The Coral Reef: Symbol of Biodiversity The term "biodiversity" inspires tiiougiits of coral reefs and rain forests, and tliese are, indeed, the two most biologically diverse ecosystems on planet Earth. One difference between the two is that we can actually "see" more of the reef diversity. A majority of rain forest inhabitants are tiny insects and animals that dwell overhead in the canopy or lower among dense foliage, usually hidden from observers. By con- trast, coral reef life forms flourish colorfully and abun- dantly in front of a human visitor's face mask. Reefs, themselves, are composed of algae, corals, gorgonians, anemones, zooanthids, and other benthic organisms that create habitat. Reef surfaces swarm with fishes and tiny Crustacea, abundant mollusks, and worms. There are, of course, many more tiny ani- mals and plants living secretly in reef crevices and deep holes that human visitors ....-^ ., .-. rarely see. Many of these tiny ^ •i^Jji^vJ-Jril?'^:^'^^ organisms, like rain forest -:'.I^^^^i^^^^^^ insects, are still scientifi- . •-•xJ^^-^?^.^^SBS*^^?-=^ss caUy unknown, and J^^^^^m^^^^ no one has yet accurately esti- mated the total numbers of different species living on reefs. We do know, however, that parts of the reef ecosys- tem depend upon one another and that the reef eco- system depends upon neighboring systems. For ex- ample, reef herbivores, such as the long-spined sea urchin, consume algae that might otherwise smother living corals. The larval stages of many reef animals drift as plankton in the open ocean, and there are also reef connections to sea grass meadows, sand flats, intertidal zones, and mangroves. Thus the complex web of reef life intertwines with the larger lattice of Earth's biodiversity. -Phillip S. Lobel, Icthyologist Boston University Marine Program r^ -: ■:/'-' T ^ G ^ ^ G ^ C ^ ^ C ^ G ^ ^ A ^ A ^ ^ A ^ A — ~ ^Se ?quence * ^2 Sequence #1 \ Sequence #1 Sequence #2 Sequence #3 % of Similar Sequences 10 15 15 25 35 %of Connmunity Sequence #3 / A A A C G T A. A A G C G C A. A T C C G C A. Groupings sequences 1,5,7,16... sequences 2,4,8,15... sequences 3,9,10,11... sequences 6,12,17,18... sequences 13,14,19,20. Identifications DNA is extracted from a phytoplankton assemblage. The gene of Interest is collectively amplified from all specimens. Once amplified, the genes are transferred (cloned) into bacterial cells from which they can be individually sequenced. 10 15 15 25 35 Melosira Coscinodiscus Dinophysis Prorocentrum Ceratium Nucleotide differences between sequences (color-coded) are used to distinguish different species. The number of uniaue sequences (genotypes) indicates the number of _ species present in the onginal sample (expressed here as a percentage of all sequences examined) and their abundances relative to one another. Identifications are based on sequence comparison with electronic data bases. Microbial diversity can be estimated by analyzing the different geno- types from a mixture of microorganisms collected in a water sample. A simple, theoretical example is illustrated here. In thefriture, as elec- tronic data bases increase in size, they will allow scientists to identify organisms by their sequence "signatures." tionary time for a variety of reasons. These changes accumulate in the genome of a species (assuming that the mutation is not a fatal one) in such a way that closely related species, hu- mans and other mammals for example, have relatively few differences between their se- quences compared to distantly related species, such as humans and bacteria, which may ex- hibit many differences. Because the number of nucleotides comprising the genetic material is large, and because of the accumulation of mu- tations, there is a very high probability of find- ing regions with nucleotide sequences that are unique to each species. Molecular biologists have devised elegant systems for determining the sequence of nucle- otide bases in a wide variety (and ever increas- ing number) of genes. The result is identifica- tion of many DNA segments that have nucle- otide sequences unique to individual species. The nucleotide sequences in these "informa- tive" areas of the DNA molecule, therefore, can then be used as a molecular signature for par- ticular species, much the same way fingerprints in humans can be used to distinguish them from one another (see figure on page 11). Searching through the milUons-to-billions of nucleotides to find unique sequences that characterize a particular species is tedious and fastidious work, and has resulted in huge accu- mulations of sequence data. The job of pro- cessing and cataloging this information has been given over to powerful computers in re- cent years, and much of the DNA sequence information knowm at this time is publically available in electronic data bases. As they grow, these data bases provide a powerful and rapid means of identifying organisms, analogous to the manner in which the national data base of fingerprints is used to rapidly identify repeat criminal offenders. The basis for assessment of biological diver- sity using molecular biology is knowledge of the genetic code of a species, that is, the se- quence of nucleotides in its DNA, and how this code varies from species to species. Assessing biological diversity implies looking at a natural assemblage of organisms and identifying each of the species composing the assemblage (or at least the more common ones). From a molecu- lar point of view, this "community level" ap- proach usually is accomplished by surveying the diversity of a gene (that is, the variations in the nucleotide sequence) in a mixture of organ- isms collected in a sample. Assuming that dif- ferent nucleotide sequences of the gene (termed genotypes) indicate the presence of different species, then the number of geno- types is indicative of the number of species in FALI7WINTER 1995 the sample (see figure at left). The frequency of each of the genotype's occurrence indicates the relative contribution of each species to the whole assemblage. There are, of course, many possible genes that can be used for investigating biodiversity, and criteria exist for choosing the most appro- priate one{s). Genes chosen for these investiga- tions often relate to a specific group of organ- isms that are the focus of the study. For ex- ample, an investigator interested in assessing phytoplankton diversity might employ a gene involved in the process of photosynthesis, w^hile a bacteriologist may use a gene found only in certain groups of bacteria. However, an investigator interested in comparing a wide spectrum of organisms might choose genes whose structures are strongly conserved through evolutionary time. These aie usually genes that serve a generalized, vital function such as those responsible for ribosomal ribo- nucleic acid (rRNA). Molecules of rRNA are essential components of the framework and function of ribosomes, cellular structures that are responsible for protein synthesis in cells. All living organisms synthesize proteins, and therefore the structure of genes that produce rRNA have been highly conserved throughout evolution. Comparison of the nucleotide se- quences of rRNA genes has been widely used for establishing evolutionary relationships among a wide variety of species. One interesting result of assessing biodiversity using molecular approaches is the finding that many nucleotide sequences ob- tained directly from natural samples are not found in existing databases. This implies that these organisms do not exist in any culture collection of marine organisms, and leads bio- logical oceanographers to conclude that we have successfully cultured only a fraction of the microorganism species living in the ocean. This exciting discovery clearly points out the limita- tions of traditional approaches for analyzing microbial diversity, and highlights the potential for molecular biology to provide new insights for understanding the diversity of natural mi- crobial communities. Genetic Diversity Among Large Organisms Many applications of molecular biology in marine science involve microbial populations. The basic tenets apply equally well to larger organisms, and molecular approaches have rapidly expanded to include marine species ranging from seaweeds to whales. For large species, molecular approaches are less useful for species identification. Traditional methods are usually easier and sufficient. Questions concerning the distribution and migratory patterns of large, mobile, or free- drifting species, however, are often very diffi- cult to answer. These issues can be addressed using the genetic variability or diversity within a species (rather than between species as de- scribed above) as a tracer of the movement and interaction of these populations. Just as genetic tests in humans can determine ancestry, ge- netic studies of marine animals can determine whether distinct (or distant) populations of SSUE S>^IVIPLES Location of collection sites Tissue samples collected iraction, Pu r if i cat i oj of DN/\ /Xnalysis for identification of genetic markers Sampling Location Genetic Markers H [2] [3] Presence of three genetic markers among sampled organisms Interpretation of 3 breeding populations based on presence of markers (golden areas encircle breeding populations) Population distribution and migratory pathways of marine animals can be studied using rapidly evolving genetic markers. For example, tissue samples collected from specimens distributed over broad geo- graphic scales can be analyzed to determine whether these popula- tions are actively interbreeding. Extraction, purification, and com- parison ofDNA markers can yield information on the degree to which separate populations of a species, such as Pacific and Atlantic popu- lations, readily exchange genetic information. OCEANUS ♦IB Each marine species possesses unique genetic information that allows it to occupy its ecological niche in the ocean. animals interbreed. For example, such studies have been conducted to examine whether geo- graphically separated populations of marine mammals interbreed, to determine the migra- tory patterns of sea turtles, and to trace the source of microcrustaceans that support juve- nile fish growrth on productive fishing grounds. This type of research often employs specific DNA sequences as genetic markers to follow the flow of genes within or between populations (see figure on page 13). By determining the size of an interbreeding population, these studies provide an index of the size of the "gene pool" for that species. This