With over 800 species, North America has the most diverse temperate freshwater fish fauna in the world. The region containing both the greatest number of species and the highest amount of endemism (i.e., native to, and restricted to, a particular geographic area) is the Central Highlands (Fig. 1), which is divided into three distinct geographical areas: the Ouachita Highlands in Oklahoma and Arkansas, the Ozarks in Missouri, and the Eastern Highlands, with the the Shawnee Hills of southern Illinois forming a "land bridge" connection between the Ozarks and Eastern Highlands.
Within the Central Highlands, the most species-rich area is a physiographic region termed the Highland Rim and Nashville Basin Drainage Realm (HRDR). The HRDR (Fig. 1) encompasses the lower Tennessee River system, the Cumberland River drainage below Cumberland Falls, most of the Green River system, and a few additional watersheds. With over 300 species of fishes, the HRDR harbors the most diverse freshwater fish fauna of any temperate region of comparable size on the planet, and probably has the most diverse mussel and crayfish fauna.
The Central Highlands of North America. (from Mayden 1987, Kansas Geological
Survey Guidebook Series. Quaternary Environments, Pleistocene Glaciation and
Historical Biogeography of North American Central-Highland Fishes. W.C. Johnson,
editor. 208 pp.)
Within the Highland Rim, as throughout the Central Highlands, the fishes showing the most complex distribution patterns are the species inhabiting the small headwater streams. These species include darters, minnows, and madtom catfishes. The adults breed and spend their whole lives in the headwater streams and, even though the young-of-the-year may get washed into larger streams, they usually move back up to the headwaters within a few months. These fishes cannot complete their life cycles in the larger riverine habitats, which are almost as much a dispersal barrier to them as dry land.
Most intriguing about the HRDR, and the Central Highlands in general, are the patchwork distributions of many of the native organisms. For example, the fringed darter, Etheostoma crossopterum, (Fig. 2) occurs throughout the lower Cumberland River system, but is limited in the Tennessee River system to small sections of the Duck River, Buffalo River, and Shoal Creek systems. Additionally, there are two disjunct populations occurring in streams draining directly into the Mississippi River in western Tennessee.
Distribution of fringed darter, Etheostoma crossopterum.
With little knowledge of the biology of the species, it is difficult to imagine how such a distribution could have resulted. One hypothesis might be that the fringed darter once was dispersed throughout the Cumberland and Tennessee River systems but has undergone widespread extirpation in the Tennessee system. Fortunately, we do have an understanding of the life history characteristics of the fringed darter, and we know that the species is incapable of long-distance dispersals. An alternative hypothesis is that the present-day drainage systems did not always exist (see Fig. 3), and that the species' disjunct distribution represents ancient stream patterns.
Hypothesized preglacial drainages in eastern North America superimposed on
existing drainage patterns. (From Mayden, 1988, Vicariance biogeography,
parsimony, and evolution of North American freshwater fishes. Systematic Zoology,
137(4).)
The factors affecting the distributions of species include biological phenomena, such as behavior and habitat selection, and geological phenomena, such as plate tectonics and stream captures. Geological events can split the range of an ancestral species and isolate populations that subsequently differentiate. Clearly, the geological evolution of drainages and biological evolution of native aquatic organisms are closely related, and studying one should provide information about the other. Unfortunately, the geologic data alone are insufficient to reconstruct detailed patterns of historical drainage connections. With this in mind, researchers at INHS are studying the "phylogeo-graphy," or the combination of biogeographic and phylogenetic information (i.e., hypotheses of species relationships) of native species, in an attempt to better understand how species came to exhibit their present-day distribution.
In phylogeographic analysis (Fig. 4), the first step is to construct phylogenies for the various groups of endemic organisms (Fig. 4A, species A-I; species B and C are most closely related, and these two form the "sister group" to species A). The second step is to develop taxon-area cladograms - simply replace the names of the species with the names of the areas in which they occur. In Fig. 4B, species A, D, and G occur in Yellow Creek, species B, E, and H occur in Blue Creek, and species C, F, and I occur in Red Creek, so the resulting taxon-area cladogram for each independent group of organisms (darters, minnows, and madtoms) looks like Fig. 4C. The final step involves combining the various taxon-area cladograms to see whether they show patterns of congruence. In our simplified example, all three groups of fishes share the identical phylogeographic pattern (Fig. 4C).
Because a vicariant event, such as stream capture, affects several taxa simultaneously, the resulting pattern should be reflected in the distributions of several species. If distribution patterns are due entirely to dispersal, similar habitats in continuous (connected) streams and drainages (Fig. 4B, Yellow and Blue creeks) should possess biotas that are most closely related to each other, whereas adjacent but discontinuous stream systems (Fig. 4B, Blue and Red creeks) can be expected to contain more distantly related groups of organisms. However, if vicariant events have contributed to distribution patterns, adjacent but discontinuous streams are expected to contain organisms that are most closely related to each other (e.g., Species E and F, H and I).
Phylogeographic analysis. See above text for explanation.
Looking again at the fringed darter, it is more parsimonious (simple) to hypothesize that populations of the fringed darter were transferred from tributaries of the Cumberland River to tributaries of the Tennessee River (or vice-versa) through headwater capture events than it is to assume the species was once widespread but has undergone numerous extirpation events. The now highly disjunct populations of the fringed darter inhabiting streams in western Tennessee are populations that survive in the only remaining suitable habitat in western Tennessee where the Tennessee and Cumberland rivers once flowed.
We must have a thorough understanding of the distributions and phylogenetic relationships of the various groups of endemic organisms in order to understand the impact of historical factors on the present-day distributions of these organisms. While examining populations of two species of fishes related to the spottail darter (Etheostoma squami-ceps), a fish present in the streams of the Shawnee Hills of southern Illinois, Survey scientists recently discovered and described five new species of darters inhabiting the HRDR. Also, Survey scientists just completed work that indicates that populations of the orangethroat darter (E. spectabile), another inhabitant of Illinois waters, actually represent a composite of six undescribed species occurring in the HRDR (one species, E. forbesi, was named after Stephen A. Forbes, first Chief of INHS). Using a combination of molecular techniques, researchers are beginning the task of examining geneological relationships that will allow them to develop phylogenies for these endemic groups of fishes. The resultant phylogeographic data will then be used to develop the first detailed explanations of the evolutionary history of the present-day river systems of the HRDR and offer explanations regarding the complex distribution patterns of fishes.
Patrick A. Ceas and Lawrence M. Page, Center for Biodiversity
Next Article
Index
to This Issue
Index to
Survey Reports
INHS Home Page
Please report any problems with or suggestions about this page to:
inhspubs@mail.inhs.uiuc.edu
Subject: INHSPUB-00416
Last Modified 7/03/96