The Guppy Project

Introduction to the Citizen Science Project Poeciliidae

The study of mode of reproduction and mating behavior in the livebearing fish family Poeciliidae – A request for your help!


 David N. Reznick & Bart J.A. Pollux

            We seek your help in a scientific endeavor. If you read this magazine then you likely know and love the livebearing fish in the family Poeciliidae. You also know that some of them have males that are brightly colored, are ornamented with structures like swords or enlarged dorsal fins, and use elaborate courtship displays to attract the attention of females. However, not all species are like this. In some species, males have the same color patterns as females and may not court at all. Species also vary in other ways that relate to mating, like the length of their gonapodia, which are modified anal fins of males that are used to transfer sperm to females. In some species, the gonapodia are quite short while in others they can be half the body length of the male. We want to build a data set that, if possible, includes every species in the family, documents whether the males have any fancy traits and whether or not they have courtship behavior. We would also like to assemble videos of mating behavior of every species and make all of this information publically available. You may wonder why we want to do this and how you can help. We seek your help because, among you, you maintain healthy stocks of many, often very rare species. You are expert at keeping them healthy and are interested in what these fish are like. This means that it might be possible for you to make observations on your fish and perhaps make videos of their mating behavior. Our interest in male appearance and behavior has its roots in an improbable place, which is our interest in how the females reproduce.



            Twenty-six years ago one of us (DR) made a remarkable scientific discovery. It was not made as you might imagine, late at night in a laboratory or out in some remote wilderness. It was made in the library or at home reading old journal articles. I was reading papers published in the 1930’s and 1940’s about how Poeciliid fish reproduce. I had agreed to write a book chapter (Reznick & Miles, 1989) about the life histories of these fish and was now spending dreary hours over my Christmas break fulfilling an unwanted obligation. Worse still, I was in the process of getting divorced and was now living by myself in a sparsely furnished townhouse. I was less inclined to read and more inclined to look at the figures. It is against this bleak background that I discovered some remarkable illustrations in papers by C. L. Turner (Turner 1937, 1939, 1940a,b).

            What Turner described and what his figures vividly illustrated was that, while all these fish give birth to live young, very different processes are going on inside of them. Most of these livebearers were really just egg-retainers. They fully provisioning their eggs with yolk, then the eggs are fertilized and retained within the females where the embryos complete development. Other species instead had the functional equivalent of a mammalian placenta. Their tiny eggs were fertilized before they had much yolk, then the offspring grew throughout development fueled by resources transferred from the mother to developing young. Turner illustrated these differences with diagrams that showed how the dry mass of young changed during development. In species that fully provisioned their eggs before fertilization, the young lost dry mass as they developed. Their birth weights were 60 to 70% of the mass of the egg before fertilization. This is the rate of weight loss that we see in egg laying species. It represents the energy used up by metabolism and the rearrangement of molecules that happens during development. In species with placentas, the embryos gain weight during development, often following the same accelerating trajectory we see in mammals. These species also have structures on the mother and the embryo not seen in the egg retaining species. The fine details of the anatomy of these structures suggested they were specialized for transferring resources from the mother to the developing young.

            Turner was shy about calling these structures a placenta, since we think of placentas as the province of mammals. He named them instead the “follicular pseudoplacenta” because the maternal side is an elaboration of the follicle, or envelope of maternal tissue that surrounds the egg. The embryonic contribution differs among species. In some it is an elaboration of the yolk sac found in non-placental species. In others, it is an externalization of a membrane, which normally surrounds the heart. In these species, the membrane is externalized through a seam that runs down the middle of the belly. The maternal and embryonic tissues are well supplied with blood circulation and both have dense, microvilli similar to what we see in the lining of the intestine.

We break from Turner’s conservatism and call these combined structures a placenta. They fulfill a generic definition – placentas are an integration of maternal and embryonic tissues that are specialized for transferring resources from the mother to the developing baby and otherwise sustaining the metabolism of the baby. At this functional level, what we see in these Poeciliids is the same as what we see in mammals, even if the tissues are not the same.

            Why is this discovery important? We are evolutionary biologists and one of our interests is in understanding how and why complex organs evolve. A complex organ, like the eye or the placenta, is the product of many individual adaptations that must be appropriately integrated for the structure to function properly. In the case of the placenta, we are talking about an organ that fulfills the same functions as all of the major organ systems of our bodies, including gas exchange, nutrient exchange and disposal of waste products. How can nature build such complexity? Many people have argued that it cannot. To believe that evolution can bring such complexity into existence is like believing that superman can leap tall buildings in a single bound. Charles Darwin, in the Origin of Species, argued instead that complexity can evolve through a series of small steps, each of which is in some small way an improvement on its predecessor. We can instead think of the evolution of complexity as being like climbing a long, spiraling staircase, one step at a time, rather than leaping over tall buildings.

