2.17 Evolution of Vertebrate Olfactory Subsystems

H L Eisthen and G Polese, Michigan State University, East Lansing, MI, USA

2.17.1 Introduction 355 2.17.1.1 What is Olfaction? 355 2.17.1.2 Components of the Vertebrate Olfactory System 357 2.17.2 Evolutionary Changes in the Vertebrate Olfactory System 360 2.17.2.1 Chordates and Basal Craniates 360 2.17.2.2 Lampreys 361 2.17.2.3 Cartilaginous Fishes: Sharks, Skates and Rays, and Chimaeras 363 2.17.2.4 Ray-Finned Fishes 365 2.17.2.5 Lobe-Finned Fishes: Lungfishes and Coelacanths 370 2.17.2.6 Tetrapods: Amphibians, Reptiles, and Mammals 372 2.17.3 Conclusions 386 2.17.3.1 Evolutionary Changes in the Organization of the Olfactory Epithelium 386 2.17.3.2 Evolutionary Changes in the Organization of the Olfactory Bulbs 387 2.17.3.3 Evolutionary Changes in the Organization of Central Olfactory Projections 389 2.17.3.4 Evolution of Vertebrate Olfactory Subsystems 390

Glossary in the respiratory, circulatory, and digestive systems that detect gasses, ions, and nutrients. In general, these olfactory The olfactory system in any animal is the chemosensory systems consist of isolated sensory cells system primary sensory system that responds to that project to the spinal cord or hindbrain. These chemical stimuli emanating from a dis- systems will not be discussed here. tant source. In this article, we will consider three interrelated pheromone A chemical cue that, when released by an olfactory subsystems: the main olfactory system, or individual, elicits specific behavioral or physiological responses from olfactory system proper; the vomeronasal system; conspecifics. and the terminal nerve. The main olfactory system terminal The most anterior cranial nerve in verte- comprises receptor neurons located in a specialized nerve brates. The terminal nerve releases sensory epithelium in the nasal cavity, as well as the compounds into the nasal epithelia, central projections of these neurons. The receptor modulating activity of sensory receptor cells of the olfactory epithelium develop from the cells. nasal placode, and the ingrowing fibers of the devel- vomeronasal A discrete olfactory subsystem present in oping sensory neurons are involved in development system tetrapods that differs morphologically of the olfactory bulb at the rostral pole of the pro- from the olfactory system. Its function spective telencephalon (Gong and Shipley, 1995; is unclear. Graziadei and Monti-Graziadei, 1992; Long et al., 2003). 2.17.1 Introduction The vomeronasal system, or accessory olfactory system, also develops from the nasal placode, and is 2.17.1.1 What is Olfaction? present as a discrete sensory system only in tetrapods The term olfaction is commonly applied to chemosen- (see The Evolution of the Vomeronasal System). The sory systems that detect chemicals emanating from a vomeronasal epithelium is sequestered in the vomer- distant source. Other chemosensory systems generally onasal organ, also known as Jacobson’s organ require physical contact with the source for detection, (Figure 1). The sensory epithelium contains and this sensory modality is called gustation. The some what different cell types than does the vertebrate olfactory and gustatory systems are anato- main olfactory epithelium, and the vomeronasal mically distinct; the latter is discussed in a separate receptor neurons terminate in microvilli, whereas article (Evolution of Taste). Fibers of the trigeminal olfactory receptor neurons can terminate in cilia nerve also detect chemical stimuli, as dochemosensors or microvilli or both. In some vertebrates, 356 Evolution of Vertebrate Olfactory Subsystems

ob aob en

ns ob vno aob en

vno ns

Figure 1 Illustration of the locations of the olfactory epithelium, inside the nasal sac (ns), and of the vomeronasal organ (vno) in a frog and a snake. The nasal sac opens to an external nostril (en) on the dorsal surface of the snout, whereas the vomeronasal organ opens to the nasal sac in frogs and oral cavity in snakes. The axons of the olfactory receptor neurons project to the olfactory bulb (ob), and those of the vomeronasal receptor neurons project to the accessory olfactory bulb (aob). the axons of the vomeronasal receptor neurons refer to this structure simply as the ‘olfactory bulb’ form a separate cranial nerve, the vomeronasal to facilitate comparisons across vertebrates. nerve, but in others these axons run alongside the The terminal nerve, or nervus terminalis, is the olfactory nerve and the two cannot be distin- most anterior of the cranial nerves. It extends guished using conventional light microscopy. between the nasal cavity and basal forebrain, and, Vomeronasal receptor cell axons project to the because of its anatomy, has been thought to serve accessory olfactory bulb, a structure that is histo- sensory function (e.g., Demski and Northcutt, 1983; logically distinct from the main olfactory bulb Rossi et al., 1972). Nevertheless, recordings from (Figure 1). The secondary projections of the olfac- the terminal nerve have failed to detect sensory tory and accessory olfactory bulbs differ. The activity (Bullock and Northcutt, 1984; White and molecular, physiological, and anatomical differ- Meredith, 1995), and the terminal nerve is now ences between the vomeronasal and olfactory thought to function in modulating activity in the system suggest that the two subsystems serve dif- olfactory epithelium (reviewed in Oka, 1992; ferent behavioral functions, but the nature of this Wirsig-Wiechmann et al., 2002a). The terminal difference is unclear. Although the vomeronasal nerve usually contains a ganglion, the location of system is often presumed to be specialized for which varies across groups of vertebrates. The detecting pheromones, it responds to a variety of neurites extending outward from these bipolar odorants with differing behavioral significance in neurons sometimes comprise a separate nerve, all groups of tetrapods, and this presumption is but the fibers of the terminal nerve can also run unwarranted (see The Evolution of the within the olfactory or vomeronasal nerve, as illu- Vomeronasal System; Baxi et al., 2006; Halpern strated in Figure 2. The cells and fibers of the and Martı´nez-Marcos, 2003; Restrepo et al., terminal nerve contain neuromodulatory com- 2004; Shepherd, 2006). A different hypothesis pounds, including gonadotropin releasing posits that the vomeronasal system is specialized hormone (GnRH) and acetylcholine (reviewed in for detecting nonvolatile molecules. Although bet- Wirsig-Wiechmann et al., 2002a). The nerve can ter supported than the pheromone hypothesis, this also be immunohistochemically labeled with anti- hypothesis is still problematic (Baxi et al., 2006). sera directed against neuropeptide Y (NPY) and Throughout this article, we will refer to the nasal the molluscan cardioexcitatory neuropeptide chemosensory system present in all vertebrates as FMRFamide, but the two antisera may cross- the ‘olfactory system’ rather than the ‘main olfac- react with a single peptide (Chiba, 2000). We tory system’. Similarly, although it is common to use include the terminal nerve in this article not only the term ‘main olfactory bulb’ to refer to the pri- because of its role as a centrifugal portion of the mary target of the olfactory receptor cells in animals olfactory system, but also because the primary that possess a separate vomeronasal system, we will neurons of the terminal nerve develop from the Evolution of Vertebrate Olfactory Subsystems 357

pumping mechanisms, and nonsensory cells with motile cilia that create constant movement. Within the nasal cavity, the pseudostratified sensory epithelium contains three basic cell types (Figure 3b): olfactory receptor cells; sustentacular cells, a class of supporting cells; and basal cells, poa the progenitor cells that give rise to new receptor ob and sustentacular cells throughout life. The olfac- on tory receptor cells are bipolar neurons with a onf dendrite that terminates in cilia or microvilli or ns both. The membrane-bound odorant receptors are en localized to these processes, which are therefore Figure 2 Location of the terminal nerve around the nasal presumed to be the site of transduction (Menco, cavities and in the brain of a salamander. Anterior is down and 1997). The odorant receptors are part of the large to the left. The fibers of the terminal nerve extend between the preoptic area and nasal cavities, wrapping around the outside of superfamily of G-protein-coupled receptors the olfactory epithelium (lower left). en, external nostril; ns, nasal (GPCRs) that have seven membrane-spanning sac; ob, olfactory bulb; onf, olfactory nerve fascicles; on, olfac- domains (reviewed in Gaillard et al., 2004; tory nerve; poa, preoptic area. Mombaerts, 2004). The large family of odorant receptor genes has been suggested to contain two nasal placode (Schwanzel-Fukuda and Pfaff, fundamentally different classes that differ in the 1989), as do the olfactory and vomeronasal recep- size of the third extracellular loop (Freitag et al., tor neurons. 1995, 1998), although others have suggested that a larger number of groupings better describes the 2.17.1.2 Components of the Vertebrate Olfactory evolutionary history of the gene family (Niimura System and Nei, 2005). As depicted schematically in Elements of the olfactory system have been Figure 3c, the odorant receptor is coupled to an described from an evolutionary or comparative per- olfactory-specific G-protein (Gaolf), with alpha spective in several reviews (Ache and Young, 2005; subunits that are expressed in few other tissues; Eisthen, 1992, 1997, 2002; Hildebrand and when stimulated, the G-protein activates type III Shepherd, 1997; Nieuwenhuys, 1967). Here, we adenylyl cyclase (Nakamura, 2000; Ronnett and will describe the general features of the olfactory Moon, 2002). The details of olfactory transduc- system in nonmammalian vertebrates, illustrated in tion are well understood for only a small number Figure 3, to provide a context for understanding the of vertebrate species, and involve myriad mechan- changes that have occurred over the course of verte- isms (Firestein, 2001; Schild and Restrepo, 1998). brate evolution, as well as variations that occur One interesting feature of olfactory transduction is within specific lineages. In the sections that follow, that it often involves several steps, in which ions we will survey the structure and function of the entering through one channel gate another olfactory system in each class of vertebrates, noting (Eisthen, 2002). features that are new, unusual, or taxon-specific. Even before odorants contact receptors, they We will not describe the results of neurobiological interact with other molecules in the nasal cavity. studies designed to investigate general features of The dendrites of the receptor neurons protrude olfactory system function in vertebrates unless the into a specialized mucus produced by a combination results have interesting implications for understand- of glands and secretory cells, including goblet and ing olfaction in the group under investigation. At the sustentacular cells as well as Bowman’s glands. This end of the article, we will discuss innovations and mucus contains a variety of compounds, including variations, and their possible functional mucopolysaccharides, peptides, and amines implications. (Getchell and Getchell, 1992; Getchell et al., 1993; In all vertebrates, the olfactory epithelium is Zancanaro et al., 1997). In some vertebrates, it also sequestered inside the nasal cavity. Because adapta- contains specialized odorant binding proteins, solution occurs fairly quickly, the odorant-containing ble lipocalins produced in the lateral nasal glands. medium must be kept moving over the surface of The role of odorant binding proteins in olfactory the sensory epithelium. A variety of mechanisms are processing is unresolved (Pelosi, 2001; Tegoni employed to achieve this end, including nasal cav- et al., 2000). ities that allow water or air to flow through when At the opposite pole of the olfactory receptor the animal is locomoting or breathing, specialized neuron, the unmyelinated axon projects to the 358 Evolution of Vertebrate Olfactory Subsystems

(c)

(b)

(d) (a) Figure 3 Diagram illustrating the basic elements of the olfactory system in nonmammalian vertebrates. a, Schematic dorsal view of the forebrain and nasal sensory epithelia of a generalized nonmammalian vertebrate. The olfactory epithelium (oe) and vomeronasal organ (vno) lie rostral to the brain, and the axons of the receptor neurons project through the olfactory nerves (on) to the olfactory bulb (ob) and accessory olfactory bulb (aob), respectively. In general, four tracts project centrally. The accessory olfactory tract (AOT) projects from the accessory olfactory bulb to the amygdala (amg). The medial olfactory tract (MOT) projects to the anterior olfactory nucleus (aon), septum (sep), and medial pallium (mp). The extrabulbar olfactory pathway (EBOP) contains axons of primary olfactory receptor neurons that bypass the olfactory bulbs, projecting directly to the preoptic area (poa). The lateral olfactory tract (LOT) projects bilaterally to the striatum (st), lateral pallium (lp), dorsal pallium (dp), and the amygdala. b, The olfactory sensory epithelium consists of three types of cells: receptor neurons (N), sustentacular cells (S), and basal cells (B). The receptor neurons in vertebrates terminate in either cilia (C) or microvilli (M). c, Transduction occurs on ciliary or microvillar membranes. The receptor protein (R) has seven membrane-spanning regions and is coupled to a G-protein (GP). Odorant (Od) binding activates adenylyl cyclase (AC), gating an ion channel (IC). Often, ions entering through this channel gate another ion channel. d, Organization of the olfactory bulb. The axons of olfactory receptor neurons enter glomeruli (G), where they interact with dendrites from mitral cells (MC), the large output neurons of the olfactory bulb. Granule cells (GC) are also present in a deeper layer of the olfactory bulb. olfactory bulb in the rostral telencephalon. In the confusion concerning the number of classes of out- olfactory bulb, each unbranching axon forms put cells in nontetrapods. Mitral cells are large, and synapses with many other cells in tangles of fibers generally possess several dendrites that project into known as glomeruli, which are characteristic of different glomeruli. Within a glomerulus, each olfactory systems in a variety of animals (Eisthen, mitral cell dendrite arborizes extensively, making 2002). The olfactory bulb has a laminar organiza- large numbers of synapses with the axons of the tion. In many vertebrates these layers are not clearly olfactory receptor neurons. The granule cell layer differentiated, but in almost all the layers form con- lies between the mitral cell layer and the layer of centric rings around the ventricular region in the ependymal cells surrounding the ventricle. The center of the olfactory bulb. The outermost layer mitral and granule cell layers may be separated by consists of the axons of the olfactory receptor cells, an internal plexiform layer. In some vertebrates, the which course over the surface of the bulb before granule cell layer contains two classes of cells: the terminating in a single glomerulus in the subjacent granule and stellate cells. Both types of cells gener- layer (Figure 3d). This glomerular layer may contain ally have axons. Stellate cells possess multiple periglomerular interneurons. Below this, an exter- dendrites that arborize in the glomeruli. In contrast, nal plexiform layer is sometimes present, overlying a the granule cells have an oval-shaped soma, and the layer containing the cell bodies of the mitral cells, dendrites interact with neurites of other cells below the output cells of the olfactory bulb. The mitre- the glomerular layer or outside glomeruli. In some shaped cell body that gives these cells their name is animals, granule cell dendrites bear spiny processes, only obvious in tetrapods, leading to some whereas those of stellate cells do not. In the Evolution of Vertebrate Olfactory Subsystems 359 following discussion, we will use these definitions in among glomeruli is unclear, probably in part due to describing cell types present in diverse animals, differences in recording methods, odorants used, and regardless of the labels used by the authors of the species examined in different studies (e.g., papers cited. Wachowiak and Cohen, 2003; Wachowiak et al., The fibers projecting centrally from the olfactory 2002). Nevertheless, the olfactory bulb does not bulb consist of the axons of mitral cells. In some appear to contain a simple map in which an groups of vertebrates, axons of other classes of odorant can be identified by the location of the cells, such as the granule and stellate cells, may glomeruli that are stimulated; instead, temporal contribute to these tracts. In most vertebrates, two features of the response also play an important tracts extend from the olfactory bulb (Figure 3a). role in odorant recognition (e.g., Friedrich and The medial tract generally projects to ipsilateral Laurent, 2001; Laurent, 2002). ventral forebrain areas such as the septum and, in The vomeronasal epithelium contains the same tetrapods, to the anterior olfactory nucleus and basic cell types found in the main olfactory epithe- medial pallium/hippocampus. The lateral olfactory lium, but lacks Bowman’s glands (Parsons, 1967). tract projects bilaterally to lateral and dorsolateral Vomeronasal receptor neurons express GPCR pallial areas and to the amygdala, and ipsilaterally genes, but the two families of GPCRs found in the to the striatum. Thus, unlike other sensory systems, vomeronasal organ do not share strong sequence olfactory projections to the cortex are not routed similarity with those expressed in the olfactory through the thalamus, contributing to Edinger’s epithelium (Dulac and Axel, 1995; Herrada and (1904) famous hypothesis that the telencephalon Dulac, 1997; Matsunami and Buck, 1997; Ryba was originally an olfactory structure that was and Tirindelli, 1997). Transduction in vomerona- invaded by other sensory systems over the course sal receptor neurons seems to involve different of vertebrate evolution. In addition to the fiber G-proteins, second messengers, and ion channels tracts that arise from output cells of the olfactory than in olfactory receptor neurons, although only bulbs, many vertebrates possess a small extrabulbar a handful of species have been examined to date olfactory pathway that consists of axons of primary (reviewed in Halpern and Martı´nez-Marcos, olfactory receptor neurons that bypass the olfactory 2003). The axons of vomeronasal receptor neu- bulb and project directly to the preoptic area rons project to the accessory olfactory bulb, (Hofmann and Meyer, 1989; Szabo et al., 1991). caudal to the olfactory bulb. In general, the acces- Because of their similar projections, many older sory olfactory bulb contains only mitral, granule, papers appear to confound the terminal nerve and and periglomerular cells, and the layers are less extrabulbar olfactory pathway (Eisthen and distinct than are those in the olfactory bulb Northcutt, 1996). In the following discussion, we (reviewed in Lohman and Lammers, 1967; will refer the projection that arises from olfactory Nieuwenhuys, 1967). The accessory olfactory receptor neurons as the extrabulbar olfactory tract projects from the accessory olfactory bulb pathway, regardless of the name the authors primarily to portions of the amygdala that differ ascribed to it. from those receiving olfactory input via the lateral The olfactory system does not seem to contain olfactory tract. simple maps or even one-to-one functional relation- Although there are theoretical reasons for envi- ships. Olfactory receptor neurons tend to be broadly sioning that the receptors in a chemosensory system tuned, and a given cell will respond to many differ- should differ in their breadth of tuning (Hildebrand ent odorants, sometimes in different ways; similarly, and Shepherd, 1997), it is not clear that vertebrates a given odorant can evoke different responses from possess any narrowly tuned receptor neurons. To different receptor neurons (e.g., Dionne, 1992; illustrate this point, we will briefly discuss two Dionne and Dubin, 1994). In mammals, olfactory examples: ciliated versus microvillar olfactory receptor neurons expressing the same receptor genes receptor neurons in teleost fishes, and olfactory ver- project to the same glomeruli, which are located in sus vomeronasal receptor neurons in mammals. relatively stable positions within the olfactory bulb Researchers studying teleost fishes have used elec- (Mombaerts et al., 1996; Schaefer et al., 2001). The tro-olfactogram (EOG) recordings to examine neurons that project to a given glomerulus tend to odorant responses in patches of epithelium that are respond to the same sets of odorant stimuli (e.g., dominated by one receptor cell type. The results Bozza and Kauer, 1998), but a given odorant can indicate that ciliated cells respond to bile acids, activate many glomeruli across the olfactory bulb which can serve as pheromones in fishes, and that (e.g., Wachowiak et al., 2002). The effect of chan- microvillar cells respond to amino acids (salmonids; ging odorant concentration on spread of activation Thommesen, 1983); or that ciliated cells respond to 360 Evolution of Vertebrate Olfactory Subsystems amino acids, and that bile acids, steroids, and pros- systems are generally labeled ‘olfactory’. taglandin pheromones are preferentially detected by Nevertheless, there is no evidence that the vertebrate microvillar cells (goldfish; Zippel et al., 1997). A olfactory system is homologous with those present similar study with channel catfish indicated that in other phyla; instead, it appears that similar fea- both ciliated and microvillar cells respond well to tures have evolved independently several times for both amino acids and bile acids (Erickson and use in sensing odorants (Eisthen, 2002). The nearest Caprio, 1984), but a later study in the same species relatives to vertebrates are the urochordates (asci- using pharmacological agents to disrupt signaling in dians or tunicates), cephalochordates (lancelets), different cell types demonstrated that ciliated olfac- and hagfishes, which are considered craniates but tory receptor neurons respond to bile acids and not vertebrates. Ascidian larvae possess unpaired amino acids, and that microvillar receptor neurons anterior sense organs, including a photosensitive respond to nucleotides and amino acids (Hansen system and a balance sensor, but no candidate et al., 2003). Whole-cell recordings from olfactory homologue of the vertebrate olfactory system receptor neurons in rainbow trout indicate that (Lacalli, 2001; Nieuwenhuys, 2002). Thus, the ver- microvillar neurons respond selectively to amino tebrate olfactory system may not have arisen from a acids, and that ciliated neurons respond more system shared with the common ancestor with broadly to amino acids, a steroid, and conspecific urochordates. urine (Sato and Suzuki, 2001). Agmatine labeling Lancelets may or may not possess an olfactory suggests that a greater proportion of microvillar system that is homologous with that of vertebrates. than ciliated olfactory receptor neurons respond to Lancelets respond to various classes of sensory sti- amino acids in zebra fish (Lipschitz and Michel, muli, including chemical cues (Parker, 1908). They 2002). Thus, the relationship between cell morphol- lack a discrete olfactory organ, but a class of poten- ogy and odorant responses is unclear, and may be tially chemosensory cells that occurs in the rostral species-specific. To consider a different comparison epithelium has been described (Lacalli and Hou, for a moment, researchers often assume that vomer- 1999). A GPCR gene that bears some sequence onasal receptor neurons in tetrapods are much more similarity to vertebrate olfactory receptor genes is narrowly tuned than are olfactory receptor neurons. expressed in bipolar rostral sensory cells in Because of this bias, few studies have investigated Branchiostoma belcheri (Satoh, 2005). Given that the breadth of tuning of vomeronasal receptor neu- these cells bear axons, they appear to be one of rons (Baxi et al., 2006). Nevertheless, a few studies the classes of type I sensory cells described by that have examined the issue conclude that vomer- Lacalli and Hou (1999), who argued on morpho- onasal receptor neurons respond to a broad array of logical grounds that these cells are likely chemicals, both naturally occurring and artificial mechanosensory. In addition, the sequence was (e.g., Sam et al., 2001; Tucker, 1971). not subjected to rigorous phylogenetic analysis The olfactory system, broadly defined, plays an nor compared with those of many other GPCRs. important role in the behavior of most vertebrates. Given that the use of GPCRs in chemosensory Olfactory cues may be involved in orientation and receptor cells appears to be a trait that has homing, habitat selection, territoriality, species evolved several times independently (Eisthen, recognition, kin recognition, individual recognition, 2002), it is difficult to accept this observation as parent–offspring interactions, mate choice, court- strong support for the presence of a vertebrate-like ship and mating behavior, predator avoidance, olfactory system in lancelets. foraging, and food choice. A thorough review of The lancelet Branchiostoma floridae possesses the behavioral functions of the olfactory system in paired anterior nerves that, based on consideration each class of vertebrates is far beyond the scope of of their external topography, have been suggested to this article. Instead, for each group, we will provide be homologues of the vertebrate olfactory nerves a brief overview in which we illustrate the beha- (Lacalli, 2002). Although the author interprets the vioral significance of olfactory input. central target of these nerves as a possible homologue of the telencephalon, this region can also be interpreted as the equivalent of the vertebrate 2.17.2 Evolutionary Changes in mesencephalon or rostral rhombencephalon; if so, the Vertebrate Olfactory System these fibers are probably not homologues of the olfactory nerves (Northcutt, 2005). If Northcutt 2.17.2.1 Chordates and Basal Craniates and Gans’s ‘new head’ hypothesis is correct, then Animals in many phyla possess a sensory system for many rostral structures are evolutionarily new in detecting chemicals at a distance, and these sensory craniates; among these structures are the ectodermal Evolution of Vertebrate Olfactory Subsystems 361 placodes, including the nasal placode, and portions olfactory bulb to terminate diffusely along a path of the forebrain, including the olfactory system extending into the diencephalon, including the (Gans and Northcutt, 1983; Northcutt, 2005; hypothalamus and dorsal thalamus (Wicht and Northcutt and Gans, 1983). Northcutt, 1993). Given its location and targets, The condition in hagfishes should be instructive, this tract may constitute the extrabulbar olfactory as features shared by hagfish and vertebrates may pathway. Immunocytochemical data suggest that have been present in earliest vertebrates. Hagfish are hagfish lack a terminal nerve (Braun et al., 1995; scavengers as well as opportunistic predators Crim et al., 1979b; Jirikowski et al., 1984; Wicht (Shelton, 1978). They respond vigorously to odor- and Northcutt, 1992). ants from dead fish, and can use chemical cues to find carrion (Greene, 1925; Tamburri and Barry, 2.17.2.2 Lampreys 1999). In addition to their olfactory system, hagfish possess Schreiner organs, unusual chemosensory Lampreys have a biphasic lifecycle. The ammocoete organs that are distributed across the body surface larvae hatch in streams and rivers, then live buried (Braun, 1995; see Evolution of Taste). The relative in sediment for years, filter feeding, before meta- contributions of the olfactory and Schreiner organ morphosing and swimming downstream to live in system to behavior have not been determined, but larger bodies of water. Juvenile lampreys are gener- odorants such as L-amino acids, GABA, and ally parasitic, attaching to other fish to consume hydroxyproline evoke strong physiological their blood and flesh. When sexually mature, adult responses from the olfactory epithelium (Døving lampreys stop feeding and migrate upstream in and Holmberg, 1974). streams and rivers to spawn, after which they Hagfish possess a single midline olfactory organ, usually die. with the sensory epithelium folded across a radial In both Ichthyomyzon fossor and Petromyzon array of lamellae (e.g., Døving and Holmberg, marinus, the complexity and relative size of the 1974). The olfactory epithelium of the Atlantic hag- nasal cavity increases greatly during metamorphosis fish Myxine glutinosa contains both ciliated and from ammocoete to the juvenile form, suggesting that microvillar receptor neurons. Instead of the 9 2 olfaction may be important for free-swimming lam- microtubule arrangement that is typical of manyþ preys than for larvae (Leach, 1951; VanDenbossche cilia, including those on vertebrate olfactory recep- et al., 1995, 1997). Metamorphosed P. marinus tor neurons, the cilia have a 9 0 arrangement increase activity in response to odorants from trout (Theisen, 1973). þ and fish-derived amines, but not amino acids In Myxine glutinosa and Eptatretus burgeri, the (Kleerekoper and Mogensen, 1960). olfactory bulb contains mitral, stellate, and granule Adult male P. marinus migrate and build nests in cells, but the layers are indistinct, with somata of the spawning area before females arrive. The males each cell type occurring in several different layers emit a unique bile acid that is attractive to females, (Iwahori et al., 1998; Jansen, 1930; Holmgren, and to which females are highly sensitive (Li et al., 1919, in Nieuwenhuys, 1967). All three cell types 2002; Siefkes and Li, 2004). Migratory lampreys do have axons, and the dendrites of the mitral cells not home to particular streams; rather, they appear extend into several glomeruli. Periglomerular cells to simply seek suitable habitat for spawning do not seem to be present, but all authors describe (Bergstedt and Seelye, 1995). The means by which intraglomerular mitral cells with a single dendrite they do this is wonderful in its simplicity, for adult that arborizes in one glomerulus, and an axon that lampreys are attracted to odorants produced by presumably exits the olfactory bulb (Holmgren, healthy ammocoetes. Specifically, adults of several 1919; Iwahori et al., 1998; Jansen, 1930). species have been shown to be extremely sensitive to Projections of the olfactory bulb have been inves- bile acids produced by ammocoetes (Fine et al., tigated in the Pacific hagfish, Eptatretus stouti 2004; Li and Sorensen, 1997; Li et al., 1995), as (Wicht and Northcutt, 1993). A short fiber tract well as to two unique steroids that are released by projects from the medial olfactory bulb to the ipsi- ammocoetes (Sorensen et al., 2005b). Adults are lateral septum and through a dorsal commissure to highly sensitive to these compounds, and both the contralateral olfactory bulb. Fibers extend from types of cues are released in sufficient quantities to the lateral olfactory bulb project bilaterally and attract them (Polkinghorne et al., 2001; Sorensen widely in the forebrain, with targets that include et al., 2005b). Interestingly, the bile acids are not the striatum, all layers of the pallium, and the cen- released by nonfeeding ammocoetes, suggesting that tral prosencephalic nucleus. An additional group of these cues serve as a good indicator of habitat suit- fibers extends from the ventrolateral portion of the ability (Polkinghorne et al., 2001). 362 Evolution of Vertebrate Olfactory Subsystems

Lampreys possess a single midline nostril and nasal with cell bodies in the mitral cell layer, which extend cavity that develops from a single nasal placode. a single dendrite to arborize in one glomerulus; and Although earlier authors suggested that monorhiny those with cell bodies in the glomerular layer, which is the ancestral condition for vertebrates, the presence have dendrites that arborize in more than one glomer- of paired olfactory organs during development in ulus (Heier, 1948; Iwahori et al.,1987; Nieuwenhuys, P. marinus suggests that monorhiny may be a derived 1967). The latter class of cells may function more like condition (Kleerekoper and Van Erkel, 1960). If so, it periglomerular cells than like mitral cells. A more arose independently in hagfishes and lampreys. recent description of the olfactory bulb of L. japonica Within the nasal cavity, the olfactory epithelia of suggests that either granule or stellate cells are absent Lampetra fluviatilis and P. marinus contain only in this species (Iwahori et al.,1987). The cells that ciliated receptor cells, indicating that lampreys lack Iwahori et al. call ‘granule’ approximate our descrip- microvillar receptor neurons (Bronshtein and Ivanov, tion of stellate cells, above. These cells lack axons, but 1965; Thornhill, 1967; VanDenbossche et al., 1995). possess several dendrites that arborize in glomeruli, In contrast with the numerous and large families often spanning the width of the bulb (Iwahori et al., of odorant receptor genes described in other groups 1987). It is difficult to reconcile these different of vertebrates, lampreys (L. fluviatilis) appear to descriptions of bulbar organization, particularly for possess only a few small families of odorant recep- two species of Lampetra, but perhaps the discrepan- tors (Berghard and Dryer, 1998; Freitag et al., cies are due to developmental or methodological 1999). Of course, it is difficult to rule out the possi- differences. bility that lampreys possess many additional The olfactory bulb gives rise to four major cen- receptor genes that have not yet been identified. tripetal pathways in silver lampreys, Although the sequences possess features that are Ichthyomyzon unicuspis, illustrated in Figure 4 characteristic of vertebrate odorant receptor genes, (Northcutt and Puzdrowski, 1988). A group of including a large third extracellular loop (Berghard fibers that may comprise a homologue of the and Dryer, 1998; Freitag et al., 1999), phylogenetic medial olfactory tract originates from the ventro- analysis indicates that the lamprey odorant receptor lateral olfactory bulb and projects to the gene family diverged before the origin of the two ipsilateral septum, preoptic area, and possibly to main classes of odorant receptor genes present in the rostral portion of the striatum (Figure 4a). A other vertebrates (Freitag et al., 1999). The odorant lateral olfactory tract projects ipsilaterally receptor genes from lampreys also share strong simi- throughout the lateral pallium and either to or larity with histamine receptors, indicating that the through the dorsal and medial pallia, as well as odorant receptor genes in lampreys and other verte- to the posterior tuberculum and hypothalamus; brates may have been independently co-opted out of some fibers project contralaterally to a dorsal por- the larger GPCR superfamily (Berghard and Dryer, tion of the lateral pallium, septum, striatum, and 1998; Niimura and Nei, 2005). The genes are the posterior tuberculum and hypothalamus expressed in the olfactory epithelium of both ammo- (Figure 4b). A third group of fibers extends coetes and adults, suggesting that the quantitative through the ventromedial olfactory bulb to the differences in the size of the olfactory system in the striatum and preoptic regions, as well as to the two forms do not necessarily indicate qualitative hypothalamus and throughout the posterior tuber- differences in the odorants detected (Berghard and culum (Figure 4c; Northcutt and Puzdrowski, Dryer, 1998). 1988). This tract has been interpreted as an extra- In Petromyzon larvae, the olfactory bulb glomer- bulbar olfactory pathway (Eisthen and Northcutt, uli are histochemically heterogeneous and form six 1996). Finally, silver lampreys possess an olfac- distinct territories, suggesting a rough functional tory pathway through a dorsal commissure to specialization (Frontini et al., 2003). Whether this the adjacent dorsomedial neuropil, a fibrous organization is particular to ammocoetes or remains region that spans the length of the olfactory throughout the lifecycle is not known. A single layer bulb, and to glomeruli in the contralateral olfac- of glomeruli encircles the bulb (Heier, 1948; tory bulb (Figure 4d; Northcutt and Puzdrowski, Iwahori et al., 1987). As in hagfish, the layers are 1988). Fibers projecting to the contralateral olfac- indistinct, and the somata of some mitral cells can tory bulb have also been described in L. fluviatilis be found in the glomerular layer. The olfactory bulb (Heier, 1948) and in hagfish (Wicht and of L. fluviatilis contains mitral, stellate, and granule Northcutt, 1992), suggesting that a dorsal com- cells, all of which have axons that exit the olfactory missure carrying secondary olfactory fibers to the bulb (Heier, 1948; Nieuwenhuys, 1967). Two contralateral olfactory bulb may have been pre- classes of mitral cells appear to be present: those sent in the earliest vertebrates. Evolution of Vertebrate Olfactory Subsystems 363

(a) (b) (c) (d)

Figure 4 Schematic dorsal view of the forebrain in the silver lamprey (Ichthyomyzon unicuspis), illustrating the major olfactory projections. a, Projections of the medial olfactory tract. b, Projections of the lateral olfactory tract. c, Projections of the extrabulbar olfactory pathway. d, Projections through the dorsal commissure to the contralateral olfactory bulb. dn, dorsomedial neuropil; gl, glomerular layer; lp, lateral pallium; mp, medial pallium; ob, olfactory bulb; oe, olfactory epithelium; poa, preoptic area; pt, posterior tubercle; sep, septum; st, striatum; vh, ventral hypothalamus. Based on Northcutt, R. G. and Puzdrowski, R. L. 1988. Projections of the olfactory bulb and nervus terminalis in the silver lamprey. Brain Behav. Evol. 32, 96–107.