feature can be important for esti- mating the resiliency of the species. Typically, populations of organisms exhibit considerable variability for any particular trait, such as eye or skin color in humans. Reducing the amount of genetic variability can have serious repercussions for a population. Individuals in a popula- tion with little genetic variability tend to be uniformly susceptible to environmental perturbations, mak- ing the entire population vulnerable to a single adverse effect. Food crops offer excellent ex- amples of this vulnerability. For many of these species, a single adverse event, such as an early frost, can destroy an entire crop that is com- posed of a single hybrid strain. The strain may have been highly inbred for a desirable quality (perhaps fruit size), but as a consequence may have little genetic variability for other traits and therefore little resistance within the population to cope with changing environmental condi- tions. Low genetic variability is one of the rea- sons why some agricultural crops have rela- tively narrow tolerances. By analogy, it can be hypothesized that dra- matic reductions in the population abundances of many large marine animals — commercially valuable finfish or marine mammals — may reduce natural genetic variability within a spe- cies. This may have serious implications for the fitness of these species. At present, however, there is litde information to determine if this is, or has been, an important consideration for the extinction of marine species. Microbial Species Identification Information obtained from the "community level" analysis of biodiversity helps researchers focus on ecological questions at the species level. The goal in many studies is to identify particular species within mixed assemblages of organisms (see Biological Oceanography from a Molecular Perspective, Oceaniis, Fall 1992). Microbial communities are most commonly studied using this approach because the organ- isms are difficult to distinguish from one an- other. Usually the species investigated are sig- nificant to environmental or human health issues, such as pathogenic or toxic species. Previously obtained information on the nucle- otide sequences of an organism can be used to identify regions of the DNA that are unique to the desired species. Short chains of nucleotides (usually about 15 to 25), called oligonucle- otides, are manufactured to be complementary to the unique sequence in the target organism. These oligonucleotides can then be used in a variety of methods to detect the presence of the target species in environmental samples. An example of the application of these tech- niques to clinical and environmental samples is the identification and detection of amoebae {Acanthamoeba) that can cause human infec- tions (see figure opposite). Gene sequences from amoebae cultured out of human infec- tions and from the environment can be em- ployed to determine genetically related groups of Acanthamoeba. Using that information, scientists can design oligonucleotides to detect the amoebae, for example, in eye infections and in biologically contaminated waters, such as sewage dump sites. From Biodiversity To Biotechnology Classical ecological principle contends that each species plays a unique role in nature. Accordingly, each marine species possesses unique genetic information that allows it to occupy its ecological niche in the ocean. These unique genetic make-ups, and the biochemis- tries and physiologies these genes manifest, allow species to cope with their environment and other species vdthin certain limits. From a practical point of view, therefore, biological diversity provides a wide array of genetically based attributes that in many cases can be employed in various human endeavors. This genetic diversity is a global inheritance avail- able to anyone who can access its untapped "resources." Molecular biology provides the needed con- ceptual approaches and practical tools. Recom- binant DNA technology provides a means of rapidly identifying genes and their specific functions in living organisms. In many cases these abilities can be transferred to other spe- cies for mass production in so-called "bioengineered" organisms, a practice that is expanding at an amazing rate. Moreover, the application of these discoveries to address needs in the biomedical and industrial sectors has aroused great interest and resulted in a FALLAA/INTER 1995 considerable number of successes. To date, however, the use of ;?7fln«f organisms in bio- technology has not been commensurate with the biological diversity present in the ocean. This inconsistency is partly a consequence of the fact that we still have little information on most marine species — the exploitation of the ocean's genetic "wealth" has barely begun. As detailed in the articles that follow, the ocean contains an enormous variety of living organisms, many still undescribed. From a bioengineering perspective, the potential for uncovering novel and useful genes within this plethora of genetic diversity is virtually certain. However, the diversity is not endless and it is not renewable. When a species becomes extinct, its unique genetic composition is lost. Eons of evolutionary change and adaptation contained in its genes vanish. While the reconstruction of an extinct species from fossilized DNA is an intriguing idea and a definite box-office hit, at present it is a scientific pipe dream. In all prob- ability, the large dinosaurs (and all other extinct species) will roam the earth again only in movie theaters and on video. Given that finality, it is imperative that we begin immediately to assess biological diversity on earth, and particularly in the ocean where the gaps in our knowledge are greatest. The acceleration of species extinctions by human activities is no longer questioned. It is now simply a matter of how fast. Human impact on biological communities in the ocean, however, cannot be fully understood until we can obtain an accurate appraisal of the number of species living in this environment that covers more than two-thirds of the earth's surface. Molecu- lar biology, in concert with traditional methods of collection and identification, will provide important new approaches for accomplishing that task, and will undoubtedly provide science with exciting new discoveries on the genetics of marine species. The authors' research on marine microorganisms is funded by National Science Foundation and the Woods Hole Oceanographic Institution Postdoctoral Fellowship Program. A graduate of the MITAA/HOI Joint Program, Dave Caron is a native of New England, where he attests to having personally sampled (on a plate at least) much of the biodiversity of the local marine environment. The marine microorganisms that he studies in his work aren't much of a culinary delight, but nonetheless they comprise an important component of marine planktonic food webs. Like so many oceanographers, Becky Gast began her career a thousand miles from the ocean in Ohio, working as a molecular biologist in a laboratory studying amoebae that cause human eye infections Understandably this work led directly to the development of her strong interest in marine ecology and her current studies of free-living (and less noxious) microorganisms Biopsy taken from infected eye. Acanthamoebas are cultured and DNA is collected. The gene of interest is amplified and sequenced. (See figure on page 1 1 .) Acanthamoeba 1 ... G Acanthamoeba 2 ...G Acanthamoeba 3 ...G Humans ... G C A C T T T C A C T T T C A C T T T A A T T A T A G G C T. A G G C T. A G G C T. A G C C C. Unique sequences are identified for designing oligonucleotide probes specific to Acanthamoeba. fSee figure on page 12.) The oligonucleotide probe (with a marker added to aid in detection) can then be used to detect or identify Acanthamoebas in CLINICAL SAMPLES or ENVIRONMENTAL SAMPLES Bottom Sample Collect cells onto a filter to concentrate them. Use oligonucleotide probe with marker to detect Acanthamoeba from the mixed population. ■/ Collect nucleic acids and amplify the gene of interest using the specific oligonucleotide. ■/ Sequence and identify the specific isolate present Human pathogens, such as Acanthamoeba, luhich is responsible for some eye infections, can be traced through the environment with molecular techniques like those described in this article. OCEANUS A natural assem- I blage of marine bacteria coloniz- ing a surface shows the diver- sity of cell shapes and growth forms. Most natural habitats support the growth of a vari- ety of different bacteria, distin- guished by their different shapes but also by dif- ferent metabolic and physiologi- cal activities in response to vari- ous physical and chemical condi- tions of the habi- tat. Interactions between the different types of bacteria, both favorable and unfavorable, also help shape the diversity of bacteria that occur together. (Magnification: 3.100.x) New Insights On Marine Bacterial Diversity MolecularTechniques Complement Culturing Paul V. Dunlap Associate Scientist, Biology Department One of the scientifically most important and challenging areas in biology today is the diversity of bacteria inhabiting the earth. Bac- teria, generally described as single-celled or- ganisms that lack a nucleus (hence the term "prokaryotic" or prenucleus) and that range in size from a few tenths to several micrometers, are recognized from fossils 3.8 billion years old as being early forms of life on the planet. Their existence as living entities has been recognized for over three hundred years, with their impor- tance as causative agents of human disease, for example, established a hundred years ago and studied intensely since that time. They are known to exist in virtually all environments throughout the biosphere. Typically, they occur in complex assemblages (above), in many cases carrying out metabohc conversions of inorganic and organic materials essential to the survival of other life forms, and in some cases reaching extraordinarily high numbers, up to 100 billion per milliliter (about one-fifth of a teaspoon of fluid). However, despite the importance of bac- teria and despite decades of active scientific study, the bacterial world today remains largely undefined, especially in the marine environ- ment. With the rare exceptions of certain well- studied disease-causing, terrestrial bacteria, most types of bacteria remain unknown and their life styles and activities obscure. Wliat accounts for this lack of knowledge? The major reason, a perplexing and long-stand- ing