We have re-evaluated, using the anterograde tracer biocytin, supraspinal efferent projections from the parabrachial complex (PBN) to gain new information about the nature of its connections and nerve terminal patterns. We selectively injected biocytin into the 3 main regions of the nucleus (lateral PBN, medial PBN and Kölliker-Fuse nucleus). We observed distinct groups of ascending and descending fibres of different calibre from the PBN running throughout the brain and reaching many brain areas involved in the regulation of autonomic function. Here we detected labelled bouton-like terminals and fibres with en-passage varicosities. The ascending efferents from the lateral PBN mainly reached the reticular, raphe and thalamic nuclei, the zona incerta (ZI), central nucleus of the amygdala (CeA) and lateral area of the periaqueductal grey (PAG). Thin descending efferents reached the ventral region of the solitary tract nucleus (STN). The ascending efferents from the medial PBN were seen in the raphe nuclei, reticular nuclei, ventral and lateral areas of the PAG, thalamic nuclei, and in the medial and lateral nuclei of the amygdala. Descending efferents were seen in the STN and in some reticular nuclei. The ascending projections from the Kölliker-Fuse targeted the ventral area of PAG, CeA, ZI, lateral hypothalamic area, ventromedial thalamic nucleus and, with only a few terminals, the ipsi and contralateral reticular area. A large number of descending efferents reached STN, caudal and paragigantocellular reticular nuclei. The higher sensitivity of biocytin compared with other types of markers allowed us to determine more effectively the distribution, nature and extent of the supraspinal PBN connections. This suggested that in several nerve circuits the PBN probably plays a more important role than previously thought.

Peripheral nerve transection induces significant changes in neuropeptide expression and content in injured primary sensory neurons, possibly due to loss of target derived neurotrophic support. This study shows that neurotrophin-3 (NT-3) delivery to the injured nerve influences neuropeptide Y (NPY) expression within dorsal root ganglia (DRG) neurons. NT-3 was delivered by grafting impregnated fibronectin (500 ng/ml; NT group) in the axotomised sciatic nerve. Animals grafted with plain fibronectin mats (FN) or nerve grafts (NG) were used as controls. L4 and L5 DRG from operated and contralateral sides were harvested between 5 and 240 d. Using immunohistochemistry and computerised image analysis the percentage, diameter and optical density of neurons expressing calcitonin gene-related peptide (CGRP), substance P (SP), vasoactive intestinal peptide (VIP) and NPY were quantified. Sciatic nerve axotomy resulted in significant reduction in expression of CGRP and SP, and significant upregulation of VIP and NPY (P<0.05 for ipsilateral vs contralateral DRG). By d 30, exogenous NT-3 and nerve graft attenuated the upregulation of NPY (P<0.05 for NT and NG vs FN). However, NT-3 administration did not influence the expression of CGRP, SP or VIP. The mean cell diameter of NPY immunoreactive neurons was significantly smaller in the NT-3 group (P<0.05 for NT vs FN and NG) suggesting a differential influence of NT-3 on larger neurons. The optical densities of NPY immunoreactive neurons of equal size were the same in each group at any time point, indicating that the neurons responding to NT-3 downregulate NPY expression to levels not detectable by immunohistochemistry. These results demonstrate that targeted administration of NT-3 regulates the phenotype of a NPY-immunoreactive neuronal subpopulation in the dorsal root ganglia, a further evidence of the trophic role of neurotrophins on primary sensory neurons.

When the tibialis anterior muscle of the rabbit is progressively stretched during surgical limb distraction, the muscle fibres lengthen by addition of new serial sarcomeres, provided that stretch is carried out at an appropriate rate. However, in spite of the apparent adaptation to the new functional length, range of joint movement is greatly decreased. In this study we have first, made measurements of the passive tension developed by distracted muscles over the range of joint movement and secondly made quantitative measurements of endomysial and perimysial connective tissue content. It was found that at all ankle joint angles greater than 90°, the passive tension developed by the distracted muscles was greater than both contralateral and sham-operated controls. Image analysis showed that the ratio of collagen to contractile material was increased in distracted muscles compared with muscles from sham-operated controls, due to increased deposition of collagen type III. Scanning electron microscopy showed the presence of a dense perimysial weave surrounding the distracted muscle fibres. These quantitative and qualitative changes in the connective tissue component could account for the increased stiffness demonstrated by the physiological measurements. It would seem that in distracted muscle the connective tissue element adapts less readily than the contractile component, with prolonged stretch leading to damage to the perimysial and endomysial network, with subsequent fibrosis and loss of muscle compliance. Such changes could help explain the loss of range of movement noted in the distracted limbs of patients undergoing surgical limb-lengthening and in other conditions that result in muscle contractures.

