The Paleontological Society Short Course this year features the history of brachiopod research, especially since the beginning of the revision of Part H, Treatise on Invertebrate Paleontology, in the early 1990s. The first version of Part H was published in 1965 and the only previous Paleontological Society Short Course to deal with brachiopods was held in 1981 at the Cincinnati meeting of the Geological Society of America. At that time, the day was split between the bryozoans and brachiopods, with a nod to the phoronids, under the rubric of lophophorates.

Brachiopods within the subphyla Linguliformea Williams et al., 1996 and Craniiformea Popov et al., 1993 comprise most, but not all, of the taxa previously grouped together in the Class Inarticulata Huxley, as defined in the first edition of the brachiopod volume of the Treatise on Invertebrate Paleontology (Rowell, 1965). The phylogeny and classification of the inarticulated organophosphatic-shelled linguliforms (typified by the modern lingulides; Fig. 1) and the inarticulated and partly articulated organocarbonate-shelled craniiforms (typified by the modern craniides; Fig. 1) have been debated during the last decade, in connection with attempts to analyze the phylogeny of the entire phylum (for summaries see Holmer et al., 1995; Carlson, 1995; Williams et al., 1996, 2000a). Much of the discussion has been centered around the conflicting results of phylogenetic analyses, either based exclusively on the anatomy and morphology of the Recent taxa (e.g., Popov et al., 1993; Carlson, 1991, 1995), or including also the more numerous, mainly Cambro-Ordovician, fossil taxa (e.g., Carlson, 1991; Holmer et al., 1995; Williams et al., 1996, 2000a). The cladistic analyses have either tended to support the monophyly of the Class Inarticulata (e.g., Carlson, 1991, 1995), or indicated that this group is paraphyletic (e.g., Popov et al., 1993, Holmer et al., 1995; Fig. 1). The resulting cladograms in the comprehensive studies of fossil and modern taxa by Williams et al. (1996, 2000a), which were the basis for the classification adopted in the latest edition of the brachiopod volume of the Treatise on Invertebrate Paleontology, largely support a monophyletic Subphylum Linguliformea; however, the phylogenetic position of the Craniiformea remains uncertain, although it may represent a monophyletic group (Fig. 1). Recent molecular phylogenies have given some support to a monophyletic Linguliformea, but also more surprisingly have indicated that the phoronids might form a clade (formally recognized as the Subphylum Phoroniformea Cohen 2000) within the Brachiopoda, possibly together with the inarticulated modern forms (Cohen and Gawthrop, 1996, 1997; Cohen, 2000).

It is a great understatement to say that our knowledge of morphologic, stratigraphic, and paleobiogeographic variation among brachiopods has expanded enormously over the past 35 years. The number of brachiopod genera discovered and named since 1965 has more than doubled the approximately 1700 genera described in Part H of the Treatise on Invertebrate Paleontology (Williams and others, 1965). 4827 genera and subgenera named as of 1995 are listed in Doescher (1996); it is likely that over 5000 exist by now. Some are certain to be reevaluated as synonyms, but we have no reason to believe that this would apply to most new genera. This remarkable increase in raw data required a thorough reexamination of the phylogenetic relationships and higher-level classification of the group, encouraged largely by the revision of Part H of the Treatise, initiated in the late 1980s by Alwyn Williams.

When Charles Darwin wrote that “Our classifications will come to be, as far as they may be so made, genealogies…we have to discover and trace the many diverging lines of descent in our natural genealogies by characters…which have long been inherited” (Darwin, 1859), he presciently laid down aims and objectives of systematics that have become attainable only since the development of genetics and of molecular approaches to systematics. Genetics elucidates the heredity of characters; molecular systematics measures the degrees of relationship between diverging lineages (Griffiths et al., 1993; Page and Holmes, 1998; Graur and Li, 2000). This article attempts to review these essential aids to the art and science of brachiopod classification as they stand today. Every step so far taken to promote molecular approaches to brachiopod systematics has yielded new, unexpected, and valuable information, but there remains massive scope for new work and new workers.

The chapter on anatomy in the Treatise on Invertebrate Paleontology (Part H, Brachiopoda, revised) (Williams et al., 1997) is the most current and comprehensive treatment that we have of reproduction and development in these animals. My contribution to this short course is a commentary on and addendum to this review. The study of the developmental biology of extant brachiopods describes a large part of their life history and defines several of the parameters that have to be taken into account when thinking about how a given set of genes will make it to the next generation (Havenhand, 1995). Some extant brachiopod genera like Discinisca and Crania (Neocrania) belong to families that first appeared in the fossil record during the Lower Ordovician or, as in the case of Glottidia, to a superfamily that first appeared during the Lower Cambrian. Studies on the development of these extant animals provide a picture of what the development of their Lower Paleozoic ancestors might have been like.

