UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of sperm competition on testes size and intromittent organ morphology in waterfowl Coker, Chris R. 1998

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

Item Metadata


831-ubc_1998-0213.pdf [ 4.84MB ]
JSON: 831-1.0088514.json
JSON-LD: 831-1.0088514-ld.json
RDF/XML (Pretty): 831-1.0088514-rdf.xml
RDF/JSON: 831-1.0088514-rdf.json
Turtle: 831-1.0088514-turtle.txt
N-Triples: 831-1.0088514-rdf-ntriples.txt
Original Record: 831-1.0088514-source.json
Full Text

Full Text

THE EFFECTS OF SPERM COMPETITION ON TESTES SIZE AND INTROMITTENT ORGAN MORPHOLOGY IN WATERFOWL by CHRIS R. COKER B.Sc. (hon.), Trent University, 1993 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Animal Science) We accept this thesis as conforming to the required standard THE: UNIVERSITY OF BRITISH COLUMBIA April 1998 © Chris R. Coker, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract Waterfowl are one of very few avian taxa that possesses an intromittent organ (10). This thesis examines the adaptive significance of the 10 in waterfowl by determining the relationships between 10 morphology and the intensity of sperm competition (as reflected by frequency of extra-pair copulations (EPCs)). Intromittent organ morphological characteristics, including length and circumference (adjusted for body size), number of ridges and/or knobs (per unit area), ridge/knob height, ridge/knob length, and % area covered by ridges/knobs were measured from scaled museum drawings of freshly killed, sexually mature, specimens of 57 waterfowl species (across 33 genera). Thirty of which were ranked by frequency of EPC (1= monogamous, 2= rare EPC, 3= frequent EPC, 4= promiscuous). Testes sizes were also investigated in relation to EPCs, where testes masses (adjusted for body size) from 47 species (across 24 genera) were obtained (32 species with mating strategies). The size of the testes, the size (length) of the 10, the size (height) of the 10 ridges/knobs and the % area covered by ridges/knobs increased significantly with the frequency of EPC. These relationships exist even after the removal of phylogenetic constraints. These results are consistent with the hypothesis that waterfowl 10s are involved in sperm competition. Further research into the actual mechanism by which the 10 is involved with sperm competition would be worthwile. TABLE OF CONTENTS Page Abstract ii Table of Contents iii List of Tables v List of Figures •. '.. vi Acknowledgements vii Chapter 1 Introduction and Background 1 Intromittent Organs 2 Male Genitalia in Animals 2 Male Genitalia in Birds 4 Why Do Some Birds Possess or Lack an IO? 5 Male Genitalia in Waterfowl 10 Female Genitalia in Birds 11 Extra-Pair Copulation 14 Sperm Competition 16 Chapter 2 The relationship between testes size and mating strategy in waterfowl 21 Avian Testes 21 Materials and Methods 23 Results 29 Discussion 32 Chapter 3 The relationship between intromittent organ morphology and mating strategy in waterfowl 37 Sperm Competition 37 Hypothesis 38 Materials and Methods 39 Results 43 Discussion 55 iii Chapter 4 Conclusions and Future Directions 59 Conclusions 59 Study Limitations 59 Future Directions 61 A Final Thought 62 References 65 Appendix A Intromittent Organ Model 73 iv List of Tables Page Table 2.1 Table of testes masses, body masses, mating strategies, and testes data sources 25 Table 3.1 Principal components analysis 43 Table 3.2 List of species categorized according to mating strategy 46 Table 3.3 Correlation matrix of morphological characteristics 50 Table 3.4 Correlation matrix of morphological characteristics after phylogenetic constraints removed 52 v List of Figures Page Figure 2.1 The relationship between testes mass and mating strategy 29 Figure 2.2 The relationship between body mass and testes mass 30 Figure 2.3 The relationship between testes size and IO size 31 Figure 2.4 The relationship between testes size and IO size (removal of phylogenetic effects) 32 Figure 3.1 Sample of Harlequin duck and Ruddy duck intromittent organ drawings showing various characteristics 40 Figure 3.2 Principal components analysis of morphological measurements......... 45 Figure 3.3 PCA of initial 18 species categorized according to mating strategy 49 Figure 3.4 PCA with newly added species' mating strategies 51 Figure 3.5 The relationship between IO length and mating strategy 53 Figure 3.6 The relationship between the number of ridges and knobs (per unit) and mating strategy 54 Figure 3.7 The relationship between the area covered by ridges and/or knobs and mating strategy 54 Figure 3.8 The relationship between the height of the ridges/knobs and mating strategy... 55 Figure 3.9 Intromittent organ of the Ruddy duck, Oxyurini jamaicensis (left), and it's features compared to the Mealworm beetle, . Tenebrio molitor (right) 58 vi Acknowledgements This thesis could not have been completed without the efforts of Helen Hays from the American Museum of Natural History (New York). The unpublished illustrations of waterfowl intromittent organs were drawn and are owned by her; she kindly let me use them for the purpose of this study. I would like to thank my supervisory committee, Kim Cheng, Frank McKinney (University of Minnesota), Wayne Vogl, Harold Kasinsky, and Raja Rajamahendran for their support and tutelage. There are also many people who were involved in helping me on certain aspects of my thesis. I would like to express gratitude to Kevin Johnson at the University of Utah, for his help on the phylogenetic analysis; Wayne Vogl for assisting in the dissections and model casting, and Gary Bradfield for his guidance on the statistical analysis. I would also like to acknowledge Frank McKinney, Dan Brooks (Texas A & M University), and Peggi Rodgers (Wildlife Rehabilitator) who provided information on mating behaviours; and Sue Briggs (CSIRO, Canberra), Brad Millen (Royal Ontario Museum), Maryanne Hughes (UBC), and Trevor Pitcher (York University), who helped supply testes data, I would also like to thank my family and friends who showed their love and support (and tried not to snicker at my topic of research). Finally, I would like to thank my wife, Anne Hepplewhite, who was there all the way and never doubted my ability to achieve excellence. vii Chapter 1 Introduction and Background Rapid and divergent evolution in male genital morphology is expected in animals with internal fertilization (Eberhard 1985). This is due to a variety of adaptive pathways that enhance the success of self's ejaculate relative to rival ejaculates (i.e. to ensure paternity). The theory that the evolution of male genitalia results from sexual selection has been examined in the book, Sexual Selection and Animal Genitalia (Eberhard 1985), which extensively reviewed numerous studies of male genitalia in a vast array of vertebrates and invertebrates. The class Aves, however, is seldom mentioned in this book. Except for general descriptions of anatomy and physiology (e.g. Barkow 1829, Eckhard 1876, Muller 1908, Liebe 1914, Lake 1981, King 1981), studies on the function or adaptive significance of the intromittent organ (IO) are extremely rare in birds. Perhaps this is due to the fact that the majority of avian species do not have an IO, and consequently there has been little interest in carrying out research in this area. The majority of IO studies that have been carried out are on mammals, or insects. Mammals have external lOs and domestic and farm mammals are abundant and easily accessible for research. IO studies began in insects in association with sperm competition studies (e.g. Waage 1979, Michiels 1989, Miller 1991). Though most avian species lack an IO, some species do have one. This study attempts to examine the evolution and adaptive significance of the IO in waterfowl. I Intromittent Organs It has long been recognized that among closely related species with internal fertilization, the genitalia often show clear morphological differences (Eberhard 1985). This is particularly accentuated in the male of the species and is widespread throughout numerous animal groups. For example, species-specific IO are found in flatworms, nematodes, oligochaete worms, insects, spiders, millipedes, sharks and rays, some lizards, snakes, mites, opilionids, crustaceans, molluscs, and mammals (including rodents, bats, armadillos, and primates) (Eberhard 1990). In contrast, animals that have external fertilization, such as most fish, do not have species-specific genital morphology. External fertilizing groups of animals include echinoderms, most polychaete worms, hemichordates, brachiopods, sipunculid worms, frogs, few insects, and most fish (Eberhard 1990). Even as recently as 1990, birds have been categorized as not having an IO, and in fact, have even been classified as having external fertilization! (Eberhard 1990, pp. 134). Any structure that has evolved both rapidly and divergently (i.e. it acquires a new form in each new species) is a useful taxonomic character at the species level (Eberhard 1985). The universality of this pattern can be demonstrated below with a review of the genitalia from a variety of animal groups together with an examination of the complexity of many genitalia. Male Genitalia in Animals When fertilization is internal, the male, with few exceptions, develops intromittent 2 or copulatory organs for introducing sperm into the female reproductive tract. For example, the 10 of elasmobranchs are grooved, finger-like appendages of the pelvic fins known as claspers (Kent 1987) and in the anuran genus Ascaphus, the IO is a permanent tubular tail-like extension of the cloaca (Taylor & Guttman 1977). Male turtles, crocodilians, a few birds and mammals exhibit an unpaired erectile penis. In its simplest form, the penis is a thickening of the floor of the cloaca that consists chiefly of spongy erectile tissue, the corpus spongiosum, which bears a urethral groove on its dorsal surface and ends in a glans penis (Fox 1977). The urethral groove channels sperm and urine toward the vent. The erectile tissue consists of cavernous blood sinuses that, when engorged, cause the glans penis to be extruded through the cloacal aperture (Fox 1977). Traditional explanations regarding the function of the IO (i.e. gamete transfer) do not seem comprehensive enough, in light of the diverse structural variations found in many species. Gamete transfer is only one of many functions the IO has in a variety of animals. Several hypotheses have been put forth, and some have yet to be tested. For example, the IO of male cats and some male rodents is extremely elaborate and it has been demonstrated that the backwardly directed spines on their penes induce ovulation in the female through stimulation (Greulich 1934, Breed 1986). Other hypothesis include the lock and key hypothesis, where it is suggested that the shape and complexity of the male's IO has evolved to correspond precisely with the reproductive tract of the female so that insemination only occurs within the species (Eberhard 1985). 3 Hypotheses that have been looked at are those pertaining to insect species, where it is known that the function of the 10 involves more than just sperm deposition. The male of several damselfly species also uses its 10 to remove the sperm deposited in the female's sperm storage organs from previous matings (Waage 1979). These are but a few examples that show the wide variety of functional uses for the IO in a diverse collection of animals. Male Genitalia in Birds Very few avian taxa have an IO: approximately three percent (Briskie & Montgomerie 1997). Among avian species that possess an IO, there are two main anatomical varieties: the true IO (found in ratites, tinamous, cracids, screamers and waterfowl), and the non-intromittent organ (found in galliforms) (King 1981). The true IO can be further divided into two types depending on the absence or presence of a blind tubular cavity within the phallus (King 1981). Anatomically dissimilar 10s can be found in the Vasa parrots {Coracopsis spp.) (Wilkinson & Birkhead 1995), and the buffalo weaver {Bubalornis spp.) (Bentz 1983). In both genera, the 10s differ from each other and from the 10s found in waterfowl and ratites. Since the structure of the IO in Vasa parrots has not been studied in detail, it is unknown whether it is comparable to other 10s (i.e. a true IO). Similarly in the buffalo weaver, the exact role it plays in copulation is unknown; but it is structurally discimilar as it is found directly above the cloacal opening (Birkhead, et al. 1993) and is therefore not considered a true IO. Various studies (see King 1981) accept the conclusion that the extracloacal 4 paired hemipenes of snakes and lizards have evolved independently from the intracloacal phallus of chelonians and crocodiles. Gadow (1887) and Jones (1915) suggest that an obvious structural association exists between the chelonian-crocodilian type of IO and that of the ratites. In fact, the IO found in Rhea, Dromaius, Casuarius, Cracidae, and the Anseriformes, strongly resembles that of the chelonian-crocodilian IO (King 1981). In all cases, both reptilian and avian lOs consist of a left and right fibrous body separated dorsally by a median phallic sulcus. The only substantial difference between the lOs of these two groups is that the left body in the avian IO is more developed than the right, which causes the IO to angle to the left (King 1981). Another difference between reptilian and avian lOs is the mechanism of erection. In reptiles, the erectile mechanism is by engorgement with blood, whereas in waterfowl it is lymphatic (King 1981). The method of erection in the ratites is currently unknown, but since the ratite phallus is almost identical in structure to that of the waterfowl, it seems probable that the erectile mechanism will also be similar (King 1981). Why Do