Corpuscular nerve endings in the nasal respiratory mucosa of the dog were investigated by immunohistochemical staining specific for protein gene product 9.5 by light and electron microscopy. In the nasal respiratory mucosa, complex corpuscular endings, which displayed bulbous, laminar and varicose expansions, were distributed on the dorsal elevated part of the nasal septum and on the dorsal nasal concha. The endings were 300–500 μm long and 100–250 μm wide. Some axons gave rise to a single ending while others branched into 2 endings. Cryostat sections revealed that the corpuscular endings were located within the nasal respiratory epithelium. On electron microscopy, immunoreactive nerve terminals that contained organelles, including mitochondria and neurofilaments, were observed within the epithelial layer near the lumen of the nasal cavity. Some terminals contacted the goblet cell. Such terminal regions were covered by the cytoplasmic process of ciliated cells and were never exposed to the lumen of the nasal cavity. These nerve endings are probably activated by pressure changes.

The vertebral artery, in its course from the subclavian artery to the basilar artery, is vulnerable to damage or distortion from external factors such as bony, ligamentous or muscular structures (Mehalic & Farhat, 1974; Parkin et al. 1978; Schellhas et al. 1980; Braun et al. 1983; Dunne et al. 1987; Fast et al. 1987). In the atlas vertebra, the retroarticular canal and the lateral bridge are examples of bony outgrowth or exostosis which may cause external pressure on the vertebral artery as it passes from the foramen transversarium of the vertebra to the foramen magnum of the skull. If this pressure is severe enough, as may occur during the extreme rotatory movements carried out during therapeutic manipulation of the cervical spine, the vertebral artery may be compressed (Lamberty & Zivanovich, 1973), reducing its cross-sectional area, and compromising its blood flow (Taitz & Nathan, 1986). Vertebrobasilar ischaemia from compression of the vertebral arteries by osteophytes is an uncommon occurrence under normal circumstances (Warlow, 1996).

There are few studies of the lateral bridge of the atlas reported in the literature (MacAlister, 1869, 1893; Lamberty & Zivanovic, 1973; Saunders & Popovich, 1978; Taitz & Nathan, 1986). The lateral bridge was first described by MacAlister (1869, 1893) as a variety of the ‘posterior glenoid process’ (the retroarticular canal), which he termed the ‘gleno-transverse bony arch’. As its name implies, it is a lateral outgrowth of bone from the superior articular facet or lateral mass to the posterior root of the transverse process of the atlas (MacAlister, 1869, 1893; Lamberty & Zivanovic, 1973; Saunders & Popovich, 1978; Taitz & Nathan, 1986). The retroarticular canal is formed by an exostosis passing from the posterior surface of the lateral mass to the posterior margin of the vertebral artery groove of the atlas. Thus the lateral bridge forms another arch, secondary to the retroarticular canal, through which the vertebral artery must pass (MacAlister, 1869).

Tibial torsion, twisting of the tibia about its longitudinal axis, varies during development and early childhood. Knowledge of the normal range of tibial torsion at various ages and its accurate clinical measurement is important in the assessment of the extent of a torsional deformity. To evaluate tibial torsion a reliable technique for its measurement in vivo is therefore required. The aim of this study was to determine which of 4 existing in vivo methods of measuring tibial torsion was the most accurate and had the highest repeatability, by comparing them with direct measurement of the tibia. A wide range of mean values for tibial torsion was observed, using the various techniques, with none of the indirect techniques employed having a strong correlation with direct measurement of tibial torsion. The repeatability of the indirect techniques was observed to be low both in cadavers (n=4) and the living (n=3). Since none of the in vivo techniques appear to measure true tibial torsion or be of a reasonable repeatability, alternative easy to use and inexpensive methods need to be developed. Accurate clinical measurement of tibial torsion is important in the assessment of the extent of a torsional deformity. It is recommended that data gained using the methods reviewed here are interpreted with caution.

