Bowie MD, November 1, 2002. This is a first for me. I've never datelined a webpage before. I never met Stephen Jay Gould (1942-2002), but for 30 years he was my teacher and mentor, through 20 books and 300 magazine articles. I would love to have seen how he would have handled this story in his column in Natural History magazine. I hope that Carl Zimmer, SJG's worthy successor, picks up this theme and writes about it.

You will find dates at the bottom of all my webpages. They record the time of last updating of the text or when I last checked all the links to make sure that you find something if you click on one. But I'm dating this page so you can look back and check the progress over the next few years that, I promise, is going to take place. I'd wager good money on this one. We are about to learn about the origin of novelty, and how evolution produces a complex organ.

Charles Darwin came close to despair over a couple of issues bearing upon his several theories of evolution. Perhaps the biggest issue to confront him was time; how to demonstrate that deep time, sufficient time to allow for the evolution of all life, had really existed. He called this problem that "odious spectre" hanging over his theory. I have never been certain whether it was time itself that he thought was the odious spectre, or whether he meant Lord Kelvin, who was then considered to be the expert on the age of the Earth. When Kelvin lowered his estimate from 100 to 20 million years, Darwin despaired. The discovery of radiation later disproved Kelvin, but Darwin did not live to taste the sweetness of  knowing that evolution has been going on for almost 4 billion years. Science (particularly geology, paleontology and paleoanthropology) underwent a paradigm shift when radiometric dating matured as a methodology and firmly established the age of the Earth at 4.56 ± 0.02 billion years.

Similarly, Darwin was frustrated by his inability to show living physical evidence for the gradual change of species or for the origin of complex organs in nature. The evolution of complex eyes, which he handled effectively by use of analogy, is but one example. (you can read Darwin on the Eye from both 1st and 6th editions of The Origin of Species). Darwin proposed that complex eyes could have been formed with a succession of photosensitive organs, each a bit more complex than its predecessor and each favored by natural selection because of the advantage that accrued to the possessor. Visualizing such a process would be easiest if steps in this sequence were preserved in closely related living organisms; however, no such sequence exists for eyes because the intermediate stages have been lost through extinction. Our best alternative has been to compare distantly related species that display what appear to be intermediate stages in evolution, as was done by Darwin with examples from across the mollusk or arthropod phyla. However, a suitable complex organ has now been found in the genus Poeciliopsis, small fishes known as guppies, topminnows or livebearers.

Placentas, Like Eyes, Are Complex

Fishes are ancient animals. Amphibians, reptiles, birds, mammals -- all the Tetrapods descend from them. Placental reproduction has evolved repeatedly in all these groups. The mammalian placenta has been far better characterized than that of any other placental organism and serves as a model for the complexity of the adaptation. More than 50 genetic loci regulate the development of the placenta while additional adaptations include loci that are members of gene families that produce protein hormones and hemoglobin but are only expressed in the fetus and are specifically adapted to fetal development. Other adaptations include facilitation of gas exchange, the transfer of nutrients, disposal of waste products, and the suppression of immunological interactions or other forms of intergenomic conflict between the mother and embryo. The placenta is also like the eye because it appears that all of these features must be present in order for it to function successfully.

Gila Topminnow -- Poeciliopsis occidentalis

There are more than 25,000 species of fishes living in all the aquatic environments on this planet. You'd think that they had long ago settled every aspect of organic design, perfected in every way for the environments in which they live. But, they can't stop evolving any more than can other plants and animals. Fishes have been tinkering with mating and reproductive systems for over 500,000 years. The Cichlid Fishes provide us with many examples of recent rapid evolution and radiation of species (and of mating systems and behaviors). However, catching a species in the act of organic change is difficult. Speciation, even in "fast" lineages, might take thousands of years, while researchers have careers spanning only a few decades. But the Poeciliids now present us with the opportunity to find out something about how they did it.