Heier (1948) first suggested that lampreys lack a oceans. Some species enter freshwater, and a few, terminal nerve, and since that time its presence in like the rays Paratrygon motoro and Himantura lampreys has been questionable. In other vertebrates, signifer, live exclusively in freshwater (Compagno the terminal nerve can be labeled with antisera direc- and Roberts, 1982; Mu¨ ller and Henle, 1841). ted against GnRH, NPY, or FMRFamide (Wirsig- Olfaction plays a major role in the life of cartilagi- Wiechmann et al.,2002a). These antisera fail to nous fishes. Their olfactory ability is legendary, and label any structures reminiscent of the terminal olfaction is important for prey detection nerve in larval or adult I. unicuspis, Lampetra japo- (Kleerekoper, 1978; Sheldon, 1911). Nevertheless, nica, L. planeri, L. richardsoni, Lethenteron japonica, the popular notion that sharks can detect even a or Petromyzon marinus (Chiba, 1999; Crim et al., small amount of blood diluted in the ocean over 1979b, 1979a; Eisthen and Northcutt, 1996; King many miles is an exaggeration. Electrophysiological et al.,1988; Meyer et al., 1987; Ohtomi et al.,1989; experiments demonstrate that sharks are able to Tobet et al., 1995; Wright et al.,1994). Fibers that detect components of human and bovine blood, as project from the nasal sac through olfactory bulb to well as other stimuli like amino acids and crab or the ventral forebrain have been labeled with injections squid extract, but responses to blood are no stronger of horseradish peroxidase or cobalt lysine into the than the responses to other stimuli (Hodgson and nasal sac of larval or adult lampreys (L. planeri, Mathewson, 1978; Kajiura et al.,2004b; Silver, Meyer et al., 1987; I. unicuspis, Northcutt and 1979; Zeiske et al.,1986). Puzdrowski, 1988; von Bartheld et al., 1987; von The role of olfaction in prey localization in sharks Bartheld and Meyer, 1988). Although this projection has been the subject of much study. Some species was originally interpreted as a terminal nerve, it may approach prey from downstream by swimming into instead be the extrabulbar olfactory pathway, which the current (rheotaxis). Tests in the open sea indi- was unknown at the time these studies were con- cate that lemon sharks (Negaprion brevirostris) use ducted (Eisthen and Northcutt, 1996). rheotaxis to find odorants from prey, but will con- tinue to swim against the current past the stimulus 2.17.2.3 Cartilaginous Fishes: Sharks, Skates source, suggesting that additional sensory cues are and Rays, and Chimaeras involved in short-range localization (Hodgson and The class chondrichthyes consists of cartilaginous Mathewson, 1971). Other species use a strategy fishes (sharks, ratfish or chimaeras, and skates and called klinotaxis in which they turn in the direction rays) that are widely distributed in the world’s of the nostril that is more strongly stimulated. Early 364 Evolution of Vertebrate Olfactory Subsystems naris-occlusion experiments with smooth dogfish (Mustelus canis) demonstrated that the animals turn persistently toward the open nostril when activated by an olfactory stimulus (Sheldon, 1911). Subsequent experiments with other sharks and rays indicate that klinotaxis is common in cartilaginous fishes (reviewed in Kleerekoper, 1978). Prey localization through klinotaxis has been suggested to constitute an important pressure driving evolution of increasing head width in hammerhead sharks (Sphyrnidae). Recent work by Kajiura et al. (2004a) demonstrates that the ability to localize odorants increases with increasing head width in sphyrnid sharks, although their olfactory sensitivity does not differ significantly from that of other Figure 5 Schematic ventral view of a shark, showing the char- families of sharks (Kajiura et al., 2004b). In addition acteristic nostrils with prominent nasal flap, and the direction of to its role in foraging, olfaction has also been impli- water flow (arrows). cated in reproductive behavior in sharks (e.g., Bleckmann and Hofmann, 1999; De Martini, 1978; Forlano et al., 2000). Nevertheless, this area Theisen et al., 1986; Zeiske et al., 1986, 1987). The of research remains largely unexplored. spiny dogfish Squalus acanthias was originally The olfactory organ of cartilaginous fishes con- reported to possess both ciliated and microvillar sists of a chamber inside a cartilaginous capsule. receptor cells (Bakhtin, 1976, 1977), but a later Incurrent and excurrent nostrils on the ventral study concluded that only microvillar receptor neu- snout allow water to flow over the olfactory rons are present (Theisen et al., 1986). Similarly, the mucosa. The anterior margin of the nasal flap is olfactory epithelium of the ratfish Chimaera mon- sufficiently diverse that it was used in species identi- strosa and those of skates and rays contain only fication until researchers learned that within-species microvillar receptor cells (Holl, 1973; Meng and variability is also high (Tester, 1963). In chimaeras, Yin, 1981). Nevertheless, one study with Oman the two olfactory chambers are adjacent, with a sharks demonstrates the presence of unusual olfac- cartilaginous septum that separates them down the tory receptor neurons that resemble the crypt-type midline. The incurrent nostrils are located medially olfactory receptor neurons present in ray-finned above the mouth, and the excurrent nostrils open fishes (Fishelson and Baranes, 1997; see discussion laterally near the edge of the mouth (Zeiske et al., in Hansen and Finger, 2000). 1992). In pelagic sharks, water is driven over the In cartilaginous fishes, the olfactory bulb is adja- olfactory epithelium by swimming movements, as cent to the olfactory sac, and is connected by short illustrated in Figure 5. In sedentary animals, flow olfactory nerves; thus, the tracts are much longer is driven by respiratory movements. Some species, than the nerves. The length of these tracts varies such as spiny dogfish (Squalus acanthias) and the across species, and in all cases they are formed by small-spotted catshark (Scyliorhinus canicula), use mitral cell axons with a characteristic black myelin specialized valve mechanisms to regulate flow over sheath (Dryer and Graziadei, 1996). The olfactory the olfactory epithelium (Theisen et al., 1986). bulb is somewhat laminated, but no plexiform Inside the nasal cavity, the olfactory organ con- layers separate the cell layers (Dryer and sists of a rosette composed of two rows of olfactory Graziadei, 1993; Franceschini and Ciani, 1993; lamellae situated on each side of a transverse raphe. Nieuwenhuys, 1967). Periglomerular cells appear In some species, such as Oman sharks (Iago oma- to be lacking. Sterzi (1909, in Nieuwenhuys, 1967) nensis), secondary lamellae greatly increase the described the olfactory bulb of spiny dogfish surface area of the sensory epithelium (Fishelson (Squalus acanthias) as containing a clear layer of and Baranes, 1997). In addition to the usual recep- mitral cells with primary dendrites that arborize in tor neurons, supporting cells, and basal cells present multiple glomeruli as well as secondary dendrites in all vertebrates, the olfactory epithelium of some that interact with fibers of other cells outside the cartilaginous fishes also contains goblet cells (Zeiske glomeruli. Granule cells have both axons and et al., 1986). In sharks, the ancestral condition smooth dendrites, and stellate cells also appear to appears to be the presence of only microvillar olfac- be present; Sterzi referred to the latter as ‘triangular tory receptor neurons (Reese and Brightman, 1970; cells’. In contrast, a more recent study of sharks Evolution of Vertebrate Olfactory Subsystems 365

(Sphyrna tiburo and Rhizoprionodon terranovae)and fibers have been described within the olfactory rays (Dasyatis sabina) found two types of mitral cells nerve in sharks and rays (Wu et al., 1992), but it is present, one with a loose dendritic arborization and not yet clear whether these fibers constitute part of the other with a more dense one (Dryer and the terminal nerve pathway. Graziadei, 1993, 1994, 1996). The dendrites of both Electrical stimulation of the terminal nerve gang- types arborize in a single glomerulus. Given the irre- lion in Atlantic stingrays (Dasyatis sabina) leads to gular laminar borders described by Dryer and an increase in GnRH levels in the brain (Moeller and Graziadei, it seems possible that their mitral cells Meredith, 1998), but electrical stimulation of the with loose dendritic arbors are actually stellate cells. terminal nerve does not significantly alter activity The central projections of the olfactory bulb have in the olfactory bulb (Meredith and White, 1987). not been examined in detail in cartilaginous fishes. Recordings from the terminal nerve in sharks and The lateral olfactory tract in sharks and rays projects rays have failed to detect sensory activity, although ipsilaterally to the area retrobulbaris, striatum, lat- efferent activity from the brain has been recorded eral pallium, and area superficialis basalis (Ebbesson (Bullock and Northcutt, 1984; Demski and and Heimer, 1970; Smeets, 1983). The existence of a Schwanzel-Fukuda, 1987; White and Meredith, medial olfactory tract is uncertain, and an earlier 1995). Further, activity of terminal nerve ganglion description of a widely projecting pathway (Smeets, cells can be modulated by application of acetylcho- 1983) has been reinterpreted as an artifact of the line or norepinephrine, suggesting that centrifugal degeneration technique used (Northcutt, 1995). An fibers may regulate activity of these cells (White and extrabulbar olfactory pathway has not been Meredith, 1993, 1995). Overall, the available data described in this group (Hofmann and Meyer, 1995). indicate that the terminal nerve in cartilaginous The terminal nerve was originally discovered in a fishes is not sensory, although it has not yet been small shark (Mustelus asterias) by Fritsch (1878), demonstrated to serve a modulatory function. who referred to it as a ‘supernumerary nerve’. Locy (1905) later described this structure in a vari- 2.17.2.4 Ray-Finned Fishes ety of sharks and rays, and named it the nervus terminalis because the nerve enters the brain Actinopterygii, the ray-finned fishes, is the largest through the lamina terminalis. The exact point of class of vertebrates, comprising nearly 25 000 spe- entry in the brain varies somewhat among species; cies. One division of ray-finned fishes, the teleost for example, in Squalus acanthias the terminal nerve fishes, contains over 23 000 species (Nelson, 1994). enters the telencephalon dorsally (Locy, 1905). In In addition to teleosts, the class Actinopterygii con- most cartilaginous fishes, the terminal nerve is com- tains paddlefish and sturgeons, gars, bowfins, and pletely separate from the olfactory nerve, and bichirs or reed fishes. Ray-finned fishes occupy an includes a visible ganglion located medial and exter- enormous diversity of aquatic ecological niches and nal to the nasal cavity. In some species, additional possess many specialized adaptations. We will ganglia occur along the nerve (Locy, 1905). describe here the main features of the olfactory sys- Electron microscopic examination of the terminal tem that are shared by the species that have been nerve in sharks and rays demonstrates that the cells examined to date. are heterogeneous. Most fibers are unmyelinated, Olfactory cues are critical for foraging in some but a few fibers in the proximal part of the nerve species of ray-finned fishes (Atema, 1980; Bateson, are myelinated (Demski and Schwanzel-Fukuda, 1890), and electrophysiological recordings from 1987; White and Meredith, 1987). The ganglion goldfish and carp (Carassius) reveal a high sensitiv- consists largely of unipolar cells, with a few bipolar ity to food-related odorants, with thresholds for 6 9 and multipolar cells present (White and Meredith, amino acids generally in the range of 10À –10À M 1995; Wu et al., 1992). Cells and fibers of the termi- (Goh and Tamura, 1978; Zippel et al., 1993). This nal nerve display GnRH-like immunoreactivity in sensitivity is greater than that typically measured in many species of sharks and rays (Chiba, 2000; other aquatic vertebrates (Hamdani et al., 2001). In Chiba et al., 1991; Demski and Schwanzel- addition to locating food by following trails of Fukuda, 1987; Stell, 1984; White and Meredith, extremely dilute odorants, many salmonid species 1995). Data from two species of sharks (Sphyrna return to their natal streams to spawn after spending tiburo and Scyliorhinus torazame) suggest that in years developing in oceans, homing in part based on cartilaginous fishes, separate populations of term- olfactory cues. In anadromous species, the develop- inal nerve neurons contain GnRH and FMRFamide- ing fish undergo behavioral, anatomical, and like compounds (Chiba, 2000; White and Meredith, physiological changes necessary to survive the mar- 1995). Additional FMRFamide immunoreactive ine environment as they migrate downstream. 366 Evolution of Vertebrate Olfactory Subsystems

During this period, olfactory sensitivity increases, cannot control (e.g., Sorensen and Scott, 1994). In allowing the animals to imprint on odorants that addition to sex pheromones, teleost fishes appear to will guide their homing behavior when they are produce alarm pheromones, which also raise inter- reproductively mature (reviewed in Nevitt and esting evolutionary issues. The phenomenon of Dittman, 2004). alarm cues was first described by von Frisch (1938, The complex nature of pheromonal signaling dur- 1941), who noted that a minnow, Phoxinus phox- ing courtship is understood better in goldfish inus, releases a chemical (‘Schreckstoff’) when (Carassius auratus) than in any other vertebrate. injured; the release of the chemical results in a beha- Vitellogenic females release 17b-estradiol, which vioral alarm response by conspecifics. This attracts males (Kobayashi et al., 2002). As the phenomenon has since been documented in many female approaches ovulation, it begins to release teleost species, particularly in Ostariophysi sex steroids. During the 12 h period leading up (reviewed in Døving et al., 2005). Nevertheless, to ovulation, the ratio of the three released steroids because its adaptive value is difficult to understand, changes, allowing males to assess the female’s repro- some have questioned the very existence of the phe- ductive state with great temporal precision (Scott nomenon (e.g., Magurran et al., 1996). The and Sorensen, 1994). The steroid that dominates problem is that production of alarm substances is the mixture early in this process inhibits male energetically costly (Wisenden, 2000), but of ques- responses to one of the steroids that dominates tionable benefit to the animal that produces them later, perhaps as a mechanism to prevent inap- while being consumed by a predator, even if kin are propriate responses to a signal that is present in nearby (Williams, 1992). A resolution to this appar- small quantities throughout the ovulatory period ent paradox may be that release of alarm substances (Stacey and Sorensen, 2002). At ovulation, the may attract secondary predators that will chase or female begins to release prostaglandins, attracting consume the initial predator, allowing the injured males, which are extremely sensitive to this mixture individual to escape (Mathis et al., 1995). This (Sorensen et al., 1988). Males also release a phero- hypothesis has received some empirical support mone that attracts females, which has been (Chivers et al., 1996a; but see Cashner, 2004). suggested to function in sex recognition during com- An interesting adaptation in some teleosts is the petition for mates (Sorensen et al., 2005a). use of the olfactory system for detecting changes in The mechanisms underlying olfactory transduc- external salinity; for example, Atlantic salmon tion of prostaglandins and sex steroids are (Salmo salar) have polyvalent cation-sensing recep- substantially similar in goldfish, and differ from tors in the olfactory epithelium (Nearing et al., those used to transduce amino acid odorants 2002). Other species can compensate for changes (Sorensen and Sato, 2005). In zebra fish (Danio in salinity. In sea bream (Sparus auratus), a euryha- rerio), pheromonal cues stimulate a different region line marine species, olfactory receptor neurons of the olfactory bulb than do other odorants increase their firing rate to compensate for reduc- (Friedrich and Korsching, 1998), and processing of tion in extracellular calcium levels (Hubbard et al., these two types of information may be segregated in 2000). Olfactory responses to odorants are similar goldfish as well (Hanson et al., 1998). In addition, in in freshwater and seawater in rainbow trout goldfish and its congener the Crucian carp (C. car- (Oncorhynchus mykiss), suggesting that the com- assius), the lateral olfactory tract carries information pensatory mechanism resides within the olfactory related to foraging behavior, whereas the medial epithelium (Shoji et al., 1996). tract is involved in pheromonal mediation of court- Ray-finned fishes usually have two pairs of nos- ship and spawning behavior (Dulka, 1993; Hamdani trils on the dorsal surface of the snout, with the et al., 2001; Kyle et al., 1987; Sorensen et al., 1991). anterior and posterior nares separated by a strip of Many teleost pheromones also function as hor- skin. Water flows into the anterior (incurrent) nares, mones, or are metabolites of hormones, and may be passes over the olfactory epithelium in the nasal sac, passively released through diffusion from the blood and then exits through the posterior (excurrent) in gills during respiration (e.g., Vermeirssen and nares. Water flow can be generated by swimming Scott, 1996). In contrast to the common conception movements. In teleosts with an accessory nasal sac, of pheromones as cues specifically produced as com- opercular movements during respiration alternately munication signals, such observations have led to compress and expand the accessory nasal sacs, the hypothesis that pheromonal communication in pumping water through the nasal sac (Døving fishes originally evolved as a mechanism by which et al., 1977; Nevitt, 1991). animals could assess the reproductive status of con- As in other fishes, the olfactory epithelium in ray- specifics through detection of cues that the releaser finned fishes is organized into a rosette of lamellae Evolution of Vertebrate Olfactory Subsystems 367 radiating outward from a central raphe. Both the terminate in the olfactory bulb, indicating that they shape and complexity of lamellar organization dif- are true olfactory receptor neurons (Hansen et al., fers among species (Yamamoto, 1982; Zeiske et al., 2003). Crypt cells are present in the olfactory 1992). The number of lamellae is also variable, epithelium of many, but not all, species of teleosts increasing during development to a species-typical that have been examined to date, and are present in level, such as the 168 lamellae present in adult a polypteriform fish, Polypterus senegalus,butnot undulated moray eels (Gymnothorax undulatus) or in shortnose gars, Lepisosteus plastostomus the 230 lamellae in adult barred snappers (Hansen and Finger, 2000). Like teleosts, sturgeons, (Hoplopagrus guenteri)(Fishelson, 1995; Pfeiffer, and paddlefish possess ciliated, microvillar, and 1964). Some groups, such as salmonids, have sec- crypt olfactory receptor neuron (Bakhtin, 1976; ondary lamellae that greatly increase the size of the Hansen and Finger, 2000; Pyatkina, 1976; Zeiske epithelial surface (Yamamoto and Ueda, 1977). et al., 2003). Overall, these data suggest that As illustrated in Figures 6a and 6c, the olfactory ciliated, microvillar, and crypt-type olfactory recep- epithelium of teleost fishes contains both ciliated tor neurons are broadly present in ray-finned fishes, and microvillar receptor cells (extensively reviewed but that the distribution of each cell type is some- in Eisthen, 1992). Within these two broad cate- what variable. gories, many subtypes can be distinguished; these In addition to these receptor cell types, both ‘rod’ subtypes differ morphologically and project to dis- and ‘rodlet’ cells have been proposed to function as tinct regions of the olfactory bulb (Morita and olfactory receptor neurons in teleosts. Instead, the Finger, 1998). former are probably unhealthy or degenerating cells Ray-finned fish are the only group of vertebrates (Muller and Marc, 1984; Zeiske and Hansen, 2005), in which the olfactory epithelium clearly contains and the latter may be migrating secretory cells, a type receptor neurons that do not fit into the preceding of white blood cell, or even parasites (Bielek, 2005). categories. Hansen et al. have described ‘crypt’ Both olfactory-type and vomeronasal-type odor- receptor neurons with a dendritic surface that ant receptor genes are expressed in the olfactory bears apical microvilli as well as a large concave epithelium of teleosts, although the size of the gene section that is filled with cilia (Figure 6b; Hansen families appears to be smaller than in mammals. In and Zeiske, 1998). Another unusual feature of crypt channel catfish (Ictalurus punctatus) fewer than 100 cells is that they express two types of G-proteins members of the olfactory receptor gene family have (Hansen et al.,2004). A recent study of the electro- been found (Ngai et al., 1993b, 1993a). In zebra physiological properties of crypt cells in the Pacific fish, the entire odorant receptor gene repertoire jack mackerel (Trachurus symmetricus) found that, appears to consist of 143 genes, with even smaller among the relatively small number of neurons exam- numbers found in pufferfish (44 genes in Takifugu ined, some responded to amino acid odorants, but rubripes and 42 in Tetraodon nigroviridis)(Alioto that none responded to polyamines or bile acids and Ngai, 2005). Members of the V2R gene family (Schmachtenberg, 2006). The axons of crypt cells have been sequenced from both pufferfish (Takifugu

Figure 6 Electron micrographs of the three classes of olfactory receptor neurons in ray-finned fishes. a, Transmission electron micrograph of dendrites of ciliated (C) and microvillar (M) olfactory receptor neurons in goldfish, Carassius auratus. b, Transmission electron micrograph of a crypt cell (Y) in zebra fish, Danio rerio. Crypt cells are characterized by cilia sunken within the cell (arrowheads), as well as microvilli on the surface. c, Scanning electron micrograph of the dendritic surface of ciliated (black arrowhead) and microvillar (white arrowhead) olfactory receptor neurons in a fathead minnow (Pimephales promela). Micrographs provided courtesy of Dr. Anne Hansen. 368 Evolution of Vertebrate Olfactory Subsystems rubripes) and goldfish (Cao et al., 1998; Naito et al., as a key criterion for homology, any attempt to 1998). A single member of the V1R gene family has compare the similarity of olfactory projection pat- been sequenced in zebra fish and found to be terns between teleosts and other vertebrates can expressed in the olfactory epithelium, and V1R- become circular; therefore, independent evidence, like genes with high sequence variability are also such as histochemical data, is necessary to corrobo- present in the genomes of several other teleost spe- rate hypotheses of homology. The central cies (Oryzias latipes, Danio malabaricus, Takifugu projections of the olfactory bulb have been exam- rubripes, and Tetraodon nigroviridis)(Pfister and ined in some detail in goldfish, Carassius auratus Rodriguez, 2005). In goldfish, the olfactory sensory (Levine and Dethier, 1985; Northcutt, 2006; von neurons that bear cilia express olfactory receptor Bartheld et al., 1984). As depicted in Figure 7a, the genes and Gaolf, and those with microvilli express medial olfactory tract projects bilaterally to the ven- V2R genes and Gao, although some are immunor- tral forebrain areas Vs, Vl, and Vv, which may be eactive for Gai-3 or Gaq instead (Hansen et al., equivalent to part of the septum (Northcutt and 2004). V2Rs are also expressed in crypt cells Braford, 1980); to Vd, which may be the equivalent (Hansen et al., 2004). In zebra fish, ciliated olfac- of part of the striatum (Northcutt and Braford, tory receptor neurons express the olfactory-type 1980); to Dm, which may be the equivalent of the receptor genes also express olfactory marker protein amygdala (Northcutt, 2006); and to the preoptic and the cyclic nucleotide-gated channel A2 subunit area, with some fibers terminating as far caudally as that is typical of mammalian olfactory receptor neu- the hypothalamus. The lateral olfactory tract of gold- rons; microvillar neurons express V2R-type fish projects mainly to dorsolateral pallial areas, as receptors and the TRPC2 channel that are typical well as projecting bilaterally to the hypothalamic of mammalian vomeronasal receptor neurons (Sato region and nucleus tuberis (Figure 7b). One of the et al., 2005). Taken together, these data suggest that targets of the lateral olfactory tract is Dl, which may a vomeronasal-type system is present in teleosts, but be the equivalent of the medial pallium/hippocampus that the receptor neurons are mixed together with (Northcutt, 2006). Topographically similar projec- olfactory receptor neurons in the epithelium (dis- tions have been described in a salmonid, cussed in Alioto and Ngai, 2005; Eisthen, 2004). Oncorhyncus mykiss (Folgueira et al., 2004). The The organization of the olfactory bulb is moder- presence of an extra-bulbar olfactory pathway is ately laminar. As in other fishes, distinct plexiform well-established in teleosts (Anado´n et al., 1995; layers are lacking, and the cell bodies are somewhat Bazer et al., 1987; Hofmann and Meyer, 1995; intermingled across layers (Nieuwenhuys, 1967). Szabo et al., 1991) as well as in Amia, sturgeons, Periglomerular cells may be lacking. Each mitral and polypteriform species (Hofmann and Meyer, cell has from one to five primary dendrites, and 1995; Huesa et al., 2000). The details of the targets each dendrite ends in one or more glomeruli differ among species, but the fibers generally project (Nieuwenhuys, 1967). Another class of output cell to ventral forebrain areas (Figure 7c). is called the ‘ruffed’ cell, due the presence of a ruff of Among nonteleost ray-finned fishes, the olfactory processes at the base of the axon (Kosaka and projections have been most carefully examined in Hama, 1979a, 1979b). This type of cell may be the bichir, Polypterus palmas (Braford and broadly present in teleosts, but has not been Northcutt, 1974; von Bartheld and Meyer, 1986). described in other classes of vertebrates (Alonso In Polypterus, a lateral olfactory tract projects lar- et al., 1987; Fuller and Byrd, 2005; Kosaka and gely to the pallial area P3 and may contain some Hama, 1980). In his study of the olfactory bulb of contralaterally projecting fibers, and a medial tract teleosts, Catois (1902) referred to an additional projects ipsilaterally to the pallial areas P1 and bilat- class of large output cells with an elongated, hori- erally to ventral telencephalic regions, with some zontally oriented cell body as ‘fusiform’ cells. fibers extending as far caudal as the hypothalamus. Morphologically similar cells are apparent in illus- Similar projections have been described in the stur- trations of the olfactory bulb of sturgeons geon Acipenser baeri (Huesa et al., 2000). Based on (Johnston, 1898), and may be present in other ray- its position and its massive input from the olfactory finned fishes as well. Granule cells and stellate cells bulb, P1 is generally considered to be the homologue appear to be present in both teleosts and sturgeons of the lateral pallium (Northcutt and Davis, 1983; (Johnston, 1898; Nieuwenhuys, 1967). von Bartheld and Meyer, 1986), and topological The forebrain of ray-finned fish is everted and considerations have led to the suggestion that P3 is contains many discrete cell groups, the homologies the homologue of the medial pallium / hippocampus of which are difficult to establish. Because connec- (Braford, 1995; Northcutt and Davis, 1983; von tivity is often used by comparative neuroanatomists Bartheld and Meyer, 1986). If these interpretations Evolution of Vertebrate Olfactory Subsystems 369

(a) (b) (c)

