THE ANATOMICAL RECORD PART A 288A:1201–1215 (2006) Terminal Nerve in the Mouse-Eared Bat (Myotis myotis): Ontogenetic Aspects HOLGER JASTROW,1* AND HELMUT HERMANN AUGUST OELSCHLÄGER2 Department of Anatomy and Cell Biology, Histology, Johannes Gutenberg University, Mainz, Germany 2 Department of Anatomy III (Dr. Senckenbergische Anatomie), J.W. Goethe University, Frankfurt am Main, Germany 1 ABSTRACT As in other mammals, ontogenesis of the terminal nerve (TN) in the mouse-eared bat (Myotis myotis) starts shortly after the formation of the olfactory placode, a derivative of the ectoderm. During development of the olfactory pit, proliferating neuroblasts thicken the placodal epithelium and one cell population migrates toward the rostroventral tip of the telencephalon. Here they accumulate in a primordial terminal ganglion, which successively divides into smaller units. Initial ﬁbers of the TN can be distinguished from olfactory ﬁbers in the mid-embryonic period. The main TN ﬁber bundle (mfb) originates from the anteriormost ganglion in the nasal roof, whereas one or more inconstant smaller ﬁber bundles (sfb) originate from one or more smaller ganglia in the basal part of the rostral nasal septum. The ﬁbers of the mfb and sfbs join in the posterior quarter of the nasal roof before reaching the central ganglion (M) located in the meninges medial to the olfactory bulb. From the mid-fetal period onward, a thin TN ﬁber bundle with some intermingled perikarya connects M to the brain by penetrating its wall rostral to the olfactory tubercle. Additional smaller ganglia may occur in this region. The TN and its ganglia persist in postnatal and adult bats but the number of perikarya is reduced here. Moreover, the different potential functions of the TN are discussed brieﬂy. Anat Rec Part A, 288A:1201–1215, 2006. Ó 2006 Wiley-Liss, Inc. Key words: Myotis myotis; nervus terminalis; olfactory placode; development; neuron count The terminal nerve (TN; nervus terminalis or cranial nerve zero) is located medial to the olfactory bulb and the olfactory ﬁbers. It was found in all mammals investigated so far (with few questionable exceptions) (Brown, 1987). Whether or not the gonadotropin-releasing hormone-immunoreactive (GnRH) neurons arise from the olfactory placode or neural crest, they are the main source of GnRH (Krey and Silverman, 1978; King et al., 1984; Schwanzel-Fukuda et al., 1985, 1988, 1994; Caldani et al., 1987; Schwanzel-Fukuda and Pfaff, 1988, 1989, 1990; Zheng et al., 1988; Jennes, 1989; Wray et al., 1989a, 1989b; Demski et al., 1990; Ronnekeiv and Resco, 1990; Oelschläger and Northcutt, 1992; Muske, 1993; Northcutt and Muske, 1994; Jastrow et al., 1998). Recently, it was demonstrated in zebraﬁsh that all hypothalamic GnRH cells arise from the neural crest (Whitlock et al., 2003). Here, the neuroendocrine population originates from the adenohypophyseal region of the developing anterior neural plate, whereas their neuroÓ 2006 WILEY-LISS, INC. modulatory neurons originate from the cranial neural crest. The latter population becomes associated with the olfactory placode and from there migrates into the diencephalon as part of the TN. The terminal nerve participates in the innervation and function of the nasal apparatus in general and hence, among other aspects, in olfaction and in the autonomic regulation of blood ﬂow in the nasal mucosa. Thus, evolutionary changes in this area may result in modiﬁcations of the TN system. *Correspondence to: Holger Jastrow, Raiffeisenstrasse 11, 55239 Gau-Odernheim, Germany. Fax: 49-6733-949540. E-mail: email@example.com Received 23 December 2005; Accepted 8 August 2006 DOI 10.1002/ar.a.20390 Published online 9 October 2006 in Wiley InterScience (www.interscience.wiley.com). 1202 JASTROW AND OELSCHLÄGER TABLE 1. List of investigated microslide series of the mouse-eared bat (Myotis myotis) No. CRL (mm) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 6.5 9 10 11 12 12 12 12 13 13 14 14 15 16 18.3 19 21 22 22 23 23 23 27 30 subadult 1 subadult 2 adult HL (mm) Stage Age (D) Orientation 4 5 5 5 6 6 6 6 6 6 6 6 7 7 8 8 8 8 8 8 8 8 9 H E E E 33 37 39 40 41 41 41 42 42 43 43 43 44 45 47 47 50 52 52 52 53 53 57 60 younger subadult older subadult adult sagittal transverse transverse sagittal transverse transverse transverse transverse transverse transverse transverse transverse transverse sagittal transverse sagittal transverse transverse sagittal transverse transverse sagittal sagittal frontal sagittal transverse sagittal – – 6 6.5 7 7 7 8 7.5 8.5 9 9 10 13 13 13 14 15 15 14 15 15 16 18 – – – CRL, crown-rump length; HL, head length; Stage, ontogenetic stage according to Štěrba (1990); Age (D), estimated days after fertilization using tables from Štěrba (1990); H, ‘‘Hairs on body’’ stage according to Štěrba (1990); E, ‘‘Eyelids open’’ stage according to Štěrba (1990). Considering the close topographical relationship of the TN to the olfactory and vomeronasal systems, it is interesting to note that some bats such as Myotis motis completely lack a vomeronasal organ (Mann, 1961), as is known of toothed and baleen whales. In these species, the TN may have taken over functions of the vomeronasal system, particularly in the whales, which lack olfactory ﬁla and an olfactory bulb from the early fetal period onward (Sinclair, 1951a, 1966; Oelschläger and Buhl, 1985a, 1985b; Oelschläger et al., 1987; Ridgway et al., 1987; Oelschläger, 1989). In this article, the development of the TN in the mouse-eared bat (Myotis myotis) from the mid-embryonic period to the adult stage is monitored. The data of this investigation conﬁrm the high variability of the terminal nerve (Huber and Guild, 1913; McCotter, 1913, 1915; Brookover, 1914, 1917; Johnston, 1914; Larsell, 1918, 1950; Humphrey, 1940; Pearson, 1941; Sinclair, 1951a, 1951b, 1951c; Bojsen-Møller, 1975; Cooper and Bhatnagar, 1976; Brown, 1980, 1987; Haymaker et al., 1982; Wirsig, 1985; Buhl and Oelschläger, 1986; Wirsig and Leonard, 1986a, 1986b; Jastrow, 1995), including the fact that the TN may be unilateral (Jastrow et al., 1998). Analysis of many ontogenetic stages shows a concentration of neurons in some ganglia and along ﬁber bundles and the persistence of a considerable number of TN neurons up to the adult mouse-eared bat. MATERIALS AND METHODS The terminal nerve and its ganglia were analyzed in 27 complete microslide series of the mouse-eared bat (Myotis myotis; Table 1). Most of these series belong to the collection of the late Professor Dietrich Starck. The deeply anesthetized animals had been ﬁxed in formaldehyde, dehydrated in ethanol, transferred to xylene, and embedded in parafﬁn wax. Blocks were cut on a sled microtome at a thickness of 10–40 mm. Sections were consecutively mounted and stained mostly with azocarmine and aniline blue (Azan) or with hematoxylin-eosin (HE). Using an Olympus BHS microscope, all TN perikarya were drawn with the help of a camera lucida at a magniﬁcation of 400–1,0003. Only a few specimens [18.3, 19, 22 mm crown-rump length (CRL), including the adult] were not entirely drawn since here the neurons of the terminal nerve could be unambiguously identiﬁed as a continuum of cells and connecting ﬁbers. TN neurons were distinguished from other tissues by their large, intensely stained acidophilic perikarya as well as by the size and morphology of their nuclei and nucleoli. Nevertheless, from mid-embryonic to early fetal stages, it was not always possible to make a clear distinction between single perikarya of the TN and immature cells of Bowman’s glands in the nasal mucosa. In sagittal series, graphic reconstructions of the terminal nerve and its ganglia were drawn by adding relevant structures from parasagittal microslides to the image of a mid-sagittal section with a Makropromar microprojector (Leitz, Germany). In case of other sectional planes (horizontal, transverse), such reconstructions obtained from the preceding and the following sagittal series were superimposed on one paper and adapted to each other in order to create a new sagittal view. Thereby outlines were TERMINAL NERVE IN BATS interpolated regarding the size of other structures, e.g., the nasal septum as well as differences in CRL. The new mediosagittal image was then modiﬁed according to the topography of TN structures in the microslide series under scrutiny. Finally all drawings were digitized using an HP ScanJet 4c. For better comparison, all animals had been classiﬁed as to their ontogenetic stage with the method of Štěrba (1985, 1990). This method uses the CRL, total length, head length, and morphological/histological criteria as, e.g., branchial arches, the development of hands and feet, eye and ear, as well as hair and cartilage/bone formation. The relevant criteria were described in detail for Myotis myotis by Štěrba (1990). The main characteristics of Štěrba’s ontogenetic levels in correlation with the staging of the Carnegie system (referring to embryos only) are as follows: level 1 stage 6, primitive streak; level 2 stage 9, ﬁrst somites (1–7); level 3 stage 13, limb buds, four branchial arches, otic vesicle closed; level 4 stage 16, retinal pigment present, handplate, distinct hemispheres, detached lens vesicle; level 5 stage 19, marked pinna, indented handplate, obliterated lens cavity, follicles of tactile hair (vibrissae) on upper lip; level 6 stage 23, fused palate, separated toes, ﬁrst ossiﬁcation, hair follicles on body; level 7 (fetal period), eyelids fused, retroposition of umbilical hernia; level 8, numerous skinfolds present; level 9, eruption of tactile hairs on lip; level H, hair growth all over body; level E, eyelids separated; level N ¼ newborn. In addition, developmental criteria of the brain and the olfactory bulb were used (Jastrow, 1995). 