issue in bacteriology, is that only a tiny percentage of the bacterial types seen by mi- croscopic examination of an environmental sample, such as seawater or soil, will grow un- der laboratory conditions. Growth of an indi- FALiywiNTER 1995 vidual bacterial type, leading to its isolation in culture, has been a key step in identify- ing a bacterium and analyzing its meta- bolic capability. However, laboratory growth media present conditions representative, at best, of only a tiny fraction of the multitude of microhabitats likely to be present in an envi- ronmental sample. The inability to define and re-create these mi- crohabitats in the labo- ratory has left the vast majority of bacteria inac- cessible for study. Indeed, much of the art of bacteriology, and much of its progress, has arisen from attempts to over come this problem. An his- torically successful approach, the enrichment cultivation method, is based on creating spe- cific conditions in the laboratory that elicit die growth of individual types of bacteria sus- pected to be present and having specific meta- bolic or physiological attributes. Once isolated, these bacteria can then be studied in detail. A modern and quite different approach to overcoming this problem is the recent develop- ment and use of mo- lecular biological methods to direcdy identify individual types of bacteria in natural samples with- out culturing them. In general terms, these methods are based on the DNA sequences of genes common to all bacteria. The se- quences of these genes have varied relatively slowly over evolution- ary time through the gradual accumulation of inconsequential mutations. Detailed analysis of many types of cultured bacteria over the past several years has revealed that the sequences of these genes have both highly conserved and more variable regions, with the sequence of the variable regions often specific to, and therefore indicative of, individual types of bacteria. By designing molecular probes that recognize appropriate "signature" portions of the variable regions, scientists can create pow- erful tools for directly identifying individual types of bacteria, even from complex assem- blages in the natural environment. This meth- odology allows them to determine in some cases what bacterial types are present or to study certain aspects of their ecology while sidestepping the culturing process. For ex- ample, WHOl microbiologist John Waterbury is using this approach udth gene sequences from Syiiechococcus to identify and study the popu- lation ecology of these important unicellular marine cyanobacteria (photosynthetic bacteria that evolve oxygen as higher plants do). The combination of molecular and classical meth- ods is particularly powerful. WHOl microbiolo- gist Holger Jannasch, for example, uses a com- bined approach to study bacteria discovered at deep-sea hydrothermal vents. He compares genetic sequences of bacteria isolated through the enrichment cultivation method with se- quences from natural samples of bacteria col- lected at the same hydrothermal vent sites to Sulfiir-oxidizing, carbon-dioxide- fixing bacteria within the trophosome of the hydrother- mal vent tube worm Riftia pachyptila, drawn with outer coverings removed. The symbiotic bacte- ria receive oxy- gen, hydrogen sulfide, and carbon dioxide from the blood stream of the animal host and provide in turn organic nutri- ents for the animal's growth. (Magnification: 302x) A virus of the cyanobacterium Synechococcus. The head con- tains the genetic material, which is injected into the cyano- bacterial cell, subverting its cellular machin- ety and forcing it to produce sev- eral copies of the virus. Viruses, as parasites infect- ing and killing marine bacteria, may play key roles in the spe- cies composition of marine micro- bial populations. (Magnification: 232,000.x) OCEANUS ♦ 17 Author Paul Dunlap views colonies of a new species of luminous Photobacterium being described in his laboratory photographed by the light they produce. Each colony contains many thousand bacteria. Like many other marine lumi- nous bacteria this new species induces light production only after it attains a high population density. relate and compare the cultured bacteria to the natural bacterial assemblages. These molecular approaches, by also increasing the number of bacteria whose key gene sequences are known, help to establish a genetic-sequence-based framework for bacterial diversity at the phylo- genetic (evolutionary relationship) level. The foundation and driving force for the development of bacterial diversity is the inter- action of the bacterial cell with the environ- ment. Bacteria are generally found in an envi- ronment because they are able to survive and acquire the nutrients necessary for growth under the specific physical, chemical, and bio- logical conditions occurring there. Survival and growth are dependent on the integrated activi- ties of several cellular processes, and the differ- ences in these activities and in their integration are at the scientific heart of bacterial diversity. These processes include: • specific metabolic capabilities, such as the ability of certain bacteria to gain energy and carbon for growth from reduced sulfur com- pounds and carbon dioxide, • the individual biochemical pathways in- volved in a metabolic capability, and • the regulated expression of genes that direct the synthesis of enzymes whose reactions form the biochemical pathways. Integration of these processes occurs at the physiological level, through sensing and adapta- tional responses to specific environmental con- ditions. Because these conditions change with time, physiological integration of the processes of growth and survival provides the bacterium with a sensitive and highly coordinated means of responding appropriately to new conditions. Ultimately, the biological foundation for diver- sity is the genome, with its sequences of genes encoding the proteins that carry out the differ- ent metabolic activities and physiological adap- tations necessary for the bacterium to interact successfully with its environment. Studies over the past 100 years have shown that bacteria as a group, in contrast to plants and animals for example, exhibit a tremendous diversity of metabolic capabilities and physi- ological attributes. What accounts for this metabolic and physiological diversity? Cer- tainly, one factor is the diversity of environ- ments created by chemical and physical changes in the planet over the billions of years since its origin. In addition, at a bacterial level, environments can be extremely complex, with conditions differing dramatically within a few micrometers. This environmental diversity brought opportunities for bacteria, but exploi- tation of these opportunities required inven- tion of the means to utilize the newly available chemical components as nutrients for energy and growth, while also finding ways to survive the physical conditions present, such as in- tense ultraviolet radiation or extremely high temperature. These opportunities and needs, through selection for genetic mutations leading to suitable new traits, undoubtedly drove de- velopment of the tremendous metabolic and physiological diversity we now find in bacteria. Similarly, there are also important biological factors, including competition for nutrients and other resources and the activities of preda- tors and parasites. For example, viruses specific to individual types of bacteria may play key roles in the species composition of marine microbial populations. Along with the use of DNA sequence-based probes to identify' bacteria in nature, molecular biological approaches are rapidly improving the ability to explore and define the metabolic and physiological diversity of bacteria. Impor- tantly, scientists now apply molecular ap- proaches to marine bacteria, a group that tradi- FALL7WINTER 1995 tionally has not been studied extensively at the molecular level. Among the various approaches they use are the cloning and characterization of genes for metabolic or physiological traits spe- cial to marine bacteria, the mutation of those genes, and the transferring of the mutation back into the bacterial genome to construct defined genetic mutants through recombina- tion. These mutants are invaluable tools for rigorously studying a trait of interest. Another approach involves delivering transposons (special DNA sequences that can insert randomly at new locations in the chro- mosome) as mutagenic tools to help identify genes involved in specific traits. A full array of supporting biochemical, genetic, and molecu lar technology is also now being applied to marine bacteria. Our work on the bacte- rium Vibrio flscheri gives some examples of how effec- tive molecular biological approaches can be in exploring metabolic and physiological diversity in marine bacteria. This bacterium is globally distributed in coastal waters and exliibits many traits representative of commonly encountered marine bacteria. It is unusual, however, in be- ing one of approximately a dozen bacteria with the metabolic capability to produce light, and it is one of a select group of three identified light- producing species able physiologically to form a bioluminescent mutualism, or light-organ symbiosis, with certain marine fish and squids. We have focused over the past several years on these unusual traits, using molecular ap- proaches to gain insight into the mechanism that controls light production and the genetic basis for bacterial adaptations to symbiosis. That work has led to discoveries that in turn have opened up new lines of research. One is the discovery of a novel signaling and gene regulatory system in V.fischeri that senses the population density of the bacterium and that helps control light production. Presently, we are interested in learning whether other meta- bolic activities in the bacterium are also con- trolled by the new signaling system. Another is the discovery of a novel metabolic trait in V. fisclieri and a few other marine bacteria, the ability to use an intracellular signal molecule, known as 3':5'-cyclic AMP, as a source of phos- phorus, nitrogen, and carbon for growth. Deg- radation of extracellular 3':5'-cyclic AMP may be a new bacterial niche, one important in the ecology of marine bacteria. A third discovery, achieved in work conducted in collaboration with Edward Ruby and Jorg Graf at the Univer- sity of Southern California, is identification of the bacterium's ability to swim as a trait necessary for V.fischeri to establish its