A macroscopic, microscopic and scanning electron microscope study was performed on the pathological bone changes of the mandibles of wild red deer (n=61) exhibiting severe dental fluorosis. The animals originated from a highly fluoride polluted area in Central Europe (Ore mountains and their southern foreland, Czech-German border region) and constituted 11.2% of the studied red deer sample (n=545) from this area. Pathologically increased wear and fracture of fluorosed teeth caused a variety of mandibular bone alterations, including periodontal breakdown, periostitis, osteitis and chronic osteomyelitis. As a further consequence of severe dental attrition, opening of the pulp chamber and formation of periapical abscesses were occasionally observed. In case of severe periodontal breakdown, loss of teeth from the mandibles was found. In addition to the inflammatory bone changes, the occurrence of osteofluorotic alterations was also diagnosed in the specimens with the highest bone fluoride concentrations (>4000 mg F−/kg dry wt). These changes comprised extended apposition of periosteal bone onto the mandibular cortex as well as deformation of the mandibular body, which was attributed to a fluoride-induced osteomalacia. The present study provided circumstantial evidence that, in addition to fluoride induced dental lesions, the occurrence of marked periodontal disease and tooth loss is an important factor responsible for a reduction of life expectancy in severely fluorotic wild red deer.

The mucosubstances in the epithelium lining the segment from gizzard to duodenum during development of the chick embryo was studied histochemically using monoclonal antibodies against gizzard mucus and lectins, with attention to the regional differentiation of the epithelium in this segment. The anterior limit of epithelial CdxA mRNA expression detected by in situ hybridisation, which served as the position of the gizzard-duodenal boundary, was clearly found from d 3. Granules positive for some antibodies or lectins were found in the region ranging from the posterior part of the gizzard to the duodenum at d 3, which was followed by an increase in the number of granules and a gradual enlargement of the granule-positive area to the anterior part of the gizzard over 4–6 d. From d 4, the epithelia of the gizzard body and of the pyloric or duodenal region came to be differently stained with some antibodies or lectins. From d 10, each region showed a specific pattern of staining. The epithelia of the gizzard body and pyloric region contained abundant mucus granules with a different staining pattern. In the duodenum the number of stained granules was low except in occasional goblet cells. Thus the epithelia of the gizzard body, pyloric region and duodenum may produce different mucosubstances and the regional differentiation in these epithelia may start at rather early stages soon after the formation of digestive tube.

A class of primitive progenitor cells with high proliferative potential in vitro (HPP-CFC), has been identified in fetal liver and adult bone marrow both in murine and human systems. The kinetic properties of HPP-CFC2 and the more mature granulocyte-macrophage colony forming units (CFU-GM) derived from murine fetal liver on d13, d15 and d19 of gestation, newborn liver and neonatal liver on d3 and d8 postpartum have been evaluated and compared with the kinetic properties of these progenitor cell populations derived from adult bone marrow. The frequency of HPP-CFC2 in fetal liver was found to be greatest on d15 of gestation then subsequently declined in newborn and neonatal liver. Similarly, the highest proportion of HPP-CFC2 engaged in DNA synthesis (53±3%) was detected in d15 fetal liver. This proportion decreased to 13±2% in the liver 1 wk after birth, which is comparable to the number of HPP-CFC2 derived from adult BM which were in S-phase (10±1%). Production of CFU-GM was found to be greater in adult bone marrow than in either fetal or newborn liver. While the proportion of CFU-GM in S-phase was high in all 3 tissue samples, the greatest proportion of cycling CFU-GM (50±2%) was detected in d15 fetal liver. These results suggest that HPP-CFC2 derived from fetal liver are actively cycling while HPP-CFC2 derived from adult bone marrow are relatively quiescent. In contrast, a high proportion of CFU-GM derived from fetal, newborn liver and adult bone marrow are engaged in DNA synthesis.

