Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A comparative morphological study of two human facial muscles : the orbicularis oculi and the corrugator… Goodmurphy, Craig Wayne 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1996-0582.pdf [ 16.13MB ]
JSON: 831-1.0087212.json
JSON-LD: 831-1.0087212-ld.json
RDF/XML (Pretty): 831-1.0087212-rdf.xml
RDF/JSON: 831-1.0087212-rdf.json
Turtle: 831-1.0087212-turtle.txt
N-Triples: 831-1.0087212-rdf-ntriples.txt
Original Record: 831-1.0087212-source.json
Full Text

Full Text

A C O M P A R A T I V E M O R P H O L O G I C A L S T U D Y OF TWO H U M A N F A C I A L M U S C L E S T H E ORBICULARIS OCULI A N D T H E C O R R U G A T O R SUPERCILII By CRAIG W A Y N E G O O D M U R P H Y H.B.A., The University of Western Ontario, 1992  A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Anatomy)  We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH C O L U M B I A August 1996 © CRAIG W A Y N E G O O D M U R P H Y ,  1996  In presenting degree  at the  this  thesis  in partial fulfilment of  University of British Columbia,  freely available for reference copying  the  requirements  I agree that the  or  by  his  or  an advanced  Library shall make it  and study. I further agree that permission for extensive  of this thesis for scholarly purposes may be granted  department  for  her  representatives.  It  is  by the head of my  understood  that  copying  or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department of The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  (QoX  Qjr  °l  ABSTRACT Facial muscles have two unique functions: as sphincters and dilators controlling the orifices of the face, and as movers of the skin of the face to produce facial expressions. It was hypothesized that human facial muscles sharing the same innervation and embryology, but having different functions, would posses morphologic differences in architecture, histology, cytochemistry and ultrastructure. To test this hypothesis, two periorbital muscles, the palpebral part of the orbicularis oculi (OO) and the corrugator supercilii (CS), were compared. The O O is a sphincteric muscle, whereas the CS is a muscle of facial expression. Whole muscle samples from human cadavers and biopsies from cosmetic surgery procedures were used. Quantitative measures of fiber sizes, fiber shapes and fiber-type distributions were performed along with measures of capillary area per unit of contractile area (capillary index). Qualitative analyses of nerve and motor end-plate distributions were also undertaken. Architectural differences were elucidated by stereoscopic dissection, conventional histological stains, and electron microscopy. Innervation patterns and motor end-plate regions were demonstrated with a pararosanaline (PIA) stain, and with an antibody to neurofilament protein. Fiber-type profiles were visualized by immunofluorescent microscopy using antibodies to fast and slow myosin. The OO was shown to differ significantly from the CS on the basis of fiber shapes, sizes and types. The OO fibers were small, rounded and 89% of them were type II. The CS fibers were larger, pleomorphic in shape, and only 49% of them were type II. The capillary index of the CS was 2.4 times that of the OO. The innervation of the OO was more intricate and arose from the deep surface of the muscle, whereas that of the CS was less ordered and arose from the superficial surface. Ulfrastructurally, the CS and OO shared many features, but also demonstrated differences in mitochondrial content and distribution, as well as motor end-plate structure. The observed differences between the two muscles support the contention that the function of human facial muscles influences their morphology. ii  T A B L E OF CONTENTS  ABSTRACT  ii  T A B L E OF CONTENTS  iii  LIST OF FIGURES  vi  ACKNOWLEDGMENTS  viii  INTRODUCTION  1  FACIAL MUSCLES  1  Historical review  2  Development  4  FACIAL N E R V E  6  Special sensory  6  General sensory  6  Visceral motor  7  Branchial motor  7  Course of the facial nerve  8  A N A T O M Y OF PERIORBITAL M U S C L E S Orbicularis oculi  9 9  Corrugator supercilii  11  CLINICAL R E S E A R C H  11  Surgical reanimation  12  Pharmacological therapy  15  MORPHOLOGICAL RESEARCH  17  Light microscopy  17  Electron microscopy  20  FIGURES 1-3  22  THESIS OBJECTIVES A N D R A T I O N A L E  28  iii  MATERIALS A N D METHODS  32  Gross anatomical dissection  32  Conventional brightfield microscopy  32  Histochemistry  34  Immunocytochemistry  35  Morphometric analysis  36  Transmission electron microscopy  37  Statistical analysis  :  OBSERVATIONS  38  39  MACROSCOPIC A R C H I T E C T U R E  39  MICROSCOPIC A R C H I T E C T U R E  41  Fiber sizes and shapes  42  Fiber typing  44  Mitochondrial staining profiles  45  Capillary index  46  Innervation patterns  47  E L E C T R O N MICROSCOPY  49  Fine structure  49  Mitochondria  51  Capillaries  51  Motor end-plates  52  Other features  52  FIGURES 4-26  55  DISCUSSION  101  U L T R A S T R U C T U R A L DIFFERENCES Mitochondrial distribution  iv  102 103  Lamellar inclusions  106  LIGHT MICROSCOPIC DIFFERENCES  108  Fiber type distribution  108  Muscle fiber shapes and sizes  109  GROSS A N A T O M I C A L DIFFERENCES  112  Motor end-plate zones  112  Motor nerve distribution  115  Sensory innervation  116  CONCLUSIONS A N D F U T U R E DIRECTIONS  121  REFERENCES  126  c  V  LIST OF FIGURES Fig. 1  Drawing of Human Facial Muscles  22  Fig. 2  Diagram Showing the Distribution Pattern of the Facial Nerve  Fig. 3  Drawing Showing Bony Attachment Sights of the Corrugator and Orbicularis  26  Fig. 4  Drawing and Four Stereoscopically Dissected Regions of the Orbicularis  55  Fig. 5  Drawing and Four Stereoscopically Dissected Regions of the Corrugator  57  Fig. 6  Light Micrographs of the Corrugator and Orbicularis Stained with H & E  59  Fig. 7  Histogram Comparing the Mean Muscle Fiber Shape of the Corrugator and the  Pes Anserinus  Orbicularis Fig. 8  Fig. 9  24  61  Immunofluorescence Micrographs of Anti-Fast Myosin Staining of the Corrugator and the Orbicularis  63  Histogram Showing Fiber Type Distributions in the Corrugator and Orbicularis  65  Fig. 10 Histogram Comparing the Mean Cross Sectional Area of Type I and Type II Muscle Fibers in the Corrugator and Orbicularis Fig. 11 Histogram Comparing the Mean Maximum Diameter of Type I and Type II Muscle Fibers in the Corrugator and Orbicularis Fig. 12 Light Micrographs of the Corrugator and Orbicularis Stained with NADH-tr  67  69 71  Fig. 13 Micrographs of Semi-thin Epon-embedded Sections of the Corrugator and Orbicularis Stained with Toluidine Blue Fig. 14 Histogram Comparing the Capillary Index of the Corrugator and Orbicularis  73 75  Fig. 15 Immunofluorescence Micrographs of Anti-NFP Staining of the Corrugator and the Orbicularis Fig. 16 Light Micrographs of Serial Sections of Corrugator Stained with H & E , NADH-tr Anti-NFP and Anti-Slow Myosin  vi  77 79  Fig. 17 Light Micrographs of Serial Sections of Orbicularis Stained with H & E , NADH-tr Anti-NFP and Anti-Slow Myosin  81  Fig. 18 Three Views of Orbicularis Motor End-Plates: Distribution Drawing, PIA Stained Clusters and a Single Motor End-Plate  83  Fig. 19 Three Views of Corrugator Motor End-Plates: Distribution Drawing, PIA Stained Clusters and a Single Motor End-Plate  85  Fig. 20 Light Micrographs of PIA Stained Motor End-Plates from Murine E D L and Gastrocnemius Controls  87  Fig. 21 Longitudinal and Cross Sectional Transmission Electron Micrographs of Normal Myofibrillar Arrangements in the Corrugator and Orbicularis  89  Fig. 22 Four Electron Micrographs of the Ultrastructural Features of the Corrugator  91  Fig. 23 Four Electron Micrographs of the Ultrastructural Features of the Orbicularis  93  Fig. 24 Electron Micrographs of Two Types of Capillaries Seen in the Corrugator and Orbicularis  95  Fig. 25 Electron Micrographs of the Ultrastructural Features of a Corrugator and an Orbicularis Motor End-Plate  97  Fig. 26 Micrographs of Four Structures of Interest Observed in the Corrugator and the Orbicularis  99  vii  ACKNOWLEDGMENTS  My first thank you must go to the many unnamed people who have assisted me throughout my program, for they supplied the daily fuel I needed in order to grow as an individual. Next I wish to thank the people in the Department of Anatomy for cultivating an environment that is personable and encouraging to graduate students. I would like to thank my lab partner Patrick Nahirney for his technical expertise, and Dr. Benjamin Gelfant for supplying me with human tissue samples. My sincere gratitude is extended to my supervisor Dr. William Ovalle, and the rest of my committee, Dr. Pierre Dow, Dr. Wayne Vogl and Professor Elizabeth Akesson. It has been a pleasure and a privilege to learn from each of them. Their personalized support and encouragement along with their combined wisdom and patience allowed me to develop more fully as a researcher and a person. As teachers, researchers and people they all cast a brilliant shadow. Heartfelt thanks goes to all of my family, including the Osborns, my family away from family. Finally, I would like to dedicate this thesis to my mother, Ann, and my sister, Angela, for whom there are no words adequate enough to express my thanks, admiration and appreciation. Their patience, support, and unconditional love have been instrumental in all of my successes.  viii  INTRODUCTION FACIAL MUSCLES The human face is an evolutionary feature of profound importance. The mobility of the face is owed entirely to a unique set of flat skeletal muscles which are called either the muscles of facial expression or the mimetic muscles. Mimetic muscles share several common features. A l l of the muscles have a common embryonic origin, and are innervated by branches of the same cranial nerve (Patten, 1968; Hazelton, 1969; Hamming, 1983; Sataloff, 1990; Ammirati et al., 1993). They are all located in the subcutaneous tissue of the face where they share a complex spatial arrangement, originating from the bone or fascia of the skull and inserting into the dermis of the face. In addition, all but one of the facial muscles, the buccinator, lack a demonstrable deep fascial covering (Stennert et al., 1985; Williams et al., 1989). This unique set of muscles performs two anatomic functions: 1) as dilators and sphincters to control the opening and closing of facial orifices, and 2) as movers of facial skin, to make facial expressions and to assist with verbal and non-verbal communication (Cruveilhier, 1844; Duchenne, 1876; Leibgott, 1982). With roughly 21 paired facial muscles, humans enjoy a repertoire of expressions that greatly exceeds that of any other animal (Freilinger et al., 1987; Williams et al., 1989). Our expressive ability has led to the development of a distinct method of emotional non-verbal communication that utilizes our fine motor control over this muscle group to its fullest. The development of facial communication has resulted in intimate links being formed between the voluntary motor control of the facial muscles and the involuntary emotional centers of the human brain. The link between the two centers and the development of emotional language have placed a social dependence and importance  1  on facial muscles such that there can be devastating personal and social reprisals associated with the loss of their control or function.  Historical review The first documented mention of the muscles of the human face was by the Graeco-Roman physician, Galen (131-201 AD) in his book entitled De Usu Partium or "the Usefulness of the Parts" (Tallmadge, 1968). In De Usu Partium, Galen commented on the intimate nature of the attachments of some of the facial muscles into the skin by writing: "The skin is even more marvelously joined to the muscles of the lips [than the frontalis muscle is]. For you could not say that here the muscles are stretched underneath and the skin grows above them, as on the forehead, on both jaws in many places and even on the inner sides of the hands andfeet, since in these places one can separate them and set definite limits where muscle ends and the skin begins. Throughout the lips, however, a blending has occurred, the muscles and the skin being so lost in one another and so intermingled that you cannot call the product of the two either muscle or skin, and you cannot separate the whole into its parts but must in justice call the lips of animals either skin-like muscle or musclelike skin." Galen's descriptions of the facial muscles were sparse and in some ways erroneous, but they were insightful and historically significant. Before his keen observations were recorded, all knowledge about facial muscles was passed from person to person and eventually became lost in the mists of prehistory. Following De Usu Partium, there was a lengthy lull in anatomical study as learning was submerged into the Dark Ages (Andrew and Ernest, 1972). Whatever the reasons, original investigation lost popularity in almost all fields of endeavor. It seemed as if Galen had said the last word, and for 14 centuries there was nothing more to be said either about facial muscles or their anatomy. Yet, despite the loss of interest in academic pursuits, the records of Galen's work remained until the Renaissance brought the rebirth of learning and the Greek spirit of scientific inquiry.  2  Andreas Vesalius (1514-1564) took up anatomical study where Galen had left off. Considered the "Father of Modern Day Anatomy", Vesalius made over 200 corrections to Galen's descriptions (Lind, 1949; Saunders and OMalley, 1950). Beyond being the person to rekindle anatomical studies, his contribution to the anatomy of the facial musculature was one of detail. He was the first to present sketches of the facial muscles and to functionally distinguish them from the muscles of mastication (Lind, 1949). Through observation and skilled dissection, Vesalius delineated several of the individual muscles of facial expression. He also made reference to the special function of the buccinator and its "somewhat different" role, recognizing that it is involved in mastication and has only a minor role in producing facial expressions (Lind, 1949). Vesalius' work added greatly to the work began by Galen, even though his descriptions of facial muscles were still very sketchy and reflected his lack of understanding of the mimic system. Some of the missing details camefromthe anatomical investigations carried out by Santorini (1681-1737), who discovered a number of the smaller facial muscles that were previously grouped as part of a larger mass (Andrew and Ernest, 1972). One such muscle, the risorius santorini, bore his name until it was shortened to risorius in the mid-20th century (Williams et al., 1989). Still more details camefromthe German anatomist Bernard Siegfried Albinus (1697-1770). The eldest son of a professor of medicine in Frankfort, Albinus sought his education at Leyden University in the Netherlands (Hale and Terence, 1979). Leyden was a great academic and cultural center known for its artistic influences, such as those of Rembrandt. Albinus recognized the need for artistic accuracy in representing the human form and enlisted the help of artist Jan Wandelaar. The anatomical drawings resulting from their collaboration possessed artistic brilliance and illustrated facial muscles with an accuracy and clarity far beyond those of their predecessors. Their work served  3  as a great impetus for those who would follow, and many of their illustrative techniques have influenced some of the more current works in anatomy (Hale and Terence, 1979). During the mid-19th century, French anatomist J. Cruveilhier commented on the unsatisfactory state of knowledge regarding, "the muscle group to which our emotions are trusted." (Andrew and Ernest, 1972). He made mstrumental efforts to categorize the facial muscles into functional units in his book System ofAnatomy, republished in English as The Anatomy of the Human Body (Cruveilhier, 1844). First, Cruveilhier classed the muscles of the face as either dilators or constrictors. He further grouped the muscles around the three openings of the face into three distinct regions, namely the palpebral region, the nasal region and the buccal region (Cruveilhier, 1844). Cruveilhier, however, lacked an understanding of the nervous supply to the facial muscles. None-theless, his classification of the muscles along with his presentation of their actions and their basic morphology was a substantial contribution towards elucidating the human facial muscles. The link between facial muscles and their nerve source, the facial nerve (or seventh cranial nerve), is credited to the pioneering work of Sir Charles Bell, after whom the condition known as Bell's palsy is named (Bell, 1821; Hoffman, 1992). Through elegant dissection and talented artistry, he was the first person to associate facial paralysis with injury of the facial nerve (Bell, 1821; Duchenne, 1876; Cuthbertson, 1990). His work was essential in elevating the understanding of facial gross anatomy to a level that is still relevant today.  