Figure 7 Schematic dorsal view of the forebrain in the goldfish, Carassius auratus. a, Projections of the medial olfactory tract. b, Projections of the lateral olfactory tract. c, Projections of the extrabulbar olfactory pathway. dl, lateral division of dorsal telencephalic area; dm, medial division of dorsal telencephalic area; dp, posterior division of dorsal telencephalic area; hyp, hypothalamus; nt, nucleus tuberis; ob, olfactory bulb; oe, olfactory epithelium; poa, preoptic area; vd, dorsal nucleus of the ventral telencephalic area; vl, lateral nucleus of the ventral telencephalic area; vs, supracommissural nucleus of the ventral telencephalic area; vv, ventral nucleus of the ventral telencephalic area. Based on Dulka (1993), Levine and Dethier (1985), Northcutt (2006), von Bartheld et al. (1984), and Polese (2004). are correct, then the lateral olfactory tract of the of the lateral olfactory tract of other vertebrates. bichir Polypterus palmas projects to the homologue Thus, the relationships between the medial and lat- of the medial pallium/hippocampus and the medial eral olfactory tracts in ray-finned fishes and other tract projects to the homologue of the lateral pal- vertebrates are unresolved, and may bear no one-to- lium, which is the reverse of the general vertebrate one correspondence with those in other vertebrates. pattern. Perhaps the apparent medial and lateral Unlike in cartilaginous fishes, the terminal nerve tracts have been reversed in Polypterus, in conjunc- in ray-finned fishes projects alongside the primary tion with eversion of the telencephalon. If this were olfactory fibers in the olfactory nerve, and is not the case, then we would expect that the tracts in visible externally. The ganglion cells of the terminal other ray-finned fish, such as goldfish, would be nerve develop from the nasal placode (Parhar, 2002; similarly reversed. Unfortunately, the data from Whitlock and Westerfield, 2000). In most teleosts goldfish are ambiguous: the lateral olfactory tract the terminal nerve ganglion consists of a cluster of projects to a lateral pallial area, as is typical of the cells located between the olfactory bulb and the lateral tract, but also to a region that may be the ventral telencephalon, with fibers that extend cen- equivalent of the septum, as is typical of the medial trally to the ventral forebrain and peripherally to olfactory tract in tetrapods. Further, the medial both the olfactory epithelium and retina olfactory tract of goldfish projects to ventral areas (Brookover and Jackson, 1911; Kim et al., 1995; that are suspected homologues of portions of the Oka et al., 1986; Parhar et al., 1996; Rossi et al., septum and medial pallium, which is typical of the 1972; Yamamoto et al., 1995). Because of this asso- medial tract of other vertebrates, but also to a ciation with both the olfactory system and retina, potential homologue of the amygdala, as it is typical the terminal nerve ganglion in teleosts is also 370 Evolution of Vertebrate Olfactory Subsystems frequently called the nucleus olfacto-retinalis (Mu¨ nz terminal nerve ganglion cells (Fujita et al., 1991), et al., 1981). The exact location and the morphol- and that lesions of the terminal nerve impair initia- ogy of the ganglion cells varies somewhat among tion of nest-building behavior, but do not abolish species (reviewed in Mu¨ nz and Claas, 1987). In reproduction, in dwarf gouramis (Yamamoto et al., some teleosts, such as the dwarf gourami (Colisa 1997). In addition, compounds present in the term- lalia), the fibers of the terminal nerve project to inal nerve alter the activity of retinal ganglion cells, many disparate regions of the brain (Oka, 1992; suggesting that the retinopetal branch of the term- Oka and Matsushima, 1993). inal nerve in teleost fishes is neuromodulatory The teleost terminal nerve contains at least one (Huang et al., 2005; Maaswinkel and Li, 2003; form of GnRH (Yamamoto et al., 1995), and many Walker and Stell, 1986). Interestingly, a tract-tra- of the cells and fibers of the terminal nerve are cing study with dwarf gouramis and tilapia GnRH-immunoreactive (e.g., Oka and Ichikawa, (Oreochromis niloticus) indicates that the terminal 1990; Schreibman et al., 1979). Some of these cells nerve receives input from olfactory areas in the fore- also display immunoreactivity to NPY or to brain, as well as from the nucleus tegmento- FMRFamide (Ekstro¨m et al., 1988; Mathieu et al., olfactorius, a midbrain region that receives input 2002; O¨ stholm et al., 1990; Pinelli et al., 2000; Rama from the reticular formation as well as areas Krishna and Subhedar, 1992; Walker and Stell, involved in visual and somatosensory processing 1986). The terminal nerve displays similar immunor- (Yamamoto and Ito, 2000). Taken together, the eactivity in polypteriform fishes (Polypterus palmas, results of these studies suggest that the teleost term- P. senegalus, and Calamoichthys calabaricus), stur- inal nerve functions to modulate activity in the geons (Acipenser ruthenus), and gars (Lepisosteus olfactory epithelium and retina, in part in response oculatus)(Chiba, 1997, 2005; Pinelli et al., 2000; to visual and olfactory input. Wright and Demski, 1996). Curiously, GnRH immunoreactivity has also been described in primary olfactory receptor neurons in the carp Cirrhinus mri- 2.17.2.5 Lobe-Finned Fishes: Lungfishes and Coelacanths gala; expression is seasonal, and occurs only in adult females (Biju et al., 2003, 2005). Perhaps these cells Three genera of lungfishes and one of coelacanths are evolutionarily derived from terminal nerve cells are alive today. Although they are dispersed around that did not migrate properly during development, the globe, each species lives in a relatively restricted and differentiated into olfactory receptor neurons area. Lungfishes live in freshwater in Africa under the influence of local cues. Nevertheless, (Protopterus), South America (Lepidosiren), and given that the neurons involved are not ganglion Australia (Neoceratodus), and the two known cells, they do not appear to be part of the terminal extant species coelacanths (Latimeria) live in the nerve system. deep ocean near Madagascar and Indonesia. In The function of the terminal nerve has been the addition to their wide geographical separation, the subject of much study in teleosts, particularly in the living lobe-finned fishes share only distant common dwarf gourami, Colisa lalia. Electrophysiological ancestors, and each group possesses unique features experiments demonstrate that the majority of the that are poorly understood. Protopterus, for exam- ganglion cells fire spontaneous action potentials ple, has a reduced olfactory system that is thought to brought about by the interaction of a tetrodotoxin- be the result of adaptation to drought and starvation resistant, persistent sodium current that depolarizes (Derivot, 1984). the cell and a persistent potassium current that repo- In lungfish, the olfactory organ is located ven- larizes the cell (Abe and Oka, 1999; Oka, 1992, trally. The incurrent nostril opens on the dorsal 1996). The firing frequency of the cells is modulated surface of the snout, but, unlike other fishes the by the same form of GnRH that is present in the excurrent nostril opens inside the mouth (Huxley, nerve, causing an initial decrease in firing rate, fol- 1876). This internal naris functionally connects the lowed by a later increase (Abe and Oka, 2000, nasal cavity and respiratory system, causing water 2002). This modulation of firing rate by GnRH to flow across the surface of the olfactory epithelium may function to synchronize firing (Abe and Oka, during breathing (Huxley, 1876). The degree to 2000). Studies such as these were the first to suggest which the gills and lungs are involved in respiration that the terminal nerve may function as a neuromo- varies among groups. dulatory system, rather than as a sensory system, as As in other fishes, the nasal cavity of lungfishes originally thought. Additional evidence for this contains a series of lamellae, on the surface of which hypothesis comes from studies showing that expo- lies the olfactory epithelium (Derivot, 1984; sure to odorants does not alter the firing rate of Pfeiffer, 1969; Theisen, 1972). The morphology of Evolution of Vertebrate Olfactory Subsystems 371 olfactory receptor neurons varies among taxa. The medial olfactory tract projects to the septum. A African lungfish (Protopterus annectans) has both contralateral projection is present, but the termina- ciliated and microvillar olfactory receptor neurons, tions of decussating fibers has not been determined. but the Australian lungfish (Neoceratodus forsteri) The presence of primary olfactory fibers that project lacks ciliated olfactory receptor neurons (Derivot bilaterally to the di- and mesencephalon has been et al., 1979; Theisen, 1972). The microvillar olfac- detected in Protopterus using horseradish peroxi- tory receptor neurons in these animals are unusual: dase injections into the nasal cavity (von Bartheld the cells lack centrioles, and the microvilli contain and Meyer, 1988), and a similar projection has been microtubules, which have not been described in the described in Neoceratodus (Schober et al., 1994). microvilli on olfactory receptor cells in other verte- Thus, lungfishes appear to possess an extrabulbar brates (Theisen, 1972). Both species lack the crypt- olfactory pathway. type olfactory receptor neurons characteristic of Initial studies of the anatomy of the terminal ray-finned fishes (Hansen and Finger, 2000). nerve in lungfish (Protopterus annectans; Pinkus, The olfactory bulbs are located adjacent to the 1894 and Neoceratodus forsteri; Sewertzoff, 1902) telencephalon in Protoperus and Lepidosiren, but described a discrete ganglion close to the olfactory those in Neoceratodus are connected to the telence- sac with a nerve that runs independent of the olfac- phalon by short, hollow peduncles (Holmgren and tory nerve, entering the brain at the level of the van der Horst, 1925). The structure of the olfactory preoptic area. Because of this topology, it was bulb in Protoperus has been described by Rudebeck named ‘nervus praeopticus’ (Sewertzoff, 1902). (1945). Clear laminae are present, including plexi- The terminal nerve has now been examined in all form layers, although the internal plexiform layer is three genera of lungfish, and in each it has two not as robust as in Neoceratodus (Holmgren and roots: an anterior root that runs within the olfactory van der Horst, 1925). Periglomerular cells are scat- nerve, consisting of a fascicle of fibers and bipolar tered among the glomeruli, and their morphology is cells, and a posterior root that corresponds to unlike those in other vertebrates: each cell has a Sewertzoff’s ‘nervus praeopticus’ (Holmgren and single dendrite that arborizes in a glomerulus, and van der Horst, 1925; Rudebeck, 1945). In the axons of most periglomerular cells form a dis- Neoceratodus, the two roots are joined at the level tinct tract that passes over the dorsal surface of the of the ganglion (Fiorentino et al., 2002; Holmgren olfactory bulb and extends to the anterior pallium and van der Horst, 1925; Rudebeck, 1945). (Rudebeck, 1945). The mitral cells have primary Experimental embryological work with larval dendrites that arborize in the glomerular layer as Neoceratodus demonstrates that the terminal nerve well as secondary dendrites that extend through ganglion and both roots originate within the olfac- the external plexiform layer. The granule cells have tory placode (Fiorentino et al.,2002). In spiny dendrites and unmyelinated axons. Stellate Neoceratodus and Protopterus, only the posterior cells may not be present. root displays GnRH immunoreactivity (Schober Although Protopterus has been suggested to pos- et al., 1994), but in Neoceratodus,bothrootsdisplay sess a vomeronasal nerve and accessory olfactory FMRFamide-like immunoreactivity (Fiorentino et al., bulb (Schnitzlein and Crosby, 1967), more recent 2002). Perhaps the anterior root is homologous with studies have demonstrated that this is not the case the bundle of FMRFamide-immunoreactive fibers (Derivot, 1984; Reiner and Northcutt, 1987). Thus, that run within the olfactory nerve of some sharks, a discrete vomeronasal system is lacking in lobe- described by Wu et al. (1992). finned fishes. Two species of coelacanths are known, and The central projections of the olfactory system in because of the rarity of material the olfactory system lungfishes have not been examined using modern has been examined only superficially in Latimeria methods, but were described by Holmgren and van chalumnae. This species has an incurrent nostril on der Horst (1925) and Rudebeck (1945) based on the dorsal side of the snout, and a valvular excurrent normal material (reviewed in Nieuwenhuys, 1967). nostril that opens just in front of the eye. The olfac- A central olfactory tract consisting of axons from tory epithelium is radially organized around a both mitral and granule cells passes through the central axis where the three dorsal and two ventral telencephalic hemisphere in the pallium and subpal- lamellar lobes converge (reviewed in Zeiske et al., lium, forming internal and external fiber layers. The 1992). Although the fine structure of the olfactory internal fiber layer is confined to the pallium epithelium has not been examined, odorant receptor whereas the external fiber layer extends to the stria- genes have been sequenced from coelacanths tum and the lateral parts of the olfactory tubercle, (Freitag et al., 1998). Both classes of receptor continuing posteriorly to the stria medullaris. A genes described by Freitag et al. (1995, 1998), 372 Evolution of Vertebrate Olfactory Subsystems purportedly for odorant detection in air and in are relatively poorly understood. Amphibians lack a water, are present. diaphragm, and as they draw air into and out of the A bundle of receptor cell axons emerges from lungs by expanding and contracting the buccophar- each lobe, and the bundles merge to become a yngeal cavity, air passes across the sensory short olfactory nerve that projects into the nearby epithelium in the nasal cavity (Jørgensen, 2000). olfactory bulb (Northcutt and Bemis, 1993). The Salamanders have been observed to do the same olfactory bulb is located adjacent to the nasal cavity, underwater, allowing for olfactory sampling of the and is connected to the telencephalon by a long aqueous environment (Jørgensen, 2000). olfactory peduncle (Nieuwenhuys, 1965). The orga- The vomeronasal system is generally present nization of the olfactory bulb resembles that of throughout life in amphibians. Among salamanders, other fishes, but clear plexiform layers are lacking the system is present in both aquatic and terrestrial (Nieuwenhuys, 1965). The olfactory tracts run species, including those that never metamorphose. through the peduncle. Some tracts terminate within The vomeronasal system has been described in cryp- the olfactory bulb (corpus rostrale), whereas the tobranchids, sirenids, ambystomatids, salamandrids, others form a central olfactory tract that becomes amphiumids, and plethodontids, but is absent in thinner as it proceeds caudally through the pallium. members of the proteid family (Anton, 1908, 1911; The termination sites of these fibers are unclear Eisthen, 2000; Eisthen et al., 1994; Saito et al.,2003; (Nieuwenhuys, 1965). Neither an extrabulbar olfac- Schmidt and Roth, 1990; Seydel, 1895; Stuelpnagel tory pathway nor a terminal nerve has been and Reiss, 2005). Given the phylogenetic relation- identified in Latimeria (Northcutt and Bemis, ships among salamander families (Frost et al., 1993). 2006), either a vomeronasal-like system arose independently at least four times in salamanders, or it was 2.17.2.6 Tetrapods: Amphibians, Reptiles, and present in the last common ancestor of extant sala- Mammals manders and lost in proteids. Clearly, the latter is the most parsimonious hypothesis. The vomeronasal sys- The tetrapods include all extant amphibians, rep- tem is also present throughout life in both tiles (including birds), and mammals. Early metamorphosing and direct-developing frogs, includ- tetrapods were aquatic, and the last common ing those that are fully aquatic as adults (Cooper, ancestor of tetrapods is now generally accepted 1943; Hansen et al., 1998; Jermakowicz et al., to have been fully aquatic (Lebedev and Coates, 2004; Nezlin and Schild, 2000; Reiss and Burd, 1995; Panchen, 1991). Thus, many features in 1997b, 1997a; Scalia, 1976; Zwilling, 1940), as amphibians and amniotes (reptiles and mammals) well as in caecilians (Billo and Wake, 1987; Schmidt that represent adaptations to terrestrial life arose and Wake, 1990). independently. Although anurans are often considered to rely In early tetrapods, the external nostril shifted almost entirely on visual and acoustic cues, olfac- position to lie low on the snout, and became tion plays an important role in the lives of some enlarged (Clack, 2002). In addition, olfaction species (Waldman and Bishop, 2004). For example, became coupled to respiration, and remains so in tadpoles reduce their activity or seek refuge when many extant tetrapods (Clack, 2002). The olfactory exposed to odorants from sympatric predators, system of tetrapods is dramatically reorganized injured conspecifics, or predators that have con- compared with that of other vertebrates, as the sumed conspecifics (e.g., Laurila et al., 1997; vomeronasal system now forms a discrete sensory Marquis et al., 2004). Sustained exposure to such system. The olfactory system of terrestrial tetrapods cues can also lead to more dramatic changes in must contain features that facilitate functioning to morphology and life history (Chivers et al., 2001). detect odorants in air, instead of in water, yet few Juvenile toads (Bufo cognatus and B. microscaphus) concrete examples have been identified. For exam- avoid chemical cues from predatory garter snakes, ple, the vomeronasal system was once thought to Thamnophis (Flowers and Graves, 1997). Wood have arisen as an adaptation to terrestrial life frog tadpoles (Rana sylvatica) prefer to school (Bertmar, 1981), but this idea has since been dis- with kin (Waldman, 1982, 1984), and kin recogni- carded (Eisthen, 1992, 1997). tion is mediated by olfactory cues (Waldman, 1985). Some tree frogs that lay eggs in small pools 2.17.2.6.1 Amphibians Amphibians comprise of water in plants provide additional trophic eggs to three distinct groups: anurans, or frogs and toads; ensure adequate food supplies for tadpoles. In one urodeles, or salamanders, including newts; and apo- such species, Chirixalus eiffingeri, tadpoles become dans or caecilians, legless neotropical animals that highly active when exposed to water conditioned by Evolution of Vertebrate Olfactory Subsystems 373 adult females (Kam and Yang, 2002), suggesting attracted to odorants from females engaged in court- that tadpoles associate odorants from females with ship behavior than females that are not, even when food. male odorants are present in both situations (Park and Olfactory cues are also used by adult frogs, parti- Propper, 2001; Park et al.,2005). Nevertheless, not all cularly in relation to reproductive behavior. pheromone effects in salamanders are mediated by Olfaction is involved in orientation to breeding nasal chemosensory systems: in many plethodontid ponds (Ishii et al., 1995), and some frogs prefer salamanders, males use their teeth to inject phero- their own odorants to those from conspecifics mones through the skin of females, thereby (Waldman and Bishop, 2004). Male magnificent increasing their receptivity (Houck and Arnold, 2003). tree frogs, Litoria splendida, produce a pheromone The behavior of caecilians has not been examined that attracts females (Wabnitz et al., 1999). Males in detail. Nevertheless, Himstedt and Simon (1995) of some other Litoria species have a rostrally direc- showed that plugging the nostrils in Ichthyophis koh- ted spike on the tip of the snout, and the surface taoensis disrupts foraging, suggesting that olfaction epithelium is rich in secretory glands (Menzies, plays an important role in this behavior. Chemical 1993). Although the function of this spike is cues have been shown to play a role in both aggrega- unknown, its sexually dimorphic distribution and tion and individual recognition in Typhlonectes glandular surface suggest that it is involved in che- natans (Warbeck and Parzefall, 2001). mical communication related to reproduction The relative contributions of the olfactory and (Menzies, 1993). Adult Xenopus can also use odor- vomeronasal systems to amphibian behavior are ant cues in air to find food (Shinn and Dole, 1978). unclear. The vomeronasal organ plays a role in pre- The role of olfaction in salamander behavior has dation in the red-backed salamander, Plethodon been examined much more extensively than that in cinereus (Placyk and Graves, 2002). In both sala- frogs. Like tadpoles, salamanders respond to odor- manders and caecilians, the relative size of the ants from injured conspecifics (Chivers et al., vomeronasal organ is larger in aquatic or semi- 1996b, 1997). Larval tiger salamanders aquatic species than in terrestrial species (Dawley, (Ambystoma tigrinum) are facultative cannibals, 1998; Schmidt and Wake, 1990), suggesting that and larvae are more likely to become cannibals in aquatic amphibians may rely more on vomeronasal mixed-sibship groups than when surrounded by sib- input than do terrestrial species. The vomeronasal lings (Pfennig and Collins, 1993). Cannibals prefer organ tends to be larger in male than female pletho- to consume conspecifics that are not kin, but this dontid salamanders, and organ size increases during discrimination disappears when the nostrils are the breeding season, suggesting a role in reproduc- plugged, implicating olfaction in the kin recognition tive behavior (Dawley, 1998; Dawley et al., 2000). process (Pfennig et al., 1994). Indeed, in female salamanders (Cynops pyrrhoga- Chemical cues are involved in social behavior in ster and Plethodon jordani), male-produced salamanders, and odorants from conspecifics are attraction pheromones elicit physiological responses involved in aggregation (Secondi et al.,2005), indivi- from the vomeronasal organ, but not from the olfac- dual recognition (Jaeger, 1981; Ovaska, 1988), and tory epithelium (Kikuyama and Toyoda, 1999; marking territories (Chivers et al., 1996b; Jaeger, Wirsig-Wiechmann et al., 2002b). Nevertheless, 1986; Ovaska and Davis, 1992; Simons et al., EOG recordings demonstrate that both the olfac- 1994). Salamanders also use chemical cues to discri- tory and vomeronasal epithelia respond to minate the sex and reproductive condition of pheromones in female red-bellied newts (C. pyrrho- conspecifics (Marco et al.,1998;Parket al.,2004; gaster) and male red-spotted newts (N. viridescens), Verrell, 1985), and the use of pheromones in court- and that both epithelia respond more strongly to ship behavior appears to be widespread. For example, odorants from the opposite sex than the same sex in two species of fire-belly newts (Cynops), males in male and female axolotls, Ambystoma mexica- produce a species-specific peptide that attracts num (Park et al., 2004; Park and Propper, 2002; females (Kikuyama et al., 1995; Yamamoto et al., Toyoda et al., 1999). 2000). In plethodontid salamanders, (Desmognathus The morphology of the nasal cavity varies con- ochrophaeus and Plethodon jordani), male phero- siderably among amphibian groups, and often mones can increase female receptivity, and in changes during metamorphosis. Many of the anato- red-spotted newts (Notophthalmus viridescens) mical differences associated with life in water and unreceptive females can become receptive when on land have recently been reviewed by Reiss and exposed to chemical cues from a male (Houck and Eisthen (2006). Perhaps the most dramatic changes Reagan, 1990; Rogoff, 1927; Rollmann et al.,1999). occur during metamorphosis in frogs. For example, Interestingly, in red-spotted newts, males are less larval African clawed frogs, Xenopus laevis, possess 374 Evolution of Vertebrate Olfactory Subsystems only one olfactory chamber, which is used for olfac- 1987; Millery et al., 2005). As in mammals, odorant tion in water. This chamber contains both ciliated binding proteins in frogs appear to be lipocalins. and microvillar receptor neurons (Hansen et al., Interestingly, in both species of Xenopus, the protein 1998). During metamorphosis, this chamber is is expressed in principal cavity but not middle cavity transformed into the principal cavity, which func- (Millery et al., 2005). This observation lends support tions for detecting odorants in air, and a new middle to the hypothesis that the function of odorant binding chamber develops (reviewed in Hansen et al., 1998; proteins is related to physical constraints imposed by Reiss and Eisthen, 2006). In other frogs, this middle the problem of detecting odorants in air; for example, cavity is nonsensory, but in Xenopus the middle odorant binding proteins may function to transport cavity is lined with sensory epithelium and functions hydrophobic molecules through aqueous lymph or in olfaction underwater (Altner, 1962; Paterson, mucus to the odorant receptors (Bignetti et al., 1951). Adult Xenopus are almost entirely aquatic, 1987; Vogt, 1987). and sample odorants in both at the surface and in The nasal sac of salamanders is essentially a sim- water. A muscular valve shunts the medium into one ple oval-shaped tube that is almost completely lined chamber or the other (Altner, 1962). In all frogs with sensory epithelium, which is one of the reasons studied to date, including Xenopus, the adult prin- why some electrophysiologists favor salamanders as cipal cavity contains only ciliated receptor neurons, model animals for olfactory research (e.g., Kauer, but the middle cavity of Xenopus contains both 2002). The olfactory epithelium of most adult sala- ciliated and microvillar receptor neurons (Bloom, manders, even those that are fully terrestrial, 1954; Hansen et al., 1998; Mair et al., 1982; appears to contains both ciliated and microvillar Menco, 1980; Oikawa et al., 1998; Reese, 1965; receptor cells (Breipohl et al., 1982; Eisthen, 2000; Taniguchi et al., 1996). Because the principal cavity Eisthen et al., 1994; Farbman and Gesteland, 1974; that detects waterborne odorants in larvae is trans- Jones et al., 1994). In contrast, four types of olfac- formed into the principal cavity that detect airborne tory receptor neurons have been described in odorants in adult, a clear correlation emerges: in Dicamptodon tenebrosus: those with cilia, with Xenopus, ciliated receptor neurons are used for long microvilli, with unusual short microvilli, and olfaction in air, and both ciliated and microvillar those with both cilia and extremely short (<0.5 mm) receptor neurons are used for olfaction in water microvilli (Stuelpnagel and Reiss, 2005). This find- (Hansen et al., 1998). ing suggests that other salamanders may possess In addition to these morphological differences, additional types of neurons that have not been dis- the sensory epithelia in the principal and middle tinguished. Thirty-five putative odorant receptor cavities of Xenopus differ at the molecular level. genes have been sequenced from tiger salamanders Freitag et al. (1995, 1998) divide the odorant recep- (A. tigrinum), and all appear to be class II genes tor genes in frogs into two classes: class I genes, with (Marchand et al., 2004). Because terrestrial tiger a large loop in the third extracellular domain and salamanders were used, this result appears to sup- sequences similar to those found in teleosts, and port the Freitag et al. (1998) hypothesis that class II genes, similar to those found in mammals. odorant receptors in the class II gene family function In Xenopus, class II genes are expressed in principal to detect odorants in air. Nevertheless, mudpuppies cavity and are proposed to function for detecting (Necturus maculosus), which are members of the odorants in air, and class II are expressed in middle fully aquatic proteid family, possess both class I cavity and are proposed to function for detecting and class II odorant receptor genes (Zhou et al., odorants in water (Freitag et al., 1995; Mezler 1996), casting doubt on the distinction. Olfactory et al., 2001). The principal cavity also contains a transduction in mudpuppies also involves a novel novel form of Gas that is more closely related to the cyclic nucleotide-dependent chloride current mammalian Gaolf than to other forms of Gas in (Delay et al., 1997). The existence of this current frogs, and that induces cAMP formation. In con- in mudpuppies may represent an environmental trast, the epithelium in the middle cavity contains adaptation to life in freshwater, in which the ionic Gao1, which stimulates formation of IP3 (Mezler concentrations of the olfactory mucus may be diffi- et al., 2001). Two different homologues of mamma- cult to regulate. If so, the external calcium required lian olfactory marker protein have been found in to gate the chloride channels widely involved in Xenopus, and they are differentially expressed in olfactory transduction may not reliably be present, the two olfactory cavities (Ro¨ ssler et al., 1998). leading to the use of a different gating mechanism in Odorant binding proteins, suggested to be a mam- these animals (Delay et al., 1997). malian adaptation, have been found in Rana pipiens, Caecilians have paired nares that open into the Xenopus laevis, and Xenopus tropicalis (Lee et al., nasal cavity, which contains the olfactory Evolution of Vertebrate Olfactory Subsystems 375 epithelium. Adult caecilians also have short tenta- enters from the rostrolateral pole, and the layers cles between the naris and eye that are both progress in wide bands from rostral to caudal (e.g., chemosensory and somatosensory (Badenhorst, Herrick, 1948). In frogs, the layers are not comple- 1978). Caecilians are blind, and the tentacles are tely concentric (Scalia et al., 1991b). derived from modified eye structures (Billo and In addition, in frogs the two olfactory bulbs are Wake, 1987). The lumen of the tentacle carries fused across the midline (Hoffman, 1963), a condi- secretions from the Harderian gland and communi- tion also observed in some teleost fishes and bird cates with the vomeronasal organ (Badenhorst, species (Nieuwenhuys, 1967). Morphologically, the 1978; Schmidt and Wake, 1990), an arrangement two bulbs in frogs appear to be organized almost as that allows the animal to detect chemicals with the a single unit. For example, the axons of some olfac- vomeronasal organ even while burrowing or swim- tory receptor neurons cross the midline to project to ming, when the external nares are closed (Prabha the contralateral bulb (Hoffman, 1963). The lami- et al., 2000). The nasal cavity of Typhlonectes com- nae within the bulb are continuous across the pressicaudum contains two morphologically midline (Hoffman, 1963; Scalia et al., 1991b), and distinct types of olfactory epithelium. The dorsocau- the processes of both mitral and granule cells cross dal portion of the nasal cavity contains an the midline to interact with elements in the contral- epithelium with only ciliated olfactory receptors ateral olfactory bulb (Herrick, 1921; Ramo´n y neurons, whereas the anterior ventral epithelium Cajal, 1922; Scalia et al., 1991b). Field potential contains both ciliated and microvillar receptor neu- recordings demonstrate that the responses to unilat- rons (Saint Girons and Zylberberg, 1992). Given eral stimulation of the olfactory nerve are similar in their relative locations, these observations suggest both olfactory bulbs, suggesting that signal trans- that, as in Xenopus, caecilians use morphologically mission is fully bilateral (Andriason and Leveteau, different receptor neurons for detecting odorants in 1989). However, recordings from mitral cells indi- water and air (Reiss and Eisthen, 2006). cate that unilateral nerve stimulation leads to an As in other tetrapods, the vomeronasal receptor inhibition of evoked responses on the contralateral neurons in frogs, salamanders, and caecilians termi- side (Leveteau et al., 1993). The authors of this nate in microvilli (Eisthen, 2000; Eisthen et al., study suggest that the contralateral inhibition is a 1994; Franceschini et al., 1991; Kolnberger, 1971; mechanism for enhancing contrast between the two Kolnberger and Altner, 1971; Oikawa et al., 1998; sides, facilitating odorant localization. If so, perhaps Saint Girons and Zylberberg, 1992). In addition to the joint olfactory bulbs of frogs allow them to the receptor neurons, sustentacular cells, and basal better localize odorants than salamanders, which cells that are generally present in the vertebrate have separate olfactory bulbs. olfactory and vomeronasal epithelia, the vomerona- In both salamanders and frogs, the axons of olfac- sal epithelium of frogs and salamanders contains tory receptor neurons have been shown to branch large ciliated supporting cells that may function to upon entering the outer fiber layer of the bulb and move fluid across the surface of the epithelium innervate glomeruli that are widely spaced (Herrick, (Eisthen, 1992). The vomeronasal epithelium in 1924, 1931; Nezlin and Schild, 2005). Periglomerular Xenopus has been shown to express members of cells are present, although they do not surround the V2R family of vomeronasal receptor genes as and isolate individual glomeruli to the same extent well as Gao, the G-protein that is co-expressed with observed in mammals (Herrick, 1924, 1931; Nezlin V2Rs in mammals (Hagino-Yamagishi et al., 2004). et al., 2003; Nezlin and Schild, 2000; Ramo´n y Although a portion of the principal cavity contain- Cajal, 1890). The primary dendrites of the mitral ing olfactory epithelium was reported to express cells arborize in multiple glomeruli (Herrick, 1948; V2R genes (Hagino-Yamagishi et al., 2004), the Hoffman, 1963). Mitral cells also bear secondary authors appear to have misidentified the posterior dendrites that do not terminate in glomeruli and portion of the vomeronasal organ, which is longer in appear to make synapses with processes of granule Xenopus than in other frogs (Reiss and Eisthen, cells; these neurites are more prominent in frogs 2006). than in salamanders (Scalia et al., 1991b). Herrick The olfactory bulbs of amphibians display a mod- (1924) describes ‘atypical mitral cells’ in the exter- erate degree of lamination, and include both nal plexiform layer in tiger salamanders that could internal and external plexiform layers in all groups be homologous with the tufted cells found in mam- (Herrick, 1948; Hoffman, 1963; Nieuwenhuys, mals, and Scalia et al. (1991b) described candidate 1967). Unlike in any other group of vertebrates, tufted cells in a similar location in northern leopard the olfactory bulb of salamanders does not consist frogs (Rana pipiens). Axonless granule cells are of concentric layers; rather, the olfactory nerve present. The dendrites of these cells are spiny in 376 Evolution of Vertebrate Olfactory Subsystems frogs and tiger salamanders, but smooth in mud- lateral pallium, dorsal striatum, lateral amygdala, puppies (Herrick, 1924, 1931; Hoffman, 1963; and a region interpreted as either the dorsal pallium Scalia et al., 1991b). Amphibians appear to lack or the dorsal portion of the lateral pallium stellate cells. The structure of the accessory olfac- (Figure 8b; Moreno et al., 2005; Northcutt and tory bulb is similar to that of the olfactory bulb, Royce, 1975; Scalia et al., 1991a). Fibers of the but plexiform layers are absent and the boundaries medial and lateral olfactory tracts project in combi- of the layers are less distinct (Herrick, 1924, 1931; nation to the contralateral amygdala and lateral Nieuwenhuys, 1967). In salamanders, some mitral pallium. The accessory olfactory tract projects bilat- cells have dendrites that project to glomeruli in erally to the medial amygdala (Figure 8d), the both the olfactory and accessory olfactory bulbs rostral portion of which also receives olfactory (Herrick, 1924). input (Kemali and Guglielmotti, 1987; Moreno Central olfactory projections have been examined and Gonzalez, 2003; Moreno et al., 2005; Scalia in several anurans, including Xenopus and ranid et al., 1991a). The extrabulbar olfactory pathway and hylid frogs, as well as in tiger salamanders, of Xenopus has been examined in detail by Ambystoma tigrinum (Kemali and Guglielmotti, Hofmann and Meyer, who have found that fibers 1987; Kokoros and Northcutt, 1977; Northcutt originating in the olfactory epithelium bypass the and Royce, 1975; Roden et al., 2005; Scalia, 1972; olfactory bulb and terminate in the ipsilateral pre- Scalia et al., 1991a). As shown in Figure 8a, the optic area and bilaterally in the hypothalamus medial olfactory tract projects to the postolfactory (Figure 8c; Hofmann and Meyer, 1991a, 1991b; eminence and medial pallium as well as to the lateral 1992). The pathway that Schmidt and colleagues and medial septal nuclei (Northcutt and Royce, (Schmidt et al., 1988; Schmidt and Wake, 1990) 1975; Roden et al., 2005; Scalia et al., 1991a). The described as the terminal nerve in salamanders and lateral olfactory tract projects ipsilaterally to the caecilians is more likely the extrabulbar olfactory

(a) (c) (d) (b)

Figure 8 Schematic dorsal view of the forebrain in ranid frogs. a, Projections of the medial olfactory tract. b, Projections of the lateral olfactory tract. c, Projections of the extrabulbar olfactory pathway. d, Projections of the accessory olfactory tract. amg, amygdala; aob, accessory olfactory bulb; dp, dorsal pallium; d st, dorsal striatum; hyp, hypothalamus; lat sep, lateral septum; lp, lateral pallium; med sep, medium septum; mp, medial pallium; ob, olfactory bulb; oe, olfactory epithelium; pe, postolfactory eminence; poa, preoptic area; vno, vomeronasal organ. Based on Kemali and Guglielmotti (1987), Northcutt and Royce (1975), Scalia (1972), and Scalia et al. (1991a). Evolution of Vertebrate Olfactory Subsystems 377 pathway, as it was labeled by injection of tracer into internal state, perhaps to maximize sensitivity to the nasal cavity but not vomeronasal organ, and odorants most relevant to the animal’s condition. ganglion cells were not found. If this interpretation is correct, the extrabulbar pathway may be more 2.17.2.6.2 Reptiles The class Reptilia includes extensive in both salamanders and caecilians than rhynchocephalians, or tuatara; squamates, a group in frogs (Hofmann and Meyer, 1989). that consists of lizards, snakes, and amphisbaenians; The terminal nerve of both frogs and salamanders crocodilians; and turtles. Because birds (Aves) form has been examined in detail. The cells and fibers of the sister group to extant crocodilians, we include the nerve are immunoreactive for a variety of com- birds in this group. pounds, including GnRH (D’Aniello et al., 1994b, 1995; Muske and Moore, 1988; Northcutt and 2.17.2.6.2.(i) Tuatara The rhynchocephalians, or Muske, 1994; Sherwood et al., 1986), NPY tuatara, constitute a separate order of reptiles. The (D’Aniello et al., 1996a; La´za´r et al., 1993; two living species of tuatara, Sphenodon punctatus Mousley et al., 2006; Tuinhof et al., 1994), and and S. guntheri, are found only on small islands off FMRFamide (D’Aniello et al., 1996b; Muske and the coast of New Zealand. Because of their phylo- Moore, 1988; Northcutt and Muske, 1994). In tiger genetic importance as a sister group to squamate salamanders (Ambystoma tigrinum), the terminal reptiles, their chemosensory systems have been stu- nerve also displays acetylcholinesterase activity, died to a limited extent. indicating that acetylcholine is present (Wirsig and Unlike snakes and some lizards, tuatara do not Getchell, 1986). In frogs, the terminal nerve may have a forked tongue (Schwenk, 1986), and they do send an additional projection to the retina not tongue-flick (Walls, 1981). The role of the olfac- (Uchiyama et al., 1988; Wirsig-Wiechmann and tory and vomeronasal systems in tuatara feeding has Basinger, 1988). been reviewed by Schwenk (2000). Much of the In salamanders, terminal nerve-derived peptides available evidence is anecdotal (e.g., Walls, 1981), alter both the odorant sensitivity and excitability of but suggests that visual cues predominate and that olfactory receptor neurons. The olfactory epithelium chemoreception plays a secondary role in foraging appears to contain GnRH receptors (Wirsig- and food choice (Schwenk, 2000). Tuatara will bite Wiechmann and Jennes, 1993), and GnRH modulates cotton that has been swabbed on prey, demonstrat- both odorant responses and the voltage-dependent, ing that the animals will respond to chemical cues tetrodotoxin-sensitive sodium current in olfactory related to feeding (Cooper et al., 2001). The role of receptor neurons (Eisthen et al., 2000; Park and chemical signals in social and reproductive behavior Eisthen, 2003). Interestingly, olfactory receptor neu- is unknown. rons are more responsive to GnRH during the The nasal cavity of Sphenodon is fairly simple in breeding season, suggesting that GnRH may play organization, although some conchae are present. A a role in regulating responses to odorants in a way long choana connects the ventral nasal cavity to the that promotes reproduction (Eisthen et al.,2000). oral cavity. Olfactory epithelium lines the posterior Other studies show that estrogen affects GnRH portion of the dorsal half of the nasal cavity, and the concentrations in the terminal nerve in Xenopus vomeronasal sensory epithelium is confined to the (Wirsig-Wiechmann and Lee, 1999)andthatcon- dorsal portion of the small vomeronasal organ, centrations are higher in courted female which lies along the septum. The vomeronasal salamanders than in uncourted conspecifics organ opens into the ventromedial portion of the (Propper and Moore, 1991), lending further support nasal cavity, anterior to the choana. Unlike that of to the idea that the GnRH-containing fibers of the snakes and lizards, the vomeronasal organ has no terminal nerve play a role in reproductive behavior. direct connection to the oral cavity. The ultrastruc- In addition, FMRFamide modulates both odorant ture of the olfactory and vomeronasal epithelia has responses and the sodium current in olfactory not been described. The axons of the vomeronasal receptor neurons (Park et al.,2003). Recently, we receptor neurons join the olfactory nerve to project have also shown that the terminal nerve in axolotls to a small accessory olfactory bulb dorsomedial to (Ambystoma mexicanum) is NPY-immunoreactive, the much larger main olfactory bulb (Hoppe, 1934, and that application of synthetic axolotl NPY in Parsons, 1959a; Parsons, 1967). Cairney (1926) modulates both odorant responses and the sodium described the olfactory projections in normal mate- current, but only in hungry animals (Mousley et al., rial without distinguishing among the tracts, 2006). Taken together, these data suggest that deducing that the olfactory bulbs project to the terminal nerve modulates responsivity in the medial pallium, septum, olfactory tubercle, olfac- olfactory epithelium in concert with changes in tostriatum, lateral pallium (piriform cortex), 378 Evolution of Vertebrate Olfactory Subsystems anterior nucleus of the amygdala, and a region he Chemical cues can facilitate predator avoidance. called the ‘hypopallium posterius’. The nucleus The Texas banded gecko, Coleonyx brevis, displays sphericus of squamates has a roughly similar posi- defensive behavior when presented with odorants tion and appearance, and receives a large projection from a predatory snake; changes in the rate of buc- from the accessory olfactory bulb. The nucleus cal pulsing in response to these cues suggest that the sphericus has been thought to be present only in discrimination is based on input to the olfactory squamates (Northcutt, 1978), but without studying system (Dial and Schwenk, 1996). The ability to projections in tuatara using modern tract-tracing discriminate harmful from harmless species based techniques, the uniqueness of this structure cannot solely on chemical cues has been demonstrated in be determined. lizards (Lacertidae, Iguanidae, and Anguidae) and in amphisbaenians, Blanus cinereus (Amo et al., 2004; Bealor and Krekorian, 2002; Cabido et al., 2004; 2.17.2.6.2.(ii) Squamates: Amphisbaenians, lizards, Lopez and Martin, 1994, 2001; Van Damme and and snakes Squamate reptiles are diverse and wide- Quick, 2001). Vomeronasal input is critical to the spread, and many species rely heavily on ability of crotaline snakes to recognize predatory chemosensory input to mediate all aspects of their kingsnakes, Lampropeltis getula (Miller and behavior. In many snakes and lizards the vomero- Gutzke, 1999). Responsiveness to chemical cues nasal system is behaviorally more important than from predators has been shown to increase the prob- the olfactory system and, in some species, the ability of surviving an encounter with a predator in neural structures that comprise the vomeronasal garden skinks, Lampropholis guichenoti (Downes, system are hypertrophied relative to those of the 2002). olfactory system. Although squamates are favored Chemical cues play a role in discrimination of the model animals for neurobiologists seeking to sex of conspecifics in amphisbaenians, Blanus ciner- understand the structure and function of the eus (Cooper et al., 1994), in recognition of familiar vomeronasal system, this endeavor is paradoxically and unfamiliar conspecifics in lacertid lizards complicated by the dominance of the vomeronasal (Aragon et al., 2003; Font and Desfilis, 2002), and system: in many squamates, the behavioral rele- in recognition of mates in snow skinks, vance of olfactory information is not clear. This Niveoscincus microlepidotus (Olsson and Shine, broad generalization does not apply to all squa- 1998). Male skinks (Eulamprus heatwolei) use che- mates, however. For example, in geckos the mosensory cues to assess female receptivity (Head olfactory system appears to be the dominant che- et al., 2005). Mate choice is influenced by chemical mosensory system (Schwenk, 1993b). cues in snakes (Thamnophis sirtalis) and lizards The nasal chemosensory systems play a critical (Lacerta monticola)(Lopez et al., 2003; Shine role in foraging and feeding in many squamates. et al., 2003), and female Swedish sand lizards For example, pygmy rattlesnakes (Sistrurus miliar- (Lacerta agilis) prefer odorants from males that dif- ius), which are sit-and-wait predators, aggregate in fer from themselves at the MHC class 1 loci (Olsson areas containing chemical cues from potential prey et al., 2003). Taken together, these observations animals (Roth et al., 1999). Snakes are well-known indicate that squamates can use chemosensory to follow chemical trails left by prey (Schwenk, information to discriminate sex, receptivity, quality 1994), as can some lizards, like Gould’s monitor as a potential mate, genotype, and individual iden- lizards, Varanus gouldii, and Gila monsters, tity of conspecifics. In squamates, pheromonal cues Heloderma suspectum (Garrett et al., 1996). are not always attractants: female brown tree Rattlesnakes, which strike and release their prey, snakes, Boiga irregularis, produce a cloacal secre- use chemical cues to follow and identify enveno- tion that decreases courtship intensity and duration mated animals. These cues derive from the venom in males (Greene and Mason, 2003). itself as well as recognition of the envenomated In general, squamates sniff to draw odorants individual (Chiszar et al., 1999; Furry et al., 1991; across the olfactory epithelium and use their tongues Lavin-Murcio and Kardong, 1995). Both snakes and to physically pick up chemical stimuli that will be lizards with experimentally impaired vomeronasal deposited in the vomeronasal organ. Geckos, which systems will attack but not consume prey, and garter depend more on olfactory than on vomeronasal snakes (Thamnophis sirtalis) with sectioned vomer- input, appear to sniff via oscillations of the throat onasal nerves stop following prey trails and (Dial and Schwenk, 1996). Garter snakes eventually cease eating (Alving and Kardong, (Thamnophis sirtalis) can detect prey odorants in 1996; Graves and Halpern, 1990; Halpern et al., air, without physical contact with the tongue, 1997; Haverly and Kardong, 1996). using the olfactory system (Halpern et al., 1997). Evolution of Vertebrate Olfactory Subsystems 379

