilio species have been studied to determine their interspecific hybrid compatibility and to test for genetic differences in larval survivorship and hybrid host plant use (Scriber 1996, Scriber & Lederhouse 1988, Scriber et a1. 1988, 1991, & 1995). Hand pairings in the laboratory have allowed investigation of the genetic basis of many previously unstudied ecological traits as well as noting diagnostic genetic markers across this hybrid zone (Scriber 1996). Most animals that form hybrid zones tend to remain as separate and distinct populations with only narrow zones, or isolated pockets of hybridization marking the flow between them (Harrison, 1993). There are numerous factors that facilitate this separation. These factors are generally separated into two different categories; those that are prezygotic and those that are postzygotic. The main factors in the prezygotic category are sperm/egg incompatibility, mate recognition and mate choice by males or females, mechanical isolation (lock and key hypothesis) (Porter & Shapiro 1990) and sperm. precedence/manipulation (cryptic sexual selection). Major factors in the post-zygotic category are egg inviability, low fecundity and larval survivorship, F1 hybrid infertility or unfitness, and the Haldane effect. The Haldane effect, which causes the death of the F1 females, or the Fla and F2 hybrid females' ova or young larvae before they have a chance to develop and reproduce (Hagen & Scriber 1995), has been well studied by researchers (Coyne et a1. 1991, Sperling 1993, Coyne 1994). This study focuses on male mate choice among paired, tethered, size-matched females of P. canadensis and P. glaucus. Strong preferences for conspecific females may act as a reproductive isolating mechanism responsible for restricting gene flow across the hybrid zone. Free flying P. canadensis males in Northern Michigan and free flying P. glaucus males in Central Florida were bioassayed to determine their mating preferences (i.e. for the conspecific or heterospecific female of the pair). Active female solicitation flights have been observed in P. glaucus (Krebs & West, 1988), therefore female mate choice could certainly be considered as a possible factor in this type of mating system (Rutowski 1984). However, in the field it has been observed that males of P. glaucus or P. canadensis initiate courtship flight and behavior. Therefore, my research was focused on this primary component of mate selection by using size-matched, tethered pairs of females placed in male flyways in natural populations. If species-specific assortative mate responses are in effect (Mayr, 1963) then strong preferences for the conspecific of the pair would be predicted, both for P. glaucus populations in Florida and for P. canadensis populations in Northern Michigan. However, random mating could result if these populations are far from. the hybrid zone where reinforcement of correct choice might be relaxed due to the rarity of encounters with the heterospecific female (Howard 1993, Butlin 1995). Figure 2 illustrates the patterns of preference that might be noted if reinforcement/reproductive zuzcoumwz no :EmEouoowEom: EwEmomEflQ 335230 922369me to EmEmBoWEmz mo cozmxflmm o\e o m . 2mm: mfiozu .3 £953.53 El now 3:83th 9». (lo\e on: Ammzmv Eggm :oEEwouud 222 9:33; flmcwvacau .L EXPECTED PATTERNS OF MATE CHOICE FROM TWO MATE CHO I CE SYSTEMS FIGURE 2 Arnold & Hodges 1995). The eventual breakdown of hybrid zones may result from neutral introgression or selection of alleles that may reduce the degree of hybrid unfitness (Riesenberg & Wendel 1993, Arnold & Hodges 1995, Ritchie & Hewitt 1995). If two populations have diverged genetically and are producing hybrids that are not as fit as the original species then natural selection should favor prezygotic reproductive isolation and an increase in assortative mating, helping to move the whole population towards speciation (Mayr 1963). Prezygotic reproductive isolating mechanisms prevent wasting of gametes on production of hybrids that will be unfit and unlikely to survive, and are subject to natural selection through reinforcement (Howard 1993, Butlin 1995). Reinforcement across hybrid zones is an idea that can be attributed to Dobzhansky in his 1940 paper. He argued that as the two populations of a species evolved in allopatry that they would also evolve different ways of recognizing their mates, both behaviorally and genetically. Once these populations were back in contact with each other then the hybrid offspring that were jproduced. would. suffer. Reinforcement itself would allow for more effective mate recognition in the areas where the populations actually meet and overlap. "Thus prezygotic isolating barriers evolve after the adaptive divergence of the incipient species, and this evolution occurs primarily in areas where the two taxa overlap in