Development Embryologically, the facial muscles arise from mesenchymal cells that form in the spaces between the developing brain and the surface ectoderm, caudal to the otic vesicles. Masses of these mesenchymal cells are crowded into the four branchial (gill) arches which develop in humans. Each  4  branchial arch consists of four structures of phylogenetic significance that become modified to form ontogenetically useful structures (Patten, 1968; Hazelton, 1969; Sperber, 1981). Each pair of branchial arches consists of: 1) a central rod of hyaline cartilage forming the skeleton of the arch, 2) a muscular component, called a branchiomere, 3) an aortic arch artery, and 4)  an associated cranial nerve with sensory components to the mucosa and motor components to the skeletal muscle of the arch (Patten, 1968). The arches are named in three ways; either by arch number, by the associated cranial nerve,  or by the bony or cartilaginous structures arising from each arch. This means that the second arch, the one giving rise to the muscles of facial expression, is often referred to as either the facial arch or the hyoid arch (Patten, 1968; Sperber, 1981; Leibgott, 1982; Williams et al., 1989). In addition to forming the hyoid bone, the cartilage of the hyoid arch (Reichert's cartilage) forms the styloid process, the stylohyoid ligament, and the stapes bone. Reichert's cartilage also contributes to the malleus and the incus bones of the middle ear (Patten, 1968; Sperber, 1981; Leibgott, 1982; Williams et al., 1989). The branchiomere of the hyoid arch subdivides into deep and superficial layers that migrate onto the face during the 6th and 7th weeks in utero. The superficial sheet of developing muscle spreads to the sides of the head to give rise to thefronto-occipitalis,the auricular, and the platysma muscles. The deeper layer subdivides and organizes itself around the nasal, oral and orbital openings to give rise to the remaining muscles of facial expression (Fig. 1) (Patten, 1968). Slips of the second branchiomere also migrate to form the stapedius muscle, the stylohyoid muscle, and the posterior belly of the digastric muscle, all of which drag with them, during migration, a motor nerve branch from the facial nerve (Patten, 1968). 5  FACIAL NERVE The facial nerve, or cranial nerve VII, has two motor and two sensory components. It contains branchial and visceral motor divisions, as well as general and special sensory divisions (Proctor, 1991). The visceral motor and both sensory portions are carried separately as a distinct nerve bundle referred to as the nervus intermedius. The branchial motor portion, on the other hand, is the larger nerve bundle. Both bundles are wrapped together in an epineurial sheath and comprise the facial nerve (Wilson-Pauwels et al., 1988). Special sensory The special sensory part of the nervus intermedius transmits the signal from the chemoreceptor cells in the taste buds on the anterior two-thirds of the tongue and from the hard and soft palates. Its course follows a branch of C N V3, the lingual nerve, until it separates into a branch called the chorda tympani. Chorda tympani courses as part of the facial nerve towards the pseudounipolar cell bodies found in the geniculate ganglion within the facial canal of the petrous temporal bone. From this ganglion, processes of the neurons travel via the nervus intermedius to the nucleus solitarius in the brainstem (Vidic, 1978; Sataloff, 1990). General sensory The general sensory component of the nervus intermedius conveys sensation from a small patch of skin behind the pinnea and the skin over the auricular concha. Nerve signals are carried by a few general sensory axons that travel through the posterior auricular branch of the facial nerve and the auricular branch of the vagus nerve. The postsynaptic cell bodies of these sensory nerve fibers are also located in the geniculate ganglion. The presynaptic axons, however, travel through the nervus intermedius to the spinal portion of the trigeminal nuclei in the brainstem (Vidic, 1978; WilsonPauwels et al., 1988).  6  Visceral motor Visceral motor fibers of the nervus intermedius function to control secretion of all of the exocrine glands of the head except the integumentary and parotid glands. These include the lacrimal glands, mucous glands of the paranasal sinuses and nasal passage, glands of the soft and hard palates, and the submandibular and sublingual glands. These preganglionic parasympathetic nervefiberstravel with other nerves (such as the chorda tympani) or as separate nerves (such as the greater superficial petrosal nerve) to synapse in three different ganglia (pterygopalatine, submandibular and sublingual ganglia). From these ganglia their postsynaptic fibers travel to their designated target glands. The cell bodies for all of the preganglionic fibers are situated in the superior salivatory nucleus, which is composed of several nuclei scattered in the pontine tegmentum of the brainstem (Malone and Maisel, 1988; Fujii and Goto, 1989; Hunyor, 1994). Branchial motor Voluntary control of the stapedius, stylohyoid, posterior belly of digastric, and all muscles of facial expression is carried by branchial motor fibers iiinning in the largest part of the facial nerve, lateral to the nervus intermedius (Malone and Maisel, 1988; Monkhouse, 1990; Proctor, 1991). The signals for controlling the contraction of the facial muscles begin in various regions of the cerebral cortex. From the cerebral cortex, signals travel in the corticobulbar tract to the contralateral and ipsilateral facial motor nuclei in the brainstem. After synapsing in the motor nucleus of C N VII the axons pass posteromedially to hook around the nucleus of the abducens nerve (CN VI), causing the bulging landmark on the floor of the fourth ventricle known as the facial colliculus. Nervefibersthen exit the brainstem by traveling ventrolaterally to leave the pons along its inferior border (Podvinec and Pfaltz, 1976; Wilson-Pauwels et al., 1988).  7  Course of the facial nerve For convenience, the course of the facial nerve can be divided into three parts:  the  intracranial segment, the cranial segment, and the extracranial segment. The intracranial segment begins at the pontomedullary junction as the facial nerve exits the brainstem. Alongside the vestibulocochlear nerve, the facial nerve heads laterally and slightly forward, crosses the subarachnoid space of the posterior cranial fossa, and enters the internal acoustic meatus (Podvinec and Pfaltz, 1976; Proctor, 1991). Upon entering the internal acoustic meatus, it begins its cranial course. Cranial nerves VII and VIII wind through the petrous temporal bone until they separate, and then the facial nerve enters the facial canal. Here the facial nerve bends sharply to form the external genu and exhibits a slight swelling (the geniculate ganglion) before heading downward to exit the skull at the stylomastoid foramen (Ruskell, 1985; Williams et al., 1989). The extracranial segment begins once the facial nerve exits the temporal bone. It continues in an anterior and inferior direction for several millimeters, giving off the posterior auricular nerve and the nerve to the posterior belly of digastric and the stylohyoid muscle, before entering the substance of the parotid gland (Niccoli Filho and Varandas, 1988; Myint et al., 1992). Within the parotid gland, the facial nerve divides into five branches known as the pes anserinus. The pes anserinus consists of temporal, zygomatic, buccal, mandibular and cervical branches. However, there is great variability between the branching patterns of these nerves, ranging from simple branches with no interconnections to complex reticular patterns containing numerous interconnecting branches (Fig. 2) (Katz and Catalano, 1987; Williams et al., 1989; Monkhouse, 1990; Myint et al., 1992).  8  A N A T O M Y O F T H E PERIORBITAL MUSCLES The 37th edition of Grey's Anatomy (Williams et al., 1989) divides the facial muscles into four groups: epicranial musculature, circumorbital musculature, nasal musculature, and buccolabial musculature. The circumorbital muscles, or periorbital muscles, consist of the orbicularis oculi and the corrugator supercilii.  Orbicularis oculi The orbicularis oculi is a broad, flat, elliptical-shaped skeletal muscle surrounding the orbital opening and spreading over the eyelids. It can be subdivided into three parts: an orbital part, a palpebral part, and a lacrimal part (Zide, 1981; Leibgott, 1982; Siegel, 1993). The orbital portion is the thickest of the three regions. Its muscle fibers wrap around the anterior temporal, infraorbital, and superciliary regions in an elliptical pattern forming a continuous sphincter of muscle interrupted only by the medial canthal tendon (Furnas, 1981; Patrinely and Anderson, 1988). Some of the uppermost muscle fibers blend into the frontalis and corrugator supercilii muscles. Medially, muscle fibers blend slightly with the procerus and inferiorly with the levator labii superioris alequae nasi, levator labii superioris and zygomaticus minor muscles. The lateral side of the orbicularis oculi has no bony attachment other than a fine reticular network of collagen fibers that holds it loosely against the orbital rim. The medial side has three firm sites of attachment. It is attached to the superomedial angle of the orbit on the frontal bone, the inferomedial angle of the maxilla, and the medial palpebral ligament between the other two attachments (Fig. 3). Periosteal extensions off the orbital rim are linked to the orbicularis oculi via a fascia called the orbital septum (Furnas, 1981; Haniming, 1983). Reticular connective tissuefibersof the orbital septum pass anteriorly between the muscle fibers of the orbicularis oculi to adhere  9  mtimately to the overlying skin. Orbital muscle fibers of the orbicularis oculi receive nervous input from both the temporal and zygomatic branches of the facial nerve (Fig. 2) (Smith and Breathnach, 1990). The lacrimal portion of the orbicularis oculi attaches to the lacrimal fascia, the lacrimal crest, and the lateral surface of the lacrimal bone. Behind the lacrimal sac, it divides into upper and lower slips where some muscle fibers attach to the tarsi near the lacrimal canaliculi. The remaining muscle fibers anchor onto the medial palpebral raphe (Fig. 3) (Patrinely and Anderson, 1988). The palpebral portion is the tliinnest part of the orbicularis oculi. Its muscle fibers sweep across the palpebral fissure horizontally and anterior to the orbital septum. They anchor to the medial palpebral ligament and converge on the lateral side into the lateral palpebral raphe (Fig. 3) (Patrinely and Anderson, 1988). The lacrimal and palpebral parts of the orbicularis oculi receive their nerve supply from temporal branches of the facial nerve (Fig. 2) (Ruskell, 1985; Smith and Breathnach, 1990; Ammirati et al., 1993). As a whole, the orbicularis oculi is important as a muscle of facial expression and as the sphincter muscle of the orbit. The orbital part is primarily a voluntary muscle involved in forceful closure of the eye (Hung et al., 1977; Doane, 1980; Siegel, 1993). It also functions to shield the eyes from bright light by protruding the brow. The lacrimal portion pulls the lids and lacrimal papillae medially and dilates the lacrimal sac to assist with tear drainage. The palpebral part, in contrast, is involved in rapid involuntary blinking as well as gentle voluntary closure of the palpebral fissure (Hung etal., 1977; Doane, 1980; Furnas, 1981).  10  Corrugator supercilii The corrugator supercilii is an obliquely-positioned and pyramidal-shaped muscle that receives innervation from the temporal branch of the facial nerve (Fig. 2) (Ruskell, 1985; Siegel, 1993). Medially, the corrugator attaches to thefrontalbone by anchoring to the medial edge of the superciliary arch (Fig. 3). The muscle fibers run linearly in a superolateral direction. At the lateral edge muscle fibers blend with the more superficialfrontalismuscle and with the subcutaneous tissue of the eyebrow (Leibgott, 1982). The orientation of the corrugator supercilii allows it to exert traction on the eyebrow and to draw it medially and downward, producing the vertical glabellar creases between the brow. In addition, the corrugator assists the orbital portion of the orbicularis oculi in protruding the brow (Williams et al., 1989; Flowers et al., 1993).  CLINICAL RESEARCH Disorders involving the facial muscles range from spasmodic dystonia and blepharospasm to facial palsy and paralysis. These conditions, although not life threatening, are disfiguring and debilitating. Our dependence on facial muscles in verbal and non-verbal communication adds a dimension of social devastation above and beyond the physical impairments which result from their loss of function. Patients who have a facial dysfunction ask that clinicians address their functional and aesthetic concerns. The resulting pressure to create effective means of dealing with the many conditions affecting the face has led to an emphasis on clinical research. The focus of the research has developed in two directions. One clinical approach focuses on surgical management and reanimation of the paralyzed face (Adams, 1946; Anderl, 1973, 1979; Harii et al., 1976; O'Brien et al., 1980; Flanagan, 1984; Freilinger et al., 1987; Broniatowski et al., 1990, 1991; Dressier and Schonle, 1991; Rosenwasser et al., 1991; Hoffman, 1992; Ozcan et al., 1993). The other approach is to treat facial conditions with pharmacological agents (Laskawi et al., 1970; Frueh et al., 1984; Jankovic et al.,  11  1990; Marrning et al., 1990; Alderson et al., 1991; Harris et al., 1991; McLoon et al., 1991a; Taylor et al., 1991; Elston, 1992; Girlanda et al., 1992; Poewe et al., 1992; Flanders et al., 1993; Hassan et al., 1995).  Surgical reanimation Considering the wide range of pathalogic problems that can result in facial paralysis, there must be several ways of correcting them, with each case requiring special consideration before deterrnining the most viable option. Approaches include static methods and dynamic methods. The goal of static methods is to provide support and symmetry to the resting face, whereas dynamic methods strive to restore resting facial symmetry as well as some degree of symmetrical movement. Dynamic reanimation techniques offer the most desirable outcomes following surgical intervention. These methods include: primary nerve repair, nerve-to-nerve anastomosis, cross facial nerve grafting, neurovascular pedicle grafts, regional muscle transposition, neurovascular muscle transfer, and a combination of procedures depending on the circumstances of the case. In 1874, Drobnick attempted thefirstfacial reanimation when he anastomosed the distal part of the facial nerve with the proximal part of the spinal accessory nerve (Rosenwasser et al., 1991). Since that time, nerve-to-nerve anastomosis of the facial nerve has been done using the hypoglossal nerve, cervical plexus C3-4, phrenic nerve, ansa cervicalis, and deep temporal nerve. Of these techniques, the hypoglossal nerve is the most widely used because of its superior post-surgical results (Broniatowski et al., 1990,1991; Rosenwasser et al., 1991; Hoffman, 1992). During resection of an acoustic neuroma, there is often loss of ipsilateral facial nerve function (Hoffman, 1992). In such cases, restoration of the facial nerve is best accomplished by direct surgical anastomosis of its severed ends. In 1927, Bunnell reported the first successful primary nerve  12  graft of the facial nerve. By transposing the facial nerve out of the facial canal, he gained sufficient length to resect the neuroma and to re-approximate the facial nerve stumps. Scaramella (1971) devised a technique that allowed optimal symmetry by linking the paralyzed side to the unparalyzed side using a cross-facial nerve graft. The paralyzed face is symmetrically reanimated by anastomosing branches of the intact facial nerve to branches of the damaged facial nerve. A n autologous graft of the sural nerve is then taken from the leg in order to gain the length needed to span across to the damaged side of the face. Primary nerve grafting, nerve-to-nerve anastomosis, and cross-facial nerve grafting are techniques which can be helpful in reinnervating existing facial muscles, thereby restoring facial symmetry and function. However, these methods are most effective when injury to the facial nerve is recent, before severe muscular atrophy has occurred. In cases where long standing facial paralysis has existed, muscle transfers may be required. This procedure uses muscles from alternative areas to replace the atrophic or fibrotic local facial muscles. Local muscle-flap transfers and distant muscle-pedicle transfers have both been used to accomplish reanimation of long-term palsies (Rosenwasser et al., 1991; Hoffman, 1992). Local muscle-flap transfers have been done using pieces of the sternocleidomastoid, trapezius, frontalis, platysma, temporalis and masseter, with the temporalis and masseter being the most popular because they allow one end of the muscle flap to remain as a stable attachment site (Broniatowski et al., 1990, 1991; Rosenwasser et al., 1991; Hoffman, 1992). The masseter was first used in 1911 as a local muscle-flap transfer by Lexer and Eden (Rosenwasser et al., 1991). By leaving the zygomatic attachment intact and by splitting the anterior portion of the masseter, they were able to transfer the masseter to the lateral side of the oral commissure and attach it to the fascia of the upper  13  lip. Clenching the teeth activated the muscle and, with practice, patients could learn to elicit a reasonably dynamic and symmetrical smile (Rosenwasser et al., 1991). With the advent of microneurovascular tissue transfers came the possibility of successfully reanimating the face using innervated and vascularized distant-muscle pedicles. The first successful report of such a transfer was by Harii and coworkers (1976) after he used a portion of the gracilis muscle as the donor pedicle. O'Brien et al. (1980) used this technique to transplant the extensor digitorum brevis muscle. They found that the extensor digitorum transfers met with a 50% success rate whereas gracilis transplants were 100% successful. The variability in success between different muscles led some clinicians to postulate that inherent muscular differences made one muscle a better donor than the other (O'Brien, 1987; Hoffman, 1992). Other muscles which have also been used as donor sites are the serratus anterior, pectoralis minor and latissimus dorsi (Mackinnon and Dellon, 1988). Microneurovascular transfers in combination with cross-facial nerve grafting has become a popular method of reanimation. A study of 59 combined procedures by O'Brien (1980) reported 35 patients having full return of facial movement. Nineteen of these cases demonstrated independent function of the two sides of the face, despite both sides having a common motor nerve origin. As a primary method of dealing with facial paralysis, static methods have lost popularity. Fascia lata slips, Marlex mesh, and face lifts provide no facial motion and their optimum result is to give resting facial symmetry (Rosenwasser et al., 1991; Hoffman, 1992). However, as an adjunct to some of the dynamic techniques available, they can be extremely useful. Static methods can provide a firm suhrring base for muscle transfers and a supportive role during healing. They may also supplement final results achieved from more dynamic methods of reariimation (Rosenwasser et al., 1991; Hoffman, 1992). 