The genus Poeciliopsis may be divided into three groups of fishes based upon their reproductive behavior (they are listed below). Remember first that they are livebearers. In all cases in this genus the eggs develop within the mother until the young are ready to be released into the external environment. In some cases the eggs are created with lots of yolk -- enough to last for the entire development of the young prior to their release. Since the mother does not need to feed the young during development, she has no need for a placenta in which to house her eggs. Nine species in the genus employ this reproductive system. This group is shown in black in the Matrotrophy Index, which shows degrees of post-fertilization maternal provisioning, and which is more fully explained below.

Another group of Poeciliids might be called the full placentals. They develop eggs with little food prestored in them. The mother requires a well-developed placenta in which to house all her eggs and into which she must supply food during the period of internal development after fertilization. It is
also through the placenta that she must arrange for waste removal and gas exchange. Six species in the genus employ this arrangement. They are shown in red in the Matrotrophy Index here and below.

Finally, we have an intermediate group of five species (blue) with less demanding requirements. They produce somewhat yokey eggs but also need to supply some food during internal development. This group also has intermediate placentas that are less-fully developed in comparison to the obligate placentals. Whether they are in the process of moving toward full placentalship or have settled on an intermediate arrangement has not yet been determined. See the discussion below where the Matrotrophy Index is again presented and explained more fully and more technically.

What is evident though is the fact that, for some species of Poeciliopsis, there has been a paradigm shift with respect to reproductive behavior. The ancestors of these fishes were non-placentals with no need for that complex organ. They simply released eggs into the water and all post-fertilization development took place outside the mother's body. How is it then that their closely related descendent species have come to have a range of novel and complex placental organization? Their genes will help to tell the story. I am willing to bet that, as a result of the work done so far, many scientists will be encouraged to delve into this chance of a lifetime to find out how these fish have changed over time. I am also willing to bet that they will find that relatively small genetic changes are required to produce significant organic and behavioral changes.

Adding Complexity With Superfetation

Now, let's tip the scale of complexity discussed so far. I thank David Reznick for informing me about another feature of these fishes not discussed in the paper in Science. The Poeciliidae have a second variation on the theme of live bearing called superfetation. Females of some species carry more than one brood of young in distinctly different stages of development. As with matrotrophy, superfetation is not simply present or absent. Some species characteristically carry two litters of young in different stages of development. Others can carry three, four or more litters. Species that carry more simultaneously developing litters give birth more frequently to fewer young per litter. Superfetation thus changes the distribution of births over time. To help me describe this process Dr. Reznick has kindly supplied the photos and their descriptions, and some of the more technical narrative on this page.
 

Dr. Reznick's research group has preliminary data indicating at least five independent origins of superfetation in Poeciliopsis and closely allied species. Also, it may be found independently of or in conjunction within post-fertilization maternal provisioning. So, in a relatively few species of fishes we find another novelty -- production line reproduction -- variously combined with a variety of strategies for feeding developing young. The obvious implication is that natural selection can find numerous pathways to success. And, for the moment, we won't even consider the additional complexities suggested by hybridization among some of these fish.

Poeciliopsis gracilis -- Photo by Jacques Blanc

The subdiscipline of biology known as evolutionary development, or Evo-Devo for short, has become an important player in the discovery of how things work. By tinkering with the genes of model organisms (the lab mouse, the plant Arabidopsis and fruit flies have been popular) scientists have been learning how control genes function, the critical nature of timing in gene expression, and many other roles played by genes in the evolution of organisms. In some cases it has even been possible to discover things that transpired hundreds of millions of years ago. By a kind of reverse engineering process in which modern genes are silenced or suppressed, scientists have gained insight as to how vertebrate jaws were modified to become what they are today.

It may be expected that experiments with Poeciliopsis will tease apart the genes that distinguish the placental differences in the three groups of species, changing some to more developed placentas or others back to less developed ones. It may also be expected that, given this example of opportunity, researchers working with other fishes or with altogether different organisms will be led to discover similar opportunities never before dreamed of within their own fields of study.

Paradigm shifts occur not only in fishes. They occur also in the thinking of scientists who learn new ways of looking at nature. I wish Steve Gould could have told you this story in his maximally effective manner. We probably would have visited the Louvre or the Sistine Chapel somewhere along the way.