distribution" (Howard 1993). Reinforcement in mate choice takes place as selection pressures become greater for one or the other of a set of potential mates. This could be due to perceived mate preference (ornamentation) or due to lower survival of offspring with one or the other potential mate, leaving more offspring that may prefer the particular traits of their parents. The idea of prezygotic isolating mechanisms and reinforcement lacks convincing examples, but this is an extremely appealing concept and one that this study attempts to examine. Hopefully other studies will follow as reinforcement and reproductive isolating mechanisms are ideas that deserve more critical research focus (Howard 1993, Butlin 1995, Futuyma and Shapiro 1995, Veech et al. 1996). Rather than using inference and indirect methods, this study used direct observations of assortative mating at different field locations across both species' home ranges to test hypotheses about prezygotic isolating mechanisms. Materials and Methods Stock Acquisition and Husbandry The swallowtails used in these experiments were laboratory reared from stock collected from natural populations. Papilic> glaucus females ‘were collected from three sites over a three year period. 1) Kentucky (Daniel Boone National Forest). 2) Lawrence County Ohio (Wayne national Forest and Dean State Forest). 3) Clarke County Georgia. Papilio canadensis females were collected from wild populations in the Lower and Upper peninsulas of Michigan; primarily' Chippewa, Cheboygan, Charlevoix: and (Utsego counties. Collected females were processed by measuring the fore- wing length (distance in mm from basal attachment to wing tip), recording the County and area of capture, and issuing the female a "Mother Number" which identified the future offspring as well. For oviposition, females were housed in clear round plastic dishes (12" diameter by 4" deep) and were placed on a turntable which rotated the dishes once every 6 minutes. The insects were positioned on the turntables in front of incandescent lamps and leaves of host plant were inserted into water-filled, rubber-capped, plastic tubes called. aquapics (available from florists). Aquapics with host leaves were placed along the inside of the dish for oviposition. Each female was fed once a day on a 20% honey and water solution by drawing their proboscises into a spoon holding the solution. Females readily took food in this manner and some females lived for as long as three weeks. When possible, females were frozen (at -80° C) when near death for later genetic and enzyme analysis. Eggs from these females were counted and collected daily and new leaves were also added at this time. The new eggs were placed into 110 mm Pctri dishes which were then placed. in. growth chambers set at 25° CL Eggs generally hatched within 3—6 days and were checked daily for newly hatched larvae which were then moved to leaves of fresh host plant in new 110 mm dishes. Leaves of the preferred host plant were kept turgid by inserting freshly cut stems into aquapics. Care was taken not to damage the larvae since they are very easily wounded at this stage. A soft rolling motion with a fine, damp camel hair brush worked well. Branches of host used to feed the larvae were collected from sites in Ingham County Michigan, or were from plants raised in pots in a greenhouse. Branches of host were pruned from near the tops or on the sides of young trees to ensure a good quality diet. The cut ends were placed in water jugs which were then refrigerated. Placement of a plastic bag around the foliage helped to retain moisture while under refrigeration. Foliage was harvested every' other day' to ensure that the host material was fresh. Larvae were identified and recorded by mother number and host plant identification abbreviation. For the P. 10 glaucus used in this experiment they were raised on Black Cherry (Prunus serotina). P. canadensis was also reared on Black Cherry. Since both P. glaucus and P. canadensis were raised on Black Cherry this removes any chance the host plant differences could play a role in choice by males. Also noted on the dish were the phenotype (species, and color morph if P. glaucus), state of origin, date that larvae hatched, and the number of larvae being placed in the dish. The dishes were kept relatively dry by placement of a paper towel circle in the bottom of each dish. Food was changed every 1-2 days, or as needed, by gently removing the larvae from the old leaves and replacing them with fresh ones. Clean, dry' paper towel was added. at this time and, the aquapics were refilled. The number of larvae placed in new dishes ranged from 3-5 depending on available space in growth chambers. Larvae, in their new dishes, were then placed into growth chambers set at 25° C and.1.30 km/hr). Analysis Male preference was determined both for individual pairs and from the population standpoint by analysis of the total number of c0pulations tallied for each pair. These data were subjected to a Wilcoxon signed-rank test to determine statistical significance. This test was used to determine preference for each of the pairs and for the populations by 19 using the means of the differences in responses for each of the species and pairs. The Wilcoxon test compares the distribution and mean of the response differences to each pair used at that site with regard to the total number of copulations. Data were analyzed to test the null hypothesis that the distribution mean of the differences was equal to 0, using the JMP statistical program. 