Innervation of the guinea pig trachea: a quantitative morphological study of intrinsic neurons and extrinsic nerves

The innervation of the guinea pig trachea was studied in wholemount preparations stained for acetylcholinesterase, catecholamines, and substance P immunoreactivity and by electron microscopy. The majority of parasympathetic and afferent nerve fibres arrive from the vagus via branches of the recurrent laryngeal nerves. The recurrent laryngeal nerves are composed of several fascicles comprising 600-700 small myelinated fibres (2-5 microns diameter) and about 1,000-2,000 unmyelinated fibres; both components exit from the nerve and project in fine branches to the trachea. A separate component of 200-250 large myelinated fibres (more than 5 microns diameter) runs the full length of the nerve and innervates the striated muscles of the larynx. The recurrent laryngeal nerves are slightly asymmetric in their origin, length, number, and composition of fibres, with the right nerve being shorter but with more numerous and thinner myelinated fibres. At the distal end of the recurrent nerve, a fine branch called the ramus anastomoticus connects it to the superior laryngeal nerve. In the tracheal plexus, there are on average 222 ganglion cells (range 166-327), distributed mostly in small ganglia of 12 or fewer neurons. The ganglionated plexus is situated entirely outside the tracheal wall, overlying the smooth muscle. Ligation experiments show that sympathetic nerve fibres reach the trachea with the recurrent nerves via anastomoses between the sympathetic chain and vagus nerves, or occasionally with recurrent nerves directly, the largest being at the level of the ansa subclavia. There are also perivascular sympathetic nerve plexuses. Substance P immunoreactive fibres enter the trachea from the vagus nerves and by pathways similar to those of sympathetic nerves. There are also paraganglion cells within the recurrent laryngeal nerve that contain catecholamines and are surrounded by substance P immunoreactive fibres. After cervical vagotomy, all the large myelinated fibres of the ipsilateral recurrent laryngeal nerve degenerate and so do all but 10 or 20 small myelinated fibres and all but a few unmyelinated fibres. Degenerating fibres are found within the entire tracheal plexus, indicating bilateral innervation. The small myelinated fibres that survive cervical vagotomy probably represent sympathetic or afferent nerves with their cell bodies located in sympathetic or dorsal root ganglia.     

Evidence for a role of endogenous nerve growth factor in the functional sex difference in the adrenergic innervation of the male and female mouse submandibular gland

In view of the remarkable sex difference in nerve growth factor (NGF) content in mouse submandibular gland (SMG), a direct comparison of the adrenergic innervation as represented by noradrenaline (NA) content in SMG, rate of losing NA after superior cervical ganglionectomy, post-reserpine recovery of NA and survival and growth of implanted neurons of superior cervical ganglion into SMG was carried out. The results showed that the NA contents in the intact male SMG and in the male SMG during the recovery from reserpinization were higher than those in the female. There were marked sex differences in patterns of the time-course of post-ganglionectomy loss of NA in SMG and in the ability of SMG to support survival and growth of the implanted SCG neurons. These results indicate that NGF in the mouse SMG may play a functional role in the adrenergic neurons innervating the gland.     

Innervation of laryngeal nerve paraganglia: An anterograde tracing and immunohistochemical study in the rat

Carotid body-like organs, paraganglia, frequently occur in the superior and recurrent laryngeal nerves. The paraganglia are supplied with a rich innervation of unknown origin. In the present study, the origin of the innervation of the paraganglia of the rat was studied with two different techniques. One approach was anterograde tracing of wheat-germ agglutinin-horseradish peroxidase after injection into the nodose and jugular ganglia of the vagus and the superior cervical ganglion. The other approach was immunohistochemical staining for neuropeptides after excision of the superior cervical ganglion, or vagotomy. Antisera against neuropeptide Y, vasoactive intestinal polypeptide, and calcitonin gene-related peptide were utilized. Both the tracing method and calcitonin gene-related peptide immunohistochemistry after vagotomy showed that the paraganglia receive sensory innervation from the vagal ganglia. No labeling was detected in the paraganglia after injection of wheat-germ agglutinin-horseradish peroxidase in the superior cervical ganglion. Excision of this ganglion did not lead to a decrease in the neuropeptide-Y innervation in the paraganglia, but most of this innervation in the surrounding blood vessels disappeared. The observations show that the superior cervical ganglion does not contribute to the innervation in the paraganglia and that the neuropeptide-Y innervation of the blood vessels originates from the superior cervical ganglion whereas that of the paraganglia has another origin, most likely local ganglionic cells. The results also suggest that the vasoactive intestinal polypeptide innervation in the paraganglia arises from local ganglionic cells. The two approaches complemented each other in mapping the afferent and efferent nerve supply of the paraganglia. © 1994 Wiley-Liss, Inc.     