14  Pharmacological therapy Facial dystonias are a group of neurological disorders characterized by either involuntary, repetitive or sustained muscle contractions which can result in facial twisting, squeezing and other abnormal movements (Jankovic and Fahn, 1988). Various types of facial dystonias are hemifacial spasm (Ushiro et al., 1993), blepharospasm(Girlanda et al., 1992), and strabismus (Harris et al., 1991; Girlanda et al., 1992). The precise causes of such dystonias are often unknown or idiopathic. Besides social embarrassment, severe cases can impair driving, reading and even unaccompanied walking (Jankovic and Fahn, 1988). Pharmacological treatments of these disorders were proposed as noninvasive alternatives to surgical management (McLoon et al., 1991; Elston, 1992). Two approaches to chemically controlling facial dystonias are possible. The first method involves the use of agents that block the nerve transmission to the affected muscles, whereas the second method uses agents that chemically destroy the muscles. Botulinum A toxin is a chemical that has gained considerable use as a nerve-blocking agent. It is the most potent biological toxin known to humans, and doses of about 2mg may result in a fatal case of botulism (Jankovic et al., 1990). Botulinum A toxin is one of eight immunologically distinct types of toxin produced by the spore-forming bacterium Clostridium botulinum (Calabresi et al., 1985). However, only toxins A, B and E have been linked to human botulism (Frueh et al., 1984). The toxin acts on the nervous system by disrupting the calcium ion influx at the neuromuscular junction. It irreversibly binds to the presynaptic cholinergic nerve terminal, effectively blocking acetylcholine release (Jankovic et al., 1990). In 1973, Scott injected botulinum A locally into the facial muscles of monkeys. Eight years later he reported its usefulness in the treatment of humans (Scott, 1980, 1981). Since then, several clinical trials have reported on the effectiveness of botulinum A injections in reducing clinical  15  symptoms of many local facial dystonias (Frueh et al., 1984; Jankovic et al., 1990; Manning et al., 1990; Taylor et al., 1991; Elston, 1992; Flanders et al., 1993). Only 7-16% (Jankovic et al., 1990; Taylor et al., 1991) of treatments were ineffective, and except for transient focal weakness, there seemed to be little to no systemic effects which would contraindicate its use as a means of treating facial dystonia (Jankovic et al., 1990; Taylor et al., 1991; Girlanda et al., 1992; Poewe et al., 1992; Flanders et al., 1993). The therapeutic muscle-weakening effects of botulinum A injections can last anywhere from 6 weeks (Manning et al., 1990) to a year (Jankovic et al., 1990). Though the effects of injections seem to be transient, repeated injections have been shown to have sustained benefits along with decreased side effects (Elston, 1992). The transient effects of botulinum toxin are causing some researchers to look for more permanent pharmacological solutions for treating facial dystonias (McLoon and Wirtschafter, 1988; McLoon et al., 1991). One such alternative is the use of the cytotoxin doxorubicin to chemically remove some of the affected muscle. Doxorubicin is an antibiotic isolated from the fungus Streptomyces peucetius (Calabresi et al., 1985). It acts by binding to D N A and R N A thereby preventing mitosis. Doxorubicin is currently used in humans as a chemotherapeutic drug. As an anticancer agent, systemic use of doxorubicin has many side effects (Calabresi et al., 1985). Long-term administration of doxorubicin is contraindicated because of its cardiotoxic effects. It may cause acute dysrhythmia and electrocardiographic abnormalities that are not dose dependent. Cumulative effects include cardiomyopathy that can lead to impulse conduction disturbances and acute congestive heart failure (Calabresi et al., 1985). In preliminary studies on primates, McLoon and coworkers (1988; 1991a) locally injected doxorubicin into facial muscles of cynomolgus monkeys. They reported minimal systemic side effects but found some of the more local effects that often occur at doxorubicin injection sites. Local effects 16  of doxorubicin include alopecia, local necrosis of skin, tendon and nerve, mucositis, and a benign local allergic reaction called '"Adriamycin flare" (Calabresi et al., 1985). In humans, the local necrosis occurring from the extravasation of doxorubicin can often be severe enough to require surgical debridement and skin grafting (Calabresi et al., 1985). The local and possible systemic effects of using such a powerful cytotoxic agent require a more precise treatment protocol and norther study before introducing it as a means of treating human focal dystonias. On the other hand, doxorubicin has been shown to be an effective means of selective chemomyectomy and would offer a more permanent solution to conditions such as blepharospasm and strabismus than botulinum therapy (McLoon and Wirtschafter, 1988; McLoon et al., 1991a).  MORPHOLOGICAL RESEARCH Light microscopy Early research into human facial muscle morphology compared enzyme histochemical features between limb muscles and, masticatory and other facial muscles. Schwarting et al. (1982) compared the platysma, levator labii superioris, zygomaticus major and orbicularis oris muscles and found them to be histologically similar to human limb muscles. These workers, however, also reported several interesting quantitative findings. They claimed that the fiber sizes and diameters of facial muscles were 50% less than those of limb muscles. On the basis of ATPase histochemical staining, they also reported that facial muscles had a predominance of type IIAfibersover type I fibers with a scarcity of type IIB fibers. Stal et. al. (1987) expanded upon this study by comparing zygomatic muscles with the masseter andfirstdorsal interosseus muscle. They commented on the lack of muscle spindles in these facial muscles and on the heterogeneity of the diameters of type I and type IIfibersin these facial muscles.  17  Soon after the realization that facial muscles were a morphologically unique subset of skeletal muscles, several studies were undertaken to test for histological and histochemical differences between individual facial muscles (Happak et al., 1988; Freilinger et al., 1990; Stal et al., 1990). Happak et. al. (1988) and Freilinger et. al. (1990) morphologically compared several of the mimic muscles. Happak and coworkers (1988) were the first to report no significant differences in muscle fiber size between males and females. However, using ATPase histochemistry, both studies were able to divide the various facial muscles into three groups based on percentages of type I muscle fibers. The Freilinger et al. study (1990), which compared fourteen of the mimic muscles, termed those with 14-15% type I fibers as "phasic" muscles, those with 28-37% type I fibers as "intermediate" muscles, and those with 41-67% type I fibers as "tonic" muscles. On the basis of ATPase staining, the phasic muscles were the procerus, orbicularis oculi and nasalis. The intermediate muscle group comprised the platysma, zygomaticus major, levator labii superioris, mentalis, levator anguli oris, depressor anguli oris and the orbicularis oris. The tonic group consisted of the corrugator supercilii, depressor labii inferioris, occipitofrontalis and buccinator. In addition to ATPase activity, the mean diameters of the type I and type II musclefiberswere compared. They also reported that, unlike the limb muscles, there was no significant difference in the diameters of type I and type II fibers in facial muscles (Freilinger et al., 1990). Stal et. al. (1990) conducted a study of two muscles which are situated around the oral commissure, the buccinator and the orbicularis oris. Again, using ATPase staining profiles as indicators, they reported that the buccinator was composed of 53% type I fibers while the orbicularis oris possessed 71% type II fibers. In their discussion, they postulated that the functional requirements of facial muscles may be of significant importance in determining their morphology, and that this unique function sets them apart morphologically from limb muscles and the muscles of mastication (Stal et al., 1987,1990; Stal, 1994). 18  Fiber-type distribution patterns were found to vary within different regions of some of the individual muscles of the face, as well as between different muscles of the face. Dittert and Bardosi (1989) studied the human platysma and found that type I fiber densities increased from the medial to the lateral parts of that muscle. In addition, they demonstrated a quantifiable difference in fiber diameters, noting that there was a trend for diameters to increase as one goes from medial to lateral across the muscle. This same regionalization, though not quantified, was noticed in the buccinator and the orbicularis oculi by others (Happak et al., 1988; Stal et al., 1990). In a study by McLoon and Wirtschafter (1991) that compared facial muscles of different species, regional variations infibertype distributions of the orbicularis oculi muscle were noted. Regional difference infiber-typedistribution patterns was a feature which was conserved between rabbits and cynomolgus monkeys. These workers found that the pretarsal portion of the muscle was composed of smallerfibers,more than 95% of which were type IIfibers,whereas the fibers of the preseptal regions were larger and only 77% of them were of the type II variety. Overall percentages of type II muscle fibers witJiin the palpebral portion of orbicularis oculi differed between species. In their account, McLoon and Wirtschafter (1991) reported that in the rabbit 77% offiberswere of the type II variety, whereas, in the monkey 91% were of the type II variety. A study by Stal et al. (1994) established that differences between human oro-facial, masticatory and limb muscles could be defined based on immunohistochemical and biochemical levels. This study applied a series of 10 different antibodies, raised against various myosin isoforms, tofrozensections of the zygomaticus major and minor, orbicularis oris, buccinator, masseter and first dorsal interosseus muscles. Each type of muscle was found to express its own distinct profile of myosin heavy chains (MHC) and myosin light chains (MLC). Individual fibers of the limb and facial muscles co-expressed one or two MHC's and up to five MLC's, whereas individual masseter fibers 19  co-expressed up to four different MHC's and eight different MLC's. Of special interest was the presence of a fourth fast-twitch band revealed by the non-denaturating pyrophosphate gels of orofacial muscles. They suggested that this may be a previously unidentified M H C which exists in orofacial muscles (termed "fast facial" or "fast F MHC"), but was not present in either the masseter or the first dorsal interosseus muscles (Stal et al., 1994). These findings reinforce the unique nature of human facial muscles and may also reflect the way these muscles have evolved to meet the specialized functional requirements imposed on them.  Electron microscopy Very little is known about the ultrastucture of normal human facial muscles. The few studies that have been undertaken focus more on describing the architectural changes occurring in pathologically affected facial muscles, and give rather cursory descriptions of the normal ulttastructural features of facial muscles. Belal (1982) was one of the first to perform a comparative ulturasturctural study of normal and denervated human facial muscle He studied the morphology of denervated auricularis posterior muscle following total facial paralysis ranging in duration from ten days to six years. Unfortunately, his descriptions of normal morphological features simply involved a recapitulation of characteristics associated with axial skeletal muscles. Belal also reported seeing normal muscle spindles in facial muscles, which to this point, were reported to be absent in facial muscles (Baum, 1899; Stal et al., 1987; McLoon and Wirtschafter, 1991). Belal did not supply any photographs to support his claim that muscle spindles exist in human facial muscle. The next study of the infrastructure of normal and denervated human facial muscle (Bardosi et al., 1987) examined the orbicularis oculi, zygomaticus major, levator labii superioris, and platysma muscles of 25 non-paralytic patients undergoing facial tumor surgery and 17 paralytic patients who  20  had suffered facial paralysis over a time period ranging from one month to 36 years. Again, the only fine structural description of the normal muscle was in a single sentence stating there was no difference between normal human facial muscles and any other mammalian and human striated muscle (Bardosi et al., 1987). Feher, (1977) in another study, detailed some of the ulfrastructural changes occurring in the orbicularis oculi muscle in malposition of the eyelid associated with senile involution. He discussed one of the effects as being either a focal widening or a streaming of the Z-lines of sarcomeres. Even though these effects appeared to be more pronounced in the pathologic samples, he noted that Z-line streaming was at times also present in normal orbicularis oculi muscle fibers. It is unclear how this observation was made since the study used 55 pathological samples and made no mention of how or if normal samples were procured.  21  Fig. 1.  Schematic diagram of some of the 21 pairs of human facial muscles. The corrugator supercilii and the orbicularis oculi are indicated. Modified from C D . Clemente, Anatomy  3rd edition, 1987.  22  Fig. 1  Corrugator Supercilii Orbicularis Oculi  23  Fig. 2.  Schematic diagram of the five extracranial branches (pes anserinus) of the facial nerve (cranial nerve VII). Modified from L. Wilson-Pauwels, E Akesson and P. Stewart  Cranial Nerves, 1988.  24  Fig. 2.  Temporal Zygomatic Posterior auricular Facial motor nucleus Buccal nerve to posterior belly of digastric muscle and stylohyoid muscle Mandibular Cervical  25  Fig. 3.  Drawing showing the bony attachment sights  of the  superciliary ridge (stippled), and the three portions (*) medial rim of the orbit (pinstriped).  26  corrugator  supercilii on the  of the orbicularis oculi on the  27  THESIS O B J E C T I V E S A N D R A T I O N A L E There is a wealth of information on the morphology, histochemistry and cytochemistry of axial skeletal muscles (Ogata and Murata, 1969; Camougis, 1970; Eriksson et al., 1980; Sarnat, 1983; Gauthier, 1988; Trotter, 1993; Rowlerson, 1994). The literature includes detailed information on their gross anatomy, innervation patterns, and ultrastructure (Uehara et al., 1976; Landon, 1982; Segal, 1992; Slater et al., 1992; Cross and Mercer, 1993). However, the unique subset of skeletal muscles contorting the skin and controlling the oral and orbital orifices of the face has, in comparison, been neglected. Further elucidation of the mimic system is necessary from both clinical and basic science perspectives. The purpose of this study, therefore, was to address the paucity of information regarding facial muscles, as well as to elucidate the links between their morphological features and their functions. This study postulates that human facial muscles with different functions posses morphological differences consistent with their functions. To test this hypothesis, a comparative morphological study of two human periorbital muscles, the palpebral part of the orbicularis oculi and the corrugator supercilii, was performed. The corrugator supercilii and the palpebral portion of the orbicularis oculi were chosen for several reasons. Embryologically, these two muscles develop from the deep layer of mesenchymal cells arising from the second branchial arch (Hazelton, 1969; Hamming, 1983). They are both innervated by the temporal branch of the facial nerve and are in topographical proximity to one another around the orbit (Patten, 1968; Leibgott, 1982). Functionally, however, the corrugator supercilii and orbicularis oculi are very different muscles. The corrugator supercilii is a muscle dedicated, almost exclusively, to producing some of  28  the many facial expressions one uses daily. For example, it is known to cause the sustainable contractions that result in vertical brow narrows (Hermsen and Dreyer, 1982; Patrinely and Anderson, 1988; Williams et al., 1989). Duchenne (de Boulogne), in his electrical study of physiognomy (1876), called the corrugator supercilii the muscle responsible for producing the primordial expressions of pain and terror (Cuthbertson, 1990). The orbicularis oculi is functionally more important because of its role as the major sphincteric muscle of the orbit. Its involvement in blinking is crucial to preserving vision by spreading tears over the surface of the eye to maintain a moist surface (Holly, 1980; Zide, 1981; Siegel, 1984; Patrinely and Anderson, 1988). Along with its obvious importance as a sphincteric muscle, the palpebral part of the orbicularis oculi has been assigned several expressive roles ranging from being the muscle of disdain and contempt to being the muscle of delirium and ecstasy (Duchenne, 1876). Considering the unique functional roles of these two muscles, it was first hypothesized that the corrugator supercilii would differ morphologically from the palpebral portion of the orbicularis oculi in its architecture, histochemistry and cytochemistry. To test this hypothesis, a morphological study was designed to compare these two periorbital muscles of facial expression. The experimental approach included: 1)  conventional  histological  and histochemical  light  microscopic  techniques  for  morphological evaluation 2)  immunocytochemical techniques as a tool for visualizing muscle fiber types expressing either fast or slow myosin isoforms to gain understanding about their fiber-type profiles  3)  transmission electron microscopy to compare the ultrastructural features of the two muscles and to, thereby, gain further details of their morphological character  29  On a more clinical level, these muscles are affected by disorders such as blepharospasm and hemifacial spasms (Ballance and Duel, 1932; Flanagan, 1984; Alderson et al., 1991; Broniatowski et al., 1991). Detailed knowledge of their innervation patterns and innervation zones is of great value to physicians who treat such conditions with botulinum A toxin injections. By locating innervation zones, injections can be made more site-specific, allowing doses to be reduced. It has been suggested that there is a correlation between a muscle's architectural and functional complexity and the complexity of its motor unit organization and firing patterns (Buchthal and Schmalbruch, 1980; Brown, 1984; Thomson et al., 1991; Scott et al., 1992). Early physiological studies to locate the innervation zone(s) of the orbicularis oculi (Borodic et al., 1991) showed a diffuse motor point region. This is consistent with what may be expected due to the complexity of the orbicularis oculi in both function and organization. When compared to the corrugator supercilii, the orbicularis oculi is the more complex of the two. The corrugator supercilii is a linearly arranged muscle that pulls the skin of the forehead in a relatively straight line. The orbicularis oculi, on the other hand, has gently arcing muscle fibers which pull the eyelid in a direction that is nearly perpendicular to the muscle fiber orientation (Williams et al., 1989). Based on these observations, it was therefore hypothesized that the orbicularis oculi and corrugator supercilii muscles would show distinctive differences in both their nerve and motor end-plate distribution patterns. Presumably, the complex architecture and unique function of the orbicularis oculi muscle should be reflected in a more complex pattern of innervation than that of the corrugator supercilii muscle. To test this hypothesis the following techniques were used: 1)  stereoscopic whole-muscle dissections of unembalmed cadaver corrugator supercilii and orbicularis oculi muscles to analyze their innervation patterns on a gross anatomical level  30  2)  motor end-plate staining, using hexazotized pararosaniline and indoxyl acetate (PIA), for microscopic visualization of motor end-plate structure and distribution patterns  3)  immunocytochemical methods, using an antibody to neurofilament protein, to examine innervation patterns as they relate to the fiber-type compositions of the two muscles To our knowledge, this is the first study to use both light and electron microscopy to  compare the normal morphological features of these two human periorbital muscles.  