20 Results The data collected in this study show clearly that mate preference by the males of P. glaucus and P. canadensis is not random, even an: great distances from time hybrid zone between the two populations. We have found that the males of both P. glaucus and P. canadensis prefer the females of P. glaucus in these paired tests. The data gathered from Florida in both 1997 and 1998, and from Michigan in 1997 reveal a clear pattern of mate choice. The null hypothesis stated at the beginning of this study was: That there will be no difference in the behavior of P. glaucus and P. canadensis males from populations long distances from the hybrid zone towards heterospecific pairs of size Hatched, tethered females. Both populations chose females of P. glaucus for their mate choices, thus rejecting the null hypothesis. In 1997 P. glaucus males in Florida were particularly selective in their choice of mates (Figure 5). Analysis showed that 94.2% of the copulations occurred with P. glaucus. 67% of the touches were made towards P. glaucus (n=112) and 87.5% of the attempted copulations were with P. glaucus (n=56) (Table 1). The total number of female pairs tested in spring of 1997 was 19 (1 set of paired females = I data point). A column of data showing the difference between 21 Frequency of Behavior Touches Att. Cops. Copulations 1 (DB- sanajéfies ,!E- glaucus Figure 5: 1997 Florida Totals. Frequncy of behavioral classes exhibited by male P. glaucus towards paired females. 22 the total number of copulations for each female from the 19 pairs was generated (Table 2). ITable 1: March\Apri1 1997 Florida Data Table. I IPairs n = 19 Touches Attempts Copulations P. canadensis: 33% 16.5 5.8% P. glaucus: 67% Total encounters: Table 2: Copulation Numbers and Differences, Florida 1997. Species: Total Copulation Numbers Pairing (Pg/Pa) P.glaucus P. canadensis Difference Pgl/Pcl o o o PgZ/Pc2 6 1 5 Pgl/Pc3 5 o 5 Pg3/Pcl 3 o 3 Pg4/Pc4 1 0 1 PgB/PCS 8 0 8 Pgl/Pc4 6 1 5 PgS/Pcl 2 o 2 Pg6/PC4 1 o 1 Pg7/PC6 1 o 1 Pg3/Pc3 0 o o Pg6/Pc6 0 0 0 Pg7/Pc7 6 o 6 PgGA/PcMI 3 0 3 PgS/PCB 5 0 5 Pg6/Pc7 2 0 2 PgB/Pc7 6 1 5 PgB/Pc6 5 0 5 P07 PC8 5 l 4 In Florida, 1997, the statistical test used showed there was indeed a difference in the way that males reacting to the individual females placed out 23 that were in heterospecific pairs (Table 3). The mean difference, in number of copulations per pair, from the estimated null of 0.0 was 3.16, with a standard deviation of 2.41. Standard error was 0.55. The signed rank produced for this data was 60.0, P <.0001. Table 3: Statistical analysis of FL 1997 Difference Data. Actual Mean = 3.15789 Hypothesized value = 0.0 St. Dev. = 2.40978 St Error = 0.55284 N= 19 Signed Rank (Wilcoxon Test) = 60.000, P = 0.000 In March and April of 1998 the Florida populations were again bioassayed. It was found that the pattern seen in 1997 had not changed significantly. The population levels in 1998 were lower than in 1997 (Scriber et al 1998) and consequently not as much data per pair was collected. All of the P. glaucus males choose to copulate with the P. glaucus females of the pairs (n=22 pairs) (Figure 6). 59% of the touches were with conspecific females (n=29), and 80% of attempted copulations with P. glaucus females as well (n=20) (Table 4). Table 4: March\Apri1 1998 Florida Data Table. Pairs n = 22 Touches Attempts Copulations P. canadensis: 41% 20% 0.0% P. glaucus: 59% 80% 100% ‘ Total encounters: 29 20 30 24 30* 25s I. I2 20 > C0 .= 0 m l H- O > 0 15— C 0 3 0' 2 _ LL 10s 5- 0_ 30 Touches Att. Cops Copulations 1:1 PeaEade—nEjs IP. gangs“ Figure 6: 1998 Florida Totals. Frequncy of behavioral classes exhibited by male P. glaucus towards paired females. 