1203 Fig. 1. Mouse-eared bat (Myotis myotis) specimen of 6.5 mm CRL. Parasagittal section demonstrating the vicinity of the telencephalic wall (T) and the olfactory placode (OP). The primordial ganglion of the terminal nerve (between arrows) is located in the mesenchyme (Me) close to the immature meninx (M) and the rostrobasal wall of the telencephalic vesicle (T). A, artifacts; nc, primitive nasal cavity; bv, region of blood (vessel) formation from mesenchyme; V, primitive ventricle. RESULTS The youngest specimen [6.5 mm CRL; stage 4, after Štěrba (1990)] shows neuroblasts of the terminal nerve (Nervus terminalis) in one large irregular primordial ganglion near the rostrobasal wall of the telencephalic vesicle (Fig. 1, arrows). Some neuroblasts are leaving the olfactory placode (OP; Fig. 2), whereas neuron clusters and single cells are located in the mesenchyme between the placode and the ganglion (Fig. 1). Only one nasal pit is present in this specimen, indicating a malformation of the nasal region anlage. In principle, neurons are assembled here in a continuum from the placode to the brain wall. Whereas very few outgrowing olfactory nerve ﬁbers are obvious, still no distinct ﬁber bundle can be recognized as the terminal nerve. On the other hand, immature neurons of the TN do not show any mitotic ﬁgures; such are only seen in neuroblasts located in the superﬁcial layer of the olfactory placode as one (?) source of TN neurons. The early neuroblasts of the TN are spherical to ovoid in shape, with a maximal diameter of about 10 mm. Their nucleus represents up to 90% of the perikaryon area in transverse section; cell processes are hardly detectable and the nuclear membrane stains more intensely for Azan than the cell membrane. The average diameter of cells is about 12 mm. Apart from this, the darker cytoplasm and the higher nucleus-to-cytoplasm ratio allow the discrimination of immature TN neurons from surrounding or interposed mesenchyme cells. Nasal and meningeal portions of the TN cannot be recognized here because no anlagen of the nasal septum and the cribriform plate are detectable. Fig. 2. Mouse-eared bat specimen of 6.5 mm CRL. Evasion of potential terminal nerve neurons (between arrows) from the olfactory placode (OP) into neighboring mesenchyme (Me). Note some mitotic ﬁgures (Mi) near the surface of P bordering the nasal cavity (nc). The embryo of 9 mm CRL (stage 5) has an elongated nasal pit with the olfactory placode as a thickening of its dorsomedial epithelium. There are still many mitotic ﬁgures near its luminal surface and few TN neuroblasts were seen penetrating the basement membrane of the 1204 JASTROW AND OELSCHLÄGER Fig. 3. Reconstruction of a parasagittal section through the forehead of a 10 mm CRL embryo of Myotis myotis showing the terminal nerve (TN), its ganglia and ﬁber bundles, the anlage of the large lateral nasal gland1 (arrow), the nasal cavity (nc), and cartilage structures of the nose (ca). The main ﬁber bundle (mfb) connects the main nasal ganglion (N) in the nasal roof to the meningeal ganglion (M) located rostral to the telencephalic wall (T) and near the superior sagittal sinus (sss). Outside the sectional plane, in the submucosa of the nasal septum, the small ﬁber bundle (sfb) runs from the small nasal ganglion (S) to join the mfb in the caudal roof of the nose. The olfactory bulb and olfactory nerve ﬁbers are omitted. placode. The number of TN neuroblasts in the submucosa of the upper septal region is smaller than in the 6.5 mm embryo described above. Single cells and small cell clusters are present along distinct ﬁber bundles running in the dorsomedial submucosa toward the primordial terminal ganglion, which now exhibits a higher neuron number and density. The olfactory ﬁbers are much more abundant in this than in the preceding embryo but a complete TN is still not obvious. A slight protrusion of the rostroventral surface of the telencephalic vesicle is the ﬁrst sign of the primordial olfactory bulb, the formation of which is induced by olfactory ﬁbers entering the brain wall in this region. Some TN neurons (diameter of perikaryon about 12 mm) exhibit ﬁrst processes, some small chromatin granules in the nucleus, and one or two larger nucleoli. The embryo of 10 mm CRL (stage 5; parasagittal reconstruction in Fig. 3) has a much larger nasal cavity, cartilaginous anlagen of cranial bones, and shows the beginning of desmal ossiﬁcation (Frick, 1954; Jastrow, 1995). The volume of the developing olfactory bulbs is about three times larger than in the 9 mm CRL specimen and olfactory ﬁla are much more prominent. Mitotic ﬁgures are no longer seen in the OP, which is more difﬁcult to discriminate from surrounding respiratory epithelium and terminal neuroblasts penetrating the basal membrane of the OP are no longer encountered. A small recess is the ﬁrst sign of the developing duct of the ﬁrst large lateral nasal gland (Frick, 1954). There is one large nasal TN ganglion (N) in the submucosa dorsal and caudal to the recess just mentioned. This ganglion seems to have about the same amount of neurons as the rostral clusters of neuroblasts encountered in the 9 mm CRL specimen but there may be a lot of variability in the migration of the neuroblasts and in the formation of the TN ganglia. The TN of the 10 mm CRL embryo can be followed throughout the sections from the caudal end of N to the main ganglion (Fig. 3), which is much larger than in the previous specimen. From now on, it can be distinguished more easily from neighboring connective tissue due to the increasing concentration of neurons. Because this ganglion is located in the developing dura mater and arachnoidea, it is referred to as the meningeal ganglion (M), which corresponds to the anterior or rostral TN ganglion in the literature (e.g., Larsell, 1918, 1950; Grüneberg, 1973). The thin but distinct ﬁber bundle connecting the ganglia N and M is called the main ﬁber bundle (mfb) of the terminal nerve since it is much thicker than TN ﬁber bundles running in the submucosa of the middle and lower nasal septum. The mfb corresponds to the rostral or anterior branch of the TN in the literature. Some neurons and small TN cell clusters are present along its course. In the submucosa of the ventral nasal septum, a small TN ganglion (S) with its small ﬁber bundle (sfb) is located near the place where a vomeronasal organ is found in most vertebrates (for reviews, see Wysocki 1979). According to Frick (1954) and Mann (1961), this organ is lacking in Myotis myotis. In the deep submucosa on both sides of the nasal septum, sfbs course from S in the dorsal and caudal direction to join the mfb near the end of the nasal cavity (Fig. 3). This ﬁgure also shows that the combined ﬁber bundles (mfb, sfb) terminate in the main TN ganglion, which is now clearly delineated by a sheath of connective tissue. Ganglion M is located in the developing meninges between the mesethmoid spine (not shown in Fig. 3) (Frick, 1954) and the thickened rostromedial wall of the telencephalic hemisphere (T). The rostralmost part of the unpaired superior sagittal sinus (SSS) is encountered about 100– 250 mm dorsomedial to the ganglia M. Interestingly, the perikarya of the TN neurons seem to be more mature than those of the telencephalic wall in that they are a little larger, no longer spherical, and show distinct processes as predominantly multipolar neurons. Some TN neurons only show one or two processes indicative of (pseudo)unipolar, bipolar, or immature multipolar cells; their average cell diameter is about 14 mm. With a diameter of 7–10 mm, the nuclei make up about 70% of the perikarya in cross-section. This increase in cytoplasm volume together with a stronger basophilia is caused by an increase of RER and ribosomes, indicating the onset of neuronal function. The 11 mm CRL (late stage 5) specimen shows a tiny lumen in the blind-ending duct of the developing ﬁrst large lateral nasal gland (Lng1), which is now about 150 mm in length (Fig. 4). The olfactory placode is no longer distinct from the future respiratory epithelium, indicating that neurogenesis has stopped. At 12 mm (early stage 6), the rapidly growing Lng1 is located in the dorsorostral quarter of the nasal roof. The main nasal TN ganglion (N) is situated mediocaudal to it. Each TN ganglion comprises about 300–400 neurons TERMINAL NERVE IN BATS Fig. 4. Parasagittal section through the anterior nasal roof of the 11 mm CRL Myotis myotis embryo (A). The main ﬁber bundle (mfb) and the small nasal ganglion (S) are located close to the cartilage of the nasal septum. B shows the opening and a part of the duct of the large lateral nasal gland1 (Lng1) in close vicinity to the main TN nasal ganglion (N). A, artifacts; nc, nasal cavity. (number of nasal TN perikarya estimated in these animals). The mfb shows 40–50 mostly spindle-shaped neurons along its course, some of which lie scattered, whereas others are grouped in clusters of up to 15 perikarya. The TN ﬁber bundles run parasagittally in the gutter between the nasal septum and the nasal roof, close to the perichondrium. Here they pass the mesenchyme of the future cribriform plate and reach the developing meninges, where they end in the ganglion M. One small nasal ganglion S was present near the incisive canal dorsal to the anterior paraseptal process (Frick, 1954). On each side, an sfb with about 20 neurons along its course originates from this ganglion to join the mfb in the caudal third of the nose. The ovoid ganglia M are bordered by meningeal ﬁbroblasts; each is situated between the mesethmoid spine and the developing olfactory bulb, which is much larger in the 12 mm specimen. As to the blood vessels in this area, the rostrobasal portion of the sss is encountered about 50 mm dor- 1205 somedial to the meningeal ganglia (M) and some branches of the posterior cerebral artery (Grosser, 1904; Oelschläger, 1988) are seen in close vicinity. In all the ganglia mentioned, perikarya of the TN are associated in groups of 4–30 cells and the groups separated from each other by thin nerve or connective tissue ﬁber bundles. Most of the somata have diameters of 12–15 mm and show three to four processes. The nucleus (diameter: 10–12 mm) covers 50%–70% of the cross-sectional area in these perikarya. One larger or up to four smaller nucleoli are located in the heterogeneous nucleoplasm. Regarding the shape of their somata and processes as well as their staining characteristics, TN neurons are advanced in differentiation compared to those of the CNS. The two embryos of 13 mm CRL (stage 6) show the differentiation of additional large lateral nasal glands (Frick, 1954; Jastrow, 1995). In the olfactory bulb, some layers can be distinguished