31  M A T E R I A L S  AND METHODS  This study included corrugator supercilii biopsies taken from patients undergoing cosmetic forehead lifts, and strips of the palpebral portion of the orbicularis oculi taken from persons having cosmetic upper blepharoplasties. None of the subjects were known to have either a disease or clinical condition which would have affected the histological and histochemical characteristics of the muscles. Most biopsies were extracted under general anesthesia, and three of the orbicularis oculi biopsies were removed under local anesthetic conditions. In addition to the surgical biopsies, four whole-muscle samples of each muscle were removed from unembalmed human cadavers. After excision, the biopsies and cadaveric tissues were immediately processed for light and/or electron microscopy.  Gross anatomical dissection Whole muscle samples were dissected from unembalmed human cadavers using standard dissecting tools and a 2X power head set magnifier. After in situ pictures were taken, the region of interest was removed. More detailed dissection was carried out under a Zeiss 5X stereoscopic dissecting microscope. Photographs were taken using a 35mm camera mounted on the stereoscope.  Conventional brightfield microscopy Eight corrugator supercilii biopsies and 14 orbicularis oculi biopsies were prepared for light microscopic evaluation. After excision, biopsies were pinned between thinly sliced pieces of calf liver on small wooden blocks to cryoprotect the samples and approximate resting lengths. Blocks were immediately plunge-frozen in isopentane cooled to -196 °C with liquid nitrogen and stored in a -60 C Forma Scientific Bio Freezer (Forma Sci. Ltd., Marietta, OH) until sectioned. 0  Specimens were cut into 10 um serial-sections in a Bright Instruments 5030 cryostat microtome  32  (Huntington, England) and stored at -20 °C until processed. Frozen sections were subsequently processed in one of three ways: 1) sections were stained with hematoxylin and eosin (H&E) for morphometric analysis of fiber size, shape, and general histological organization (Dubowitz and Brooke, 1973), 2) histochemical oxidative enzyme profiles of muscle fibers were visualized with nicotinamide adenine dinucleotide-tetrazolium reductase (NADH-tr) (Novikoff et al., 1961; Dubowitz and Brooke, 1973), 3) immunofluorescent methods were utilized to detect myosin isoform expression with antibodies to fast myosin (FM) and slow myosin (SM), and an antibody to neurofilament protein (NFP) was used to demonstrate innervation patterns and motor end-plates (Roden et al., 1991; Nahirney and Ovalle, 1992, 1993). In addition to frozen sections, light microscopic evaluation of whole muscle samples procured from unembalmed human cadavers was also undertaken. Cadaver material was taken from both sides of the body within 48 hours of death, and stained with an osmiophilic azoindoxyl complex that allowed further study of motor end-plate distribution patterns in the two periorbital muscles (Strum and Hall-Craggs, 1982). Epon-embedded semithin sections stained with 1% toluidine blue (Dubowitz and Brooke, 1973) were used to evaluate the ratio of capillary area to muscle area (capillary index) (Engel, 1986) of both the corrugator supercilii and the orbicularis oculi. Semithin sections, as opposed to frozen sections, were used because capillaries were optimally resolved and better preserved in the epon-embedded specimens.  33  Histochemistry The oxidative enzyme profiles of the orbicularis oculi and the corrugator supercilii were compared using 45-minute incubations of frozen sections with a conventional NADH-tr solution (Novikoff et al., 1961; Dubowitz and Brooke, 1973) at a pH of 7.4 (Sigma, St. Louis, MO). The insoluble formazan precipitate that forms when tetrazolium salts are reduced in this oxidoreductase reaction allows visualization of the cellular oxidative enzyme distribution patterns (Dubowitz and Brooke, 1973; Padykula, 1988). It has been shown that oxidative enzyme quantities are important because they indicate the oxidative capacity and mitochondrial content of muscle fibers (Padykula, 1988). This histochemical staining protocol was used as a method of fiber typing as well as for comparing how closely the oxidative enzyme profiles of muscle fibers related to their myosin isoform expression. The motor end-plate regions of the orbicularis oculi and the corrugator supercilii were visualized by the osmiophilic property of the reaction product formed when hexazotized pararosaniline and indoxyl acetate reacted with esterases found in the synaptic cleft of the motor end-plates. This procedure was first developed by Holt and Hicks (1966) for identifying esterases in the kidney, and was further refined by Strum and Hall-Craggs (1982) for staining the motor endplates of rat skeletal muscles. In the present study, the protocol was modified to account for the unique nature of the human facial muscles. Samples were excised and promptly fixed overnight in aqueous 2% parformaldehyde and 2% gluteraldehyde. The orbicularis oculi was sufficiently thin to be used whole, but because of its thickness, the corrugator supercilii was serial sectioned into 500 i\m slices using a 752M Campden Instruments Vibroslice. After slicing, all samples were stereoscopically dissected under a Zeiss stereoscope to remove excessive amounts of connective tissue surrounding the facial muscles. The stereoscopic dissection allowed observation of the  34  branching patterns of the nerves to each of the muscles and helped expose motor end-plates better. Samples were then incubated for one hour with hexazotized pararosaniline and indoxyl acetate at pH 6.0. After several washes in fresh, ice-cold citrate buffer, also at pH 6.0, branching patterns and motor end-plate distributions were examined under the Zeiss stereoscope. Composite drawings of samples showing branching patterns and motor end-plate distributions were done at that time. Some clusters of motor end-plates were further dissected from the muscles and subsequently viewed and photographed under higher power using a Leitz Orthoplan photomicroscope. To the best of my knowledge, this is the first time human motor end-plates have been visualized in this manner. To ensure the suitability of the pararosanaline staining protocol for human material, controls using fresh mouse extensor digitiorum longus (EDL) and gastrocnemius were carried out in parallel (Fig. 20).  Immunocytochemistry Immunofluorescent double-labeling was done on frozen transverse-sections warmed to room temperature and incubated with various antibodies in a humidifying chamber. Primary polyclonal antibody to rabbit neurofilament protein (Sigma, St. Louis, MO) and primary monoclonal antibodies to either fast myosin or slow myosin (Sigma, St. Louis, MO) were diluted in phosphate-buffered saline with bovine serum albumin (PBS/BSA) (pH 7.3) to 1:100, 1:1000 and 1:1000 respectively. Goat anti-rabbit conjugated with fluorescein isothiocyanate (FITC) at a concentration of 1:32 (Sigma, St. Louis, MO) was used as a secondary antibody for the neurofilament protein primary. Goat anti-mouse (1:64) was used as the secondary antibody for the fast and slow anti-myosin primaries. The anti-myosin secondary was in turn conjugated with the fluorescent marker, streptavidin-Texas red (Amersham, Oakville, ON). A l l antibodies were  35  commercial IgG antibodies purchased from Sigma Immuno Chemicals (St. Louis, MO). Frozen sections were first incubated with a primary antibody cocktail for 1 hr and washed in PBS/BSA buffer at a pH of 7.3. Secondary antibodies were individually applied for 1 hr each, with three washes in fresh buffer between incubations. Coverslips were mounted using a 1:1 glycerol/buffer mix, and slides were subsequently examined and photographed under the mercury bulb illumination of a Zeiss Axiophot fluorescence microscope. This technique was used to quantify muscle fiber-type distribution patterns and the cross sectional areas and maximum diameters of each fiber type. It also helped to further elucidate the morphology of the motor end-plates, nerve axons and their distribution patterns within the muscles.  To ensure the specificity of the  antibody stains control slides, lacking the primary antibody, were run in parallel. In addition, slides stained with antibodies to fast and slow myosin were compared to serial sections of the same slide stained with NADH-tr to test whether or not results of the two staining protocols were in agreement.  Morphometric analysis Sections of orbicularis oculi and corrugator supercilii, stained with either H & E or toluidine blue, were examined and photographed either under a Leitz Orthoplan or Zeiss Axiophot light microscope. Images were subsequently developed into 35mm negatives which were then printed as 8"X10" photographs. Peripheral traces of both the photographed muscle fibers and the capillaries were made on clear acetate sheets. From the tracings, measurements of cross-sectional areas and maximum diameters were made with a Zeiss MOP-3 digitizing table and stylus pen. To quantitatively compare the shapes of the orbicularis oculi fibers to those of the corrugator supercilii, photographic images of muscle fields in cross section were digitized using an  36  Epson ES 1200C flat-bed scanner and a Power Mac 7500 Macintosh computer. Due to their optimal presentation and resolution, only epon-embedded specimens were used. Though tissue shrinkage is known to be a result of fixation, it was assumed that shrinkage would be consistent for both muscles and would, therefore, not interfere with a proper comparison. Once images were digitized, they were imported into the software program Optimas (Bioscan Inc., Edmons WA). A macro was created in Optimas which determined the borders, cross sectional areas, and both minimum and maximum diameters of each muscle fiber. Using these values, a measurement of the rectangularity of each fiber was taken and the means were subsequently plotted as a histogram. Rectangularity was a unitless ratio which was calculated using an algorithm that takes each fiber's cross sectional area divided by the area of an enclosing box oriented along the major axis of the fiber. The minimum value of such an algorithm approaches zero for very narrow fibers, is 0.500 for square shaped fibers, and is -n/4 (0.785) for circular fibers.  Transmission electron microscopy For electron microscopy, seven corrugator and thirteen orbicularis oculi samples were pinned to dental wax at the time of biopsy to approximate their resting lengths. Muscles were immediately placed in phosphate free HEPES-buffered fixative containing 2% paraformaldehyde and 2% gluteraldehyde (Karnovsky, 1965) at pH 7.3 for 1-2 hr. depending on the length of surgery. After initial fixation, the muscle was minced into pieces less than lrnm and placed in fresh 2  fixative for another 3-4 hr. followed by 3 washes in fresh buffer. Samples were then postfixed in 1 % buffered OSO4 for 1 hr, en block stained with 1 % uranyl acetate, and dehydrated in an ascending ethanol series before embedding pieces in either epon resin or L R white.  37  To localize points of interest, semithin (1 pm) transverse and longitudinal sections were cut with glass knives and stained with 1% aqueous toluidine blue. Once regions of interest were located, thin sections of 60 to 100 nm were cut with a Dupont diamond knife and post-stained with Reynolds (1963) lead citrate and 1% uranyl acetate. Thin sections were placed on copper grids and examined with a Philips 301 transmission electron microscope (TEM) at 60 kV.  Statistical analysis The reliability of the data gathering method was tested to determine the level of experimental error caused by the tracing of muscle fiber outlines with the digitizing pad and stylus pen. Fifteen orbicularis oculi and fifteen corrugator supercilii muscle fibers were chosen from photographs of random cross sectional fields of muscle fibers. Fibers were grouped according to their cross sectional area, as either large, medium, or small. Each grouping contained five fibers and each fiber was traced ten times on three separate days. Both an intra-day A N O V A and an inter-day A N O V A were performed to test the consistency of tracing accuracy. The method was found to be accurate within acceptable statistical levels (p's<0.01). An A N O V A was performed on all data sets followed by Neuman-Keuls post hoc tests where necessary. The data were graphed as histograms and expressed as means of the measure. All error bars are expressed as the standard error of the associated mean.  38  OBSERVATIONS MACROSCOPIC ARCHITECTURE Gross anatomical dissections of muscles of facial expression have been carried out for centuries (Bell, 1821; Cruveilhier, 1844; Virchow, 1908; Lind, 1949). The attachments, pennations and fiber-orientations of these muscles are clearly documented in numerous textbooks of anatomy (Crouch, 1978; Leibgott, 1982; Williams et al., 1989). However, a closer evaluation of their macroscopic  innervation patterns and fascicular organization can lead both to a better  understanding of the functional roles of individual facial muscles and to a clearer view of how they perform their functions. In the present study, stereoscopic dissections of fresh cadaver muscle revealed several architectural differences between the corrugator supercilii and the orbicularis oculi (Figs. 4-5). The palpebral portion of the orbicularis oculi (Fig. 4) is a thin sheet of muscle that sweeps across the palpebral fissure superficial to the pretarsal fascia and to the tarsal plate (Furnas, 1981; Siegel, 1984; Patrinely and Anderson, 1988; Siegel, 1993). The muscular sheet is organized into small fascicles that span between the medial and lateral canthal tendons. As the muscle fibers extend across the orbital fissure they arc slightly upward before arcing back toward the lateral canthal tendon. Figure 4c shows a posterior view of the lateral palpebral musclefiberswhich can be seen arcing into the lateral canthal region as the orbital fibers continue upwards around the orbit. Most muscle fibers appear to remain in their respective fascicles while spanning the palpebral fissure. However, some of the fibers in the more inferior fascicles exit from the perimysium into the subcutaneous connective tissue in the rim of the palpebrum to anchor into it. The palpebral fascicles, themselves, are somewhat tear-shaped and appear to be angled at approximately 30° from  39  vertical so that their inferior edge is deeper than their superior edge. This angulation results in the superior fascicle overlapping the fascicle below it much like the shingles of a roof (Fig. 4d). The temporal branch of the facial nerve gives off one or more branches that supply motor innervation to the palpebral fibers of the orbicularis oculi (Ruskell, 1985; Vestal et al., 1994). These palpebral nerve branches travel in a loose, fatty connective tissue layer under the lateral side of the orbicularis just superior to the lateral canthal tendon (Fig. 4c). Above the lateral canthal tendon, the palpebral nerve(s) travels along the superior edge of the palpebral portion of the orbicularis oculi superficial to, and at the same level as, the top of the tarsal plate. It gives off primary neuronal branches at intervals of 0.5-1.0 cm that gently arc downward and perpendicular to the orientation of the muscle fibers (Fig. 4a). The primary nerve branches give rise to secondary nerve branches that run parallel to the muscle fascicles. The secondary branches then travel between the overlapping muscle fascicles in the perimysial connective tissue (Fig. 4c). Smaller tertiary nerve branches emanate from the secondary nerve branches and course into the muscle fascicles before becoming too small to observe clearly with a dissecting microscope. The smallest branches disappear into motor end-plate regions at scattered points along the length of the orbicularis. These regions will be discussed later in greater detail. The corrugator supercilii has a relatively simple macroscopic organization compared to the orbicularis oculi (Fig. 5). The muscle itself is situated at an oblique angle originating from the superciliary ridge medially and blending distally and laterally into the orbicularis oculi and frontalis muscles which are superficial to it (Pitanguy, 1979; Williams et al., 1989). A n incomplete layer of connective tissue is interposed between the fibers of the corrugator and the two muscles above it. Openings in this connective tissue layer serve as windows through which bundles of corrugator muscle fibers pass as they blend into the orbital portion of the orbicularis oculi and the frontalis  40  muscles (Fig. 5a). Unlike the orbicularis, fibers of the corrugator are packed into large fascicles that run parallel to one another with no obvious pattern of overlap (Fig. 5c). However, some of the most superficial muscle fibers appear to have a small rotational component. These "rotated" muscle fibers appear to be oriented in a slightly more vertical plane than the deeper muscle fibers. Motor innervation to the corrugator supercilii is supplied by the temporal branch of the facial nerve (Ruskell, 1985; Malone and Maisel, 1988; Williams et al., 1989). The corrugator component of the temporal branch travels in a fatty connective tissue layer under the upper portion of the orbicularis. It enters the corrugator from a superficial position in the lateral portion of the muscle (Fig. 5b). The nerve runs obliquely across the muscle in an infero-medial direction. It gives rise to primary nerve branches that tend to run perpendicular to the muscle fiber orientation, though there is more variability to this than is seen in the orbicularis. The primary nerve branches can be seen coursing within muscle fascicles or within cords of loose connective tissue that are scattered throughout the muscle. The secondary nerve branches arising from the primary branches are typically found running parallel to the muscle fibers (Fig. 5c) before disappearing into motor endplate regions. These motor end-plate regions will be discussed in more detail below.  