mutualistic bioluminescent symbiosis with the Hawaiian coastal squid, Euprymna scolopes. Molecular ap- proaches are making the cellu- lar and genetic basis for bacterial mutualisms with marine animals ac- cessible to detailed study for the first time. That work, in turn, may help better under- standing of human disease by pinpointing basic principles underlying the interactions of bacteria with animals. Molecular biological approaches provide an array of powerful tools for exploring bacterial diversity. Given the relatively little attention focused on marine bacteria at the molecular level in the past, the increasing use of molecular approaches is likely to result in a multitude of unexpected discoveries over the next several years. These insights into marine bacterial diver- sity will also form a foundation for many new biotechnological products and applications. The next millennium will be an exciting time for bacteriologists as molecular biology helps open new windows on the bacterial world. Work in the author's laboratory on the molecular biology of V. fischeri is supported by the National Science Foundation. Recent publications have appeared in the Journal of Bacteriology, r/ie Journal of Biological Chem- istry, fl/jrf Archives of Microbiology. Paul Dunlap is an Associate Scientist in the Biology Department at WHO!. He has been interested in the biology of marine bacteria from the time of his undergraduate days at Oregon State University and his interests were captured by luminous bacteria and symbiosis during graduate studies at the University of California, Los Angeles. The coastal Ha- waiian squid Euprymna scolopes. The nocturnally active animal harbors a cul- ture of Vihrio fischeri as its bioluminescent symbiont in a pairof gland- like light organs within the mantle cavity. It uses the bacte- rial light for counter- illuminatioti, that is, it casts ligiit downward to match light fiom the surface as a means of hidingfrom predators below. The adult ani- mal shown here is approximately 5 cetttimeters long. OCEANUS Cyclothone, the world's most abundant fish, is 2 to 3 centime- ters long, with only a dozen species in all oceans. From Vie Fishes. Memoirs of the Mu- seum of Comparative Zoology at Harvard College, Vol. XXIV Cambridge, USA, 1899. Diversity in a Vast and Stable Habitat Midwater Is One of Earth's Least Explored Environments Laurence P. Madin Associate Scientist. Biology Department Katherine A. C. Madin Guest Investigator, Biology Department The waters of the world ocean, from the surface to the deepest reaches of the bottom trenches, have a total volume of about 1.4 billion cubic kilometers, more than 99 percent of Earth's biosphere. Only about 150 years ago, biologists were convinced that the deep ocean was an "azoic zone" where crushing pressure made life impossible. One of the most important discov- eries of early oceanographic expeditions was the great variety of deep-ocean life, well adapted to pressure and cold. Since then, sam- pling with trawl nets, and more recently with acoustics and submersibles, has increased our understand- j*. ingof the midwater fauna, but it remains one of the least explored environments on earth. Only in re- cent times have we even been able to wonder about this fauna. Wliat do we know about the diversity of life in deep water? This vast habitat is home to far fewer species than any terrestrial or nearshore marine environments. We don't actually know how many midwater species there are in the world, but if we exclude protozoans (unicellular organisms), there may be only something like 5,000, many of which occur throughout large parts of the ocean. For example, out of about 12,000 species of marine fishes, only about 1,000 are considered deep pelagic species worldwide. The great abundance of some of them belies the actual number of species present. The most abundant fishes in the world, the tiny bristlemouths {Cyclothone], live in midwater, yet there are only about a dozen species of them worldwide. Similarly, the most numerous crustaceans in the world are the marine copepods, but there are only a few thousand species, a total vanishingly small compared to the millions of species of insects that occupy analogous niches on land. <>^ This seems to be a very low diversity, but is it? Is it a lot lower than we expect for a habitat so large, or is it perhaps rather high for an envi- ronment that appears so homogeneous, with little variety in living spaces? Do we know what the biodiversity really is in these poorly ex- plored places, or wrill our perceptions change with new discoveries? And finally, is it impor- tant for us to understand the diversity of the midwaters, and the forces that regulate it? To begin answering these questions, we need to understand what the water column is like. The midwater can be divided into two major regions, based on depth, with very differ- ent physi- cal and biological attributes. The mesopelagic zone is a twilight region where sunlight is too weak to support photo- synthesis, but penetrates sufficiently to affect the behavior of animals on a diurnal schedule, that is, the animals living there can see and react to the changing light levels. In most places the mesopelagic zone extends from about 200 to 1,000 meters, and includes the thermocline, where temperature drops some 20°C from the warmth of the surface to the cold of the depths. The proximity to upper- water primary production makes it possible for many herbivores to live here, and perhaps half the species are vertical migrators, who spend the day at depth but come to the rich surface waters at night to feed on phytoplankton or other animals. Below about 1,000 meters lies the bathype- lagic zone, accounting for 88 percent of the total ocean area and covering the entire deep ocean floor. Here there is no light from the sun, the temperature is constant and cold (4°C), and food is scarce. The full time residents of these lower depths generally do not migrate out of their zone, but tend to be rather sluggish 20 • FALL7WINTER 1995 North Atlantic Western North fc Equatomr astern South Central . uatorial Indian Central South Pacific Central detritivores, predators, com mensals (two closely associated organisms), and parasites that have evolved to survive wath minimal energy expenditure in a food-poor environment. How much biodiversity could we expect in such environments? It is well known that high diversity is associated with heterogeneous, complex environments like rain forests or coral reefs, where the physical and biological com- plexity allows great specialization of plants and animals, yet tends to limit dispersal and isolate populations. Diver- sity can also be high in habitats that change little over the course of time, such as the deep seaf- loor. The midwater, though, appears to be about the least hetero- geneous environment on earth. There are few physical barriers to movement of animals from one ocean basin to another, and the major ""-' " " deep ocean currents actually cause the waters below 1 ,500 meters to be completely ex- changed every few hundred years. But it is also one of the most stable; compared to land, physical characteristics and food supply in the midwater environment are fairly constant and uniform. Seasonal changes in food production at the surface are damped down to a sparse, monotonous rain of sinking detritus, punctu- ated by occasional dead whales. So why is there any biodiversity at all in midwater? It seems plausible that over millions of years, populations of animals would be dis- tributed throughout the midwater depths by currents, and competition for limited food re- sources would have reduced the species to the fittest few. Why should there be as many as 1,000 spe- cies of midwater fishes, probably thousands of species of deep sea pelagic copepods, 200 kinds of mesopelagic medusae and ctenophores, or even 20 species of deep-water squids? What kinds of physical or biological structures create the opportunity for diverse species to evolve and maintain their genetic identity? We can look for this structure at several levels. On a global scale, the distribution of animals in the water col- umn roughly follows the major water masses of the oceans. These masses differ in tem- perature, salinity, and oxygen content, differ- ences that make those waters better suited for some species than others. Water masses also have physical boundaries, which are sometimes the edges of continents, but more often currents or sharp temperature discontinuities that form "walls" in the water. These walls can delineate large regions of the ocean, even at midwater depths, which have stable physicochemical conditions and characteristic faunas, or groups of species. Large faunal "provinces" have been identi- fied in the North Pacific, based on the distribu- tion patterns of many species of zooplankton and fish, together with knowledge of water circulation patterns. During the 1970s, WHOI scientists Richard Backus and James Craddock identified thousands of mesopelagic lantern fishes from the Atlantic Ocean, and found that particular groups of species always occurred Major water masses of the world's oceans have character- istic physical properties that are uniform over large areas, and bounded by currents or land. Large species gioups are asso- ciated with the water masses, although the picture is blurred by nu- merous cosmo- politan animals that cross boundaries. Within the At- lantic (lower map), a finer grid ofzoogeo- graphic prov- inces has been identified, based on midwater fish distribu- tions. Seventeen such provinces are numbered within larger faunal regions (solid lines) rangingfiom Subarctic to Tropical. Samples pom a multiple-net trawl fished from 0 to 1,000 meters in the GulfofAden during August 1995. The trawl is designed to catch large zoop- lankton and midwater fishes, with net mouth openings of 10 square meters, and a mesh size of 3 millimeters. Net 1 (top photo) was open throughout the water column and contains a mi.xture of all the species present. The trays below are from nets fished through specific depth ranges, and show the vertical segrega- tion of different species. Net 5, fished between 0 and 150 meters, contains mainly ctenophores, medusae, and pteropods (small black shells of planktonic snails). Net 4 fished the low- oxygen waters between 150 and 300 meters where almost nothing lives. Net 3 was open between 300 and 500 Dieters, a range occupied mainly by red sergestid shrimp. Net 2 fished 500 to 1,000 meters, catching large purple medusae and a variety