Merkel-like basal cells in the taste buds of the frog were examined by fluorescence histochemistry, immunohistochemistry and electron microscopy. There were about 16–20 basal cells arranged in a radial fashion at the base of each taste bud. These cells were strongly immunopositive for serotonin antiserum. They were characterised by the presence of numerous dense-cored granules in the cytoplasm ranging from 80 to 120 nm in diameter, and of microvilli protruding from the cell surface. For 4 mo after sensory denervation by cutting the gustatory nerves, all cell types of the taste bud were well preserved and maintained their fine structure. Even at 4 mo after denervation, the basal cells exhibited a strong immunoreaction with serotonin antiserum. To investigate the function of serotonin in the basal cells in taste bud function, serotonin deficiency was induced by administration of p-chlorophenylalanine (PCPA), an inhibitor of tryptophan hydroxylase, and of p-chloroamphetamine (PCA), a depletor of serotonin. After administration of these agents to normal and denervated frogs for 2 wk, a marked decrease, or complete absence, of immunoreactivity for serotonin was observed in the basal cells. Ultrastructurally, degenerative changes were observed in both types of frog; numerous lysosome-like myelin bodies were found in all cell types of the taste buds. The number of dense-cored granules in the basal cells also was greatly decreased by treatment with these drugs. Serotonin in Merkel-like basal cells appears to have a trophic role in maintenance of the morphological integrity of frog taste bud cells.

The sternalis muscle is an uncommon anatomical variant. It is located on the human anterior pectoral wall, superficial to pectoralis major. This muscle has been reported both in males and females, and in whites, blacks and Asians (Barlow, 1934; Kida & Kudoh, 1991; Shen et al. 1992; Bradley et al. 1996).

Although the importance of this muscle is still a mystery, various different interpretations have been made. Clemente (1985) considered sternalis to be a misplaced pectoralis major, although some embryologists have viewed it as part of a ventral longitudinal column muscle layer arising at the ventral tip of the hypomeres (Sadler, 1995). Sadler claimed that this muscle is represented by rectus abdominis in the abdominal region and by the infrahyoid musculature in the cervical region; in the thorax, this layer usually disappears but occasionally remains as a sternalis muscle. Kitamura et al. (1985) reported a case of congenital partial deficiency of pectoralis major accompanied by an enormous sternalis. Barlow (1934), on the other hand, claimed that sternalis represents the remains of a panniculus carnosus.

Metatarsal growth was studied in 780 human fetal metatarsal bones, crown-rump length ranging from 120 to 260 mm and gestational age ranging from 14 to 23 wk postconception. Total length, diaphyseal length and diameter were measured and statistical analysis performed. Means and standard deviations for each measurement were calculated at 1 wk intervals. These data were correlated with crown-rump length and gestational age and an equation was constructed in an attempt to predict gestational age.

The importance of continuing to record and discuss anatomical anomalies was addressed recently by Hicks & Newell (1997) in the light of technical advances and interventional methods of diagnosis and treatment. Two recent radiological reports (Bradley et al. 1996; Murphy & Nokes, 1996) highlighted the diagnostic dilemma posed by a sternalis muscle in the detection of breast cancer.

Frank Goldby, Emeritus Professor of Anatomy in the University of London at St Mary's Hospital Medical School (now Imperial College School of Medicine at St Mary's) died at Cambridge on 20 October 1997, aged 94, fully compos mentis as those who knew him would have expected, and content enough to go when he had to, having reached the same age as his father. He became a member of the Anatomical Society of Great Britain and Ireland in 1931, served as Secretary from 1946 to 1956 and as President from 1957 to 1959. He is survived by his wife Helen and 6 children, to whom our deepest sympathy is extended.

Prenatal feather growth development in the chicken was studied in 7 body regions in HH stages 27–45, using direct measurements, specific histological and immunohistochemical methods, and scanning electron microscopy. The results from measurements of absolute length values, and, particularly, growth rate development in each HH stage revealed a distinct phase of most intensive growth in HH stage 40–41, which was preceded by feather follicle insertion and accompanied by the occurrence of α-keratins in barbule cells. Specific regional evaluation demonstrated that growth in the feather follicles of abdominal skin generally showed the slowest progression from absolute values and that in the feather filaments of the developing wings the most rapid progression occurred during HH stage 40–41 from growth rate values.