25 The numbers of copulations for each female of each paring was placed generated (Table 5). into columns and a column of differences was Table 5: Copulation Numbers and Differences. Florida 1998. Species: Total Copulation Numbers Pairing (Pg/Po) P.glaucus P. canadensis Difference Png/Pcl4 PgS/PCZ Pg4/Pc1 PglO/Pc5 Png/Pc? PgS/Pc4 Pg2/Pc2 Pgl/Pcl PgB/Pc13 Pg21/Pc9 Pg8/Pc14 Png/Pc3 Pg6/Pc1 Pg4/Pc2 Pgl/Pc2 Pgl6/Pc2 Pg18/Pc12 PgIZ/Pc9 Pgl6/Pc7 PglO/PCIZ Png/Pc4 Png/PCS Pgl/Pc4 Pg13/Pc1 The data collected in 1998 in Florida showed a significant difference females of the pairs were chosen. HMOOOOWHOifiOOOOOOHOKONNi-‘WH in the OOOOOOOOOOOOOOOOOOOOOOOO in the way that the i—‘NOOOOWi—‘OhOOOOOOOOKONNi—‘UJH very two The number of pairs used in 1998 was 22. Data was gathered in only 12 of these pairs, but the others were included because there were males in the area and they 26 did have the opportunity to react to the females in these pairs. The mean of the differences in copulation total for each of the females in these parings was 1.25 with a standard deviation of 2.03 and a standard error of 0.41. The Wilcoxon test returned a value of 39.0, which correlates to a P value of .0061 (Table 6). Table 6: Statistical analysis of FL 1998 Difference Data. Actual Mean = 1.25 Hypothesized value = 0.0 St. Dev. = 2.03 St Error = 0.41 N = 22 Signed Rank (Wilcoxon Test) = 39.0, P = 0.000 The data collected in June of 1997 at sites in Northern Michigan did not support what was predicted (Figure 7). Of 493 total copulations observed in Northern Michigan, 82.3% were with the heterospecific female, P. glaucus. Out of 355 touches, 58.4% were for the glaucus females. The totals for attempted copulations showed that 79% of 128 encounters were for P. glaucus as well (Pairs=25) (Table 7). Table 7: 1997 Michigan Data Table. Pairs n = 25 Touches Attempts Copulations P. canadensis: 41.6% 21% 17.7% P. glaucus: 58.4% 79% 82.3% Total encounters: 355 128 493 These data were analyzed as before, using the same criteria and set up (Table 8). 27 450.— 406 Frequency of Behavior Touches Att. Cops. Copulations EIPcahadensis [:1 IP. glaucus| Figure 7: 1997 Michigan Totals. Frequency of behavioral classes exhibited by male P. glaucus towards paired females 28 Table 8: Copulation Numbers and Differences. Michigan 1997. Species: Total Copulation Numbers Pairing (Pg/Po) P.glaucus P. canadensis Difference Pgl/Pcl 10 Pg2/Pc2 20 Pg3/Pc3 29 Pg4/Pc4 12 Pg4/Pc1 7 PgS/Pcl 24 Pg5/Pc4 23 PgZ/Pc7 18 Pg2/PC3 8 Pg3/Pc2 28 Pg5/Pc6 8 Pg6/Pc4 21 pg8/Pc8 33 Pg7/Pc6 22 Pg7/Pc8 2 pr/Pc6 4 PgZ/PC9 3o PglO/PclO 16 PgZ/PclO 13 PglO/Pc9 7 Pg7/Pc1 5 PgB/Pc3 21 Pg6/Pc8 20 Pgll/Pcll 2 Pg8/pc4 6 OI—‘UTONNONKDUJHOUlibU'li—‘(DNmi—‘i-‘OOGDWO The data gathered in Michigan in 1997 was analyzed using the same test. The column of differences produced gave a mean difference of 12.32, with standard deviation of 8.08, and standard error of 1.62. There were 25 pairs of females used for data collection in Michigan. The Wilcoxon test returned a signed rank that was 162.5, giving a P value of 0.000 (See Table 9). 29 Table 9: Statistical analysis of MI 1997 Difference Data. Actual Mean = 12.32 Hypothesized value = 0.0 St. Dev. - 8.08 St Error — 0.1.62 = 25 Signed Rank (Wilcoxon Test) = 162.5, P = 0.000 30 Discussion fflua data collected 1J1 this study shows clearly that mate preference by the males of P. glaucus and P. canadensis is not a random phenomenon, even at great distances from the hybrid zone between the two populations. We have found that the males of both Papilio glaucus and P. canadensis prefer the females of P. glaucus in these paired tests. The 1997 experiments in Florida showed that 94.2% of P. glaucus males chose to mate with the conspecific female of the pairs (n=19). We observed a similar pattern in spring of 1998 even though the population levels were lower than in 1997. P. glaucus males choose to copulate with the P. glaucus females of the pairs we placed out in all cases. We tethered a total of 22 pairs in spring of 1998 but recorded data in the form of copulations on only 12 of those pairs. Results from Florida suggest that P. glaucus males are able to differentiate between females of P. glaucus and P. canadensis, and that they prefer to mate with the conspecific female. This result was not totally unexpected as species specific mate cues are often an important factor in mate selection. The data collected in June of 1997 showed that males in Michigan were as selective as those observed in Florida. Out of 493 total copulations involving free flying P. canadensis 31 males in Northern Michigan, 82.3% were with the heterospecific female, P. glaucus (Pairs=25). It can be stated with confidence that the males counted in the Florida data were not counted twice since P. glaucus in the area were being marked for a population study (Scriber 6H: al. 1998 submitted). Previously unmarked males that copulated were removed from copulation and were then marked and recorded. None of the males returned twice. This not only provided proof that the same males were not being counted twice but also allowed for a record of the sizes and wear class (approximate age) of each of the males. This was not so in Michigan as P. canadensis males that copulated in Michigan were not being marked. It was noted that all but a few males