of midwater fishes. Net #5 0-150 meters n * 9i Av / ( ^^^r^. > • ^ i * > • f I " « -. •>'» /.;. i Net #4 150-300 ni Net #2 500-1,000 meters within regions bounded by physical discon- tinuities. The walls are leaky, to be sure, and there are species that overlap more than one region, but the distribution patterns hold for several kinds of animals in both the Pacific and Adantic Oceans (see figure on page 21). And just as on land, species diversity seems to in- crease from the poles towards the tropics. Since conditions within faunal regions are fairly uniform, it is curious that we find a large number of species sharing the same water. The North Pacific gyre (the central tropical and subtropical region of the North Pacific, bounded by ocean currents travelling in clock- wise rotation), for example, is home to a persis- tent assemblage of over 200 copepod and 228 midwater fish species. Ordinarily, we expect only one species to occupy an ecological niche. How have all these species found unique slots — are there more niches than are obvious in the midwater? What we know about the animals and envi- ronment so far indicates that there are three ways of dividing the habitat, three axes, along which different species have particular slots. The first axis, spatial separation, is mainly verti- cal. In almost all parts of the ocean, there are clear faunal differences between the meso- and bathypelagic depths, and the animals of those zones have very different lifestyles. There are also differences v«thin these ranges, preferred depths that species may occupy all the time, or move within as they migrate. Movement, mainly as vertical migration, allows a sort of time-sharing division for the water column; individual patterns of movement give different species a way to avoid each other and partition their environment along a second axis of time (see figure at left). The third important axis for differentiation is the physiology and behavior of feeding and reproduction. Different species are adapted for catching certain kinds of prey; these spe- cializations sometimes occur on the level of species, and sometimes for whole families of fishes or crustaceans. Reproductive adapta- tions to insure the timely matching of mates can be even more specialized, sometimes using patterns of luminous organs or chemical pheromones. When the different species living together in a midwater habitat are mapped out on these axes, they sometimes fall into clearly separate niches. Studies of the vertical distribution and diet of midwater crustaceans and fishes in the Gulf of Mexico, for example, revealed that animals with similar diets lived at different depths and vice versa, so that these species divided their monotonous envi- ronment and avoided competition by not FALITWINTER 1995 eating the same things in the same place at the same time. However, this explanation may not always work. The 200 species of copepods in the Pa- cific gyre appear to have similar diets and over- lapping vertical distributions. We don't really know how such seemingly similar species can coexist without the type of competition that would drive out the majority of "less fit" com- petitors. Recent investigations of small scale mixing events in the water column suggest that localized turnover of the waters can produce temporary patches of higher productivity that may change the abundance and diversity of other zooplankton. If these effects filter down to the midwater depths, this transient patch- work of conditions could add heterogeneity to the environment and continually interrupt the interactions by which species compete v^ath each other. Another key to understanding these high diversities may be the proportions of different species. In midwater samples we usually find that a few species are very abundant, several are common and most are rare, with only a few individuals each. In fact, it is estimated that over 80 percent of all oceanic species are rare. Rarity is sometimes a function of sampling methods — animals not easily collected will appear rare until a better sampling method shows that they were there all along. But for animals that are well sampled, like copepods, the rarity is real, and may reduce competition simply because the species are too few and dispersed to interact very often. Whether midwater biodiversity is consid- ered high or low depends on the scale at which it is reckoned. Globally, it is obvious that the number of species of midwater animals is quite low. Yet on the local scale of a single water mass or faunal region in the midwater, the species richness may be high; hundreds of kinds of fishes, crustaceans, and gelatinous animals can be caught in a single trawl. The difference is that global biodiversity is the sum of all the species in all the habitats, and habitats in the water column are few, huge, and not strongly endemic. When species lists are added up for all midwater communities, the world totals are only a few times more than the count for any single community. In contrast, there are mil- lions of distinct and fixed habitats on land, some as small as a single rain forest tree, with highly endemic species that contribute to glo- bal terrestrial diversity. Assessment of biodiversity depends on knowing what species are present, and in the deep ocean this information is strongly fil- tered through the nets and trawls used for the past 100 years. Recent investigations in new places and with new methods have uncovered additional sources of diversity that may cause us to reexamine oceanic biodiversity. Here are three examples. • Intensive trawl sampling by oceanographers at the University of Hawaii revealed unex- pected diversity in another kind of midwater region that is associated with the edges of land. Biologists have identified a "mesopelagic- boundary" community comprising species of fish and crustaceans that are distinct from the much larger "oceanic-mesopelagic" commu- nity of the central North Pacific. Deep bound- ary communities may exist near other coasts. This crustacean is an isopod fAnuropus sp.) which lives only inside the um- brella of the large deep-sea medusae Deepstaria enigmatica. Other gelatinous animals in midwater har- bor amphipods or fish as specific symbionts. Ctenophores are beautiful but fragile gelati- nous animals that prey upon a variety of other zooplankton. Exploring with manned submersibles, WHOI biologists have discovered dozens of new species, like this as-yet-unnamed one, at depths below 500 meters. Virtually no exploration of the midwater by human ob- servers has yet gone below 1 ,000 meters, where many more new kinds of these fragile creatures may be waiting. and contribute a significant new source of di- versity to our knowledge of the global total. • In many terrestrial habitats, high diversity results partly from numerous symbiotic rela- tionships between highly specialized species. It turns out that several groups of midwater ani- mals have evolved a similar habit. In situ diving and submersible observations by the first au- thor and WHOI colleague Richard Harbison have shown that a significant number of crus- tacean species and some fishes, worms, and molluscs specialize in living as parasites, com- mensals, or micropredators around or on the surfaces of gelatinous animals, using them as floating substrates, shelters, food sources, and nurseries for the young (see photo on page 23). These relationships usually aren't detected in net samples, because the partners become separated. Often highly specific, these symbio- ses create another level of structure in the wa- ter column, like the nooks and crannies in a reef that provide niches for more species. • Over the last 20 years, the first author and colleagues at WHOI, Harbor Branch Oceano- graphic Institution, and the Monterey Bay Aquarium Research Institute have used re- search submersibles to explore the mesope- lagic zone, and collect a variety of fragile gelati- nous animals that had rarely, if ever, been seen before. This new capability has revealed a re- markable diversity of medusae, ctenophores, and siphonophores restricted to the midwater environment. During a decade of submersible dives, over 40 new species of ctenophores have been discovered, nearly doubling the number of species known in this quintessentially pe- lagic phylimi. We still have a lot to learn about midwater species and their communities. This part of the ocean is seldom visited, and only at great cost. Our samples of its fauna over the last hundred years have been like flash photographs taken in the forest at night. We see the more common, less wary animals, but often not the fast, small, or fragile ones, and we can infer relatively little about the behavior or life history of those we do see. So, for now, our conception of midwater biodiversity has to rest on tough animals caught in nets — mainly fishes and crustaceans. We probably do know the diversity of these groups pretty well, but our knowledge of the quicker, tinier, or flimsier animals will improve only when our coflecting and observation tech- niques improve. We will also learn more about midwater diversity as more sampling is done in oceans other than the North Atlantic and North Pacific, and when data from different parts of the world can be brought together and com- pared in a global taxonomic database. However remote, unfamiliar, and uncompli- cated the oceanic environment may seem com- pared to the land, we should not forget that it is the biggest biome on earth, and its biota plays a significant role in global biological and chemi- cal cycles. Our present, incomplete, knowledge of diversity and structure suggests that midwater communities are stable, but we have little basis for predicting what effect changes in global climate, carbon balance, or sea level might have on their structure and function. If these changes might be hastened by human activities, we would be wise to improve our understanding of oceanic biodiversity enough to be able to detect changes when they begin, and not after they are irreversible. The first author's research on the ecology and behav- ior of deep-sea plankton has been flinded by the National Science Foundation and theNOAA National Undersea Research Program. Larry Madin was at first attracted to gelatinous plank- tonic animals by their crystalline beauty He decided to go into the business when he discovered how poorly known they were, and consequently how easy it was to become an expert! When funding permits, he pursues studies on the behavior and ecology of "jelly animals" and other plankton in open-ocean and deep-sea environments. Katherine Madin has always found marine invertebrates more interesting than the soil nematodes she studied for her Ph.D. Currently a WHOI Guest Investigator, she divides her time (unequally) between teenage sons and marine organisms. She has taught comparative physiology and marine biology, and co-authored a forthcoming children's book on planktonic animals. 