Observations on the fine structure and distribution of presumptive baroreceptor nerves at the carotid sinus

Comparative nerve fiber counts show that the areas of the carotid bifurcation receiving a rich innervation correspond in extent to areas of the vascular wall with a thinned and elastic tunica media. Presumptive baroreceptor nerves within the carotid sinus wall were examined by electron microscopy in normal rabbits and also after superior cervical ganglionectomy and/or Hering's nerve transection. The glossopharyngeal (baroreceptor) nerve terminals are either denuded of a cellular covering or are thinly clothed in Schwann cell cytoplasm, and are intimately related over wide areas to at least three sinus wall components: to elastin at the medio-adventitial border, to collagen in the inner one-half of the adventitia, and to the surfaces of isolated adventitial smooth muscle cells. These muscle cells also receive a terminal innervation from the superior cervical ganglion. There is no nervous penetration of the sinus tunica media. The spatial arrangement of the adventitial muscle cells and their associated sensory and motor nerves simulates a smooth muscle spindle, possibly constituting a morphological basis for an efferent control of baroreceptor output. The baroreceptor nerve terminals contain large accumulations of mitochondria, and also myelin figures and small dense bodies resembling glycogen granules. Evidence is presented that baroreceptor impulses may be conveyed from the sinus wall in both myelinated and nonmyelinated nerve fibers.     

Bovine pancreatic polypeptide-like immunoreactive nerves in the rat major cerebral arteries

Immunofluorescence histochemical observation of the major cerebral arteries of the rat demonstrated the presence of an extensive plexus of bovine pancreatic polypeptide (BPP) containing nerves. The density of the nerves were similar to those previously observed in the adrenergic plexus. The density of BPP-immunoreactive fibers markedly decreased following bilateral superior cervical ganglionectomy. It is concluded that BPP-containing fibers of the cerebral arteries emanate mostly from the superior cervical ganglion and that BPP and norepinephrine coexist in the adrenergic nerves.     

The allocation of nerve fibres to the anterior eye segment and peripheral ganglia of rats. II. The sympathetic innervation

The sympathetic innervation of the peripheral ganglia related to the eye, i.e. the trigeminal ganglion, the ciliary ganglion and the pterygopalatine ganglion, and of the anterior eye segment was studied in rats. Selective labelling of sympathetic nerves was obtained by means of injection of [3H]leucine into the superior cervical ganglion. Bundles of sympathetic nerve fibres were found in the trigeminal ganglion and the pterygopalatine ganglion but were absent in the ciliary ganglion. In addition individual sympathetic nerve fibres, which may have contacts with trigeminal ganglion cells, were found between the ganglion cell bodies all over the trigeminal ganglion indicating a sympathetic innervation of this ganglion. In the anterior eye segment, there appeared to be a sympathetic innervation of the ciliary cleft, the ciliary body and the iris. Within the ciliary body sympathetic nerve fibres innervate the central stroma and the stroma of the ciliary processes. Labelled sympathetic nerve fibres were also observed in the stroma of the iris and were most abundant in its periphery. Most sympathetic fibres reach the iris and ciliary body by way of the base of the ciliary body. Only few sympathetic fibres are present in the ciliary cleft. No sympathetic innervation of the cornea was found.     

Histochemical and immunohistochemical study of noradrenergic, serotonergic and peptidergic innervation of the cerebral circulation

Using fluorescence histochemistry, we have examined the origin of noradrenaline (NA)-, 5-hydroxytryptamine (5-HT)- and neuropeptide Y (NPY)-containing perivascular nerve fibres supplying the major cerebral arteries of the gerbil, and the effect of superior cervical ganglionectomy on their cerebrovascular density and distribution. Bilateral ganglionectomy resulted in a complete loss of these perivascular nerve types in the anterior circulation, while a few fibres still persisted in the posterior circulatory bed. In addition, we demonstrate the presence of both 5-HT- and NPY-positive cell bodies in the superior cervical ganglion (SCG) of adult gerbils, suggesting that the main origin of cerebrovascular 5-HT- and NPY- (as well as NA-) containing nerve fibres is in the SCG.     