Ultrasound of the calcaneus may be used as a cheap, ionising radiation-free and easy to use indicator of skeletal status, and hence of osteoporotic fracture risk. At present ultrasound is not widely used as it suffers from high precision errors. As ultrasound parameters are determined in part by bone mineral density (BMD), an increase in the accuracy and precision of BMD measurements should reduce the precision error associated with ultrasound measurements. The aim of this study was to define an anatomical site on the calcaneus at which accurate and precise measurements of BMD can be made. Ten dry calcanei and 10 cadaveric feet were scanned using a DXA scanner; 9 anatomically defined regions (1 cm2) were selected in the posterior part of the calcaneus for analysis. The centre of region 1 was positioned halfway along the line joining the anterior border of the calcaneal tubercle and the peak of the posterior superior tubercle, and the remaining 8 regions were placed around this central area. The BMD in these 9 regions was compared with the whole bone BMD and the variability of BMD within each of the 9 regions was measured. The reproducibility of the technique was assessed by taking 10 repeated measurements of 2 bone and 2 cadaveric specimens, each specimen being removed and repositioned between measurements. Region 1 was found to be the most representative of total BMD in cadaveric feet. This region also showed the least variability of BMD and consistently gave the lowest coefficients of variation in the reproducibility study both in the bone and the cadaveric specimens. This region is hence the most suitable site on the calcaneus for measuring absolute values of and changes in BMD. The surface position of region 1 was found to be consistently 5/9 along the line at 45° to the vertical, from the lateral malleolus to the heel. The identification of the surface location of region 1 relative to anatomical landmarks of the foot has enabled the same anatomical site to be measured in all subjects. This allows meaningful intersubject comparisons to be made. Preliminary data suggest that precision errors using ultrasound are also reduced when measurements are taken at this region of the calcaneus. The reduction in the precision error of ultrasound assessment of skeletal status may provide a cheap and safe way to identify individuals at risk from osteoporotic fracture.

A joint meeting of the Anatomical Society of Great Britain and Ireland, the Nederlandse Anatomen Vereniging and the Anatomical Society of Southern Africa was held from 15 to 17 April 1998 at Rolduc, Limburg, The Netherlands. It included symposia on ‘Recent advances in human evolution research’ on Wednesday 15 April and on ‘Adaptation of cells and tissues to mechanical stimuli’ on Thursday 16 April. The first EFEM Lecture was given by Professor J. Voogd of Erasmus University Rotterdam on ‘Transformations and interactions of cerebellar maps’ on Friday 17 April. The following are abstracts of communications and demonstrations presented at the meeting.

The position of taste buds around the human mouth, pharynx and larynx is outlined in many texts. While their distribution, histological differences and taste sensitivity on the dorsum of the tongue are well described, extralingual taste buds are described in far less detail both in adults (Bannister, 1995) and neonates (Lalonde & Eglitis, 1961). Studies on a variety of grazing mammals (e.g. Bradley et al. 1980 in sheep; Shrestha et al. 1993 in buffalo), omnivores (Andrew & Oliver, 1951 in rat) and carnivores (Palmieri et al. 1983 in cat and dog) have shown abundant taste buds on the laryngeal surface of the epiglottis. Some authors have suggested that these taste buds may be important in initiating reflexes protecting the airway (Bradley et al. 1980) and in this context it is surprising that literature on their distribution on the human epiglottis is not available.

The papillary layer of the oral part of the human soft palate was studied in 31 subjects of different age by means of histological, immunohistochemical and scanning electron microscopical methods. For scanning electron microscopy a new maceration method was introduced. Results determine epithelial thickness, height and density of connective tissue papillae and their 3-dimensional architecture inside the lining epithelium as well as the collagenous arrangement of the openings of the glandular ducts. The individual connective tissue papillae of the soft palate are compared with the connective tissue boundary on the other side of the oral cavity. The connective tissue plateaux carrying a variable number of connective tissue papillae were found to be the basic structural units of the papillary body. The function of the epithelial-connective tissue interface and the extracellular matrix arrangement in the lamina propria are discussed in order to promote the comparability of normal with pathologically altered human soft palates.