left the area very rapidly once they were separated by hand from copulation with the female. However, the number of copulations by males recorded in Northern Michigan, as well as the number of pairs used, was large enough that it would likely outweigh any possible pseudo— replication caused by returning males. This study has shown that P. canadensis males have a strong preference for the heterospecific female of these pairs, and that species conspecific mate cues were not preferred here. This allows for rejection of the null hypothesis. Obviously these results are in conflict with the prediction of no clear preference for females by males in either population long distances from the hybrid zone. It 32 can be surmised that since the two species likely arose from an allopatric speciation event during the last ice age that there may be some ancestral trait in females of P. glaucus that is more attractive to the males of P. canadensis than their own females. This idea, of ancestral attractiveness, is not unprecedented. Other researchers have used this idea to explain some behaviors that they found in their systems, including ovipositional preference in P. canadensis females for a toxic (for them) host; Liriodendron tulipifera (Bossart & Scriber 1995), mating ornamentation systems in least auklets (Jones and Hunter 1998) and female mate choice in some Shizocoza wolfspiders (McClintock & Uetz 1996). Interestingly, the study by Bossart & Scriber (1995) showed that some females from a population of P. canadensis in Alaska had a strong preference for using tulip tree as an ovipositional host, especially since they are thousands of miles from any tulip tree and they are toxic to the larvae. Those females with lesser preference for tulip tree were collected from near the plant tension zone, where they could have been undergoing selection against using the toxic tulip tree. So thus, it appears that the species can exhibit behavior that is non-adaptive. It is quite possible that tethering within the hybrid zone itself will reveal that those P. canadensis males might be more likely to select conspecific females bearing in mind that there may be evolutionary pressures (e.g. via this reinforcement of 33 prezygotic isolating' mechanisms) to ix: more accurate in their selections. The results found in this study help to confirm that the males of the Papilio glaucus species group are characterized by strong visual acuity, as is commonly found in many species where the males engage in puddling behavior(Arms et al. 1974). With male mate locating behavior associated with patrolling (Lederhouse 1995) this is also expected. The males of P. canadensis and P. glaucus patrol roadsides, forest edges, steam edges, nectar source areas and hedgerows (personal observation). The Western species P. rutulus, P. eurymedon and P. multicaudatus also patrol using the canyons and hilltops as the areas that they frequent (Brower 1959, Shields 1968, Scott 1975, 1983). There are other patrolling Papilionidae that have been documented in other countries as well (Yamashita 1995). Females in the P. glaucus group are known to mate multiple times from studies counting spermatophores in some species (Lederhouse et al. 1989, Lederhouse 1995), but this is not the case in all species of Papilionidae. There are many species that have evolved complex arrangements of mating plugs designed to mechanically ensure that the female does not re-mate (Orr 1995, Matsumoto & Suzuki 1995). This investment of resources by the male ensures that no other male fertilizes the eggs of that particular female. This is important to) the ‘male that mates ‘with the female first 34 because if the female is able to mate again the new spermatophore would likely displace his sperm through sperm precedence (Parker 1970). Conversely, the investment made by males of the P. glaucus group is in the spermatophore itself . The spermatophore that these butterflies pass to the female can often contain 5% or more of their total body weight (Rutowski et al. 1984). This nuptial gift also often contains sugars, proteins and other micronutrients that the female may be able to incorporate and use in somatic maintenance and egg production (Boggs & Gilbert 1979, 80998 1981, 1990). Generally speaking, the larger the spermatophore that is passed to the female, the longer the time it will be until she mates again. This helps to ensure that more eggs will be fertilized using the sperm provided by the male producing that spermatophore (Lederhouse et al. 1989). Some other factors that may be considered as possible reasons that P. canadensis males choose heterospecific mates include; (1) the amount of blue present on the hindwings of most of the P. glaucus females. The blue is much. more extensive than that on the hindwings of P. canadensis females and may be acting as a mate attractant. (2) the