now: glomerular layer (about 25 mm thick), external plexiform layer (thickness 20 mm), internal plexiform layer and inner granular layer (together about 130 mm), and subependymal layer (60 mm). The characteristic mitral cells (mitral cell layer), however, are not yet encountered in these specimens. The duct of Lng1 turns from the mediorostral opening at the anterior third of the nasal roof laterocaudally to its blind end (Fig. 5). The main nasal TN ganglia (N), situated mediodorsal to it, cause slight protrusions of the epithelium in the rostral nasal roof into the nasal cavity (not shown). Along a single mfb, only about 15 spindleshaped terminalis neurons are seen in very small ganglia. On both sides, an sfb originates from a small ganglion (S). The TN enters the cranial vault through the anteriormost and medialmost foramen of the incipient cribriform plate, together with a thin olfactory ﬁber bundle (Foramen olfacto-terminale) and a tiny artery next to the mfb. Further caudally, each mfb enters the main ganglion (M). These two ganglia are spherical to ovoid and have several processes; among typical TN neurons they contain about 15 cells with a strikingly large perikaryon (diameter: up to 22 mm; nuclear diameter: up to 18 mm). In the two 14 mm CRL (late stage 6) specimens, the entire cribriform plate is present for the ﬁrst time. Apart from that, morphological ﬁndings are similar to those in the younger animals but the number of neurons along the nasal ﬁber bundles is smaller and the meningeal ganglia have some plump protrusions. The main nasal ganglia exhibit 10 neurons of noticeable larger perikaryon size (diameter: up to 22 mm). The 15 mm CRL (early stage 7) specimen for the ﬁrst time shows Bowman’s glands in the medial nasal submucosa. In principle, the olfactory bulb and the cerebral cortex now exhibit the characteristic layers, although typical mitral or pyramidal neurons are still not obvious. The course of the TN and the location of its ganglia resemble previous stages but the ganglia S are smaller, and there are thin extensions of the meningeal ganglia M. Somata are no more located along ﬁber bundles (mfb, sfb) in the deep nasal submucosa. When neurons are sectioned at their maximum diameter, the nucleus comprises about 60% of the whole perikaryon. Most of the TN neurons are multipolar and the few particularly large somata found in the two preceding embryos are no longer detectable. 1206 JASTROW AND OELSCHLÄGER Fig. 5. Horizontal reconstruction of the anterior nasal roof in a 13 mm CRL Myotis myotis demonstrating the course of the large lateral nasal gland1 (Lng1), the main TN nasal ganglia (N), from which the main nasal ﬁber bundles (mfb) run caudally. Whereas Lng1, N, and mfb are located in the nasal roof, the small nasal ganglia (S) as the origins of the small TN ﬁber bundles (sfb) are located close to the base of the nasal septum. From here the sfb ascends to join mfb in the caudal third of the nasal roof (asterisk). The 16 mm CRL specimen (stage 7) has a large main nasal ganglion (N) on each side with 1,236 cells on an average. Both ganglia are situated in the rostral quarter of the nasal roof immediately medial to the duct of Lng1 (Fig. 6) and are surrounded by a plexus of small blood vessels. Next to a thin artery, the mfb runs from the caudal end of N to the meningeal ganglion (M). There are not only one but even two sfbs on both sides, each with two ganglia, a larger (S1) and a smaller one (S2). The unusually large ganglion S1 with over 1,000 neurons is present at the base of the nasal septum in the right nasal submucosa. A rostral sfb, adjacent to blood vessels, ascends from the dorsocaudal end of S1 and runs rather straight to the nasal roof to join the mfb. The small ganglion S2 with only 23 neurons is attached to the rostral sfb about 200 mm caudal to N. A second smaller sfb rising from S2 is free of TN neurons and joins the mfb further caudally. A similar situation is present on the left side, with a much smaller ganglion S1 (110 neurons), whereas ganglion S2, located dorsocaudal to S1, contains 33 neurons. Both meningeal ganglia (M) are ovoid, with extensions toward the mfbs. From a ventrocaudal protrusion of the ganglia M, a few ﬁbers can be followed a short distance along medial olfactory ﬁla in the direction of the medial brain wall, but an entrance of the TN into the CNS is not detectable. The perikarya of 20 TN neurons located in M and N are probably precursors of the type 2 neurons (see below, specimens of 21 mm CRL) and are hardly stained. Whereas in younger specimens, one or two nucleoli are predominating, TN neurons of this animal show three to four nucleoli in their nucleoplasm. The majority of neurons are multipolar with three to ﬁve processes; only some are bipolar. In the 18.3 mm CRL mouse-eared bat (early stage 8, Fig. 7), the TN can be followed into the brain wall for the ﬁrst time. Moreover, initial mitral cells are now obvious in the olfactory bulbs. There are no ganglia S or sfbs in this animal. Each ganglion N (not shown in Fig. 7) gives rise to a long mfb lacking any perikarya Fig. 6. Parasagittal reconstruction of the anterolateral nasal roof of a 16 mm CRL mouse-eared bat specimen with a projection of the large lateral nasal gland1 (Lng1), which is situated lateral to the sectional plane, the main TN nasal ganglion (N) from which the main nasal ﬁber bundle (mfb) runs caudally and some Bowman’s glands (B). Note the length of the duct and the bulbous blind ending of the large lateral nasal gland1. Fig. 7. Three-dimensional reconstruction of the rostral neurocranium of a 18.3 mm CRL Myotis myotis specimen showing the meningeal ganglion (M), the central ﬁber bundle (c), its ganglion (C), and the entrance (asterisk) of the terminal nerve (TN) into the medioventral telencephalon (T). The main TN ﬁber bundle (mfb), which runs from the main nasal ganglion rostralward, is accompanied by a branch of the posterior cerebral artery (a), another branch of which penetrates the meningeal ganglion (M). The bizarre ganglion with its ﬁnger-like protrusions is located between the olfactory bulb (ob), the spina mesethmoidalis (sme, cut), and the rostroventral part of the superior sagittal sinus (sss). II, optic nerve; cc, corpus callosum; V, septal ventricle. along its course toward the bizarre ganglion M. The central part of each TN (c) originates from a posterior projection of M, runs close to the medialmost olfactory ﬁla, and enters the brain wall (asterisk) immediately caudal to a small central ganglion (C; 10 and 11 perikarya, respectively) in the area of the primordial anterior olfactory nucleus (Fig. 7). The multipolar neurons in C do not differ from other neurons of the TN and are somewhat larger than the neurons in the brain wall, which begin to develop initial processes. The fetus of 19 mm CRL (early stage 8) lacks the ganglia N and the sfbs on both sides. Other ﬁndings largely correspond to those reported for the previous specimen. However, the duct of the Lng1 is now considerably longer and extends over 15 mm from its opening in the anterior medial to the lateral nasal roof and further caudal TERMINAL NERVE IN BATS to the lateral nasal wall. It ends in the body of the gland, which has been formed in the middle third of the deep submucosa, some millimeters above the nasal ﬂoor (Frick, 1954). Five neurons of the left ganglion C seem to have entered the brain wall in the direction of the septum and hypothalamus. These immigrated peripheral TN cells differ from cerebral neurons by the larger size of their perikarya and nuclei and slightly more intense staining of the cytoplasma. The Myotis specimens of 21 mm CRL (stage 8) show some larger mucus glands in the nasal submucosa near the ganglia N. The irregular ganglia M show several ﬁnger-like extensions. The small ganglia C are located in the pia mater and comprise 10 neurons per side. In contrast to the younger specimens, there is a new cell type present in all the ganglia except for C. These type 2 cells (for potential precursor cells, cf. 16 mm specimen above) make up about 30% of all TN neurons and differ from the type 1 cells by a nearly unstained cytoplasm, a darker nucleus with condensed chromatin, only 1–2 bigger instead of 2–4 smaller nucleoli, and slightly reduced average cell and nuclear diameters. About 2% of the neurons are intermediate in appearance, with a more or less spherical perikaryon and maximal diameters of 16– 20 mm, hardly staining cytoplasm, and a karyoplasm darker than in type 1 cells. The great majority of all cells are multipolar; few are bipolar. The two fetuses of 22 mm CRL (stage 8) lack the small ganglia (S) and their ﬁber bundles (sfb). Apart from their processes and topography, it is rather difﬁcult to discriminate fetal nasal TN neurons from epithelial cells of Bowman’s glands, which now are present throughout the nasal submucosa. As in the 21 mm CRL specimen, there are no perikarya along the course of the mfb. The branched meningeal ganglia M are elongated rostrocaudally and located close to the frontal part of the superior sagittal sinus, the medialmost olfactory ﬁber bundles, and branches of the posterior cerebral artery. Central ﬁber bundles (c) are only found in one of the two specimens. One central ganglion (C) with 10 perikarya is present on the right. It is located close to the entrance of c into the brain, i.e., in the area of the prospective anterior olfactory nucleus. On the left, the situation is similar. About 60% of the TN neurons belong to type 1, 38% to type 2, and 2% are intermediate. As in the previous specimen (21 mm CRL), perikarya of a single type are not clustered in the ganglia but mixed with the other types and cell groups are separated from each other by delicate ﬁbers with small intermingled cells resembling ﬁbrocytes. The three specimens of 23 mm CRL (late stage 8) show ﬁrst mitral cells in their thickened olfactory bulbs and pyramidal neurons can now be distinguished in the prospective cortex of the enlarged cerebral hemisphere. In the rostral quarter of the nasal roof, medial to the duct of the ﬁrst large lateral nasal gland, all three animals show main nasal ganglia (N), which do not consist of one compact cluster of cells but have bizarre