The topographical distribution of different fibre types in muscles of the lower hindlimb in rabbits and mice was quantitatively determined. The results were compared to those previously obtained, using the same new quantification methods, in homologous muscles of the rat. Type I fibres (‘slow’) were identified using myofibrillar ATPase histochemistry and mapped out at the mid proximo-distal level for 11 ‘fast’ muscles in the rabbit and 7 ‘fast’ muscles in the mouse. For the slow soleus muscle the procedure was undertaken for the type II fibres. Furthermore, for 5 of the ‘fast’ muscles from each animal species (extensor digitorum longus; flexor digitorum and hallucis longus; gastrocnemius medialis; peroneus longus; tibialis anterior), several more proximal and distal cross-sectional levels were also analysed. All the investigated ‘fast’ muscles showed a significant degree of topographical eccentricity in the midlevel distribution of type I fibres. For most muscles, the direction of this ‘vector regionalisation’ of type I fibres was similar between the three animal species. For homologous muscles, the degree of vector regionalisation was significantly different: mouse > rat > rabbit. The relative area of the region containing the type I fibres, inversely related to the degree of ‘area regionalisation’, was also significantly different: mouse < rat < rabbit. Also within each animal species, muscles with a marked degree of vector regionalisation tended to show a marked area regionalisation. Proximo-distal differences in type I fibre density were observed in all the three species of animals; also these patterns showed marked inter-species differences. The findings demonstrate the general occurrence of, and systematic relationships between, different aspects of type I fibre regionalisation. The observed interspecies differences suggest that the expression of this phenomenon is adapted to differing functional needs.

The curly tail (ct/ct) mouse mutant shows a high frequency of delay or failure of neural tube closure, and is a good model for human neural tube defects, particularly spina bifida. In a previous study we defined distinct domains of gene expression in the caudal region of non-mutant embryos during posterior (caudal) neuropore closure (Gofflot et al. Developmental Dynamics 210, 431–445, 1997). Here we use BrdU incorporation into S-phase nuclei to investigate the relationship between cell proliferation and the previously described gene expression domains in ct/ct mutant embryos. The BrdU-immunostained sections were also examined for abnormalities of tissue structure; immunohistochemical detection of perlecan (an extracellular heparan sulphate proteoglycan) was used as an indicator of neuroepithelial basement membrane structure and function. Quantitation of BrdU uptake revealed that at early stages of neurulation, cell proliferation was specifically reduced in the paraxial mesoderm of all ct/ct embryos compared with wild type controls, but at later stages (more cranial levels) it was increased. Those ct/ct embryos with enlarged posterior neuropore (indicating delay of closure) additionally showed an increased BrdU labelling index within the open neuroepithelium at all axial levels; however, this tissue was highly abnormal with respect to cell and nuclear morphology. It showed cell death and loss of cells from the apical surface, basement membrane defects including increased perlecan immunoreactivity, and increased separation from the underlying mesenchyme and notochord. These observations suggest that the mechanism of delay or failure of neuroepithelial curvature that leads to neural tube defects in curly tail embryos involves abnormalities of neuroepithelial-mesenchymal interactions that may be initiated by abnormal cellular function within the neuroepithelium. Minor histological and proliferation abnormalities are present in all ct/ct embryos, regardless of phenotype.

Mesenteric lymph nodes and gut-associated lymphoid tissue (GALT) from juvenile eastern grey kangaroos were investigated. The mesenteric nodes had a similar structure to that described for eutherian mammals. They contained distinct regions of medulla and cortex, with prominent follicles and germinal centres. Gut associated lymphoid tissue consisted of areas of submucosal follicles. These varied from areas of densely packed lymphocytes with darkly staining, prominent coronas to areas with no defined follicles. The distribution of T cells in these tissues was documented by use of species-crossreactive antibodies to the surface markers CD3 and CD5; B cells were identified by antibodies to CD79b. Within the lymph nodes T cells were located mainly in the paracortex and cortex, with limited numbers observed in the follicles; B cells were located on the marginal zone of the follicles. In GALT, T cells were located in the peripheral regions of the germinal centres of secondary follicles, while B cells were abundant in primary follicles. These observations are consistent with those made in a range of other marsupials (metatherian) and eutherian mammals and are indicative of the capacity to respond to antigens entering via the mouth.

Standard immunohistochemical methods were used to detect the presence of vimentin, cytokeratin 8, cytokeratin 18, macrophages and Langerhans cells in the human tonsillar epithelium in formalin-fixed and frozen tissue specimens. Vimentin detection was restricted to infiltrating cells of the lymphoid series, dendritic and vascular endothelial cells. All epithelial cells were negative. Cytokeratin 8 and 18 were readily detected in a large proportion of epithelial cells lining the crypt, but these cells bore no resemblance to the intestinal M-cells. Langerhans cells and macrophages were seen in both the oropharyngeal and crypt epithelium and were more common in the latter. This study confirms the presence of antigen-presenting cells, macrophages and Langerhans cells in the tonsillar epithelium and shows that intermediate filament proteins, vimentin, cytokeratin 8 and 18 are unreliable markers for human tonsillar M-cells, if indeed such cells exist in human tonsils.