Some Birds Possess or Lack an IO? Due to the lack of research on avian lOs, the adaptive significance of the IO remains unknown. It is also unknown why so few bird species possess an IO; however, several hypotheses have been proposed to explain their presence or absence (see below). It is important to note that these hypotheses may not pertain to all avian species; phylogenetic constraints may have precluded the evolution of lOs in various 5 bird groups. For example, a selective force such as not permitting extra weight to be carried in flight could overshadow the selective pressures to evolve IOs. Lock and Key Hypothesis Many insects, have species specific genitalia in both females and males (Eberhard 1985). The lock and key hypothesis states that the genitalia of the male (the key) must be an exact match to copulate successfully with the female (the lock) (Eberhard 1985), thus preventing copulations with a male from a different species. A number of studies, however, have shown that it is unlikely to be true for many groups (see Eberhard 1985). Water Damage Hypothesis The water damage hypothesis suggests that an IO may be advantageous for species that copulate in water by preventing water from entering the female's cloaca and damaging or diluting the sperm (Lake 1981). This hypothesis would predict a clear pattern among aquatic versus terrestrial birds: that is, those birds without IOs would copulate on land whereas aquatic birds would copulate on the water. Although there is some support for this hypothesis in that several aquatic birds species without IOs do copulate on land, this hypothesis does not seem to be universal for all aquatic birds. For example, there are various aquatic birds (which include auklets, murrelets, pelicans, phalaropes and puffins) that copulate on water even though they lack an IO (Cramp & Simmons 1977-1983, Johnsgard 1987). The apparently successful 6 capabilities of these species, indicates that an 10 may not be necessary to prevent water damage. Sperm Transfer Assurance Hypothesis The sperm transfer assurance hypothesis implies that an 10 is favoured when copulation is difficult or inefficient because of physical constraints imposed by anatomy or environment (King 1981). For example, long-leggedness in ostriches (Struthio camelus) could be a possible physical constraint. However, this is countered behaviourally by the female crouching on the ground (Cramp & Simmons 1977-1983) to aid copulation attempts. In addition, as mentioned previously, an environmental constraint such as the inability to balance while copulating on water does not seem to pose a problem for several aquatic species without IOs. Therefore, it seems unlikely that IOs have been retained as an adaptation to counter difficulties in copulation. Sperm Competition Hypothesis It is well known that male paternal care will evolve primarily when a male's confidence of paternity is high (Winkler 1987, Knowlton & Greenwell 1984). Briskie and Montgomerie (1997) have therefore suggested that an IO would be advantageous when paternal investment is high because it would increase the male's confidence of paternity by ensuring that the sperm are deposited directly into the female's reproductive tract. There are, however, several species of waterfowl (genus Anas) where the male generally abandons the female soon after incubation begins and thus 7 does not participate greatly in paternal care (Johnsgard 1965). In contrast, many species (Jacanidae, Rostratulidae, Turnicidae and Pedionomidae) that do not possess lOs have exclusive paternal care (Cramp & Simmons 1977-1983). Flight Cost Hypothesis The flight cost hypothesis implies that lOs may have been lost as an adaptation to lessen the costs of flight (Briskie & Montgomerie 1997). This hypothesis seems to hold true when one looks at the group of ratites, a predominantly flightless group of birds (all with lOs). However, several species of waterfowl (with lOs) are strong fliers and endure annual long-distance migrations. Furthermore, on an anatomical basis, the IO is withdrawn internally when not "in use" and does not seem to weigh enough to make a significant difference to flight costs. Hohn (1960) determined that seasonal changes in IO mass occur. The IO of Mallard ducks, Anas platyrhinchos, weighs about 3 grams at it's peak: during the breeding season, but only 0.6 grams (approximately 0.05% of body weight) the rest of the year when migration would occur (Hohn 1960). Sexually Transmitted Disease Hypothesis Briskie and Montgomerie (1997) have implied that copulation in birds generally increases the chances of sexually transmitted parasites and pathogens because the openings of the gastrointestinal tract and the vagina are shared, unlike the situation in mammals where they are separate. From this, it has been assumed that the costs of contracting a sexually transmitted disease (STD) would favour the loss of the IO as a 8 way of reducing the probability of disease transmission (through a reduction in surface contact). Although little is known about the transmission and viability of STDs in birds, it seems unlikely that bird species with IOs would have benefits through intromission sufficient to outweigh any potential cost of STDs to other species without IOs. Further, although transmission of microbes occurs during copulation, both detrimental and beneficial microbes are transferred and these mostly get incorporated into the gastrointestinal tract (Lombardo et al. 1996). Therefore, it would only make sense for IOs to be lost when the costs of microbe transmission outweigh the benefits. Even so, it is very unlikely that the presence of an IO would increase the chance of and/or numbers of microbes transmitted during copulation; as they are abundant and transferred during cloacal contact regardless (Lombardo, pers. comm.). Female Choice Hypothesis The female choice hypothesis suggests that IOs have disappeared because females prefer males that do not perform forced copulations (Briskie & Montgomerie 1997). The literature indicates, however, that an IO is found in species that do not perform forced copulations (such as several species of swans, geese and shelducks; McKinney, et al. 1983). This hypothesis also seems to imply that an IO is required to perform forced copulations, but there are many species of birds that do not have IOs and are apparently capable of forcing copulation. McKinney et al. (1984), analyzed reports of extra-pair copulations in 104 different species of birds and found that 81 of 9 those species participate in forced copulations. Of those 81 species, only 39 possessed an IO (family Anatidae). Male Genitalia in Waterfowl The IO of male waterfowl is formed by an "outpocketing" of the ventral wall of the cloaca (Liebe 1914). The IO recesses into the body outside of the breeding season. The IO has been shown to increase in mass from 0.06 grams to 3 grams in Mallard ducks, during the breeding season (Hohn 1960). When not erect, it lies entirely in the cloacal cavity within a thin peritoneal sac, and when erect, is everted through the vent and is visible externally. The mechanism of erection is lymphatic and not blood-vascular (King 1981) and when erect, the base of the IO completely fills the vent (Liebe 1914). During ejaculation, semen passes along a recessed sulcus on the outside surface of the organ (Fig. 3.1) and not through an enclosed internal tube as in mammals. The IO is spiral in form and coils 3 times in the wild Mallard (pers. obs.). In many species, the surface of the IO is rough in appearance and upon closer examination transverse ridges and/or knobs appear (see Chapter 3). One obvious purpose of the IO is to guide sperm into the female's cloaca during copulation. Approximately 7 billion sperm are transferred during copulation in domestic chickens (Lake 1957). The average number of sperm in the ejaculate of several species of domestic ducks is reported to range from 2.1 to 9.3 billion (Kamar 1962). Baldassarre & Bolen (1994) hypothesized that ejaculate amounts from chickens should be similar for waterfowl, but it is likely that this assumption was based on similarity in 10 body size between domestic fowl and waterfowl. Other factors, including mating behaviours, probably have an effect on ejaculate quality and volume (Moller 1988, M0ller & Briskie 1995) (see Chapter 2). Female Genitalia in Birds In order to fully understand the adaptive significance and/or functions of the IO, one must also consider the design features of the female reproductive tract. The digestive, urinary and reproductive products of the female pass through a common chamber, the cloaca, to the opening outside, the vent. The cloaca is divided into three parts: (1) the coprodeum (next to the rectum), (2) the urodeum, and (3) the proctodeum (which empties through the vent) (Johnson 1986a). In most birds, the mucosa of the proctodeum is stratified squamous epithelium, but in waterfowl it is simply columnar (Gowaty and Buschhaus, in press). This suggests that water and other liquids can be absorbed here. Another difference between most birds and waterfowl is that in female waterfowl, a membrane covers the opening of the oviduct that is not absorbed until sexual maturity at the beginning of the first breeding season (Gowaty and Buschhaus, in press). In species without IOs, sperm is deposited directly into the urodeum (the middle cloacal chamber) as the female and male urodeums come into contact during copulation. Although it hasn't been investigated, it can be assumed that in species with IOs, the sperm is deposited closer to the opening of the oviduct. 11 The primary reproductive organ of the female is the ovary, which is responsible for the production of eggs. Only the left ovary normally develops into a functional oviduct in birds, with the right ovary being vestigial (Johnson 1986a). During the breeding season, the ovary enlarges enormously, but remains quite small at other times of the year. At its maximum size, the ovary resembles a bunch of grapes, with each "grape" forming the yolk of a complete egg. A layer of supporting cells known collectively as the follicle (Johnson 1986a) surrounds each ovum. Anatomically and functionally the oviduct is divided into five distinct segments. When the ovarian follicle reaches maturity (i.e., all the yolk is deposited), the outer cell wall ruptures and the ovum is moved into the anterior end of the oviduct, the infundibulum. The infundibulum possesses the fimbria used in grasping and guiding the ovulated ovum to the ostium of the oviduct. In ducks and many other birds, this process of ovulation occurs 15-75 minutes after the previous egg is laid (Johnson 1986b). The ovum remains in the infundibulum for about 18 minutes, which is the only time it can be fertilized (Johnson 1986b). Just distal to the infundibulum is the longest and widest segment of the oviduct, the magnum. Its surface epithelium and subepithelial tubular glands synthesize and secrete the proteins comprising the egg albumen (or "egg white"). After the ovum passes from the infundibulum into the magnum, it remains for about 4 hours (in the domestic chicken) to acquire albumen, which is deposited around the yolk (Johnson 1986a). 12 The next sequence of movements begins with passage of the ovum into the isthmus, where it remains for about 15 minutes while acquiring shell membranes, which function to shield the embryo, to conserve food and water, and to facilitate gas exchange (Gill 1990). The ovum then moves to the uterus, where it acquires a shell, and hence spends the most time (18-20 hours) (Johnson 1986a). The egg finally moves into the vagina, where mucous glands and a strong muscular wall aid its deposition into the nest. The vagina serves as a conduit between the uterus and the cloaca. When visualized grossly, the vagina appears as a thickened mass of connective tissue extending from the uterus (Bakst 1987). Though it is not considered a segment, the uterovaginal junction is significant as it contains the primary sperm storage site in the bird oviduct. Sperm Storage and Sperm Precedence Prolonged storage of sperm by female birds after mating occurs in several species, ranging from about 6 to 45 days (Birkhead 1988), and is believed to be a universal feature of avian reproduction. Following insemination sperm are stored in sperm storage tubules, located in the uterovaginal junction of the oviduct (Bakst 1987). Sperm storage tubules are typically long, narrow tubular structures with a single opening into the lumen of the oviduct (Briskie 1996). A few hours after copulation, the sperm enter the SSTs where they remain until they are used to fertilize the eggs, they die and are absorbed, or are flushed out by the egg (Johnson 1986b, Briskie 1996). Each SST can retain several hundred sperm (Birkhead et al. 1990, Briskie & 13 Montgomerie 1993) and a female can possess from 300 to 20,000 SSTs depending on the species (Birkhead & M0ller 1992b, Briskie & Montgomerie 1992, 1993). The number of SSTs a female possesses is partly (50% of the variation) due to body mass (i.e. the larger the bird, the more SSTs it has). The size of SSTs also varies across species and is correlated with sperm size (Birkhead & M0ller 1992b, Briskie & Montgomerie 1992, 1993). The adaptive significance of prolonged sperm storage in birds seems to be related to the fact that insemination and fertilization are rarely synchronized (Birkhead & M0ller 1992b). Although mammals can store sperm, they do so for shorter lengths of time than do birds (Birkhead & Moller 1992, Gomendio & Roldan 1993). An evolutionary consequence of sperm storage is that it enhances the opportunity for sperm competition: if sperm are stored prior to being used to fertilize eggs the potential exists for another male to displace sperm and replace them with his own (Parker 1970). Extra-Pair Copulation Until recently, most species of birds were considered to be monogamous (Lack 1968). Monogamy was assumed to benefit paired males the most, since they contribute to the care of eggs and young. In recent years, however, both sexes in many apparently monogamous bird species have been observed copulating with more than one partner (e.g. Cheng et al. 1983, McKinney et al. 1983, Birkhead & M0ller 1992a, Briskie & Montgomerie 1993, Gomendio & Roldan 1993). As a result, it is now known that in waterfowl, mating strategies range from exclusive monogamy to 14 promiscuity. Many species of waterfowl participate in extra-pair copulations (EPCs) in addition to maintaining a pair bond, thus practicing a Mixed Reproductive Strategy (MRS) (Trivers 1972). EPCs may be either apparently unforced or forced by the male (McKinney et al. 1983). Past studies (Cheng, era/. 