Analysis of Developing Tooth Germ Innervation Using Microfluidic Co-culture Devices.

Innervation plays a key role in the development, homeostasis and regeneration of organs and tissues. However, the mechanisms underlying these phenomena are not well understood yet. In particular, the role of innervation in tooth development and regeneration is neglected. Several in vivo studies have provided important information about the patterns of innervation of dental tissues during development and repair processes of various animal models. However, most of these approaches are not optimal to highlight the molecular basis of the interactions between nerve fibres and target organs and tissues. Co-cultures constitute a valuable method to investigate and manipulate the interactions between nerve fibres and teeth in a controlled and isolated environment. In the last decades, conventional co-cultures using the same culture medium have been performed for very short periods (e.g., two days) to investigate the attractive or repulsive effects of developing oral and dental tissues on sensory nerve fibres. However, extension of the culture period is required to investigate the effects of innervation on tooth morphogenesis and cytodifferentiation. Microfluidics systems allow co-cultures of neurons and different cell types in their appropriate culture media. We have recently demonstrated that trigeminal ganglia (TG) and teeth are able to survive for a long period of time when co-cultured in microfluidic devices, and that they maintain in these conditions the same innervation pattern that they show in vivo. On this basis, we describe how to isolate and co-culture developing trigeminal ganglia and tooth germs in a microfluidic co-culture system.This protocol describes a simple and flexible way to co-culture ganglia/nerves and target tissues and to study the roles of specific molecules on such interactions in a controlled and isolated environment.

Sema3A chemorepellant regulates the timing and patterning of dental nerves during development of incisor tooth germ.

Semaphorin 3A (Sema3A) axon repellant serves multiple developmental functions. Sema3A mRNAs are expressed in epithelial and mesenchymal components of the developing incisor in a dynamic manner. Here, we investigate the functions of Sema3A during development of incisors using Sema3A-deficient mice. We analyze histomorphogenesis and innervation of mandibular incisors using immunohistochemistry as well as computed tomography and thick tissue confocal imaging. Whereas no apparent disturbances in histomorphogenesis or hard tissue formation of Sema3A (-/-) incisors were observed, nerve fibers were prematurely seen in the presumptive dental mesenchyme of the bud stage Sema3A (-/-) tooth germ. Later, nerves were ectopically present in the Sema3A (-/-) dental papilla mesenchyme during the cap and bell stages, whereas in the Sema3A (+/+) mice the first nerve fibers were seen in the pulp after the onset of dental hard tissue formation. However, no apparent topographic differences in innervation pattern or nerve fasciculation were seen inside the pulp between postnatal and adult Sema3A (+/+) or Sema3A (-/-) incisors. In contrast, an abnormally large number of nerves and arborizations were observed in the Sema3A (-/-) developing dental follicle target field and periodontium and, unlike in the wild-type mice, nerve fibers were abundant in the labial periodontium. Of note, the observed defects appeared to be mostly corrected in the adult incisors. The expressions of Ngf and Gdnf neurotrophins and their receptors were not altered in the Sema3A (-/-) postnatal incisor or trigeminal ganglion, respectively. Thus, Sema3A is an essential, locally produced chemorepellant, which by creating mesenchymal exclusion areas, regulates the timing and patterning of the dental nerves during the development of incisor tooth germ.

Developmentally regulated expression of Sema3A chemorepellant in the developing mouse incisor.

OBJECTIVE: Semaphorin 3A (Sema3A) is an essential chemorepellant controlling peripheral axon pathfinding and patterning, but also serves non-neuronal cellular functions. Incisors of rodent are distinctive from molars as they erupt continuously, have only one root and enamel is present only on the labial side. The aim of this study is to address putative regulatory roles of Sema3A chemorepellant in the development of incisor innervation and formation. MATERIALS AND METHODS: This study analyzed expression of Sema3A mRNAs during embryonic and early post-natal stages of mouse mandibular incisor using sectional radioactive in situ hybridization. RESULTS: Although Sema3A mRNAs were observed in condensed dental mesenchyme during the early bud stage, they were absent in dental papilla or pulp at later stages. Sema3A mRNAs were observed in the dental epithelium including the cervical loops and a prominent expression was also seen in alveolar bone. Interestingly, transcripts were absent from the mesenchymal dental follicle target area (future periodontal ligament) throughout the studied stages. CONCLUSION: The expression patterns of Sema3A indicate that it may control the timing and patterning of the incisor innervation. In particular, Sema3A appears to regulate innervation of the periodontal ligament, while nerve penetration into the incisor dental pulp appears not to be dependent on Sema3A. Moreover, Sema3A may regulate the functions of cervical loops and the development of alveolar bone. Future study with Sema3A deficient mice will help to elucidate the putative neuronal and non-neuronal functions of Sema3A in incisor tooth development.