Olfactory input appears to be important for elevat- Both cilia and short microvilli are present ing the rate of tongue-flicking, which then allows together on the olfactory receptor cells of the blue- the animal to sample with the vomeronasal organ tongued lizard, Tiliqua scincoides (Kratzing, 1975), (Halpern et al., 1997; Zuri and Halpern, 2003). although only ciliated receptor cells were found in a Although tongue-flicking is widespread among scanning electron microscopic investigation of the squamates, not all have forked tongues (Schwenk, olfactory epithelium in the garter snakes 1993a, 1994). Forked tongues have evolved repeat- Thamnophis sirtalis and T. radix (Wang and edly among squamates, particularly in groups that Halpern, 1980b). The morphology of receptor neu- forage widely (Schwenk, 1994). This arrangement rons may vary between species, but it also seems may be used to enhance two-point comparisons, for possible that short microvilli could have been over- example to facilitate trail following by improving looked or difficult to see in the preparations from edge detection (Schwenk, 1994). snakes. Both lizards and snakes possess only micro- The vomeronasal organ of squamate reptiles villar vomeronasal receptor neurons (Altner and opens directly into the oral cavity and has no con- Brachner, 1970; Altner and Mu¨ ller, 1968; nection with the nasal cavity, a condition that is Bannister, 1968; Kratzing, 1975; Takami and different than in any other tetrapod (Schwenk, Hirosawa, 1990; Wang and Halpern, 1980a, 1993a). Tongue-flicking brings molecules into the 1980b). The olfactory and vomeronasal receptor vomeronasal organ (reviewed in Schwenk, 1995; see genes from squamates have not yet been sequenced. also Graves and Halpern, 1989; Halpern and Kubie, In snakes and lizards, the vomeronasal epithelium 1980). Although the tongue tips are essential for expresses Gao and Gai2, which are also found in the vomeronasal sampling, the mechanism by which vomeronasal epithelium of mammals, but lack compounds move from the tongue to the vomero- Gaolf,Ga11, and Gaq, which are found in the mam- nasal ducts and into the organ is not understood malian olfactory epithelium (Labra et al., 2005; Luo (reviewed in Schwenk, 1994). The sublingual plicae et al., 1994). The vomeronasal epithelium in garter have been suggested to play a key role, perhaps by snakes expresses type IV adenylyl cyclase (Liu et al., creating suction within the vomeronasal organ or 1998), as does that of rats (Ro¨ ssler et al., 2000). duct (Young, 1993), but snakes with cauterized pli- The olfactory bulb of squamates is distinctly cae are able to find and consume food, and transport laminated, with both external and internal plexi- of molecules into the vomeronasal organs does not form layers present. In adult lizards (Podarcis differ between lesioned and control animals hispanica), periglomerular cells have a single den- (Halpern and Borghjid, 1997). Instead, the heavily drite, oriented parallel to the bulbar surface vascularized vomeronasal organ may use a pumping (Garcıá-Verdugo et al., 1986). Each mitral cell mechanism to create suction, similar to a mechan- has one primary dendrite that arborizes in a single ism used by some mammals (Halpern and Martıńez- glomerulus, as well as secondary dendrites that Marcos, 2003). travel through the external plexiform layer. In The squamate nasal cavity has a relatively simple contrast, in embryonic snakes (Elaphe quadrivir- organization, and in most species is essentially an gata), the primary dendrites of mitral cells elongated sac, the posterior dorsal portion of which arborize in many glomeruli (Iwahori et al., is lined with olfactory epithelium (Parsons, 1959b, 1989a). In both animals, the spiny dendrites of 1967). The dorsal wall of the spherical vomeronasal the axonless granule cells extend into the plexi- organ is lined with sensory epithelium, and the form layer, but do not arborize in the glomeruli lumen is made narrow by a ventral protruding struc- (Garcıá-Verdugo et al., 1986; Iwahori et al., ture called the mushroom body. In garter snakes 1989a). (Thamnophis sirtalis), as in other squamates, fluid The accessory olfactory bulb lacks the clear lami- from Harderian glands flows into vomeronasal nar organization of the olfactory bulb, and organ (Rehorek, 1997; Rehorek et al., 2000a, plexiform layers are absent (Iwahori et al., 1989b; 2000b). The products of the gland play a critical Llahi and Garcıá-Verdugo, 1989). In Podarcis, role in solubilizing a lipophilic pheromone in this incoming axons of vomeronasal receptor neurons species, allowing the pheromone to be detected by branch and enter more than one glomerulus, which vomeronasal receptor neurons (Huang et al., 2006). does not appear to be the case for olfactory receptor Animals from which the gland has been removed cell axons entering the main olfactory bulb (Garcıá- display impaired courtship behavior and prey cap- Verdugo et al., 1986; Llahi and Garcıá-Verdugo, ture, demonstrating the essential role of Harderian 1989). The dendrites of periglomerular cells arbor- gland secretions in signal detection in the vomero- ize in more than glomerulus, as do those of the nasal system in snakes (Mason et al., 2006). mitral cells (Iwahori et al., 1989b; Llahi and 380 Evolution of Vertebrate Olfactory Subsystems

Garcıá-Verdugo, 1989). Mitral cells also possess amygdala (Figure 9b). The intermediate olfactory secondary dendrites that extend through the exter- tract projects to the olfactory tubercle and olfactory nal and internal plexiform layers (Iwahori et al., gray, and joins the lateral olfactory tract in a projec- 1989b; Llahi and Garcıá-Verdugo, 1989). A class tion to the contralateral hemisphere (Figure 9c). The of cells that Llahi and Garcıá-Verdugo call ‘small accessory olfactory tract, carrying information from mitral cells’ have a soma in the outer mitral cell the vomeronasal organ, projects to three portions of layer or external plexiform layer and a single the amygdala: the nucleus sphericus, medial amyg- dendrite that arborizes sparsely in the glomerular dala, and nucleus of the accessory olfactory tract layers; these cells might be equivalent to the tufted (Figure 9d). The homology of the nucleus sphericus cells present in the mammalian olfactory bulb with regions of the amygdala in other tetrapods is (Llahi and Garcıá-Verdugo, 1989). The granule unclear, because it has been suggested to be the only cells resemble those in the main olfactory bulb amygdalar target of the accessory olfactory bulb (Iwahori et al., 1989b; Llahi and Garcıá- that does not project to the hypothalamus (Bruce Verdugo, 1989). Stellate cells do not seem to be and Neary, 1995; Lanuza and Halpern, 1997). This present in the main or accessory olfactory bulbs in interpretation is complicated by newer data demon- either species. strating a small projection from the nucleus Among squamates, the central olfactory projec- sphericus to the hypothalamus, and a much larger tions have been described in most detail in garter projection to the olfactostriatum (Martıńez-Marcos snakes, Thamnophis sirtalis and T. radix (Halpern, et al., 1999, 2002). The latter structure in turn 1976; Lanuza and Halpern, 1997, 1998). Three projects to the lateral posterior hypothalamic central olfactory projections are present in these nucleus, and may be homologous with the nucleus animals, comprising the lateral, intermediate, and accumbens (Martıńez-Marcos et al., 2005a, 2005b). medial olfactory tracts, illustrated in Figure 9. The Thus, the nucleus sphericus may be unique to squa- medial olfactory tract projects ipsilaterally to the mates, may be unique to lepidosaurs (squamates and anterior olfactory nucleus (Figure 9a). The lateral tuatara), or may be homologous with amygdalar olfactory tract projects bilaterally to the lateral cor- regions in other vertebrates but have reorganized tex, as well as to the external and ventral anterior connections in squamates.

Figure 9 Schematic dorsal view of the forebrain in garter snakes, Thamnophis sirtalis. a, Projections of the medial olfactory tract. b, Projections of the lateral olfactory tract. c, Projections of the intermediate olfactory pathway. d, Projections of the accessory olfactory tract. aob, accessory olfactory bulb; aon, anterior olfactory nucleus; ea, external amygdala; lc, lateral cortex; ma, medial amygdala; naot, nucleus of the accessory olfactory tract; ns, nucleus sphericus; ob, olfactory bulb; oe, olfactory epithelium; ot, olfactory tubercle; vaa, ventral anterior amygdala. Based on Halpern (1976), Lanuza and Halpern (1997, 1998), and Lohman and Smeets (1993). Evolution of Vertebrate Olfactory Subsystems 381

The medial amydala in garter snakes receives a terminal nerve during development in Chalcides direct projection from the accessory olfactory bulb, chalcides, but not in adults (D’Aniello et al., as well as indirect vomeronasal input via the nucleus 2001). Overall, the available data indicate either sphericus. The olfactory bulb has an indirect projec- that the immunoreactive characteristics of the term- tion as well, via the external and ventral anterior inal nerve in some squamate species differ from amygdaloid nuclei (Lanuza and Halpern, 1997, those in other jawed vertebrates, perhaps a way 1998; Martı´nez-Marcos et al., 1999). Because the that varies with developmental stage, or that the medial amygdala projects to the lateral posterior nerve is not present at all developmental stages in hypothalamic nucleus, which in turn projects to all squamate species. the hypoglossal nucleus, this pathway has been suggested to function in control of tongue-flicking in 2.17.2.6.2.(iii) Crocodilians Crocodiles, alliga- response to chemosensory input in snakes tors, caimans, and gharials are a relatively small (Martı´nez-Marcos et al., 2001). group, comprising only 20–25 species. The nasal Olfactory projections have also been examined in chemosensory systems have not been extensively several lizards, including Gekko gecko. The lateral studied in this group, and much of the information olfactory tract projects bilaterally to the entire available concerning the role of chemoreception in length of the lateral cortex, as well as to the anterior behavior is reviewed by Weldon and Ferguson olfactory nucleus, external amygdaloid nucleus, and (1993). perhaps the central amygdaloid nucleus. An inter- The nares of crocodilians are closely situated on mediate olfactory tract projects to the olfactory the dorsal portion of the snout and can protrude tubercle and joins the lateral olfactory tract to pro- even while most of the animal is submerged under ject to the contralateral hemisphere. A short medial water, suggesting that olfactory cues may serve olfactory tract is present. Finally, an accessory olfac- important functions in crocodilians. Crocodilians tory tract projects ipsilaterally to the nucleus draw air through the nasal cavity through buccal sphericus, external and central amygdaloid nuclei, oscillations that do not contribute to respiration and bed nucleus of the stria terminalis (Lohman and (Gans and Clark, 1976). Electroencephalograph Smeets, 1993; Lohman et al., 1988). In Tupinambis recordings from the olfactory bulb demonstrate teguixin, the projections are similar, although the activity coincident with buccal oscillations, demon- projections to the contralateral hemisphere differ strating that this behavior is equivalent to sniffing in somewhat between the two species (Lohman and crocodilians (Huggins et al., 1968; Naifeh et al., Smeets, 1993). To our knowledge, an extrabulbar 1970). Because of this demonstrated correlation, olfactory pathway has not been described in buccal pumping is used as a metric of olfactory squamates. sampling in some studies. For example, analysis of The anti-peptide antisera often used to identify buccal pumping demonstrates that juvenile alliga- the terminal nerve in other vertebrates produce tors (Alligator mississippiensis) respond to odorants mixed results with squamates. In adult garter snakes from food (Weldon et al., 1992). Other studies (Thamnophis sirtalis), a GnRH-immunoreactive further demonstrate that olfactory cues play a role terminal nerve is present, with a discrete ganglion in localization of food (Scott and Weldon, 1990; at the ventral border between the olfactory bulb and Weldon et al., 1990). rostral telencephalon (Smith et al., 1997). Crocodilians have prominent paracloacal and Intraventricular administration of a GnRH antago- gular glands, which are suspected to play a role in nist interferes with courtship behavior in these territorial and sexual behavior (Weldon and animals, although the relative contributions of the Ferguson, 1993). Recent work has described the terminal nerve and central GnRH neurons to this chemistry of the secretions from these glands (e.g., effect cannot be determined (Smith and Mason, Garcia-Rubio et al., 2002; Ibrahim et al., 1998; 1997). Among lizards, GnRH-immunoreactive Wheeler et al., 1999; Yang et al., 1999), but the cells and fibers are also present in the terminal behavioral significance of these compounds has not nerve of adult Eumeces laticeps and Sceloporus yet been elucidated. Behavioral observations suggest undulatus, but not in adult Anolis carolinensis that crocodilians rub scent glands during courtship (Rosen et al., 1997). In Podarcis sicula, GnRH- and nesting (Weldon and Ferguson (1993) and refer- immunoreactive cells and fibers cannot be detected ences therein), suggesting that pheromonal cues may in a terminal nerve, nor in the olfactory epithelium play a role in crocodilian reproductive behavior. and bulb, in either embryos or adults (D’Aniello The vomeronasal system has been lost in crocodi- et al., 1994a; Masucci et al., 1992). FMRFamide- lians. The vomeronasal organ begins to develop in immunoreactive cells and fibers are present in the the embryos of some alligators and crocodiles, but 382 Evolution of Vertebrate Olfactory Subsystems regresses by the time of hatching (Parsons, 1959a). embryos, but not adults (D’Aniello et al., 1999, The accessory olfactory bulb is similarly present in 2001). crocodilian embryos but not adults (Parsons, 1967, 1970). 2.17.2.6.2.(iv) Birds Although olfaction used to To our knowledge, the ultrastructure of the olfac- be considered unimportant or even absent in birds, tory epithelium in crocodilians has not been birds possess a robust olfactory system that med- described. The detailed anatomy of the olfactory iates many types of behavior. Notably, olfaction has bulb and tracts has been examined in young been shown to play a role in food finding in brown A. mississippiensis (Crosby, 1917). In these animals, kiwis (Apteryx australis; Wenzel, 1968, 1971), and clear laminae, including external and internal plexi- in some, but not all, species of vultures (Graves, form layers, are present. The glomeruli are 1992; Houston, 1984), as well as in ravens surrounded by somata of periglomerular cells, (Corvus corax; Harriman and Berger, 1986), par- which are clearly present. The mitral cells project rots (Strigops habroptilus; Hagelin, 2004), and primary dendrites into two or more glomeruli, but procellariiforms, the group of Antarctic seabirds also possess secondary dendrites that extend later- that includes shearwaters, petrels, and albatrosses ally through the external plexiform layer. Somata in (Hutchison and Wenzel, 1980). Some procellariid the external plexiform layer may belong to tufted species are highly sensitive to specific odorants cells, but could also belong to displaced mitral cells. such as dimethyl sulfide and 3-methyl pyrazine The granule cell layer contains three classes of cells. that are associated with krill, an important and The first are anaxonal intrinsic cells, with spiny patchily distributed food source (Nevitt, 2000; dendrites that extend in all directions within the Nevitt and Haberman, 2003; Nevitt et al., 1995, layer. The second are granule cells with spiny den- 2004). Olfaction plays a role in both feeding and drites that project to the glomerular layer but do not predator avoidance in chickens (reviewed in Jones seem to arborize inside glomeruli; Crosby (1917) and Roper, 1997). called these ‘stellate’ cells. The axons of these cells Olfaction has also been implicated in homing and enter the olfactory tracts and may project out of the navigation by passerines, which may use stable fea- olfactory bulb. The stellate cells have the smooth tures such as the presence of airborne hydrocarbons dendrites that arborize in glomeruli, and axons that to orient within landscapes (reviewed in Wallraff, project out of the olfactory bulb; Crosby (1917) 2003, 2004). Nevitt and Bonadonna (2005) suggest called these cells ‘‘goblet’’ cells. that Procellariiforms may use dimethyl sulfide in a The central projections of the olfactory bulb have similar fashion for navigating to small islands in been described in A. mississippiensis based on large open ocean areas. Golgi-stained material (Crosby, 1917) and in The role of olfaction in reproduction has been Caiman sklerops based on a degenerating fiber examined in only a handful of species. Crested auk- stain (Scalia et al., 1969). In Alligator, as in garter lets (Aethia cristatella) produce a scent that is snakes, three centripetal tracts emerge from the attractive to conspecifics, which humans perceive olfactory bulb. The individual tracts could not be as resembling tangerine odor (Hagelin et al., visualized in Caiman, but where the same target was 2003). Antarctic prions, Pachiptila desolata, can noted in both species we will assume that the same discriminate between chemical cues from their part- tract carries the fibers to it. The lateral tract projects ners and from other birds in the breeding colony, bilaterally to the lateral cortex, amygdala, and lat- demonstrating that this species is capable of indivi- eral portion of the olfactory peduncle. The medial dual recognition based on odorant cues (Bonadonna tract projects to the anterior hippocampus and med- and Nevitt, 2004). ial septum. The intermediate tract projects to the Many vertebrates show learned preferences for nucleus of the diagonal band of Broca and bilater- odorants experienced before birth or hatching, as ally to the olfactory tubercle. An additional small do chicks, G. domesticus, which could use such projection to the internal plexiform layer of the cues to recognize and orient to the nest (Sneddon contralateral olfactory bulb was observed in et al., 1998). Prions and petrels (Procellariidae), that Caiman. nest in burrows and return to them at night, use Medina et al. (2005) report that Nile crocodiles olfactory cues to find their own burrow, but olfac- (Crocodylus niloticus) possess a GnRH-immuno- tion appears to be unimportant for nest recognition reactive terminal nerve, but a detailed description in diurnal and surface-nesting species (Bonadonna of the pathway has not been published. The terminal and Bretagnolle, 2002; Bonadonna et al., 2001, nerve of the spectacled caiman (Caiman crocodilus) 2003a, 2003b, 2004). The use of acoustic cues to displays FMRFamide-like immunoreactivity in recognize partners and chicks might attract avian Evolution of Vertebrate Olfactory Subsystems 383 predators; thus, the use of chemical cues by some bulbs whereas others have large, obvious olfactory species may represent an adaptation for avoiding bulbs (Nieuwenhuys, 1967). Lamination appears to predation by other birds (Bonadonna et al., be highly variable across taxa, with small, indistinct 2003a). In some species, olfactory cues play an addi- layers in species that have tiny olfactory bulbs tional role in nesting: both European starlings (Nieuwenhuys, 1967) but clear cellular laminae (Sturnus vulgaris) and blue tits (Parus caeruleus) with internal and external plexiform layers visible use olfactory cues to select plant leaves that are even in species with moderately sized olfactory used as nest fumigants (Clark, 1991; Clark and bulbs, such as chickens and pigeons (McKeegan, Mason, 1985, 1987; Petit et al., 2002). 2002; Rieke and Wenzel, 1978). We are unaware Physiological studies demonstrate that the com- of any Golgi studies of the olfactory bulbs in birds; ponents of the olfactory system function similarly in thus, the cell types present and the details of cellular birds and other groups of vertebrates. At the level of morphology are largely unknown. Rieke and the olfactory epithelium and nerve, both excitatory Wenzel (1978) speculate that somata visible in the and inhibitory responses can be observed in a wide external plexiform layer of pigeons may belong to range of odorants (Jung et al., 2005; Shibuya and tufted cells. Tucker, 1967; Tucker, 1965). Similarly, single-unit An early study with pigeons (Columba livia) using recordings from the olfactory bulb demonstrate a combination of electrophysiology and degenerat- both excitatory and inhibitory responses to odor- ing fiber stains demonstrated ipsilateral projections ants (McKeegan, 2002; and references therein), to the piriform cortex, hyperstriatum ventrale and and an electroencephalography study with chickens the medial striatum, as well as projections to the found no significant differences in responses to contralateral globus pallidus and caudal portion of odorants relative to those recorded in mammals the medial striatum (Rieke and Wenzel, 1978; ter- using the same technique (Oosawa et al., 2000). minology after Reiner et al., 2004). The authors of Birds lack a vomeronasal organ and accessory this study did not distinguish among the different olfactory bulb (Huffman, 1963; Parsons, 1959b). olfactory tracts. The results of later study with the The olfactory epithelium is located on a single same species indicate that the medial olfactory tract spiral-shaped turbinate bone inside the nasal cavity, projects ipsilaterally to the septum and to a region and air is drawn across it as the animal breathes in dorsal to this, that an intermediate tract projects to and out (Bang and Wenzel, 1985). The anatomy of the olfactory tubercle and medial striatum, and that the olfactory epithelium has been examined in mem- the lateral olfactory tract projects bilaterally to the bers of five orders of birds (Anseriformes, piriform cortex and nucleus taeniae of the amygdala Charadriiformes, Ciconiiformes, Columbiformes, (Reiner and Karten, 1985; terminology after Reiner and Galliformes), and all possess unusual olfactory et al., 2004). The authors note that the projection to receptor neurons that are capped with cilia sur- the amygdala in birds is more restricted than in rounded by short microvilli (Bedini et al., 1976; some other groups, including turtles, which could Brown and Beidler, 1966; Drenckhahn, 1970; be related to the lack of a vomeronasal system in Graziadei and Bannister, 1967; Matsuzaki et al., birds. The differences in results obtained in the two 1982; Mu¨ ller et al., 1979; Okano and Kasuga, studies are difficult to understand, as they do not 1980). appear to be attributable to simple differences in Of the 20 000–23 000 genes in the genome of nomenclature or mistaken identification of cell jungle fowl (Gallus gallus), 283 odorant receptor groups. In young ducks (Anas platyrhynchos), the genes and ,100 odorant receptor pseudogenes medial olfactory tract projects to the dorsomedial have been identified (Hillier et al., 2004); however, hippocampus and superior frontal lamina, whereas a study of domesticated chickens (Gallus domesti- the lateral tract terminates in the pallial–subpallial cus) reported only 12 odorant receptor genes (Nef lamina, medial striatum, dorsomedial telencephalic and Nef, 1997). It is not clear whether this large wall, and posterior pallial amygdala (Teuchert et al., discrepancy is due to methodological differences, or 1986; terminology after Reiner et al., 2004). An to loss of genes as a result of domestication. The intermediate olfactory tract has not been described odorant receptor genes in galliforms are evolutiona- in this species. rily more closely related to those found in mammals The existence of an extrabulbar olfactory path- than to those found in aquatic vertebrates, like tele- way in birds has not been directly demonstrated, but ost fishes (Niimura and Nei, 2005). a study in which horseradish peroxidase was The size of the olfactory bulb relative to the brain injected into the nasal cavity in ducks (Anas platyr- varies considerably across birds (Bang and Cobb, hynchos) labeled a pathway that proceeded through 1968), and some species have small, conjoined the ventral olfactory bulb (Meyer et al., 1987; von 384 Evolution of Vertebrate Olfactory Subsystems

Bartheld et al., 1987). The terminations of these coriacea, in a laboratory setting (Constantino and fibers were not observed, but ganglion cell bodies Salmon, 2003). were not found and the injection is likely to have Olfactory cues play a role in courtship and repro- labeled primary olfactory receptor neurons. duction in turtles. Many turtles are endowed with As in other jawed vertebrates, the terminal nerve secretory glands (Ehrenfeld and Ehrenfeld, 1973), in birds arises from the nasal placode and, during although the behavioral significance of these secre- development, demonstrates immunoreactivity to tions has not been thoroughly studied. Males of both GnRH and FMRFamide (Norgren and many species sniff or bob their heads when stimu- Lehman, 1991; Norgren et al., 1992; Wirsig- lated by odorants from female scent glands (e.g., Wiechmann, 1990; Yamamoto et al., 1996). A Auffenberg, 1978; Kaufmann, 1992; Rose, 1970). FMRFamide-immunoreactive terminal nerve has Some species, such as the musk turtle Sternotherus been described in adult Japanese quail (Coturnix odoratus, produce chemicals that probably serve as japonica), but the anatomy and chemical character- a deterrent or aposematic signal to predators, but istics of the terminal nerve have not been studied in could also function in intraspecific communication detail in adult birds (Fujii and Kobayashi, 1992). (Eisner et al., 1977). The size of chin glands in male desert tortoises (Gopherus agassizii) is testosterone 2.17.2.6.2.(v) Turtles Turtles live in diverse habi- dependent, varying seasonally and with dominance tats, ranging from completely terrestrial tortoises to status (Alberts et al., 1994). Both males and females fully aquatic sea turtles. Many freshwater species can discriminate individuals based on secretions are semi-aquatic. Turtles sample the olfactory envir- from these glands (Alberts et al., 1994). In male onment both on land and under water, and in both Berlandier’s tortoise, Gopherus berlandieri, fatty cases, sniffing involves throat movements that acids from chin glands elicit aggressive behavior in resemble those used by crocodilians and amphibians males and cause females to approach and bob their (Belkin, 1968; McCutcheon, 1943; Root, 1949). As heads at a model painted with these compounds an adaptation for prolonged diving, many species (Rose, 1970; Rose et al., 1969). In contrast, in desert are remarkably tolerant of anoxia (Lutz and Milton, tortoises (Gopherus agassizii), males prefer to use 2004). This trait makes turtles well suited for elec- burrows scented with chin gland rubbings from con- trophysiological experiments, and many studies specific males compared with untreated burrows have used turtles as model animals for understand- (Bulova, 1997). Male stripe-necked terrapin ing general principles of olfactory system function in (Mauremys leprosa) have been shown to avoid vertebrates. These studies will not be comprehen- water conditioned by other males and prefer water sively reviewed here. conditioned by conspecific females, but only during Sea turtles can swim thousands of miles to parti- the breeding season (Mun˜ oz, 2004). The results of cular nesting beaches, and olfactory cues have been these studies clearly indicate that chemical signals suggested to play a key role in this behavior. The play a role in social and reproductive behavior in available data indicate that olfactory-based homing some turtles. cannot account for long-distance migration in tur- The organization of the nasal cavity and vomer- tles (reviewed in Lohmann et al., 1999). However, onasal organ varies considerably among species. In some studies indicate that hatchlings may imprint some, such as Testudo or Emys, the vomeronasal on odorants specific to their local environment, epithelium is not contained in a separate organ, but suggesting that such cues could play a role in lies along the ventromedial wall or in grooves in the short-range orientation or selection of nesting sites floor of the nasal cavity; the olfactory and vomer- (Grassman, 1993; Grassman and Owens, 1987; onasal regions of the nasal cavity are separated only Grassman et al., 1984). In addition, freshwater tur- by a slight horizontal ridge (Parsons, 1959b, tles (Chrysemys picta) can use chemical cues to 1959a). In loggerhead turtles, Caretta caretta, the discriminate water from home ponds versus that vomeronasal epithelium is more widely distributed from ponds with and without conspecifics (Quinn in the nasal cavity than is the olfactory epithelium and Graves, 1998). (Saito et al., 2000). On the other hand, in The importance of chemical cues in foraging and Dipsochelys and Dermochelys coriacea, the vomer- feeding has not the subject of extensive study in onasal organ is a discrete structure that is turtles. Nevertheless, Honigmann (1921) showed encapsulated in bone and opens into the oral cavity that both aquatic and semi-aquatic species will bite (Gerlach, 2005). A comparison of axon counts indi- at a bag filled with fish, but not at one filled with cates that olfactory receptor neurons outnumber sand. Olfaction has also been shown to play a role in vomeronasal receptor neurons in the Russian tor- foraging in leatherback turtles, Dermochelys toise (Testudo horsfieldii), which is terrestrial, but Evolution of Vertebrate Olfactory Subsystems 385 that two semi-aquatic species (Chinemys reevesii As in other reptiles, three central olfactory tracts and Mauremys japonica) have more vomeronasal have been described in turtles. The medial olfactory than olfactory receptor neurons (Hatanaka and tract projects ipsilaterally to the septum and to a Matsuzaki, 1993). medial cortical area that may be the homologue of As in some other reptiles, the dendrites of the the medial pallium/hippocampus of other verte- olfactory receptor neurons in Hermann’s tortoise, brates (Johnston, 1915; Reiner and Karten, 1985). Testudo hermanni, bear both cilia and numerous The lateral olfactory tract has a massive bilateral short microvilli (Delfino et al., 1990). Like other projection to a lateral cortical area, and an inter- tetrapods, turtles possess only microvillar vomero- mediate olfactory tract that may be a subdivision of nasal receptor neurons (Graziadei and Tucker, the lateral tract projects to the olfactory tubercle 1970; Hatanaka et al., 1982). Electrophysiological and the basal amygdaloid nucleus (Gamble, 1956; recordings from both the vomeronasal epithelium Reiner and Karten, 1985). In addition, a large olfac- and accessory olfactory bulb in Geoclemys reevesii tory projection to the entire pial surface of the demonstrate that the vomeronasal system in turtles amygdala was described in Trachemys scripta responds to general odorants with no inherent beha- (Reiner and Karten, 1985). Given that the accessory vioral significance, such as amyl acetate, geraniol, olfactory bulb in turtles is sometimes included in cineole, and citral (Hatanaka and Matsuzaki, 1993; injections or lesions of the main olfactory bulbs Hatanaka and Shibuya, 1989; Shoji et al., 1993; (Chkheidze and Belekhova, 2005; Gamble, 1956), Shoji and Kurihara, 1991). Recordings from disso- the description of the large amygdalar projection in ciated vomeronasal receptor neurons in stinkpot Trachemys may be due in part to inclusion of pro- turtles, Sternotherus odoratus, demonstrate that jections from the accessory olfactory bulb (see turtle vomeronasal cells also respond to a variety discussion in Eisthen, 1997). Because the separate of complex natural odorants, including urine and projections of the accessory olfactory bulb in turtles musk from both males and females, as well as odor- have not been described, it is not clear whether the ants derived from food pellets (Fadool et al., 2001). main and accessory olfactory systems project to Gao is expressed in different vomeronasal receptor different portions of the amygdala, as in other tetra- neurons than is Gai1–3, although the zonal segrega- pods (Chkheidze and Belekhova, 2005). tion of expression seen in mammals does not occur The development of the terminal nerve in turtles in Sternotherus (Murphy et al., 2001). In the same has been described by Johnston (1913) and Larsell species, females show higher levels of Gai1–3 expres- (1917), who were able to visualize the nerve as it sion and lower levels of TRP2 immunoreactivity courses over the surface of the olfactory nerve due to than do males (Murphy et al., 2001), and odorant its several conspicuous ganglia. To our knowledge, responses recorded from vomeronasal receptor neu- the histochemical characteristics of the nerve have rons from females are larger than those from males not been examined in detail in turtles, although (Fadool et al., 2001). Taken together, these data FMRFamide-immunoreactive cells and fibers do demonstrate that the vomeronasal system in turtles not appear to be present in or around peripheral responds to a wide range of chemicals, including olfactory structures in adult Trachemys scripta cues that may be involved in intraspecific commu- (D’Aniello et al., 2001, 1999). nication, and that the functioning of the system may be sexually dimorphic. The turtle olfactory bulb is highly laminar, with 2.17.2.6.3 Mammals The comparative neuro- both external and internal plexiform layers separat- biology of the olfactory system in mammals will ing the cell layers (Johnston, 1915; Orrego, 1961). not be described in detail here. Nevertheless, the The glomeruli are surrounded by periglomerular cells textbook view of the organization of the vertebrate (Orrego, 1961). Mitral cells usually extend primary olfactory system typically includes many features dendrites into two glomeruli, and possess long sec- that are unique to mammals, which should be ondary dendrites that project through the external noted by those interested in understanding the struc- plexiform layer (Mori et al., 1981; Orrego, 1961). ture and function of the olfactory system in Tufted cells may be present; Johnston (1915) called vertebrates in general. For example, the olfactory these ‘brush cells’. Two classes of granule cells are epithelium in mammals contains only ciliated olfac- present, one of which has processes that arborize in tory receptor neurons (reviewed in Eisthen, 1992), glomeruli (Johnston, 1915; Orrego, 1961). As in whereas in other vertebrates, the morphology of frogs, the olfactory bulbs are fused in some species, olfactory receptor neurons varies considerably. As with continuous layers and some neurites interacting in other tetrapods, the vomeronasal receptor neu- across the midline (Skeen and Rolon, 1982). rons in mammals are microvillar (Eisthen, 1992). 386 Evolution of Vertebrate Olfactory Subsystems

The organization of the olfactory bulb of mam- proteins are too energetically expensive for use mals is similar to that of other tetrapods, with a by aquatic vertebrates, as the hydrophilic proteins distinct laminar organization that includes large could easily dissolve in the water flowing over plexiform layers. In mammals, the mitral cells pos- through the nasal sac (discussed in Eisthen, sess a single primary dendrite that arborizes in one 2002). Although their presence in Xenopus and glomerulus, and prominent secondary dendrites that mammals suggests that odorant binding proteins extend orthogonal to the primary dendrite through should be broadly present in terrestrial tetrapods, the external plexiform layer (reviewed in Eisthen, Baldaccini et al. (1986) were unable to find bind- 1997; Nieuwenhuys, 1967). In addition to mitral ing proteins in birds (Columba livia and Cairina cells, mammals possess tufted cells, a second class moschata) and turtles (Testudo hermanni), despite of output cell (Nieuwenhuys, 1967; Pinching and being able to sequence them from several mam- Powell, 1971). Although the somata of most tufted malian species. Given that the odorant binding cells are found in the external plexiform layers, proteins in both Xenopus and mammals are lipo- another group, the external tufted cells, have cell calins, it is surprising that these proteins have not bodies in the glomerular layer; the glomerular been found in other classes of vertebrates. Perhaps layer also contains short-axon cells with no clear different types of molecules are used as binding counterpart among nonmammalian vertebrates proteins in other groups of vertebrates, or perhaps (Pinching and Powell, 1971). The lateral olfactory the presence of these molecules in both Xenopus tract projects ipsilaterally, and not bilaterally as in and mammals is convergent and not informative other vertebrates (Skeen et al., 1984). Mammals about tetrapods generally. may lack a true medial olfactory tract that arises Although vomeronasal receptor neurons invari- from the olfactory bulb; instead, the tract of the ably terminate in microvilli, vertebrate olfactory same name in mammals appears to arise from the receptor neurons are morphologically diverse; anterior olfactory nucleus (Lohman and Lammers, their phylogenetic distribution is shown in 1967; Nieuwenhuys, 1967). The vomeronasal sys- Figure 10. Researchers have long considered the tem is generally present in mammals, but has been possibility that the microvillar olfactory receptor lost in cetaceans as well as in some bats and primates (reviewed in The Vomeronasal Organ and Its Evolutionary Loss in Catarrhine Primates). s

2.17.3 Conclusions In this section, we integrate the preceding information, both to describe patterns of change and to Hagfishes Lampreys Cartilaginous fishe fishes Ray-finned Coelacanths Lungfishes Salamanders Frogs Turtles Squamates Crocodilians Birds Mammals consider the functional implications of these changes. In general, we will not cite references for information presented earlier, as details are provided in the sections pertaining to each taxonomic group.