ﬁngerlike protrusions, and all of them show slender, elongated nasal ganglia S on both sides. The latter ganglia occur in the deep submucosa at about half-length of the basal nasal septum close to larger vessels and in the vicinity of larger glands (Fig. 8a). The bizarre meningeal ganglia M consist of type 1 cells (about 60%) and type 2 cells (40%; Fig. 8b), whereas the ganglia N and S have 1207 Fig. 8. Oblique sagittal section of the nose of a 23 mm CRL Myotis specimen with the cartilaginous nasal septum (center) as a bridge between the two nasal cavities (nc). A: Overview of the nose showing part of the duct of the large lateral nasal gland1 (Lng1), a small portion of the main TN nasal ganglion (N), the small nasal ganglion (S), the small ﬁber bundle (sfb), and the meningeal ganglion (M) of the other side. B: Detail of M. Note that type 2 TN neurons (2) have more condensed nuclei and nearly unstained cytoplasm in comparison to type 1 neurons (1). T, Telencephalon. slightly more type 1 neurons ( 70%). No TN perikarya are obvious along any TN ﬁber bundles here (mfb, sfb, c). Small central ganglia (C) are situated on both sides in all animals, in close vicinity to the entrance of the central TN ﬁber bundle (c) into the brain. This entrance is located either in the ventral lamina terminalis, i.e., medial to the basis of the olfactory bulb and corresponding with the situation shown in Figure 7 (asterisk), or a few hundred micrometers further rostralward in the area of the anterior olfactory nucleus. The situation in the 27 mm CRL bat fetus (stage 9) is quite similar to that of the previous specimens, but with the following exceptions: there are two and three ganglia S (left, right), respectively, in the middle part of the nasal septum and two sfbs on the right side (Fig. 9). The fan-shaped elongate main nasal ganglia (N) are extremely large. On both sides, each ganglion is situated medial to the opening and initial part of the Lng1, which now has a total length of about 2,5 mm. Processes of the large meningeal ganglia (M; Fig. 10) with densely packed neurons extend forward to the medialmost ﬁla olfactoria and backward to the central ﬁber bundle of the TN (c), which lacks ganglia on both sides. Compared to the preceding animals, there are no major differences in the 30 mm CRL specimen (stage H). A small nasal ganglion (S) is only present on the left 1208 JASTROW AND OELSCHLÄGER Fig. 9. Parasagittal reconstruction of the nose of the 27 mm CRL Myotis myotis specimen showing the large TN nasal ganglion (N), the two small nasal ganglia (S1, S2), and their two ﬁber strands (sfb1, sfb2) joining the main ﬁber bundle (mfb), which enters the cranial vault through the cribriform plate (cp) to reach the meningeal ganglion (M). From here, a small central ﬁber bundle (c) runs to the entrance of the terminal nerve into the rostromedioventral telencephalon (asterisk). Fig. 10. Myotis myotis of 27 mm CRL. A: Original parasagittal section with the maximal sectional area of the largest TN meningeal ganglion (M) of all specimens investigated. Note the duct of the large lateral nasal gland1 (Lng1) sectioned as well. B: Detail. A, cutting artifacts; a, branch of A. cerebri posterior; cp, cribriform plate; nc, nasal cavity; of, olfactory ﬁla; sss, superior sagittal sinus; T, telencephalon. and lacking on the right side, and so is its ﬁber bundle (sfb). The larger TN ganglia contain less neurons than in the 27 mm specimen. The shape of the bizarre ganglion M, situated close to a rostral (nasal) branch of the Fig. 11. Myotis myotis of 30 mm CRL. Three-dimensional reconstruction of the right meningeal ganglion (M) and its topographical relations seen from rostral, dorsal, and the right. Note some terminal nerve (TN) neurons (asterisk) located along the main ﬁber bundle (mfb) near the Foramen olfacto-terminale of the cribriform plate (cp), which contains bundles of terminal and medialmost olfactory ﬁbers (of) and a small artery (a; nasal branch of the posterior cerebral artery). The central terminal ﬁber bundle (c) enters the brain some millimeters further caudally and more ventrally (not shown). Olfactory bulbs were omitted for clarity. cd, caudal; d, dorsal. posterior cerebral artery, is shown in Figure 10. Central ganglia (C) are lacking. Type 2 TN neurons are no longer detectable in the nasal ganglia N and S and there are only a few of them left in the meningeal ganglion M. Some spindle-shaped TN neurons appear mainly in the ganglia N and S. On the right side, a few TN neurons are scattered along the main ﬁber bundle (mfb) rostral to a ﬁnger-like extension of the ganglion M. Some of these cells are located in the rostralmost and medialmost foramen of the cribriform plate (Fig. 11, Foramen olfacto-terminale), in close vicinity to a small artery and the rostral- and medialmost olfactory ﬁber bundles. A few other TN neurons lie further rostrally in the caudalmost part of the nasal roof (Fig. 11, asterisk). The two subadult bats (stage E) have their main nasal ganglia (N) located in the rostral sixth of the nose, which is considerably elongated in comparison with the preceding fetuses. The duct of Lng1 extends further caudally, where its body is located some millimeters above the nasal ﬂoor on the basal lateral wall of the nose. The initial segment of the duct is located lateral to the ganglia N for a few mm but does not seem to have any connections to it. Small nasal ganglia and sfbs are not detectable in both animals. Two of the main ganglia (M) are of a bizarre appearance, whereas the main nasal ganglia (N) seem to have dissolved into 3–4 smaller ganglia of different size, all of them being located in close vicinity to a nasal branch of the posterior cerebral artery. Central ﬁber strands (c) and their ganglia, for the most part, are not obvious in the sections, some of which are destroyed in the region of interest in both specimens. In one animal strand c can be followed as far as the mid-basal forebrain; TERMINAL NERVE IN BATS 1209 DISCUSSION Topography and Development of TN Fig. 12. Original parasagittal section of the adult Myotis myotis. A: Survey demonstrating the topography of the TN meningeal ganglion (M) with the adjacent posterior cerebral artery (Acp) and olfactory ﬁbers (of). B: Detail of the meningeal ganglion with the entrance of the main TN ﬁber bundle (mfb; arrows) and a few scattered TN neurons (asterisks). Note that there are no longer different types of TN neurons. A, artifact; nc, nasal cavity; T, telencephalon. however, the area of entry into the brain wall is destroyed. No type 2 TN neurons are present here. In the adult bat (late stage E), the main TN nasal ganglia (N) are still located in the rostral sixth of the medial nasal roof, medial to the ﬁrst few mm of the duct of Lng1. The ganglia N, which are usually fan-like in younger animals, on both sides seem to have divided into two neighboring ganglia. Each ganglion N is elongated toward the mfb, which joins its neighbor a few hundred mm further caudally. Together, the ﬁbers run along the medial nasal roof and enter the cranial vault through the Foramen olfacto-terminale (cf. Fig. 11). No neurons are encountered along the course of the mfbs. Small nasal ganglia or ﬁber bundles are not present. Only in this adult specimen, the large meningeal ganglion (M) on both sides is directly attached to the posterior cerebral artery (Fig. 12). Whereas on the right, the ganglion is nearly spherical, it is more elongate on the left. In addition, some single TN neurons are scattered in the meninges close to the ganglia M. A central ﬁber strand (c) is not obvious. All of the multipolar TN neurons belong to cell type 1 (Fig. 12b). In mammals, olfactory placodes become evident as oval-shaped, thickened areas of the anterolateral ectoderm of the head in the late ontogenetic stage 2 of Štěrba’s (1985, 1990) classiﬁcation system. In the mouseeared bat, Myotis myotis, this should happen at about gestational day (D) 25 and at a CRL of about 3.8 mm [calculated on the basis of Štěrba (1990)]. In early stage 3, the OPs show horseshoe-like ridges and begin to develop simple olfactory pits. In Myotis myotis, this presumably takes place at about D27 and a CRL of 4 mm. In recent studies, Whitlock et al. (2003, 2004) demonstrated the origin of at least the GnRH- immunopositive subset of TN cells from the anteriormost edge of the neural crest in D3 zebraﬁsh [probably corresponds to stage 3 of Štěrba (1990)]. In human embryos, the neural crest was ﬁrst observed by Müller and O’Rahilly (2004) in the Carnegie stage 13 [about D27, which corresponds to stage 3 of Štěrba (1990)]. Müller and O’Rahilly (2004) investigated over 300 serially sectioned human embryos and clearly demonstrated nasal crest material in several Štěrba stage 4 specimens directly underneath the basement membrane of the olfactory placode. This material originates from the nasal plate at the neurosomatic junction, as does the neural crest, and contains LHRH precursor cells (Verney et al., 2002). Müller and O’Rahilly (2004) observed that crest cells began to adhere to each other and to form cords, which a little later began to migrate toward the region of the olfactory tubercle. In this context, one can assume that the majority of TN neurons, particularly the endocrine cell populations of the TN in humans, originate from crest material of the neurosomatic junction, which is a major source of endocrine cells in general (von Bartheld and Baker, 2004; Whitlock, 2004). In contrast to this, a most recent investigation of the Indian major carp (Cirrhinus mrigala) by Biju et al. (2005) very clearly demonstrated GNRH peptide only in surprisingly well-differentiated bipolar epithelial cells of the olfactory placode and not in the neural crest on D1 [probably corresponding to stage 3 of Štěrba (1990)]. In this context, it should be noted that neural crest cells may also invade the placode (von Bartheld and Baker, 2004). Almost certainly the cells that were noted to leave the placode in the youngest specimens (6.5 and 9 mm CRL) of this investigation belong to another, probably nonendocrine cell population of the TN that probably is by far larger than the endocrine one. In bats, initial terminal nerve neuroblasts (TNn) and Schwann