MICROSCOPIC ARCHITECTURE When viewed by light microscopy, the corrugator supercilii and orbicularis oculi muscles show many features common to each other, and to other muscles of facial expression. Conventional H & E sections of both muscles reveal that they are composed of multinucleated muscle fibers with peripherally located nuclei. In addition, tendinous insertions and deep fascial coverings are discrete (Happak et al., 1988; Williams et al., 1989). In this study, encapsulated sensory receptors such as muscle spindles (Kennedy, 1970; Sahgal et al., 1985; Barker and Banks, 1986; Ovalle and Dow,  41  1986; Stal et al., 1987; Gauthier, 1988) and tendon organs (Stuart et al., 1972; Ovalle and Dow, 1983), as they are typically described in other muscles, were not seen in either the corrugator supercilii or the orbicularis oculi. Despite these similar features, the two muscles could be further distinguished by their microscopic architecture and their histochemical staining patterns.  Fiber sizes and shapes Initial observations of the two muscles showed that there were distinctive variations in the shapes of the muscle fibers and fascicles. The corrugator supercilii is composed of pleomorphic muscle fibers that are organized into large, tightly-packed fascicles with little intervening endomysial and perimysial connective tissue (Fig. 6a). In contrast, the orbicularis oculi is characterized by smaller muscle fascicles. The fascicles are composed of loosely-packed and rounded muscle fibers with substantial amounts of endomysial and perimysial connective tissue between muscle fibers and fascicles (Fig. 6b). Differences in muscle fiber shape have previously either been reported in qualitative terms (Engel, 1986; Stal et al., 1990; Gans and Gaunt, 1991), or they have not been reported, possibly due to the difficulty in quantifying such difference. To determine whether or not the muscle fibers of the orbicularis were quantifiably rounder than those in the corrugator, ten muscle fields of semi-thin sections stained with toluidine blue were scanned into a Power Macintosh 7500 computer with an Epson ES 1200C flatbed scanner. Images were then analyzed using the Optimas imaging data analysis program (Bioscan, Edmonds WA). Using an algorithm within the Optimas program, relative rectangularity of the fibers was measured. Rectangularity, defined in the Methods section under the subheading Morphometric Analysis, is a unitless value calculated by dividing the area of a muscle fiber by the area of a surrounding box. This unitless ratio was used to compare the shapes  42  of muscle fibers in the corrugator supercilii with those in the orbicularis oculi. Rectangularity ratios approaching a value of 0.500 are representative of square-shaped fibers, whereas values approaching 0.785 (71 divided by 4) indicates rounded fibers (Optimas rectangularity menu, Bioscan). A mean fiber rectangularity value for each muscle was calculated from the rectangularity values of each scanned fiber in the two muscles. A total of 281 muscle fibers were measured in the orbicularis oculi, whereas 237 fibers were measured in the corrugator supercilii. Figure 7 summarizes these results and shows that the orbicularis oculi has rounder fibers, at a mean of 0.698, than the corrugator supercilii, at a mean of 0.687 (P < 0.02). Microscopic evaluation of H & E sections also revealed distinctive differences in muscle fiber sizes. Both facial muscles had muscle fibers that were relatively small compared to previously published reports of other axial skeletal muscles (Brooke and Engel, 1969; Edstrom and Torlegard, 1969). This is similar to reports that have compared other facial muscles with axial skeletal muscles or muscles of mastication (Stal et al., 1987, 1990; Happak et al., 1988; Dittert and Bardosi, 1989). The muscle fibers in the corrugator were fairly uniform in size, whereas fibers in the orbicularis exhibited more variation in size, ranging from very small to quite large (Fig. 6). To quantify these differences in size, transverse fields of fibers were digitized, and maximum diameter and area measurements were gathered and analyzed from each digitized fiber. Results showed that muscle fibers from the corrugator had a mean cross sectional area of 802 pm ( ± 1 8 pm ) and a maximum 2  2  diameter of 34 pm ( ± 0 . 5 pm), whereas the fibers in the orbicularis had a mean cross sectional area of 624 um .(± 26 pm ) and a maximum diameter of 28 pm ( ± 0.6 pm) (Figs. 10-11). To further 2  2  elucidate these size differences, fibers were categorized by fiber type (either Type I or Type II)  43  using immunocytochemical staining techniques. The mean cross sectional areas and maximum diameters of these fiber types were then compared (Figs. 10-11).  Fiber typing Frozen transverse-sections  of the corrugator supercilii and orbicularis oculi were  immunolabelled with either slow or fast myosin antibody (Sigma Irnmuno Chemicals, St. Louis, MO) tagged with a fluorescent Texas Red marker (Amersham, Oakville, ON) (Fig. 8). Antibodies to fast and slow heavy chain myosin isoforms were used to verify the specificity of each stain. Quantitative analysis of sections included fiber type distributions in the two muscles (Fig. 9) and size measurements of each fiber type population (Figs. 10-11). Fiber-type distribution patterns were markedly different in the corrugator than those in the orbicularis (Fig. 8). The corrugator was composed of a heterogeneous mixture of type II (fasttwitch) fibers and type I (slow-twitch) fibers (49% type II, 5 1 % type I). The orbicularis, on the other hand, was composed primarily of type II fibers with only a few type I fibers scattered throughout the muscle fascicles (89% type II, 11% type I) (Fig. 9). This distribution pattern was similar to what was reported earlier by Freilinger et al. (1990). These workers used ATPase histochemistry to fiber type fourteen human mimic muscles, and reported that the corrugator had over 4 0 % type I fibers while the orbicularis had only 13% type I fibers. Once the fiber type populations were separated, morphometric size measurements were taken, and graphic representation of these size differences is shown in Figures 10 and 11. The histogram in Figure 10 shows that the mean cross sectional area of corrugator type II fibers (909 um ± 3 1 nm ) was larger than that of type T fibers (738 um ± 22 um ). It also shows that 2  2  2  2  orbicularis type II fibers (688 um ± 38 um ) possessed a larger mean cross sectional area than 2  2  44  orbicularis type I fibers (542 um ± 33 urn ). When comparing the fiber sizes between the 2  2  corrugator and orbicularis muscles, the data not only support the earlier finding that corrugator fibers are, in general, significantly larger than orbicularis fibers but they also show that type I corrugator fibers are significantly larger than orbicularis type II fibers (p<0.02) (Figs. 10-11). This is in contrast to the Freilinger et al. study (1990) which reported no significant size difference between different fiber types of human facial muscles.  Mitochondrial staining profiles Histochemical staining of frozen serial-sections of the corrugator and orbicularis was carried out in parallel with the immunocytochemical staining. Mitochondrial enzyme profiles in the muscle fibers were visualized and compared by using the NADH-tr staining reaction described earlier in the Methods section. The usefulness of this staining method was two-fold. First, it permitted comparative examination of mitochondrial staining densities (Fig. 12), and second, it provided a method of assessing how closely oxidative enzyme profiles compared with myosin heavy-chain expression (Figs. 16-17). Oxidative enzyme staining of the corrugator supercilii muscle differed from that of the orbicularis oculi in two respects. In the corrugator there was: 1. an almost even distribution of dark-staining and pale-staining muscle fibers, and 2. regardless of a muscle fiber's staining intensity, mitochondria appeared to be evenly distributed throughout each muscle fiber, giving the appearance of a uniform and punctate profile in most muscle fibers (Fig. 12a).  45  In contrast, the orbicularis oculi was composed predominantly of pale-staining muscle fibers with a paucity of mitochondria which were often clustered in groups around the periphery of the fiber giving it a granular-staining profile (Fig. 12b). Since NADH-tr is a staining method which has often been used in fiber typing (Dubowitz and Brooke, 1973; Fawcett, 1986; Padykula, 1988), it was compared with the immunolabelled slow myosin sections that this study used for fiber typing, in order to see how closely the two methods matched one another (Figs. 16-17). Upon examination, it was apparent that the bright, red-staining type I fibers in the immunoflourescent sections were also the dark, mitochondria-dense fibers in the NADH-tr stained sections (Fig. 16b, d). Though it was more commonly seen in the orbicularis, both the corrugator and the orbicularis possessed some fibers that stained darkly for NADH-tr, but did not express the slow myosin heavy chain isoform (Fig. 17b, d).  Capillary index The high levels of mitochondrial enzyme (NADH-tr) staining in the corrugator relative to the orbicularis, coupled with some early indications of vascular differences seen in H & E sections (Fig. 6), led to a comparison of capillary content in the two facial muscles. The superior preservation of morphological detail in the semi-thin transverse sections made these sections ideal for such a comparison. Microscopic examination of these semi-thin sections revealed that the corrugator supercilii contained an abundance of capillaries compared to the orbicularis oculi (Fig. 13). Digitization of six representative random fields of each muscle was carried out using the MOP 3 Image Analyzer. Measurements of the capillary area and the muscle fiber area in each field were taken. The total area of the capillaries in the six orbicularis fields was divided by the total area  46  of muscle fibers in the same six fields to calculate a unitless value, called the capillary index. The same procedure was performed on the six corrugator fields. The resulting capillary index values compare the units of contractile area to units of capillary area, while simultaneously discounting any differences in connective tissue or nervous tissue quantities that may have existed between the orbicularis and the corrugator. Analysis revealed that the corrugator, with a capillary index value of 0.019 ± 4 % , is 2.4 times better supplied with capillaries than the orbicularis, with a capillary index value of 0.008 ± 3 % (Fig. 14).  Innervation patterns Macroscopic nerve branching patterns in the two facial muscles were visualized by stereoscopic dissection of cadaver material (Figs. 4-5). Further detail of their innervation was gained by subsequent light microscopic study. Frozen sections of biopsied muscles were immunolabelled with FITC-tagged antibody to neurofilament protein (Sigma, St. Louis, MO), a 200 kD member of the intermediate filament family (Nahirney and Ovalle, 1992, 1993). Sections from both muscles consistently showed strong reactivity in nerve fibers and at neuromuscular junctions (Fig. 15). The corrugator supercilii typically had numerous intramuscular nerve fascicles that traveled within the muscle fascicles. These intramuscular nerves exhibited a wide range of sizes (Fig. 15a). On the other hand, nerve fascicles associated with the orbicularis tended to travel in the perimysial connective tissue between muscle fascicles rather than within the muscle fascicles themselves. These intramuscular nerves traveled in the connective tissue, sending off smaller and smaller branches into the muscle fascicles (Fig. 15b). These small secondary and tertiary nerves would then branch into the muscle fascicles close to their motor end-plate regions where they  47  would terminate as clusters of motor end-plates, most of which appeared to have several terminal boutons (Fig. 15). Concomitant immunolabelling with either fast or slow myosin antibodies tagged with Texas red (Sigma, St. Louis, MO) (Figs 16-17c, d) demonstrated that the motor end-plates within a cluster were all terminating on fibers of the same type (either type I or type II). This could be seen with double labeling, and was clearly visualized in serial sections of the two muscles stained with the various histological and immunohistochemical stains used in this study (Figs. 16-17). Motor end-plate regions were also visualized in fresh cadaver whole-muscle samples using the pararosanaline and indoxyl acetate staining technique described earlier (Strum and HallCraggs, 1982). Composite drawings of the distribution patterns of the motor end-plates showed distinct pattern differences between the corrugator and the orbicularis (Figs. 18-19a). The palpebral portion of the orbicularis showed clusters of motor end-plates scattered along the muscle in a diffuse manner (Fig. 18a). Clusters usually contained 4-30 motor end-plates (Fig. 18b). A higher concentration of motor end-plate clusters were located along the lateral portions of the muscle with fewer and smaller clusters tending towards the medial side (Fig. 18a). The motor end-plate regions of the corrugator supercilii were also organized in a diffuse pattern (Fig. 19a). The clusters were typically smaller than those of the orbicularis with 4-20 motor end-plates in a group (Fig. 19b). Even though clusters were scattered throughout the corrugator, there appeared to be higher concentrations of clusters organized in an " H " pattern that ran across the transverse plane of the muscle (Fig. 19b). When examined by light microscopy at higher magnification, the motor end-plates of the orbicularis and the corrugator appeared very similar. The motor end-plates of the two human periorbital muscles were rounded and exhibited deep and convoluted synaptic clefts. From a cross48  sectional perspective, they appeared as in Figure 18c, and from a superior perspective, they appeared more rounded as in Figure 19c. For morphological comparison, fresh mouse E D L and gastrocnemius muscles were also stained with PIA as a control. Murine motor end-plates visualized by this method were similar in appearance to those reported by Ogata (1965) who demonstrated murine motor end-plates with succinic dehydrogenase and cholinesterase. Mouse motor end-plates in the E D L and in the gastrocnemius (Fig. 20) did not exhibit the same degree of synaptic infolding as did the motor end-plates seen in the two human periorbital muscles. Moreover, motor end-plates in the mouse did not appear as compact as those in the human (Fig. 18-20).  ELECTRON MICROSCOPY Due to the paucity of electron microscopic (EM) data on normal human facial muscles, the goal of the E M portion of this study was to confirm features observed at the light microscopic level and expand our understanding of their normal ulfrastructure.  