Development of the pioneer sympathetic innervation into the dental pulp of the mouse mandibular first molar.

OBJECTIVE: Our goal was to study the development of pioneer sympathetic innervation of dental pulp of mouse mandibular first molar. DESIGN: We used double fluorescent immunohistochemistry with tyrosine hydroxylase (TH) and anti-medium-chain neurofilament (2H3) antibodies to detect sympathetic and sensory nerve fibres. Serial sections of whole teeth from postnatal days (PN) 0-14, trigeminal and sympathetic superior cervical ganglia of PN 15 mice were examined with confocal microscope. RESULTS: There were two main findings. The unexpected finding was that 2H3 antibody was specific only for sensory nerve fibres and neurons and failed to stain either sympathetic nerve fibres or neurons. The main finding was that although both sympathetic and sensory nerve fibres were already seen near the tooth germ at the newborn stage, the pioneer sympathetic nerve fibres were first observed in the dental pulp only after the onset of root formation on day 9, in contrast to sensory nerve fibres which entered the tooth already on day 4. CONCLUSION: Pioneer sympathetic innervation of dental pulp starts on postnatal day 9 and follows sensory innervation. This indicates differential developmental regulation of the initial sensory and sympathetic innervation of teeth and provides essential background data for further studies on the molecular regulation of pulp innervation.

[Dynamics in the content of neuromediators in amine-containing structures during development of tooth hard tissues]. / Dinamika soderzhaniia neiromediatorov v aminosoderzhashchikh strukturakh v period razvitiia tverdykh tkanei zuba.

Using luminescent-histochemical methods the contents of neurotransmitters (catecholamines, serotonin, histamine) was determined during the development of dental hard tissues in man. In the dental papilla the luminescent blood vessels were initially demonstrated, while single nerve fibers branching around the odontoblasts appeared later. It was found that enamel organ differentiation at the early stages took place without the participation of granular luminescent cells and mast cells. These cells appeared starting at week 22 of fetal development. Their appearance coincided with the process of enameloblast and odontoblast differentiation. The contents of neurotransmitters in the dental structures studied was found to increase with the fetal age.

Development of innervation in primary incisors in the foetal period.

Sections from the frontal part of the mandible of 43 human foetuses from 9 to 39 weeks of prenatal age, which contained two, three and sometimes four lower incisors were immunohistochemically examined using protein gene product and neuron specific enolase (NSE) antibodies in order to establish the time of appearance of nerve fibres in the developing tooth germ and to define their topography. Nerve fibres were first detected in the dental follicle in the 11th week of intrauterine life. Their presence in the dental papilla was confirmed in the 18th week when the first layers of dentine and enamel were deposited. In the 24th week of intrauterine life, the nerve fibres first reached the subodontoblastic region. In the subsequent weeks, an increase in the number of nerve fibres accompanying blood vessels in the central portion of the dental papilla resulted in the formation of neuro-vascular bundles. Moreover, the progressive deposition of enamel and dentine was accompanied by branching of papillary nerves, which thereby formed a fan-pattern. In the foetal period, no evidence was found for the formation of a subodontoblastic plexus. However, we did observe single nerve fibres in close proximity to the odontoblast layer at the end of intrauterine life. Nerve fibres were not detected in either predentine or dentine throughout foetal life.

Mouse rudimentary diastema tooth primordia are devoid of peripheral nerve fibers.