cells, ensheathing the olfactory and TN ﬁbers later, leave the medial wall of the OP in stage 4. Some immature neurons penetrate the internal limiting membrane of the OP in the two youngest specimens investigated by Brown (1980). The smallest embryo of this study (stage 4; 6.5 mm CRL) already shows an aggregation of TNn near the surface of the telencephalic vesicle; thus, the onset of TNn migration must have happened earlier in ontogenesis, probably early in stage 4. This is conﬁrmed by Brown (1980), who could not detect any TN material in smaller/younger embryos of other bats (Tadarida brasiliensis, Myotis velifer, and Myotis lucifugus) of 6 mm CRL (stage 3), whereas bats of more than 7 mm CRL (stage 4 and later) exhibit large 1210 JASTROW AND OELSCHLÄGER numbers of TNn adjacent to the OP (Brown, 1980; Oelschläger, 1988). In case neural crest material would be an important source of the TN cells in bats, the relevant neuroblasts should have appeared before stage 4 and Brown should have encountered them; therefore, a placodal origin of the TN material seems more convincing for the moment. In the next stage (5), when the ﬁrst ﬁla olfactoria can be detected (Brown, 1980; Oelschläger, 1988; Jastrow, 1995; this study), outgrowing TN ﬁbers gather in the mesenchyme rostral, ventral, and medial to the location where the primordial olfactory bulb will protrude from the telencephalic wall after being induced by olfactory ﬁbers (Oelschläger and Buhl, 1985a, 1985b; Oelschläger, 1989). At small pit is found close to the rostral end of the OP (Fig. 3), when the majority of the TNn have left the OP. It indicates the formation of the ﬁrst large lateral nasal gland. In the 9 mm CRL specimen (stage 5), most TN neurons seem to be present. Some of them are still on their way to the ganglion situated medial and rostroventral to the telencephalic vesicle. For the ﬁrst time, the TN is traceable as a distinct ﬁber bundle in the 11 mm CRL embryo, whereas the formation of the main nasal ganglia begins at 10 mm CRL with the association of cell clusters (Brown, 1980; Oelschläger, 1988; this study). Additional smaller ganglia (S) are present on both sides of the nasal septum in the region where a vomeronasal organ normally would be expected. The central portion of the terminal nerve develops slightly later; in this study, it was ﬁrst seen at 18.3 mm CRL (early stage 8): TN perikarya migrate and ﬁbers grow out from the meningeal ganglion (M) in the direction of the future attachment site of the terminal nerve. Finally, TN neurons penetrate the leptomeninx. In part, the migration of neurons and the extension of the nerve are supplemented by a passive shift due to allometric phenomena since structures exceedingly growing next to the TN (e.g., the nasal septum) tend to stretch its ﬁbers. Thus, the more or less spherical ganglion M of younger specimens may lengthen mainly by considerable proliferation of neighboring connective tissue in older specimens (Jastrow, 1995). Brown (1980) followed two central ﬁber bundles in embryos of 8 mm CRL (early stage 5) in the bat species mentioned above, one of the bundles entering the brain close to the primordial medial septal nucleus and the other in the area of the olfactory tubercle. In the material of this investigation, the ﬁrst reliable central ﬁbers are encountered at 18.3 mm CRL (early stage 8). In contrast to the ﬁndings of Brown (1980), who investigated Tadarida brasiliensis, Myotis velifer, and Myotis lucifugus, there is always one central TN ﬁber strand on each side in the mouse-eared bat. From stages 5 to 8, the transformation of migrating young round cells into differentiated neurons seems to be achieved [cell processes much more prominent, increased volume of perikaryon, LHRH synthesis (Schwanzel-Fukuda and Pfaff, 1990)]. The ability of the cell somata to migrate is reduced and seems to come to an end by the development of cell processes, retaining the mostly multipolar TNn in their locations. It is surprising that Brown (1980) found Tadarida embryos of 13 mm CRL (stage 6) to be the oldest animals showing a TN in this species. Maybe this author did not investigate the anterior nasal roof in older bats where, in the present material, the majority of the TNn are found. There is consid- erable variation in the small septal nasal ganglia (S) and their ﬁber bundles (sfb), which may be lacking, even on both sides (18.3, 19, 22 mm CRL specimens), or can contain up to 1,357 TN neurons. Usually, one ganglion S is seen per side, but two ganglia of different size in combination with two sfbs are encountered as well. Most probably these ganglia persist even if a vomeronasal organ is not present in its vicinity because the neurons are involved in quick operation of the local (respiratory) nasal mucosa. For instance, they could be involved in local reﬂex mechanisms as discussed later in Functional Aspects. The meningeal ganglion (M) and its main ﬁber bundle (mfb) were found in all bats investigated in this study. In contrast to other species such as, e.g., the human or cat (Brookover, 1914, 1917; Larsell, 1918; Jastrow, 1995), there is no nasal ﬁber plexus with integrated small ganglia in adult Myotis myotis. In the bat species investigated, the meningeal ganglia (M) seem to change in shape during ontogenesis; the initially irregular aggregation of cells (stages 4 and 5) becomes spherical (stage 6), develops processes of increasing length, and ﬁnally dissolves into two or more compact ganglia in the adult bat. No Vomeronasal Organ But a Large Gland The vomeronasal organ (VNO) is lacking in Myotis myotis (Frick, 1954), as is the case in other bats of the family Vespertilionidae (Mann, 1961; Baron et al., 1996; Wible and Bhatnagar, 1996) but not in Miniopterinae (Wible and Bhatnagar, 1996), such as, e.g., in Miniopterus schreibersi, where the organ exhibits a well-developed sensory epithelium (Cooper and Bhatnagar, 1976). In other mammals, TN ﬁbers usually reach the VNO running next to vomeronasal nerve ﬁbers, e.g., in the human (Brookover, 1917; Jastrow, 1995), cat, horse, and cattle (Larsell, 1918), rat (Bojsen-Møller, 1975), as well as rabbit (Huber and Guild, 1913). In the material scrutinized in this study, neither accessory olfactory bulbs (AOB) nor vomeronasal nerves are present. Conﬂicting data in the literature regarding the existence of a VNO in bats (Humphrey, 1936; Schneider, 1957; Mann, 1961; Cooper and Bhatnagar, 1976; Kämper and Schmidt, 1977; Bhatnagar et al., 1982) will not be discussed here. For reviews on the VNO and AOB, most of which are focused on bats, see Mann (1961), Wysocki (1979), Frahm and Bhatnagar (1980), Baron et al. (1996), Wible and Bhatnagar (1996), Bhatnagar and Meisami (1998), and Meisami and Bhatnagar (1998). On both sides, the investigated Myotis specimens measuring more than 10 mm CRL show a small duct in the anterior third of the nasal roof, which opens into the rostromedial nasal cavity, extends caudalward and lateralward, widens slightly, and has a blind end (Figs. 3–6, 8, and 10). This duct persists up to the adult and belongs to the large lateral nasal gland I (Frick, 1954), a fact conﬁrmed in this investigation. In an earlier study (Jastrow, 1995), the ending of the duct was not scrutinized in later fetal stages up to the adult and erroneously suggested to be a vomeronasal organ in an unusual location. The nasal opening of this duct is situated some micrometers away from the main nasal ganglion (N) of the TN, but there is no obvious connection to the latter. Later the distance between the duct and N increases to some hundred micrometers. TERMINAL NERVE IN BATS Functional Aspects Criteria for deﬁning what neural components can be included as constituents of the TN were discussed by von Bartheld (2004) with respect to the multitude of potential markers for its cells. When considering the ontogenetic development, topographical relationships, cytological criteria, and immunohistochemical data, it is obvious that the terminal nerve must have different functional implications, which in part may overlap with those of the main olfactory and accessory olfactory systems. Moreover, regarding these aspects, there are differences between species and groups of animals and there is even considerable inter- and intraindividual variation; thus, e.g., the complete TN may be absent on one side (Jastrow, 1995; Jastrow et al., 1998; this study). It is interesting to note that TN neurons precede those of the telencephalon as to developmental criteria: up to a CRL of about 19 mm, these peripheral neurons have larger perikarya, develop processes much earlier, and stain more intensely than central neurons (Jastrow, 1995; Müller and O’Rahilly, 2004). It was only for this reason that in the mouse-eared bat of 18.3 mm CRL, some TN neurons that were continuous with the central TN ﬁber tract could be detected inside the telencephalic wall with routine stain. As to their morphology, terminal nerve neurons comprise unipolar cells with pericellular baskets of gliocytes or ﬁbrocytes, bipolar or multipolar perikarya, which may have two nuclei (Larsell, 1918; Filogamo and Robecchi, 1969; Bojsen-Møller, 1975; Mendoza et al., 1982; Wirsig and Leonard, 1986b; Zheng et al., 1988, 1990; WirsigWiechmann, 1990; Oelschläger and Northcutt, 1992). In the material scrutinized in this investigation, apart from some spindle-shaped bipolar cells, multipolar TNn clearly predominated. This situation may be typical for bats in general (Brown, 1980, 1987; Oelschläger, 1988; this study), and such neurons may have a sensory or vegetative function. In their study on the TN of different mammals, Huber and Guild (1913) suggest that TN neurons mostly resemble those of cervical sympathetic ganglia. Accordingly, Larsell (1918) mentions a morphological similarity of the TNn to sympathetic neurons in that they also have pericellular baskets. In contrast to this, Brown (1980), who compared TN neurons to those of other ganglia in bats, found the best correspondence with small somata in the ganglia of cranial nerves V and IX. As a consequence, he favors a sensory function of the TN, which is supported by their placodal origin and ﬁber terminations between epithelial cells of the nasal mucosa. Ridgway et