1983, McKinney, et al. 1983) have suggested that in unforced EPC situations the female either passively accepts, or even actively solicits the copulations. Although there may be benefits for females to solicit EPCs (e.g. increasing genetic diversity or reduced risk of failed fertilization), the literature provides no evidence of this in waterfowl; females resist and avoid EPCs when possible (McKinney & Evarts 1998). Resistance may have advantages; when resisting EPCs, females can test male quality (Sorenson 1994). It has also been hypothesized that females resist EPCs either because they are already paired with a high-quality male, or to preserve the pairbond and investment of the mate (Sorenson 1994). The chance of fertilization by a specific male when multiple mating occurs is proportional to the relative number of sperm inseminated into a specific female; i.e. male-male competition (Miller 1991). One way for a male to increase the number of eggs fertilized by his sperm would be to mate with as many females as possible. Extra-pair copulations provide this opportunity. By maintaining the pair bond along with participating in EPCs, a male can be sure, at minimum, that he sired his mate's brood (providing he can keep other males from fertilizing her eggs); Similarly, since the costs of providing paternal care are high, strong selective pressure exists for males to fertilize the eggs of paired females other than their mate 15 (Miller 1991). When a male inseminates a paired female that is not his own mate, he increases his reproductive success since he does not have to assist in the care of those offspring (McKinney, et al. 1983). This increase in reproductive success is contingent on the progeny from the extra-pair copulation surviving to breeding age without his help. Sperm Competition The competition between sperm from two or more males to fertilize the eggs of a single female during a reproductive cycle is called sperm competition (Parker 1970). This phenomenon is widespread among both vertebrates and invertebrates and occurs when females mate with more than one male during a single reproductive cycle. The variety of mating strategies in many waterfowl species results in varying intensities of sperm competition. Recently, the term "sperm competition" has been generalized to refer to the competition between males to fertilize a specific female. Therefore, sperm competition occurs both before and after copulation or sperm transfer. Many characteristics are involved in sperm competition; which include the size, number and structure of sperm, sperm storage, mating systems, and the morphology of the reproductive systems of both sexes (Birkhead & M0ller 1992a). Before copulation sperm competition is often characterized by aggression by which paired males actively prevent other males from mating with their mate (Cade 1979). Sperm competition also influences many aspects of sexuality, not only concerning sperm and the ejaculate, but many other aspects of 16 male and female anatomy and physiology. Following copulation, sperm competition usually refers to the physiological processes occurring in the female's reproductive tract after multiple mating. An important consideration in sperm competition is the uncertainty of paternity. Increased paternal investment will result from certainty of paternity. In other words, if a male is confident that he has fathered his mate's offspring, he will put more (parental) effort into the brood to ensure their survival. Therefore, males will evolve strategies in order to assure paternity and evolve to be sensitive to cues that tell them otherwise. Several different adaptations exist in the animal kingdom. Adaptations may be behavioural, such as mate guarding (Birkhead et al. 1987); mechanical, like copulatory plugs (Thornhill & Alcock 1983) or sperm removal (Waage 1979); or a combination of both, such as frequent copulations* (Gowaty & Plissner 1987, M0ller 1988a). It has been presumed that mate guarding and frequent copulations are the methods used in birds. In waterfowl, however, the variety of mating strategies (and the resulting varied intensities of sperm competition that exist) could increase selective pressures for males to adapt other methods of assuring paternity. Sperm Precedence and Displacement Several experiments have shown that the last male to mate with a female before fertilization, is most likely to father the majority of the offspring (e.g. Compton et al. 1978, Cheng et al. 1983, Birkhead etal. 1988). This mechanism of sperm precedence * Birds that copulate frequently (behaviour) have relatively larger testes (morphology) than species which use other paternity guarding strategies (see Chapter 2). 17 (i.e. the last male to mate has precedence in fertilizations) has been interpreted in four ways: (1) stratification of sperm (Compton et al. 1978), (2) sperm displacement and/or removal (Birkhead & Hunter 1990), (3) passive sperm loss (Birkhead etal. 1988), and (4) pattern of sperm storage (Briskie 1996). Poultry biologists have long favoured the idea of stratification and a last in-first out system, because it is consistent with most observations (Van Krey et al. 1981). However, more recent studies involving the examination of the distribution of sperm within the SSTs of finches provide little empirical evidence for stratification (Birkhead & Hunter 1990). Moreover, studies on the ways in which sperm orient in the SSTs in several species shows substantial variation (Birkhead et al. 1990, Briskie & Montgomerie 1993), indicating that female reproductive tracts may differ among species in effecting sperm precedence. Using blackbirds as a model, Briskie (1996) provided evidence that last-male sperm precedence may be attributed to the patterns by which sperm is stored. Results indicated that sperm is stored in the vaginal-end of the uterovaginal junction first, and sperm from copulations occurring last will be stored at the uterine-end of the reproductive tract and will therefore be closest to the infundibulum to fertilize the next egg(s). This mechanism of sperm precedence seems to indicate that a female can control which male will sire the majority of the young by mating with a "preferred" male last. One sperm precedence mechanism that hasn't been studied in birds is sperm displacement. Sperm displacement may occur when the sperm from the most recent insemination forces out the sperm from previous inseminations (Lessels & Birkhead 18 1990). If sperm are forced into a part of the reproductive tract where they suffer a higher mortality, this will cause elevated losses of sperm from earlier inseminations and lead to last male precedence. The mechanisms by which sperm precedence occurs are known for some insects and other arthropods. It is often the morphology of the females' spermathecae and the ducts serving them that determine whether it is the first or last sperm entering which fertilizes the eggs (e.g. Ischnura graellsii (Cordero & Miller 1992)). There are, however, some interesting insect species (such as the damselfly, Calopteryx maculata) where the male "scoops" out the sperm from previous inseminations using specially modified parts of their IOs (Waage 1979). The resulting sperm displacement (removal) is approximately 88-100 percent complete (Waage 1979). Further, the IO of the beetle, Tenebrio molitor, comprises a central shaft enclosed within a flexible sheath covered with chitinous spines (Gage 1992). As the shaft extends within the female's copulatory bursa, the sheath and its covering of spines rolls back producing a 'scouring' effect (Gage 1992). A similar system is also found in several species of Odonata (dragonflies and damselflies) (Cordero & Miller 1992). Electron micrographs have shown the presence of genetically marked rival sperm amongst the spines and ridges of the male's IO after copulation (Cordero & Miller 1992, Gage 1992). This thesis examines the adaptive significance of the IO in waterfowl. It is hypothesized that relationships exist between sperm competition and both testis size and IO morphology. In this hypothesis, it is predicted that larger testes and larger, 19 more complex IOs should be found in species where sperm competition is more intense. The hypothesis is first tested with an investigation of the relationship between testis size and mating strategies (Chapter 2). The hypothesis is further tested by examining the relationship between 10 morphology and mating strategies found in waterfowl (as a measure of sperm competition). The presence of ridges and knobs on the surface of the 10 (Chapter 3) that may suggest the ability to remove sperm from previous copulations was also investigated 20 Chapter 2 The relationship between testes size and mating strategy in waterfowl Sperm competition results in various selective pressures acting on both sexes. If a particular male can increase the probability of fertilization by increasing the number of sperm transferred to the female (Martin et al. 1974), selection should favour those individuals who produce more sperm. As a result, individuals producing more sperm should possess relatively larger testes, due to the increase in seminiferous tissue (Hohn, 1947). Variations in testes size have been found in numerous invertebrate, mammal, amphibian and bird species where species involved in more intense sperm competition have larger testes than related species in less intense competition situations (Brownell & Ralls 1986, Moller 1988a, 1988b, Moller 1991, M0ller & Briskie 1995). Specifically in birds, a wide variety of species with high frequencies of extra-pair paternity have been found to have relatively larger testes than those with little or no extra-pair paternity (M0ller 1991, Moller & Briskie 1995). Avian Testes Unlike mammals, the paired testes of the male bird are located within the body cavity. Each testis contains many seminiferous tubules, which produce the sperm. During development, sperm pass through three stages. First, small cells known as spermatogonia line the wall of each seminiferous tubule (Baldassarre & Bolen 1994). These cells multiply (by mitosis) to form millions of cells within the tubules. The older 21 spermatogonia then move toward the central cavity within each tubule and begin a period of growth in which they approximately double their diameter (Johnson 1986b); at this second stage, the cells are called spermatocytes. Finally, after undergoing meiosis, the cells become spermatids that then mature into spermatazoa, each with a tail and carrying a halved load of genetic material (Johnson 1986b). The vas deferens also increases greatly in size, especially near the cloaca, during the breeding season. The spermatozoa move from the testes and accumulate in the ciliated vas deferens until ejaculated (Baldassarre & Bolen 1994). Past studies on relative testis size in birds have involved a wide variety of species from several diverse genera (M0ller 1991, Mailer & Briskie 1995), but none have focused on one particular group. Mailer's study (1991) examined testis size in relation to sperm competition, where one measure of sperm competition included mating system. The study included 18 Anseriform (waterfowl) species, but the majority of them were categorized as monogamous for mating system. Many waterfowl species, however, participate in a Mixed Reproductive Strategy (Trivers 1972), where the male not only maintains a pair bond with one female throughout the reproductive cycle, but also participate in extra-pair copulations (EPCs) with other females. A good example of the difficulty with Mailer's (1991) mating system classifications for waterfowl is found in the mallard {Anas platyrhinchos). Mailer placed this species into a monogamous mating system category - which is only partially correct. Evarts (1990) showed that mallards frequently participate in EPCs. By participating in EPCs, the intensity of sperm competition would be increased further 22 than in a species that does not participate in EPCs. It is for this reason that, in order to accurately determine the relationship between sperm competition and testes size in birds, a group like the waterfowl must be examined in more detail. The goal of this section of study, then, is to determine the relationship between sperm competition and testes size in waterfowl. With the knowledge that most species are only apparently monogamous and participate in different frequencies of EPCs, one can accurately determine the intensity of sperm competition within the group. A strong positive relationship is predicted; where as the frequency of EPCs increases across species, so does the size of the testes. The main goal of this thesis is to examine the relationships between intromittent organ (10) morphology and mating strategies in waterfowl (see Chapter 3). If associations exist between 10 morphology and mating strategies, and between testes size and mating strategies, there is conceivably a relationship between testes size and 10 morphology. Therefore, this potential relationship is examined in this section. If found, it may show that variation in testes size and 10 morphology are involved in some common role. Materials and Methods Testes Data Data were collected from the literature and museum collections on the size of testes for waterfowl species (Table 2.1). All data used were from birds that were in breeding condition. Some of the data collected, as well as some information from the 23 literature, were obtained as length and width measurements of the testes. These measurements were converted into testes mass using the equation from Mailer (1991) (revised according to Mailer & Briskie (1995)): total testes mass (g) = 2 x 1.087 g cm"3 1.33TT a 2 (cm2) b (cm), where 1.087 g cm"3 is the mean specific gravity for testes (Mailer 1991), a is the smallest and b is the largest radius of the testis (Mailer & Briskie 1995). Note that testes mass includes both testes. Waterfowl Mating Strategy Classifications Reliable references on mating strategy for waterfowl are difficult to uncover. Studies in the last 15 years have found that in many apparently monogamous bird species, both males and females copulate with more than one partner (Briskie & Montgomerie 1993, Gomendio & Roldan 1993, Birkhead & Mailer 1992b, Cheng et al. 1983, McKinney, et al. 1983). Before the MRS in waterfowl came to light, most observations on extra-pair copulations were considered unusual and rare occurrences that were either not recorded or only mentioned anecdotally in the literature. Because of this, mating strategies for all 47 waterfowl species in this study are not available and there are reliable references for only 32 species (Table 3.2). The 32 species of known mating strategy were assigned to one of 4 categories (Table 2.1). The 4 categories were (1) monogamous, (2) rare EPCs, (3) frequent EPCs, and (4) promiscuous. Species with no mating strategy information were not used (see below). 