An understanding of the origin of the sympathetic innervation of the upper limb is important in surgical sympathectomy procedures. An inconstant intrathoracic ramus which joined the 2nd intercostal nerve to the ventral ramus of the 1st thoracic nerve, proximal to the point where the latter gave a large branch to the brachial plexus, has become known as the ‘nerve of Kuntz’ (Kuntz, 1927). Subsequently a variety of sympathetic interneuronal connections down to the 5th intercostal space were reported and also described as the nerve of Kuntz. The aim of this study was to determine: (1) the incidence, location and course of the nerve of Kuntz; (2) the relationship of the nerve of Kuntz to the 2nd thoracic ganglion; (3) the variations of the nerve of Kuntz in the absence of a stellate ganglion; (4) to compare the original intrathoracic ramus with sympathetic variations at other intercostal levels; and (5) to devise an appropriate anatomical classification of the nerves of Kuntz.

Bilateral microdissection of the sympathetic chain and somatic nerves of the upper 5 intercostal spaces was undertaken in 32 fetuses (gestational age, 18 wk to full term) and 18 adult cadavers. The total sample size comprised 99 sides.

Sympathetic contributions to the first thoracic nerve were found in 60 of 99 sides (left 32, right 28). Of these, 46 were confined to the 1st intercostal space only. The nerve of Kuntz (the original intrathoracic ramus) of the 1st intercostal space had a demonstrable sympathetic connection in 34 cases, and an absence of macroscopic sympathetic connections in 12. In the remaining intercostal spaces, intrathoracic rami uniting intercostal nerves were not observed. Additional sympathetic contributions (exclusive of rami communicantes) were noted between ganglia, interganglionic segments and intercostal nerves as additional rami communicantes. The eponym nerve of Kuntz should be restricted to descriptions of the intrathoracic ramus of the 1st intercostal space. Any of these variant sympathetic pathways may be responsible for the recurrence of symptoms after sympathectomy surgery.

The collagen fibrillar architectures in the general matrix of cartilage slices removed from both normal and osteoarthritic femoral heads were examined by both differential interference light microscopy and scanning electron microscopy. Whereas the normal general matrix contained a finely differentiated pseudo-random weave of fibrils developed from an interconnected array of radial elements, the osteoarthritic general matrix was characterised by the presence of structurally distinct regions consisting of strongly aligned radial bundles of fibrils and associated intense tangles or ‘knotted’ features. Simple structural models were developed to explore possible transformation structures based on two different types of interconnectivity in the three-dimensional fibrillar network. These models support the hypothesis that the distinctive ultrastructural features of the osteoarthritic general matrix can develop as a consequence of largely passive degradative changes occurring in the fibrillar weave originally present in the normal matrix. This could, in principle, occur independently of any new structure that might develop as a consequence of any upregulation of collagen associated with the osteoarthritic process.

The ultrastructure of Betz cells in the 5th layer of the primary motor cortex of 17 neurologically and psychiatrically normal control individuals was studied. Normal-appearing Betz cells showed a wide range of features including novel electron-dense inclusion bodies (Bunina-like bodies) resembling Bunina bodies characteristic of amyotrophic lateral sclerosis (ALS), accumulations of neurofilaments (10 nm in diameter), bundles of filaments (20–25 nm in diameter) thicker than neurofilaments, lamellar structures, lamellar bodies and structures similar to Hirano bodies. Among these ‘abnormal’ features, the presence of Bunina-like bodies may be an age-related nonspecific degenerative change, since they appeared more frequently in elderly individuals. The presence of these abnormal features—particularly the Bunina-like bodies—in the Betz cells of normal human brains must be considered in the assessment of the pathognomonic significance of such structures in ALS and other neurological diseases that affect the motor cortex.