Fine structure The orbicularis oculi and the corrugator supercilii showed many ultrastructural features similar to those previously reported in human axial skeletal muscle elsewhere in the body (Murata and Ogata, 1969, 1988; Ovalle, 1987; Gauthier, 1988; Cross and Mercer, 1993). The components of each sarcomere including the two sets of myofilaments, Z-lines, I-bands, A-bands, and M-lines were readily identified (Fig. 21). In longitudinal sections of the two periorbital muscles (Fig. 21a) distinctive H-bands and well developed M-lines were present. Preliminary measurements of myofilament lengths are within previously reported normal limits for mammalian muscle (Page and Huxley, 1963; Dubowitz and Brooke, 1973; Uehara et al., 1976; Eisenberg, 1983; Craig, 1986; Gauthier, 1988) with myosin filament lengths of approximately 1.5-1.6 um and actin filament  49  lengths of approximately 1.0 pm. Transverse sections of the two periorbital muscles (Fig. 21b) also demonstrated myofilament arrangements that were in keeping with previous reports of mammalian skeletal muscle found elsewhere (Murata and Ogata, 1969; Pepe, 1971; Carlson and Wilkie, 1974; Craig, 1986; Gauthier, 1988; Ogata, 1988). At the junction between the A-band and I-band, six thin (actin) filaments could be seen around each thick (myosin) filament. Each thick filament was in turn surrounded by six other thick filaments. This pattern is commonly called the "double hexagonal array" (Fig. 21b) (Dubowitz and Brooke, 1973; Landon, 1982; Craig, 1986). The muscle fibers in the corrugator were characterized by a poorly developed sarcotubular system and contained numerous mitochondria (Fig. 22b). As is typical of mammalian skeletal muscle (Ogata and Yamasaki, 1985; Ogata, 1988; Cross and Mercer, 1993), the mitochondria of the corrugator were often arranged in doublets that straddled the Z-lines of each sarcomere (Fig. 22a). In transverse section, myofibrillar patterns were often disrupted by mitochondrial aggregates (Fig. 22c). In addition to the doublets, mitochondria were regularly seen in subsarcolemmal clusters (Fig. 22c) and in myofibrillar chains (Fig. 22d). As was the case with the corrugator, transverse sections of fibers of the orbicularis showed mitochondrial doublets at the Z-lines (Fig. 23a). However, fibers of the orbicularis typically had a better developed sarcotubular system (Fig. 23b, c). In longitudinal sections, triads were identifiable at the A-I junctions and could be seen regularly due to the consistency of the undisturbed myofibrillar patterns (Fig. 23a). The Z-lines of the orbicularis were thin, relative to the corrugator, and mitochondria surrounding the Z-lines as doublets often appeared smaller than those seen in the corrugator. Moreover, there was a paucity of mitochondria in most fibers of the orbicularis. Occasionally, however, there were small clusters of mitochondria that were seen in subsarcolemmal  50  pockets (Fig. 23d), but the Z-lines were the most common distribution site of the mitochondria in fibers of the orbicularis.  Mitochondria Besides the corrugator having fibers with many more mitochondria than those in the orbicularis,, the shapes and sizes of these mitochondria also appeared to be different. Indeed the variance in the distribution, sizes and shapes of mitochondria were some of the most notable ultrastructural differences between the orbicularis and the corrugator. Mitochondria at the Z-lines were rounded in both the corrugator and orbicularis, but they tended to be larger in the corrugator than in the orbicularis. The occasional pockets of subsarcolemmal mitochondria seen in fibers of the orbicularis were usually rounded (Fig. 23d), whereas the mitochondria of the corrugator were often either rounded, pretzel-shaped or elongated (Fig. 22c). It is important to note that branching mitochondria and interfibrillar chains of mitochondria were a feature of both muscles, but they were more commonly seen in the corrugator than in the orbicularis (Fig. 22d).  Capillaries As was previously mentioned in the light microscopic portion of this study, the corrugator had a 2.4 fold greater capillary index than the orbicularis. Not surprisingly, this difference was observed ultrastructurally as well. Capillaries of the orbicularis oculifrequentlyappeared to be located further away from the muscle fibers than the capillaries of the corrugator supercilii. Corrugator fibers were often seen surrounded by 1-4 capillaries which were distributed closely around the periphery of the fibers, whereas the orbicularis fibers were more often seen either without an associated capillary or with one capillary separated from the fiber by intervening connective tissue (Fig. 13).  51  Continuous (or tight) capillaries such as the one shown in Figure 24a were the most common type of capillary observed in both muscles. The endothelium of these capillaries was often filled with numerous pinocytotic vesicles. This feature is thought to be a sign of high levels of cellular activity involving exchange between endothelia and surrounding tissues (Williams et al., 1989; Hudlicka et al., 1992). Besides the typical tight capillaries, some fenestrated (or discontinuous) capillaries were observed in the orbicularis (Fig. 24b).  Motor end-plates At the light microscopic level, the morphological architecture of the motor end-plates of the two periorbital muscles appeared structurally similar (Figs. 18c-19c). At the ultrastiiictural level, however, differences in motor end-plate morphology were more apparent (Fig. 25). As seen by electron microscopy, both muscles possessed small motor end-plates with complicated synaptic infolds. However, the synaptic clefts of the orbicularis oculi appeared to be more complex than those of the corrugator supercilii. Primary synaptic clefts were deeper, and secondary synaptic clefts were longer and exhibited more extensive branching in the orbicularis (Fig. 25b inset). In contrast, the secondary synaptic clefts in motor end-plates of the corrugator were shorter, wider, and appeared to have little to no branching (Fig. 25a inset). Other features Various cellular components of connective tissue could be seen in the perimysium of both the orbicularis and the corrugator. Mast cells were sometimes seen in close proximity to capillaries and muscle fibers. They were filled with numerous, electron-dense cytoplasmic granules which have long been known to be filled with histamine, heparin and various other substances important in the inflammatory response (Fig. 26a) (Goldberg and Rabinovitch, 1988; Cross and Mercer,  52  1993). At times, connective tissue macrophages with multiple pseudopodia and pinocytotoic vesicles were also observed (Fig. 26b). Besides these few, expected transient cellular residents, there were other more unexplainable observations. The lamellar-like sarcoplasmic inclusions pictured in Figure 26c were seen with surprising regularity in muscle fibers of both the corrugator and the orbicularis. They showed great variability in size with some of these inclusions nearing the size of nearby myonuclei. Their lamellar spacing was irregular with some of the lamellae appearing to fuse together at points only to separate again. More often than not they were located under the sarcolemma within a region of cytoplasmic swelling, but they were also commonly seen in perinuclear regions. Lamellar-like bodies similar to these have been reported previously by Ovalle and Dow (1986) who suggested that they may be remnants of degenerating mitochondria. Others have suggested that they are lysosomal remnants found in pathologically affected muscle cells that have had their ability to eliminate accumulating degradation products compromised (Schiaffmo et al., 1977; Cullen and Mastaglia, 1982; Engel and Banker, 1986). As was mentioned previously, typical muscle spindles were not seen in either the corrugator supercilii or the orbicularis oculi. However, the toluidine-blue stained semi-thin section of the orbicularis, shown in Figure 26d, demonstrates several spindle-like features. A myelin covered peripheral nerve fascicle wraps around the perimeter of two muscle fibers appearing to completely separate it from neighboring extrafusal fibers. One of the sequestered muscle fibers has a diameter of approximately 42 um, while the other is substantially smaller having a diameter of approximately 22 um. The two sequestered muscle fibers are rounded in shape with a slightly flattened area where they come in close contact with each other. The surrounding nerve fascicle is  53  in close proximity to both muscle fibers and appears to be spiraling around their longitudinal axis. The spindle-like structure, however, does not have an outer and inner connective tissue capsule with an expanded periaxial space intervening between them as has been reported in muscle spindles elsewhere (Barker, 1974; Ovalle and Dow, 1983,  1985; Sahgal et al., 1985). However, the  perineurium of the surrounding peripheral nerve fascicle does seem to exhibit complete inner and outer portions which may serve as a perineurial capsular equivalent. Some of the possible functional considerations of such a structure will be detailed further in the discussion.  54  Fig. 4.  Idealized drawing of the palpebral portion of the orbicularis oculi showing temporal nerve branching pattern and muscle fascicle orientation. Inset pictures a to d are from regions indicated on the drawing. a) Posterior view of a stereoscopically dissected palpebral portion of orbicularis oculi. Arrows indicate primary nerve branches from the palpebral portion of the temporal branch of cranial nerve VII. M=Medial, S=superior. b) Small artery (A) and primary palpebral nerves (N) running perpendicular to muscle fibers giving off secondary nerve branches (pin tips) that run parallel to the muscle fibers. c) Postero-lateral side of palpebral muscle fibers (P) joining into orbital muscle fibers (O) deep to the lateral canthal tendon. Temporal nerve branch (N) and associated artery (A) are also shown. d) Stereoscopic view of the palpebral muscle fascicles teased apart to show their overlapping organization.  55  Fig. 5.  Idealized drawing of the corrugator supercilii showing temporal nerve branching pattern and muscle orientation. Inset pictures a to d are from regions indicated on the drawing. a) Stereoscopic dissection showing a bundle of corrugator muscle fibers (B) exiting from a window in the connective tissue (CT) to blend into the orbital portion of the orbicularis oculi andfrontalismuscles that have already been removed. b) Picture of dissected corrugator muscle with the corrugator portion of the temporal branch of the facial nerve (arrow) entering from the supero-lateral side (S=superior, L=Lateral) The orbicularis oculi is reflected forward (OO). c) Teased apart muscle fascicles of the corrugator. Notice that they do not exhibit a special overlapping pattern such as the one seen in the orbicularis (see Fig. 4d). d) Primary nerve (N) giving rise to a secondary nerve (arrow) parallel to muscle fibers.  57  Fig. 6.  Transverse frozen sections of portions of the corrugator supercilii (a) and orbicularis oculi (b) stained with H & E . The corrugator fibers are tightly packed and pleomorphic in shape with little intervening connective tissue. In the orbicularis, fibers are rounded in shape with large variations in fiber size and increased amounts of connective tissue. X460.  59  Fig. 6.  60  Fig. 7.  Histogram showing the mean shape of muscle fibers within each muscle by using an algorithm that measures relative rectangularity of the fibers. To the right of the graph the inset diagram indicates that fibers which are round in shape will have a rectangularity value that approaches the value of a perfect circle (0.785). Fibers that tend to be square in shape will have a rectangularity value that approach the value of a perfect square (0.500). The graph shows that the mean rectangularity value of the orbicularis is closer to the 0.785 value than the mean of the corrugator, indicating that the orbicularis is composed of rounder fibers than the corrugator (p < 0.05).  When rectangularity is defined as: Rectangularity = Area of muscle fiber Area of surrounding box  61  Fig. 7.  Shape of Muscle Fibers  0.72  0.62  CS  62  oo  Fig. 8.  Transverse sections of corrugator supercilii (a) and orbicularis oculi (b) stained with fluorescent labeled anti-fast myosin. One red-staining fast-twitch type II fiber (II) and one unstained slow-twitch type I fiber (I) are indicated in each micrograph. A reverse staining pattern occurs when anti-slow myosin is used. X460.  63  64  Fig. 9.  Stacked bar graph showing fiber type distribution as a percentage of type I and type II fibers counted in the two muscles. In toto 1227 fibers of the orbicularis and 1307 fibers of the corrugator were counted. The corrugator was composed of 49% type II fibers and 51% type I fibers, whereas the orbicularis was composed of 89% type II fibers and only 11% type I fibers.  65  9. Fiber Type Distribution  66  Fig. 10. Histogram graphically demonstrating the mean cross sectional area of type I and type II fibers in the two muscles. In the corrugator, type I fibers had a mean area of 738 um (±22 um ) and type II fibers had a mean area of 909 um (±31 um ). Type I and type II fibers were smaller in the orbicularis with mean areas of 542 um (±33 um ) and 688 um (±38 um ), respectively. All values were significant when a Newman-Keuls post hoc test was performed (p<0.02). 2  2  2  2  2  2  67  2  2  g. 10.  Fiber Type Area  68  Fig. 11. Histogram demonstrating the mean maximum diameters of type I and type II fibers in the two muscles. In the corrugator, the mean maximum diameter of type I fibers was 32 pm (+0.5 pm) while that of type II fibers was 38 pm (±0.8 pm). In the orbicularis, type I fibers had a mean maximum diameter of 27 pm (±0.8 pm) while type II fibers had a mean diameter of 29 pm (±0.9 pm). With Newman-Keuls post hoc analysis all values were significant (p<0.02).  69  11. Fiber Type Maximum Diameter  70  Fig. 12. Transverse sections of portions of the corrugator supercilii (a) and orbicularis oculi (b) stained with NADH-tr to demonstrate mitochondrial enzyme profiles. The corrugator has many mitochondria-rich (type I) fibers while the orbicularis is predominantly composed of mitochondria-poor (type II) fibers. This suggests that the corrugator has a higher oxidative capacity than the orbicularis. X460.  71  72  Fig. 13. Semithin sections of portions of the corrugator supercilii (a) and orbicularis oculi (b) stained with toluidine blue. The corrugator has an abundance of capillaries (arrows) compared to the orbicularis. Furthermore, capillaries in the corrugator often appear to be located closer to the muscle fibers than those in the orbicularis. X460.  73  74  Fig. 14. Graphic comparison of the capillary index of the two muscles reveals that the corrugator, index (0.019) is approximately 2.4 fold larger than the orbicularis index (0.008).  Where capillary index is defined as: capillary index = capillary area muscle area  75  Fig. 14.  Capillary Index  0.20  «  0.16 h  S3 0.12 -  0.08 h  0.04 h  0.00  76  Fig. 15. Fluorescent micrographs of portions of the corrugator supercilii (a) and orbicularis oculi (b) immunolabelled with anti-neurofilament protein. Peripheral nerve fascicles (arrows) and fibers with motor end-plates (*) possessing multiple terminal boutons are indicated. Notice that the nerve fascicles in the orbicularis are located in the connective tissue outside of the muscle fascicles, while those of the corrugator course through the muscle fascicles. X460.  77  Fig. 15. -•  * p  |  / i  •  «  ^  *  /••• #  1  i  •  •  78  •  50 nm  Fig. 16. Serial transverse sections of the corrugator stained with H & E (a), NADH-tr (b), anti-NFP (c) and anti-slow myosin (d). Two fibers with motor end-plates are indicated (*) in each section. In panel (a) the nuclei associated with motor end-plates are visible. In panel (b) the mitochondrial clusters associated with motor end-plates can be seen. Panel (c) shows the nerve axons and terminal boutons of the motor end-plates, and panel (d) reveals that the two motor end-plates are located on two unstained fast-twitch (type II) fibers. In panel (b) note that the mitochondria-rich fibers (dark-staining) are also the fibers that express the slow myosin isoform (red-staining). All panels X330.  79  80  Fig. 17. Serial transverse sections of orbicularis oculi stained with H & E (a), NADH-tr (b), antiNFP (c) and anti-slow myosin (d). The same fiber is marked (*) in all four panels. In panel (a) the marked fiber has not yet received the brightly staining motor end-plate that can be seen terminating on it in panel (c). In panel (b) the marked fiber is a dark mitochondria-rich fiber that can also be seen to express slow myosin in panel (d). Notice that the dark mitochondria-rich fiber marked with the arrow in panel (b) does not express slow myosin in panel (d). All panels X330.  81  Fig. 17.  82  F i g . 18.  Fig. 19. Three plates showing some of the salient features of the motor end-plate organization in the corrugator supercilii. a) Composite drawing of the motor end-plate distribution pattern of the corrugator supercilii. Motor end-plates appear throughout the muscle in clusters of 4 to 20. A predominant H pattern (arrows) emerges when motor end-plate sites are positioned on a diagram. b) High power dissecting microscope view of the corrugator after being stereoscopically dissected and stained with PIA. Motor end-plate clusters are easily visible (black dots). A small secondary nerve and its associated vessel can be seen between the two clusters of motor end-plates (arrow). c) A single motor end-plate seen from above. From this vantage, it appears punctate and rounded, revealing many complex synaptic folds. XI 000.  85  ,20. a) Light micrograph of mouse E D L stained with PIA. Motor end-plates are stained darkly showing their large size and relative lack of synaptic infolding compared to that in the human facial muscles (see Figs. 18c, 19c). X930. b) Light micrograph of mouse gastrocnemius muscle stained with PIA. The large motor end-plates in this muscle lack the complexity of synaptic infolds compared to human motor end-plates (see Figs. 18c, 19c). X930.  87  Fig. 20  e  88  Fig. 21. Electron micrographs detailing the typical arrangement of myofilaments in the two periorbital muscles. a) Section of corrugator supercilii demonstrating the typical longitudinal arrangement of actin and myosin filaments within the sarcomeres. Aligned directly below it is a schematic diagram of the sarcomere organization. The same arrangement was also seen in the orbicularis oculi. Mitochondrial doublets (arrows) straddling the Z-line and a triad (circled) at the A-I junction are indicated. X50 400. b) High magnification view of a transversely sectioned myofibril cut through the outer A band. The thin (actin) and thick (myosin) filaments are in a typical double hexagonal array. A schematic diagram of the myofilament arrangement can be seen to the right of the micrograph. X121 000.  