The tooth is a well-defined peripheral target organ for trigeminal nerve fibers. However, only limited information is available regarding pioneer axon guidance to the developing tooth target field. In rodents there is a toothless diastema region between incisors and molars that in the mouse maxilla contains three rudimentary tooth anlagen. Their development stop at the early bud stage when the primary nerve axons grow towards the developing first molar tooth germs. In order to provide background information for studies of regulatory mechanisms of pioneer axon guidance to the developing tooth germs, we investigated the distribution of nerve fibers in the mouse diastema tooth buds, and compared it to the axon growth to the maxillary and mandibular first molar tooth germs by immunohistochemical localization of peripherin and PGP9.5. Analysis of serial sections showed that trigeminal nerve fibers emerging from the trigeminal maxillary and mandibular nerve trunks started to grow towards the developing molar tooth germ at the early bud stage, and subsequently they diverged into buccal and lingual branches next to the condensed dental mesenchyme. During the cap stage, nerve fibers were observed around the tooth germ in the dental follicle region. In contrast, no nerve fibers were located in the vicinity of the diastema tooth primordia at any stage studied, nor did any nerve fibers appear to grow towards this region. Our results show that the development and subsequent disappearance of the diastema tooth primordia takes place without peripheral trigeminal innervation. The diastema tooth primordia may therefore be a useful model system for future studies on molecular regulatory mechanisms of pioneer axon guidance to the tooth germs, and possibly also for evolutionary studies of peripheral axon guidance mechanisms.

Molecular signaling and pulpal nerve development.

The purpose of this review is to discuss molecular factors influencing nerve growth to teeth. The establishment of a sensory pulpal innervation occurs concurrently with tooth development. Epithelial/mesenchymal interactions initiate the tooth primordium and change it into a complex organ. The initial events seem to be controlled by the epithelium, and subsequently, the mesenchyme acquires odontogenic properties. As yet, no single initiating epithelial or mesenchymal factor has been identified. Axons reach the jaws before tooth formation and form terminals near odontogenic sites. In some species, local axons have an initiating function in odontogenesis, but it is not known if this is also the case with mammals. In diphyodont mammals, the primary dentition is replaced by a permanent dentition, which involves a profound remodeling of terminal pulpal axons. The molecular signals underlying this remodeling remain unknown. Due to the senescent deterioration of the dentition, the target area of tooth nerves shrinks with age, and these nerves show marked pathological-like changes. Nerve growth factor and possibly also brain-derived neurotrophic factor seem to be important in the formation of a sensory pulpal innervation. Neurotrophin-3 and -4/5 are probably not involved. In addition, glial cell line-derived neurotrophic factor, but not neurturin, seems to be involved in the control of pulpal axon growth. A variety of other growth factors may also influence developing tooth nerves. Many major extracellular matrix molecules, which can influence growing axons, are present in developing teeth. It is likely that these molecules influence the growing pulpal axons.

Prenatal traces of aberrant neurofacial growth.

The interrelation between the development of the brain/peripheral nerves and that of the surrounding bone tissue is termed neuro-osteology. In orthodontic and pediatric practice the development of the hard tissues is evaluated radiographically, but the development of the neural tissue within the bone tissue is not evaluated. In this review the emphasis is placed on two neuro-osteologic interrelations that can be observed on profile radiographs and orthopantomograms, respectively. One is the connection between the pituitary gland of the central nervous system and the sella turcica (profile radiograph), and the other is the association between the peripheral nerves and the development of the dentition (orthopantomogram). Pituitary gland/sella turcica: The correlation between prenatal malformation in the pituitary gland/sella turcica and the postnatal morphology of the sella turcica in holoprosencephaly, spina bifida/myelomeningocele, and cri-du-chat syndrome is demonstrated. Peripheral nerves/dentition: The prenatal innervation of the dentition is presented. Agenesis and tooth malformation occur in constant patterns within the dental arch fields that share the same innervation. The findings demonstrate that in postnatal diagnosis of the cranium and the teeth, traces of prenatal aberrations can be found that are important for neurofacial growth.

Immunohistochemical localization of nerve fibres during development of embryonic rat molar using peripherin and protein gene product 9.5 antibodies.

Nerve fibres were localized during the initiation and early morphogenesis of the first molar tooth in rat embryos by immunoperoxidase detection of the intermediate-filament protein peripherin and protein gene product 9.5 (PGP 9.5). Nerve fibres from the trigeminal ganglion were detected in the developing first branchial arch of E12-14 embryos. Nerves were not seen in the vicinity of the developing tooth germ before the buid stage (E15), when they were seen around the condensed dental mesenchyme. During transition from the bud to the cap stage (E15), nerve fibres were detected not only in the area of the future dental follicle but also in the mesenchyme next to dental epithelium on the buccal side of the tooth germ. During later cap and bell stages nerve fibres persisted in the dental follicle, but they were not seen in the epithelial dental organ or dental papilla mesenchyme. Absence of trigeminal nerve fibres from the presumptive tooth-bearing area indicates that they are not involved in the initiation of rat tooth development. In addition, the localization of nerve fibres shows that there are some differences in the innervation of rat teeth compared with human and mouse teeth. These results provide data for further studies on the regulation of embryonic rat tooth innervation.