2.17.3.1 Evolutionary Changes in the Organization of the Olfactory Epithelium

Before odorants contact the sensory receptor cells, Figure 10 Phylogenetic distribution of receptor cell types and they pass through a mucous layer. This mucus con- correlation with expression of olfactory (OR) and vomeronasal tains odorant binding proteins, which are known to (V2R) receptor genes in the olfactory (olf) and vomeronasal be present in terrestrial mammals and in frogs. (vom) epithelia. The expression of genes is demonstrated with Interestingly, in Xenopus, the binding proteins are electron microscopy; otherwise, it is inferred. Boldface letters indicate cell types: C, ciliated receptor neuron; M, microvillar expressed in the principal cavity, which is used for receptor neuron; X, receptor neuron with both cilia and microvilli; detecting odorants in air, but not in the middle Y, crypt-type receptor neuron. ? indicates that the morphology of cavity, which is used for detecting odorants in the receptor neurons is unknown. Salamanders possess both water. Odorant binding proteins have been sug- ciliated and microvillar receptor neurons, but we do not yet know gested to represent an adaptation for detecting whether one or both express olfactory-type odorant receptor genes. Crocodilians, birds, and all fishes lack a vomeronasal odorants in air, to transport hydrophobic molecules organ. This diagram is simplified, as it assumes homogeneity through the mucous layer to the receptor neurons. within large taxonomic groups, such as teleost fishes. See text Another possibility is that odorant binding for references. Evolution of Vertebrate Olfactory Subsystems 387 neurons in teleosts could be homologous with 2.17.3.2 Evolutionary Changes in the mammalian vomeronasal receptor neurons (dis- Organization of the Olfactory Bulbs cussed in Dulka, 1993; Eisthen, 1992). Recent The organization of olfactory bulb circuits differs data demonstrating that the olfactory epithelium among vertebrate groups. Figure 11 illustrates the of teleost fishes contains ciliated receptor neurons phylogenetic distribution of the cellular elements of that express genes typical of olfactory receptor the olfactory bulb. In interpreting the significance of neurons, and microvillar neurons that express these patterns, it is important to note that we genes typical of vomeronasal receptor neurons assigned names to the different cell types based on may bolster this impression. Nevertheless, the their morphology alone, and did so independently of expression of olfactory- or vomeronasal-typical the names used by the authors of the source mate- genes may not correlate tightly with cell morphol- rial. Thus, we use the term ‘stellate cell’ for any cells ogy (Figure 10). Further, if one assumes that the correlation applies to all craniates, then unlikely patterns are predicted to emerge. For example, although hagfish have both ciliated and microvillar olfactory receptor neurons, lampreys possess only ciliated olfactory receptor neurons, and those of sharks and skates terminate in microvilli. If the association between receptor cell morphol- Hagfishes Lampreys Cartilaginous fishes fishes Ray-finned Coelacanths Lungfishes Salamanders Frogs Turtles Squamates Crocodilians Birds Mammals ogy and gene expression arose with the earliest Lamination craniates, we would expect to see only olfactory- Periglomerular cells type genes expressed in lampreys, and only vomer- Mitral cells Stellate cells onasal-type genes expressed in sharks and skates. Granule cells Thus, the apparent correlation between receptor cell morphology and gene expression observed in teleosts and mammals may be a coincidence, may represent examples of convergent evolution, or may represent a derived condition that pertains only to bony vertebrates. Figure 11 Phylogenetic distribution of cellular elements of Alternatively, the morphological categories we are olfactory bulb circuitry. The hypothesized ancestral condition, using may be too crude. For example, distinct subsets illustrated by hagfish, includes a low degree of lamination, with of microvillar and ciliated olfactory receptor neurons cell bodies frequently located across laminar boundaries and a lack of plexiform layers; periglomerular cells absent; mitral cells project to different regions of the olfactory bulb in with primary dendrites that arborize in more than one glomerulus channel catfish, Ictalurus punctatus (Morita and and no secondary (basal) dendrites; stellate cells present, with Finger, 1998). Similarly, in goldfish (Carassius aura- smooth dendrites that arborize in glomeruli; and granule cells tus), the microvillar receptor neurons that are bearing smooth dendrites, with an axon present. Boxes with crosses indicate cases for which information is unavailable. immunoreactive for Gaq are shorter and have more Lamination: white box low degree of lamination, with cell ¼ stiff microvilli than other microvillar receptor neu- bodies frequently located across laminar boundaries and a rons (Hansen et al.,2004). If ‘ciliated’ and lack of plexiform layers; gray moderate degree of lamination, ¼ ‘microvillar’ receptor neurons in vertebrates actually with cell bodies confined to clear layers, and a lack of plexiform layers; black high degree of lamination, with cell bodies con- comprise several subclasses of cells, cell morphology ¼ and receptor gene expression may be tightly corre- fined to clear layers, separated by one or more plexiform layers. Periglomerular cells: white absent; gray ambiguous condi- ¼ ¼ lated within specific subclasses of receptor neurons. If tion, in which candidate periglomerular cells have been so, some of the vomeronasal-type microvillar recep- described by some authors; black periglomerular cells pre- ¼ tor neurons in goldfish could be homologous with sent. Mitral cells: white primary dendrites arborize in more ¼ vomeronasal receptor neurons in mammals. On the than one glomerulus, with no secondary dendrites; white/ gray primary dendrites arborize in only one glomerulus, with other hand, microvillar olfactory and vomeronasal ¼ no secondary dendrites; gray primary dendrites arborize in ¼ receptor neurons often contain centrioles and basal more than one glomerulus, with secondary dendrites that extend bodies, suggesting that they derive evolutionarily laterally; black primary dendrites arborize in only one glomer- ¼ from ciliated cells (Delfino et al.,1990). If ciliogenesis ulus, with secondary dendrites that extend laterally. Stellate cells: white absent; gray stellate cells present, with spiny is suppressed in microvillar cells (Kolnberger and ¼ ¼ dendrites that arborize in glomeruli; black stellate cells pre- Altner, 1971; Pyatkina, 1976), then receptor cell ¼ sent, with smooth dendrites that arborize in glomeruli. Granule morphology might be evolutionarily labile, with no cells: white smooth dendrites, with an axon present; gray - ¼ ¼ fixed relationship between receptor cell morphology spiny dendrites, with axon present; black spiny dendrites, with ¼ and receptor gene expression. no axon. See text for references. 388 Evolution of Vertebrate Olfactory Subsystems with a soma in the deep layers of the bulb, and with cells have multiple primary and secondary den- multiple dendrites that arborize in glomeruli. The drites. The breadth of inputs to mitral cells somata of stellate cells are generally star-shaped, as therefore differs considerably among groups. In the name implies. A ‘granule cell’ is a cell with an mammals, each olfactory receptor neuron is oval-shaped soma located in a deep layer of the believed to express only one odorant receptor bulb, and with dendrites that project upward but gene, and receptor neurons that express the same that do not enter or arborize in glomeruli. The den- gene project to the same glomerulus (reviewed in drites of granule cells generally bear spines, whereas Mombaerts, 2004). Thus, mitral cells with dendrites those of stellate cells are generally smooth. In many in a single glomerulus receive inputs from a homo- groups, both types of cells have a long axon. geneous population of receptor neurons. In contrast, ‘Periglomerular’ cells have a soma in the glomerular olfactory receptor neurons in teleost fishes may layer, with dendrites that arborize in several glomer- express more than one receptor gene (Ngai et al., uli. In many groups, these cells have an axon that 1993a; Speca et al., 1999), and mitral cells in these projects at least to deeper layers of the bulb. animals extend dendrites to several glomeruli. Ideally, additional criteria would be used to recog- Perhaps these glomeruli receive inputs from receptor nize cell types, such as data concerning the neurons that express one receptor gene in common, neurotransmitter characteristics of the cells. in which case the type of coding occurring in the Unfortunately, such data are available for an olfactory bulbs of teleosts and mammals may be extremely limited set of species, and cannot be similar. If not, the nature or location of odorant used for broad phylogenetic comparisons. Thus, information processing in teleosts may differ con- the categories we are using are quite broad, and siderably from that in mammals, which seems quite in at least some groups probably encompass sev- possible given that teleosts possess stellate cells, eral recognizably different types of cells; for which mammals lack, and that the stellate and gran- example, Golgi data indicate that multiple classes ule cells in teleosts have long axons that may project of mitral cells are present in some animals, and to secondary olfactory regions in the telencephalon. histochemical data indicate that several types of Other aspects of bulbar circuitry differ among periglomerular cells are present in some species. groups, although the functional consequences are With these caveats in mind, the pattern depicted not easy to predict. For example, morphologically in Figure 11 indicates that the organization of the distinct periglomerular cells are present only in tet- olfactory bulb did not undergo any sudden, dra- rapods, but other cells, such as the displaced mitral matic shifts over the course of vertebrate cells in the glomerular layer of hagfish and lam- evolution. Rather, changes in organization occurred preys, or even stellate cells, may serve similar as a series of steps. Laminar organization of brain functions. Similarly, a dorsal commissure connect- structures is generally regarded as indicative of mul- ing the two olfactory bulbs is unique to hagfish and tiple stages of integration, with cells in one layer lampreys, but other types of connections between receiving processed input from cells in other layers the bulbs exist in other groups. For example, in frogs and then processing these signals in more extensive and some turtles and birds, the two olfactory bulbs ways. What are the functional consequences of the are fused across the midline, with mitral and granule almost complete lack of lamination observed in hag- cell dendrites apparently integrating inputs from fish or lampreys, or the high degree found in both sides. In other animals, such as the hedgehog squamates or mammals? It is difficult to make pre- Erinaceus europaeus, the olfactory bulb projects dictions based on this feature alone, particularly ipsilaterally to the anterior olfactory nucleus, because the numbers of cell types with axons that which sends fibers to the mitral cell layer of the project to secondary processing areas differs consid- contralateral olfactory bulb (De Carlos et al., erably among these groups. Perhaps more 1989). These different anatomical arrangements processing occurs in the olfactory bulbs in some may facilitate integration of inputs to the two animals, and in secondary olfactory regions in nares, or perhaps bilateral comparison of inputs to others. enhance localization of odorant sources. The textbook view of a mitral cell is one with a Granule cells lack dendritic spines in hagfish, single primary dendrite that extends to one glomer- lampreys, and cartilaginous fishes, and functionally ulus, and prominent secondary dendrites that equivalent circuitry is probably lacking. The pre- extend laterally to interact with dendrites of granule sence of spines varies even within groups: for cells as well as secondary dendrites of other mitral example, although granule cell dendrites are spiny cells. Nevertheless, this type of mitral cell is present in most amphibians, Herrick’s (1931) studies indi- only in mammals, and in most vertebrates, mitral cate that the dendrites are smooth in mudpuppies Evolution of Vertebrate Olfactory Subsystems 389

(Necturus). What are the functional consequences of this diversity? Given that dendritic spines are often involved in plasticity in other regions of the central nervous system, does this suggest that olfactory bulb circuits in animals lacking spines are more hard-wired? Alternatively, or in addition, perhaps Hagfishes Cartilaginous fishes fishes Ray-finned Coelacanths Turtles Birds Mammals Lampreys Lungfishes Salamanders Frogs Squamates Crocodilians less compartmentalization of processing occurs in the olfactory bulbs of animals with granule cells that lack dendritic spines (e.g., Woolf et al., 1991). Finally, additional types of output cells have been described in some animals, such as the ruffed cells present in teleost olfactory bulbs, or the tufted cells present in reptiles and mammals. Without intensive electrophysiological and neurochemical studies of Figure 12 Evolutionary changes in the central projections of the olfactory bulb circuits in diverse vertebrates, the olfactory bulbs. The hypothesized ancestral condition the functional significance of these differences in includes the presence of a lateral olfactory tract that projects organization are likely to remain mysterious. bilaterally to lateral pallial/cortical areas and the striatum; the presence of a medial olfactory tract that projects ipsilaterally Given that we also lack detailed psychophysical to the septum; and an extrabulbar olfactory pathway. data concerning the relative capabilities of the olfac- Numbers indicate hypothesized changes in connectivity. 1, tory system in diverse vertebrates, the behavioral Reduction in the extent of projections of the lateral olfactory consequences of differences in circuitry cannot be tracts. 2, Possible loss of the medial olfactory tract and bilat- predicted. eral projections of the lateral tract. 3, Origin of a distinct vomeronasal system, including an accessory olfactory tract that projects to the amygdala. Origin of an olfactory projection to separate portions of the amygdala via the lateral olfactory 2.17.3.3 Evolutionary Changes in the tract. Origin of an olfactory projection to the medial pallium / Organization of Central Olfactory Projections hippocampus via the medial olfactory tract. 4, Origin of an Broad patterns of change in the organization of intermediate olfactory tract that projects to the olfactory tubercle, and loss of olfactory projection to the striatum via the central olfactory pathways are illustrated in lateral olfactory tract. 5, Loss of the vomeronasal system. 6, Figure 12. Most vertebrates possess a medial olfac- Loss of the medial olfactory tract and the contralateral projec- tory tract that projects to the septum. The tract has tion of the lateral olfactory tract. See text for references. been lost in mammals and may also be lost in cartilaginous fishes, although the central projections in this group must be examined using modern tract- tract methods before strong conclusions can be lateral olfactory tract has extensive bilateral projec- drawn. In tetrapods, the tract may acquire a projections to both dorsal and ventral telencephalic tion to the medial pallium/hippocampus. Such a regions in hagfish and lampreys, and that these pro- projection may also be present in ray-finned fishes, jections are more restricted in jawed vertebrates. In but given the confusion concerning pallial homolo- addition, the lateral olfactory tract projects to the gies in this group, we cannot reach a conclusion striatum in most groups discussed in this review, but concerning this matter at present. Nevertheless, the not in reptiles; instead, an intermediate olfactory medial pallium/hippocampus is generally involved tract projects to the olfactory tubercle in the ventral in memory and spatial perception, and one might striatum. This shift in connectivity suggests that the expect that olfactory input to this region would be intermediate olfactory tract may be a branch of the behaviorally important for many vertebrates. lateral olfactory tract, or may have arisen from it In general, a lateral olfactory tract is present and evolutionarily. Some support for this idea comes projects bilaterally to lateral and dorsal pallial or from data demonstrating that the intermediate and cortical areas. Although it is tempting to interpret lateral olfactory tracts project together to the con- this connectivity as a conserved feature that must be tralateral hemisphere in snakes (Lanuza and critical for processing olfactory information in all Halpern, 1998), and that the two tracts run in par- vertebrates, in many animals the homologies of pal- tial continuity with each other through the rostral lial areas are based largely on their receipt of forebrain in turtles (Reiner and Karten, 1985). olfactory input. Thus, it would be circular to argue A discrete vomeronasal system, including an that this projection is conserved or that the lateral accessory olfactory bulb and tract, is present only olfactory tract projects to homologous areas in in tetrapods. The main central target of the acces- diverse vertebrates. One clear trend is that the sory olfactory tract is the amygdala. The lateral 390 Evolution of Vertebrate Olfactory Subsystems olfactory tract also projects to the amygdala, but the in jawed vertebrates (Northcutt and Puzdrowski, specific regions of the amygdala that receive input 1988; Wicht and Northcutt, 1993). Overall, how- from the lateral and accessory olfactory tracts show ever, the available data clearly demonstrate that the little or no overlap in the species studied to date. invasion scenario is incorrect (reviewed in Clear olfactory projections to the amygdala have Northcutt, 1981). been recognized only in tetrapods; thus, the origin Two large-scale changes in the organization of the of an olfactory bulb projection to the amygdala olfactory system have occurred over the course of correlates roughly with the origin of the vomerona- vertebrate evolution: the origin of the terminal sal system. Given that the portions of the amygdala nerve, and the origin of the vomeronasal system. that receive olfactory and vomeronasal input are Both hagfish and lampreys appear to lack a terminal interconnected (reviewed in Halpern and nerve, as no projection has been described that com- Martıńez-Marcos, 2003), perhaps this olfactory prises a peripheral ganglion and fibers that display projection arose to facilitate integration of informa- the types of immunoreactivity that characterize the tion from the two systems. If so, integration of terminal nerve (reviewed in Wirsig-Wiechmann chemosensory inputs cannot be the only current et al., 2002a). If so, then the terminal nerve arose function of the olfactory projection to the amyg- in jawed vertebrates. As described above, studies of dala, as this projection is retained in crocodilians the terminal nerve ganglion and retina in teleost and birds, which have lost the vomeronasal system. fishes strongly indicate that the terminal nerve An olfactory projection to the amygdala may also serves a modulatory function, and studies with sal- exist in ray-finned fishes, but the homologies of amanders demonstrate that terminal nerve-derived possible amygdala-equivalent areas in ray-finned peptides modulate activity in the olfactory epithe- fishes are controversial (reviewed in Northcutt, lium. In teleost fishes, the terminal nerve ganglion 2006). Given that the vomeronasal-like and olfac- receives input from the olfactory, visual, and soma- tory-like receptor neurons in teleosts project to tosensory systems. In amphibians, the extent to different portions of the olfactory bulb, perhaps which terminal nerve peptides modulate olfactory tracing the central projections of these two regions epithelial activity depends on the animal’s physiolo- would provide new insight into the organization of gical or behavioral state, as both hunger and the amygdala in teleosts. reproductive condition appear to play a role. Similar centrifugal modulation occurs in the retina and cochlea of many vertebrates (reviewed in 2.17.3.4 Evolution of Vertebrate Olfactory Akopian, 2000; Manley, 2000, 2001). Do hagfish Subsystems and lampreys lack this modulation, or do they pos- A common assertion is that the olfactory system is sess alternate mechanisms for regulating olfactory phylogenetically ancient, or that olfaction is the old- responses with regard to their physiological needs? est sensory system. The basis of such statements is What are the overall functional consequences for unclear. Olfactory systems in animals in several animals that possess or lack such mechanisms? phyla, including nematodes, mollusks, arthropods, Perhaps the more active foraging and courting beha- and vertebrates, possess olfactory systems with simi- viors of jawed vertebrates benefit from more central lar features, but these features probably arose control of olfactory epithelial function, which is independently in each group, in response to similar unnecessary in hagfishes and lampreys. constraints and as adaptations for similar tasks A discrete vomeronasal system is present only in (Eisthen, 2002). Perhaps such statements simply tetrapods, but recent work, described above, clearly indicate that the ability to sense chemicals in the indicates that the elements of the vomeronasal sys- external environment is widespread among animals, tem are present in teleost fishes: the olfactory and although this ability is by no means restricted to vomeronasal receptor genes, as well as their asso- metazoans. A third possibility is that such state- ciated G-proteins and ion channels, are expressed in ments are an oblique reference to a long-standing different receptor neurons in the olfactory epithe- idea that among vertebrates, the forebrain was ori- lium. In goldfish (Carassius auratus), the ginally an olfactory structure, and that inputs from vomeronasal-type elements are expressed in micro- other sensory systems ‘invaded’ the telencephalon villar olfactory receptor neurons, whereas the via thalamus over the course of vertebrate evolution olfactory-type elements are expressed in ciliated (Arie¨ns-Kappers et al., 1936; Edinger, 1904; receptor neurons (Hansen et al., 2004). Although Herrick, 1948). As described above, the olfactory these morphological cell types are superficially simi- projections in hagfish and lampreys distribute to lar to the vomeronasal and olfactory receptor larger portions of the telencephalon than do those neurons in mammals, as discussed above, it appears Evolution of Vertebrate Olfactory Subsystems 391 unlikely that the two cell types in teleosts simply Alonso, J. R., Lara, J., Miguel, J. J., and Aijoń, J. 1987. Ruffed segregated into two epithelia to give rise to a sepa- cells in the olfactory bulb of freshwater teleosts. I: Golgi rate vomeronasal organ (Eisthen, 2004). In zebra impregnation. J. Anat. 155, 101–107. Altner, H. 1962. Untersuchungen u¨ ber Leistungen und Bau der fish (Danio rerio), the two classes of receptor neu- Nase des su¨ dafrikanischen Krallenfrosches Xenopus laevis rons send their axons to different portions of the (Daudin, 1803). Z. Vergl. Physiol. 45, 272–306. olfactory bulb (Sato et al., 2005), an arrangement Altner, H. and Brachner, I. 1970. The ultrastructure of the vomer- similar to that in tetrapods, in which the olfactory onasal organ in reptilia. Z. Zellforsch. Mikrosk. Anat. 105, and vomeronasal receptor neurons send axons to 107–122. Altner, H. and Mu¨ ller, W. 1968. Electrophysiological and elec- distinct olfactory and accessory olfactory bulbs. In tron microscopial investigations of the sensory epithelium in tetrapods, the projections of the olfactory and acces- the vomero-nasal organ in lizards (Lacerta). Z. Vergl. Physiol. sory olfactory bulb differ, and the next step is to 60, 151–155. determine whether the portions of the olfactory bulb Alving, W. R. and Kardong, K. V. 1996. The role of the vomer- that receive input from the two cell types in teleosts onasal organ in rattlesnake (Crotalus viridis oreganus) predatory behavior. Brain Behav. Evol. 48, 165–172. also have distinct projections. If so, it would appear Amo, L., Lopez, P., and Martin, J. 2004. Chemosensory recogni- that teleost fishes have a complete vomeronasal sys- tion and behavioral responses of wall lizards, Podarcis tem intermingled with the olfactory system. It muralis, to scents of snakes that pose different risks of preda- would be interesting to know whether the same is tion. Copeia 691–696. true of other classes of fishes; perhaps the vomero- Anadoń, R., Manso, M. J., Rodriguez-Moldes, I., and Becerra, M. 1995. Neurons of the olfactory organ projecting nasal system has been present since the origin of to the caudal telencephalon and hypothalamus: A carbocya- vertebrates, and only became separate from the nine-dye labeling study in the brown trout (Teleostei). olfactory system in tetrapods. Because the func- Neurosci. Lett. 191, 157–160. tional differences between the olfactory and Andriason, I. and Leveteau, J. 1989. Projection bilate´rale de la muquese olfactive chez la grenouille. 309, vomeronasal systems are unclear (reviewed in Baxi C. R. Acad. Sci. III 513–518. et al., 2006; Halpern and Martıńez-Marcos, 2003), Anton, W. 1908. Beitrag zur morphologie des Jacobsonschen it is difficult to speculate about the causes or con- organs und der Nasenho¨ hle der cryptobranchiaten. sequences of the origin of a separate vomeronasal Morphol. Jahrb. 38, 448–470. system in tetrapods. Anton, W. 1911. Die Nasenho¨ hle der Perennibranchiaten. Morphol. Jahrb. 44, 179–199. Aragon, P., Lopez, P., and Martin, J. 2003. Differential avoidance responses to chemical cues from familiar and unfamiliar con- Acknowledgments specifics by male Iberian Rock Lizards (Lacerta monticola). J. Herpetol. 37, 583–585. The authors thank Anne Hansen for providing them Arie¨ns-Kappers, C. U., Huber, C. G., and Crosby, E. C. 1936. with the lovely electron micrographs in Figure 6. The Comparative Anatomy of the Nervous System of Their research is supported by a grant from the Vertebrates, Including Man. Macmillan. National Institutes of Health (DC05366). Atema, J. 1980. Chemical senses, chemical signals, and feeding behavior in fishes. In: Fish Behavior and its Use in the Capture and Cultivation of Fishes (eds. J. E. Bardach, J. J. Magnuson, References R. C. May, and J. M. Reinhart), pp. 57–94. ICLARM. Auffenberg, W. 1978. Courtship and breeding behavior in Abe, H. and Oka, Y. 1999. Characterization of Kþ currents Geochelone radiata (Testudines: Testudinidae). underlying pacemaker potentials of fish gonadotropin-releasing Herpetologica 34, 277–287. hormone cells. J. Neurophysiol. 81, 643–653. Badenhorst, A. 1978. The development and the phylogeny of the Abe, H. and Oka, Y. 2000. Modulation of pacemaker activity by organ of Jacobson and the tentacular apparatus of Ichthyophis salmon gonadotropin-releasing hormone (sGnRH) in term- glutinosus (Linne´). Ann. Univ. Stellenbosch Ser. A2 1, 1–26. inal nerve (TN)-GnRH neurons. J. Neurophysiol. 83, Bakhtin, E. K. 1976. Morphology of the olfactory organ and 3196–3200. possible functional interpretation. J. Ichthyol. 16, 786–804. Abe, H. and Oka, Y. 2002. Mechanisms of the modulation of Bakhtin, E. K. 1977. Peculiarities of the fine structure of the pacemaker activity by GnRH peptides in the terminal nerve- olfactory organ of Squalus acanthias. Tsitologija 14, 725–731. GnRH neurons. Zool. Sci. 19, 111–128. Baldaccini, N. E., Gagliardo, A., Pelosi, P., and Topazzini, A. Ache, B. W. and Young, J. M. 2005. Olfaction: Diverse species, 1986. Occurrence of a pyrazine binding protein in the nasal conserved principles. Neuron 48, 417–430. mucosa of some vertebrates. Comp. Biochem. Physiol. B 84, Akopian, A. 2000. Neuromodulation of ligand- and voltage- 249–253. gated channels in the amphibian retina. Microsc. Res. Tech. Bang, B. G. and Cobb, S. 1968. The size of the olfactory bulb in 50, 403–410. 108 species of birds. The Auk 85, 55–61. Alberts, A. C., Rostal, D. C., and Lance, V. A. 1994. Studies on Bang, B. G. and Wenzel, B. M. 1985. Nasal cavity and olfactory the chemistry and social significance of chin gland secretions system. In: Form and Function in Birds (eds. A. S. King and J. in the desert tortoise, Gopherus agassizii. Herpetol. Monogr. McClelland) vol. 3, pp. 195–225. Academic Press. 8, 116–124. Bannister, L. H. 1968. Fine structure of the sensory endings in the Alioto, T. S. and Ngai, J. 2005. The odorant receptor repertoire vomero-nasal organ of the slow-worm Anguis fragilis. Nature of teleost fish. BMC Gen. 6, 173. 217, 275–276. 392 Evolution of Vertebrate Olfactory Subsystems

Bateson, W. 1890. The sense-organs and perceptions of fishes; Bonadonna, F., Villafane, M., Bajzak, C., and Jouventin, P. 2004. with remarks on the supply of bait. J. Marine Biol. Assoc. UK Recognition of burrow’s olfactory signature in blue petrels, 1, 225–256. Halobaena caerulea: An efficient discrimination mechanism Baxi, K. N., Dorries, K. M., and Eisthen, H. L. 2006. Is the in the dark. Anim. Behav. 67, 893–898. vomeronasal system really specialized for detecting phero- Bozza, T. C. and Kauer, J. S. 1998. Odorant response properties mones? Trends Neurosci. 29, 1–7. of convergent olfactory receptor neurons. J. Neurosci. 18, Bazer, G. T., Ebbesson, S. O. E., Reynolds, J. B., and Bailey, R. P. 4560–4569. 1987. A cobalt-lysine study of primary olfactory projections Braford, M. R., Jr. 1995. Comparative aspects of forebrain orga- in king salmon fry (Oncorhynchus shawytscha walbaum). nization in the ray-finned fishes: Touchstones or not? Brain Cell Tissue Res. 248, 499–503. Behav. Evol. 46, 259–275. Bealor, M. T. and Krekorian, C. O. 2002. Chemosensory identi- Braford, M. R., Jr. and Northcutt, R. G. 1974. Olfactory bulb fication of lizard-eating snakes in the desert iguana, projections in the bichir, Polypterus. J. Comp. Neurol. 156, Dipsosaurus dorsalis (Squamata: Iguanidae). J. Herpetol. 36, 165–178. 9–15. Braun, C. B. 1995. Schreiner organs: A new craniate chemosen- Bedini, C., Fiaschi, V., and Lanfranchi, A. 1976. Olfactory nerve sory modality in hagfishes. J. Comp. Neurol. 392, 135–163. reconstitution in the homing pigeon after resection: Braun, C. B., Wicht, H., and Northcutt, R. G. 1995. Distribution Ultrastructural and electrophysiological data. Arch. Ital. of gonadotropin-releasing hormone immunoreactivity in the Biol. 114, 1–22. brain of the Pacific hagfish, Eptatretus stouti (Craniata: Belkin, D. A. 1968. Aquatic respiration and underwater survival Myxinoidea). J. Comp. Neurol. 353, 464–476. of two freshwater turtle species. Resp. Physiol. 4, 1–14. Breipohl, W., Moulton, D., Ummels, M., and Matulionis, D. H. Berghard, A. and Dryer, L. 1998. A novel family of ancient 1982. Spatial pattern of sensory cell terminals in the olfactory vertebrate odorant receptors. J. Neurobiol. 37, 383–392. sac of the tiger salamander. I: A scanning electron microscope Bergstedt, R. A. and Seelye, J. G. 1995. Evidence for lack of study. J. Anat. 134, 757–769. homing by sea lampreys. Trans. Am. Fish. Soc. 124, 235–239. Bronshtein, A. A. and Ivanov, V. P. 1965. Electron microscopic Bertmar, G. 1981. Evolution of vomeronasal organs in verte- investigation of the olfactory organ in the lamprey. Zh. brates. Evolution 35, 359–366. Evolut. Biochem. Physiol. 1, 251–261. Bielek, E. 2005. Development of the endoplasmic reticulum in Brookover, C. and Jackson, T. S. 1911. The olfactory nerve and the rodlet cell of two teleost species. Anat. Rec. A 283, the nervus terminalis of Ameiurus. J. Comp. Neurol. 22, 239–249. 237–259. Bignetti, E., Damiani, G., De Negri, P., et al. 1987. Specificity of Brown, H. E. and Beidler, L. M. 1966. The fine structure of the an immunoaffinity column for odorant-binding protein from olfactory tissue in the black vulture. Fed. Proc. 25, 329. bovine nasal mucosa. Chem. Senses 12, 601–608. Bruce, L. L. and Neary, T. J. 1995. The limbic system of tetra- Biju, K. C., Singru, P. S., Schreibman, M. P., and Subhedar, N. pods: A comparative analysis of cortical and amygdalar 2003. Reproduction phase-related expression of GnRH-like populations. Brain Behav. Evol. 46, 224–234. immunoreactivity in the olfactory receptor neurons, their Bullock, T. H. and Northcutt, R. G. 1984. Nervus terminalis in projections to the olfactory bulb and in the nervus terminalis dogfish (Squalus acanthias, Elasmobranchii) carries tonic in the female Indian major carp Cirrhinus mrigala (Ham.). efferent impulses. Neurosci. Lett. 44, 155–160. Gen. Comp. Endocrinol. 133, 358–367. Bulova, S. J. 1997. Conspecific chemical cues influence burrow Biju, K. C., Gaikwad, A., Sarkar, S., Schreibman, M. P., and choice by desert tortoises (Gopherus agassizii). Copeia Subhedar, N. 2005. Ontogeny of GnRH-like immunoreac- 802–810. tive neuronal systems in the forebrain of the Indian major Cabido, C., Gonzalo, A., Galan, P., Martin, J., and Lopez, P. carp, Cirrhinus mrigala. Gen. Comp. Endocrinol. 141, 2004. Chemosensory predator recognition induces defensive 161–171. behavior in the slow-worm (Anguis fragilis). Can. J. Zool. Billo, R. and Wake, M. H. 1987. Tentacle development in Rev. Can. Zool. 82, 510–515. Dermophis mexicanus (Amphibia, Gymnophiona) with an Cairney, J. 1926. A general survey of the forebrain of Sphenodon hypothesis of tentacle origin. J. Morphol. 192, 101–111. punctatum. J. Comp. Neurol. 42, 255–348. Bleckmann, H. and Hofmann, M. H. 1999. Special senses. Cao, Y., Oh, B. C., and Stryer, L. 1998. Cloning and In: Sharks, Skates, and Rays (ed. W. C. Hamlett), localization of two multigene receptor families in goldfish pp. 300–328. The John Hopkins University Press. olfactory epithelium. Proc. Natl. Acad. Sci. USA 95, Bloom, G. 1954. Studies on the olfactory epithelium of the frog 11987–11992. and toad with the aid of the light and electron microscope. Cashner, M. F. 2004. Are spotted bass (Micropterus punctulatus) Z. Zellforsch. Mikrosk. Anat. 41, 89–100. attracted to Schreckstoff? A test of the predator attraction Bonadonna, F. and Bretagnolle, V. 2002. Smelling home: A good hypothesis. Copeia 592–598. solution for burrow-finding in nocturnal petrels? J. Exp. Biol. Catois, E. M. 1902. Recherches sur l’histologie et l’anatomie 205, 2519–2523. microscopique de l’e´ncephale chez les poissons. Bull. Sci. Bonadonna, F. and Nevitt, G. A. 2004. Partner-specific odor France Belg. 36, 1–166. recognition in an Antarctic seabird. Science 306, 835–835. Chiba, A. 1997. Distribution of neuropeptide Y-like immunor- Bonadonna, F., Spaggiari, J., and Weimerskirch, H. 2001. Could eactivity in the brain of the bichir, Polypterus senegalus, with osmotaxis explain the ability of blue petrels to return to their special regard to the terminal nerve. Cell Tissue Res. 289, burrows at night? J. Exp. Biol. 204, 1485–1489. 275–284. Bonadonna, F., Hesters, F., and Jouventin, P. 2003a. Scent of a Chiba, A. 1999. Immunohistochemical distribution of neuropep- nest: Discrimination of own-nest odours in Antarctic prions, tide Y-related substance in the brain and hypophysis of the Pachyptila desolata. Behav. Ecol. Sociobiol. 54, 174–178. arctic lamprey, Lethenteron japonica. Brain Behav. Evol. 53, Bonadonna, F., Cunningham, G. B., Jouventin, P., Hesters, F., 102–109. and Nevitt, G. A. 2003b. Evidence for nest-odour recognition Chiba, A. 2000. Immunohistochemical cell types in the in two species of diving petrel. J. Exp. Biol. 206, 3719–3722. terminal nerve ganglion of the cloudy dogfish, Scyliorhinus Evolution of Vertebrate Olfactory Subsystems 393

torazame, with special regard to neuropeptide Y/FMRFamide- Crim, J. W., Urano, A., and Gorbman, A. 1979a. immunoreactive cells. Neurosci. Lett. 286, 195–198. Immunocytochemical studies of luteinizing hormone-releasing Chiba, A. 2005. Neuropeptide Y-immunoreactive (NPY-ir) struc- hormone in brains of agnathan fishes. II: Patterns of immu- tures in the brain of the gar Lepisosteus oculatus noreactivity in larval and maturing Western brook lamprey (Lepisosteiformes, Osteichthyes) with special regard to their (Lampetra richardsoni). Gen. Comp. Endocrinol. 38, anatomical relations to gonadotropin-releasing hormone 290–299. (GnRH)-ir structures in the hypothalamus and the terminal Crim, J. W., Urano, A., and Gorbman, A. 1979b. nerve. Gen. Comp. Endocrinol. 142, 336–346. Immunocytochemical studies of luteinizing hormone-releasing Chiba, A., Oka, S., and Honma, Y. 1991. Immunocytochemical hormone in brains of agnathan fishes. I: Comparisons of distribution of FMRFamide-like substance in the brain of the adult Pacific lamprey (Entosphenus tridentata) and the cloudy dogfish, Scyliorhinus torazame. Cell Tissue Res. 265, Pacific hagfish (Eptatretus stouti). Gen. Comp. Endocrinol. 243–250. 37, 294–305. Chiszar, D., Walters, A., Urbaniak, J., Smith, H. M., and Crosby, E. C. 1917. The forebrain of Alligator mississippiensis. J. Mackessy, S. P. 1999. Discrimination between envenomated Comp. Neurol. 27, 325–402. and nonenvenomated prey by western diamondback rattle- D’Aniello, B., Pinelli, C., King, J. A., and Rastogi, R. K. 1994a. snakes (Crotalus atrox): Chemosensory consequences of Neuroanatomical organization of GnRH neuronal systems in venom. Copeia 640–648. the lizard (Podarcis sicula) brain during development. Brain Chivers, D. P., Brown, G. E., and Smith, R. J. F. 1996a. The Res. 657, 221–226. evolution of chemical alarm signals: Attracting predators ben- D’Aniello, B., Masucci, M., di Meglio, M., Iela, L., and efits alarm signal senders. Am. Nat. 148, 649–659. Rastogi, R. K. 1994b. Immunohistochemical localization of Chivers, D. P., Kiesecker, J. M., Anderson, M. T., Wildy, E. L., GnRH in the crested newt (Triturus carnifex)brainandterminal and Blaustein, A. R. 1996b. Avoidance response of a terres- nerve: A developmental study. J. Neuroendocrinol. 6, 167–172. trial salamander (Ambystoma macrodactylum) to chemical D’Aniello, B., Pinelli, C., Di Fiore, M. M., Tela, L., King, J. A., alarm cues. J. Chem. Ecol. 22, 1709–1716. and Rastogi, R. K. 1995. Development and distribution of Chivers, D. P., Kiesecker, J. M., Wildy, E. L., Anderson, M. T., gonadotropin-releasing hormone neuronal systems in the frog and Blaustein, A. R. 1997. Chemical alarm signalling in ter- (Rana esculenta) brain: Immunohistochemical analysis. Brain restrial salamanders: Intra- and interspecific responses. Res. Dev. Brain Res. 89, 281–288. Ethology 103, 599–613. D’Aniello, B., Vallarino, M., Pinelli, C., Fiorentino, M., and Chivers, D. P., Kiesecker, J. M., Marco, A., DeVito, J., Rastogi, R. K. 1996a. Neuropeptide Y: Localization in the Anderson, M. T., and Blaustein, A. R. 2001. Predator-induced brain and pituitary of the developing frog (Rana esculenta). life-history changes in amphibians: Egg predation induces Cell Tissue Res. 285, 253–259. hatching. Oikos 92, 135–142. D’Aniello, B., Fiorentino, M., Pinelli, C., di Meglio, M., Chkheidze, D. D. and Belekhova, M. G. 2005. Projections of Vallarino, M., and Rastogi, R. K. 1996b. Distribution of olfactory bulbs and non-olfactory telencephalic structures in FMRFamide-like immunoreactivity in the brain and pituitary amygdaloid complex of the turtle Testudo horsfieldi: A study of Rana esculenta during development. Brain Res. Dev. Brain using anterograde tracer technique. J. Evol. Biochem. Physiol. Res. 95, 194–204. 41, 561–569. D’Aniello, B., Pinelli, C., Jadhao, A. G., Rastogi, R. K., and Clack, J. A. 2002. Gaining Ground: The Origin and Early Meyer, D. L. 1999. Comparative analysis of FMRFamide- Evolution of Tetrapods. Indiana University Press. like immunoreactivity in caiman (Caiman crocodilus) and Clark, L. 1991. The nest protection hypothesis: The adaptive use turtle (Trachemys scripta elegans) brains. Cell Tissue Res. of plant secondary compounds by European starlings. 298, 549–559. In: Bird–Parasite Interactions: Ecology, Evolution and D’Aniello, B., Fiorentino, M., Pinelli, C., Guarino, F. M., Behaviour (eds. J. E. Loye and M. Zuk), pp. 205–221. Angelini, F., and Rastogi, R. K. 2001. Localization of Oxford University Press. FMRFamide-like immunoreactivity in the brain of the vivipar- Clark, L. and Mason, J. R. 1985. Use of nest material as insecti- ous skink (Chalcides chalcides). Brain Behav. Evol. 57, 18–32. cidal and anti-pathogenic agents by the European starling. Dawley, E. M. 1998. Species, sex, and seasonal differences in Oecologia 67, 169–176. VNO size. Microsc. Res. Tech. 41, 506–518. Clark, L. and Mason, J. R. 1987. Olfactory discrimination of plant Dawley, E. M., Fingerlin, A., Hwang, D., John, S. S., and volatiles by the European starling. Anim. Behav. 35, 227–235. Stankiewicz, C. A. 2000. Seasonal cell proliferation in the Compagno, L. J. V. and Roberts, T. R. 1982. Fresh water stin- chemosensory epithelium and brain of red-backed salaman- grays (Dasyatidae) of Southeast Asia and New Guinea, with a ders, Plethodon cinereus. Brain Behav. Evol. 56, 1–13. description of a new species of Himantura and reports of De Carlos, J. A., Lo´pez-Mascaraque, L., and Valverde, F. 1989. unidentified species. Env. Biol. Fishes 7, 321–339. Connections of the olfactory bulb and nucleus olfactorius Constantino, M. A. and Salmon, M. 2003. Role of chemical and anterior in the hedgehog (Erinaceus europaeus): Fluorescent visual cues in food recognition by leatherback posthatchlings tracers and HRP study. J. Comp. Neurol. 279, 601–618. (Dermochelys coriacea L.). Zoology 106, 173–181. De Martini, E. E. 1978. Copulation and possible olfaction- Cooper, R. S. 1943. An experimental study of the development of mediated pair formation in two species of carcharinid sharks. the larval olfactory organ of Rana pipiens Schreber. J. Exp. Copeia 3, 539–542. Zool. 93, 415–451. Delay, R. J., Dubin, A. E., and Dionne, V. E. 1997. A cyclic Cooper, W. E., Lopez, P., and Salvador, A. 1994. Pheromone detec- nucleotide-dependent chloride conductance in olfactory tion by an amphisbaenian. Anim. Behav. 47, 1401–1411. receptor neurons. J. Mem. Biol. 159, 53–60. Cooper, W. E., Ferguson, G. W., and Habegger, J. J. 2001. Delfino, G., Cerbai, P., and Calloni, C. 1990. Apical specializa- Responses to animal and plant chemicals by several iguanian tions of receptor cells in the olfactory epithelium of Testudo insectivores and the tuatara, Sphenodon punctatus. hermanni (Gmelin) (Reptilia, Testudinidae). Z. Mikrosk. J. Herpetol. 35, 255–263. Anat. Forsch. 104, 689–704. 394 Evolution of Vertebrate Olfactory Subsystems