            How can we test Darwin’s proposal? We cannot do so by studying mammalian placentas. All mammals inherited their placenta from a single common ancestor that lived over 100 million years ago. That common ancestor does not represent the origin of the placenta. That event occurred sometime between 100 and 200 million years ago and dates to the common ancestor of placental mammals and marsupials, like opossums or kangaroos. Whatever was happening and however the transition occurred has long since been lost to history.

            The fish that Turner was describing were something else again. Some of the fish that either did or did not have placentas were close relatives in the genus Poeciliopsis. Some varied in how elaborate their placentas were. Others, like Heterandria formosa, were distantly related to Poeciliopsis and had structurally distinct plancentas, suggesting that they had evolved independently from the placentas in Poeciliopsis. When Turner did this work, many of the species he worked on had not even been named. His paper refers instead to Poeciliopsis a, b, c and d for some. Even now we are not sure who some of them were. Nevertheless, his discovery was remarkable and of inestimable value. Here was diversity in a complex organ among close relatives. Here also was what seemed to be complexity that had evolved independently in different lineages within one family of fish. Here was the kind of raw material that might make it possible to study the evolution of complexity. What was more amazing is that all this had been known for over 40 years, yet nothing had been done to exploit the opportunity.

            For me (DR), this was also a chance to take advantage of my new circumstances. I had more time on my hands and was more free to travel. I decided that I would adopt a new program of research. I would describe maternal provisioning in all species in the family Poeciliidae. I would take advantage of the then new (in the late 1980’s) and emerging technology of DNA sequencing to describe how they were related to one another. I would then integrate the data on how mothers provisioned their young with the family tree to learn what the patterns of evolution were throughout the family. The endeavor lead to many adventures. It became a hobby to fly into Latin American capitals, rent a car and travel the countryside collecting fish. Since fish collecting happens during the day, I could spend nights pursuing reptiles and amphibians. I filled my brain with many vivid images of exotic places, animals and plants that I would otherwise would never have seen. I saw the Amazonian rainforests of Ecuador, which I visited because I was sure there were poeciliids to be found there, even though none had been reported. I found instead only guppies and Gambusia affinis in polluted streams in some small towns. I also got to experience what it feels like to be tall and tower over the heads of a crowd when I was wandering through a market filled with Quechua speaking natives. When pursuing Xenodexia ctenolepus in Guatemala, I witnessed the horror of a newly clear cut rainforest, where every stem more than a few centimeters thick had been cut in pursuit of the one tree in a hundred that is marketable hardwood. When pursuingPhallichthys quadripunctata in Costa Rica, I saw the bright, ruby colored eyes of tree frogs reflecting in my headlamp while wandering through a swamp after a rain storm. Other adventures happened in museums, where I worked with some of the same collections Turner worked with. I can recall dodging the night guards after closing hours in the Smithsonian so that I could get more time to work with their collections, then sleeping on a lawn chair in a colleagues office so that I could start work again before opening time. All of these images and the science that followed (Reznick et al., 2002) were my reminder that good things can emerge even from the bleakest of times.

            The other of us (BP) joined the project nearly 20 years later. By that time the DNA-based family tree was near complete and we had maternal provisioning data on over 150 species of Poeciliidae. We had discovered that there was, indeed, repeated evolution of placentas within the family. We identified three groups of species that contained close relatives that either did or did not have placentas or in other ways varied in how extensively the placentas were developed. We completed a series of experiments in which we compared close relatives with and without placentas to characterize the biological consequences of having a placenta in ways that would never be possible for mammals (Banet and Reznick 2008; Banet, Au et al. 2010; Meredith, Pires et al. 2010; Pires, Arendt et al. 2010; Meredith, Pires et al. 2011; Pires, Bassar et al. 2011; Pollux and Reznick 2011; Bassar, Auer et al. 2014). It was also now possible to use this big data set to address big questions about how placentas affect other aspects of biology. Here is where the bridge was built between the evolution of placentas, the properties of males and how these fish mate with one another.