possibility that males are choosing the females of P. glaucus simply because they are novel to them, (3) That there is a certain amount of hybrid vigor present in the F1 35 crosses of these species (Scriber & Deering pers. obs.). Hybrid vigor is a well known phenomena (Collins 1984) and may manifest as increased survivorship due in part to faster developmental rates. This may account for the large proportions of hybrids in some areas such as Hyalophora hybrids in California (Collins 1984). Other evidence of hybrid vigor that often appear are increased disease resistance (Scriber & Deering pers. obs.) or the fact that hybrids between P. glaucus and P. canadensis are able to utilize the host ranges of both populations as they can detoxify both the hosts of P. canadensis, mainly Populus sp., and those of Pa glaucus, Liriodendron tulipifera, and Magnolia Virginiana (Scriber et al. 1995). Finally, (4) given that the females are multiple maters (Drummond 1984) they do not have to be selective, as their mate choices may have little overall effect as far as fertilization goes. The females of both P. glaucus and P. canadensis are known to be multiple maters (Lederhouse & Scriber 1987, 1989). Other researchers have looked at the role of sperm precedence and cryptic sexual selection in this and other mating systems (Eberhardt 1996), and are currently looking into the possibility that cryptic sexual selection by females may override sperm precedence (Stump & Scriber, unpublished). If a female can choose to use a spermatophore from a conspecific ‘male even. though. she ‘may' have a inixture of 36 spermatophores in her bursa, then male mate choice becomes less critical in terms of reinforcement. If reinforcement is taking' place between these two species it must be occurring only in the narrow confines of the hybrid zone or very near to it. We were not able to test for this reinforcement due to lack of suitable field sites in and around the hybrid zone in Michigan and this is an area that needs to be more closely scrutinized in the future. A new way that researchers may be able to tell if males of P. canadensis and P. glaucus are choosing heterospecific or conspecific mates in the hybrid zone is by looking at individual specimens collected from these areas. By running the specimens through an electrophoretic analysis designed to identify’ diagnostic allozyme markers, or Iby using' diagnostic larval traits (Scriber' 1998) it may' be determined if the individual was a hybrid or not. Additionally, maternally inherited mitochondrial DNA from the specimen would reveal the direction of the hybrid cross and gene flow. Since there are differences in the mitochondrial DNA from these two species, and since this DNA is always passed maternally we can tell if the mother of an individual was of P. glaucus or P. canadensis lineage. The data that has been recorded so far disproves the null hypothesis that mate choice by male P. glaucus and P. canadensis would be an effective form of prezygotic reproductive isolation by itself. If there were no other 37 barriers to gene flow involved in this system the hybrid zone between the two species would be spreading Southward as males of both species chose P. glaucus females for mates. In effect, there would facilitated introgression to the South as males of both P. glaucus and P. canadensis will choose P. glaucus females over P. canadensis females to mate with. Since the hybrid zone appears to have been stable for at least a 100 years (Edwards 1884), corresponding closely with the Boreal/Temperate plant transition zone and with Rothschild and Jordans (1906) and Remingtons' observations (1968), it may be more likely that other prezygotic, postzygotic or ecological factors are constraining the movement of these Papilio species across the hybrid zone. 38 LI TERATURE CITED 39 Literature Cited Arms, K., Feeny, P. & Lederhouse, R.C. 1974. Sodium: Stimulus for puddling behavior by tiger swallowtail butterflies, Papilio glaucus. Science 185:372-374 Arnold, M. L. & Hodges, S. A. 1995. Are natural hybrids fit or unfit relative to their parents? Trends in Ecology and Evolution 10:67-71. Boggs, C. L. 1981. Selection pressures affecting male nutrient investment at mating in Heliconiine butterflies. Evolution 35:931-940. Boggs, C. L. 1990. A general model of the role of male- donated nutrients in female insects' reproduction. American Naturalist 136:598-617. Boggs, C. L. & Gilbert, L. E. 1979. male contribution to egg production in butterflies: Evidence for transfer of nutrients at mating. Science 206:83-84. Bossart, J. L., & J. M. Scriber. 1995. Maintenance of ecologically significant genetic variation in the tiger swallowtail butterfly through differential selection and gene flow. Evolution 49:1163-1171 Boyden, T. C. 1976. Butterfly palatability and mimicry: Experiments with Ameiva lizards. Evolution 30:73-81. Brower, L. P. 1959. Speciation in butterflies of the Papilio glaucus group II. Ecological relationships and interspecific sexual behavior. Evolution. 