Demski, L. S. and Northcutt, R. G. 1983. The terminal nerve: A Eisthen, H. L. 1997. Evolution of vertebrate olfactory systems. new chemosensory system in vertebrates? Science 220, Brain Behav. Evol. 50, 222–233. 435–437. Eisthen, H. L. 2000. Presence of the vomeronasal system in Demski, L. S. and Schwanzel-Fukuda, M. (eds.) 1987. The aquatic salamanders. Philos. Trans. R. Soc. Lond. B 355, Terminal Nerve (Nervus Terminalis): Structure, Function, 1209–1213. and Evolution. New York Academy of Sciences. Eisthen, H. L. 2002. Why are olfactory systems of different Derivot, J. H. 1984. Functional anatomy of the peripheral olfac- animals so similar? Brain Behav. Evol. 59, 273–293. tory system of the African lungfish Protopterus annectans Eisthen, H. L. 2004. The goldfish knows: Olfactory receptor cell Owen: Macroscopic, microscopic, and morphometric aspects. morphology predicts receptor gene expression. J. Comp. Am. J. Anat. 169, 177–192. Neurol. 477, 341–346. Derivot, J. H., Mattei, X., Godet, R., and Dupe´, M. 1979. Etude Eisthen, H. L. and Northcutt, R. G. 1996. Silver lampreys ultrastructurale de la re´gion apicale des cellules de (Ichthyomyzon unicuspis) lack a gonadotropin-releasing hor- l’e´pithe´lium olfactif de Protopterus annectens Owen mone- and FMRFamide-immunoreactive terminal nerve. (Dipneustes). J. Ultrastruct. Res. 66, 22–31. J. Comp. Neurol. 370, 159–172. Dial, B. E. and Schwenk, K. 1996. Olfaction and predator detec- Eisthen, H. L., Sengelaub, D. R., Schroeder, D. M., and tion in Coleonyx brevis (Squamata: Eublepharidae), with Alberts, J. R. 1994. Anatomy and forebrain projections of comments on the functional significance of buccal pulsing in the olfactory and vomeronasal organs in axolotls geckos. J. Exp. Zool. 276, 415–424. (Ambystoma mexicanum). Brain Behav. Evol. 44, 108–124. Dionne, V. E. 1992. Chemosensory responses in isolated olfac- Eisthen, H. L., Delay, R. J., Wirsig-Wiechmann, C. R., and tory receptor neurons from Necturus maculosus. J. Gen. Dionne, V. E. 2000. Neuromodulatory effects of gonadotro- Physiol. 99, 415–433. pin releasing hormone on olfactory receptor neurons. Dionne, V. E. and Dubin, A. E. 1994. Transduction diversity in J. Neurosci. 20, 3947–3955. olfaction. J. Exp. Biol. 194, 1–21. Ekstro¨ m, P., Honkanen, T., and Ebbesson, S. O. 1988. Døving, A., Dubois-Dauphin, M., Holley, A., and Jourdan, F. 1977. FMRFamide-like immunoreactive neurons of the nervus ter- Functional anatomy in the olfactory organ of fish and the ciliar minalis of teleosts innervate both retina and pineal organ. mechanism of water transport. Acta Zool. 58, 245–255. Brain Res. 460, 68–75. Døving, K. B. and Holmberg, K. 1974. A note on the function of Erickson, J. R. and Caprio, J. 1984. The spatial distribution of the olfactory organ of the hagfish Myxine glutinosa. Acta ciliated and microvillous olfactory receptor neurons in the Physiol. Scand. 91, 430–432. channel catfish is not matched by a differential specificity to Døving, K. B., Hamdani, E. H., Ho¨ gland, E., Kasumyan, A., and amino acid and bile salt stimuli. Chem. Senses 9, 127–141. Tuvikene, A. O. 2005. Review of the chemical and physiolo- Fadool, D. A., Wachiowiak, M., and Brann, J. H. 2001. Patch- gical bassis of alarm reactions in cyprinids. In: Fish clamp analysis of voltage-activated and chemically activated Chemosenses (eds. K. Reutter and B. G. Kapoor), currents in the vomeronasal organ of Sternotherus odoratus pp. 133–163. Science Publishers. (stinkpot/musk turtle). J. Exp. Biol. 204, 4199–4212. Downes, S. J. 2002. Does responsiveness to predator scents affect Farbman, A. I. and Gesteland, R. C. 1974. Fine structure of the lizard survivorship? Behav. Ecol. Sociobiol. 52, 38–42. olfactory epithelium in the mud puppy, Necturus maculosus. Drenckhahn, D. 1970. Untersuchungen an regio olfactoria und Am. J. Anat. 139, 227–244. nervus olfactorius der Silbermo¨ ve (Larus argentatus). Z. Fine, J. M., Vrieze, L. A., and Sorensen, P. W. 2004. Evidence that Zellforsch. Mikrosk. Anat 106, 119–142. petromyzontid lampreys employ a common migratory pher- Dryer, L. and Graziadei, P. P. C. 1993. A pilot study on morpho- omone that is partially comprised of bile acids. J. Chem. Ecol. logical compartmentalization and heterogeneity in the 30, 2091–2110. elasmobranch olfactory bulb. Anat. Embryol. 188, 41–51. Fiorentino, M., D’Aniello, B., Joss, J., Polese, G., and Rastogi, R. K. Dryer, L. and Graziadei, P. P. C. 1994. Projections of the olfactory 2002. Ontogenetic organization of the FMRFamide immunor- bulb in an elasmobranch fish, Sphyrna tiburo:Segregationof eactivity in the nervus terminalis of the lungfish, Neoceratodus inputs in the telencephalon. Anat. Embryol. 190, 563–572. forsteri. J. Comp. Neurol. 450, 115–121. Dryer, L. and Graziadei, P. P. 1996. Synaptology of the olfactory Firestein, S. 2001. How the olfactory system makes sense of bulb of an elasmobranch fish, Sphyrna tiburo. Anat. Embryol. scents. Nature 413, 211–218. 193, 101–114. Fishelson, L. 1995. Comparative morphology and cytology of the Dulac, C. and Axel, R. 1995. A novel family of genes encoding olfactory organs in moray eels with remarks on their foraging putative pheromone receptors in mammals. Cell 83, 195–206. behavior. Anat. Rec. 243, 403–412. Dulka, J. G. 1993. Sex pheromone systems in goldfish: Fishelson, L. and Baranes, A. 1997. Ontogenesis and cytomor- Comparisons to vomeronasal systems in tetrapods. Brain phology of the nasal olfactory organs in the Oman shark, Iago Behav. Evol. 42, 265–280. omanensis (Triakidae), in the Gulf of Aqaba, Red Sea. Anat. Ebbesson, S. O. E. and Heimer, L. 1970. Projections of the Rec. 249, 409–421. olfactory tract fibers in the nurse shark (Ginglymostoma cir- Flowers, M. A. and Graves, B. M. 1997. Juvenile toads avoid ratum). Brain Res. 17, 47–55. chemical cues from snake predators. Anim. Behav. 53, Edinger, L. 1904. Bau der nervo¨ sen Zentralorgane des Menschen 641–646. und der Tiere. Vogel. Folgueira, M., Anado´n, R., and Yanez, J. 2004. An experimental Ehrenfeld, J. G. and Ehrenfeld, D. W. 1973. Externally secreting study of the connections of the telencephalon in the rainbow glands of freshwater and sea turtles. Copeia 1973, 305–314. trout (Oncorhynchus mykiss). I: Olfactory bulb and ventral Eisner, T., Conner, W. E., Hicks, K., et al. 1977. Stink of stinkpot area. J. Comp. Neurol. 480, 180–203. turtle identified: o-Phenylalkanoic acids. Science 196, Font, E. and Desfilis, E. 2002. Chemosensory recognition of 1347–1349. familiar and unfamiliar conspecifics by juveniles of the Eisthen, H. L. 1992. Phylogeny of the vomeronasal system and of Iberian wall lizard Podarcis hispanica. Ethology 108, 319–330. receptor cell types in the olfactory and vomeronasal epithelia Forlano, P. M., Maruska, K. P., Sower, S. A., King, J. A., and of vertebrates. Microsc. Res. Tech. 23, 1–21. Tricas, T. C. 2000. Differential distribution of gonadotropin- Evolution of Vertebrate Olfactory Subsystems 395

releasing hormone-immunoreactive neurons in the stingray hispanica: An electron microscopic and Golgi study. brain: Functional and evolutionary considerations. Gen. J. Hirnforsch. 27, 87–100. Comp. Endocrinol. 118, 226–248. Garrett, C. M., Boyer, D. M., Card, W. C., Roberts, D. T., Franceschini, F., Sbarbati, A., and Zancanaro, C. 1991. The Murphy, J. B., and Chiszar, D. 1996. Comparison of chemo- vomeronasal organ in the frog, Rana esculenta. An electron sensory behavior and prey trail-following in the varanoid microscopy study. J. Submicr. Cytol. Pathol. 23, 221–231. lizards Varanus gouldii and Heloderma suspectum. Zoo Franceschini, V. and Ciani, F. 1993. Lectin binding to olfactory Biol. 15, 255–265. system in a shark, Scyliorhinus canicula. Folia Histochem. Gerlach, J. 2005. The complex vomeronasal structure of Cytobiol. 31, 133–137. Dipsochelys giant tortoises and its identification as a true Freitag, J., Krieger, J., Strotmann, J., and Breer, H. 1995. Two Jacobson’s organ. Herpetol. J. 15, 15–20. classes of olfactory receptors in Xenopus laevis. Neuron 15, Getchell, M. L. and Getchell, T. V. 1992. Fine structural aspects 1383–1392. of secretion and extrinsic innervation in the olfactory mucosa. Freitag, J., Ludwig, G., Andreini, I., Ro¨ ssler, P., and Breer, H. Microsc. Res. Tech. 23, 111–127. 1998. Olfactory receptors in aquatic and terrestrial verte- Getchell, T. V., Su, Z., and Getchell, M. L. 1993. Mucous brates. J. Comp. Physiol. A 183, 635–650. domains: Microchemical heterogeneity in the mucociliary Freitag, J., Beck, A., Ludwig, G., von Buchholtz, L., and Breer, H. complex of the olfactory epithelium. Ciba Foundation Symp. 1999. On the origin of the olfactory receptor family: Receptor 179, 27–40. genes of the jawless fish (Lampetra fluviatilis). Gene 226, Goh, Y. and Tamura, T. 1978. The electrical responses of the 165–174. olfactory tract to amino acids in carp. Bull. Jpn. Soc. Sci. Fish. Friedrich, R. W. and Korsching, S. I. 1998. Chemotopic, combi- 44, 341–344. natorial, and noncombinatorial odorant representations in Gong, Q. and Shipley, M. T. 1995. Evidence that pioneer the olfactory bulb revealed using a voltage-sensitive axon olfactory axons regulate telencephalon cell cycle kinetics tracer. J. Neurosci. 18, 9977–9988. to induce the formation of the olfactory bulb. Neuron 14, Friedrich, R. W. and Laurent, G. 2001. Dynamic optimization of 91–101. odor representations by slow temporal patterning of mitral Grassman, M. 1993. Chemosensory orientation behavior in juve- cell activity. Science 291, 889–894. nile sea turtles. Brain Behav. Evol. 41, 224–228. Fritsch, G. 1878.Untersuchungen u¨ ber den feineren bau des fisch- Grassman, M. and Owens, D. 1987. Chemosensory imprinting in gehirns mit besonderer beru¨ cksichtigung der homologien bei juvenile green sea turtles, Chelonia mydas. Anim. Behav. 35, anderen wirbelthierklassen. Verlag der Gutmann’schen 929–931. Buchhandlung. Grassman, M. A., Owens, D. W., McVey, J. P., and Frontini, A., Zaidi, A. U., Hua, H., et al. 2003. Glomerular Marquez, M. R. 1984. Olfactory-based orientation in artifi- territories in the olfactory bulb from the larval stage of the cially imprinted sea turtles. Science 224, 83–84. sea lamprey Petromyzon marinus. J. Comp. Neurol. 465, Graves, B. M. and Halpern, M. 1989. Chemical access to the 27–37. vomeronasal organs of the lizard Chalcides ocellatus. J. Exp. Frost, D. R., Grant, T., Faivovich, J. N., et al. 2006. The amphi- Zool. 249, 150–157. bian tree of life. Bull. Am. Mus. Natl. Hist. 297, 1–370. Graves, B. M. and Halpern, M. 1990. Roles of vomeronasal Fujii, K. and Kobayashi, H. 1992. FMRFamide-like immunor- chemoreception in tongue-flicking, exploratory and feeding eactivity in the brain of the Japanese quail, Coturnix coturnix behaviour of the lizard, Chalcides ocellatus. Anim. Behav. 39, japonica. Neuroendocrinol. Lett. 14, 97–102. 692–698. Fujita, I., Sorensen, P. W., Stacey, N. E., and Hara, T. J. 1991. Graves, G. R. 1992. The greater yellow-headed vulture The olfactory system, not the terminal nerve, functions as the (Cathartes melambrotus) locates food by olfaction. J. Raptor primary chemosensory pathway mediating responses to sex Res. 26, 38–39. pheromones in the male goldfish. Brain Behav. Evol. 38, Graziadei, P. P. C. and Bannister, L. H. 1967. Some observations 313–321. on the fine structure of the olfactory epithelium in the domes- Fuller, C. L. and Byrd, C. A. 2005. Ruffed cells identified in the tic duck. Z. Zellforsch. Mikrosk. Anat. 80, 220–228. adult zebrafish olfactory bulb. Neurosic. Lett. 379, 190–194. Graziadei, P. P. C. and Monti-Graziadei, A. G. 1992. The Furry, K., Swain, T., and Chiszar, D. 1991. Strike-induced che- influence of the olfactory placode on the development of mosensory searching and trail following by prairie the telencephalon in Xenopus laevis. Neuroscience 46, rattlesnakes (Crotalus viridis) preying upon deer mice 617–629. (Peromyscus maniculatus): Chemical discrimination among Graziadei, P. P. C. and Tucker, D. 1970. Vomeronasal receptors individual mice. Herpetologica 47, 69–78. in turtles. Z. Zellforsch. Mikrosk. Anat. 105, 498–514. Gaillard, I., Rouquier, S., and Giorgi, D. 2004. Olfactory recep- Greene, C. W. 1925. Notes on the olfactory and other physiolo- tors. Cell Mol. Life Sci. 61, 456–469. gical reactions of the California hagfish. Science 61, 68–70. Gamble, H. J. 1956. An experimental study of the secondary Greene, M. J. and Mason, R. T. 2003. Pheromonal inhibition of olfactory connexions in Testudo graeca. J. Anat. 90, 15–29. male courtship behaviour in the brown tree snake, Boiga Gans, C. and Clark, B. 1976. Studies on ventilation of Caiman irregularis: A mechanism for the rejection of potential mates. crocodilus (Crocodilia: Reptilia). Resp. Physiol. 26, Anim. Behav. 65, 905–910. 285–301. Hagelin, J. C. 2004. Observations on the olfactory ability of the Gans, C. and Northcutt, R. G. 1983. Neural crest and the origin Kakapo Strigops habroptilus, the critically endangered parrot of vertebrates: A new head. Science 220, 268–274. of N. Z. Ibis 146, 161–164. Garcia-Rubio, S., Attygalle, A. B., Weldon, P. J., and Hagelin, J. C., Jones, I. L., and Rasmussen, L. E. L. 2003. A Meinwald, J. 2002. Reptilian chemistry: Volatile compounds tangerine-scented social odour in a monogamous seabird. from paracloacal glands of the American crocodile Proc. R. Soc. Lond. Ser. B 270, 1323–1329. (Crocodylus acutus). J. Chem. Ecol. 28, 769–781. Hagino-Yamagishi, K., Moriya, K., Kubo, H., et al. 2004. Garcı´a-Verdugo, J. M., Llahi, S., Farin˜ as, I., and Martı´n, V. 1986. Expression of vomeronasal receptor genes in Xenopus laevis. Laminar organization of the main olfactory bulb of Podarcis J. Comp. Neurol. 472, 246–256. 396 Evolution of Vertebrate Olfactory Subsystems

Halpern, M. 1976. The efferent connections of the olfactory bulb organized and sexually dimorphic distribution. Cell 90, and accessory olfactory bulb in the snakes, Thamnophis sirta- 763–773. lis and Thamnophis radix. J. Morphol. 150, 553–578. Herrick, C. J. 1921. The connections of the vomeronasal nerve, Halpern, M. and Borghjid, S. 1997. Sublingual plicae (anterior accessory olfactory bulb, and amygdala in Amphibia. processes) are not necessary for garter snake vomeronasal J. Comp. Neurol. 33, 213–280. function. J. Comp. Psychol. 111, 302–306. Herrick, C. J. 1924. The amphibian forebrain. II: The olfactory Halpern, M. and Kubie, J. L. 1980. Chemical access to the bulb of Amblystoma. J. Comp. Neurol. 37, 373–396. vomeronasal organs of garter snakes. Physiol. Behav. 24, Herrick, C. J. 1931. The amphibian forebrain. V: The olfactory 367–371. bulb of Necturus. J. Comp. Neurol. 53, 55–69. Halpern, M. and Martı´nez-Marcos, A. 2003. Structure and func- Herrick, C. J. 1948. The Brain of the Tiger Salamander. tion of the vomeronasal system: An update. Prog. Neurobiol. University of Chicago Press. 70, 245–318. Hildebrand, J. G. and Shepherd, G. M. 1997. Mechanisms of Halpern, M., Halpern, J., Erichsen, E., and Borghjid, S. 1997. olfactory discrimination: Converging evidence for common The role of nasal chemical senses in garter snake response to principles across phyla. Ann. Rev. Neurosci. 20, 595–631. airborne odor cues from prey. J. Comp. Psychol. 111, Hillier, L. W., Miller, W., Birney, E., et al. 2004. Sequence and 251–260. comparative analysis of the chicken genome provide unique Hamdani, E. H., Alexander, G., and Døving, K. B. 2001. perspectives on vertebrate evolution. Nature 432, 695–716. Projection of sensory neurons with microvilli to the lateral Himstedt, W. and Simon, D. 1995. Sensory basis of foraging olfactory tract indicates their participation in feeding beha- behaviour in caecilians. Herpetol. J. 5, 266–270. viour in crucian carp. Chem. Senses 26, 1139–1144. Hodgson, E. S. and Mathewson, R. F. 1971. Chemosensory Hansen, A. and Finger, T. E. 2000. Phyletic distribution of crypt- orientation in sharks. Ann. NY Acad. Sci. 188, 175–182. type olfactory receptor neurons in fishes. Brain Behav. Evol. Hodgson, E. S. and Mathewson, R. F. 1978. Electrophysiological 55, 100–110. studies of chemoreception in elasmobranchs. In: Sensory Hansen, A. and Zeiske, E. 1998. The peripheral olfactory organ Biology of Sharks, Skate, and Rays (eds. E. S. Hodgson and of the zebrafish, Danio rerio: An ultrastructural study. Chem. R. F. Mathewson), pp. 269–329. Office of Naval Research Senses 23, 39–48. Department of the Navy. Hansen, A., Reiss, J. O., Gentry, C. L., and Burd, G. D. 1998. Hoffman, H. H. 1963. The olfactory bulb, accessory olfactory Ultrastructure of the olfactory organ in the clawed frog, bulb, and hemisphere of some anurans. J. Comp. Neurol. 120, Xenopus laevis, during larval development and metamorpho- 317–368. sis. J. Comp. Neurol. 398, 273–288. Hofmann, M. H. and Meyer, D. L. 1989. Central projections of Hansen, A., Rolen, S. H., Anderson, K., Morita, Y., Caprio, J., the nervus terminalis in four species of amphibians. Brain and Finger, T. E. 2003. Correlation between olfactory recep- Behav. Evol. 34, 301–307. tor cell type and function in the channel catfish. J. Neurosci. Hofmann, M. H. and Meyer, D. L. 1991a. Subdivisions of the 23, 9328–9339. terminal nerve in Xenopus laevis. J. Exp. Zool. 259, Hansen, A., Anderson, K. T., and Finger, T. E. 2004. Differential 324–329. distribution of olfactory receptor neurons in goldfish: Hofmann, M. H. and Meyer, D. L. 1991b. Functional subdivi- Structural and molecular correlates. J. Comp. Neurol. 477, sions of the olfactory system correlate with lectin-binding 347–359. properties in Xenopus. Brain Res. 564, 344–347. Hanson, L. R., Sorensen, P. W., and Cohen, Y. 1998. Sex pher- Hofmann, M. H. and Meyer, D. L. 1992. Peripheral origin of omones and amino acids evoke distinctly different spatial olfactory nerve fibers by-passing the olfactory bulb in patterns of electrical activity in the goldfish olfactory bulb. Xenopus laevis. Brain Res. 589, 161–163. Ann. NY Acad. Sci. 855, 521–524. Hofmann, M. H. and Meyer, D. L. 1995. The extrabulbar olfac- Harriman, A. E. and Berger, R. H. 1986. Olfactory acuity in the tory pathway: Primary olfactory fibers bypassing the olfactory common raven (Corvus corax). Physiol. Behav. 36, 257–262. bulb in bony fishes? Brain Behav. Evol. 46, 378–388. Hatanaka, T. and Matsuzaki, O. 1993. Odor responses of the Holl, A. 1973. Feinstruktur des Riechepithels von Chimaera vomeronasal system in Reeve’s turtle, Geoclemys reevesii. monstrosa (Holocephali). Mar. Biol. 23, 59–72. Brain Behav. Evol. 41, 183–186. Holmgren, N. 1919. Zur Anatomie und Histologie des Vorder- Hatanaka, T. and Shibuya, T. 1989. Odor response patterns of und Zwischenhirns der knochenfische. Acta Zool. 60, 1–96. single units in the accessory olfactory bulb of the turtle, Holmgren, N. and van der Horst, C. J. 1925. Contribution to Geoclemys reevesii. Comp. Biochem. Physiol. A 92, the morphology of the brain of Ceratodus. Acta Zool. 6, 505–512. 59–165. Hatanaka, T., Matsuzaki, O., and Shibuya, T. 1982. Fine struc- Honigmann, H. 1921. Zur Biologie der Schildkro¨ ten. Biol. tures of vomeronasal receptor cells in the Reeve’s turtle, Zentralbl. 41, 241–250. Geoclemys reevesii. Zool. Sci. 91, 190–193. Hoppe, G. 1934. Das Geruchsorgan von Hatteria punctata. Haverly, J. E. and Kardong, K. V. 1996. Sensory deprivation Z. Anat. Entwicklungs. 102, 434–461. effects on the predatory behavior of the rattlesnake, Houck, L. D. and Arnold, S. J. 2003. Courtship and mating. Crotalus viridis oreganus. Copeia 419–428. In: Phylogeny and Reproductive Biology of Urodela Head, M. L., Keogh, J. S., and Doughty, P. 2005. Male southern (Amphibia) (ed. D. M. Sever), pp. 383–424. Science water skinks (Eulamprus heatwolei) use both visual and Publishers. chemical cues to detect female sexual receptivity. Acta Houck, L. D. and Reagan, N. L. 1990. Male courtship phero- Ethol. 8, 79–85. mones increase female receptivity in a plethodontid Heier, P. 1948. Fundamental principles in the structure of the salamander. Anim. Behav. 39, 729–734. brain: A study of the brain of Petromyzon fluviatilis. Acta Houston, D. C. 1984. Does the king vulture Sarcorhamphus papa Anat. 5, 1–213. use a sense of smell to locate food? Ibis 126, 67–69. Herrada, G. and Dulac, C. 1997. A novel family of putative Huang, G.-Z., Zhang, J.-J., Wang, D., Mason, R. T., and pheromone receptors in mammals with a topographically Halpern, M. 2006. Female snake sex pheromone induces Evolution of Vertebrate Olfactory Subsystems 397

membrane responses in vomeronasal sensory neurons of male Johnston, J. B. 1915. The cell masses in the forebrain of the turtle, snakes. Chem. Senses. 31, 521–529. Cistudo carolina. J. Comp. Neurol. 25, 393–468. Huang, L. X., Maaswinkel, H., and Li, L. 2005. Olfactoretinal Jones, F. M., Pfeiffer, C. J., and Asashima, M. 1994. centrifugal input modulates zebrafish retinal ganglion cell Ultrastructure of the olfactory organ of the newt, Cynops 2 activity: A possible role for dopamine-mediated Ca þ signal- pyrrhogaster. Ann. Anat. 176, 269–275. ling pathways. J. Physiol. Lond. 569, 939–948. Jones, R. B. and Roper, T. J. 1997. Olfaction in the domestic Hubbard, P. C., Barata, E. N., and Canario, V. M. 2000. fowl: A critical review. Physiol. Behav. 62, 1009–1018. 2 Olfactory sensitivity to changes in environmental [Ca þ] in Jung, Y., Wirkus, E., Amendola, D., and Gomez, G. 2005. the marine teleost Sparus aurata. J. Exp. Biol. 203, Characteristics of odorant elicited calcium fluxes in acutely- 3821–3829. isolated chick olfactory neurons. J. Comp. Physiol. A 191, Huesa, G., Anado´n, R., and Yanez, J. 2000. Olfactory projections 511–520. in a chondrostean fish, Acipenser baeri: An experimental Jørgensen, C. B. 2000. Amphibian respiration and olfaction and study. J. Comp. Neurol. 428, 145–158. their relationships: From Robert Townson (1794) to the pre- Huffman, H. J. H. 1963. The olfactory bulb, accessory olfactory sent. Biol. Rev. Camb. Philos. Soc. 75, 297–345. bulb and hemispheres of some animals. J. Comp. Neurol. 120, Kajiura, S. M., Forni, J. B., and Summers, A. P. 2004a. 317–368. Olfactory morphology of carcharhinid and sphyrnid sharks: Huggins, S. E., Parsons, L. C., and Pen˜ a, R. V. 1968. Further Does the cephalofoil confer a sensory advantage? study of the spontaneous electrical activity of the brain of J. Morphol. 264, 253–263. Caiman sclerops: Olfactory lobes. Physiol. Zool. 41, Kajiura, S. M., Tricas, T. C., Forni, J. B., and Summers, A. P. 371–383. 2004b. Olfaction in sphyrnid sharks. Int. Comp. Biol. 44, Hutchison, L. V. and Wenzel, B. M. 1980. Olfactory guidance in 580. foraging by procellariiforms. Condor 82, 314–319. Kam, Y. C. and Yang, H. W. 2002. Female-offspring communi- Huxley, T. H. 1876. On Ceratodus forsteri, with observation on cation in a Taiwanese tree frog, Chirixalus eiffingeri (Anura: classification of fishes. Proc. Zool. Soc. Lond. 1876, 23–59. Rhacophoridae). Anim. Behav. 64, 881–886. Ibrahim, S. A., Avery, J. W., Weldon, P. J., and Wheeler, J. W. Kauer, J. S. 2002. On the scents of smell in the salamander. 1998. Age-class differences in lipids from the paracloacal Nature 417, 336–342. glands of the American alligator (Alligator mississippiensis). Kaufmann, J. H. 1992. The social behavior of wood turtles, Z. Naturforsch. 53, 201–209. Clemmys insculpta, in central Pennsylvania. Herpetol. Ishii, S., Kubokawa, K., Kikuchi, M., and Nishio, H. 1995. Monogr. 6, 1–25. Orientation of the toad, Bufo japonicus, toward the breeding Kemali, M. and Guglielmotti, V. 1987. A horseradish peroxidase pond. Zool. Sci. 12, 475–484. study of the olfactory system of the frog, Rana esculenta. Iwahori, N., Kiyota, E., and Nakamura, K. 1987. A Golgi study J. Comp. Neurol. 263, 400–417. on the olfactory bulb in the lamprey, Lampetra japonica. Kikuyama, S. and Toyoda, F. 1999. Sodefrin: A novel sex pher- Neurosci. Res. 5, 126–139. omone in a newt. Rev. Reprod. 4, 1–4. Iwahori, N., Nakamura, K., and Mameya, C. 1989a. A Golgi Kikuyama, S., Toyoda, F., Ohmiya, Y., Matsuda, K., Tanaka, S., study on the main olfactory bulb in the snake, Elaphe quad- and Hayashi, H. 1995. Sodefrin: A female-attracting peptide rivirgata. Neurosci. Res. 6, 411–425. pheromone in newt cloacal glands. Science 267, 1643–1645. Iwahori, N., Nakamura, K., and Mameya, C. 1989b. A Golgi Kim, M. H., Oka, Y., Amano, M., et al. 1995. study on the accessory olfactory bulb in the snake, Elaphe Immunocytochemical localization of sGnRH and cGnRH-II quadrivirgata. Neurosci. Res. 7, 55–70. in the brain of goldfish, Carassius auratus. J. Comp. Neurol. Iwahori, N., Baba, J., and Kawawaki, T. 1998. Neuronal orga- 356, 72–82. nization of the olfactory bulb in the hagfish, Eptatretus King, J. C., Sower, S. A., and Anthony, E. L. P. 1988. burgeri: A Golgi study. J. Hirnforsch. 39, 161–173. Neuronal systems immunoreactive with antiserum to lam- Jaeger, R. G. 1981. Dear enemy recognition and the costs prey gonadotropin-releasing hormone in the brain of of aggression between salamanders. Am. Nat. 117, Petromyzon marinus. Cell Tissue Res. 253, 1–8. 962–974. Kleerekoper, H. 1978. Chemoreception and its interaction with Jaeger, R. G. 1986. Pheromonal markers as territorial advertise- flow and light perception in the locomotion and orientation of ment by terrestrial salamanders. In: Chemical Signals in some elasmobranchs. In: Sensory Biology of Sharks, Skate, Vertebrates (eds. D. Duvall, D. Muller-Schwarze, and R. M. and Rays (eds. E. S. Hodgson and R. F. Mathewson), Silverstein), pp. 191–203. Plenum. pp. 269–329. Office of Naval Research Department of the Jansen, J. 1930. The brain of Myxine glutinosa. J. Comp. Neurol. Navy. 49, 359–507. Kleerekoper, H. and Mogensen, J. 1960. Role of olfaction in the Jermakowicz, W. J., 3rd, Dorsey, D. A., Brown, A. L., et al. 2004. orientation of Petromyzon marinus. I: Response to a single Development of the nasal chemosensory organs in two terres- amine in prey’s body odor. Physiol. Zool. 36, 347–360. trial anurans: The directly developing frog, Eleutherodactylus Kleerekoper, H. and Van Erkel, G. A. 1960. The olfactory appa- coqui (Anura: Leptodactylidae), and the metamorphosing ratus of Petromyzon marinus L. Can. J. Zool. 38, 209–223. toad, Bufo americanus (Anura: Bufonidae). J. Morphol. 261, Kobayashi, M., Sorensen, P. W., and Stacey, N. E. 2002. 225–248. Hormonal and pheromonal control of spawning behavior in Jirikowski, G., Erhart, G., Grimmelikhuijzen, C. J., Triepel, J., the goldfish. Fish Physiol. Biochem. 26, 71–84. and Patzner, R. A. 1984. FMRF-amide-like immunoreactivity Kokoros, J. J. and Northcutt, R. G. 1977. Telencephalic efferents in brain and pituitary of the hagfish Eptatretus burgeri of the tiger salamander Ambystoma tigrinum tigrinum (Cyclostomata). Cell Tissue Res. 237, 363–366. (Green). J. Comp. Neurol. 173, 613–628. Johnston, J. B. 1898. The olfactory lobes, forebrain, and habe- Kolnberger, I. 1971. U¨ ber die Zuga¨ngglichkeit des nular tracts of Acipenser. Zool. Bull. 1, 221–241. Interzellularraumes und Zellontakte im riechepithel des Johnston, J. B. 1913. Nervus terminalis in reptiles and mammals. Jabonsonschen organs. Z. Zellforsch. Mikrosk. Anat. 122, J. Comp. Neurol. 23, 97–120. 564–573. 398 Evolution of Vertebrate Olfactory Subsystems