24 Table 2.1 Table of testes masses, body masses, mating strategies*, sample sizes, and testes data sources. Species Testes Mass Body Mass Mating Sample Source (g) (g) Strategy* Size (Testes Mass) Aix sponsa 0.9810 681 2 2 Royal Ontario Museum Anas acuta 9.4760 899.72 3 6 Royal Ontario Museum Anas americanas 0.9900 770 2 • Wishart 1983 Anas castanea 2.0100 595 3 25 Norman & Hurley 1984 Anas clypeata 2.2900 636 2 • M0ller 1991 Anas crecca 0.7473 360 3 • Jallageas et al. 1978 Anas cyanoptera 5.5640 408 3 1 Royal Ontario Museum Anas discors 0.6800 383 3 • M0ller 1991 Anas fulbigula 2.4000 1030 N/A 12 Allen & Perry 1979 Anas georgica 21.3400 590 N/A • M0ller 1991 Anas gibberifrons 0.6900 507 N/A • M0ller 1991 Anas platyrhinchos (Mallard) 21.4500 1144 3 5 Hohn 1947 Anas platyrhinchos (Peking) 163.0700 2921 4 6 Hughes (unpublished data) Anas rhychotis 1.3280 667 N/A 6 Braithwaite & Norman 1974 Anas superciliosa 5.7800 1114 N/A • M0ller 1991 Anas undulata 19.0000 965 3 13 Douthwaite 1977 Anseranser 4.0000 3460 1 • M0ller 1991 Anser caerulescens 17.9000 2744 3 • M0ller 1991 Anseranas semipalmata 1.5990 3300 N/A • Pitcher (unpublished data) Aytha australis 4.3620 902 N/A 2 Braithwaite & Norman 1974 Aythya affinis 4.7000 850 3 353 Anderson & Warner 1969 Aythya americana 0.0034 1080 2 • Pitcher (unpublished data) Aythya collaris 0.6950 752 2 105 Anderson & Warner 1969 Branta bernicla 1.3000 1575 2 • • M0ller 1991 Branta canadensis 4.2600 3814 1 • M0ller 1991 Branta sandvicensis 1.3700 2155 1 • M0ller 1991 Bucephala clangula 1.1000 1000 1 4 Royal Ontario Museum Table continues... 25 Species Testes Mass Body Mass Mating Sample Source (g) (g) Strategy* Size (Testes Mass) Chenonetta jubata 5.7000 815 2 7 Braithwaite & Norman 1974 Clangula hyemalis 9.4700 746 N/A • Mailer 1991 Cygnus atratus 9.1000 6270 1 12 Braithwaite 1976 Cygnus olor 7.1000 11070 1 • Breucker 1982 Dendrocygna bicolor 10.0000 680 2 11 Douthwaite 1977 Dendrocygna viduata 0.1990 670 2 1 Royal Ontario Museum Heteronetta atricapilla 11.5000 513 N/A 9 Hohn1975 Histrionicus histrionicus 1.2130 799 1 4 Royal Ontario Museum Malacorhynchus membranaceous 1.5000 404 N/A 3 Braithwaite 1976 Melanitta nigra 4.4970 950 N/A 1 Royal Ontario Museum Merganetta armata 0.4260 440 1 • Niethammer 1953 Mergus merganser 5.2500 1671 N/A 4 Royal Ontario Museum Mergus serrator 14.1900 1140 N/A • Mailer 1991 Nettapus auritus 0.0140 285 1 2 Royal Ontario Museum Oxyurini jamaicensis 10.7550 610 4 • Niethammer 1953 Plectopterus gambiensis 21.0000 5150 N/A • Mailer 1991 Somateria mollissima 5.8000 2262 1 18 Gorman 1974 Stictonetta naevosa 2.7470 842 N/A 3 Braithwaite & Norman 1974 Tadorna cana 0.2500 1527 1 1 Royal Ontario Museum * Mating strategies: 1=monogamous, 2=rare EPCs, 3=frequent EPCs, 4=promiscuous, and N/A = unknown. • Not available from reference. 26 Statistical Analyses Following a correlation test between testes mass and body mass, the testes mass measurements were standardized to unit body size (mass). Testes mass was then analyzed (ANOVA) using JMP (version 3.2, SAS Institute Inc. 1997) to determine if it was affected by mating strategy (frequency of EPCs). In addition, since a relationship between IO size and mating strategy has been demonstrated (see Chapter 3), the relationship between testes size (mass) and IO size (length) was analyzed. Phylogenetic Analysis Related species may share similar character(s) solely because they share a recent ancestor. Most statistical methods (e.g. regressions, correlations, etc.) assume that the elements are drawn independently from a common distribution (Felsenstein 1985). But, when species being analyzed arise from a common lineage, they are clearly non-independent. Using a statistical method that assumes independence will therefore cause an exaggeration of the significance in hypothesis tests (Felsenstein 1985). It is important, therefore, to circumvent this difficulty by using a statistical procedure to correct for phylogeny. Felsenstein (1985) developed a procedure (further developed into CAIC software package (version 2.0.0) by Purvis & Rambaut (1995)) that involves both the tree topology and the branch lengths of the phylogeny, and it allows the characters to be modeled by Brownian motion on a linear scale. With this procedure, the phylogeny specifies a set of contrasts among species that are 27 statistically independent and can be used in standard statistics (Purvis & Rambaut 1995). An accurate and complete phylogenetic tree of all the species in this study is not available. The phylogenies of several groups of waterfowl have been published by Livezey (1986, 1991, 1995a, 1995b, 1995c, 1996) based on morphological characters. A phylogenetic tree of dabbling ducks based on mitochodrial DNA sequence variations is also available (Johnson and Sorenson, in press). A molecular phylogeny tree describes the relations among species more accurately (Johnson, personal communication) and was used in this study. Since dabbling ducks are the biggest group of waterfowl for which the detailed molecular phylogeny is available (containing both the tree topology and the branch lengths), Felsenstein's (1985) procedure was applied to this tree and the results were generalized to other species in this study. A limitation of the CAIC software package is that it can correlate evolution between a continuous trait and a categorical one (such as mating strategy) with only two classes. For this reason, the mating strategy ranking had to be reduced to just two states. In order to accommodate all strategies, analysis was carried out twice with two different classifications of mating strategy. Mating strategies were categorized as 1) monogamous species (mating strategy 1) against all species that perform EPCs (irrespective of frequency) (mating strategies 2, 3, and 4); and 2) species which don't or rarely participate in EPCs (mating strategies 1 and 2) against species which participate in EPCs frequently (mating strategies 3 and 4). 28 The analysis yields one set of contrasts, and the Wilcoxon sign rank test was used to determine if these contrasts were significantly negative or positive indicating whether a correlated evolution between the traits exists. Results Testes Mass and Mating Strategy The effect of mating strategy on testes mass was significant (p< 01): testes mass increased with the frequency of EPCs (Fig. 2.1). There were no dabbling duck species to contrast in the monogamous versus EPC grouping of mating strategy. However, independent contrasts for monogamous and rare EPC species against frequent EPC and promiscuous species were significant and positive (Wilcoxon sign rank test: z=6.536, p<001). 0.04 -, Mating Strategy Figure 2.1 The relationship between testes mass and mating strategy. The effects of body mass removed (p< 0001). 29 80 n -10 I i i i i i i i i i 0 100 200 300 400 500 600 700 800 90 Body Mass Contrasts Figure 2.2 The relationship between body mass and testes mass. Derived from comparisons within taxa at each node of a phylogeny (removal of phylogenetic effect) (p<0001). Relationship Between Testes Mass and Body Mass Testes mass and body mass were not significantly correlated (Pearson correlation coefficient = 0.3471, p=0.251). After removing the effects of phylogeny, however, a significant positive relationship between testes mass and body mass (p<0001) was found (Fig. 2.2). Although the relationship seems to be skewed because of two contrasts, the relationship is still significant when that datapoint is removed (p<01). 30 Relationship Between Testes Size and 10 Size ANOVA results showed that a significant relationship exists between testes size (mass) and 10 size (length) (p< 001). Figure 2.3 illustrates the relationship, where testes mass increased with IO length. Analysis (ANOVA) of independent contrasts between testes size and IO size are also significant (p<0001). Figure 2.4 illustrates the relationship where evolutionary changes in testes size increased with evolutionary changes in IO size. Again, one contrast seems to be producing the significant relationship, but if removed, the relationship is still significant (p< .01). 0.06-^  0.05-<n 0.04-tn CO E >. •o 1 0.03. tn n E w £> 0.02. 0.01. i ® ® ^ ® Rfll V* tflJP w HSJ I I 1 1 1 1 1 •-— — 0.1 0.2 0.3 0.4 0.5 0.6 IO Length/Body Length Figure 2.3The relationship between testes size and IO size. Effect of body size removed (p<001). 31 0.02 r -0.002 1 1 1 1 1 1 1 1 1 1 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 Independent Contrasts of IO Length/Body Length Figure 2.4 The relationship between testes size and IO size in waterfowl. Testes mass and IO length are linear contrasts derived from comparisons within taxa at each node of a phylogeny (p<0001). Discussion In a range of waterfowl species, irrespective of body size, males that participate more frequently in EPCs have larger testes than species that either do not participate in EPCs or rarely do so (Fig. 2.1). This result is consistent with the hypothesis that an increase in sperm competition (as indicated by frequency of EPCs) results in increased testes size. The relationship can also be looked at in a different way, such that males with bigger testes are capable of performing EPCs more frequently. Whether the behaviour (EPCs) dictates the adaptation (larger testes) or the adaptation dictates the behaviour is currently unknown. 32 The Peking drake, a domestic mallard, has the largest testes and one of the largest IOs relative to its body size (see Chapter 3). Under domestication and high density rearing in large groups, many environmental constraints that affect reproduction, such as predation and food availability, are removed (Clayton 1972, Nichols et al. 1992). In addition, the pair bond breaks down, females lay up to 300 eggs a year and the males stay fertile all year. Sperm competition under this situation is very intense and would be a strong selective force to change the physiology, anatomy and behaviour of the males (Kawahara 1972, Boyce 1973). My hypothesis predicts that the Peking drake should have large testes and a large IO. Even if the Peking drake is taken out of the analysis and only wild species are considered, the effect of mating strategy on relative testes size and IO size is still significant. For birds, Mailer (1988b) determined that an increase in testes size between species produce ejaculates with more sperm per ejaculate. Mailer (1989) also showed that for mammals, two positive relationships exist: 1) between testis size and sperm production rate, and 2) between testis size and the size of the sperm reserve. Data on birds shows a similar relationship (Birkhead et al. 1993). Because testis size has a heritable component (Jones & Lamoreaux 1942), it can therefore be affected by selection (Mailer 1991). These studies seem to indicate that testis size can be used as a measure of ejaculate volume (and number of sperm) (Mailer 1991). One study that has been overlooked in much of the recent literature is a study on the seasonal changes in gonads and adrenals of the mallard duck (Hohn 1947). In this study, the weight and histological state of mallard testes were examined at various 33 periods throughout a year and described in detail. In wild species of ducks, which, unlike domestic ducks, have a restricted annual breeding season, a seasonal enlargement of the testes occurs during the breeding season (Hohn 1947). This enlargement in the testes is due to spermatogenesis (production and maturation of sperm) which does not occur outside the breeding season and only begins with an increase in testosterone levels as the breeding season commences (Hohn 1960). More specifically, the weight of the testes varies directly with seminiferous tubular activity and spermatogenesis (Hohn 1947). Other studies have also found positive relationships between testes size (mass) and sperm competition in birds in general (Mailer 1991, Briskie & Montgomerie 1992, Birkhead et al. 