A dissertation submitted in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

(Biology)

The University of Michigan

1999

 326 pp.

The dense, raucous, mass emergences of North American periodical cicadas (Magicicada spp.) have attracted scientific study since the 17^th century. Periodical cicadas are unique in their combination of long life cycles (13- or 17- years) and synchronous development, leading to periodical emergences of adults. Although their long life cycles make it difficult to conduct a longitudinal study of any one periodical cicada population, because populations in different regions of eastern North America are divided into asynchronous "broods," or year classes, in almost any year, it is possible to study Magicicada. Over four years, I collected natural history information and studied the mating behaviors of 13- and 17- year Magicicada and two species in a non-periodical genus, Okanagana.

I. Mate choice in invertebrates, with special reference to periodical cicadas (Magicicada spp.).
Chapter 1 is a review chapter, and it owes a major intellectual debt to my experiences as a co-author of a book chapter about the evolution of insect mating systems. As I became involved in that project, and as I started to construct my thesis, I realized that I would have difficulty discussing mate choice unless I clarified for myself, at least, the nature of mate choice. In this chapter, I review the reasons, starting with asymmetries in parental investment, why mates should be choosy. I continue with discussions of how sexual conflicts of interest shape mating systems, and the kinds of choice mechanisms and criteria invertebrates are most likely to employ. I conclude with some predictions about possible mechanisms and functions of mate choice in periodical cicadas.

II. Experimental methods in studies of periodical cicadas: Examinations of possible methodological biases in the study of periodical cicadas (Magicicada spp.).
Chapter 2 is largely a methodological chapter. I present data demonstrating that the methods I used to mark and confine cicadas were not likely to cause mortality or behavioral changes that would bias my studies.

III. Mate guarding and the nature and possible functions of "seminal plugs" in periodical cicadas (Magicicada spp.)
Chapter 3 is a brief examination of "seminal plugs" in Magicicada. These structures are usually left behind in the female genital opening after mating. I demonstrate that these structures contain DNA and are likely composed of dried ejaculate, not other specially constructed or secreted materials. Seminal plugs do not prevent future mating attempts, but I suggest that they may be the products of antagonistic coevolution between females who are time-limited and males who can reduce threats to their paternity by imposing time-costs on females’ future mating attempts.

IV. Female multiple mating and mate choice in a periodical cicada, Magicicada septendecim.
Chapter 4 tests two hypotheses for female multiple mating in Magicicada: Females remate either to replenish depleted supplies or to effect postcopulatory mate choice. I demonstrate that interrupting a female’s mating tends to promote remating, while matings with an inappropriate, conspecific mate do not. Further, mated females do not actively seek additional mating opportunities, nor are they sexually attractive to males. Interrupted females, however, solicit second matings. These results are most consistent with the hypothesis that females remate for replenishment, and they disprove the hypothesis that females use postcopulatory choice mechanisms to make species-level mating decisions.

V. Sexual signaling in periodical cicadas, Magicicada spp.
Chapter 5 documents a previously unknown female wing flick signal, used by females to signal sexual receptivity. No such signal has previously been reported in North American cicadas, although similar signals are known in some Australian and New Zealand species. This chapter documents the signal in different Magicicada species and also describes a bizarre effect (behavioral bisexuality) of a fungus disease in males, as well as a specialized, competitive signal males use to "jam" the acoustical signals of interloping rivals. The newly described behaviors allow formulation and testing of a hypothesis for the two-part calling songs of Magicicada species such as M. septendecim . The two-part structure of these songs facilitate female discrimination of an individual male’s call against a background chorus.

VI. Reproductive character displacement and allochronic speciation in periodical cicadas with a new species, Magicicada neotredecim.
Chapter 6 describes a new 13- year species, Magicicada neotredecim. M. neotredecim male calling songs and female mate acceptance criteria are distinct from those of the similar, synchronic and partially sympatric species, M. tredecim. Perhaps the most compelling piece of evidence that M. neotredecim and M. tredecim are separate species is that M. neotredecim has undergone reproductive character displacement as a result of its interactions with M. tredecim. The existence of this new species suggests two general hypotheses for speciation in Magicicada, the "nurse brood" hypothesis, and the "canalization" hypothesis. Each could explain the origin of M. neotredecim, and each leads to specific predictions that may someday clarify how M. neotredecim came to be. This chapter ends with a prediction, based on the "nurse brood" hypothesis and patterns of reproductive character displacement, suggesting what other undiscovered Magicicada species might be like and where they might be found.

VII. Asymmetry and mating success in a periodical cicada, Magicicada septendecim (Homoptera: Cicadidae).
Chapter 7 explores the possibility that females discriminate among conspecifics on the basis of phenotypic symmetry. Symmetry has been promoted as a universal and reliable mate choice criterion. I begin by reviewing asymmetry theory in order to clarify the nature of the relationship between symmetry and mating success. I conclude that asymmetry has no special status as a mate choice criterion, but rather, it is like any other condition-dependent trait. I present the results of a study of symmetry and mating success in periodical cicadas demonstrating that males with more symmetrical forelegs are most likely to mate. However, the differential success of some males is more likely due to indirect mate discrimination than to evolved female mate choice mechanisms.

VIII. The geographic distribution and mating behaviors of two cicada species (Okanagana spp.) in northeast Michigan.
Chapter 8 compares the behaviors of Magicicada to members of another genus, Okanagana. I present the results of a 3- year biogeographic survey of the distributions of Okanagana canadensis and O. rimosa in northern Michigan. I then describe the mating behaviors of both species, which involve females approaching stationary calling males, and contrast them to those of Magicicada, which involve males searching for females. From this contrast, I suggest hypotheses for how the unique mating system of Magicicada may have evolved from mating systems such as those of Okanagana.