Brachiopods are one of the few phyla to have lived in the world's oceans for the last 550 million years, since early Cambrian times. For much of that time they were the dominant shelled form in marine aquatic environments. Throughout all of this time their physiological structure has remained unchanged, and for some extant species their external anatomy is remarkably similar to early Cambrian forms (e.g., some inarticulated lingulids and some articulated rhynchonellides). They perform all of the basic physiological functions common to the vast majority of marine ectotherms, namely feeding, growth, reproduction, and metabolic and excretory processes. Their bodies are enclosed by two unequal shells, which are oriented dorsally and ventrally, as opposed to the lateral, equal shells in bivalve molluscs. The ventral valve in articulated species has a characteristic opening for the pedicle to pass through. In some articulated brachiopods the shells are traversed by many extensions of the mantle tissue called caeca (e.g., in Waltonia inconspicua, or Terebratulina retusa), while in others the shells form a continuous solid integument (e.g., rhynchonellides such as Notosaria nigricans). The former are the punctate brachiopods and the latter are impunctate species.

Biominerals are produced in all five kingdoms. Calcium carbonate and calcium phosphate are the most abundant biominerals performing many functions including protection and skeletal support. The phylum Brachiopoda is divided into three subphyla: Linguliformea, Craniiformea, and Rhynchonelliformea (Williams et al., 1996). The Linguliformea possess inarticulated phosphatic valves. Articulation is also lacking in the calcitic valves of the Craniiformea while the calcitic valves of the Rhynchonelliformea are articulated. The paired valves of the brachiopod shell are one of the earliest examples of biomineralization. The existence of different mineral regimes and shell ultrastructures within the phylum makes the brachiopods ideal candidates for the study of biomineralization. The formation of brachiopod valves is an example of organic controlled mineralization, a term introduced by Lowenstam (1981) to describe biomineralization which is under genetic control via specific organic material controlling the precipitation and formation of the biomineral. In organically induced biomineralization (Lowenstam, 1981), organic molecules provide a nucleating surface on which mineral precipitates. Such precipitation continues as long as the solution is saturated with respect to the mineral ions. Stromatolite formation is an example of organically induced biomineralization. In brachiopod shell formation, organic molecules are not solely involved in nucleation. By binding to specific crystal faces, organic molecules inhibit growth along certain crystal axes and enhance growth in other directions, influencing the growth and formation of organically controlled biominerals. Finally, organic molecules inhibit biomineral growth. Thus, a suite of organic molecules is involved in brachiopod shell formation, their spatial and temporal presentation resulting in the formation of species-specific valves.

The Brachiopod shell provides more than protection for the organism dwelling within: it provides a structural record of growth and a chemical record of fluctuations in the environment in which it formed. Typically, a shell formed by an accretionary process will have growth lines visible on the shell surface that can be used to measure chronological age directly. This is common in many fossil and modern bivalve molluscs where growth bands are prominent (see, e.g., Lutz and Rhoads, 1980; Jones, 1983; Tanabe, 1988). However, growth increments are not always so easily identified in other shelled organisms. For instance, growth lines in brachiopod shells are often indistinct, and the time interval represented by each growth line is uncertain. When growth information can't be obtained directly, we must turn to an alternative method (Jones and Gould, 1999). Fortunately, periodicities in the chemical composition of a shell may serve as proxies for age (e.g., Buening and Carlson, 1992). The shell chemistry of brachiopods reflects, for the most part, the changes in chemistry of the surrounding seawater. For instance, partitioning of oxygen isotopes (18O/16O ratio) in shell carbonate is sensitive to temperature (e.g., Hoefs, 1997). Thus, if brachiopods precipitate shells in water that is affected by seasonal temperature variations, then fluctuations of the 18O/16O ratio of the shell should provide a record of annual temperature variation in ambient seawater. Consequently, this seasonal record may be used in conjunction with structural growth lines to reconstruct the life history of the individual, as well as the local environmental conditions in which the brachiopod lived.

In the first Paleontological Society short course dealing with functional morphology of extinct brachiopod taxa, Grant (1981, p. 127) emphasized analogy with living species in reconstruction of life habits, but then cited many pitfalls of taxonomic uniformitarianism. The present is hardly the key as to how brachiopod shells functioned in the past, given the plethora of morphologic structures that have vanished since the Paleozoic diversity climax of articulates. Although direct observations (Table 1), such as clasping spines encircling a blastoid columnal (Grant, 1963), or biomechanical tests (e.g., Thayer, 1975a) and flume experiments (e.g., LaBarbera, 1978) on living articulate brachiopods, enable convincing assertion of a morphologic structure's function, indirect methodologies (Savazzi, 1999)(Table 1) have supplanted mere analogies with living brachiopods. Indirect methodologies (Savazzi, 1999, p. 6) reconstruct and infer the function of skeletal structures from theoretical (morphospace) and actual scaled models of extinct taxa, hydrodynamic and biomechanical tests on empty shells, commensal associations with fossilized epibionts, clinal variations in structures over paleogeographic and paleobathymetric gradients, and post-mortem, post-burial orientational evidence. The common denominator of indirect methodologies is that inferences are made on dead shells usually, but not always, from extinct taxa. Although such inferences of skeletal function of extinct taxa vary in the rigor by which they are deduced, indirect methodologies have been increasingly refined over the last 20 years (Table 1).