            We were motivated by the “viviparity driven conflict hypothesis” presented by David and Jean Zeh in a few articles, the first appearing in 2000 (Zeh and Zeh 2000; Zeh and Zeh 2008). They speculated about the broader consequences of maternal provisioning. If a female fully provisions an egg before it is fertilized and if she wants to choose a mate that will produce high quality children, then her only recourse is to base the choice on the appearance of the male she mates with. If instead a female provides most provisioning after the egg is fertilized, then there is a broad overlap in time between when a female provisions an egg and when the father’s genes are active participants in the development of the embryo. It becomes possible, at least in theory, for a female to choose a mate based on the genotype of the male before she has made most of her investment in the baby. Because of the differences in the timing of maternal investment, Zeh and Zeh predicted that species without placentas would place more emphasis on choosing mates before they mate with them. Such choice should favor the evolution of male display traits, like the brilliant colors, ornaments and courtship displays we see in some species of Poeciliids. In species with placentas, females should instead seek multiple mates so that they have a diversity of sperm and fertilized eggs to “choose” from. This shift in provisioning is thus predicted to cause a shift in the emphasis from choosing fathers before or after mating, but also a shift in how males differ from females and whether or not males use elaborate courtship displays to attract the attention of females.

            The possibility of such a connection became more real to one of us one day when we were wandering around a large commercial aquarium, looking at the fancy Poecillids for sale. By that time we had a good idea of what the mode of maternal provisioning was like throughout the family. What was striking was that all of the species for sale lacked placentas. This was not because they were easier to breed and maintain. Some placental species are what we describe as “hard to kill”. The real reason is that only the non-placental species have the fancy males that make them commercially attractive. The reason the males are more attractive is that their gaudy colors, ornaments and courtship behavior are the product of sexual selection, meaning that they were shaped by the females selecting who they would mate with.

            We have since completed a family-wide analysis in which we combined data on whether or not a species had males with bright coloration, ornamentation, or courtship displays. We also quantified the relative length of the male’s gonapodium and the size differences between males and females. We found, as Zeh and Zeh predicted, that the males of species with placentas were less likely to be brightly colored, have ornaments or courtship behavior. They also tended to be smaller than females and to have longer gonapodia. All of these traits are associated with species that mate covertly, by sneaking up on females. Species with placentas are also more likely to have superfetation, which means that they have multiple broods of young in different stages of development. A consequence of having superfetation is that they produce litters of young more often, but also tend to have fewer babies in each litter. There is reason to think that superfetation promotes mating with multiple males, again as predicted by Zeh and Zeh. The data we gathered so far do indeed show a relationship between sexual selection and placental reproduction, however strange that may seem.  Species without placentas are more likely to have males with gaudy colors, ornaments, courtship displays and short gonapodia.  Species with placentas are more likely to have males that have the same coloration as females, lack ornaments, lack courtship and have long gonapodia. 

            To get these data, we have had to rely on what is in the literature and on the species available in our lab. Many species are not yet accounted for. We also know that some of what is reported in the literature is not correct. For example, the published result for Poecilia gilli, a molly from Costa Rica, reports that the males have some bright coloration but that they are not ornamented nor do they have courtship behavior. We have since found that the males are ornamented, because they have enlarged dorsal and caudal fins, and that they do have courtship behavior. We realize that whether or not you see courtship might depend on the conditions in the aquarium or on how many males and females are present. There might also be differences among individual males in appearance and behavior. If only some individuals are genetically predisposed to court and if a collection was initiated with only a few males, then it is possible that the collection does not include any courting individuals. We must also remember that not all populations of a species are necessarily the same in all traits. It is possible that some populations have males that actively court females but others do not.

            We have good reasons for wanting and needing this information. Our analyses so far are compelling, in fact compelling enough for us to be able to publish them in Nature (Pollux et al., 2014), one of the most exclusive journals of science in the world, but they are also incomplete. If we can assemble more complete data, then we can address important scientific questions. One such question pertains to speciation. One surprising aspect of our results so far is that we found that the speciation rate in lineages without placentas is nearly three times faster than in lineages with placentas. Answering how and why organisms form new species is one of the most fundamentally important questions in biology. The differences in speciation rate in groups with and without placentas offers an opportunity to find out why speciation happens. It has been suggested that sexual selection causes speciation, so the non-placental lineages may have higher rates of speciation because they also have more highly developed sexual selection.