Reorganization of cranial sympathetic pathways following neonatal ganglionectomy in the rat

Postganglionic sympathetic innervation normally is distributed ipsilaterally to lateral cranial targets. However, contralateral outgrowth occurs following unilateral ganglionectomy in neonatal rats. This study was conducted to determine the prevalence, morphological features, ganglionic derivations, and temporal sequence of sympathetic reinnervation of denervated cranial targets. Unilateral superior cervical ganglionectomy of mature rats revealed exclusively ipsilateral distributions of catecholaminergic histofluorescent fibers to orbital targets (Meibomian gland, tarsal muscle, orbital muscle, iris, ciliary body, vasculature) and the circle of Willis, with the exception of the anterior cerebral artery. In mature rats following neonatal unilateral ganglionectomy, all targets were reinnervated by fibers displaying morphologies and target relationships similar to normal innervation, but with lower densities. Acute excision of the remaining superior cervical ganglion eliminated all fibers in 7 of 8 preparations. In adult rats receiving neonatal bilateral superior cervical ganglionectomies, cerebral vasculature was reinnervated consistently, and orbital targets contained fluorescent fibers in 6 of 16 cases, indicating that reinnervation can derive from other sources when superior ganglion outgrowth is prevented. Observations in developing rats revealed fibers along the cranial portion of the contralateral optic nerve sheath at 2-3 days postganglionectomy, and within the orbit at later ages, reaching the most distal targets by 14 days. It is concluded that widespread sympathetic reinnervation of orbital and cerebrovascular targets derives primarily from the contralateral superior ganglion. Orbital ingrowth apparently originates intracranially and enters the orbit by an atypical pathway within the optic foramen.     

Organization of the sympathetic postganglionic innervation of the rat heart

The origins and organization of cardiac sympathetic postganglionic nerves in the rat were identified in the present investigation. The retrograde tracer, Diamidino Yellow, was injected into the right or left ventricles to label somata in the sympathetic chain. Analysis of all sympathetic ganglia from superior cervical ganglion through the 10th thoracic ganglion indicated that the postganglionic innervation of the rat cardiac ventricles originates bilaterally. The majority of these somata were located in the middle and inferior cervical ganglia (middle cervical-stellate ganglion complex) (approximately 92% of all labelled cells), with lesser contributions from the superior cervical and 4th through 6th thoracic ganglia. To confirm and further quantitate these findings, the middle cervical-stellate ganglion complex was removed (MC-S ganglionectomy) bilaterally or ipsilaterally from the left or right sides, and regional cardiac norepinephrine concentration (left and right atrial appendages and left and right ventricles) was analysed 7 or 28 days later. At both times after bilateral MC-S ganglionectomy, regional cardiac norepinephrine was reduced by 89% to 100%, indicating the removal of almost all cardiac noradrenergic cells of origin and possibly fibers of passage. The results of unilateral MC-S ganglionectomy experiments indicated that the atrial appendages and the left ventricle receive bilateral innervation from the middle cervical-stellate ganglion complex. However, the left middle cervical-stellate ganglion complex appears to contribute a majority of the norepinephrine to the right ventricle. Furthermore, between 7 and 28 days after contralateral MC-S ganglionectomy, atrial appendages, but not ventricles, display significant recovery of norepinephrine content. The present data demonstrate: (1) a bilateral locus of origin of cardiac sympathetic postganglionic neurons, limited longitudinally to cervical through mid-thoracic ganglia, and (2) the ability of the cardiac postganglionic innervation to regenerate after partial denervation. These results demonstrate anatomical evidence for significant bilateral integration of cardiac sympathetic activity at the level of the sympathetic ganglion in the rat.     