Ecology is the study of the relationships between organisms and their environments. For brachiopods living in marine habitats the main external factors of interest are temperature, water chemistry, light characteristics, and oxygen availability. However, organisms also modify their environments and in many cases the biotic environment may dictate changes or organism responses. Factors of importance here for animals are resources consumed (usually food items), predators that consume the species under study, organisms that compete for resources, organisms that provide benefits such as symbiotic arrangements or new habitat for colonisation, and those organisms which reduce the quality of the environment such as pathogens or parasites.

Because of their great abundance, widespread distribution, excellent preservation potential (Foote and Sepkoski, 1999), and a tendency not to disarticulate after death, brachiopods are ideal subjects for paleoecological research involving morphometrics, population analysis, and phylogenetics. Paleoecology is a subdiscipline that demands large data sets and statistical tests, and brachiopods provide the opportunity to create such databases. I fully expect to see brachiopods play a major role in the coming years in studies on the cutting edge of paleoecology. My approach in this chapter is to provide some background and tools that hopefully will inspire many new ideas for using brachiopods in the study of paleoecology. My intent is not to convince anyone of the correctness of my ideas, but rather to encourage future research in these directions.

Over two hundred years ago the Swedish scientist Carl Linnæus (1781), in an analysis of the biogeographic patterns of living organisms, suggested that all species originated in Paradise. Although there has been considerable progress in the understanding of biogeographical patterns during the intervening two centuries, modern debate has focused on the general applicability of the concept of faunal realms together with the relevance of dispersal, panbiogeographic, and vicariance models (Nelson and Platnick, 1981). To date, studies of Paleozoic brachiopod biogeography have no strong theoretical base; rather the various numerical techniques available, including both cladistic and phenetic methodologies, have helped organize the growing amount of distributional data into recognizable and useful structures.

The main focus here is the paleobiogeography and paleoecology of Mesozoic-Cenozoic Brachiopoda, mostly the forms referred to as articulated brachiopods. These are the forms I have worked on primarily for the past two decades. To put these creatures in context some attention should be paid to their evolutionary developments during this time (Fig. 1). The articulated brachiopods that continued into the Mesozoic from the Paleozoic are the rhynchonellides (internally with two processes or prongs, crura, to support the lophophore, Fig. 2.1; examples of external morphology shown in Figs. 3.1–3.5, 4.1); the terebratulides with a brachidium in the form of a loop to support the lophophore (“short-looped” in the Terebratulidina, Figs. 2.2, 2.3, 3.6–3.8, 4.2–4.5; “long-looped” in the Terebratellidina, Figs. 2.4, 2.5, 3.9–3.13); and two forms with a spiral lophophore support, the athyridides and spiriferides (these latter two both become extinct during the Mesozoic; Figs. 2.6, 2.7, 3.14–3.16). The thecideides are micromorphic, cryptic, articulated brachiopods (Fig. 3.17) that have much-debated origins (Baker, 1990; Jaecks, 2000).

Broad patterns of originations and extinctions of genera, as well as families and higher groups, have always interested those who study the fossil record (e.g., Sepkoski, 1984). They record an important part of the major changeovers, and thus the dynamics, of marine ecosystems over time (Droser et al., 1996; Droser and Sheehan, 1997). This seems especially true for the Paleozoic, when brachiopods were the dominant shelly animals on the seafloor in tropical, temperate, and even cold water settings. Attempts have also been made to determine turnover patterns at the species level (Patzkowsky and Holland, 1997), though this is a much more difficult task, as the validity of species depends a great deal on the skills of the taxonomist. A similar problem is the comparative analysis of diversification data based on a single continent, e.g., North America, as related to others (Miller, 1997a, b); though Laurentia is probably better studied than most areas except western Europe. The exercise of studying broad-scale generic gains and losses for the brachiopods is at the present time preliminary (only three volumes of the revised Treatise are published). The 1965 Treatise contains fewer than 25% of the genera known in detail and described today, with an almost exponential increase in taxonomic description since the 1960s (Williams, 1996). Since then, there have been dramatic revisions and re-interpretations of the evolutionary history of the major brachiopod families, as a new generation of brachiopod workers arrived and matured. We also have a considerably improved knowledge of molecular relationships within the Brachiopoda (Cohen and Gawthrop, 1996). Sound taxonomy is the fundamental basis for sound theoretical discussion of the nature and origins of major changeovers in phyla such as the Brachiopoda. Unfortunately, there are presently relatively few, active brachiopod specialists, as taxonomy has given way to other, more general interests.