Can the location of tooth agenesis and the location of initial bone loss seen in juvenile periodontitis be explained by neural developmental fields in the jaws?

Recent studies on prenatal innervation of the jaws have shown that three separate main innervation paths, constituting three bilateral neural developmental fields (incisor field, canine/premolar field, molar field) exist in each jaw. In this communication the sequences in which the fields are innervated are indicated. These correspond to the sequences of formation of teeth and jawbone. The normal pattern of tooth agenesis is closely related to the neural fields, as the region within a single field were innervation occurs last is always the area most often affected by tooth agenesis. The initial manifestations of juvenile periodontitis also appear at the sites within the different fields where innervation occurs last. It is suggested that the pubertal growth of the alveolar process does not occur in these regions due to deficient innervation, and that the infection in juvenile periodontitis might be secondary to this regional lack of bone apposition.

Isolation and culture of rat sensory neurons having distinct sensory modalities.

We have recently published papers in which sensory neurons that innervate either the tooth pulp or masseter muscle spindles were labelled in vivo and later identified and studied in primary tissue culture (Taddese et al., 1995; Cook et al., 1997). Here, we provide detailed descriptions of cell labelling and tissue culture methods that we used. The purpose of the preparations is to compare nociceptive and non-nociceptive sensory neurons in vitro. The spindles in mastication muscles are the only muscle afferents whose cell bodies reside in the mesencephalic nucleus (MeN5) of the fifth nerve (Corbin and Harrison, J Neurophysiol, 1940; Cody et al., J Physiol, 1972). Thus, labelling neurons projecting to the masseter muscle and dissecting the MeN5 isolates muscle spindle afferents. Pain is the only conscious sensation elicited by physiological stimulus of tooth pulp (Anderson and Matthews, 1967; Edwall and Olgart, 1977; Ahlquist et al., 1984; Narhi et al., 1994); there may be unconscious sensations that arise from the pulp, but these have never been demonstrated. Thus, tooth pulp afferents represent at least a highly enriched, and possibly a pure, population of nociceptors. In broad outline, the methods of labelling and tissue culture are standard, but we have honed many details in order to obtain practical yields.

Immunohistochemical and electron microscopic demonstration of nerve fibres in relation to gingiva, tooth germs and functional teeth in the lower jaw of the cichlid Tilapia mariae.

Immunohistochemistry revealed the presence of numerous neurofilament (NF)-like immunoreactive axons in relation to gingiva and dental follicles surrounding mineralizing tooth germs. The gingival nerve fibres frequently approached the prospective papilla of early tooth primordia. Electron microscopic (EM) analysis revealed the presence of bundles of unmyelinated axons immediately below the epithelial-proprial junction of the gingiva. Bundles of nerve fibres were also present in the border zone between the prospective papilla of bud-stage tooth germs and surrounding mesenchyme and in close proximity to blood vessels of the follicles surrounding older tooth germs, but no axons were observed within the emerging dental papilla. In the individual functional tooth, a bundle of NF-like immunoreactive nerve fibres entered the apical part of the pulp forming a subodontoblastic plexus at mid-pulpal levels. EM analysis showed that the apical bundle consisted of many unmyelinated and a few myelinated axons invested by Schwann cell processes. The subodontoblastic plexus contained unmyelinated axons only. Thin, axon-like profiles were also seen in predentinal tubules. Nerve fibres were not observed at pulpal horn levels and in the ligamentous attachment. It is concluded that both immature and mature parts of the lower-jaw dentition of the cichlid T. mariae are innervated and that the microscopic anatomy of this innervation is partly similar to the pattern seen in developing and adult mammals.

Evidence for a neural influence on tooth germ generation in a polyphyodont species.