13:212-228 Butlin, R. K. 1995. Reinforcement-and idea evolving. trends in Ecological Evolution 10:432-434. Collins, M. M. 1984. genetics and Ecology of hybrid zone in Hyalophora (Lepidoptera: Saturniidae). Univ. Calif. Publ. Entom. 104:1-93 Coyne, J.A. 1994. Rules for Haldane's Rule. Nature 369:189- 190. Coyne, J.A., Charlesworth, B. & Orr, H.A. 1991. Haldane's Rule Revisited. Evolution 45:1710-1714. Drummond, B. A. 1984. Multiple mating and sperm competition in the Lepidoptera. In: Sperm competition and the evolution of animal mating systems (Ed. By R. L. Smith), pp. 291-370. New York Academic Press. 40 Eberhardt, W. G. 1996. Female control: sexual selection by cryptic female choice. Princeton University Press, NJ. 501 pp. Edwards, W. H. 1884. The Butterflies of North America (Vol. II). Houghton Mifflin Company. Futuyma, D. J. & Shapiro, L. H. 1995. Hybrid Zones. Evolution 49:222-226. Hagen, R. H. 1990. Population structure and host use in hybridizing subspecies of Papilio glaucus (Lepidoptera: Papilionidae) Evolution 44:1914-1930. Hagen, R. H. and Scriber, J. M. 1989.Sex linked diapause, color and allozyme loci in Papilio glaucus: linkage analysis and significance in a hybrid zone. Journal of Heredity 80:179-185. Hagen, R.H., Lederhouse, R.C., Bossart, J.L. & Scriber, J.M. 1991.Papilio canadensis and P. glaucus are distinct species. Journal of the Lepidopterists Society. 45(4): 245-258. Hagen, R. H. & J.M. Scriber 1995. Sex Chromosomes and speciation in the Papilio glaucus group. In: The Swallowtail Butterflies: Their ecology and evolutionary biology, (Eds. J. M. Scriber, R. C. Lederhouse & Y. Tsubaki), pp. 211—228. Scientific Publishers, Inc., Gainesville, Florida. Harrison, R. G., Hybrids and hybrid zones: Historical perspective. 1993. in:Hybrid Zones and the Evolutionary Process, 1993. Edited by Richard G. Harrison. Cornell University. Oxford University Press. 364 pp. Howard, D. J. 1993. Reinforcement: origin, dynamics, and fate of an evolutionary hypothesis. In: Hybrid Zones and the Evolutionary Process, 1993. Edited by Richard G. Harrison. Cornell University. Oxford University Press. 364 pp. Jones, I. L. & Hunter, Fiona M. 1998. Heterospecific mating preferences for a feather ornament in least auklets. Behavioral Ecology. 9:187-192 Karlsson, B. 1995. Resource allocation and mating systems in butterflies. Evolution. 49(5):955-961. King, M. 1993. Species Evolution. Cambridge University Press. 41 Krebs, R. A. & West, D. A. 1988. Female mate preference and the evolution of female limited Batesian mimicry. Evolution 42: 1101-1104 Lederhouse, R.C. 1995. Comparative mating behavior and sexual selection in North American swallowtail butterflies. In: Swallowtail butterflies: Their ecology and evolutionary biology. 1995. (Eds. Scriber, J. M., Y. Tsubaki & R.C. Lederhouse) Scientific Publishers Inc., Gainesville, FL. Lederhouse, R. C. & Scriber, J. M. 1987. Ecological significance of a postmating decline in egg viability in the tiger swallowtail. Journal of the Lepidopterists Society 41(2):83-93 Lederhouse, R. C., Ayres, M. P. & Scriber, J. M. 1989. Evaluation of spermatophore counts in studying mating systems of Lepidoptera. Journal of the lepidopterists Society. 43:93-101. Lederhouse, R.C. & Scriber, J.M. 1998. Differential male preference for mimetic females in two populations of the tiger swallowtail butterfly, Papilio glaucus. Unpublished. Luebke, H. J., Scriber, J.M. & Yandell, B.S. 1988. Use of multivariate discriminate analysis of male wing morphometrics to delineate a hybrid zone for Papilio glaucus glaucus and P.g. canadensis in Wisconsin. American Midland Naturalist. 119(2):366-379. Matsumoto, K. & Suzuki, N. 1995. The nature of mating plugs and the probability of reinsemination in Japanese Papilionidae. In: Swallowtail butterflies: Their ecology and evolutionary biology. 1995. pp. 145-154 (Eds. Scriber, J. M., Y. Tsubaki & R.C. Lederhouse) Scientific Publishers Inc., Gainesville, FL. Mayr, E. 1963. Animal species and evolution. Belknap Press, Cmabridge, MA. McClintock, William J. & Uetz, George W. 1996. Female Choice and Pre-existing bias: visual cues during courtship in two Schizocosa wolf spiders (Araneae: Lycosidae). Animal Behavior 52:167-181, Part 1, July 1996 . 42 Orr, A. G. 1995. The evolution of the sphragi in the Papilionidae and other butterflies. In: : Swallowtail butterflies: Their ecology and evolutionary biology. 1995. pp. 155-164 (Eds. Scriber, J. M., Y. Tsubaki & R.C. Lederhouse) Scientific Publishers Inc., Gainesville, FL. Parker, G. A. 1970. Sperm competition and its' evolutionary consequences in insects. Biological Rev. 45: 525-567. Porter, A & Shapiro, A. 1990. Lock and key hypothesis: Lack of mechanical isolation in a butterfly (Lepidoptera: Pieridae) hybrid zone. Annals of the Ent. Soc. of America. Vol. 83, #2. Remington, C. L . 