Frequency histogram of MI values for all species in the genus Poeciliopsis. Mothers that bear live young can provision those young either before or after the egg is fertilized. Many organisms that bear live young actually provide all resources before fertilization, so it is as if they were egg layers that simply retained the eggs inside them during development. Others provide varying amounts of resources to the young after fertilization. The placenta is an elaborate organ that transfers resources from the mother to the young after fertilization. These scientists developed a metric called the "matrotrophy index" (MI) to quantify the way mothers provision their offspring. MI is equal to the estimated dry mass of young at birth divided by the dry mass of the egg at fertilization (they dry the eggs and embryos so that changes in the amount of water present does not affect the results). If all resources are provided prior to fertilization, then the birth weight is around 60-70% of the egg weight (MI = 0.6 to 0.7). Weight loss is caused by the costs of metabolism. If mothers provide resources after fertilization, then less weight may be lost and there may even be a weight gain. These scientists applied this index to all species in this genus and found that there were species with MI values that ranged from around 0.6 to 1.6, then the next highest value was 5.4. Those with values of 5.4 and higher were classified as having extensive post-fertilzation provisioning (and highly evolved placentas), those with values that were significantly greater than 0.7 and less than 5.4 as having rudimentary placentas, and those with values that did not differ significantly from 0.7 as having no post-fertilization maternal provisioning. There is an overlap in the MI values of those that do not differ significantly from 0.7 and those that do because the statistic is also a function of sample size and of the degree to which individuals within a sample vary from one another.

Maximum likelihood phylogram for the genus Poeciliopsis, based on mitochon-drial gene sequences (see Science paper for genes used and references cited). Bootstrap values for clades supported above the 50% level are shown adjacent to branches. MI values for all taxa, not including outgroups, are included in parentheses next to each species. Each species was coded as either having no evidence of post-fertilization maternal provisioning (gray), moderate amounts of maternal provisioning (blue; MI range from 0.86 to 1.58), or extensive maternal provisioning (MI > 5). Species coded in red or blue have MI values that are significantly greater than 0.7. The red branches associated with P. prolifica, P. presidionis, and P. turneri are those associated with estimates of the rate of evolution of extensive maternal provi-sioning. Asterisks denote branches on which extensive maternal provisioning evolved in the genus Poeciliopsis.

Note that the species coded in red (fully placental) are nested in clades within clades. The ancestral clades do not have placentas. Under a theory of descent with modification placentas evolved at least three times in these fishes. Superfetation also appears to have evolved independently in these fishes. See the text and photos above for details.

D. Reznick et al. : Independent Origins and Rapid Evolution of the Placenta in the Fish Genus Poeciliopsis. Science 298 (5595):1018-1020, 1 Nov 2002. This is the original publication that stimulated the creation of this webpage. Temporary .pdf file 110k.

Placentas May Nourish Complexity Studies : by V. Morell, Science 298 (5595):945. This is a news article discussing the importance of the paper by Reznick et al. Available in libraries and through online subscription.

Mateos et al. : Historical Biogeography of the Livebearing Fish Genus Poeciliopsis. Evolution 56 (5):972-984 (2002). This is the original paper presenting the phylogram data shown above. Free access online.

Evaluation of the Rate of Evolution in Natural Populations of Guppies : Reznick et al., Science 275 (5308):1934-1937 (1997). Another research project by David Reznick and colleagues in which natural populations of guppies were subjected to an episode of directional selection that mimicked natural processes. The resulting rate of evolution of age and size at maturity was similar to rates typically obtained for traits subjected to artificial selection in laboratory settings and up to seven orders of magnitude greater than rates inferred from the paleontological record. Temporary .pdf file 347k.

Mating Systems and Energetics of Reproduction : This is an outline for a college course in which the genera Goodeidae and Poeciliidae are indicated as examples of true viviparity, and exhibit vascularized pericardial tissue loops around necks of embryos, maintaining close contact with the ovarian wall.