Correlation with superior cervical sympathetic ganglion and sympathetic nerve innervation of intracranial artery-electron microscopical studies

When the superior cervical ganglion was resected in dogs, nerve degeneration in arterial walls began after about 28 h and marked degenerative substance was shown after 40–48 h; after 4 days the small cored vesicles of adrenergic axons disappeared. The same condition was seen after 3 months, but after 6 months the small cored vesicles were again visible. When the middle cerebral artery was examined by separating it into the perforating artery near to the internal carotid artery and the peripheral portion of the middle cerebral artery, degeneration of the nerve fibers of the arterial walls occurred earlier in the more proximal portion.The distribution of adrenergic nerve fibers from the superior cervical ganglion is bilateral in the anterior cerebral artery from the anterior communicating artery to the peripheral region, basilar artery, and vertebral artery, but ipsilateral only in the anterior cerebral artery as far as the anterior communicating artery, middle cerebral artery, posterior communicating artery, posterior cerebral artery and superior cerebellar artery.Degeneration of nerve fibers of the walls of these cerebral arteries was not seen ever after stellate ganglionectomy in both sides.     

Afferent projections of the superior and recurrent laryngeal nerves.

The central afferent projections of the superior and recurrent laryngeal nerves were investigated in the rat, utilizing the transganglionic transport of WGA-HRP. Labelled superior laryngeal nerve terminal fields were found bilaterally in the interstitial and medial subnuclei of the nucleus tractus solitarii with the ipsilateral being more dense. The distribution of the recurrent laryngeal nerve terminals were similar to that of the SLN with two major differences: the projections were ipsilateral, and there was a marked decrease in the terminal field density. The terminal field density differences were confirmed by quantitatively identifying the labelled ganglion cells of the vagus nerve. These findings accurately delineate the first integrative components in the mediation of the complex laryngeal reflexes.     

Substance P in the superior cervical ganglion and the submaxillary gland of the rat

Substance P has been detected in the superior cervical ganglion and submaxillary gland of the rat by radioimmunoassay. Decentralization, section of the carotid sinus nerve, or deefferentation does not affect the level of substance P in the rat superior cervical ganglion. Superior cervical ganglionectomy does not affect the amount of substance P in the submaxillary gland; however, section of the chorda tympani or duct ligation significantly reduces substance P in the submaxillary gland. It is concluded that substance P in the superior cervical ganglion is not located in the preganglionic sympathetic neurons, afferents from the glossopharyngeal nerve or post-ganglionic neurons projecting to the submaxillary gland or carotid body. Substance P does not appear to be associated with sympathetic innervation of the submaxillary gland, but it is anatomically and functionally related to the chorda tympani and possibly involved in the parasympathetic innervation of the gland. Substance P may also be present in gland cells of the submaxillary gland.     

Cerebral perivascular serotonergic fibres have a peripheral origin in the gerbil

The origin of serotonergic nerves supplying the large cerebral vessels of the gerbil has been investigated after bilateral superior cervical ganglionectomy. Immunohistochemical techniques were applied to whole-mount stretch preparations of the cerebral vessels. Removal of both ganglia resulted in the complete loss of immunoreactive fibres in the vessels supplied by the internal carotid artery and in a marked reduction of fibres innervating vessels of the vertebro-basilar system, indicating that most of the cerebrovascular serotonergic nerves have a peripheral sympathetic origin in the gerbil. The contrast with the central origin of serotonergic perivascular nerves claimed in the rat is discussed.     

Sensory nerves impair sympathetic reinnervation and recovery of smooth muscle function.

Neuronal populations projecting to a common target may compete for neurotrophic substances. To determine if competition impairs target reinnervation, we examined the effect of capsaicin-induced sensory denervation on sympathetic nerve ingrowth to the sympathectomized rat superior tarsal smooth muscle. In tarsal muscles with intact sympathetic innervation, capsaicin injection on Day 2 reduced numbers of perimuscular CGRP-ir sensory nerves by 68% at 3-4 months; however, it did not alter dopamine-beta-hydroxylase-ir nerve density, response to nerve stimulation, or tarsal muscle adrenoceptor-mediated contraction. Tarsal muscles denervated by ipsilateral superior cervical ganglionectomy on Postnatal Day 4 were partially reinnervated by fibers from the contralateral ganglion, as noted in previous studies. Sensory denervation by capsaicin improved sympathetic reinnervation, as evidenced by a 174% increase in numbers of DBH-ir nerves and a 62% increase in neurally mediated smooth muscle contraction evoked by electrical stimulation of the contralateral pathway relative to reinnervated muscles of vehicle-injected rats; smooth muscle function was also influenced, as indicated by a decrease toward normal in adrenoceptor sensitivity. Tarsal muscles denervated at 30 days were not reinnervated in either vehicle-injected or capsaicin-treated rats, indicating that sensory denervation does not extend the developmental window during which contralateral reinnervation can occur. Both the vehicle-injected and capsaicin-treated preparations with sustained juvenile sympathectomy showed sensory hyperinnervation as adults; thus, a chronic reduction in competition from sympathetics is a sufficiently powerful stimulus to overcome the decreased nerve density induced by neonatal capsaicin treatment. We conclude that sensory nerves limit the extent of sympathetic reinnervation and functional recovery that can occur following neonatal sympathetic denervation.     