1993, Mailer & Briskie 1995). From these, certain hypotheses have arisen. Mailer (1991) and Birkhead et al. (1993) determined that testes size (and male sperm reserves - Birkhead et al. 1993) was positively correlated with copulation frequency per female. This hypothesis, the sperm competition hypothesis, states that when sperm competition is intense (as measured by copulation frequency), males will have relatively large sperm reserves and testes size (Mailer 1991, Birkhead et al. 1993). The increase in the size of the testes and sperm reserve is required to maintain a large sperm store for high copulation rates (Birkhead et al. 1993). Another hypothesis that has arisen, the sperm size hypothesis, examines the possibility that increased testes size and sperm reserve size is due to an increase in the size of spermatozoa (reflecting the need for a larger storage area) (Briskie & Montgomerie 1992). To date, however, with little data on sperm size in birds, there is 34 no evidence to support or contradict this hypothesis. Birkhead et al. (1993) point out that it is unlikely that sperm size alone can account for variations in testes size since there is also a correlation between testes size and sperm number. Finally, Moller and Briskie (1995) have suggested that testes size results from more intense sperm competition through extra-pair paternity. They found that testes size increased with the level of extra-pair paternity (M0ller & Briskie 1995). Similar to the sperm competition hypothesis, this hypothesis assumes that the increased testes size is due to increased sperm number. This increase in sperm number is said to aid in competing against the sperm from rival males in direct sperm competition situations. Moller & Briskie (1995) also determined that paired males, however, are not always successful in competing against rival males' ejaculates. They suggested that females may have some control over paternity and the male's are "making the best of a bad situation." My results seem to support both the sperm competition and the extra-pair paternity hypotheses (although I did not examine any of the sperm size hypotheses) for waterfowl, where intensity of sperm competition affects the size of the testes. It is difficult to compare studies, however, because the measure of sperm competition is different in all cases (i.e. copulation frequency, extra-pair paternity and frequency of EPCs). Further, my results indicate that a fourth hypothesis is possible for waterfowl. A positive relationship exists between testes size and the size (length) of the IO in waterfowl (Fig. 2.4). There is also a positive relationship between the size of the IO and frequency of EPCs (see Chapter 3). It seems likely, then that these two features 35 (testes size and 10 size) are in conjunction with each other in mating situations. Further, sperm precedence studies (e.g. Lessels & Birkhead 1990) have shown that the ejaculate from the most recent insemination may force out the sperm from previous inseminations. Once the sperm are displaced, the IO could remove them (see Chapter 3). In conclusion, the diversity of mating strategies in waterfowl species results in varying intensities of sperm competition. As a result, males in more intense sperm competition situations have larger testes, and therefore larger volumes of sperm, in order to compete with rival males. Since IO size is also positively related to both testes size and EPC frequency, the volume of the ejaculate along with the size and features of the IO may be involved in sperm removal. These ideas are examined in more detail in Chapter 4. 36 Chapter 3 The relationship between intromittent organ morphology and mating strategy in waterfowl It is clear that the main function of the intromittent organ (10) in birds is for sperm transfer, but the question of why so few bird species have an 10 remains unanswered. Several hypotheses have been suggested to explain the existence or absence of avian IOs (see Chapter 1 for more details). Unfortunately, there is very little empirical data that either supports or refutes any of these hypotheses. Furthermore, all of these hypotheses deal with why some species possess or lack an IO. In this thesis, I am concentrating on the fact that waterfowl presently have IOs and am examining them for functional significance. Sperm Competition Sperm competition, the competition between sperm from two or more males to fertilize the eggs of a single female (Parker 1970), is widespread among animals and occurs when females mate with more than one male during a single reproductive cycle. The variety of mating strategies found in waterfowl species consequently results in varying intensities of sperm competition. In situations where sperm competition is more profound, a male may be skeptical as to whether he has sired the offspring of his mate. This uncertainty of paternity has considerable bearing on paternal care. A male will be more likely to increase his parental effort if he is confident that he has sired his mate's offspring. Accordingly, 37 males will evolve tactics in order to assure paternity. In waterfowl, a Mixed Reproducive Strategy (and not monogamy) could enhance the evolution of such tactics of securing paternity. One adaptation that seems to have occurred not only in birds, but also in a variety of mammals and amphibians, is an increase in testes size with increased sperm competition (e.g. Harcourt et al. 1981, Kenagy & Trombulak 1986, Jennions & Passmore 1994). Studies have shown that large testes generally produce proportionately larger ejaculates (Moller 1988), have higher overall ejaculate production (Moller 1989), and maintain larger reserves of sperm (Birkhead et al. 1993). Chapter 2 examined this phenomenon in waterfowl, and showed that there was a positive relationship between testes size and frequency of EPCs. Hypothesis The size of the IO greatly varies between species of waterfowl and also shows a variety of surface structural differences (see below). For example, the Ruddy duck (Oxyura jamaicensis) has a very large IO with various features on the surface, while the Harlequin duck (Histrionicus histrionicus), a larger bird, has a smaller, smooth surfaced IO. It is interesting to note that the Harlequin duck is strictly monogamous (with limited number of observations) and does not participate in EPCs (Derrickson 1977) while the Ruddy duck is very promiscuous (Cramps & Simmons 1977). Since a positive relationship exists between testes size and the frequency of EPCs in waterfowl (Chapter 2), it is possible that a similar relationship may also be found between the 38 characteristics of the 10 and EPC frequency. The purpose of this section of study is to examine whether a relationship exists between the morphological features of the 10 and mating strategies in waterfowl. Materials and Methods Museum Drawings Of Waterfowl IOs Dr. Helen Hays at the American Museum of Natural History in New York, has made a set of unpublished drawings of the IOs of 57 different species of waterfowl from 33 genera. These (e.g. Fig. 3.1) were drawn to scale from freshly killed specimens in breeding condition. The "width" of the drawing is actually the circumference of the IO due to the tissue being split during dissection (Fig. 3.1). Each drawing has between 1 and 3 subsets which are magnifications of various sections along the IO (e.g. Fig. 3.9). As mentioned previously, a variety of features are found on the surface of the IO. Some species have soft cross-ridges along the length of the IO, others have cone-shaped knobs, or a combination of both ridges and knobs (Fig. 3.1). There is only one set of drawings for each species, so I assume that the specimen drawn is representative of the species. 10 Surface Morphological Measurements Several morphological measurements were taken from the drawings. 1. The length of the IO was measured as a straight line from the most proximal point on the drawing (top of Fig. 3.1) to the most distal (bottom of Fig. 3.1). 39 Figure 3.1 Sample of Harlequin duck (left) and Ruddy duck (right) intromittent organ drawings showing various characteristics (not to scale). 40 2. The circumference is simply a mean value of three measurements of the width of the drawing (see above) taken at 3 locations along the length (2 cm from the proximal end, 2 cm from the distal end, and the exact centre of the 10). 3. The number of ridges: at two random locations along the length of the 10 (1 distally, and 1 proximally), the number of ridges were counted in a centimetre and then averaged. 4. The number of knobs: from a square centimetre at two random locations (1 distally, and 1 proximally), the number of knobs were counted and the mean of the two areas was used. 5. Ridge/knob height: an index of the prominence of the ridges and knobs. This was based on a scale from 0-3, where 0 = no ridges/knobs and 3 = very prominent ridges/knobs (each value of the scale represents an equal gain in height). 6. Ridge/knob length: for some species the ridges and knobs do not completely span the circumference of the IO. Therefore, an index (from 0-2) of this ridge/knob length was used, where 0 = no ridges/knobs, 1 = "incomplete" ridges/knobs, and 2 = ridges/knobs that go around the circumference of the IO completely. 7. Area covered: this is the proportion (%) of the amount of surface area taken up by ridges and/or knobs. This measure was taken for the proximal and distal halves of the IO. 41 Classification of Waterfowl Mating Strategies As discussed in Chapter 2, reliable references on mating strategy for waterfowl are difficult to uncover. As a result, the mating strategies for all 57 waterfowl species in this section of study are not available. There are reliable references for only 30 species which were each assigned to one of the 4 mating strategy categories (Table 3.2). Statistical Analyses Principal Component Analysis (PCA) A correlation matrix only compares two variables. PCA analyzes the relationships between all of the morphological characteristics at once. Therefore, PCA provides a more complete analysis of morphological measurements. PCA was performed using NTSYS (version 1.8, Rohlf 1993). After an initial correlation between IO length and body size was calculated, the length and circumference measurements of the IO were standardized to unit body size (length). A correlation matrix was calculated between the morphological measurements and each relationship was examined for significance. PCA was carried out on the morphological measurements, followed by graphically incorporating the assigned mating strategy categories. In addition, each morphological measurement was independently analyzed (ANOVA) to determine if it was affected by mating strategy. 42 Phvlogenetic Analysis Multi-species comparative analyses are difficult as variables may be affected because species are linked through their ancestry, such that species may be similar in character solely because they share a recent ancestor. It is therefore imperative to account for and correct for phylogeny. Using Felsenstein's (1985) method of independent contrasts, it is possible to remove the "phylogenetic effect" given an accurate phylogeny with evolutionary branch lengths (see Chapter 2). As in Chapter 2, the mating strategy categories had to be reduced to two states for the analyses comparing the mating strategy and each individual morphological character. In all cases, the classification where monogamous species are ranked separately from those who perform EPCs, seemed to represent the change in EPC frequency more accurately, and is therefore displayed in the results. Results Relationship of 10 Morphological Features by Principal Component Analysis Results of PCA are presented in Table 3.1. Only the first two axes are presented because they accounted for just over 80% of the variation. The first principal component (PC1) indicated a strong relationship between the length and the circumference of the IO. The relationship between IO length and Table 3.1 Principal components analysis. Principal Component (axis) Factors I II Length 0.8994 0.2434 Circumference 0.8511 0.3942 # Ridges 0.3470 -0.7986 # Knobs -0.3666 0.7564 Variation (%) 44.7 35.6 43 circumference was expected as these two types of measurements are often analogous in anatomical features. Graphically, PC1 is represented by the X axis and accounted for just under 45% of the variance (Fig. 3.2). Increasing values on the X axis from left to right indicates an increase in the overall size of the IO (i.e. a combination of length and circumference). The second principal component (PC2), which is the Y axis, indicated a strong inverse relationship between the number of ridges and the number of knobs on the surface of the IO and made up another 36% of the variance. The negative correlation between the number of knobs (per cm2) and the number of ridges (per cm) was also predictable, since in the majority of species, there were either knobs or ridges, and rarely both. In Fig 3.2, the number of knobs increases in a positive direction along the Y axis and, inversely, the number of ridges increases in a negative direction. Relationship Between 10 Morphological Features and Mating Strategies Preliminary statistical analysis was carried out with only 18 species as 12 were added at a later date as information became available (Table 3.2). Table 3.2 illustrates all of the species in this analysis and their corresponding mating strategies. By taking the mating strategies of the initial 18 species and overlaying them on the PCA graph an apparent pattern was formed (Fig. 3.3). The monogamous species (represented by the pink triangles) fell under the small to medium IO size range and were relatively tight to the X-axis indicating that there were few to no ridges and/or knobs present on their IOs. As the frequency of EPCs increased from monogamous to promiscuous (represented 44 PCA of 10 lyjorpholoqy 37 O 2 # Knobs 4 23 48 5? © 13® 53 9 -1.5 21 51 P ^ O " 1 17 45 -»- 43 © -0.5 * 25 47 © 4 4 12 , 9 1.5 52 ©27 O 41 & 0.5 $ 15 0 36 © 35 & 11 22 49G?