24 ♦ FALLAA/INTER 1995 The Deep Sea: Desert AND Rainforest Debunking the Desert Analogy Paul V. R. Snelgrove Postdoctoral Fellow, Institute of Marine and Coastal Sciences, Rutgers University* J. Frederick Grassle Director, Institute of Marine and Coastal Sciences, Rutgers University** The title of this article may surprise some and offend others, but we chose it to highlight a common misconception. In sources ranging from the popular press to university textbooks, the deep sea is often likened to a desert with large expanses of monotonous landscape de- void of life. Most panoramic photographs of the deep sea bottom are indeed reminiscent of deserts, with gently rolling contours of mud or sand and little visible life. Until the 1960s, most impressions of the deep sea were based on photographic observa- tions and ineffective sampling techniques, and both supported the view of life in the deep oceans as species poor. Thus, there arose the analogy of the "ocean desert," a per- spective that persists even today among most people who do not actually study deep-sea biology. In the 1960s, WHOI bi- ologists Howard Sand- ers and Robert Hessler (Hessler is now at the Scripps Institution of Oceanography) began to use a sampling device called an epibenthic sled (see photo on page 28). This device was dragged across the bottom to provide more quantitative and complete samples of bottom- living organisms (benthos). They sampled a number of deep-sea sites between Martha's Vineyard and Bermuda, and provided the first evidence that deep-sea communities are actu- ally extremely varied. A tremendous diversity of tiny invertebrates (macrofaunal benthos) lives within the bottom sediment. This community includes polychaetes, crustaceans, and mollusks that had been missed in photographs and by the relatively primitive sampling equipment used up until that time. The magnitude of this diversity was not fully appreci- ated until extensive sampling of the Atlantic continental slope of the United States was un- dertaken in the 1980s by author Grassle's lab at WHOI and Nancy Maciolek's and Jim Blake's lab at Battelle Ocean Sciences. These samples were collected using a device called a box corer, (see photo on page 27), which collects quantitative samples of benthic organisms, including fauna that were not effectively sampled using previous gear. This sampling revealed that the deep sea may, in fact, rival tropical rainforests in terms of numbers of species present. Thus, the deep sea may physically resemble a desert, but in terms of species composition it is more like a tropical rainforest! The continental slope and rise from New England to South Carolina is the most extensively sampled region of the deep sea. On eight cruises during the period from 1983 to 1985 we collected 556 box-core samples at depths ranging from 600 to 3,500 meters. Each encompassed a 30-by-30-centimeter-square section of ocean bottom and included the sedi- ment to 10 centimeters depth. At a single sam- pling site off Charleston, South Carolina, at about 800 meters depth, 436 species were taken from an area of less than one square meter of seafloor (nine samples pooled). A total of 1,597 species were identified in the 556 box cores combined. These sampling squares together total a little over 7 by 7 meters, an area about the size of a large living room, but nevertheless represent (by far) the most extensive sample collection from an area of the deep sea! The figure overleaf summarizes the diversity of taxa present in a subset of these samples (collected from between 1,500 and 2,500 Photos like this one, showing only two species, a sea urchin and brittle stars, supported a species-poor view of the ocean floor. However, begin- ning in the 1960s, improved sampling tech- niques began to show the spe- cies-rich nature of smaller ani- mals on and in the sediments beneath these animals. •Also Guest Investiga- tor, Department of Applied Ocean Physics & Engineer- ing, WHOI "Also Adjunct Scien- tist. Department of Biology, WHOI OCEANUS • 25 These drawings represent the diversity of Ufe found in seafloor samples col- lected between 1,500 and 2,500 meters at 14 stations along a 180-kilometer- long contour of continental slope off the eastern United States. The au- thors note that the numbers of species in this sampling area alone invite comparison with rainforests. meters in a 180-kilometer section of continen- tal slope). Very few individuals are qualified to undertake identification of species in even one of the groups of deep-sea animals, and we were fortunate to have a high proportion of the world's deep-sea taxonomic experts working on this project. It is very difficult to find support for deep-sea systematists, and this is one rea- son for a critical shortage of trained taxono- mists. The numbers of species in this sampling area alone invite comparison writh rainforests. Initial estimates by Terry Erwin (Smithsonian Institution) of tens of millions of species of insects and spiders in rainforests were based on his finding 1,080 species of beetles from 50- meter transects in four different types of forest within a 70-kilometer radius of Manaus, Brazil. We must point out, however, that most of the species found in both deep-sea and rainforest samples are very rare, while in most ecosys- tems a given sample will often yield a number of individuals of each species. It is very difficult to estimate total numbers of species for both the deep sea and the rainforest (see Estimating Biodiversity on page 9). Because the composition of macrofaunal species has been looked at in very few other quantitative or qualitative samples from the deep-sea floor, and the deep sea is so vast, it is not easy to contemplate what we may eventu- ally find. Recent quantitative studies off south- eastern Australia and off California, although FAL17WINTER1995 less extensive, indicate that similarly diverse communities, made up of almost completely different species from those found off the US East Coast, occur in other regions of the deep sea. In addition, John Lambshead (Museum of Natural History, UK) suggests that the number of deep-sea species of smaller multi-celled animals (the meiofauna) is even greater than the numbers of macrofauna. Not all deep-sea biological communities are species-rich. Hydrothermal vents are often described, again in sources ranging from the popular literature to university texts, as "oases in the ocean desert." When they were first dis- covered in 1977, hydrothermal vents generated great excitement because they indeed appeared as "oases" with high densities of extraordinarily large individuals. Organisms included new families and genera of organisms such as tube worms over 2 meters long and clams with shell lengths in excess of 25 centimeters. Nothing like it had ever been observed in the deep oceans, and the hydrothermal vent ecosystems remain among the most spectacular and fasci- nating on earth. Although the life forms discov- ered there are extremely unusual, the high temperatures and hydrogen-sulfide content of active-vent water combine to create a habitat that is inhospitable to most organisms. Thus, species diversity at hydrothermal vents is quite low, with only a handful of extraordinarily adapted species able to thrive under the harsh conditions. Hydrother- mal vent ecosystems are an "oasis" in terms of biomass and density of organisms, but in terms of species diver- sity they are extremely poor relative to other ecosystems. Given the physically harsh nature of hydro- thermal vents, it is hardly surprising that relatively few species are able to live there. The same is true for other deep-ocean habitats that present physically harsh or extreme environ- ments. Deep-sea trenches, which are subject to frequent catastrophic sediment slumping, have rela- tively low diversities. David Thistle and co- The box corer is a principal tool for deep-sea faiinal studies. As it nears the seafloor the winch is slowed to ensure gentle Impact and limited sediment disturbance. When the tripod rests on the bottom, a release mechanism allows the box within the framework to slide slowly Into the sediment. Another release then swings the flat spadelike bottom of the box down to enclose a cube of sediment for its trip to the surface. Author Grassle, at left, signals the fl/l^Oceanus winch operator to lift the corer to begin Its descent. workers (Florida State University) sampled an area of the bottom south- east of Nova Scotia subject to intensive underwater storms and strong bottom currents and found generally low diver- sity (though meiofaunal organ- isms were some- what diverse). So what is it about some deep- sea communities that makes them so diverse? In the past it was thought that deep-sea environ- ments were ex- tremely homoge- neous and stable. But more recent observations show that deep-sea habitats have significant hetero- geneity with respect to both space and time. In 1982, a series of time-lapse photographs col- lected by David Billett and co-workers (Insti- tute of Oceanographic Sciences, UK) in the northeast Atlantic revealed strong seasonal pulses of phytoplank- ton detritus sinking from surface waters to the sediment at depths of up to 4,100 meters. In addition to the sea- sonality of detrital input, these research- ers also observed spa- tial variability in patch distribution. As sam- pling frequency and spatial coverage in the deep sea increases, so does our perception of the heterogeneity of deep-sea ecosystems. For example, samples collected from the equatorial Pacific by Craig Smith (Univer- sity of Hawaii) have revealed a similar heterogeneity in food supply to the benthos. This heterogeneity is now thought to be a One of the capti- vating hydro- thermal vent photos taken over the last 18 years shows tube worms, clams, mussels, crabs, and some smaller vent dwellers. Though the vent areas are densely populated, the Inhospitable high tempera- tures and hydro- gen-sulfide con- tent of active- vent water keep the species diver- sity low. OCEANUS Bob Hessler, right, and George Hampson, of Howard Sanders's lab, mug for the camera near epibenthic sleds used for