89  Fig. 22. Series of four electron micrographs demonstrating some of the ultrastructural features of the corrugator supercilii. a) Longitudinal section showing rounded mitochondrial doublets (arrows) that straddle the Z-lines of the sarcomeres. Triads (circled) comprise a central T-tubule and two cisternae of the sarcoplasmic reticulum. X21 350. b) Cross sectional view of myofibrils and mitochondria (Mt). Note the poorly developed sarcotubular system surrounding the myofibrils. X53 500. c) Longitudinal section showing mitochondria (Mt) located just under the sarcolemma (S). The mitochondria seem to exhibit tubulovesicular cristae. Note the poorly developed sarcotubular system and the triad (circled). XI5 550. d) Longitudinal section showing rows of intermyofibrillar mitochondria (Mt). The Z-lines in this fiber appear thicker than those in the orbicularis at the same magnification (see Fig. 22d).Xll 900.  91  92  Fig 23.  Series of electron micrographs demonstrating some of the more commonly seen ultrastmctural features of the orbicularis oculi. a) Micrograph showing mitochondrial doublets (arrows) at the Z-lines and triads (circled) at the A-I junction. X23 000. b) Transverse section of myofibrils showing a small rounded mitochondrion (Mt) amidst a well developed sarcotubular system (arrows). X36 800. c) Micrograph showing a triad (circle) with two portions of the sarcoplasmic reticulum (r) between an extensive network of T-tubules (t). X37 200. d) Longitudinal section of a type II muscle fiber showing a peripheral nucleus (N) and a group of perinuclear mitochondria (Mt). The sarcolemma (S) and Z-lines (Z) are identified. X9 300.  93  94  . 24. a) Electron micrograph of a continuous capillary with a red blood cell (RBC) in its lumen. This type of capillary was the most common type seen in the orbicularis and the corrugator. Tight junctional complexes bind the endothelia together (circled). XI2 125. The inset shows a junctional region at higher magnification (arrows). Inset X34 800. b) Electron micrograph of a fenestrated capillary with a red blood cell (RBC) in its lumen. This type of capillary was only observed in the orbicularis muscle. The fenestrated region (circled) is much more permeable to macromolecules. XI2 125. The inset shows the diaphragms (arrows) of the fenestrated region at higher magnification. Inset X34 800.  95  Fig. 24.  Fig. 25. Electron micrographs of motor end-plates in the corrugator (a) and the orbicularis (b). a) The terminal bouton of the nerve axon (A) sits within the primary synaptic cleft (arrows) with secondary synaptic clefts (arrow heads) projecting into the fiber. There are numerous mitochondria (Mt) within the muscle fiber. XI6 300. A portion of the neuromuscular junction (*) is enlarged to show the wide unbranched secondary synaptic clefts (arrow heads). Inset X27 000. b) Two terminal boutons (A) sit deep in the primary synaptic cleft (arrows). Several slightly degenerated axonal mitochondria (M), and some cytoplasm of the insulating Schwann cells (S) are seen.. Mitochondria (Mt) and a portion of a myonucleus (N) are indicated within the muscle fiber. XI6 300. A region of the motor end-plate (*) is magnified to show the narrow and highly branched secondary synaptic clefts (arrow heads). Inset X27 000.  97  Fig. 26. a) Electron micrograph of a tissue macrophage (M) beside a capillary (C). The large number of pseudopodia suggest that this macrophage may be active in phagocytosis or cell movement. XI2 100. b) Electron micrograph of a mast cell beside a capillary (C). Its irregular and central nucleus (N) is surrounded by numerous electron-dense granules. XI1 000. c) Electron micrograph of a lamellar inclusion in a muscle fiber from the corrugator. XI7 000. d) Light micrograph of a semithin transverse section of a spindle-like structure in the orbicularis. Two muscle fibers (*) are surrounded by a myelinated peripheral nerve with its perineurium (arrows). X850.  99  Fig. 26.  100  DISCUSSION To our knowledge this is the first study to compare the two human periorbital facial muscles, corrugator supercilii and orbicularis oculi, at gross anatomical, light microscopic and electron microscopic levels. Results of the morphological comparison between the palpebral portion of the orbicularis oculi and the corrugator supercilii strongly suggests that the unique functions of these two facial muscles, and the complexity of their functions, are reflected in their architecture at both macroscopic and microscopic levels. Their differences and similarities are factors to consider for those either studying or treating human facial muscles and the various conditions affecting them. When comparing the morphological features of the orbicularis with those in the corrugator, it is important to consider their respective functions. The palpebral portion of the orbicularis contracts an average of 12.5 times per minute to close the orbital opening in the form of a spontaneous blink lasting approximately 70 to 100 msec. (Safran, 1989). The blink, itself, is a complex movement crucial to maintaining the integrity of the corneo-scleral epithelium by sweeping a thin layer of protective tear fluid across its surface (Hung et al., 1977; Doane, 1980; Furnas, 1981; Karson, 1989). Corrugator contractions, on the other hand, tend to be of a more tonic and voluntary nature, occurring sporadically for variable lengths of time. When corrugator contractions do occur, they serve to pull the glabellar skin medially and downward producing the vertical furrows on either side of the nasion, as well as the numerous shades of expression. The primary role of the corrugator is to produce facial expressions, while the orbicularis is a sphincteric muscle involved in the closing of the orbital orifice (Hojo and Shinoda, 1982; Bradley et al., 1993; Foster et al., 1996). Despite having very different functions, the two muscles both developfromthe  101  deep layer of the second branchial arch, attach to the rim of the orbit, and receive motor nerve supply from the temporal branch of cranial nerve VII (the facial nerve).  ULTRASTRUCTURAL DIFFERENCES In the present study, a preliminary survey of the ultrastructure of the corrugator supercilii and the orbicularis oculi revealed several differences between the two muscles. In the corrugator there was a relative abundance and variety of shapes of mitochondria, and thicker Z-lines . In the orbicularis there was a well developed sarcotubular system, and more complex synaptic folds in the motor end-plates. Many of these same differences have been reported to exist between type I and type II fibers elsewhere in the body (Ogata, 1965, 1988a; Ogata and Murata, 1969; Salmons et al., 1978; Fardeau and Tome, 1994). For example, A well developed sarcotubular system is reported to be linked with speed of contraction and the enhanced depolarization/repolarization of type II fibers (Buller, 1976; Ogata and Yamasaki, 1985; Park-Matsumoto et al., 1992). The increased complexity of primary and secondary synaptic clefts of type II fibers is thought to be related to faster rates of sarcolemmal membrane depolarization when compared to the slower and less complex type I fibers (Murata and Ogata, 1969; Deschenes et al., 1994). Indeed, some of the observed differences between the corrugator and orbicularis may be partly due to the differences in fiber-type populations within the two periorbital muscles. If the apparent fine structural differences are compared within fiber types (i.e. either type I or II orbicularisfiberscompared to either type I or II corrugator fibers, respectively) most of the differences become more subtle. Nonetheless, differences in mitochondrial number and their distribution in muscle fibers still exist.  102  Mitochondrial distribution In this study, it was shown that type II fibers in the corrugator and type II fibers in the orbicularis both possessed sparse populations of mitochondria. However, the type II fibers of the orbicularis appeared to have more subsarcolemmal clusters of mitochondria whereas the corrugator appeared to have a more homogeneous distribution of mitochondria throughout the muscle fiber. This preferential and subsarcolemmal distribution of mitochondria in the orbicularis may represent a strategy the muscle has developed to meet its unique functional demands. Based on a contraction rate of 12-13 times per minute (Karson, 1989; Safran, 1989), the orbicularis muscle contracts over 750 times in every waking hour. As a primarily fast-twitch skeletal muscle (having only 11% slow-twitch fibers) it would use stored glycogen as its primary fuel source (Ogata and Mori, 1964; Beatty and Bocek, 1970; Buchthal and Schmalbruch, 1980; Williams et al., 1989; Hudlicka et al., 1992) which it must replenish via aerobic pathways. Its high activity rate, coupled with its relatively low oxidative capacity, might conceivably create a selective pressure whereby the fast-twitch fibers of the orbicularis would need to maximize their oxidative efficiency in order to meet their constant energy drain. By concentrating its limited quantities of mitochondria around the periphery of the fiber, the orbicularis may maximize its oxidative ability both by reducing diffusion times and by increasing the local concentration gradients of oxygen and other necessary components of aerobic metabolism. Since the corrugator is not contracting 750 times per hour, and because it is not faced with a lack of slow-twitch fibers (51% type I fibers), it would not be under the same pressure to localize mitochondria near the periphery. This would also help explain the lack of subsarcolemmal mitochondria and the homogeneous staining pattern seen in type II fibers of the corrugator when NADH-tr staining techniques are used.  103  Should this proposed oxidative maximization theory be the explanation for the peripheral distribution of mitochondria in the orbicularis, it could conceivably help explain why some mitochondria-rich fibers of the orbicularis also express fast myosin. Indeed, these fibers may exist as a unique, fast-contracting, yet highly oxidative, fiber type specially adapted for the palpebral portion of the orbicularis oculi and its unusual function of blinking. A recent study by Stal and coworkers (1994a) reported the existence of a unique and previously undetected fast myosin heavychain isoform in some of the human oro-facial muscles. It is conceivable that this "fast facial" myosin might also be expressed in the specialized muscle fibers of the palpebral portion of the orbicularis oculi. Besides the oxidative enzyme quantities and the mitochondrial distribution, there may be an additional indicator of the differential oxidative capacities of the two periorbital muscles. It is well established that the oxidative capacity of a muscle is reflected in its capillary blood supply (Simionescu and Simionescu, 1988). The capillary index of the corrugator is 2.4 times that of the orbicularis. This difference in capillary density is reflective of the corrugator's greater oxygen uptake ability. However, in addition to the difference in the number of capillaries, the placement of capillaries within the muscle was also different. In the orbicularis, capillaries appeared to be further from the muscle fibers than those in the corrugator. It has been shown that the diffusion of oxygen into the muscle cell can be greatly affected by increased diffusion distances (Wissig, 1964; Simionescu and Simionescu, 1988). Even small increases in diffusion distance can have deleterious effects in tissues such as the myocardium (Cross and Mercer, 1993). Therefore, the arrangement of capillaries further from the muscle fibers of the orbicularis may require that the mitochondria, with their compliment of oxidative enzymes, be placed closer to the periphery of the muscle fibers to help minimize diffusion times.  104  The appearance of fenestrated capillaries in the orbicularis may also be noteworthy. Continuous capillaries are rather ubiquitous throughout the body. They are characteristically found in muscle tissue mnning parallel to the longitudinal axis of the muscle fibers, and many histological texts site skeletal muscle as an area whereby typical continuous capillaries can be seen (Fawcett, 1986; Simionescu and Simionescu, 1988; Cross and Mercer, 1993). Each endothelial cell in a continuous, or tight capillary, links to those around it by occluding (tight) junctions. These intercellular junctions do not occupy the entire margin of the endothelia. Between them there are intercellular clefts of approximately 10-20 nm through which fluid and macromolecules can pass (Wissig, 1964; Karnovsky, 1967; Fawcett, 1986). Fenestrated capillaries, on the other hand, are typically found in areas that have high and faster rates of exchange between the plasma and surrounding tissues, such as in the mucosa of the gastrointestinal tract and in endocrine glands (Simionescu and Simionescu, 1988). The endothelia of fenestrated capillaries are essentially similar to those of the continuous variety except for an attenuated region of the cell cytoplasm which has several permanent and circular openings called fenestrae (Fawcett, 1986; Simionescu and Simionescu, 1988; Cross and Mercer, 1993). Most endothelial fenestrae have a diameter of approximately 60-80 nm, with each one being closed by a thin diaphragm with a central thickening of 10-15 nm (closed fenestrae) (Fawcett, 1986; Simionescu and Simionescu, 1988; Cross and Mercer, 1993). However, in locations such as the liver there are endothelia without diaphragms across each fenestra (open fenestrae) (Cross and Mercer, 1993). The fenestrated region, whether open or closed, is believed to facilitate the extensive exchange of materials between blood plasma and surrounding tissues (Fawcett, 1986; Simionescu and Simionescu, 1988; Cross and Mercer, 1993).  105  The fenestrated capillaries seen in the orbicularis may indicate that there is a special need for macromolecular exchanges to occur between the plasma and either the muscle fibers, or the surrounding connective tissue cells. Alternatively, the appearance of fenestrated capillaries might indicate the need for faster and more efficient molecular exchanges between the blood and the surrounding muscle tissue than can be achieved with an exclusive population of tight capillaries. Perhaps fast molecular exchanges are required in the orbicularis muscle because of the fast (70-100 msec/blink) and repetitive nature of spontaneous blinking (Hung et al., 1977; Doane, 1980; Karson, 1989; Safran, 1989). The precise scientific reason for the existence of a population of fenestrated capillaries within the orbicularis oculi requires further and more detailed investigation than was possible in this study.  Lamellar Inclusions Lamellar-like cytoplasmic inclusions, or myeloid figures, have been reported previously in skeletal muscle fibers (Schiaffino et al., 1977; Cullen and Mastaglia, 1982; Ovalle and Dow, 1986). There is currently no known reference to these lamellar inclusions in normal human facial muscle. They have, however, been reported to occur in association with a range of muscle pathologies (Schiaffino et al., 1977; Cullen and Mastaglia, 1982; Ovalle and Dow, 1986). Bardosi and coworkers (1987) reported their appearance in denervated human platysma muscle fibers along with numerous other cytoplasmic inclusions. Ovalle and Dow (1986) reported sighting myeloid figures in the intrafusal fibers of dystrophic mice, and suggested that they may be remnants of degenerating mitochondria. Others have suggested that they are lysosomal remnants found in pathologically affected muscle fibers which have compromised their ability to eliminate accumulating degradation products (Schiaffino et al., 1977; Cullen and Mastaglia, 1982; Engel and  106  Banker, 1986). The appearance of these inclusions in normal facial muscles of healthy persons (most of whom were in their fifth decade) may implicate them as a normal age-related pathologic change. Little is known about the structure of muscle lysosomes because they are seldom seen in normal skeletal muscle (Cullen and Mastaglia, 1982). The degradation of muscle fibers which occurs in a plethora of myopathic conditions points to there being some form of lysosomal involvement in these disorders. However, the precise role of lysosomes in skeletal muscle is yet unknown. The apparent lack of primary lysosomes in striated muscle has led some to suggest that the sarcotubular system has specialized regions for lysosomal enzyme storage (Christie and Stoward, 1977). These investigators were able to show that the sarcotubular system reacts positively for acid phosphatase which is a known lysosomal enzyme. It has also been suggested that the membranes of autophagic vacuoles are derived from the sarcoplasmic reticulum (SR) (Cullen and Mastaglia, 1982). If lysosomal enzymes in muscle fibers are produced by the SR, and if they are sequestered by the SR membranes, then it may prove to be that the lamellar bodies are residual lysosomal storage areas for the membranes of the sarcotubular system which have been "dismantled". It would stand to reason that the sarcotubular membranes be resilient to digestion from the enzymes which it normally produces. A lack of these residual lysosomes in normal muscle may indicate that there is a minimal amount of sarcotubular restructuring required under normal circumstances. If there were to be an increase in the amount of sarcotubular breakdown associated with a particular pathology, it would increase the population of the residual lysosomes. If this were the case, it would help explain why lamellar bodies were seen in the normal facial muscles of older persons. The aging process may gradually increase the need for storage of the undigestable, or slowly digestible, sarcotubular membranes which could be provided by these lamellar inclusions.  