The allocation of nerve fibres to the anterior eye segment and peripheral ganglia of rats. I. The sensory innervation

The distribution of sensory trigeminal nerve fibres in the anterior eye segment and the autonomic eye related ganglia, i.e. the parasympathetic ciliary and pterygopalatine ganglia and the sympathetic superior cervical ganglion, was studied in rats. For this the trigeminal ganglion was injected with tritiated leucine and wheat germ agglutinin coupled to horseradish peroxidase (WGA-HRP). After injection of WGA-HRP into the trigeminal ganglion, ganglion cell somata in the superior cervical and the pterygopalatine ganglion were labelled. As labelling of these cell bodies with WGA-HRP is the result of retrograde transport it must be assumed that cell bodies in these ganglia project to the trigeminal ganglion. [3H]Leucine injection into the trigeminal ganglion revealed the presence of labelled nerve fibres in the pterygopalatine ganglion and the nodose ganglion i.e. the sensory ganglion of the vagus nerve. Labelled nerve fibres were absent in the ciliary and superior cervical ganglion. As [3H]leucine labelling of nerve fibres is the result of anterograde transport exclusively, it can be concluded that trigeminal nerve fibres project to the nodose ganglion and the pterygopalatine ganglion, but not to the ciliary and superior cervical ganglion. In the retrobulbar structures, sensory nerve fibres occurred between the inferior oblique and the lateral rectus muscle and were present medial to the medial rectus muscle. Within the anterior eye segment, sensory nerve fibres were found in the cornea epithelium, stroma and adjacent to the endothelium. In addition, labelled fibres were found in the anterior stroma of the ciliary body, throughout the iris up to the pupillary border and in the conjunctiva. Most sensory nerve fibres which innervate the cornea, the iris and the ciliary body traverse the ciliary cleft.     

Disappearance of noradrenaline from different parts of the rabbit external ear following superior cervical ganglionectomy

Following superior cervical ganglionectomy the noradrenaline tissue levels in the base of the rabbit external ear started to decrease after about 12 hours. About 4 hours later noradrenaline started to decrease about simultaneously in the middle piece and the tip. Within 24 hours after ganglionectomy the noradrenaline decreased rapidly in all three pieces. Thereafter the noradrenaline of the base piece was levelling off at about 15% of the normal whereas the noradrenaline in the two other pieces remained at about 40% of the normal. The results indicate that the adrenergic innervation of the ear is complex, but that the base piece is supplied with adrenergic axons that are shorter than those of the middle and tip pieces. Based on 1: the delay of 4 hours observed in the pattern of the noradrenaline disappearance between the base and the two other pieces, and II: the difference in distance between the three pieces of the ear and the superior cervical ganglion, a transport rate of a "protecting factor" could be discussed to be about 1-2 cm/h. The rate of axonal transport of dopamine-beta-hydroxylase in the rabbit sciatic nerve, was found to be within the same range, being about 2 cm/h.     

Synapse formation by autonomic nerves in the previously denervated neuromuscular junctions of the feline intrinsic laryngeal muscles

Nerve terminals of unknown origin at the previously denervated neuromuscular junctions (NMJ) of the feline intrinsic laryngeal muscles were investigated. Until 3 weeks after the transection of the recurrent laryngeal nerve (RLN), no axons were observed in the Büngner's bands and the NMJ, accompanied by a marked decrease of autonomic nerves around the blood vessels. At 5-6 weeks nerve varicosities labeled by 5-hydroxydopamine (5-OHDA) were observed in the Büngner's bands together with an increase of autonomic nerves around the blood vessels. At 9-30 weeks (8 cases) nerve terminals were found at the NMJ in all cases. Even if the ipsilateral vagosympathetic trunk was transected at 17 weeks, nerve terminals were found at all the NMJ 3 weeks after this treatment, indicating that nerve terminals were not from the original RLN. These nerve terminals were considered to be autonomic because nerve terminals labeled by 5-OHDA were observed. Furthermore, in the case of the removal of the superior cervical ganglion (SCG) and the nodose ganglion at 21 weeks, though nerve varicosities were found in the Büngner's bands, nerve terminals were not found 9 days after this treatment, suggesting that the ipsilateral SCG possibly played an important role. In electromyographic findings fibrillation-like activities were recognized in 8/11 cases after 5 weeks. The relationship of this phenomenon to fibrillation and muscle atrophy was discussed.     