al. (1987) depicted TNn in the adult bottlenose dolphin with large diameters, sparse processes, and a dense sheath of potential gliocytes around their perikarya. These features are indicative of pseudounipolar neurons and may have a (somato?) sensory function (Oelschläger et al., 1987; Oelschläger and Northcutt, 1992). Interestingly, between these numerous large and rounded perikarya, a few bipolar LHRH-immunoreactive (ir) cells exist (Ridgway et al., 1987; Oelschläger and Northcutt, 1992). During ontogenesis, the TN is an important source of LHRH (¼ mGnRH) neurons that innervate/invade the hypothalamus to establish the hypothalamo-hypophysial-gonadal axis (Schwanzel-Fukuda et al., 1981, 1985; Oelschläger and Northcutt, 1992; Jastrow, 1995; Schwanzel- 1211 Fukuda, 1999). This outstanding developmental phenomenon seems to be obligatory for all mammals; disturbances in the formation of the TN result in sterile individuals with the olfacto-genital (Kallmann) syndrome (Kallmann et al., 1944; Gauthier, 1960; SchwanzelFukuda et al., 1989). The claim that GnRH neurons of the terminal nerve arise from the neural crest, however, should be viewed with caution as it is impossible to discriminate neural crest from placodal ectoderm in teleosts and because this is in contrast to the ﬁndings of Biju et al. (2005), who report that, in the Indian major carp, GnRH ﬁrst and only appears in cells of the olfactory placode. It was shown in zebraﬁsh that only the neuromodulatory mGnRH-positive cells are part of the terminal nerve (Whitlock et al., 2003). As mentioned above, these cells originate from the rostralmost edge of the neural crest (Whitlock, 2004). In early embryonal stages they associate with the material of the olfactory placode before it is morphologically distinct, and they migrate along TN ﬁbers into the hypothalamus during early fetal development. In contrast, the immediately endocrine hypothalamic GnRH neurons that express species-dependent forms of GnRH (Dubois et al., 2002) constitute another cell population originating from the anteriormost neural crest (adenohypophysial placode) and reach the hypothalamus independent of the TN (Kouki et al., 2001; Whitlock et al., 2003; Whitlock, 2004). Among the prenatal bats under scrutiny, there were no specimens from ontogenetic stages early enough to check what happens to cranial neural crest material. In teleost GnRH- and FMRFamide immunoreactive cells of the TN form a plexus in the upper inner plexiform layer of the retina before terminating on dopaminergic interplexiform amacrine cells or ganglion cells. By this and the fact that locally secreted GnRH causes horizontal processes to invaginate deeper into terminals (of cones only), the TN exerts a modulatory effect on the retina reducing its sensitivity during light adaptation processes (Behrens and Wagner, 2004). In this context, the topography of the TN ﬁbers can explain why odorants of food extracts inﬂuence the b-wave of electroretinograms, demonstrating an olfactory inﬂuence on the visually guided behavior in ﬁsh (Weiss and Meyer, 1988). Other implications of the terminal nerve may come via colocalized GnRH/FMRFamide ﬁbers, which innervate the basal forebrain and hypothalamus, including the hypothalamo-hypophyseal-gonadal axis, and inﬂuence the mating behavior (Oelschläger et al., 1998; Helpert, 2006). The GnRH component of the TN alone is not sufﬁcient to explain how this system manages peripheral and central sensory and motor functions related to behavior and reproduction (Wirsig-Wiechmann, 2004). In fact, the GnRH-immunoreactive cells comprise only a minor population (10%) of all TN neurons while other TN cell populations are labeled by a large variety of other markers as reviewed by von Bartheld (2004). This is not surprising since the neural crest is known to generate several types of endocrine/neuroendocrine cells (von Bartheld and Baker, 2004; Whitlock, 2004). Due to the fact that the present investigation could not involve any immunohistochemistry, the discussion was not extended beyond the essential issue of LHRH neurons and their migration. 1212 JASTROW AND OELSCHLÄGER A TN function related to olfaction is highly probable since the areas where the nerve enters the brain, i.e., the medial septal nucleus or its primordium, olfactory tubercle (Brown, 1980; Buhl and Oelschläger, 1986; Oelschläger et al., 1987; Oelschläger, 1988; Oelschläger and Northcutt, 1992; this study) are part of or related to the olfactory system. Free TN endings in and immediately beneath the mucosa of the nasal septum and in the vomeronasal organ were interpreted as part of a general visceroafferent system consisting of uni- and bipolar neurons (Huber and Guild, 1913; Larsell, 1918, 1950; Stewart, 1920; Simonetta, 1932; Simonetta and Magnoni, 1939; Pearson, 1941). Allen (1936) assumed that TN neurons are receptive to stimuli other than those adequate for trigeminal nerve terminals. Most probably, TN-speciﬁc odorants do not bind to superﬁcial cell surface receptors reaching into the nasal mucus as is the case in olfactory sensory cells but, in this case, agents must be able to penetrate somewhat deeper to reach free nerve endings between epithelial cells. It is possible that in Myotis myotis and other bat species that lack a vomeronasal organ, TN terminals can perceive pheromones so that a vomeronasal system is not required. However, it seems premature to suggest that the TN takes over the vomeronasal function when the main olfactory system is another candidate and by far more similar to the VN system. All primates have a functional main olfactory system, and this appears to be true for bats as well. Thus, the main olfactory system may well take over for the missing VNO in both Myotis (and many other bats) as well as in Old World primates (Smith et al., 2001). In this context, Biju et al. (2005) note that the occurrence of GnRH in olfactory receptor neuron-like cells of Indian major carp larvae coincides with the onset of feeding and suggest a food intake-related chemosensory perception of this TN cell population. Apart from the suggested shift of postnatal VNO function to the TN, the latter could also take over the prenatal functions of the VNO system, e.g., the generation of LHRH neurons usually located in the VNO. One could consider this since this investigation revealed that Myotis myotis not only lacks a VNO, but moreover not even forms a VNO primordium or an accessory olfactory bulb (at least in the investigated microslide series) as this is the case in many other mammals. There is a parallel provided by Catarrhine primates: adult Old World monkeys, including apes, and some humans lack a VNO, and lower Old World monkeys lose the VNO earlier in development than apes or humans, which appear to retain a vestige (Smith et al., 2001; for a comprehensive summary of VNO system regression in humans, cf. Trotier et al. 2000). Some studies have reported LHRH neurons in the human VNO or entangled with vomeronasal nerves during embryonic development (Boehm et al., 1994; Kjaer and Fisher-Hansen, 1996; SchwanzelFukuda, 1999). These neurons most probably belong to the terminal nerve. Further, LHRH neurons are associated with the TN of macaques (Quanbeck et al., 1997). These authors even describe two migratory ‘‘waves’’ of GnRH neurons in the macaque and related both of them to the TN. Since the TN provides a documented source of LHRH migration and a pathway for these neurons, it is certainly reasonable to suggest that it may make up for the absence of the VNO in Myotis. The scenario presented above relies on comparison to other taxa with similar characteristics. In a study on adult humans, no neurons or vomeronasal nerve ﬁber bundles were observed by Trotier et al. (2000). This suggests that in this special case, the VNO may not function as sensory organ. In contrast, all Old World monkeys studied to date do not develop a VNO vestige (Smith et al., 2001) and yet still have a TN. On the other hand, blood pressure monitoring and modulation as a function of afferent TN ﬁbers was inferred from potential sensory terminals that were encountered between smooth muscle cells in the Tunica media of adjacent cerebral arteries (Larsell, 1918; Pearson, 1941). In this context, the regulation of blood ﬂow to the olfactory bulb and the nasal submucosa could be realized by TN branches along the supplying blood vessels. Apart from this, it is most probable that there are anastomoses between the TN and autonomic nerve ﬁbers along these vessels. In the present study, nasal TN ﬁber bundles were observed in close vicinity of or directly attached to blood vessels. Therefore, it can be assumed that thin nerve terminals enter their walls. In the adult mouseeared bat, the immediate contact of meningeal ganglia with blood vessels on both sides also implies an innervation of the latter (in the case of M. myotis: posterior cerebral arteries) (Grosser, 1904). Unfortunately, investigated sections were too thick to follow potential TN ﬁber endings to their target structures. The swelling of the nasal mucosa in correlation with air ﬂow and composition is certainly inﬂuenced by free endings of the TN in the submucosa that may detect odorants or irritating agents and quickly mediate a reaction via efferent TN ﬁbers (perhaps of the same perikaryon in a nasal ganglion) to relevant glands and vessels (Holmgren, 1920; Larsell, 1950; Graziadei, 1976). Potential vasomotor ﬁbers terminating at smooth muscle cells of the tunica media of blood vessels were demonstrated by Larsell (1950) and are also seen along vessels of the septal mucosa and the anterior brain (Oelschläger, 1988). Haymaker et al. (1982) therefore suggest the involvement of the TN also in the cyclic swelling of the nasal mucosa. Although quite probable, with the small nasal ganglia and their ﬁbers located close to larger glands of the nasal septum (Fig. 6), a direct innervation of Bowman’s glands by terminals of multipolar TNn could not be traced in the sections of the Myotis myotis presently examined. Perhaps the TN houses the parasympathetic (efferent?) component of innervation leading to vasoconstriction. Free TN ﬁber endings were also considered to serve thermoreception (Grüneberg, 1973; Bojsen-Møller, 