Kolnberger, I. and Altner, H. 1971. Ciliary-structure precursor Lee, K. H., Wells, R. G., and Reed, R. R. 1987. Isolation of an bodies as stable constituents in the sensory cells of vomero- olfactory cDNA: Similarity to retinol-binding protein suggests nasal organ of reptiles and mammals. Z. Zellforsch. Mikrosk. a role in olfaction. Science 235, 1053–1056. Anat. 118, 254–262. Leveteau, J., Andriason, I., and MacLeod, P. 1993. Interbulbar Kosaka, T. and Hama, K. 1979a. New type of neuron with a reciprocal inhibition in frog olfaction. Behav. Brain Res. 54, distinctive axon initial segment. Brain Res. 163, 151–155. 103–106. Kosaka, T. and Hama, K. 1979b. Ruffed cell: A new type of Levine, R. L. and Dethier, S. 1985. The connections between the neuron with a distinctive initial unmyelinated portion of the olfactory bulb and the brain in the goldfish. J. Comp. Neurol. axon in the olfactory-bulb of the goldfish (Carassius aura- 237, 427–444. tus).1: Golgi impregnation and serial thin sectioning studies. Li, W. and Sorensen, P. W. 1997. Highly independent olfactory J. Comp. Neurol. 186, 301–319. receptor sites for naturally occurring bile acids in the sea Kosaka, T. and Hama, K. 1980. Presence of the ruffed cell in the lamprey, Petromyzon marinus. J. Comp. Physiol. A 180, olfactory bulb of the catfish, Parasilurus asotus, and the sea 429–438. eel, Conger myriaster. J. Comp. Neurol. 193, 103–117. Li, W., Sorensen, P. W., and Gallaher, D. D. 1995. The olfactory Kratzing, J. 1975. The fine structure of the olfactory and vomer- system of migratory adult sea lamprey (Petromyzon marinus) onasal organs of a lizard (Tiliqua scincoides scincoides). Cell is specifically and uniquely sensitive to bile acids released by Tissue Res. 156, 239–252. conspecific larvae. J. Gen. Phys. 105, 569–578. Kyle, A. L., Sorensen, P. W., Stacey, N. E., and Dulka, J. G. 1987. Li, W., Scott, A. P., Siefkes, M. J., et al. 2002. Bile acid secreted by Medial olfactory tract pathways controlling sexual reflexes male sea lamprey that acts as a sex pheromone. Science 296, and behavior in teleosts. In: The Terminal Nerve (Nervus 138–141. Terminalis): Structure, Function, and Evolution (eds. L. S. Lipschitz, D. L. and Michel, W. C. 2002. Amino acid odorants Demski and M. Schwanzel-Fukuda), pp. 60–68. New York stimulate microvillar sensory neurons. Chem. Senses 27, Academy of Sciences. 277–286. Labra, A., Brann, J. H., and Fadool, D. A. 2005. Heterogeneity of Liu, W. M., Wang, D., Liu, J. M., Chen, P., and Halpern, M. voltage- and chemosignal-activated response profiles in 1998. Chemosignal transduction in the vomeronasal organ of vomeronasal sensory neurons. J. Neurophysiol. 94, garter snakes: Cloning of a gene encoding adenylate cyclase 2535–2548. from the vomeronasal organ of garter snakes. Arch. Biochem. Lacalli, T. C. 2001. New perspectives on the evolution of proto- Biophys. 358, 204–210. chordate sensory and locomotory systems, and the origin of Llahi, S. and Garcı´a-Verdugo, J. M. 1989. Neuronal organization brains and heads. Philos. Trans. R. Soc. Ser. B 356, of the accessory olfactor bulb of the lizard Podarcis hispanica: 1565–1572. Golgi study. J. Morphol. 202, 13–28. Lacalli, T. C. 2002. Sensory pathways in amphioxus larvae. I: Locy, W. A. 1905. On a newly recognized nerve connected with Constituent fibres of the rostral and anterodorsal nerves, their the fore-brain of selachians. Anat. Anz. 26, 111–123. targets and evolutionary significance. Acta Zool. 83, Lohman, A. H. and Smeets, W. J. A. J. 1993. Overview of the 149–166. main and accessory olfactory bulb projections in reptiles. Lacalli, T. C. and Hou, S. F. 1999. A reexamination of the Brain Behav. Evol. 41, 147–155. epithelial sensory cells of amphioxus (Branchiostoma). Acta Lohman, A. H. M. and Lammers, H. J. 1967. On the structure Zool. 80, 125–134. and fibre connections of the olfactory centers in mammals. Lanuza, E. and Halpern, M. 1997. Afferent and efferent connec- In: Progress in Brain Research (ed. Y. Zotterman), pp. 65–82. tions of the nucleus sphericus in the snake Thamnophis Elsevier. sirtalis: Convergence of olfactory and vomeronasal informa- Lohman, A. H. M., Hoogland, P. V., and Witjes, R. J. G. M. tion in the lateral cortex and amygdala. J. Comp. Neurol. 385, 1988. Projections from the main and accessory olfactory 627–640. bulbs to the amygdaloid complex in the lizard Gekko gecko. Lanuza, E. and Halpern, M. 1998. Efferents and centrifugal In: The Forebrain of Reptiles (eds. W. K. Schwerdtfeger and afferents of the main and accessory olfactory bulbs in the W. J. A. J. Smeets), pp. 41–49. Karger. snake Thamnophis sirtalis. Brain Behav. Evol. 51, 1–22. Lohmann, K. J., Hester, J. T., and Lohmann, C. M. F. 1999. Larsell, O. 1917. Studies on the nervus terminalis: Turtle. Ann. Long-distance navigation in sea turtles. Ethol. Ecol. Evol. Otol. Rhinol. Laryngol. 30, 423–443. 11, 1–23. Laurent, G. 2002. Olfactory network dynamics and the coding of Long, J. E., Garel, S., Depew, M. J., Tobet, S., and multidimensional signals. Nat. Rev. Neurosci. 3, 884–895. Rubenstein, J. L. 2003. DLX5 regulates development of per- Laurila, A., Kujasalo, J., and Ranta, E. 1997. Different antipre- ipheral and central components of the olfactory system. dator behaviour in two anuran tadpoles: Effects of predator J. Neurosci. 23, 568–578. diet. Behav. Ecol. Sociobiol. 40, 329–336. Lopez, P. and Martin, J. 1994. Responses by amphisbaenian Lavin-Murcio, P. A. and Kardong, K. V. 1995. Scents related to Blanus cinereus to chemicals from prey or potentially harmful venom and prey as cues in the poststrike trailing behavior of ant species. J. Chem. Ecol. 20, 1113–1119. rattlesnakes, Crotalus viridis oreganus. Herpetol. J. 51, Lopez, P. and Martin, J. 2001. Chemosensory predator recogni- 39–44. tion induces specific defensive behaviours in a fossorial La´za´r, G., Maderdrut, J. L., Trasti, S. L., et al. 1993. Distribution amphisbaenian. Anim. Behav. 62, 259–264. of proneuropeptide Y-derived peptides in the brain of Rana Lopez, P., Aragon, P., and Martin, J. 2003. Responses of female esculenta and Xenopus laevis. J. Comp. Neurol. 327, lizards, Lacerta monticola, to males’ chemical cues reflect 551–571. their mating preference for older males. Behav. Ecol. Leach, W. J. 1951. The hypophysis of lampreys in relation to the Sociobiol. 55, 73–79. nasal apparatus. J. Morphol. 89, 217–255. Luo, Y., Lu, S., Chen, P., Wang, D., and Halpern, M. 1994. Lebedev, O. A. and Coates, M. I. 1995. The postcranial skeleton Identification of chemoattractant receptors and G-proteins of the Devonian tetrapod Tulerpeton curtum Lebedev. Zool. in the vomeronasal system of garter snakes. J. Biol. Chem. J. Linnean Soc. 114, 307–348. 269, 16867–16877. Evolution of Vertebrate Olfactory Subsystems 399

Lutz, P. L. and Milton, S. L. 2004. Negotiating brain anoxia Matsuzaki, O., Shibuya, T., and Hatanaka, T. 1982. Number survival in the turtle. J. Exp. Biol. 207, 3141–3147. of olfactory receptor cells and fine structure of olfactory Maaswinkel, H. and Li, L. 2003. Olfactory input increases visual nerve fibers in the pigeon, Columba livia. Zool. Mag. 91, sensitivity in zebrafish: A possible function for the terminal 221–229. nerve and dopaminergic interplexiform cells. J. Exp. Biol. McCutcheon, F. H. 1943. The respiratory mechanism in turtles. 206, 2201–2209. Physiol. Zool. 16, 255–269. Magurran, A. E., Irving, P. W., and Henderson, P. A. 1996. Is McKeegan, D. E. F. 2002. Spontaneous and odour evoked activ- there a fish alarm pheromone? A wild study and critique. ity in single avian olfactory bulb neurones. Brain Res. 929, Proc. R. Soc. London Ser. B 263, 1551–1556. 48–58. Mair, R., Gesteland, R. C., and Blank, D. L. 1982. Changes in Medina, M., Reperant, J., Miceli, D., Ward, R., and Arckens, L. morphology and physiology of olfactory receptor cilia during 2005. GnRH-immunoreactive centrifugal visual fibers in the development. Neuroscience 7, 3091–3103. Nile crocodile (Crocodylus niloticus). Brain Res. 1052, Manley, G. A. 2000. Cochlear mechanisms from a phylogenetic 112–117. viewpoint. Proc. Natl. Acad. Sci. USA 97, 11736–11743. Menco, B. P. 1997. Ultrastructural aspects of olfactory signaling. Manley, G. A. 2001. Evidence for an active process and a Chem. Senses 22, 295–311. cochlear amplifier in nonmammals. J. Neurophysiol. 86, Menco, B. P. M. 1980. Qualitative and quantitative freeze- 541–549. fracture studies on olfactory and nasal respiratory structures Marchand, J. E., Yang, X., Chikaraishi, D., Krieger, J., Breer, H., of frog, ox, rat, and dog. I: A general survey. Cell Tissue and Kauer, J. S. 2004. Olfactory receptor gene expression in Res. 207, 183–209. tiger salamander olfactory epithelium. J. Comp. Neurol. 474, Meng, Q. and Yin, M. 1981. A study on the olfactory organ 453–467. of skates, rays, and chimaeras. J. Fish. (China) 5, Marco, A., Chivers, D. P., Kiesecker, J. M., and Blaustein, A. R. 209–228. 1998. Mate choice by chemical cues in Western redback Menzies, J. I. 1993. Systematics of Litoria iris (Anura, Hylidae) (Plethodon vehiculum) and Dunn’s (P. dunni) salamanders. and its allies in New Guinea and a note on sexual dimorphism Ethology 104, 781–788. in the group. Aust. J. Zool. 41, 225–255. Marquis, O., Saglio, P., and Neveu, A. 2004. Effects of predators Meredith, M. and White, J. 1987. Interactions between the olfac- and conspecific chemical cues on the swimming activity of tory system and the terminal nerve. Electrophysiological Rana temporaria and Bufo bufo tadpoles. Arch. Hydrobiol. evidence. In: The Terminal Nerve (Nervus Terminalis): 160, 153–170. Structure, Function, and Evolution (eds. L. S. Demski and Martıńez-Marcos, A., Lanuza, E., and Halpern, M. 1999. M. Schwanzel-Fukuda), pp. 349–368. New York Academy Organization of the ophidian amygdala: Chemosensory path- of Sciences. ways to the hypothalamus. J. Comp. Neurol. 412, 51–68. Meyer, D. L., von Bartheld, C. S., and Lindo¨ rfer, H. W. 1987. Martıńez-Marcos, A., Ubeda-Banon, I., and Halpern, M. 2001. Evidence for the existence of a terminal nerve in lampreys and Neural substrates for tongue-flicking behavior in snakes. J. in birds. In: The Terminal Nerve (Nervus Terminalis): Comp. Neurol. 432, 75–87. Structure, Function, and Evolution (eds. L. S. Demski and Martıńez-Marcos, A., Lanuza, E., and Halpern, M. 2002. Neural M. Schwanzel-Fukuda), pp. 385–391. New York Academy substrates for processing chemosensory information in of Sciences. snakes. Brain Res. Bull. 57, 543–546. Mezler, M., Fleischer, J., Conzelmann, S., et al. 2001. Martıńez-Marcos, A., Ubeda-Banon, I., Lanuza, E., and Identification of a nonmammalian Golf subtype: Functional Halpern, M. 2005a. Chemoarchitecture and afferent connec- role in olfactory signaling of airborne odorants in Xenopus tions of the ‘olfactostriatum’: A specialized vomeronasal laevis. J. Comp. Neurol. 439, 400–410. structure within the basal ganglia of snakes. J. Chem. Miller, L. R. and Gutzke, W. H. N. 1999. The role of the vomer- Neuroanat. 29, 49–69. onasal organ of crotalines (Reptilia: Serpentes: Viperidae) in Martıńez-Marcos, A., Ubeda-Banon, I., Lanuza, E., and predator detection. Anim. Behav. 58, 53–57. Halpern, M. 2005b. Efferent connections of the ‘olfactos- Millery, J., Briand, L., Bezirard, V., et al. 2005. Specific triatum’: A specialized vomeronasal structure within the expression of olfactory binding protein in the aerial olfac- basal ganglia of snakes. J. Chem. Neuroanat. 29, 217–226. tory cavity of adult and developing Xenopus. Eur. J. Mason, R. T., Wang, D., Chen, P., and Halpern, M. 2006. Neurosci. 22, 1389–1399. Evidence for a perireceptor role for Harderian gland secre- Moeller, J. F. and Meredith, M. 1998. Increase in gonadotropin- tions in garter snakes: Delivery of pheromone molecules to the releasing hormone (GnRH) levels in CSF after stimulation of vomeronasal organ. Chem. Senses 31, A89–A90. the nervus terminalis in Atlantic stingray, Dasyatis sabina. Masucci, M., D’Aniello, B., Iela, L., Ciarcia, G., and Brain Res. 806, 104–107. Rastogi, R. K. 1992. Immunohistochemical demonstration Mombaerts, P. 2004. Genes and ligands for odorant, vomerona- of the presence and localization of diverse molecular forms sal and taste receptors. Nat. Rev. Neurosci. 5, 263–278. of a gonadotropin-releasing hormone in the lizard (Podarcis s. Mombaerts, P., Wang, F., Dulac, C., et al. 1996. Visualizing an sicula) brain. Gen. Comp. Endocrinol. 86, 81–89. olfactory sensory map. Cell 87, 675–686. Mathieu, M., Tagliafierro, G., Bruzzone, F., and Vallarino, M. Moreno, N. and Gonzalez, A. 2003. Hodological characteriza- 2002. Neuropeptide tyrosine-like immunoreactive system in the tion of the medial amygdala in anuran amphibians. J. Comp. brain, olfactory organ and retina of the zebrafish, Danio rerio, Neurol. 466, 389–408. during development. Brain Res. Dev. Brain Res. 139, 255–265. Moreno, N., Morona, R., Lopez, J. M., Munoz, M., and Mathis, A., Chivers, D. P., and Smith, R. J. F. 1995. Chemical Gonzalez, A. 2005. Lateral and medial amygdala of anuran alarm signals: Predator deterrents or predator attractants? amphibians and their relation to olfactory and vomeronasal Am. Nat. 145, 994–1005. information. Brain Res. Bull. 66, 332–336. Matsunami, H. and Buck, L. B. 1997. A multigene family encod- Mori, K., Nowycky, M. C., and Shepherd, G. M. 1981. ing a diverse array of putative pheromone receptors in Electrophysiological analysis of mitral cells in the isolated mammals. Cell 90, 775–784. turtle olfactory bulb. J. Physiol. 314, 281–294. 400 Evolution of Vertebrate Olfactory Subsystems

Morita, Y. and Finger, T. E. 1998. Differential projections of Nevitt, G. A. and Haberman, K. 2003. Behavioral attraction of ciliated and microvillous olfactory receptor cells in the catfish, Leach’s storm-petrels (Oceanodroma leucorhoa) to dimethyl Ictalurus punctatus. J. Comp. Neurol. 398, 539–550. sulfide. J. Exp. Biol. 206, 1497–1501. Mousley, A., Polese, G., Marks, N. J., and Eisthen, H. L. 2006. Nevitt, G. A., Veit, R. R., and Kareiva, P. 1995. Dimethyl sulfide Terminal nerve-derived neuropeptide Y modulates odorant as a foraging cue for Antarctic procellariiform seabirds. responses in hungry axolotls (Ambystoma mexicanum). Nature 376, 680–682. J. Neurosci. 26, 7707–7717. Nezlin, L. P. and Schild, D. 2000. Structure of the olfactory Muller, J. F. and Marc, R. E. 1984. Three distinct morphological bulb in tadpoles of Xenopus laevis. Cell Tissue Res. 302, classes of receptors in fish olfactory organs. J. Comp. Neurol. 21–29. 222, 482–495. Nezlin, L. P. and Schild, D. 2005. Individual olfactory sensory Mu¨ ller, H., Drenckhahn, D., and Haase, E. 1979. Vergleichend neurons project into more than one glomerulus in Xenopus quantitative und ultrastrukturelle Untersuchungen am laevis tadpole olfactory bulb. J. Comp. Neurol. 481, 233–239. Geruchsorgan von vier Haustaubenrassen. Z. Mikrosk. Nezlin, L. P., Heermann, S., Schild, D., and Rossler, W. 2003. Anat. Forsch. 93, 888–900. Organization of glomeruli in the main olfactory bulb of Mu¨ ller, J. and Henle, F. G. J. 1841. Systematische Beschreibung Xenopus laevis tadpoles. J. Comp. Neurol. 464, 257–268. der Plagiostomen. Ngai, J., Dowling, M. M., Buck, L., Axel, R., and Chess, A. Mun˜ oz, A. 2004. Chemo-orientation using conspecific chemical 1993a. The family of genes encoding odorant receptors in cues in the stripe-necked terrapin (Mauremys leprosa). J. channel catfish. Cell 72, 657–666. Chem. Ecol. 30, 519–530. Ngai, J., Chess, A., Dowling, M. M., Necles, N., Macagno, E. R., Mu¨ nz, H. and Claas, B. 1987. The terminal nerve and its devel- and Axel, R. 1993b. Coding of olfactory information: opment in teleost fishes. In: The Terminal Nerve (Nervus Topography of odorant receptor expression in the catfish Terminalis): Structure, Function, and Evolution (eds. L. S. olfactory epithelium. Cell 72, 667–680. Demski and M. Schwanzel-Fukuda), pp. 50–59. New York Nieuwenhuys, R. 1965. The telencephalon of the crossopterygian Academy of Sciences. Latimeria chalumnae Smith. J. Morphol. 117, 1–24. Mu¨ nz, H., Stumpf, W. E., and Jennes, L. 1981. LHRH systems in Nieuwenhuys, R. 1967. Comparative anatomy of the olfactory the brain of platyfish. Brain Res. 221, 1–13. centers and tracts. In: Progress in Brain Research (ed. Y. Murphy, F. A., Tucker, K., and Fadool, D. A. 2001. Sexual Zotterman), pp. 1–64. Elsevier. dimorphism and developmental expression of signal-trans- Nieuwenhuys, R. 2002. Deuterostome brains: Synopsis and com- duction machinery in the vomeronasal organ. J. Comp. mentary. Brain Res. Bull. 57, 257–270. Neurol. 432, 61–74. Niimura, Y. and Nei, M. 2005. Evolutionary dynamics of olfac- Muske, L. E. and Moore, F. L. 1988. The nervus terminalis in tory receptor genes in fishes and tetrapods. Proc. Natl. Acad. amphibians: Anatomy, chemistry and relationship with the Sci. USA 102, 6039–6044. hypothalamic gonadotropin-releasing hormone system. Norgren, R. B., Jr. and Lehman, M. N. 1991. Neurons that Brain Behav. Evol. 32, 141–150. migrate from the olfactory epithelium in the chick express Naifeh, K. H., Huggins, S. E., and Hoff, H. E. 1970. The nature luteinizing hormone-releasing hormone. Endocrinology 128, of the ventilatory period in crocodilian respiration. Resp. 1676–1678. Physiol. 10, 338–348. Norgren, R. B., Jr., Lippert, J., and Lehman, M. N. 1992. Naito, T., Saito, Y., Yamamoto, J., et al. 1998. Putative phero- Luteinizing hormone-releasing hormone in the pigeon term- 2 mone receptors related to the Ca þ-sensing receptor in Fugu. inal nerve and olfactory bulb. Neurosci. Lett. 135, Proc. Natl. Acad. Sci. USA 95, 5178–5181. 201–204. Nakamura, T. 2000. Cellular and molecular constituents of olfac- Northcutt, R. G. 1978. Forebrain and midbrain organization in tory sensation in vertebrates. Comp. Biochem. Physiol. A 126, lizards and its phylogenetic significance. In: Behavior and 17–32. Neurology of Lizards (eds. N. Greenberg and P. D. Nearing, J., Betka, M., Quinn, S., et al. 2002. Polyvalent cation MacLean), pp. 11–64. NIMH. receptor proteins (CaRs) are salinity sensors in fish. Proc. Northcutt, R. G. 1981. Evolution of the telencephalon in non- Natl. Acad. Sci. USA 99, 9231–9236. mammals. Ann. Rev. Neurosci. 4, 301–350. Nef, S. and Nef, P. 1997. Olfaction: Transient expression of a Northcutt, R. G. 1995. The forebrain of gnathostomes: In search putative odorant receptor in the avian notochord. Proc. Natl. of a morphotype. Brain Behav. Evol. 46, 275–318. Acad. Sci. USA 94, 4766–4771. Northcutt, R. G. 2005. The new head hypothesis revisited. J. Exp. Nelson, J. S. 1994. Fishes of the World, 3rd edn. Wiley. Biol. 304B, 274–297. Nevitt, G. and Dittman, A. 2004. Olfactory imprinting in salmon: Northcutt, R. G. 2006. Connections of the lateral and medial New models and approches. In: The Senses of fish. divisions of the goldfish telencephalic pallium. J. Comp. Adaptations for the Reception of Natural Stimuli (ed. G. Neurol. 494, 903–943. von der Emde, J. Mogdans, and B. G. Kapoor), Northcutt, R. G. and Bemis, W. E. 1993. Cranial nerves of the pp. 109–127. Kluwer. coelacanth Latimeria chalumnae (Osteichthyes: Sarcopterygii: Nevitt, G., Reid, K., and Trathan, P. 2004. Testing olfactory Actinistia) and comparisons with other craniates. Brain Behav. foraging strategies in an Antarctic seabird assemblage. J. Evol. 42(Suppl. 1), 1–76. Exp. Biol. 207, 3537–3544. Northcutt, R. G. and Braford, M. R. J. 1980. New observa- Nevitt, G. A. 1991. Do fish sniff? A new mechanism of olfactory tions on the organization and evolution of the sampling in pleuronectid flounders. J. Exp. Biol. 157, 1–18. telencephalon of the actinopterygian fishes. Nevitt, G. A. 2000. Olfactory foraging by Antarctic procellarii- In: Comparative Neurology of the Telencephalon form seabirds: Life at high Reynolds numbers. Biol. Bull. 198, (ed. S. O. E. Ebbesson), pp. 41–98. Plenum. 245–253. Northcutt, R. G. and Davis, R. E. 1983. Telencephalic organiza- Nevitt, G. A. and Bonadonna, F. 2005. Sensitivity to dimethyl tion in ray-finned fishes. In: Fish Neurobiology (ed. R. E. sulphide suggests a mechanism for olfactory navigation by Davis and R. G. Northcutt), vol. 2, pp. 203–235. University seabirds. Biol. Lett. 1, 303–305. of Michigan Press. Evolution of Vertebrate Olfactory Subsystems 401

Northcutt, R. G. and Gans, C. 1983. The genesis of neural crest Ovaska, K. and Davis, T. M. 1992. Fecal pellets as burrow and epidermal placodes: A reinterpretation of vertebrate ori- markers: Intraspecific and interspecific odor recognition by gins. Q. Rev. Biol. 58, 1–28. Western plethodontid salamanders. Anim. Behav. 43, Northcutt, R. G. and Muske, L. E. 1994. Multiple embryonic 931–939. origins of gonadotropin-releasing hormone (GnRH) immu- Panchen, A. L. 1991. The early tetrapods: Classification and the noreactive neurons. Dev. Brain Res. 78, 279–290. shapes of cladograms. In: Origins of the Higher Groups of Northcutt, R. G. and Puzdrowski, R. L. 1988. Projections of the Tetrapods: Controversy and Consensus (ed. H.-P. Schultze olfactory bulb and nervus terminalis in the silver lamprey. and L. Trueb), pp. 100–144. Cornell University Press. Brain Behav. Evol. 32, 96–107. Parhar, I. S. 2002. Cell migration and evolutionary significance of Northcutt, R. G. and Royce, G. J. 1975. Olfactory bulb projec- GnRH subtypes. Prog. Brain Res. 141, 3–17. tions in the bullfrog Rana catesbeiana. J. Morphol. 145, Parhar, I. S., Pfaff, D. W., and Schwanzel-Fukuda, M. 1996. 251–268. Gonadotropin-releasing hormone gene expression in teleosts. Ohtomi, M., Fujii, K., and Kobayashi, H. 1989. Distribution of Mol. Brain Res. 41, 216–227. FMRFamide-like immunoreactivity in the brain and neurohy- Park, D. and Eisthen, H. L. 2003. Gonadotropin releasing pophysis of the lamprey, Lampetra japonica. Cell Tissue Res. hormone (GnRH) modulates odorant responses in the per- 256, 581–584. ipheral olfactory system of axolotls. J. Neurophysiol. 90, Oikawa, T., Suzuki, K., Saito, T. R., Takahashi, K. W., and 731–738. Taniguchi, K. 1998. Fine structure of three types of olfactory Park, D. and Propper, C. R. 2001. Repellent function of male organs in Xenopus laevis. Anat. Rec. 252, 301–310. pheromones in the red-spotted newt. J. Exp. Biol. 289, Oka, Y. 1992. Gonadotropin-releasing hormone (GnRH) cells of 404–408. the terminal nerve as a model neuromodulator system. Park, D. and Propper, C. R. 2002. The olfactory organ is acti- Neurosci. Lett. 142, 119–122. vated by a repelling pheromone in the red-spotted newt Oka, Y. 1996. Characterization of TTX-resistant persistent Naþ Notophthalmus viridescens. Kor. J. Biol. Sci. 6, 233–237. current underlying pacemaker potentials of fish gonadotro- Park, D., Zawacki, S. R., and Eisthen, H. L. 2003. Olfactory pin-releasing hormone (GnRH) neurons. J. Neurophysiol. 75, signal modulation by molluscan cardioexcitatory tetrapeptide 2397–2404. (FMRFamide) in axolotls (Ambystoma mexicanum). Chem. Oka, Y. and Ichikawa, M. 1990. Gonadotropin-releasing hor- Senses 28, 339–348. mone (GnRH) immunoreactive system in the brain of the Park, D., McGuire, J. M., Majchrzak, A. L., Ziobro, J. M., and dwarf gourami (Colisa lalia) as revealed by light microscopic Eisthen, H. L. 2004. Discrimination of conspecific sex and immunocytochemistry using a monoclonal antibody to com- reproductive condition using chemical cues in axolotls mon amino acid sequence of GnRH. J. Comp. Neurol. 300, (Ambystoma mexicanum). J. Comp. Physiol. A 190, 511–522. 415–427. Oka, Y. and Matsushima, T. 1993. Gonadotropin-releasing hor- Park, D., Propper, C. R., and Eisthen, H. L. 2005. The mone (GnRH)-immunoreactive terminal nerve cells have pheromonal repelling response in red-spotted newts intrinsic rhythmicity and project widely in the brain. J. (Notophthalmus viridescens). In: Chemical Signals in Neurosci. 13, 2161–2176. Vertebrates 10 (ed. R. Mason, M. LeMaster, and D. Mu¨ ller- Oka, Y., Munro, A. D., and Lam, T. J. 1986. Retinopetal projec- Schwarze), pp. 42–48. Springer. tions from a subpopulation of ganglion cells of the nervus Parker, G. 1908. The sensory reactions of amphioxus. Proc. Am. terminalis in the dwarf gourami (Colisa lalia). Brain Res. Acad. Arts Sci. 43, 416–455. 367, 341–345. Parsons, T. S. 1959a. Studies on the comparative embryology of Okano, M. and Kasuga, S. 1980. Developmental studies on the the reptilian nose. Bull. Mus. Comp. Zool. 120, 101–277. olfactory epithelium in domestic fowls. In: Biological Rhythms Parsons, T. S. 1959b. Nasal anatomy and the phylogeny of rep- in Birds: Neural and Endocrine Aspects (ed. Y. Tanabe, K. tiles. Evolution 13, 175–187. Tanaka, and T. Ookawa), pp. 235–248. Springer. Parsons, T. S. 1967. Evolution of the nasal structures in the lower Olsson, M. and Shine, R. 1998. Chemosensory mate recognition tetrapods. Am. Zool. 7, 397–413. may facilitate prolonged mate guarding by male snow skinks, Parsons, T. S. 1970. The origin of Jacobson’s organ. Forma Niveoscincus microlepidotus. Behav. Ecol. Sociobiol. 43, Functio 3, 105–111. 359–363. Paterson, N. F. 1951. The nasal cavities of the toad Hemipipa Olsson, M., Madsen, T., Nordby, J., Wapstra, E., Ujvari, B., and carvalhoi Mir.-Rib. and other Pipidae. Proc. Zool. Soc. 121, Wittsell, H. 2003. Major histocompatibility complex and 381–415. mate choice in sand lizards. Proc. R. Soc. Lond. Ser. B 270, Pelosi, P. 2001. The role of perireceptor events in vertebrate S254–S256. olfaction. Cell. Mol. Life Sci. 58, 503–509. Oosawa, T., Hirano, Y., and Tonosaki, K. 2000. Petit, C., Hossaert-McKey, M., Perret, P., Blondel, J., and Electroencephalographic study of odor responses in the Lambrechts, M. M. 2002. Blue tits use selected plants and domestic fowl. Physiol. Behav. 71, 203–205. olfaction to maintain an aromatic environment for nestlings. Orrego, F. 1961. The reptilian forebrain. I. The olfactory path- Ecol. Lett. 5, 585–589. ways and cortical areas in the turtle. Arch. Ital. Biol. 99, Pfeiffer, W. 1964. Morphology of the olfactory organ of 425–445. Hoplopagrus guentheri Gill 1862. Can. J. Zool. 42, 235–237. O¨ stholm, T., Ekstrom, P., and Ebbesson, S. O. E. 1990. Pfeiffer, W. 1969. Das Geruchsorgan der rezenten Actinista und Distribution of FMRFamide-like immunoreactivity in the Dipnoi (Pisces). Z. Morphol. Tiere 64, 309–337. brain, retina and nervus terminalis of the sockeye salmon Pfennig, D. W. and Collins, J. P. 1993. Kinship affects morpho- parr, Oncorhynchus nerka. Cell Tissue Res. 261, 403–418. genesis in cannibalistic salamanders. Nature 362, 836–838. Ovaska, K. 1988. Recognition of conspecific odors by the Pfennig, D. W., Sherman, P. W., and Collins, J. P. 1994. Kin Western red-backed salamander, Plethodon vehiculum. Can. recognition and cannibalism in polyphenic salamanders. J. Zool.-Rev. Can. Zool. 66, 1293–1296. Behav. Ecol. 5, 225–232. 402 Evolution of Vertebrate Olfactory Subsystems