It has been suggested that nerve endings emanating from the dental nerve plexus of the jaw might be involved in the formation of tooth germs. In the present study we examine the effect of unilateral denervation on the formation of tooth germs in the lower jaw of a polyphyodont teleost--the cichild Tilapia mariae. Repeated inspection of the lower jaw dentition in normal animals over a period of about 300 days showed that the functional time of an average individual tooth is 101 days. In operated animals, the functional time was normal on the unoperated side, but on the denervated side tooth turnover ceased about 100 days after surgery. Radiographic plates from lower jaw specimens revealed that mineralized replacement teeth were present on the unoperated side, but not on the denervated side, 300 days after denervation. Light microscopic examination of semi-thin transverse sections from decalcified plastic-embedded lower jaws showed that soft-tissue tooth primordia and nerves were lacking on the denervated side, while present within the undisturbed half-jaw. It is concluded that the local presence of mandibular nerve branches is necessary for the formation of tooth germs in the lower jaw of the cichlid T. mariae.

Nerve fibres and cells immunoreactive to neurochemical markers in developing rat molars and supporting tissues.

The distribution of nerve fibres immunoreactive to calcitonin gene-related peptide (CGRP), substance P (SP) and neuropeptide Y (NPY) was compared to the general neurochemical markers for nerves and neuroendocrine cells protein gene product 9.5 (PGP 9.5) and neurone-specific enolase (NSE), by use of the avidin-biotin peroxidase complex method in developing dental structures in rats aged 13 to 27 days. A substantially greater part of the nerve fibres was immunoreactive to CGRP and SP than to NPY. In the bell stage, nerve fibres immunoreactive to PGP 9.5, CGRP and SP were found in the dental follicle but not in the dental papilla and stellate reticulum. In the advanced bell stage, after initiation of dentine and enamel formation, PGP 9.5, CGRP- and SP-immunoreactive fibres were found in the dental papilla, while the first NPY-immunoreactive fibres were observed in the papilla when root formation started. Concomitant with the beginning of root development, a subodontoblastic nerve plexus was gradually formed and PGP 9.5-, CGRP- and SP-immunoreactive fibres were found within the dentinal tubules. From the start of root formation, CGRP-, SP- and NPY-immunoreactive nerves were shown in the developing periodontal ligament, although a mature distribution pattern was not observed until root formation was nearly completed. Ameloblasts, odontoblasts and cell-like structures in the outer enamel epithelium and within the dental lamina were PGP 9.5-immunoreactive at the bell stage. As the tooth matured, the immunolabelling gradually decreased, but was still present in some odontoblasts after tooth eruption. NSE-immunoreactive, cell-like structures were found in the periphery of the dental follicle, and persisted close to alveolar bone in the periodontal ligament when the tooth reached occlusion. Hence, it may be concluded that sensory nerves containing SP and CGRP are present in the pulp in advance of sympathetic nerves immunoreactive to NPY.

Immunohistochemical evidence for the presence of nerve fibres with substance P- or calcitonin gene-related peptide-like immunoreactivity in the proliferating epithelium in the developing teeth of rats.

By immunohistochemistry using the avidin-biotin peroxidase complex method, nerve fibres with substance P- or calcitonin gene-related peptide (CGRP)-like immunoreactivity (IR) were examined in the dental lamina, cells external to the reduced ameloblasts and oral epithelium in the developing upper first molars of postnatal rats. At birth, very few varicosities with substance P- or CGRP-IR were found in the dental lamina over the mesial cusp of the dental germ. In 5-day-old rats, nerve fibres with substance P- or CGRP-IR in the dental lamina over the mesial cusp gradually increased in number. In 7-day-old rats, over the mesial portion of the dental germ, the oral half of the dental lamina began to thicken on the buccopalatal side, in which many nerve fibres with substance P- or CGRP-IR were observed. A few nerve fibres began to penetrate the cells external to the reduced ameloblasts over the middle buccal cusp. In 10-day-old rats, the oral epithelium over the mesial cusp gradually thickened, and nerve fibres with substance P- or CGRP-IR were found especially in its basal layer. In 13-15-day-old rats, a great many nerve fibres with substance P- or CGRP-IR were distributed all over the fused, thickened dental lamina and cells external to the reduced ameloblasts proliferated over the middle and distal cusps. Nerve fibres with substance P- or CGRP-IR formed a plexus in the thickened oral epithelium, which spread to the mesiopalatal end of the mesial cusp.(ABSTRACT TRUNCATED AT 250 WORDS)

Immunohistochemical analysis of pulpal innervation in developing rat molars.