pioneer- ing work on deep-sea diver- sity begun in the ' 1960s. Towed across the bot- tom in the posi- tion they appear here, the sleds scraped the top few centimeters of sediment into a fine-mesh container critical factor in sustaining the diversity of deep-sea life. There is little doubt that food is in short supply in deep-sea communities, but it now appears that what is available may be extremely patchy in space and time. One theory holds that the deep sea may be species rich because of small-scale patches created by events such as phytoplankton blooms; the sinking offish carcasses, pieces of wood, or seaweed; small-scale physical distur- bances created by fish feeding; and polychaete fecal mounds. These patches create microhabi- tats that certain species may be able to utilize better than other spe- cies. The shifting mo- saic of small-scale patches that occurs over the deep-sea floor may allow coexistence of all sorts of different species that would otherwise be compet- ing for extremely lim- ited food resources. In shallow water, physical events such as storms and tides tend to oblit- erate patches quickly so they cannot offer the same habitat het- erogeneity as in deep- sea ecosystems. In 1989, we con- ducted a series of ex- periments that were designed to determine whether different types of potential food patches would attract different organisms. A number of past studies show that pulses of organic matter attract a specialized fauna, but we reasoned that if small-scale patches were to serve as a mecha- nism for enhancing diversity, then different patch types would attract different species of organisms. Working south of St. Croix at 900 meters depth, we created artificial sediment patches that contained no organic material or that contained one of two different types of algae. We found that a type of seaweed [Sargas- sum sp.) attracted relatively low densities of a moderately diverse fauna over the 23 days of the experiment, whereas a type of single-celled phytoplankton (Thalassiosii-a sp.) attracted extremely high densities of only a few species. The patches containing no organic matter at- tracted a fauna that differed from both algal treatments. The fauna in all of the artificial patches was quite different from the natural fauna in nearby undisturbed areas. Additional experiments conducted in 1991 demonstrated that patches attract different faunas as organic material in the patches ages. These experi- ments support the hypothesis that small-scale patches create microhabitats on which differ- ent species may specialize. Thus, it is the het- erogeneity of a habitat once thought to be ho- mogeneous that appears to be the key to its remarkable diversity. We anticipate that in- creased sampling of natural ephemeral patches will support the notion of a patch mosaic in the deep sea. But only time will tell — the deep sea continues to provide more and more sur- prises as we are able to look more closely! Aspects of the research described have been sup- ported by the National Undersea Research Pro- gram of the National Oce- anic and Atmospheric Administration, the Na- tional Science Foundation, and the Minerals Manage- ment Service of the Depart- ment of the Interior More detailed information on this work may be found in the February 1992 issue of American Naturalist in an article entitled "Deep-sea species ricliness: regional and local diversity esti- mates from quantitative bottom samples" by J. F. Grassle and N. J. Maciolek, and in the November 1992 issue o/Limnology and Oceanography in an article entitled "The role of food patches in maintaining high deep-sea diversity: Field experiments with hydrody- immically unbiased colonization trays." byP.V.R. Snelgrove, J. F. Grassle, and R. F. Petrecca. Paul Snelgrove is a displaced Newfoundlander who completed a Ph.D. in the MITA/VHOI Joint Program in Bio- logical Oceanography in 1993 with advisors Cheryl Ann Butman and Fred Grassle. In 1993 he followed Fred to Rutgers University to become a postdoctoral fellow in Fred's lab, where he has been ever since. He continues his affilia- tion with WHOI as a guest investigator in Cheryl Ann's lab. (A note to the editor said that he was getting married in Woods Hole as soon as he sent off this manuscript, so any typos should be blamed on this coincidence of datesl) Fred Grassle is a displaced Clevelander who was a member of the WHOI Biology Department for 20 years. He moved to Rutgers University in 1989 to become the first Director of the Institute of Marine and Coastal Sciences. He retains his affiliation with WHOI as an adjunct scientist. (Fred also traveled to Woods Hole to attend Paul's marriage to Joint Program graduate Michele DuRand when this manuscript was completed.) FALiyWINTER 1995 What of the Deep Sea's Future Diversity? Given that we now know that the deep sea is a tre- mendously diverse habitat, how should we treat this environment in the future? In many respects, deep- sea ecosystems are among the most pristine and least- threatened habitats on earth. Despite the fact that deep-sea bottom covers some 300,000,000 cubic kilometers of the earth's surface, most of it is far enough from land that coastal pollution has had minimal impact. Still, because deep-sea habitats are so vast and "out of sight," lawmakers and some scientists have sug- gested that deep-sea envi- ronments might be an ideal repository for various types of waste. Indeed, several deep-ocean sites on both coasts of the United States have been subject to various forms of ocean dumping over the years, ranging in composi- tion from radioactive waste to sewage sludge. One such site is 106-Mile Deepwater Sludge Dump Site (DWD-106), an area off the New Jersey coast where sludge was dumped from 1986 to 1993. Very little is known about the vulnerability of deep-sea ecosystems to human disturbance, but our re- search at DWD-106 sug- gests a cautious approach to deep-ocean dumping. At DWD-106, currents spread sludge dumped in surface waters over a large area before the sludge particles reached the deep-sea floor, but a mea- surable impact was still observed. Cindy Van Dover (now at the University of Alaska) demonstrated that ani- mals accumulated sludge- (Above) Up to tlieir ankles in mud from Deep Water Dumpsite 106 (DWD 106), Fred Grassle and co-workers examine benthic animals from a bottom trawl collected on a 7989 fl/V Atlantis II voyage. (Below) A free descent "elevator" fidl of coring devices is launched from M/V Betty Chouest during 1991 DWD 106 work with the remotely operated vehicle Jason. / k M [ I !*¥ iWBBJniw^l li i M ^^m ^m '^ '^^K^^^^^* -*'"^^i i 1 1 1^ L-.1' derived material in their tissues in an area of highest sludge concentration. Our own data, which are still being analyzed, indicate that several species normally not present or present in very low numbers increased in areas of high sludge deposition. It is likely that if sludge dumping continued over a long period of time the diversity of species would decrease as a few species became more and more abundant and rarer species disappeared. At present, we are trying to determine how (and if) benthic communities at DWD-106 v^rill recover now that sludge dumping has terminated. There are definite threats to the tremendous biodiversity of deep-sea ecosystems. In addition to serving as a potential re- pository for various types of waste, some areas of the ocean may someday be mined for valuable miner- als such as manganese. Despite their remote loca- tion, areas of the deep sea are already being subjected to increases in anthropo- genic chemicals and nutri- ents as a result of atmo- spheric and river input. As human population densi- ties continue to increase along our coastlines, an- thropogenic impacts on the highly diverse, deep- sea communities are likely to increase. Previous stud- ies of the deep-sea bottom of the western Adantic slope and rise, including DWD-106, help to provide baselines for continuing studies of these changes through repeated visits to areas that are now rela- tively well known. — Paul Snelgrove and Fred Grassle OCEANUS • 29 Manatees belong to the smallest marine mam- mal order, the Sirenia, and they are related to the elephants. Sire- nians share a heairy-boned frame, a hori- zontally flat- tened tail, a tough wrinkled hide, and bristly hairs around the lips. In addition to teeth, they have special plates for chew- ing estuarine and coastal vegetation. •NMFS. NOAA: Na- tional Marine Fislieries Service, National Oceanic and Atmo- spheric Administration Marine Mammal Biodiversity Three Diverse Orders Encompass I 19 Species Robert L. Brownell, Jr. Research Zoologist. NMFS. NOAA* Katherine Ralls Research Zoologist, National Zoological Park William F. Perrin Senior Scientist. NMFS, NOAA* Marine mammals — species that spend much of their time in the water — are descended from terrestrial ancestors and have evolved many interesting adaptations to their aquatic habitat. Thick layers of blubber, in the pinnipeds and cetaceans, or fur, in the sea otter, protect them from cold. Their bodies have become streamlined for efficient swim- ming and their forelimbs modified into fins or flippers, except for the otters. The cetaceans are the most highly spe- cialized for aquatic life, with no hind limbs, greatly modified skulls, and many physiological adapta- tions that allow them to remain underwater for extended periods. Male sperm whales [Physeter catodon) are the champion divers among marine mammals, diving for more than an hour at depths possibly to 2,000 meters. Marine species occur in three diverse orders of mammals: • manatees and dugongs in the order Sirenia, • sea otters, polar bears, and pinnipeds in the order Carnivora, and • whales, dolphins, and porpoises in the order Cetacea. There are only four extant species in the Living Marine Mammalian Species by Order Sirenia Carnivora Cetacea dugongs (1) sea otters (1) baleen whales (11) manatees (3) polar bears (1) beaked whales (19) pinnipeds (34) sperm whales (3) river dolphins (5) dolphins and porpoises (41) FALLAWINTER 1995 Sirenia: the dugong (Dugong diigon) and three manatee spe- cies {Trichechus spp.). All occur in tropical to warm temperate regions. The pinnipeds, with approximately 34 species, are the most numerous group of marine mammals within the Carnivora. The most diverse pinniped family is the Phocidae or true seals with 18 species. The fur seals and sea lions are represented by 15 species. The thirty- fourth species, the walrus [Odobenus rosmarus), is the only remaining member of the family Odobenidae (pinnipeds with tusklike canines) that flourished from the early Miocene to the Pliocene (24 to 2 million years ago). The diver- sity of many animal groups increases in habitats nearer the equator; however, the pinnipeds show the reverse pattern — only the monk seals {Monacluis spp.) occur and breed in the tropics. The cetaceans, with some 80 species, are the largest group of marine mammals. The great baleen whales are well described, in part be- cause they were commercially hunted, and the last of the 1 1 species now recognized, Bryde's whale [Balaenoptera edeni). was named in 1879. A sperm whale, Physeter catodon, blows at the surface off Dominica in 1991. This adult male is about 15 meters long. The northern right whale dolphin, Lissodelphis borealis, Is a morphulugiccdly highly derived denizen of cool North Pacific waters. Lacking the usual stabilizing dorsal fin of other dolphins, it still manages to swim at high speeds, perhaps using its laterally compressed tail stock for hydro- dynamic stabilization. In contrast, new species are still being discov- ered among the more numerous toothed whales. Eleven species have been described since 1900. The most diverse family of marine mammals is the Delphinidae with about 35 species that fill many ecological niches. The beaked whales, family Ziphiidae, are the sec- ond most diverse family, with 19 species, but are also the most poorly known group because of their secretive habits. Seven new species of beaked whales have been described in this cen- tury. Six of these spe- cies are in the genus Mesoplodon, including the newest species, which was described in 1991 (see photo overleaf). Another new species of this genus may be described in the near future. Morphological diversity in cetaceans is as great or greater than what is seen in most terrestrial mam- mal groups, and the toothed whales are more morphologically diverse than the baleen whales. Toothed whales range from generalized forms, such as the bottlenose dolphin {Tursiops truncatus) to highly specialized forms such as the right whale dolphin [Lissodelphis borealis], which has no dorsal fin and is almost snakelike in its streamlining. The toothed whales range in weight from the tiny river dolphins weighing 3 kilograms to the giant sperm whales weighing over 100,000 kilograms. Substantial morphological diversity also occurs vdthin species of cetaceans; common observation is that animals of a given species living in open ocean tend to be larger than members of the same species in closed seas. Similarly, inshore forms of the same species are usually larger than offshore forms. This has been shown, for example, for the Atlantic spot- ted dolphin (Stenella frontalis) , the pantropical spotted dolphin (S. attenuata], and the spinner dolphin (S. longirostris) . A baleen whale, Eubalaena glacialis, known as the right whale, shows the baleen plates it uses to filter its food from sea- water. The photo was taken in Cape Cod Bay in 1974. The right whale is on the endangered list. OCEANUS Recent DMA studies of what was thought to be a single spe- cies, the com- mon dolphin. Delphinus delphis, revealed instead two species: Delphi- nus delphis, top, and Delphinus capensis, bot- tom. The subtle morphological dijferences in- clude length of the beak, size and shape of the flipper stripe, and overall contrast of coloration. The recent explosion of molecular tech- niques, in particular those utilizing the Poly- merase Chain Reaction (PCR), has led to what may be a quantum jump in our knowledge of marine mammal species diversity. For example, recent DNA studies of what was thought to be a single species, the common dolphin (Delphi- nus delphis), revealed the existence of two spe- cies, now known as the short-beaked common dolphin (D. delphis) and the long-beaked com- mon dolphin [D. capensis). As yet uncompleted DNA work indicates that similar hidden diver- sity may lie within other supposed cetacean species, for example, the bottlenose dolphin and Bryde's whale, so the most basic taxonomy of even the more common cetaceans is not yet fully resolved. The new molecular techniques, together with the development of equipment for biopsying live marine mammals at sea, are also opening new vistas in defining populations below the species level. For example. National Marine Fisheries Service scientists are presently using both nuclear and mitochondrial DNA to examine population structure in white whales (Delphinapterus leucas), harbor seals (Phoca vitulina), spinner dolphins, and other species. Historically, marine mammal diversity was threatened mainly by direct human exploita- tion. Almost all the large species of ceta- ceans and many smaller ones, many species of pinnipeds, and all sirenians were reduced to very small numbers. Humans have exterminated three species and at least one subspecies of marine mammal within the last 225 Tlw newest species of beaked whale, Mesoplodon peruvianus, was described in 1991. This photo was taken in a Peruvian fish market. years. The best known case is the Steller's sea cow {Hydrodamalis gigas) , which was discovered in 1741 and extirpated by 1768. The lapanese sea lion [Zalophus californianus japonicus) has not been reported since the end of World War II. The most recent extinc- tion was the Caribbean monk seal (M. tropicalis) in the 1950s. In addition, two North Atlantic gray whale populations [Eschrichtius robustus) were extinct by the early 18th century. Several other overexploited populations of whales still have very low population levels (in the hundreds of animals) throughout various parts of the Northern Hemisphere. These in- clude Okhotsk Sea bowheads [Balaena mysticetus), western North Pacific gray whales, and two North Pacific and two North Atlantic right whale [Eubalaena glacialis) populations. Fortunately, a few large whales, such as the California gray whale, have recovered to near historical population levels. Today, marine mammal populations are threatened or endangered by numerous human activities including direct exploitation and culling, incidental catch in fisheries, and habi- tat loss and degradation. Much can still be done to conserve marine mammals, but time is running out for some species and populations. The most endangered seals are the two re- maining species of monk seals. The Mediterra- nean monk seal (M. monachus) population contains less than 500 individuals dispersed in small remnant groups across the broad range of the species. Counts of the Hawaiian monk seal (M schauinsiandi) have declined 60 percent since the late 1950s and have been declining at about 5 percent annually since 1984. The cur- rent population is estimated at less than 1,500. Some other pinni- peds have also under- gone dramatic de- clines in recent years. The population of SteUer sea lions [Eumetopias jubatus) in US waters was close to 190,000 animals in the 1960s but was just over 50,000 in 1994, more than a 70 per- cent decline. Similarly, ♦ FALLAA/INTER 1995 MBI, WHcil I.IHKARY WH IflBP 0 'iJil' ,jliHfJ^iMiiiiiiA^, '^^L the Kodiak Archipelago in Alaska was once a major breeding area for harbor seals (Plioca vitiilina], but the population declined by 90 percent between the late 1970s and 1992. These declines are due to poor juvenile survival, which may be related to a reduced food supply or a natural fluctuation of the resource, entangle- ment in fishing gear, or other human activities. The baiji or Chinese river dolphin [Lipotes vexiller) is the most endangered cetacean. These freshwater dolphins are often killed by entanglement in fishing gear and collisions with boat propellers. As a last effort to save the baiji, the lUCN (World Conservation Union) Cetacean Specialist Group and the Institute of Hydrobiology in Wuhan have started a program to capture as many of the remaining dolphins as possible and move them to the Shishou Seminatural Reserve. During the past three years, a small group of finless porpoises were used to test the suitability of the reserve. They appear to be surviving well and several calves have been born. How- ever, the extremely small size of the sur- viving baiji population does not bode well for the continued exist- ence of the species. In summary, just as we are gaining a fuller understanding of diversity in marine mammals, that diver- sity is being rapidly and progressively eroded by human activities. Scientists must accelerate their efforts on the conser- vation biology of this group ifwe are to provide the informa- tion needed to stem this erosion. A few references for readers interested in additional information on marine mammal biodiversity include: 1) "The plight of the forgotten' whales". Oceanus 32(1):5-11, by the three authors of this article, 1989: 2) "Evidence for tu'o species of common dolphins (genus De\ph\nus) from the eastern North Pacific,"Los Angeles County Museum Contributions in Science, 442:1-35, by J. E. Heyning. and W. F. Perrin, 1994: "A whale of a new species," Nature 350:560, by A.'. Ralls and R. L Brownell. ]r, 1991: and "Genetic analysis ofsympatric morphotypes of common dolphins fge«i/5Delphinusj, "Marine Biology 119:159- 167, by R Rosel, A. Dizon, and J. E. Heyning 1994. Robert Brownell, Jr, and William Pernn work for the National Marine Fisheries Service in La Jolla, CA, They are, says Katherine Ralls, bonafide experts on marine mammals, particularly small cetaceans, Ralls works for the National Zoological Park in Washington, DC. She originally studied terrestrial mammals, but the other authors, especially Brownell, have involved her in so many marine mammal projects over the last 20 years that she can now pass as a marine mammal expert. All three authors have worked on marine mammal conservation issues. They have collabo- rated on a previous article for Oceanus and a workshop on the biology and conservation of the river dolphins, held in Wuhan, People's Republic of China. The first sighting ofSteller's sea cow ii'as re- corded in 1741 in the Bering Sea off what is now Bering Island. The survivors of the Russian explorer Vitus Bering's second expedition to North America reported that the sea cow's meat ivas comparable to beef and its fat tasted like sweet almond oil. Fur-hunting expeditions killed off some 1,500 sea cows over the next 27 years, and by 1 768, the species was extinct. The song of the humpback whale, Megaptera novaeangliae, is the most ivell- known whale sound, and this animal is also among the most familiar to whale ivatch participants. OCEANUS ,t^N0,Gfl4, 1930 Woods Hole Oceanographic Institution Woods Hole, MA 02543 508-457-2000