References to histochemistry are extensive for human limb muscles but occur less frequently in relation to vertebral muscle. Most vertebral muscle literature has been concerned with muscle fibre characteristics in the lumbar and thoracic spine, due in large part to the incidence of low back pain and idiopathic scoliosis. However few studies have investigated the histochemical composition of neck muscles in humans: and, to our knowledge, no previous study has examined the antagonistic longus colli and multifidus muscle pair. In addition, while age-related segmental degeneration is most prominent between C5 and C7, it is not known whether these osteoligamentous changes are paralleled by changes in muscle fibre ratio. Tissue blocks comprising muscle and bone from C5–C7 segments were harvested at autopsy from 16 subjects with ages ranging from 4 to 77 years. The prevertebral longus colli and postvertebral multifidus muscle pairs were randomly selected from one or other side in each subject. The tissue was frozen, sectioned and histochemically stained for myofibrillar adenosine triphosphatase. Analysis of muscle fibre types was performed by light microscopy. Wilcoxon paired t-tests were used to ascertain whether intramuscular and intermuscular differences in fibre composition were significant. In addition, correlation and regression analyses were used to determine whether fibre type proportions changed in either muscle with increasing age. The present study has revealed histochemical differences between longus colli and multifidus at the level of the C5–C7 vertebral segments. Multifidus comprises a significantly greater proportion of type I than type II fibres. Longus colli comprises a significantly greater proportion of type II fibres than multifidus. Further there were no changes in fibre type proportion in either muscle with increasing age. These observations suggest that longus colli responds equally to postural and phasic demands, whereas multifidus is predominantly postural. Also it would appear that age-related structural alterations in lower cervical segments are not paralleled by changes in muscle fibre ratio.

The aim of this study was to investigate the risk and to analyse the significance of laceration of the sural and superficial fibular nerves during the surgical approach to the lateral malleolus. The sural and the superficial fibular nerves, and their branches were dissected under ×3 magnifying lenses in 68 embalmed leg-ankle-foot specimens. The specimens were measured, drawn and photographed. In 35% of specimens the superficial fibular nerve branched before piercing the crural fascia, and in all these specimens the medial dorsal cutaneous nerve of the foot was located in the anterior compartment while the intermediate dorsal cutaneous nerve of the foot was located in the lateral compartment. In 35% of specimens the intermediate dorsal cutaneous nerve of the foot was absent or did not innervate any toe. The deep part of the superficial fibular nerve was in contact with the intermuscular septum. Its superficial part was parallel with the lateral malleolus when the nerve pierced the fascia more proximally and oblique to the lateral malleolus when the nerve pierced the fascia distally. In one case the intermediate dorsal cutaneous nerve of the foot was in danger of laceration during a subcutaneous incision to the lateral malleolus. In 7 cases (10%) the sural nerve overlapped or was tangent to the tip of the malleolus. Malleolar nerve branches were identified in 76% of the cases (in 28% from both sources). The sural nerve supplies the lateral 5 dorsal digital nerves in 40% of cases. Our study indicates that during the approach to the lateral malleolus there is a high risk of laceration of malleolar branches from both the sural and the superficial fibular nerves. There is less risk of damage to the main trunk of these nerves, but the 10% chance of laceration of sural nerve at the tip of the malleolus is significant. As the sural nerve supplies the superficial innervation to the lateral half of the foot and toes in 40% of cases, the risk of its laceration is even more important than indicated by the common anatomical teaching.

After a surgically induced partial obstruction of the small intestine (ileum) in adult rats there is an accumulation of ingesta and a progressive enlargement of the lumen accompanied by wall thickening: over a period of 2–3 wk the circumference of the hypertrophic intestine increases by a factor of 2·7 and the thickness of the musculature increases more than threefold, while the length of the ileum (measured at the mesenteric attachment) remains unchanged. The villi become markedly larger and more elongated in the circumferential direction, and have a greater separation between one another. The number of villi per unit surface is markedly reduced but the number of villi per unit length of ileum, whilst appearing to show a small increase, was not significantly altered. The component epithelial cells (absorptive cells) appear unchanged in morphology and size (height). The microvilli of the epithelial cells have the same appearance, size (height) and packing density in the control and the hypertrophic ileum. Glands of Lieberkühn, Peyer's patches and single lymphatic follicles constituting the Peyer's patches are significantly increased in size in the hypertrophic intestine. The serosal surface of the hypertrophic ileum, in spite of the great expansion, remains regularly covered by mesothelial cells; these are much larger than in the controls and have an altered distribution of their microvilli.

Previous light microscopic investigations have indicated that, in the rat, the innervation of mesenteric lymph nodes may be less dense than that of axillary nodes. However, nerves of the enteric system are difficult to visualise by light microscopy. Therefore we quantified the density of innervation of axillary and mesenteric lymph nodes at the ultrastructural level. The results show a highly significant difference in the density of innervation between these 2 groups of lymph nodes, but morphologically the type of the innervation does not seem to differ. In previous studies, nerves were found predominantly in regions characterised by aggregations of plasma cells. In view of this association, we suggest that the difference in innervation may reflect the migration of plasma cells out of mesenteric nodes and into the mucosa of the gut wall. By contrast, in peripherally located nodes, plasma cells tend to remain within the lymph nodes, and hence the density of innervation of these nodes is greater.