® PC 1 0.5© ©56 1 57 1.5 16 © 4 2 5 # Ridges 18 329 @ 19@ 50 9 PC 2 -L-1.5 ' Size (Inth/circumference) 39j ©29 33 10 © 40 -1 38 Figure 3.2 Principal component analysis of morphological measurements. IO size (length and circumference) increases from left to right, the number of knobs increases in a positive direction along the Y axis, and the number of ridges increases in a negative direction. For species labels see Table 3.2. 45 Table 3.2 List of all species studied; showing labels used in graphs, mating strategy, and mating strategy sources. Species Label used Mating Strategy* Sources Aix galericulata (Mandarin duck) • 25 3 Brooks (pers. comm.) Aix sponsa (Wood duck) 55 2 Miller (1977) Anas acuta (Northern pintail) 31 3 Derrickson (1977) Anas americanas (American wigeon) 2 McKinney (pers. comm.) Anas bahamensis (Bahama pintail) • 7 3 Sorenson (1994) Anas castanea (Chestnut teal) 3 McKinney (pers. comm.) Anas clypeata (Nothern shoveler) 32 2 McKinney (1967) Anas crecca (Green-winged teal) 3 McKinney (pers. comm.) Anas cyanoptera (Cinnamon teal) • 15 3 Brooks (pers. comm.) Anas discors (Blue-winged teal) 3 McKinney (pers. comm.) Anas flavirostris (Yellow-billed teal) 57 5 McKinney etal (1983) Anas fulbigula (Mottled duck) 28 5 Paulus (1984) Anas georgica (Yellow-billed pintail) 56 N/A Weller(1968) Anas gibberifrons (Grey teal) 20 5 McKinney et al (1983) Anas penelope (European wigeon) 19 5 Ugelvick(1986) Anas platyrhinchos (Mallard duck) 3 Evarts (1990) Anas platyrhynchos (Peking duck) 35 4 Cheng (pers. comm) Anas sibilatrix (Chiloe wigeon) 14 5 Brewer (in press) Anas sparsa (African black duck) 1 1 McKinney etal (1978) Anas superciliosa (Australian black duck) 3 5 McKinney & Evarts (1997) Anas undulata (Yellow-billed duck) 3 McKinney (pers. comm.) Anas versicolor (Silver teal) 44 5 McKinney & Evarts (1997) Anser anser (Greylag goose) 1 McKinney (pers. comm.) Anser caerulescens (Snow goose) • 46 3 Mineau & Cooke (1979) Anser indicus (Bar-headed goose) 8 1 McKinney etal (1983) Anseranas semipalmata (Magpie goose) 24 5 Davies (1990) Table continues... 46 Aytha australis (Australian white-eye duck) 6 N/A Aythya affinis (Lesser scaup) 3 Afton (1985) Aythya americana (Redhead) 39 2 McKinney & Evarts (1997) Biziura lobata (Musk duck) 29 4 Frith (1967) Branta bernicla (Brant Goose) 2 McKinney (pers. comm.) Branta Canadensis (Canada goose) 12 1 Ewaschuk & Boag (1972) Branta ruficollis (Red-breasted goose) 37 N/A Branta sandvicensis (Hawaiian goose) 1 McKinney (pers. comm.) Bucephala clangula (Common goldeneye) 17 1 Afton SSayler (1982) Cairina moschata (Muskovy duck) • 30 3 Rodgers (pers. comm.) Cereopsis novaehollandiae (Cape barren goose) 13 5 Veselovsky (1970) Chenonetta jubata (Maned goose) 26 2 McKinney (pers. comm.) Chloephaga picta (Magellan goose) • 23 1 Brooks (pers. comm.) Clangula hyemalis (Oldsquaw) 33 N/A Cygnus atratus (Black swan) 1 Kear(1972) Dendrocygna arcuata (Wandering whistling duck) 52 N/A Dendrocygna autumnalis (Black-bellied tree duck) 9 N/A Dendrocygna bicolor (Fulbous whistling duck) 2 McKinney (pers. comm.)' Dendrocygna viduata (White-faced whistling duck) • 54 1 Brooks (pers. comm.) Heteronetta atricapilla (Black-headed duck) 10 N/A Histrionicus histrionicus (Harlequin duck) 21 1 Derrickson (1977) Malacorhynchus membranaceous (Pink-eared duck) 36 N/A Melanitta nigra (Scoter) 43 5 Bengston (1966) Merganetta armata (Torrent duck) • 50 1 McKinney (pers. comm.) Mergellus albellus (Smew) • 45 3 Brooks (pers. comm.) Mergus merganser (Common merganser) 18 N/A Neochen jubata (Orinoco goose) 34 1 Brooks (pers. comm.) Netta peposaca (Rosy-billed pochard) 40 N/A Netta rufina (Red-crested pochard) 38 3 Heinroth (1911), Lind (1962) Table continues... 47 Nettapus auritus (African pigmy goose) • 2 3 Brooks (pers. comm.) Nettapus pulchellus (Australian pigmy goose) 4 N/A Nomonyx dominica (Masked duck) 27 N/A Olor columbianus (Tundra swan) 51 1 Kear(1972) Oxyura australis (Blue-billed duck) 11 5 Carbonell (1983) Oxyura jamaicensis (Ruddy duck) 41 4 Cramps & Simmons (1977) Plectopterus gambiensis (Spur-winged goose) 48 N/A Polystica stelleri (Stellers eider) 49 N/A Salvadorina waigiuensis (Salvadori duck) 42 N/A Somateria mollissima (Common eider) 16 1 McKinney (pers. comm.) Somateria spectabilis (King eider) 22 1 Ugelvick(1986) Stictonetta naevosa (Freckled duck) N/A Tadorna cana (South african shelduck) • 47 1 Brooks (pers. comm.) Tadorna tadornoides (Australian shelduck) 5 1 Cramps & Simons (1977) Thalassornis leuconotus (White-backed duck) 53 N/A * Species were placed into 1 of 4 categories based on the frequency of EPCs: (1) Monogamous, (2) Rare EPCs, (3) Frequent EPCs, and (4) Promiscuous. • Species that were added after initial analysis. 48 PCA Incorporating 18 Species with Mating Strategies # Knobs -2 # Ridges -1.5 -1 21 -0.5 "5T 16 . 1 2 A * 9-0.5 32 • 1.5 0.5 22 5 A0.5 I B31 26 3 9 ^ 9 38 -1.5 -L-2 35 : 1 41 • 1.5 A Monogamous • Rare EPCs • Frequent EPCs • Promiscuous IO Size (length/circumference)-Figure 3.3 PCA of initial 18 species categorized according to mating strategy. For species labels see Table 3.2. 49 by the red circles), the 10's became larger and had more ridges and/or knobs. After the initial PCA analysis, information on the mating strategies of 12 more species became available (see Table 3.2). By fitting these 12 species into the PCA chart, one can see if the criteria mentioned in the previous paragraph can be used to predict where the new species should fit. If predictions were true, the newly added species should fall into corresponding areas on the PCA graph. Figure 3.4 illustrates the newly added species where additions did seem to fit into the predicted areas, indicating that IO size and the number of structures roughly predicts mating strategy. Intromittent Organ Length and Body Length IO length and body length are not significantly correlated (r= -0.1139, p=0.399). A subsequent correlation test between independent contrasts of IO length and independent contrasts of body length (Phylogenetic Analysis) was also not significant (r = 0.0867, p=0.732). Therefore, waterfowl IO size did not co-evolve with body size, and is likely acted on by other factor(s). Table 3. 3 Correlation matrix of morphological characteristics. IO Length IO Circumf. # Ridges # Knobs R/K Height R/K Length Cover (D) Cover (P) IO Length 1 IO Circumf. 0.736* 1 * significant (p<0.05) # Ridges 0.194 -0.084 1 # Knobs -0.042 -0.105 -0.467* 1 R/K Height 0.466* 0.517* -0.061 0.225 1 R/K Length 0.571* 0.450* 0.115 0.242 0.681* 1 % Cover (D) 0.584* 0.459* 0.348* 0.068 0.600* 0.665* 1 % Cover(P) 0.490* 0.387* 0.212* 0.122 0.675* 0.718* 0.843* 1 50 # Knobs -1.5 # Ridges PCA with New Species 1 23 54 T -1 -0.5 45 46! 34 47 •25 • ^ 50* • 1.5 0.5 15 " 7*0.5 • _ -0.5 -1.5 1 1.5 A Monogamous T Rare EPCs • Frequent EPCs • Promiscuous IO Size (length/circumference) Figure 3.4 PCA with newly added species' mating strategies. New species are labelled and have a "*". See Table 3.2 for species labels. 51 Further Correlations of 10 Morphological features A correlation matrix of the morphological characteristics showed several correlations between the surface features and 10 size (Table 3.3). Similar to results of the PCA, there was a positive correlation between the length and the circumference of the 10. There was also a negative correlation between the number of knobs (per cm2) and the number of ridges (per cm) on the surface of the 10. Table 3.4 shows the correlation matrix of the morphological characteristics after the removal of phylogenetic effects. The relationship between 10 length and 10 circumference was still significant, but because none of the dabbling duck species have knobs, correlations involving the number of knobs (per cm2) were not estimated. Associations that differed after the removal of phylogenetic constraints were: (1) the relationship between the number of ridges (per cm) and IO size, and (2) correlation between feature size and the number of ridges (per cm). There was a significant negative correlation between these two parameters after the effects of phylogeny were removed. These relationships along with a positive correlation between feature size Table 3.4 Correlation matrix of morphological characteristics after phylogenetic contstraints removed. IO Length IO Circumf. # Ridges # Knobs R/K Height R/K Length Cover (D) Cover (P) 10 Length 1 10 Circumf. 0.782* 1 * significant (p<0.05) # Ridges -0.565* -0.827* 1 • significant (p<0.1) #Knobs R/K Height 0.873* 0.451* -0.509* 1 R/K Length 0.946* 0.668* -0.408* 0.911* 1 % Cover (D) 0.425* 0.388* -0.051 0.570* 0.566* 1 % Cover(P) 0.695* 0.451* -0.331 0.885* 0.826* 0.612* 1 52 and 10 size indicate that, at least for dabbling ducks, there were fewer but larger ridges (per cm) on larger 10s. Several characteristics were highly correlated with each other and one from each group was picked to represent a general feature. Specifically, 10 length was picked to represent 10 size, ridge/knob height to represent feature size, and % cover (distal end) to represent area covered. Further Examination Of Relationships Between 10 Morphology And Mating Strategy All morphological characteristics were independently analyzed in relation to mating strategy. Intromittent Organ Size Figure 3.5 illustrates that the size (length) of the 10 in waterfowl significantly increases as the frequency of EPCs increases (across species) (p<0001). After the removal of phylogenetic constraints, however, Wilcoxon's sign rank tests on independent contrasts for IO length and mating strategy only tended to be significant (z=1.514, p=0.06). Figure 3.5 The relationship between IO length and mating strategy (p<.0001). 53 Number of Ridges (per cm) and Knobs (per cm2) No significant relationships were found between EPC frequency and either the number of ridges (p=0.257) or the number of knobs (p=0.905) an 10 has (Fig. 3.6). The number of ridges and knobs also did not relate to mating strategy after the phylogenetic constraints were removed. Area Covered by Ridges/Knobs Figure 3.7 illustrates the significant (p=0.0117) positive relationship between area covered by ridges and/or knobs (at the distal end) and frequency of EPCs. As the frequency of EPCs increased, the area covered by knobs and/or ridges also enlarged. This relationship persisted after the removal of c 3 1_ 0) a in JQ o c * •a c ra V) a> a •o 11 10 9 8 7 6 5 4 3 2 1 0 Number of Ridges/Knobs I# Ridges !# Knobs 3 o o fi) 3 o c 73 si —i CD m o (A m 2 •v -a o 3 5' o c o Figure 3.6 The relationship between the number of ridges (p=0.257) or knobs (p=0.905) (per unit) and mating strategy. Figure 3.7 The relationship between the area covered by ridges and/or knobs (distal end) and mating strategy (p=0.0117). 54 phylogenetic constraints (Wilcoxon sign rank test of independent contrasts: z=1.935, p< 05). Ridge/knob Size The size of the ridges/knobs was not significantly related to the frequency of EPCs (p=0.085) and only indicated a trend (Fig. 3.8). A significant relationship was uncovered, however, once the constraints from phylogeny were removed (Wilcoxon sign rank test of independent contrasts: z=2.102, p<.05). Discussion When one compares the IO of the Ruddy duck with that of the Harlequin duck (Fig. 3.1), one cannot help but be amazed at the divergence in relative size and morphological features. A previous hypothesis was that the presence of the IO in birds functions to assure paternity when there is sperm competition (Briskie & Montgomerie 1997). This study presented data showing that, in waterfowl, both the relative size of the IO, the relative size of the ridges/knobs and the surface area of the IO covered with ridges and/or knobs significantly increased with the increase in intensity of sperm o o (Q 3 o c V) 73 •n a m ro ro TJ JO EP o (A uen O 01 o 3 w o c o Figure 3.8 The relationship between the height of ridges/knobs and mating strategy (p=0.085). 55 competition (as indicated by mating strategy). The last two results held even after phylogenetic constraints were removed. The relationship between the relative 10 size and mating strategy only approached significance after the removal of phylogenetic constraints. This was probably due to the small number of independent contrasts (n=4). If a larger phylogenetic tree including species outside the dabbling duck group was available for the phylogenetic analysis, the positive relationship would likely become statistically significant (Felsenstein 1985). Nevertheless, the current study demonstrated that IO size evolved independently from body size and its evolution is likely affected by a separate set of factors. The results from this study also do not support the lock and key hypothesis (Eberhard 1985); which would predict no relationship between the morphological features and intensity of sperm competition. The exact role that the IO plays in sperm competition in waterfowl is unknown. It is possible, however, that the IO may be involved in manipulating the sperm of rival males. It has been demonstrated that the last male to mate with a female is likely to father the majority of offspring in birds (e.g. Compton et al. 1978, Birkhead et al. 1988, see also Birkhead & Mailer 1992a) and certain insects (e.g. Waage 1979). Last-male sperm precedence in insects such as the damselfly, Calopteryx maculata, is controlled by the male's ability to remove sperm from previous inseminations using specially modified parts of his IO (Waage 1979). Other insect species, such as Calopteryx maculata and Tenebrio molitor, also accomplish sperm removal by virtue of spines covering the male's IOs (Waage 1979, Gage 1992). When the IO is removed from the female's copulatory bursa, sperm from rival males gets trapped within and beneath the 56 spines on the surface of the 10 (Waage 1979, Cordero & Miller 1992, Gage 1992). A comparison between the structures found on insect IOs and those found on waterfowl IOs reveals noticeable similarities (Fig. 3.9). The ridges and knobs of the waterfowl IO are all pointed towards the base of the IO, facilitating a scraping action with the withdrawal of the IO after copulation. It seems conceivable that waterfowl IOs may also be involved in sperm removal. Further research in this direction in certainly warranted. 