            This preamble brings us back to why we are asking for your help. We would like your help in adding to our Poeciliid database. We ask you to spend some time watching your fish, describe their different mating behaviors and make a video record of the behaviors you see. We have prepared a standard data sheet on which you can record your observations so that you can send them to us for inclusion in the database. The data sheet is printed below; a digital version of the data sheet can be downloaded from the website of Dr. Pollux and the website of Dr. Reznick’s Guppy project (details given below). The addresses where you can send the filled in data sheets are given on the standard data form. Please do not hesitate to contact one of us should you have any further questions about the data form. We will gather all submissions and compile a summary database for the project. We shall share all of our discoveries as well as photos and videos of courtship behaviors and make them available on our websites. We thank you and hope to hear back from many of you!


David and Bart 

The standard data sheet can be downloaded from the following websites:

http://www.bartpollux.nl/_Citizens_Science_Project_Poeciliidae_.html orcnas.ucr.edu/guppy/citizenscienceprojectpoeciliidae.docx


Literature Cited 

Banet, A. I., A. G. Au, et al. (2010). "Is mom in charge?  Implications of resource provisioning on the evolution of the placenta." Evolution 64(11): 3172-3182.

Banet, A. I. and D. N. Reznick (2008). "Do placental species abort offspring? Testing an assumption of the Trexler-DeAngelis model." Functional Ecology 22(2): 323-331.

Bassar, R. D., S. K. Auer, et al. (2014). "Why do placentas evolve? A test of the life history facilitation hypothesis in two clades of the genus Poeciliopsis representing two independent origins of placentas." Functional Ecology in press.

Meredith, R. W., M. N. Pires, et al. (2010). "Molecular phylogenetic relationships and the evolution of the placenta in Poecilia (Micropoecilia) (Poeciliidae: Cyprinodontiformes)." Molecular Phylogenetics and Evolution 55(2): 631-639.

Meredith, R. W., M. N. Pires, et al. (2011). "Molecular phylogenetic relationships and the coevolution of placentotrophy and superfetation in Poecilia (Poeciliidae: Cyprinodontiformes)." Molecular Phylogenetics and Evolution 59(1): 148-157.

Pires, M. N., J. Arendt, et al. (2010). "The evolution of placentas and superfetation in the fish genus Poecilia (Cyprinodontiformes: Poeciliidae: subgenera Micropoecilia and Acanthophacelus)." Biological Journal of the Linnean Society 99(4): 784-796.

Pires, M. N., R. D. Bassar, et al. (2011). "Why do placentas evolve? An evaluation of the life-history facilitation hypothesis in the fish genus Poeciliopsis." Functional Ecology25(4): 757-768.

Pollux, B. J. A. and D. N. Reznick (2011). "Matrotrophy limits a female's ability to adaptively adjust offspring size and fecundity in fluctuating environments." Functional Ecology 25(4): 747-756.

Pollux, B.J.A., R.W. Meredith, M.S. Springer MS and D.N. Reznick (2014) The evolution of the placenta drives a shift in sexual selection in livebearing fish. Nature In press.

Reznick, D.N., M. Mateos and M.S. Springer (2002) Independent origins and rapid evolution of the placenta in the fish genus Poeciliopsis. Science 298:1018-20.

Reznick, D.N. and D.B. Miles DB (1989) Review of life history patterns in poeciliid fishes. In: (Eds GK Meffe & FF Snelson Jr.) Ecology & Evolution of Livebearing Fishes (Poeciliidae), pp. 125-148.

Turner CL. 1937. Reproductive cycles and superfetation in Poeciliid fishes. Biol. Bull.72:145-64

Turner CL. 1939. The pseudoamnion, pseudochorion, pseudo-placenta and other foetal structures in vivparous Cyprinodont fishes. Science 90:42-43

Turner CL. 1940a. Pseudoamnion, pseudochorion, and follicular pseudoplacenta in Poeciliid fishes. J. Morphol. 67:59-87

Turner CL. 1940b. Superfetation in viviparous Cyprinodont fishes. Copeia 1940:88-91.

Zeh, D. W. and J. A. Zeh (2000). "Reproductive mode and speciation: the viviparity-driven conflict hypothesis." Bioessays 22(10): 938-946.

Zeh, J. A. and D. W. Zeh (2008). Viviparity-driven Conflict More to Speciation than Meets the Fly. Year in Evolutionary Biology 2008126-148.

More Information

General Campus Information

University of California, Riverside
900 University Ave.
Riverside, CA 92521
Tel: (951) 827-1012

Career OpportunitiesUCR Libraries
Campus StatusDirections to UCR

Project Information

Foundations in Integrative Biological Research (FIBR)
Funded by: National Science Foundation

Email: gupy@ucr.edu

Related Links