Pfister, P. and Rodriguez, I. 2005. Olfactory expression of a single Reiner, A. and Northcutt, R. G. 1987. An immunohistochemical and highly variable V1r pheromone receptor-like gene in fish study of the telencephalon of the African lungfish, species. Proc. Natl. Acad. Sci. USA 102, 5489–5494. Protopterus annectens. J. Comp. Neurol. 256, 463–481. Pinching, A. J. and Powell, T. P. S. 1971. The neuron types of the Reiner, A., Perkel, D. J., Bruce, L. L., et al. 2004. Revised nomen- glomerular layer of the olfactory bulb. J. Cell Sci. 9, 305–345. clature for avian telencephalon and some related brainstem Pinelli, C., D’Aniello, B., Sordino, P., Meyer, D. L., nuclei. J. Comp. Neurol. 473, 377–414. Fiorentino, M., and Rastogi, R. K. 2000. Comparative immu- Reiss, J. O. and Burd, G. D. 1997a. Metamorphic remodeling of nocytochemical study of FMRFamide neuronal system in the the primary olfactory projection in Xenopus: Developmental brain of Danio rerio and Acipenser ruthenus during develop- independence of projections from olfactory neuron sub- ment. Dev. Brain Res. 119, 195–208. classes. J. Neurobiol. 32, 213–222. Pinkus, F. 1894. U¨ ber einen noch nicht beschriebenen Hirnnerven Reiss, J. O. and Burd, G. D. 1997b. Cellular and molecular des Protopterus annectens. Anat. Anz. 9, 562–566. interactions in the development of the Xenopus olfactory Placyk, J. S., Jr. and Graves, B. M. 2002. Prey detection by system. Semin. Cell Dev. Biol. 8, 171–179. vomeronasal chemoreception in a plethodontid salamander. Reiss, J. O. and Eisthen, H. L. 2006. Chemical senses: J. Chem. Ecol. 28, 1017–1036. Comparative anatomy and physiology in amphibians. Polese, G. 2004. Sistemi peptidergici (FMRFamide, neuropeptide In: Senses on the Threshold, Function and Evolution of Y, miostatina), nervo terminale e sistema di fibre olfattive the Sense Organs in Secondarily Aquatic Vertebrates extrabulbari in due popolazioni (mediterranea e nordalantica) (eds. J. G. H. Thewissen and S. Nummela). University of di spigola: Analisi ontogenetica e comparativa. Unpublished California Press (in press). doctoral dissertation, Universita´ delgi studi di Napoli Restrepo, D., Arellano, J., Oliva, A. M., Schaefer, M. L., and ‘‘Federico II.’’ Lin, W. 2004. Emerging views on the distinct but related Polkinghorne, C. N., Olson, J. M., Gallaher, D. G., and roles of the main and accessory olfactory systems in respon- Sorensen, P. W. 2001. Larval sea lamprey release two unique siveness to chemosensory signals in mice. Horm. Behav. 46, bile acids to the water at a rate sufficient to produce detectable 247–256. riverine pheromone plumes. Fish Physiol. Biochem. 24, 15–30. Rieke, G. K. and Wenzel, B. M. 1978. Forebrain projections of Prabha, K. C., Bernard, D. G., Gardner, M., and Smatresk, N. J. pigeon olfactory bulb. J. Morphol. 158, 41–55. 2000. Ventilatory mechanics and the effects of water depth on Roden, K., Endepols, H., and Walkowiak, W. 2005. breathing pattern in the aquatic caecilian Typhlonectes Hodological characterization of the septum in anuran natans. J. Exp. Biol. 203, 263–272. amphibians: I. Afferent connections. J. Comp. Neurol. Propper, C. R. and Moore, F. L. 1991. Effects of courtship on 483, 415–436. brain gonadotropin hormone-releasing hormone and plasma Rogoff, J. L. 1927. The hedonic glands of Triturus viridescens:A steroid concentrations in a female amphibian (Taricha gran- structural and functional study. Anat. Rec. 34, 132–133. ulosa). Gen. Comp. Endocrinol. 81, 304–312. Rollmann, S. M., Houck, L. D., and Feldhoff, R. C. 1999. Pyatkina, G. A. 1976. Receptor cells of various types and their Proteinaceous pheromone affecting female receptivity in a proportional interrelation in the olfactory organ of larvae and terrestrial salamander. Science 285, 1907–1909. adults of acipenserid fishes. Tsitologiya 18, 1444–1449. Ronnett, G. V. and Moon, C. 2002. G proteins and olfactory Quinn, V. S. and Graves, B. M. 1998. Home pond discrimination signal transduction. Ann. Rev. Physiol. 64, 189–222. using chemical cues in Chrysemys picta. J. Herpetol. 32, Root, R. W. 1949. Aquatic respiration in the musk turtle. Physiol. 457–461. Zool. 22, 172–178. Rama Krishna, N. S. and Subhedar, N. K. 1992. Distribution of Rose, F. L. 1970. Tortoise chin gland fatty acid composition: FMRFamide-like immunoreactivity in the forebrain of the Behavioral significance. Comp. Biochem. Physiol. 32, catfish, Clarias batrachus (Linn.). Peptides 13, 183–191. 577–580. Ramoń y Cajal, P. 1922. El cerebro de los batracios. Libr. Hon. Rose, F. L., Drotman, R., and Weaver, W. G. 1969. Ramo´ n y Cajal 1, 13–59. Electrophoresis of chin gland extracts of Gopherus Ramoń y Cajal, S. 1890. Origen y terminacioń de las fibras (Tortoises). Comp. Biochem. Physiol. 29, 847–851. nerviosas olfatorias. Gazz. Sanit. Barcelona 1–21. Rosen, G., Sherwood, N., and King, J. A. 1997. Immunoreactive Reese, T. S. 1965. Olfactory cilia in the frog. J. Cell Biol. 25, gonadotropin-releasing hormone (GnRHir) is associated with 209–230. vestibular structures in the green anole (Anolis carolinensis). Reese, T. S. and Brightman, M. W. 1970. Olfactory surface and Brain Behav. Evol. 50, 129–138. central olfactory connexions in some vertebrates. In: Taste Rossi, A., Basile, A., and Palombi, F. 1972. Speculations on the and Smell in Vertebrates (eds. G. E. W. Wolstenholme and J. function of the nervus terminalis in teleosts. Riv. Biol. 65, Knight), pp. 115–143. J & A Churchill. 385–409. Rehorek, S. J. 1997. Squamate Harderian gland: An overview. Ro¨ssler, P., Mezler, M., and Breer, H. 1998. Two olfactory marker Anat. Rec. 248, 301–306. proteins in Xenopus laevis. J. Comp. Neurol. 395, 273–280. Rehorek, S. J., Firth, B. T., and Hutchinson, M. N. 2000a. The Ro¨ ssler, P., Kroner, C., Krieger, J., Lobel, D., Breer, H., and structure of the nasal chemosensory system in squamate rep- Boekhoff, I. 2000. Cyclic adenosine monophosphate signaling tiles. 2. Lubricatory capacity of the vomeronasal organ. in the rat vomeronasal organ: Role of an adenylyl cyclase type J. Biosci. 25, 181–190. VI. Chem. Senses 25, 313–322. Rehorek, S. J., Hillenius, W. W., Quan, W., and Halpern, M. Roth, E. D., May, P. G., and Farrell, T. M. 1999. Pigmy rattle- 2000b. Passage of Harderian gland secretions to the vomer- snakes use frog-derived chemical cues to select foraging sites. onasal organ of Thamnophis sirtalis (Serpentes: Colubridae). Copeia 772–774. Can. J. Zool. – Rev. Can. Zool. 78, 1284–1288. Rudebeck, B. 1945. Contributions to forebrain morphology in Reiner, A. and Karten, H. J. 1985. Comparison of olfactory bulb Dipnoi. Acta Zool. 26, 1–156. projections in pigeons and turtles. Brain Behav. Evol. 27, Ryba, N. J. and Tirindelli, R. 1997. A new multigene family of 11–27. putative pheromone receptors. Neuron 19, 371–379. Evolution of Vertebrate Olfactory Subsystems 403

Saint Girons, H. and Zylberberg, L. 1992. Histologie compare´e hormone (LH-RH) in the brain and the pituitary gland of a fish des glandes ce´phaliques exocrines et des fosses nasales des by immunocytochemistry. J. Exp. Zool. 210, 153–160. Lissamphibia. II. E´pitheliums des fosses nasales. Ann. Sci. Schwanzel-Fukuda, M. and Pfaff, D. W. 1989. Origin of luteiniz- Nat. Zool. (Paris) 13, 121–145. ing hormone-releasing hormone neurons. Nature 338, Saito, K., Shoji, T., Uchida, I., and Ueda, H. 2000. Structure of 161–164. the olfactory and vomeronasal epithelia in the loggerhead Schwenk, K. 1986. Morphology of the tongue in the tuatara, turtle Caretta caretta. Fish. Sci. 66, 409–411. Sphenodon punctatus (Reptilia: Lepidosauria), with com- Saito, S., Matsui, T., Kobayashi, N., et al. 2003. Lectin histo- ments on function and phylogeny. J. Morphol. 188, chemical study on the olfactory organ of the newt, Cynops 129–156. pyrrhogaster, revealed heterogeneous mucous environments Schwenk, K. 1993a. The evolution of chemoreception in squa- in a single nasal cavity. Anat. Embryol. 206, 349–356. mate reptiles: A phylogenetic approach. Brain Behav. Evol. Sam, M., Vora, S., Malnic, B., Ma, W., Novotny, M. V., and 41, 124–137. Buck, L. B. 2001. Odorants may arouse instinctive beha- Schwenk, K. 1993b. Are geckos olfactory specialists? J. Zool. viours. Nature 412, 142. 229, 289–302. Sato, K. and Suzuki, N. 2001. Whole-cell response characteristics Schwenk, K. 1994. Why snakes have forked tongues. Science 263, of ciliated and microvillous olfactory receptor neurons to 1573–1577. amino acids, pheromone candidates and urine in rainbow Schwenk, K. 1995. Of tongues and noses: Chemoreception in trout. Chem. Senses 26, 1145–1156. lizards and snakes. Trends Ecol. Evol. 10, 7–12. Sato, Y., Miyasaka, N., and Yoshihara, Y. 2005. Mutually exclu- Schwenk, K. 2000. Feeding in lepidosaurs. In: Feeding: Form, sive glomerular innervation by two distinct types of olfactory Function and Evolution in Tetrapod Vertebrates (ed. K. sensory neurons revealed in transgenic zebrafish. J. Anat. 25, Schwenk), pp. 175–291. Academic Press. 4889–4897. Scott, A. P. and Sorensen, P. W. 1994. Time course of release Satoh, G. 2005. Characterization of novel GPCR gene coding of pheromonally active gonadal steroids and their conju- locus in amphioxus genome: Gene structure, expression, and gates by ovulatory goldfish. Gen. Comp. Endocrinol. 96, phylogenetic analysis with implications for its involvement in 309–323. chemoreception. Genesis 41, 47–57. Scott, T. P. and Weldon, P. J. 1990. Chemoreception in the Scalia, F. 1972. The projection of the accessory olfactory bulb in feeding behavior of adult American alligators, Alligator mis- the frog. Brain Res. 36, 409–411. sissippiensis. Anim. Behav. 39, 398–400. Scalia, F. 1976. Structure of the olfactory and accessory olfactory Secondi, J., Haerty, W., and Lode, T. 2005. Female attraction to systems. In: Frog Neurobiology (ed. R. Llina´s and W. Precht), conspecific chemical cues in the palmate newt Triturus helve- pp. 213–233. Springer. ticus. Ethology 111, 726–735. Scalia, F., Halpern, M., and Riss, W. 1969. Olfactory bulb pro- Sewertzoff, A. N. 1902. Zur Entwicklungsgeschichte des jections in the South American caiman. Brain Behav. Evol. 2, Ceratodus forsteri. Anat. Anz. 21, 593–608. 238–262. Seydel, O. 1895. U¨ ber die Nasenho¨ hle und das Jacobson’sche Scalia, F., Gallousis, G., and Roca, S. 1991a. Differential projec- Organ der Amphibien: Eine vergleichend-anatomische tions of the main and accessory olfactory bulb in the frog. Untersuchung. Morphol. Jahrb. 23, 453–543. J. Comp. Neurol. 305, 443–461. Sheldon, R. E. 1911. The sense of smell in Selachians. J. Exp. Scalia, F., Gallousis, G., and Roca, S. 1991b. A note on the Zool. 10, 51–62. organization of the amphibian olfactory bulb. J. Comp. Shelton, R. G. J. 1978. On the feeding of the hagfish Myxine Neurol. 305, 435–442. glutinosa in the North Sea. J. Marine Biol. Assoc. UK 58, Schaefer, M. L., Finger, T. E., and Restrepo, D. 2001. Variability 81–86. of position of the P2 glomerulus within a map of the mouse Shepherd, G. M. 2006. Behaviour: Smells, brains and hormones. olfactory bulb. J. Comp. Neurol. 436, 351–362. Nature 439, 149–151. Schild, D. and Restrepo, D. 1998. Transduction mechanisms in Sherwood, N. M., Zoeller, R. T., and Moore, F. L. 1986. vertebrate olfactory receptor cells. Physiol. Rev. 78, 429–466. Multiple forms of gonadotropin-releasing hormone in amphi- Schmachtenberg, O. 2006. Histological and electrophysiological bian brains. Gen. Comp. Endocrinol. 61, 313–322. properties of crypt cells from the olfactory epithelium of the Shibuya, T. and Tucker, D. 1967. Single unit responses of olfac- marine teleost Trachurus symmetricus. J. Comp. Neurol. 495, tory receptors in vultures. In: Olfaction and Taste (ed. T. 113–121. Hayashi), vol. II, pp. 219–234. Pergamon. Schmidt, A. and Roth, G. 1990. Central olfactory and vomero- Shine, R., Phillips, B., Waye, H., LeMaster, M., and nasal pathways in salamanders. J. Hirnforsch. 31, 543–553. Mason, R. T. 2003. Chemosensory cues allow courting Schmidt, A. and Wake, M. H. 1990. Olfactory and vomeronasal male garter snakes to assess body length and body condi- systems of caecilians (Amphibia: Gymnophiona). J. Morphol. tion of potential mates. Behav. Ecol. Sociobiol. 54, 205, 255–268. 162–166. Schmidt, A., Naujoks-Manteuffel, C., and Roth, G. 1988. Shinn, E. A. and Dole, J. W. 1978. Evidence for a role for Olfactory and vomeronasal projections and the pathway of olfactory cues in the feeding response of leopard frogs, Rana the nervus terminalis in ten species of salamanders. Cell Tissue pipiens. Herpetologica 34, 167–172. Res. 251, 45–50. Shoji, T. and Kurihara, K. 1991. Sensitivity and transduction Schnitzlein, H. N. and Crosby, E. C. 1967. The telencephalon of mechanisms of responses to general odorants in turtle vomer- the lungfish, Protopterus. J. Hirnforsch. 9, 105–149. onasal system. J. Gen. Physiol. 98, 909–919. Schober, A., Meyer, D. L., and von Bartheld, C. S. 1994. Central Shoji, T., Enomoto, S., Taniguchi, M., and Kurihara, K. 1993. projections of the nervus terminalis and the nervus praeopti- Transduction mechanisms in the olfactory and vomeronasal cus in the lungfish brain revealed by nitric oxide synthase. J. organs of turtles. Brain Behav. Evol. 41, 192–197. Comp. Neurol. 349, 1–19. Shoji, T., Fujita, K. I., Furihata, E., and Kurihara, K. 1996. Schreibman, M. P., Halpern, L. R., Goos, H. J., and Margolis- Olfactory responses of a euryhaline fish, the rainbow trout: Kazan, H. 1979. Identification of luteinizing hormone-releasing Adaptation of olfactory receptors to sea water and 404 Evolution of Vertebrate Olfactory Subsystems

salt-dependence of their responses to amino acids. J. Exp. neurons in the terminal nerve of the spiny dogfish, Squalus Biol. 199, 303–310. acanthias. Anat. Rec. 208, 173A–174A. Siefkes, M. J. and Li, W. 2004. Electrophysiological evidence for Sterzi, G. 1909. Il Sistema Nervoso Centrale dei Vertebrati. detection and discrimination of pheromonal bile acids by the Draghi. olfactory epithelium of female sea lampreys (Petromyzon Stuelpnagel, J. T. and Reiss, J. O. 2005. Olfactory metamorphosis marinus). J. Comp. Physiol. A 190, 193–199. in the coastal giant salamander (Dicamptodon tenebrosus). J. Silver, W. L. 1979. Olfactory responses from a marine elasmo- Morphol. 266, 22–45. branch, the Atlantic stingray, Dasyatis sabina. Marine Behav. Szabo, T., Bla¨hser, S., Denizot, J.-P., and Ravaille-Ve´ron, M. Physiol. 6, 297–305. 1991. Projection olfactive primaire extrabulbaire chez cer- Simons, R. R., Felgenhauer, B. E., and Jaeger, R. G. 1994. taines poissons te´le´oste´ens. C. R. Acad. Sci. Paris 312, Salamander scent marks: Site of production and their role in 555–560. territorial defense. Anim. Behav. 48, 97–103. Takami, S. and Hirosawa, K. 1990. Electron microscopic Skeen, L. C. and Rolon, R. R. 1982. Olfactory bulb interconnec- observations on the vomeronasal sensory epithelium of a tions in the soft shell turtle. Neurosci. Lett. 33, 223–228. crotaline snake, Trimeresurus flavoviridis. J. Morphol. 205, Skeen, L. C., Pindzola, R. R., and Schofield, B. R. 1984. 45–63. Tangential organization of olfactory, association, and com- Tamburri, M. N. and Barry, J. P. 1999. Adaptations for scaven- missural projections to olfactory cortex in a species of reptile ging by three diverse bathyal species, Eptatretus stouti, (Trionyx spiniferus), bird (Aix sponsa), and mammal (Tupaia Neptunea amianta, and Orchomene obtusus. Deep-Sea Res. glis). Brain Behav. Evol. 25, 206–216. 46 (Pt. I), 2079–2093. Smeets, W. J. A. J. 1983. The secondary olfactory connections in Taniguchi, K., Toshima, Y., Saito, T. R., and Taniguchi, K. 1996. two chondrichthians, the shark Scyliorhinus canicula and the Development of the olfactory epithelium and vomeronasal ray Raja clavata. J. Comp. Neurol. 218, 334–344. organ in the Japanese reddish frog, Rana japonica. J. Vet. Smith, M. T. and Mason, R. T. 1997. Gonadotropin antagonist Med. Sci. 58, 7–15. modulates courtship behavior in male red-sided garter snakes, Tegoni, M., Pelosi, P., Vincent, F., et al. 2000. Mammalian Thamnophis sirtalis parietalis. Physiol. Behav. 61, 137–143. odorant binding proteins. Biochim. Biophys. Acta 1482, Smith, M. T., Moore, F. L., and Mason, R. T. 1997. 229–240. Neuroanatomical distribution of chicken-I gonadotropin- Tester, A. L. 1963. Olfaction, gustation, and the common chemi- releasing hormone (cGnRH-I) in the brain of the male red- cal sense in sharks. In: Sharks and Survival (ed. P. W. Gilbert), sided garter snake. Brain Behav. Evol. 49, 137–148. pp. 255–282. Heath. Sneddon, H., Hadden, R., and Hepper, P. G. 1998. Teuchert, G., Reissmann, T., and Vockel, A. 1986. Olfaction in Chemosensory learning in the chicken embryo. Physiol. Peking ducks (Anas platyrhynchos): A comparative study of Behav. 64, 133–139. centrifugal and centripetal olfactory connections in young Sorensen, P. W. and Sato, K. 2005. Second messenger systems ducks and in embryos and ducklings (Aves). mediating sex pheromone and amino acid sensitivity in gold- Zoomorphology 106, 185–198. fish olfactory receptor neurons. Chem. Senses 30(Suppl. 1), Theisen, B. 1972. Ultrastructure of the olfactory epithelium in the i315–i316. Australian lungfish Neoceratodus forsteri. Acta Zool. 53, Sorensen, P. W. and Scott, A. P. 1994. The evolution of hormonal 205–218. sex pheromones in teleost fish: Poor correlation between the Theisen, B. 1973. The olfactory system in the hagfish Myxine pattern of steroid release by goldfish and olfactory sensitivity glutinosa. I: Fine structure of the apical part of the olfactory suggests that these cues evolved as a result of chemical spying epithelium. Acta Zool. 54, 271–284. rather than signal specialisation. Acta Physiol. Scand. 152, Theisen, B., Zeiske, E., and Breucker, H. 1986. Functional mor- 191–205. phology of the olfactory organs in the spiny dogfish (Squalus Sorensen, P. W., Hara, T. J., Stacey, N. E., and Goetz, F. W. acanthias L.) and the small-spotted catshark (Scyliorhinus 1988. F prostaglandins function as potent olfactory stimu- canicula L.). Acta Zool. 67, 73–86. lants that comprise the postovulatory female sex pheromone Thommesen, G. 1983. Morphology, distribution, and specificity in goldfish. Biol. Reprod. 39, 1039–1050. of olfactory receptor cells in salmonid fishes. Acta Physiol. Sorensen, P. W., Hara, T. J., and Stacey, N. E. 1991. Sex pher- Scand. 117, 241–249. omones selectively stimulate the medial olfactory tracts of Thornhill, R. A. 1967. The ultrastructure of the olfactory epithe- male goldfish. Brain Res. 558, 343–347. lium of the lamprey Lampetra fluviatilis. J. Cell Sci. 2, Sorensen, P. W., Pinillos, M., and Scott, A. P. 2005a. Sexually 591–602. mature male goldfish release large quantities of androstene- Tobet, S. A., Nozaki, M., Youson, J. H., and Sower, S. A. 1995. dione into the water where it functions as a pheromone. Gen. Distribution of lamprey gonadotropin-releasing hormone-III Comp. Endocrinol. 140, 164–175. (GnRH-III) in brains of larval lampreys. Cell Tissue Res. 279, Sorensen, P. W., Fine, J. M., Dvornikovs, V., et al. 2005b. 261–270. Mixture of new sulfated steroids functions as a migratory Toyoda, F., Hayakawa, Y., Ichikawa, M., and Kikuyama, S. pheromone in the sea lamprey. Nature Chem. Biol. 1, 1999. Olfactory responses to a female-attracting pheromone 324–328. in the newt, Cynops pyrrhogaster. In: Advances in Chemical Speca, D. J., Lin, D. M., Sorensen, P. W., Isacoff, E. Y., Ngai, J., Signals in Vertebrates (eds. R. E. Johnston, D. Mu¨ ller- and Dittman, A. H. 1999. Functional identification of a gold- Schwarze, and P. W. Sorensen), pp. 607–615. Kluwer/ fish odorant receptor. Neuron 23, 487–498. Plenum. Stacey, N. E. and Sorensen, P. W. 2002. Fish hormonal phero- Tucker, D. 1965. Electrophysiological evidence for olfactory mones. In: Hormones, Brain, and Behavior (ed. D. Pfaff, D. function in birds. Nature 207, 34–36. Arnold, A. Ategen, S. Farbach, and R. Rubin), vol. 2, Tucker, D. 1971. Nonolfactory responses from the nasal cavity: pp. 375–435. Academic Press. Jacobson’s organ and the trigeminal system. In: Handbook of Stell, W. K. 1984. Luteinizing hormone-releasing hormone Sensory Physiology (ed. L. M. Beidler) vol. 4, Part 1, (LHRH)- and pancreatic polypeptide (PP)-immunoreactive pp. 151–181. Springer. Evolution of Vertebrate Olfactory Subsystems 405

Tuinhof, R., Gonzalez, A., Smeets, W., and Roubos, E. W. 1994. Waldman, B. 1985. Olfactory basis of kin recognition in toad Neuropeptide Y in the developing and adult brain of the South tadpoles. J. Comp. Physiol. A 156, 565–577. African clawed toad Xenopus laevis. J. Chem. Neuroanat. 7, Waldman, B. and Bishop, P. J. 2004. Chemical communication in 271–283. an archaic anuran amphibian. Behav. Ecol. 15, 88–93. Uchiyama, H., Reh, T. A., and Stell, W. K. 1988. Walker, S. E. and Stell, W. K. 1986. Gonadotropin-releasing Immunocytochemical and morphological evidence for a reti- hormone (GnRF), molluscan cardioexcitatory peptide nopetal projection in anuran amphibians. J. Comp. Neurol. (FMRFamide), enkephalin and related neuropeptides affect 274, 48–59. goldfish retinal ganglion cell activity. Brain Res. 384, Van Damme, R. and Quick, K. 2001. Use of predator chemical 262–273. cues by three species of lacertid lizards (Lacerta bedriagae, Wallraff, H. G. 2003. Olfactory navigation by birds. J. Ornithol. Podarcis tiliguerta, and Podarcis sicula). J. Herpetol. 35, 144, 1–32. 27–36. Wallraff, H. G. 2004. Avian olfactory navigation: Its empirical VanDenbossche, J., Seelye, J. G., and Zielinski, B. S. 1995. The foundation and conceptual state. Anim. Behav. 67, 189–204. morphology of the olfactory epithelium in larval, juvenile and Walls, G. Y. 1981. Feeding ecology of the tuatara Sphenodon upstream migrant stages of the sea lamprey, Petromyzon mar- punctatus on Stephens Island, Cook Strait. N. Z. J. Ecol. 4, inus. Brain Behav. Evol. 45, 19–24. 89–97. VanDenbossche, J., Youson, J. H., Pohlman, D., Wong, E., and Wang, R. T. and Halpern, M. 1980a. Light and electron micro- Zielinski, B. S. 1997. Metamorphosis of the olfactory organ of scopic observations on the normal structure of the the sea lamprey (Petromyzon marinus L): Morphological vomeronasal organ of garter snakes. J. Morphol. 164, 47–67. changes and morphometric analysis. J. Morphol. 231, 41–52. Wang, R. T. and Halpern, M. 1980b. Scanning electron micro- Vermeirssen, E. L. M. and Scott, A..P. 1996. Excretion of free and scopic studies of the surface morphology of the vomeronasal conjugated steroids in rainbow trout (Oncorhynchus mykiss): epithelium and olfactory epithelium of garter snakes. Am. J. Evidence for branchial excretion of the maturation-inducing Anat. 157, 399–428. steroid, 17,20 beta-dihydroxy-4-pregnen-3-one. G. Comp. Warbeck, A. and Parzefall, J. 2001. Mate recognition via water- Endocrinol. 101, 180–194. borne chemical cues in the viviparous caecilian Typhlonectes Verrell, P. A. 1985. Male mate choice for large fecund females in natans (Amphibia: Gymnophiona). In: Chemical Signals in the red-spotted newt, Notophthalmus viridescens: How is size Vertebrates 9 (eds. A. Marchlewska-Koj, J. J. Lepri, and D. assessed? Herpetologica 41, 382–386. Mu¨ ller-Schwarze), pp. 263–268. Kluwer Academic/Plenum. Vogt, R. G. 1987. The molecular basis of pheromone reception: Weldon, P. J. and Ferguson, M. W. J. 1993. Chemoreception in Its influence on behavior. In: Pheromone Biochemistry crocodilians: Anatomy, natural history, and empirical results. (eds. G. D. Prestwich and G. L. Blomquist), pp. 385–431. Brain Behav. Evol. 41, 239–245. Academic Press. Weldon, P. J., Swenson, D. J., Olson, J. K., and Brinkmeier, W. G. von Bartheld, C. S. and Meyer, D. L. 1986. Central connections 1990. The American alligator detects food chemicals in aqua- of the olfactory bulb in the bichir, Polypterus palmas, reex- tic and terrestrial environments. Ethology 85, 191–198. amined. Cell Tissue Res. 244, 527–535. Weldon, P. J., Brinkmeier, W. G., and Fortunato, H. 1992. Gular von Bartheld, C. S. and Meyer, D. L. 1988. Central projections of pumping responses by juvenile American alligators (Alligator the nervous terminalis in lampreys, lungfishes, and bichirs. mississippiensis) to meat scents. Chem. Senses 17, 79–83. Brain Behav. Evol. 32, 151–159. Wenzel, B. M. 1968. Olfactory prowess of the kiwi. Nature 220, von Bartheld, C. S., Meyer, D. L., Fiebig, E., and 1133–1134. Ebbesson, S. O. E. 1984. Central connections of the olfactory Wenzel, B. M. 1971. Olfactory sensation in kiwi and other birds. bulb in the goldfish, Carassius auratus. Cell Tissue Res. 238, Ann. NY Acad. Sci. 188, 183–193. 475–487. Wheeler, J. W., Ibrahim, S. A., and Weldon, P. J. 1999. 2,3- von Bartheld, C. S., Lindo¨ rfer, H. W., and Meyer, D. L. 1987. Dihydrofarnesyl and citronellyl esters in the paracloacal The nervous terminalis also exists in cyclostomes and birds. gland secretions of the brown caiman (Caiman crocodilus Cell Tissue Res. 250, 431–434. fuscus) from Costa Rica. Biochem. Syst. Ecol. 27, 27–32. von Frisch, K. 1938. Zur psychologie des Fishe-Schwarmes. White, J. and Meredith, M. 1987. The nervus terminalis of the Naturwissenschaften 26, 601–606. shark: The effect of efferent impulses on ganglion cell activity. von Frisch, K. 1941. U¨ ber einen Shreckstoff der Fischhaut and Brain Res. 400, 159–164. seine biologishe Bedeutung. Z. Vergl. Physiol. 29, 46–145. White, J. and Meredith, M. 1993. Spectral analysis and modelling Wabnitz, P. A., Bowie, J. H., Tyler, M. J., Wallace, J. C., and of ACh and NE effects on shark nervous terminalis activity. Smith, B. P. 1999. Aquatic sex pheromone from a male tree Brain Res. Bull. 31, 369–374. frog. Nature 401, 444–445. White, J. and Meredith, M. 1995. Nervous terminalis ganglion of Wachowiak, M. and Cohen, L. B. 2003. Correspondence the bonnethead shark (Sphyrna tiburo): Evidence for choli- between odorant-evoked patterns of receptor neuron input nergic and catecholamingeric influence on two cell types and intrinsic optical signals in the mouse olfactory bulb. J. distinguished by peptide immunocytochemistry. J. Comp. Neurophysiol. 89, 1623–1639. Neurol. 351, 385–403. Wachowiak, M., Cohen, L. B., and Zochowski, M. R. 2002. Whitlock, K. E. and Westerfield, M. 2000. The olfactory pla- Distributed and concentration-invariant spatial representa- codes of the zebrafish form by convergence of cellular fields tions of odorants by receptor neuron input to the turtle at the edge of the neural plate. Development 127, olfactory bulb. J. Neurophysiol. 87, 1035–1045. 3645–3653. Waldman, B. 1982. Sibling association among schooling tad- Wicht, H. and Northcutt, R. G. 1992. FMRFamide-like immu- poles: Field evidence and implications. Anim. Behav. 30, noreactivity in the brain of the Pacific hagfish, Eptatretus 700–713. stouti (Myxinoidea). Cell Tissue Res. 270, 443–449. Waldman, B. 1984. Kin recognition and sibling association Wicht, H. and Northcutt, R. G. 1993. Secondary olfactory pro- among wood frog (Rana sylvatica) tadpoles. Behav. Ecol. jections and pallial topography in the Pacific hagfish, Sociobiol. 14, 171–180. Eptatretus stouti. J. Comp. Neurol. 337, 529–542. 406 Evolution of Vertebrate Olfactory Subsystems

Williams, G. C. 1992. Natural Selection: Domains, Levels, and Yamamoto, N., Oka, Y., and Kawashima, S. 1997. Lesions of Challenges. Oxford University Press. gonadotropin-releasing hormone-immunoreactive terminal Wirsig, C. R. and Getchell, T. V. 1986. Amphibian terminal nerve cells: Effects on the reproductive behavior of male nerve: Distribution revealed by LHRH and AChE markers. dwarf gouramis. Neuroendocrinology 65, 403–412. Brain Res. 385, 1–21. Yang, Z. C., Whyte, A., Attygalle, A. B., Weldon, P. J., Eisner, T., Wirsig-Wiechmann, C. R. 1990. The nervus terminalis in the and Meinwald, J. 1999. Reptilian chemistry: Characterization chick: A FMRFamide-immunoreactive and AChE-positive of a family of dianeackerone-related steroidal esters from a nerve. Brain Res. 523, 175–179. crocodile secretion. Proc. Natl. Acad. Sci. USA 96, Wirsig-Wiechmann, C. R. and Basinger, S. F. 1988. FMRFamide- 12251–12256. immunoreactive retinopetal fibers in the frog, Rana pipiens: Young, B. A. 1993. Evaluating hypotheses for the transfer of Demonstration by lesion and immunocytochemical techni- stimulus particles to Jacobson’s organ in snakes. Brain ques. Brain Res. 449, 116–134. Behav. Evol. 41, 203–209. Wirsig-Wiechmann, C. R. and Jennes, L. 1993. Gonadotropin- Zancanaro, C., Caretta, C. M., Bolner, A., Sbarbati, A., releasing hormone agonist binding in tiger salamander nasal Nordera, G. P., and Osculati, F. 1997. Biogenic amines in cavity. Neurosci. Lett. 160, 201–204. the vomeronasal organ. Chem. Senses 22, 439–445. Wirsig-Wiechmann, C. R. and Lee, C. E. 1999. Estrogen regulates Zeiske, E. and Hansen, A. 2005. Development and phylogeny of gonadotropin-releasing hormone in the nervous terminalis of the olfactory organ in gnathostome fish. In: Fish Chemosenses Xenopus laevis. Gen. Comp. Endocrinol. 115, 301–308. (eds. K. Reutter and B. G. Kapoor), pp. 1–26. Oxford/IBH Wirsig-Wiechmann, C. R., Wiechmann, A. F., and Eisthen, H. L. Publishing Co. 2002a. What defines the nervous terminalis? Neurochemical, Zeiske, E., Caprio, J., and Gruber, S. H. 1986. Morphological developmental, and anatomical criteria. Prog. Brain Res. 141, and electrophysiological studies on the olfactory organ of the 45–58. lemon shark, Negaprion brevirostris (Poey). In: Indo-Pacific Wirsig-Wiechmann, C. R., Houck, L. D., Feldhoff, P. W., and Fish Biology: Proceedings of the Second International Feldhoff, R. C. 2002b. Pheromonal activation of vomeronasal Conference on Indo-Pacific Fishes (eds. T. Uyeno, R. Arai, neurons in plethodontid salamanders. Brain Res. 952, T. Taniuchi, and K. Matsuura), pp. 381–391. Ichthyological 335–344. Society of Japan. Wisenden, B. D. 2000. Olfactory assessment of predation risk in Zeiske, E., Theisen, B., and Gruber, S. H. 1987. Functional the aquatic environment. Philos. Trans. R. Soc. Lond. Ser. B morphology of the olfactory organ of two carcharhinid 355, 1205–1208. shark species. Can. J. Zool. 65, 2406–2412. Woolf, T. B., Shepherd, G. M., and Greer, C. A. 1991. Local Zeiske, E., Theisen, B., and Breucker, H. 1992. Structure, devel- information processing in dendritic trees: Subsets of spines in opment, and evolutionary aspects of the peripheral olfactory granule cells of the mammalian olfactory bulb. J. Neurosci. system. In: Fish Chemoreception (ed. T. J. Hara), pp. 13–32. 11, 1837–1854. Chapman and Hall. Wright, D. E. and Demski, L. S. 1996. Organization of GnRH Zeiske, E., Kasumyan, A., Bartsch, P., and Hansen, A. 2003. and FMRF-amide systems in two primitive bony fishes (order Early development of the olfactory organ in sturgeons of the Polypteriformes). Brain Behav. Evol. 47, 267–278. genus Acipenser: A comparative and electron microscopic Wright, G. M., McBurney, K. M., Youson, J. H., and Sower, S. A. study. Anat. Embryol. 206, 357–372. 1994. Distribution of lamprey gonadotropin-releasing hor- Zhou, Q., Hinkle, G., Sogin, M. L., and Dionne, V. E. 1996. mone in the brain and pituitary gland of larval, Cloning of olfactory receptor genes from the mudpuppy, metamorphic, and adult sea lampreys, Petromyzon marinus. Necturus maculosus. Chem. Senses 21, 694. Can. J. Zool. 72, 48–53. Zippel, H. P., Lago-Schaaf, T., and Caprio, J. 1993. Ciliated Wu, C. C., Yoshimoto, M., and Ito, H. 1992. The selachian olfactory receptor neurons in goldfish (Carassius auratus) terminal nerve. Kaibogaku Zasshi 67, 317–332. partially survive nerve axotomy, rapidly regenerate and Yamamoto, K., Kawai, Y., Hayashi, T., et al. 2000. Silefrin, a respond to amino acids. J. Comp. Physiol. A 173, 537–547. sodefrin-like pheromone in the abdominal gland of the Zippel, H. P., Sorensen, P. W., and Hansen, A. 1997. High sword-tailed newt, Cynops ensicauda. FEBS Lett. 472, correlation between microvillous olfactory receptor cell abun- 267–270. dance and sensitivity to pheromones in olfactory nerve- Yamamoto, M. 1982. Comparative morphology of the peripheral sectioned goldfish. J. Comp. Physiol. A 180, 39–52. olfactory organ in teleosts. In: Chemoreception in Fishes Zuri, I. and Halpern, M. 2003. Differential effects of lesions of (ed. T. J. Hara), pp. 39–59. Elsevier. the vomeronasal and olfactory nerves on garter snake Yamamoto, M. and Ueda, K. 1977. Comparative morphology of (Thamnophis sirtalis) responses to airborne chemical stimuli. fish olfactory epithelium. I: Salmoniformes. Bull. Jpn. Soc. Sci. Behav. Neurosci. 117, 169–183. Fish 43, 1163–1174. Zwilling, E. 1940. An experimental analysis of the development Yamamoto, N. and Ito, H. 2000. Afferent sources to the ganglion of the anuran olfactory organ. J. Exp. Zool. 55, 291–323. of the terminal nerve in teleosts. J. Comp. Neurol. 428, 355–375. Yamamoto, N., Oka, Y., Amano, M., Aida, K., Hasegawa, Y., Further Reading and Kawashima, S. 1995. Multiple gonadotropin-releasing hormone (GnRH)-immunoreactive systems in the brain of Finger, T. E., Silver, W. L., and Restrepo, D. (eds.) 2000. The the dwarf gourami, Colisa lalia: Immunohistochemistry and Neurobiology of Taste and Smell. 2nd edn. Wiley-Liss. Wilson, D. A. and Stevenson, R. J. 2006. Learning to Smell: radioimmunoassay. J. Comp. Neurol. 355, 354–368. Yamamoto, N., Uchiyama, H., Ohki-Hamazaki, H., Tanaka, H., Olfactory Perception from Neurobiology to Behavior. Johns and Ito, H. 1996. Migration of GnRH-immunoreactive neu- Hopkins University Press. rons from the olfactory placode to the brain: A study using Wyatt, T. D. 2003. Pheromones and Animal Behaviour: Communication by Smell and Taste. Cambridge University avian embryonic chimeras. Brain Res. Dev. Brain Res. 95, 234–244. Press.