The beginning of pulpal innervation was examined in these developing teeth. Mandibular first molars removed from newborn to 7-day-old rats were cryosectioned and nerve fibres were localized by the peroxidase-antiperoxidase technique using a neuropeptide-specific (200 kDa) antibody. Some pulps from 5- and 7-day-old rats were also examined by conventional electron microscopy. In newborn to 4-day-old rats, molars in the initial stages of the dentine deposition were innervated in the follicles but not in the pulps. In molars from 5-day-old rats, nerve fibres were found in the pulps of 11 out of 16 samples. The fibres were mainly located along the blood vessels in the basal part of the pulp, with some arborizations. In rats of 6 and 7 days old, nerve fibres were found in all the pulps examined. These had gradually extended deep into the cuspal area and were increasingly arborized with increasing age. Nerve fibres were also found along the basal laminae of blood vessels in some dental pulps from 5-day-old rats when examined by electron microscopy. At 7 days, nerve fibres were composed of bundles of axons, some of which were covered with Schwann cell processes and basal laminae. These observations indicate that the innervation of rat molar dental pulps begins after the start of the deposition of enamel, in animals of around 5 days of age, which is at the same stage as in mouse molars, as others have shown by a silver-staining method.

Innervation of mouse molars during the early states of tooth germ development.

The topography of nerves and first molar tooth germs in 11-14-day embryos was studied using silver-impregnated serial sections. Nerve fibers growing toward the developing tooth germ became visible on the 12th day, while the first sign of molar tooth differentiation was found as a thickening of the oral epithelium in 11-day embryos. From the 12th to 13th day the nerve fibers spread, forming a plexus close to the base of the tooth bud, and on the 14th day some entered into the dental follicle of the tooth germ at the early cap stage. However, no nerve fiber was found growing into the dental papilla during the observation period. The observations showed that the earliest nerve fibers running toward the tooth forming area appeared after the histogenesis of the tooth germ started, and the timing and pattern of the innervation of the tooth germs revealed that tooth germs are a useful model for investigating the mechanism of nerve growth into developing peripheral organs.

Nerve growth to tooth buds after homotopic or heterotopic autotransplantation.

Feline permanent incisor tooth buds (bell stage) were autotransplanted to mandibular alveolar sockets (homotopic site) or to the submandibular subcutis or the leg (heterotopic sites). This was done in 34 kittens aged 1-2 months. After survival times of 3-8 months the animals were fixed by glutaraldehyde perfusion. A total of 56 mineralized teeth, which had developed at the recipient sites, were removed, demineralized and processed for light microscopic (LM) general evaluation. Fourty-four teeth, which were judged to be grossly normal in the LM, were selected for electron microscopic (EM) analysis with respect to the occurrence of pulpal nerve fibres. The highest proportion of normal teeth (16 of 16) was obtained from the alveolar site, followed by the submandibular (11 of 14) and hindlimb (17 of 26) sites. Most of the grossly normal grafts possessed pulpal axons (37 of 44). The alveolar grafts were all innervated and exhibited a largely normal appearance qualitatively and in terms of percentage of myelinated fibres. The proportion of innervated pulps was lower among the heterotopic mandibular (10 of 11) and hindlimb (11 of 17) grafts. In addition, signs of nerve fibre degeneration appeared more frequently at the heterotopic sites. On the basis of these findings, and in view of the results of other workers, we conclude that tooth germs are attractive targets for all divisions of the trigeminal nerve and for cutaneous nerves outside the trigeminal system. However, the morphological picture tends to become increasingly abnormal with increasing distance from the normal locus.

A histological study of the innervation of developing mouse teeth.

The innervation of developing mouse teeth between initial formation and crown formation was investigated using silver-stained serial sections. The developing innervation correlated with the stage of development of individual teeth rather than the chronological age of the mice. Nerves approached the developing dental papilla during the bud stage and formed a basal plexus below the dental papilla in the early cap stage. Nerve fibres from this plexus spread into the dental follicle as it began to develop. However, nerves did not enter the dental papilla until crown formation commenced, when the innervation was fairly rapid. Innervation commenced in the incisor teeth as soon as dentinogenesis started but not until a thin layer of enamel had been formed in the molar teeth. Although some of the early fibres were associated with blood vessels, many nerves lay free in the pulp. The absence of nerves in intimate relationship to the presumptive dental regions during the inductive phase of tooth development suggests that neural induction plays no part in the initiation of odontogenesis. However, it is not possible, from a purely histological study such as this, to attribute any function to the nerves at other stages of tooth development until the neurotransmitter content, and hence the type and likely function of the nerves, is established.