1968. Suture-zones of hybrid interaction between recently joined biotas. Evolutionary Biology, 2:321-428 Riesenberg,L.F. & Wendel, J. F. 1993. Introgression and its consequences. In: Hybrid Zones and the Evolutionary Process, 1993. Edited by Richard G. Harrison. Cornell University. Oxford University Press. 364 pp. Ritchie, M. G. & Hewitt, G. M. 1995. Outcomes of negative heterosis. In: Speciation and the recognition concept: Theory and application. Eds. Lambert, D. M. & Spencer, H. G.. john Hopkins University Press. Baltimore, MD. 502 pp. Rockey, S. J., Hainze, J.H. & Scriber, J. M. 1987. Evidence of sex-linked diapause response in Papilio glaucus subspecies and their hybrids. Physiological Ecology. 12:181-184. Rothschild, W., & Jordan, K. 1906. A revision of the American Papilios. Novitates 2001. 13:411-744. Rutowski, R. L. 1984. Sexual selection and the evolution of butterfly mating behavior. Journal of Research on the Lepidoptera. 23 (2): 125-142. Scott, J. A. 1975. Mate-locating behavior of western North American Butterflies. Journal of Research Lepidoptera. 14:1-40 Scott, J. A. 1983. Mate-locating behavior of western North American Butterflies. II. New observations and morphological adaptations. Journal of Research Lepidoptera. 21:177-187. 43 Scriber, J.M., 1996. Tiger Tales: Natural history of native North American swallowtails. American Entomologist. spring 1996, pp.19-32 Scriber, J. M. 1998. The inheritance of diagnostic larval traits for interspecific hybrids of Papilio canadensis and P. glaucus. Great Lakes Entomologist (In press) Scriber, J. M. & Bossart, J. L. 1995. Genetic variation in ovipositional preference in tiger swallowtail butterflies: interspecific, interpopulation and interindividual comparisons. In: The Swallowtail Butterflies: their Ecology and Evolutionary Biology (J. M. Scriber, Tsubaki, Y, and R. C. Lederhouse. Eds) pp. 319-344. Scientific Publishers, Gainesville Florida. Scriber, J.M., Evans, M.H. & Lederhouse, R.C. 1988. Hybridization of the Mexican tiger swallowtail, Papilio alexiares garcia (Lepidoptera: Papilionidae) with other P. glaucus group species an survival of pure and hybrid larvae on potential host plants. Journal of Research on the Lepidoptera. 27(3-4):222-232. Scriber, J. M. & Gage, S. 1995. Pollution and global climate change: Plant ecotones, butterfly hybrid zones and biodiversity. In: The Swallowtail Butterflies: their Ecology and Evolutionary Biology (J. M. Scriber, Tsubaki, Y, and R. C. Lederhouse. Eds) pp. 319-344. Scientific Publishers, Gainesville Florida. Scriber, J.M. & Lederhouse, R.C. 1988. Hand pairing of Papilio glaucus glaucus and Papilio pilumnus (Papilionidae) and hybrid survival on various food plants. Journal of Research on the Lepidoptera. 27(2):96-103. Scriber, J. M., Lederhouse, R. C. & Brown , K.. 1991. Hybridization of Brazilian Papilio Pyrrhostica)(Section V) with North American Papilio (Pterourus)(Section III). Journal of Res. Lepidoptera 29:21-32. Scriber, J. M., Lederhouse, R. C. & Dowell, R. 1995. Hybridization studies with North American swallowtails. In: Swallowtail Butterflies: Their Ecology and Evolutionary Biology. (J. M. Scriber, Y. Tsubaki and R.C. Lederhouse. Eds.) pp. 269-282. Scientific Publishers, Gainesville, Florida Scriber, J. M., Deering, M. D., Francke, L, Wehling, w. & Lederhouse, R. C. 1998. Notes on the butterfly population dynamics of three Papilio species in south central Florida (Highlands County) (Holarctic Lepidoptera, manuscript). Shields, O. 1968. Hilltopping. Journal of Research Lepidoptera. 6:69-178. Sperling, F.A.H. 1993. Mitochondrial DNA variation and Haldane's rule in the Papilio glaucus and P. troilus species groups. Heredity 71:227-233. Tidwell, P.A. 1995. Latitudinal comparisons of diapause induction between populations of Papilio glaucus and Papilio troilus (Lepidoptera: Papilionidae). MS Thesis, Michigan State University, East Lansing, MI. 80pp. Valella, P. A. & Scriber, J. M. 1998. Critical photoperiods for diapause induction in Papilio troilus and P. glaucus from different latitudes. (Manuscript). Veech, J. A., Benedix, J.H., Jr. & Howard, D.J. 1996. lack of calling song displacement between two closely related ground crickets. Evolution 50:1982-1989. Wagner, W. H., Jr. 1970. The role and taxonomic treatment of hybrids. taxon 19:146-151. White, M. J. D. 1978. Modes of Speciation. W. H. Freeman. Yamashita, K. 1995. Comparison of visual cues in mating behavior in four species of swallowtail butterflies. Pp. 133-144. In: Swallowtail butterflies: their ecology and evolutionary biology. 1995. J.M. Scriber, Y. Tsubaki & R. C. Lederhouse (eds.). Scientific Publishers, Gainesville, FL. 45 113111111111311111111111111121112111|