57 Figure 3.9 Intromittent organ of the Ruddy duck, Oxyurini jamaicensis (left), and it's features compared to the Mealworm beetle, Tenebrio molitor (right). Images of beetle IO after Gage (1992). 58 Chapter 4 C o n c l u s i o n s and Future Directions Conclusions 1. Testis size is correlated with mating strategy (measured by EPC frequency) in waterfowl. This finding is consistent with the hypothesis that larger testes should be found in species where sperm competition is more intense. 2. In waterfowl, IO size and the relative size of the ridges and knobs on the IO are D correlated with mating strategy. Results presented are consistent with the hypothesis that sperm competition is a selective force behind IO morphology. 3. Waterfowl IOs may be involved in sperm removal; however, this hypothesis warrants more study. Study Limitations Drawings and Sample Sizes One limitation of this study is that only one drawing of the IO exists per species. It was therefore assumed that the specimen the drawing was taken from was a random sample that represented the species as a whole. It would be advantageous to collect more specimens of as several species from this analysis to determine whether the original was a true representative. This increase in sample size (of drawings) should not only be within species, but also between species as there are many species for which drawings do not exist. 59 The drawings of the intromittenforgans of waterfowl used in this study were very detailed (see Fig. 3.1) and provided sufficient information for analysis. However, two dimensional drawings of three dimensional objects may not give an adequate perspective when dealing with a situation such as determining a physical capability (i.e. sperm removal). For example, Mallard lOs (pers. obs.) are actually spiral shaped and not straight as indicated in the illustrations. Therefore, a three dimensional analysis of the IO may provide a more advanced quantitative analysis. Mating Strategies Another limitation of the current study is the lack of information on the mating strategies of many species. Acquisition of more mating strategy details will enhance the analysis and provide more information. Phylogeny The lineage used for phylogenetic analysis of the data consisted of only the dabbling duck species. Although this represents roughly 33% of all species examined, an obvious need for accurate phylogenies of all groups of waterfowl exists. Acquiring more detailed phylogenies as well as incorporating more information on the mating strategies of all species in question will notably strengthen the analysis. 60 Future Directions 3-D Modelling Discussions with the Biomedical Imaging and Processing Lab at the University of Minnesota revealed that quantitative analysis of three dimensional objects is conceivable using computers (Jerry Sedgewick, pers. comm.). Resin 10 models can be made from waterfowl specimens (see Appendix A). Such models could be very thinly sliced horizontally, "photographed", and then re-assembled by a computer. Once this has been accomplished, the computer will be able to quantify and analyze the size, shape, and aspect of the various surface features on the 10. Ideally, this will aid in determining the function of such features. Labelled Sperm Experiments Experiments should be conducted to determine whether sperm removal occurs in waterfowl. It has been demonstrated that subcutaneous injections of low specific activity 3H-thymidine in domestic fowl will label sperm adequately for detection with autoradiography (Van Krey, et al. 1981). Therefore, male mallard ducks can be injected with 3H-thymidine and at 21-25 days post injection, the semen should be near peak labelling of the spermatozoa (Van Krey, et al. 1981). Various mating combinations could be set up to determine where an individual's sperm from multiple males remains within the female. For example, the radioactively labelled males would be allowed to copulate with one specific female each followed by a second copulation from a different male who's sperm is not labelled. Using. 61 autoradiography techniques, it can then be determined whether the labelled spermatozoa are being displaced. As a control, several models of 10s (see Appendix A for instructions) made from rubber latex could be used in place of the second unlabelled male. Various models could also be developed to determine the precise capacity of the surface features (eg. comparing a smoothed model against an accurate casting). 10 Movement 10 morphology alone may not be able to sufficiently remove the sperm from the female. Instead, a combination of morphology and motion may be required for more "complete sperm removal. Using 10 models in experiments similar to those mentioned above, could be carried out to determine the potential of various movements of the 10. While performing artificial insemination experiments, Cheng (pers. comm.) has observed such a potential. To collect semen from waterfowl, it is ideal to stimulate the bird to ejaculate without producing an erection in the 10. The reason this is ideal, is that once the bird reaches full erection, the 10 rotates rapidly in a spiral (Cheng, pers. comm.). Therefore, it is quite possible that this spiral action is occurring within the female and needs to be investigated. It is easy to see that the results from this study are only preliminary and much more research needs to be completed. Perhaps more emphasis on the importance of studies of both male and female genitalia would prove beneficial for a better understanding on the sexual process as a whole 62 A Final Thought One question that remains unanswered is: why do the relatively "ancient" species (waterfowl and ratites) have an intromittent organ whereas more "modern" species (passerines, raptors, etc.) do not (King 1981)? This is also true in the anuran genus Ascaphus: an "ancient" group which has an IO (Taylor & Gluttman 1977). Interestingly, by examining a phylogenetic tree of the relationship between the Crocodilia and Aves (Sibley & Ahlquist 1990), it is curious to note that the Vasa parrots and the buffalo weavers seem to have "regained" the IO (Wilkinson & Birkhead 1995, Bentz 1983). King (1981) suggested that the presence of lOs in relatively "ancient" species (and the lack of lOs in most species) could be explained by phylogenetic inertia. This explanation, however, does not provide a functional explanation for the loss or retention of an IO during avian evolution. It seems plausible that waterfowl may have retained the IO for sperm competition. Whether IO morphology in other avian species is similar to that in waterfowl has not been investigated. An illustration of the IO of the Greater Rhea, Rhea americana, seems to resemble waterfowl lOs with respect to surface structures (see King 1981). Only an examination into ratite mating strategies and IO morphology can provide clues as to whether secondary IO functions similar to waterfowl exist. My results are the first to provide evidence for the adaptive significance of the intromittent organ in waterfowl. Whether male waterfowl are capable of removing the sperm of a rival male from the female's reproductive tract is unknown at this point, but 63 seems entirely possible. Prospective research with more advanced analyses of 10 morphology and mating experiments will shed light on this issue. Perhaps future research in this area will also resolve the question of why certain (and so few) species possess 10s. 64 References Allen, J.A. and H.R. Perry. 1979. Breeding chronology of Louisiana mottled ducks as indicated by gonads. Proc. Ann. Conf. S.E. Assoc. Fish & Wildl. Agencies. 33: 159-164. Anderson, B.W. and D.W. Warner. 1969. A morphological analysis of a large sample of lesser scaup and ring-necked ducks. Bird Banding. 40 (2): 85-94. Bakst, M.R. 1987. Anatomical basis of sperm-storage in the avian oviduct. Scan. Microsc. 1: 1257-1266. Baldassarre, G.A. and E.G. Bolen. 1994. Reproductive ecology. In Waterfowl Ecology and Management. John Willey & Sons, Inc.: New York. pp. 125-164. Barkow, H. 1829. Anatomisch-physiologische Untersuchungen, vorzuglich uber das Schlagadersystem der V0gel. Arch. Anat. Physiol. 306-450. Bentz, G. D. 1983. Myology and histology of the phalloid organ of the buffalo weaver (Bubalornis albirostris). Auk 100: 501-504. Birkhead, T.R., L. Atkin and A.P. Moller. 1987. Copulation behaviour of birds. Behaviour 101: 101-138. Birkhead, T.R. 1988. Behavioural aspects of sperm competition in birds. Adv. Study Behav. 18: 35-72. Birkhead, T.R. and F.M. Hunter. 1990. Mechanisms of sperm competition. Trends Ecol. Evol. 5: 48-52. Birkhead, T.R. and A.P. M0ller. 1992a. Sperm Competiiton In Birds: Evolutionary Causes And Consequences. Academic Press: London. Birkhead, T.R. and A.P. M0ller. 1992b. Numbers and size of sperm storage tubules and the duration of sperm storage in birds: a comparative study. Biol. J. Linn. Soc. 45: 363-372. Birkhead, T.R., K. Clarkson, and R. Zann. 1988. Extra-pair courtship, copulation and mate guarding in wild zebra finches Taeniopygia guttata. Anim. Behav. 35: 1853-1855. Birkhead, T.R., J.E. Pellatt, and F.M. Hunter. 1990. Numbers and distribution of sperm in the uterovaginal sperm storage tubules of the zebra finch. Condor 92: 508-516. 65 Birkhead, T.R., M.T. Stanback, and R.E. Simmons. 1993. The phalloid organ of buffalo weavers Bubalornis. Ibis 135:326-331. Boyce, R. 1973. Domestication. Psychol. Bull. 80:215-230. Braithwaite, L.W. 1976. Breeding seasons of waterfowl in Australia. Proc. 16th International Ornithological Congress, pp. 235-247. Braithwaite, L.W. and F.I. Norman. 1974. The 1972 open season on waterfowl in south-eastern Australia: Division of Wildlife Research Technical Paper. No. 29. Breed, W.G. 1986. Comparative morphology and evolution of the male reproductive tract in the Australian hydromysine rodents (Muridae). J. Zool. A209: 607-629. Breucker, H. 1982. Seasonal spermatogenesis in the mute swan (Cygnus olor). Adv. Anat., Embr. & Cell Biol. 72: 456-460. Briskie, J.V. 1996. Spatiotemporal patterns of sperm storage and last-male sperm precedence in birds. Func. Ecol. 10: 375-383. Briskie, J.V. and R. Montgomerie. 1992. Sperm size and sperm competition in birds. Proc. R. Soc. Lon. B 247: 89-95. Briskie, J.V. and R. Montgomerie. 1993. Patterns of sperm storage in relation to sperm competition in passerine birds. Condor 95: 442-454. Briskie, J.V. and R. Montgomerie. 1997. Sexual selection and the intromittent organ of birds. J. Avian Biol. 28: 73-86. Brody, S. 1945. Bioenergetics and Growth. Reinhold: New York. Brownell, R.L. Jr. and K. Ralls. 1986. Potential for sperm competition in baleen whales. Rep. Int. Whal. Comm. 8:97-112. Cade, W. 1976. Accoustically orienting parasitoids: fly phonotaxis to cricket song. Science. 190: 1312-1313. Cheng, KM., J.T. Burns, and F. McKinney. 1983. Forced copulation in captive mallards. III. Sperm competition. Auk 100: 302-310. Clayton, G.A. 1972. Effects of selection on reproduction in avian species. J. Reprod. Fert. Supplement 15: 1-21. 66 Compton, M.M., HP. Van Krey, and P.B. Siegel. 1978. The filling and emptying of the uterovaginal sperm-host glands in the domestic hen. Poultry Sci. 57: 1696-1700. Cordero, A. and P.L. Miller. 1992. Sperm transfer, displacement and precedence in Ischnura graellsii (Odonata: Coenagrionidae). Behav. Ecol. Sociobiol. 30: 261-267. Cramp, S. and K.E.L. Simmons. 1977-1983. Handbook of the Birds of Europe, the Middle East and North Africa, Vols. 1-5. Oxford University Press: Oxford. Derrickson, S.R. 1977. Aspects of Breeding Behavior in the Pintail (Anas acuta). PhD. Thesis, University of Minnesota, Mineapolis. Douthwaite, R.J. 1977. Filter-feeding ducks of the Kafue Flats, Zambia, 1971-1973. Ibis. 119: 44-66. Eberhard, W.G. 1985. Sexual Selection And Animal Genitalia. Harvard University Press: Cambridge, Massachusetts and London. Eberhard, W.G. 1990. Animal genitalia and female choice. Am. Sci. 78:134-141. Eckard, C. 1876. Ueber die Erection der Vogel. Beitr. Anat. Physiol. 7:116-125. Evarts, S. 1990. Male Reproductive Strategies In A Wild Population Of Mallards (Anas platyrhynchosj. PhD Thesis, University of Minnesota, Minneapolis. Felsenstein, J. 1985. Phylogenies and the comparative method. Am. Nat. 125: 1-15. Fox, H. 1977. The urinogenital system of reptiles. In C. Gans and T.S. Parsons (eds.) Biology of the Reptilia, vol. 6. Academic Press: New York. Gadow, H. 1887. Remarks on the cloaca and copulatory organs of the amniota. Phil. Trans. R. Soc. Ser. B 178: 5-37. Gage, M.J.G. 1992. Removal of rival sperm during copulation in a beetle, Tenebrio molitor. Animal Behav. 44: 587-589. Gill, F.B. 1990. Ornithology. W.H. Freeman and Co.: New York. Gomendio, M. and E.R.S. Roldan. 1993. Mechanisms of sperm competition: linking physiology and behavioural ecology. Trends Ecol. Evol. 8:95-100. Gorman, M.L. 1974. The endocrine basis of paired-formation behaviour in the male eider Somateria mollissima. Ibis. 116: 451-465. 67 Gowaty, P.A. and J.H. Plissner. 