1975). Such an additional afferent innervation of nasal glands would allow the moistening of air by a reﬂex via the TN and thus serve thermoregulation. In vertebrates, thin ﬁber bundles of the TN were reported to be in close association with the nasopalatine nerve via anastomoses (Brookover, 1917). By this, the TN may modulate the activity of structures as different as glands and blood vessels in the nasal septum as well as the naris constrictor muscle (Wirsig-Wiechmann and Ebadifar, 2002). From the fact that, in some of the bats investigated in the present work, thin TN ﬁbers run from the large ganglion M into the meninges, it may be assumed that there is a contribution to their innervation with possible anastomoses to those branches of the trigeminal nerve being responsible for the meninges. TERMINAL NERVE IN BATS In addition to the TN, autonomic nerve ﬁbers carried by other nerves (mainly trigeminal branches) reach the nasal submucosa and inﬂuence its swelling, nasal air ﬂow, and secretion. The persistence of the TN into the adult stage could be a sign that this innervation alone is not sufﬁcient for the required protective mechanisms. Despite its connections particularly with the olfactory nerve, the TN has to be regarded an independent cranial nerve (Jastrow 1995; von Bartheld, 2004; Wirsig-Wiechmann, 2004). On the one hand, this is obvious from its origin, development, course, and attachment to the brain; on the other hand, from its presence in toothed whales. These animals completely lack the vomeronasal system and lose the anterior olfactory system (olfactory nerves, bulbs, and peduncles) during early fetal development and the concomitant total reconstruction of the nasal region as a consequence of their phylogenetic adaptation to aquatic habitats (Johnston, 1914; Sinclair, 1951a, 1951c, 1966; Oelschläger and Buhl, 1985a, 1985b; Buhl and Oelschläger, 1986; Oelschläger et al., 1987; Ridgway et al., 1987; Oelschläger, 1992; Cranford et al., 1996; Klima, 1999; Cranford, 2000; Huggenberger, 2003). In toothed whales, which echolocate by means of their highly modiﬁed upper respiratory tract (epicranial complex, nasal complex) (Cranford et al., 1996, Huggenberger, 2003), the terminal nerve attains a maximal number of neurons among the Mammalia (Buhl and Oelschläger, 1986; Oelschläger et al., 1987; Ridgway et al., 1987). The coincidence of this hypertrophy with the loss of the nasal chemoreceptor systems clearly demonstrates the considerable degree of independence of the TN from these systems. As is the case in the adult bottlenose dolphin (Ridgway et al., 1987), the terminal neurons in the adult Big Brown Bat (Eptesicus fuscus) for the most part do not express LHRH but may have to be attributed to other functional systems rather in microchiropterans (Oelschläger and Northcutt, 1992). It was suggested earlier that in toothed whales the TN innervates parts of the upper respiratory tract (Buhl and Oelschläger, 1986; Oelschläger et al., 1987) involved in the generation of sonar signals by means of a secondary pneumatically driven vocalization mechanism (Cranford and Amundin, 2004; Goodson et al., 2004). At the moment, however, it is unclear whether the TN can possibly play a similar role in the vocalization of bats. Microchiropterans use ultrasound signals of comparable intensity, directivity, and frequency range (Cranford and Amundin 2004). However, as the mouse-eared bat produces ultrasound signals with the larynx and emits them through the mouth (open-mouth vocalizer) and not the nose, it appears rather unlikely that the TN should be involved here in sound emission. ACKNOWLEDGMENTS Dedicated to the memory of Prof. Dr. Dietrich Starck (1908–2001), eminent authority of comparative morphological and embryological research. The authors thank Dr. Pavel Nĕmec (Biodiversity Research Group, Department of Zoology, Charles University, Prague, Czech Republic) for helpful discussion and the reviewers for constructive comments. LITERATURE CITED Allen WF. 1936. A comparative study of the olfactory and trigeminal reﬂexes elicited by various vapours in different mammals. J Wash Acad Sci 26:466–473. 1213 Baron G, Stephan H, Frahm HD. 1996. Comparative neurobiology in chiroptera, vols. 1–3. Basel: Birkhäuser Verlag. Behrens U, Wagner H-J. 2004. Terminal nerve and vision. Microsc Res Tech 65:25–32. Bhatnagar KP, Matulionis DH, Breipohl W. 1982. Fine structure of the vomeronasal neuroepithelium of bats: a comparative study. Acta Anat 112:158–177. Bhatnagar KP, Meisami E. 1998. Vomeronasal organ in bats and primates: extremes of structural variability and its phylogenetic implications. Microsc Res Tech 43:465–475. Biju KC, Gaikwad A, Sarkar S, Schreibman MP, Subhedar N. 2005. Ontogeny of GnRH-like immunoreactive neuronal systems in the forebrain of the Indian major carp, Cirrhinus mrigala. Gen Comp Endocrinol 141:161–171. Boehm N, Roos J, Gasser B. 1994. Luteinizing hormone-releasing hormone (LHRH)-expressing cells in the nasal septum of human fetuses. Brain Res Dev Brain Res 82:175–180. Bojsen-Møller F. 1975. Demonstration of terminalis, olfactory, trigeminal and perivascular nerves in the rat nasal septum. J Comp Neurol 159:245–256. Brookover C. 1914. The nervus terminalis in adult man. J Comp Neurol 24:131–135. Brookover C. 1917. The peripheral distribution of the nervus terminalis in an infant. J Comp Neurol 28:349–360. Brown JW. 1980. Developmental history of nervus terminalis in embryos of insectivorous bats. Anat Rec 196:23–35. Brown JW. 1987. The nervus terminalis in insectivorous bat embryos and notes on its presence during human ontogeny. Ann NY Acad Sci 519:184–200. Buhl EH, Oelschläger HA. 1986. Ontogenetic development of the nervus terminalis in toothed whales. Anat Embryol 173:285–294. Caldani M, Batailler M, Jourdan F. 1987. The sheep terminal nerve: coexistence of LHRH- and AChE-containing neurons. Neurosci Lett 83:221–226. Cooper JG, Bhatnagar KP. 1976. Comparative anatomy of the vomeronasal organ complex in bats. J Anat 122:571–601. Cranford TW, Amundin M, Norris KS. 1996. Functional morphology and homology in the odontocete nasal complex: implication for sound generation. J Morphol 228:223–285. Cranford TW. 2000. In search of impulse sound sources in odontocetes. In: Au WWL, Popper AN, Fay RR, editors. Hearing by whales and dolphins. New York: Springer. p 109–155. Cranford TW, Amundin M. 2004. Biosonar pulse production in odontocetes: the state of our knowledge. In: Thomas JA, Moss CF, Vater M, editors. Echolocation in bats and dolphins. Chicago: University of Chicago Press. p 27–35. Demski LS, Ridgway SH, Schwanzel-Fukuda M. 1990. The terminal nerve of dolphins: gross structure, histology and luteinizing hormone-releasing hormone immunocytochemistry. Brain Behav Evol 36:249–261. Dubois EA, Zandbergen MA, Peute J, Goos HJT. 2002. Evolutionary development of three gonadotropin-releasing hormone (GnRH) systems in vertebrates. Brain Res Bull 57:413–418. Filogamo G, Robecchi MG. 1969. Neuroblasts in the olfactory pits in mammals. Acta Anat 56:182–187. Frahm HD, Bhatnagar KP. 1980. Comparative morphology of the accessory olfactory bulb in bats. J Anat 130:349–365. Frick H. 1954. Die Entwicklung und Morphologie des Chondrocraniums von Myotis Kaup. Stuttgart: Thieme. Gauthier G. 1960. La dysplasie olfacto-génitale. Vienna: Springer. p 345–407. Goodson AD, Flint JA, Cranford TW. 2004. The harbor porpoise (Phocoena phocoena): modeling the sonar transmission mechanism. In: Thomas JA, Moss CF, Vater M, editors. Echolocation in bats and dolphins. Chicago: University of Chicago Press. p 64– 85. Graziadei PPC. 1976. Functional anatomy of the mammalian chemoreceptor system. In: Müeller-Schwarze D, Morell MM, editors. Chemical signals in vertebrates. New York: Plenum. p 435– 454. Grosser O. 1904. Zur Anatomie und Entwicklungsgeschichte des Gefäßsystems der Chiropteren. Anat Hefte 17:203–424. 1214 JASTROW AND OELSCHLÄGER Grüneberg H. 1973. A ganglion probably belonging to the N. terminalis system in the nasal mucosa of the mouse. Z Anat Entw Gesch 140:39–52. Haymaker W, Azcoaga JE, Kelly MA. 1982. The olfactory system. In: Haymaker W, Adams DR, editors. Histology and histopathology of the nervous system, vol. 2. Springﬁeld, IL: C. Thomas. p 2396–2527. Helpert C. 2006. Immunzytochemische Untersuchungen am Gehirn der großen braunen Fledermaus (Eptesicus fuscus) mit ‘‘molluscan cardioexcitatory Neuropeptide’’ (FMRFamid). Frankfurt am Main: J.W. Goethe-University. Holmgren N. 1920. Zur Kenntnis des Nervus terminalis bei Teleostiern. Folia Neurobiol 11:16–36. Huber GC, Guild SR. 1913. Observations on the peripheral distribution of the nervus terminalis in mammalia. Anat Rec 7:253–272. Huggenberger S. 2003. Functional morphology, development, and evolution of the upper respiratory tract in toothed whales. PhD thesis. Frankfurt am Main: Faculty of Biology, Johann Wolfgang Goethe University. Humphrey T. 1936. The telencephalon of the bat: I, the noncortical nuclear masses and certain pertinent ﬁber connections. J Comp Neurol 65:603–711. Humphrey T. 1940. The development of the olfactory and the accessory olfactory formations in human embryos and fetuses. J Comp Neurol 73:431–468. Jastrow H. 1995. Der Nervus terminalis der Säugetiere und des Menschen: ontogenetische, immunhistochemische und quantitative Untersuchungen. Frankfurt am Main: J.W. Goethe-University. Jastrow H, Burda H, Oelschläger HHA. 1998. Unilateral absence of the terminal nerve and distribution of gonadotropin-releasing hormone immunoreactive neurons in the brain of the common mole-rat (Cryptomys, Rodentia). Brain Res 813:229–240. Jennes L. 1989. Prenatal development of the gonadotropin-releasing hormone-containing systems in rat brain. Brain Res 482:97–108. Johnston JB. 1914. The nervus terminalis in man and mammals. Anat Rec 8:185–198. Kallmann F, Schönfeld WA, Barrera SA. 1944. The genetic aspects of primary eunuchoidism. Am J Ment Deﬁc 48:203–236. Kämper R, Schmidt U. 1977. Die Morphologie der Nasenhöhle bei einigen neotropischen Chiropteren. Zoomorphology 87:3–19. King JC, Anthony ELP, Gustafson AW, Damassa DA. 1984. Luteinizing hormone-releasing hormone (LH-RH) cells and their projections in the forebrain of the bat Myotis lucifugus lucifugus. Brain