Trans-synaptic regulation of the development of end organ innervation by sympathetic neurons.

To examine the regulation of development of end organ innervation the superior cervical ganglion (SCG), and two of its target organs, the iris and pineal gland, were studied using biochemical and histofluorescent approaches. During postnatal ontogeny the activity of tyrosine hydroxylase (T-OH), which is localized to adrenergic neurons, increased 50-fold in iris, and 34-fold in pineal nerve terminals of the rat. These increases paralleled the in vitro rise in iris [3H]norepinephrine ([3H]NE) uptake, a measure of the presence of functional nerve terminal membrane. These biochemical indices of end organ innervation correlated well with developmental increases in density of innervation, adrenergic ground plexus ramification and nerve fiber fluorescence intensity as determined by fluorescence microscopy. Unilateral transection of the presynaptic cholinergic nerves innervating the SCG in 2-3-day-old rats prevented the normal development of end organ innervation: T-OH activity, [3H]NE uptake, innervation density, plexus ramification and fluorescence intensity failed to develop normally in irides innervated by decentralized ganglia. It is concluded that trans-synaptic factors regulate the maturation of adrenergic nerve terminals, and the development of end organ innervation by SCG.     

Galanin immunoreactivity in autonomic innervation of the cat eye

In an immunohistochemical study, we find that galanin is much more widely distributed in the peripheral innervation of the cat eye than in other animals so far examined. Previous studies of rat and pig eyes have revealed sparse galanin-positive nerves that presumbably originate in the trigeminal ganglion. In contrast, the cat has a rich supply of galanin-containing nerve fibers throughout the uvea. Galanin-positive varicose nerves concentrate densely in iris muscles and distribute more sparsely in the ciliary muscle. The ciliary processes have a plexus of galanin-positive nerve underlying the ciliary epithelium at their base and positive nerve fibers coursing within their stroma. The ciliary artery and its branch vessels in the uvea are invested with a dense plexus of galanin-positive nerves. All autonomic ganglia supplying the eye contain cells that express galanin. It is present in 97% of superior cervical ganglion cells, coexisting with both tyrosine hydroxylase and neuropeptide Y; in 80% of pterygopalatine ganglion cells, most of whiclh also contain vasoactive intestinal peptide; and in approximately 25% of ciliary ganglion celis. After unilateral superior cervical ganglionectoiny, galanin-positive nerves almost totally disappear from the iris muscles, demonstrating that they are predominantly of sympathetic origin. Galanin-positive nerves investing the ciliary artery and choroidal blood vessels are not detectably reduced by sympathectomy, indicating that perivascular parasympathetic nerves from the pterygopalatine ganglion also express galanin. Other galanin-containing nerves in the eye can originate from the trigeminal and ciliary ganglia. The prominence of galanin in the ocular autonomic innervation of the cat provides an opportunity to explore the physiological effects of this neuropeptide in the eye. © 1994 Wiley-Liss, Inc.     

Neuronal localization of pancreatic polypeptide (PP) and vasoactive intestinal peptide (VIP) immunoreactivity in the earthworm (Lumbricus terrestris)

Nerve fibres and cell bodies displaying vasoactive intestinal polypeptide (VIP) or pancreatic polypeptide (PP) immunoreactivity were demonstrated in ganglia of the earthworm (Lumbricus terrestris). VIP cell bodies were found in the most anterior ganglion of the ventral nerve cord, the subpharyngeal ganglion. Immunoreactive nerves were seen running in the center of the cord until about the 10th segment. PP cell bodies were found in the cerebral ganglion where VIP was lacking, in the subpharyngeal ganglion and in more posteriorly located ganglia of the ventral nerve cord. PP nerve fibres could be followed below the 10th segment of the cord.