1987. Association of male and female American robins (Turdus migratorius) during the breeding season: paternity assurance by sexual access or mate-gurading? Wilson Bull. 99: 56-62. Gowaty, P.A. and N. Buschhaus. In press. Female resistance mechanisms to nonpreferred fertilizations predict never-before-observed traits and suggests an alterntive explanation for forced copulation in birds. Am. Zoologist. Greulich, W.W. 1934. Artificially induced ovulation in the cat {Felis domestica). Anat. Rec. 58:217-223. Hohn, E.O. 1947. Sexual behaviour and seasonal changes in the gonads and adrenals of the mallard. Proc. Zool. Soc. Lond. 117: 281-304. Hohn, E.O. 1960. Seasonal changes in the mallard's penis and their hormonal control Proc. Zool. Soc. Lond. 147: 547-555. Hohn, E.O. 1975. Notes on black-headed ducks, painted snipe, and spotted tinamous Auk. 92: 566-575. Harcourt, A.H., P.H. Harvey, S.G. Larson, and R.V. Short. 1981. Testes weight, body weight and breeding system in primates. Nature 293:55-57. Jennions, M.D. and N.I. Passmore. 1994. Sperm competition in frogs: testis size and £ 'sterile male' experiment on Chiromantis xerampelina (Rhacophoridae). Biol. J. Linn. Soc. 50: 211-220. Johnsgard, P. A. 1987. Diving Birds of North America. University of Nebraska Press: Lincoln, Nebraska. Johnsgard, P.A. 1965. Handbook of Waterfowl Behavior. Cornell University Press: Ithaca, NY. Johnson, A.L. 1986a. Reproduction in the female. In P.D. Sturkie (ed.) Avian Physiology, 4th ed. Springer-Verlag: New York. pp. 403-431. Johnson, A.L. 1986a. Reproduction in the male. In P.D. Sturkie (ed.) Avian Physiology, 4th ed. Springer-Verlag: New York. pp. 432-451. Johnson, K.P. and M.D. Sorenson. In press. Comparing molecular evolution in two mitochondrial protein coding genes (Cytochrome b and ND2) in the dabbling ducks (Tribe: Anatini). Molecular Phylogenetics and Evolution. 68 Jones, D.G. and W.F. Lamoreaux. 1942. Semen production of white leghorn males from strains selected for high and low fecundity. Poultry Science 21: 173-184. Jones, F. Wood 1915. The chelonian type of genitalia. J. Anat. Physiol. 49: 383-406. Kamar, G.A.R. 1962. Semen characteristics of various breeds of drakes in the subtropics. J. Reprod. Fertil. 3: 405. Kawahara, T. 1972. Genetic changes occuring in wild quails due to "natural selection" under domestication. Ann. Report National Inst. Genetics (Japan) 22: 111-112. Kenagy, G.J., and S.C. Trombulak. 1986. Size and function of mammalian testes in relation to body size. J. Mammal. 67:1-22. Kent, G.C. 1987. Comparative Anatomy of the Vertebrates. Times Mirror/Mosby College Publishing: St. Louis, Toronto and Santa Clara. King, A.S. 1981. Phallus. In A.S. King and J. McLelland (eds) Form and Function in Birds. Academic Press: London, pp. 107-147. Knowlton, N. and Greenwell, S.R. 1984. Male sperm competition avoidance mechanisms: the influence of female interests. In R.L. Smith (ed.) Sperm Competition and the Evolution of Animal Matings Systems. Academic Press: Orlando, pp. 61-84. Lack, D. 1968. Ecological Adaptations for Breeding Birds. Chapman & Hall: London. Lake, P. E. 1957. The male reproductive tract of the fowl. J. Anat. 91:116-129. Lake, P.E. 1981. Male genital organs. In Form and Function in Birds, (eds. A.S. King and J. McLelland) Academic Press: London, pp. 1-61. Lessels, CM. and T.R. Birkhead. 1990. Mechanisms of sperm competition in birds: mathematical models. Behav. Ecol. Sociobiol. 27:325-337. Liebe, W. 1914. Die mannliches Begattungsorgane der Hausente. Jena Z. Naturw. 51:627-696. Livezey, B.C. 1986. A phylogenetic analysis and classification of recent anseriform genera using morphological characters. Auk 103:737-754. Livezey, B.C. 1991. A phylogenetic analysis and classification of recent dabbling ducks (Tribe Anatini) based on comparative morphology. Auk 108: 471-507. 69 Livezey, B.C. 1995a. Phylogeny and evolutionary exology of modem seaducks (Anatidae: Mergini). Condor 97: 233-255. Livezey, B.C. 1995b. A phylogenetic analysis of the whistling and white-backed ducks (Anatidae: Dendrocygninae) using morphological characters. Ann. Carnegie. Mus. 64: 65-97. Livezey, B.C. 1995c. Phylogeny and comparative exology of stiff-tailed ducks (Anatidae: Oxyurini). Wilson Bull. 107:214-234. Livezey, B.C. 1996. A phylogenetic analysis of modern pochards (Anatidae: Aythyini). Auk 113: 74-93. Lombardo, M.P., P.A. Thorpe, R. Cichewicz, M. Henshaw, C. Millard, C. Steen and T.K. Zeller. 1996. Communities of cloacal bacteria in tree swallow families. Condor 98(1): 167. Martin, P.A., T.J. Reimers, J.R. Lodge and P.J. Dziuk. 1974. The effect of ratios and numbers of spermatozoa mixed from males on the proportion of offspring. J. Reprod. Fertil. 39: 251-258. McKinney, F., S.R. Derrickson and P. Mineau. 1983. Forced copulation in waterfowl. Behaviour 86: 250-294. McKinney, F., K.M. Cheng and D.J. Bruggers. 1984. Sperm competition in apparently monogamous birds. In R.L. Smith (ed.) Sperm Competition and the Evolution of Animal Matings Systems. Academic Press: Orlando, pp. 523-545. McKinney, F. and S. Evarts. 1998. Sexual coercion in waterfowl and other birds. In P. Parker & N. Birley (eds.) Male and Female Extra-pair Mating Strategies in Birds. Ornith. Monogr. 49, Am. Omithol. Union. Michiels, N.K. 1989. Morphology of male and female genitalia in Sympetrum danad (Sulzer), with special reference to the mechanism of sperm removal during copulation (Anisoptera: Libellulidae). Odonatologica 18: 21-31. Miller, P.L. 1991. The structure and function of the genitalia in Libellulidae (Odonata). Zool. J. Linn. Soc. 102: 43-73. Moller, A.P. 1988a. Ejaculate quality, testes size and sperm competition in primates. J. Human Evol. 17: 479-488. M0ller, A.P. 1988b. Testes size, ejaculate quality and sperm competition in birds. Biol. J. Linn. Soc. 33: 273-283. 70 Mailer, A.P. 1989. Ejaculate quality, testes size and sperm production in mammals. Functional Ecology. 3: 91-96. Mailer, A.P. 1991. Sperm competition, sperm depletion, paternal care, and relative testis size in birds. Am. Nat. 137 (6): 882-906. Mailer, A.P. and J.V. Briskie. 1995. Extra-pair paternity, sperm competition and the evolution of testes size in birds. Behav. Ecol. Sociobiol. 36: 357-365. Muller, R. 1908. Liber zwei verscheidene Typen in dem Bau der erectilen mannlichen Geschlechtsorgane bei den straussartigen Vbgeln. Gelesen in den kgl. Akad. Wiss. Physikal. Abhandl. 137-177. Nichols, CP., C.A.F. Fobinson and K.M.Cheng. 1992. Influence of domestication on fecundity.and reproductibe behaviour of Japanese quail {Corurnix japonica). Gibier Faune Sauvage 9: 743-755. Niethammer, G. 1953. Volewelt boliviens. Bonn. Zool. Beitre. 4: 195-305. Norman, F.I. and V.G. Hurley. 1984. Gonad measurements and other parameters from chestnut teal, Anas castanea, collected in the Gippsland Lakes region, Victoria. Emu. 84: 52-55. Parker, G.A. 1970. Sperm competition and its evolutionary consequences in the insects. Biol. Rev. 45: 525-567. Purvis, A. and A. Rambaut. 1995. Comparative analysis by independent contrasts (CAIC): an Apple Macintosh application for analysing comparative data. Computer Appl. Biosci. 11:247-251. Rohlf, F.J. 1993. NTSYS-pc, Version 1.8. Applied Biostatistics: Setauket, New York. SAS Institute, Inc. 1997. MP, Version 3.2. SAS Institute, Inc.: Cary, NC. Short, R.V. 1977. Sexual selection and the descent of man. In J.H. Calabya and CH. Tyndale-Biscoe (eds.) Proceedings of the Canberra Symposium on Reproduction and Evolution. Australian Academy of Sciences: Canberra, pp. 3-19. Sibley, C.G. and J.E. Ahlquist. 1990. Phylogeny and Classification of Birds. Yale University Press: New Haven and London. Taylor, D.H. and S.I. Guttman. 1977. The Reproductive Biology of Amphibians. Plenum Press: New York. 71 Thornhill, R. and J. Alcock. 1983. The Evolution of Insect Mating Systems. Harvard University Press: Cambridge, Massachusetts and London. Trivers, R.L. 1972. Parental investment and sexual selection. In B. Campbell (ed.) Sexual Selection and the Descent of Man, 1871-1971. Aldine, Atherton: Chicago, pp. 136-179. Van Krey, H.P., R.J. Balander and M.M. Compton. 1981. Storage and evacuation of spermatozoa from the uterovaginal sperm-host glands in the domestic fowl. Poultry Sci. 60: 871-878. Waage, J.K. 1979. Dual function of the damselfly penis: sperm removal and transfer. Science 203 (2): 916-918. Welty, J.C. 1982. The Life of Birds. Saunders College Publishing: New York, NY. Wilkinson, R. and T.R. Birkhead. 1995. Copulation behaviour in the Vasa parrots Coracopsis vasa and C. nigra. Ibis 137: 117-119. Winkler, D.W. 1987. A general model for paternal care. Am. Nat. 130: 526-543. Wishart, R.A. 1983. Pairing chronology and mate selection in the American wigeon (Anas americana). Can. J. Zool. 61: 1733-1743. 72 Appendix A Intromittent Organ Model Chapter 4 discussed the need for three dimensional (3-D) quantification of waterfowl IOs. The shape of the 10 in waterfowl is quite complex (see Chapter 3) and there are many structures (ridges and/or knobs) on the surface. Waterfowl have IOs that are corkscrew shaped with at least three rotations (personal observations). It is because of these various elaborate characteristic that an analysis of a two dimensional (2-D) drawing is subject to inaccuracy. A 3-D model can be analyzed on a precise, detailed level through digital imaging (Jerry Sedgewick, pers. comm.), where the model can be sliced, photographed and re-built into a computer. This thesis has examined the adaptive and functional significance of the 10 in waterfowl. A hypothesis that has been proposed is that the 10 and it's surface morphology are involved in sperm removal (Chapter 3). In order to determine whether sperm removal is occurring, mating experiments need to be performed. In these experiments, several controls could be employed to determine the precise function of the surface features as well as possible roles of movement during copulation. For this, 3-D models made from resin or latex.could be used. Models could also be altered, for example, to increase the intensity of the corkscrew, produce a straight 10, or produce a smooth surfaced 10. Through the course of this thesis, I have determined that 3-D models of waterfowl IOs are possible. The details on making 10 models are described below. 73 Knowledge of the methods used to collect semen for artificial insemination purposes provided the technique with which to produce an erection on a (live) Mallard* . Erection of the 10 simply occurs when muscular contractions around two lymph bodies cause lymph fluid to be forced into the 10 (King 1981). The resulting pressure causes the organ to evert (like an inverted glove) and become erect. By gently massaging the back and the abdomen at the same time and gradually putting pressure around the base of the cloaca, erection can be induced (Cheng et al. 1983) (Fig. 1). Once erection was achieved, the drake was euthanized, while the IO remained everted. This specimen was then used for casting the model. After death of the bird, the IO becomes flaccid due to lack of muscular pressure on the lymph nodes. Since this pressure • cannot be easily replicated, the cavity within the IO can be injected with clear liquid plastic casting resin, which will harden and provide the "erection" from which a,cast can be made. Materials and Methods The following describes the materials and procedures needed to create a model of a Mallard IO. Preparing 10 for Mold Casting • Massage the drake to produce erection according to method described in Cheng et al., 1983 (Fig. 1). * This technique, although very successful on domestic species and birds which are acclimatized to being handled by humans, may not be successful on wild birds (Cheng, pers. comm.). 74 • Inject the bird through the wing vein with a lethal dose of Euthanol (1 ml_ per Kg body mass) and ensure the IO doesn't regress back into the cloaca. • Mix the liquid plastic with the catalyst (liquid hardener) (25 drops per ounce of resin) to start the chemical reaction which ultimately cures (hardens) the resin. • While the resin is still in liquid form, inject inside the IO using a syringe (10cc, 19 G needle) (Fig. 2). • Completely fill the IO with resin and maintain pressure with the syringe (Fig. 2). Note that if the syringe is removed before the resin thickens, there will be seepage (through the hole made by the syringe) and ultimately reduced size of the IO. • As the resin begins to cure, position the IO as required for final model. The syringe can be removed after approximately 2 hours. • Allow to completely cure for 24 hours (minimum). Casting the Mold • Once cured, carefully disect the IO from the bird's body to allow for unconstrained manipulation while casting the mold. • The mold is made using liquid latex rubber. Brush 10 to 12 coats of liquid.rubber onto IO (allowing each coat to dry) and allow to set for 24 hours (minimum) after last coat is applied. The IO can also be dipped into the liquid rubber, but will require a much longer drying time (approx. 72 hours). 75 • Carefully peel the mold off of the 10. Because of the complex shape of the 10, I split the mold down one side to ease removal (this will eventually have to be joined. together again for the model casting). Producing the Final 10 Model • Once a mold is created, it can be filled with casting resin (liquid plastic as mentioned above). • Simply mix the resin with the catalyst and fill the mold. • Allow to completely cure (approx. 72 hours) and carefully remove model from mold (Fig. 3). Discussion A resin model of a Mallard 10 (Fig. 3) can be used for further morphological quantification (see above). If required, the above steps can be altered in order to create a model for a specific use. For example, the casting resin can be coloured (although still translucent) or a non-translucent form of resin can be used. It is also likely that a plaster mould could be taken from the 10 from which a latex model of the 10 could be created and manipulated for mating experiments (see above). 76 Figure 2 Injecting resin into Mallard 10. Figure 3 Model of Mallard 10. 77 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items