Res 298:289–301. Kjaer I, Fischer Hansen B. 1996. The human vomeronasal organ: prenatal developmental stages and distribution of luteinizing hormone-releasing hormone. Eur J Oral Sci 104:34–40. Klima M. 1999. Development of the Cetacean Nasal Skull. Adv Anat Embryol Cell Biol 149:1–143. Kouki T, Imai H, Aoto K, Eto K, Shioda S, Kawamura K, Kikuyama S. 2001. Developmental origin of the rat adenohypophysis prior to the formation of Rathke’s pouch. Development 128:959–963. Krey LC, Silverman AJ. 1978. The luteinizing hormone-releasing hormone (LH-RH) neuronal networks of the guinea pig brain: II, the regulation of gonadotropin secretion and the origin of terminals in the median eminence. Brain Res 157:247–255. Larsell O. 1918. Studies on the nervus terminalis: mammals. J Comp Neurol 30:3–68. Larsell O. 1950. The nervus terminalis. Ann Otol Rhinol Laryngol 59:414–438. Mann G. 1961. Bulbus olfactorius accessorius of Chiroptera. J Comp Neurol 116:135–144. McCotter RE. 1913. The nervus terminalis in adult dog and cat. J Comp Neurol 23:145–152. McCotter RE. 1915. A note on the course and distribution of the nervus terminalis in man. Anat Rec 9:243–246. Meisami E, Bhatnagar KP. 1998. Structure and diversity in mammalian accessory olfactory bulb. Microsc Res Tech 43:476–499. Mendoza AS, Breipohl W, Miragall F. 1982. Cell migration from the olfactory placode: a light and electron microscopic study. J Embryol Exp Morphol 69:47–59. Müller F, O’Rahilly R. 2004. Olfactory structures in staged human embryos. Cell Tiss Org 178:93–116. Muske LE. 1993. Evolution of gonadotropin-releasing hormone (GnRH) neuronal systems. Brain Behav Evol 42:215–230. Northcutt RG, Muske LE. 1994. Multiple embryonic origins of gonadotropin-releasing hormone (GnRH) immunoreactive neurons. Dev Brain Res 78:279–290. Oelschläger HA, Buhl EH. 1985a. Development and rudimentation of the peripheral olfactory system in the harbor porpoise Phocoena phocoena (Mammalia: Cetacea). J Morphol 184:351–360. Oelschläger HA, Buhl EH. 1985b. Occurrence of an olfactory bulb in the early development of the harbor porpoise (Phocoena phocoena L.). In: Duncker HR, Fleischer G, editors. Functional morphology of vertebrates. Stuttgart: G. Fischer. p 695–698. Oelschläger HA, Buhl EH, Dann JF. 1987. Development of the nervus terminalis in mammals including toothed whales and humans. Ann NY Acad Sci 519:447–464. Oelschläger HA. 1988. Persistence of the nervus terminalis in adult bats: a morphological and phylogenetical approach. Brain Behav Evol 32:330–339. Oelschläger HA. 1989. Early development of the olfactory and terminalis systems in baleen whales. Brain Behav Evol 34:171–183. Oelschläger HA. 1992. Development of the olfactory and terminalis systems in whales and dolphins. In: Doty RL, Müller-Schwarze D. editors. Chemical signals in vertebrates VI. New York: Plenum Press. p 141–147. Oelschläger HA, Northcutt RG. 1992. Immunocytochemical localization of luteinizing hormone-releasing hormone (LHRH) in the nervus terminalis and brain of the big brown bat, Eptesicus fuscus. J Comp Neurol 315:344–363. Oelschläger HA, Helpert C, Northcutt RG. 1998. Coexistence of FMRFamide-like and LHRH-like immunoreactivity in the terminal nerve and forebrain of the Big Brown Bat, Eptesicus fuscus. Brain Behav Evol 52:139–147. Pearson AA. 1941. The development of the nervus terminalis in man. J Comp Neurol 75:39–66. Quanbeck C, Sherwood NM, Millar RP, Terasawa E. 1997. Two populations of luteinizing hormone-releasing hormone neurons in the forebrain of the rhesus macaque during embryonic development. J Comp Neurol 380:293–309. Ridgway SH, Demski LS, Bullock TH, Schwanzel-Fukuda M. 1987. The terminal nerve in odontocete cetaceans: dolphins and other toothed whales. Ann NY Acad Sci 519:201–212. Ronnekeiv O, Resco JA. 1990. Ontogeny of gonadotropin-releasing hormone-containing neurons in early fetal development of rhesus macaques. Endocrinology 126:498–511. Schneider R. 1957. Morphologische Untersuchungen am Gehirn der Chiroptera (Mammalia). Abh Senckenberg Naturforsch Ges 496:1–92. Schwanzel-Fukuda M, Robinson JA, Silverman AJ. 1981. The fetal development of the luteinizing hormone-releasing hormone (LHRH) neuronal systems of the guinea pig brain. Brain Res Bull 7:293–315. Schwanzel-Fukuda M, Morell JI, Pfaff DW. 1985. Ontogenesis of neurons producing luteinizing hormone-releasing hormone (LHRH) in the nervus terminalis of the rat. J Comp Neurol 238:348– 364. Schwanzel-Fukuda M, Pfaff DW. 1988. A proposed migratory route for luteinizing hormone-releasing hormone (LHRH) neurons from the medial olfactory placode to the forebrain in the mouse. Soc Neurosci Abstr 14:984. Schwanzel-Fukuda M, Fadem BH, Garcia MS, Pfaff DW. 1988. Immunocytochemical localization of luteinizing hormone-releasing hormone (LHRH) in the brain and nervus terminalis of the adult and early neonatal gray short-tailed opossum (Monodelphis domestica). J Comp Neurol 276:44–60. Schwanzel-Fukuda M, Pfaff DW. 1989. Origin of luteinizing hormone-releasing hormone neurons. Nature 338:161–164. Schwanzel-Fukuda M, Bick D, Pfaff DW. 1989. Luteinizing hormone-releasing hormone (LHRH)-expressing cells do not migrate normally in an inherited hypogonadal (Kallmann) syndrome. Mol Brain Res 6:311–326. TERMINAL NERVE IN BATS Schwanzel-Fukuda M, Pfaff DW. 1990. The migration of luteinizing hormone-releasing hormone (LHRH) neurons from the medial olfactory placode into the medial basal forebrain. Experientia (Basel) 46:956–961. Schwanzel-Fukuda M, Reinhard GR, Abraham S, Crossin KL, Edelman GM, Pfaff DW. 1994. Antibody to neural cell adhesion molecule can disrupt the migration of luteinizing hormone-releasing hormone neurons into the mouse brain. J Comp Neurol 342:174– 185. Schwanzel-Fukuda M. 1999. Origin and migration of luteinizing hormone-releasing hormone neurons in mammals. Microsc Res Tech 44:2–10. Simonetta B. 1932. Origine e sviluppo del nervo terminale nei mammiferi: sua funzione e suoi rapporti con l’organo di Jacobson. Z Anat EntwGesch 97:425–463. Simonetta B, Magnoni A. 1939. Ricerche sulla presenze e sullo sviluppo del nervo terminale e dell’organo di Jacobson nei chirotteri. Arch Ital Anat Embriol 41:343–356. Sinclair JG. 1951a. The terminal olfactory complex in the porpoise. Texas J Sci 3:251–258. Sinclair JG. 1951b. A ganglionic plexus in the human falx cerebri. Tex Rep Biol Med 9:348–352. Sinclair JG. 1951c. Development of the anterior cerebral nerve plexus. Tex Rep Biol Med 9:805–810. Sinclair JG. 1966. The olfactory complex of dolphin embryos. Tex Rep Biol Med 24:426–431. Smith TD, Siegel MI, Bhatnagar KP. 2001. Reappraisal of the vomeronasal system of catarrhine primates: ontogeny, morphology, functionality, and persisting questions. Anat Rec 265:176– 192. Štěrba O. 1985. Ontogenetic levels in mammals. In: Mlı́kovský J, Novák VJA, editors. Evolution and morphogenesis. Prague: Academia. p 567–571. Štěrba O. 1990. Prenatal development of Myotis myotis and Miniopterus schreibersi. Folia Zool 39:73–83. Stewart FW. 1920. On the origin of the ganglion cells of the nervus terminalis of the albino rat. J Comp Neurol 32:99–111. Trotier D, Eloit C, Wassef M, Talmain G, Bensimon JL, Døving KB, Ferrand J. 2000. The vomeronasal cavity in adult humans. Chem Senses 25:369–380. Verney C, Lebrand C, Gaspar P. 2002. Changing distribution of monoaminergic markers in the developing human cerebral cortex with special emphasis on the serotonin transporter. Anat Rec 267:87–93. von Bartheld CS. 2004. The terminal nerve and its relation with extrabulbar ‘‘olfactory’’ projections: lessons from lampreys and lungﬁshes. Microsc Res Tech 65:13–24. 1215 von Bartheld CS, Baker CV. 2004. Nervus terminalis derived from the neural crest? A surprising new turn in a century-old debate. Anat Rec (New Anat) 278B:12. Weiss O, Meyer DL. 1988. Odor stimuli modulate retinal excitability in ﬁsh. Neurosci Lett 93:209–213. Whitlock KE, Wolf CD, Boyce ML. 2003. Gonadotrophin-releasing hormone (GnRH) cells arise from cranial neural crest and adenohypophyseal region of the neural plate in the zebraﬁsh, Danio rerio. Dev Biol 257:140–152. Whitlock KE. 2004. Development of the nervus terminalis: origin and migration. Microsc Res Tech 65:2–12. Wible JR, Bhatnagar KP. 1996. Chiropteran vomeronasal complex and the interfamilial relationships of bats. J Mamm Evol 3:285–314. Wirsig CR. 1985. The terminal nerve: a neuroanatomical and functional study in the golden hamster (Mesocricetus auratus). PhD dissertation, Gainsville, FL: University of Florida. Wirsig CR, Leonard CM. 1986a. The terminal nerve projects centrally in the hamster. Neuroscience 19:709–717. Wirsig CR, Leonard CM. 1986b. Acetylcholinesterase and luteinizing hormone-releasing hormone distinguish separate populations of terminal nerve neurons. Neuroscience 19:719–740. Wirsig-Wiechmann CR. 1990. The nervus terminalis in the chick: a FMRFamide-immunoreactive and AChE-positive nerve. Brain Res 523:175–179. Wirsig-Wiechmann CR, Ebadifar B. 2002. The naris muscles in tiger salamander: II, innervation as revealed by enzyme histochemistry and immunocytochemistry. Anat Embryol 205:181–186. Wirsig-Wiechmann CR. 2004. Introduction to the anatomy and function of the nervus terminalis. Microsc Res Tech 65:1. Wray S, Nieburgs A, Elkabes S. 1989a. Spatiotemporal cell expression of luteinizing hormone-releasing hormone in the prenatal mouse: evidence for an embryonic origin in the olfactory placode. Dev Brain Res 46:309–318. Wray S, Grant P, Gainer H. 1989b. Evidence that cells expressing luteinizing hormone-releasing hormone mRNA in the mouse are derived from progenitor cells in the olfactory placode. Proc Natl Acad Sci USA 86:8132–8136. Wysocki C. 1979. Neurobehavioral evidence for the involvement of the vomeronasal system in mammalian reproduction. Neurosci Biobehav Rev 3:301–341. Zheng LM, Caldani M, Jourdan F. 1988. Immunocytochemical identiﬁcation of luteinizing hormone-releasing hormone-positive ﬁbres and terminals in the olfactory system of the rat. Neuroscience 24:567–578. Zheng LM, Pfaff DW, Schwanzel-Fukuda M. 1990. Synaptology of luteinizing hormone-releasing hormone (LHRH)-immunoreactive cells in the nervus terminalis of the gray short-tailed opossum (Monodelphis domestica). J Comp Neurol 295:327–337.