107  LIGHT MICROSCOPIC DIFFERENCES Besides their oxidative capacities, some of the main light microscopic differences seen between the corrugator supercilii and the orbicularis oculi were the shapes and sizes of the muscle fibers, and their fiber-type distributions. Again, the principle consideration for examining these differences is to explain the link between the fiber shapes, sizes and types with the functions of these periorbital muscles.  Fiber type distributions Fiber-type distribution patterns can be of use when attempts are made to characterize how the form of a muscle relates to its function. Much of the early histological data on human facial muscles have come from fiber typing using ATPase staining techniques (Happak et al., 1988; Freilinger et al., 1990; Stal et al., 1990; McLoon and Wirtschafter, 1991). In the present study, however, myosin isoform expression was used to determine the fiber-type distribution patterns of the two periorbital muscles. This was chosen because myosin isoform expression has been considered to be a more effective correlate to the physiological contraction speed of a muscle fiber (Burke et al., 1974; Buchthal and Schmalbruch, 1980; Williams et al., 1989; Hoh, 1992; Stal et al., 1994). The palpebral portion of the orbicularis, with 89% of fibers expressing fast myosin, is clearly a skeletal muscle well suited to high speeds of contraction, and poorly suited to slower and more sustained contractions. The corrugator, however, with 51% of fibers expressing slow myosin, is better suited to slow contractions. This conclusion supports the contention of Freilinger et al. (1990) who, using ATPase staining, classified the corrugator as a physiologically "tonic" muscle and the palpebral orbicularis as a physiologically "phasic" muscle.  108  Functionally, phasic skeletal muscles, such as the orbicularis, and tonic skeletal muscles, such as the corrugator, can be expected to exhibit characteristics in keeping with their fiber-type distribution patterns. One characteristic that may be influenced by their fiber type distribution is the size of their motor units. Phasic muscles tend to have larger motor units than tonic muscles (Olson and Swett, 1966; Buchthal and Schmalbruch, 1980; Brown, 1984; Hijikata et al., 1992). Moreover, it has been reported that the size of a muscle's motor units is related to the degree of controlled movement, or finesse, required by that muscle (Olson and Swett, 1966; Buchthal and Schmalbruch, 1980; Brown, 1984). By recruiting smaller motor units, the central nervous system can gradually increase the force production of a muscle rather than the abrupt increases in force that are associated with the recruitment of larger motor units (Buchthal and Schmalbruch, 1980; Brown, 1984; Lev-tov et al., 1988). This being the case, it may be expected that the corrugator, a muscle which produces a myriad of fine movements associated with facial expression, would have smaller motor units than the orbicularis, a muscle in which partial bilateral lid closure causes lid tremor — an indication of lack of fine control (Schmidtke and Buttner-Ennever, 1992). This may help explain why, when stained with PIA, the corrugator supercilii demonstrated fewer motor end-plates (with approximate groups of 4-20) in the motor end-plate clusters than the orbicularis oculi muscle (with approximate groups of 4-30).  Muscle fiber shapes and sizes Both the corrugator supercilii and the orbicularis oculi are non-pennated muscles that have their fibers organized in what is often called a "parallel array" (Williams et al., 1989). The obvious fundamental difference between the two is that the corrugator fibers are arranged linearly, while the orbicularis fibers are arranged in an arc spanning the palpebral fissure. The shortening of muscle  109  fibers, as a result of contraction, produces tensile forces parallel to the long axis of the muscle fibers (Loeb et al., 1987; Gans and Gaunt, 1991; Trotter, 1993). In addition to the tensile forces, contraction of the arced fibers of the orbicularis produces vector forces perpendicular to their longitudinal axis which are also known as "shearing forces" (Williams et al., 1989). Fibers of the corrugator contract linearly, exerting the usual tensile forces parallel to their longitudinal axis, but they are not likely to produce the added forces associated with contracting in a non-linear plane. Since the orbicularis is composed predominantly of fast-twitch (type II) fibers that produce shearing forces perpendicular to the fiber axis, there may be an increased selection pressure towards creating fibers that best diffuse those forces and minimize the risk of producing fiber damage. One possible explanation for why the muscle fibers of the orbicularis are smaller and rounder than those in the corrugator could be based on the effective dispersal of contractile forces. It has been shown that the force which a muscle fiber produces during contraction is transferred into the surrounding extracellular connective tissues through cytoskeletal connections linking the outside of the cell to the force-producing myofibrils inside the cell (Trotter, 1990, 1991, 1993; Scott et al., 1992). Through such links, forces can be summated to create movement (Brown, 1984; Williams et al., 1989). In most skeletal muscles, this summation of forces is directed through a discernible tendon to produce movement across a joint. In facial muscles, however, forces are passed more diffusely into facial skin by muscle fibers blending discretely into the subdermal connective tissues. It is more advantageous to distribute forces in a predictable and controlled fashion. It would be easier to control force distribution into connective tissues when thefibersare geometrically regular in their shapes, such as the rounded fibers of the orbicularis, rather than the irregular and pleomorphic-shaped fibers of the corrugator. Because of the risk that excessive shearing forces may cause sarcolemmal damage, uniform force dispersal would be particularly  110  important in fibers which produce shear forces across their longitudinal axis, as do the orbicularis fibers. According to Trotter (1990), interfiber tension-transmission occurs between muscle fiber and endomysium across most, or all, of the sarcolemmal interface. It stands to reason that force transmission would be better in a muscle fiber with more sarcolemmal interface than in a fiber with less sarcolemmal interface. Perhaps by maximizing its endomysial contact, a muscle fiber increases its effectiveness in transmitting forces outward into the surrounding connective tissue. This may explain the need for the orbicularis oculi to maintain larger amounts of perimysial and epimysial connective tissues to ensure that there is sufficient contact to assist with the dispersal of the shearing forces. In the corrugator, a muscle which is not likely to experience the same internal shearing forces due to its linear fiber arrangement, the need to have complete connective tissue coverage for each fiber may be reduced. This same type of physics-based reasoning may also help explain why muscle fibers of the orbicularis oculi have a relatively small cross-sectional area when compared to those of the corrugator supercilii. If one were to construct a system that would mimic the function of the palpebral portion of the orbicularis using the same components that are in the system now, it would be limited by standard structural principles. For instance, when a system is required to undergo shear stresses, it is best to reduce the shearing force exerted on any one component of the system (Williams et al., 1989). One strategy to accomplish this is to distribute the force in small increments, over several, smaller components rather than across fewer, larger components. As an example of the application of this principle, one might consider a steel cable. Such a structure is purposefully composed of several wires of small cross-sectional area as opposed to one large wire. These wires help maintain the cable's flexibility and also reduce its susceptibility to shearing  111  damage during motion. In order to minimize the shear stress upon the libers of the orbicularis, it may be necessary for them to retain a relatively small cross-sectional area when compared to the fibers of a muscle, such as the corrugator, which are unlikely to have any added stress resulting from contracting in a non-linear path.  GROSS ANATOMICAL DIFFERENCES Human facial muscles are truly unique in their gross anatomical arrangements because of the complex spatial relationship they maintain just under the skin of the face. They function as movers of skin rather than as movers of joints, and they do so in a manner that allows a plethora of facial expressions unmatched by any other living creature. Despite the fact that human facial muscles have been dissected, and their spatial relationships have been detailed and studied for centuries, there are still many organizational questions which remain unanswered.  Motor end-plate zones Several human and subhuman skeletal muscles are characterized by having motor endplate, or innervation zones, arranged in narrow bands which are located perpendicular to the muscle fiber orientation, and midway between their origins and insertions (Brown, 1984; Loeb et al., 1987; Borodic et al., 1991). In the present study, this typical pattern was not seen in either the corrugator or the orbicularis when whole muscle samples of each muscle were stained with PIA. Instead, the corrugator demonstrated a somewhat random scattering of motor end-plates with a tendency for them to be organized in a loosely structured " H " pattern. On the other hand, the orbicularis showed an almost entirely random scattering of motor end-plates except for the appearance of slightly higher concentrations located towards the lateral portion of the muscle. The precise reason for this varying distribution pattern is uncertain, however, there are several plausible explanations.  112  One  such explanation relies on previously reported findings that more complex  innervation patterns exist in skeletal muscles with more complex arrangements of muscle fibers (Brown, 1984; Gans and Gaunt, 1991). For instance, the human sartorius and iliacus muscles both have been reported to exhibit numerous innervation zones scattered throughout them (Gans and Gaunt, 1991). They are also short-fibered muscles that have several tendinous insertions between parallel bundles of short muscle fibers. The short-fiber arrangement is one in which long strap-like muscles with fibers in parallel arrays are composed of several shorter fibers in series, rather than long muscle fibers which span the length of the entire muscle (Williams et al., 1989; Gans and Gaunt, 1991). Short-fibered muscles can either be arranged with fibers abutting end-to-end or they can overlap one another. If there is a specific pattern to the alignment of the short fibers, muscles will appear to have several bands of innervation zones. If there is no alignment because the short fibers within the muscle are overlapped in a more complex manner, then it may appear that the innervation zones are scattered throughout the muscle. Should this be the case in the two periorbital muscles examined in this study, it would be logical to assume, due to their motor-end plate distribution patterns, that the fibers in the corrugator exhibit a more regular overlap of muscle fibers than the orbicularis giving the subtle "H-like" distribution pattern of its motor end-plates. Likewise, the random appearance of the motor end-plate distribution pattern in the orbicularis would lead to the belief that its short fibers have a more intricate and less-ordered overlap. In this study, there was some evidence that the orbicularis and the corrugator muscles were both composed of overlapping short muscle fibers. However, this line of observation was not pursued stringently enough to conclude that the corrugator and orbicularis are true short-fibered muscles. Indeed, determining whether or not the muscle fibers of these two human periorbital muscles do overlap would be an interesting future directive. 113  Despite the lack of empirical evidence, there is reason to suspect that these two periorbital muscles  are short-fibered skeletal muscles.  From structural  and functional  perspectives, it would be a distinctive disadvantage for the fibers of the orbicularis to run the entire length of the palpebral fissure. From a functional perspective, long and singly-innervated muscle fibers would reach maximal force production more slowly than shorter fibers. Since the orbicularis is such a fast-contracting muscle (Safran, 1989), and because there is no evidence to suggest that fibers of the orbicularis have multiple end-plates, it would seem unlikely that this muscle is composed of fibers in which maximal force-production times would be lengthy. From a structural perspective, shear forces would again be lessened if the orbicularis were comprised of short-fibers because each of these fibers would be less capable of producing damaging levels of force due to the reduced number of sarcomeres per muscle fiber. In addition, the arcing within each fiber would be lessened if the fibers were shorter which would also reduce the production of damaging shear forces. Clinically, it has been shown by that the human gracilis muscle is a better donor muscle for use as a distant muscle-pedicle in facial muscle reanimation surgeries than muscles such as the extensor digitorum brevis (O'Brien, 1980). It has also been shown by others that the gracilis muscle exhibits a short-fibered architecture (Gans and Gaunt, 1991), whereas the human extensor digitorum brevis ends in four tendons and is a pennated muscle (Williams et al., 1989). Though it is still speculative, it may be that the success of the human gracilis muscle as a donor muscle for facial muscle transplants is due, in part, to its being a short-fibered skeletal muscle.  114  Motor nerve distribution In the present study, the motor nerve supply to the corrugator supercilii was found to come from the superficial surface and appeared relatively simple in the manner in which it branches throughout the muscle. The palpebral portion of the orbicularis oculi, on the other hand, demonstrated some very complex features in its branching pattern. The branches of the temporal nerve (from C N VII) that supply the orbital portions of the orbicularis oculi have been shown by others to exhibit several variations (Malone and Maisel, 1988; Safran, 1989; Monkhouse, 1990; Ammirati et al., 1993). It is often expected in the branching patterns of peripheral nerves that there will be a significant amount of variation, especially in the smaller-diameter nerves (Ruskell, 1985; Proctor, 1991; Myint et al., 1992). However, the small, arcing-nerve distribution pattern seen on the palpebral portion of the orbicularis oculi appeared with regularity. This may indicate that the manner in which these small nerves are organized has functional significance. Perhaps the axons of each of the motor units is organized so that their activation occurs in a sequence which is important to orchestrate the complex series of events involved in a spontaneous blink. There are, to our knowledge, no published studies reporting that the lateral portion of the palpebrum either initiates or completes movement before the medial portion. It has been shown that the lower palpebrae contract with a horizontal and medially directed component which serves to sweep dust, debris and tear fluid toward the lacrimal punctae (Doane, 1980). Though it has not yet been shown, there is an organizational indication that the upper lid might also have some sort of horizontal component in its movements. With the lacrimal gland being located in the upper lateral region of the orbit, and the lacrimal sac being in a lower medial location, it seems as if the most effective way to spread tears would be with a tangential motion across the eye. It is possible that the arcing nerves of the superior palpebrum play a role  115  in mediating the spread of tears across the eye in a superolateral to inferomedial direction by ensuring the proper distribution and firing of motor units.  Sensory innervation Sound, light and several types of mechanical stimuli such as glabellar tapping, corneal, lid and eyelash touch, or electrical stimulation of cutaneous nerves (such as the supraorbital nerve) can all elicit the blink reflex (Brown, 1984; Karson, 1989; Safran, 1989). There is evidence that most of these reflexes are mediated centrally by the trigeminal nerve (CN V) as either monosynaptic or polysynaptic reflexes (Brown, 1984; Monkhouse, 1990). There is also some evidence to show that components of the blink reflex can be produced by electrical stimulation of afferent fibers in the facial nerve, but it is not clear if some or all of these afferents pass by way of connecting branches to C N V prior to reaching the brainstem (Brown, 1984). Evidence relating to facial muscle reflexes, and the fact that humans have the ability to create fine variations in tone and contraction of facial muscle when producing facial expressions, leads one to intuitively postulate that some form of sensory and proprioceptive feedback exists in human facial muscles. Using histological techniques, Baum (1899) was the first to search for muscle spindles in muscles innervated by the facial nerve. He reported that there were no muscle spindles in either the auricular muscles, posterior belly of digastric or stylohyoid (Baum, 1899; Blevins, 1964). In 1927, Hines, concluded that there were no muscle spindles in the muscles of facial expression of rabbits. Since then, several researchers have reported on the existence of spindles in human facial muscles (Kadanoff, 1956; Belal, 1982). However, none of these reports were accompanied either by detailed morphological descriptions or by photographs of the spindles which were apparently observed.  116  In some subhuman vertebrates, there have been substantiated reports on the existence of muscle spindles in muscles innervated by the facial nerve. Lovell et al. (1977) found typical muscle spindles in the auricular muscles of two species of monkeys and the baboon, and others (Kubota and Masegi, 1972; Kubota et al., 1975a, b; Kubota et al., 1978) also found muscle spindles in the snout musculature of the mole. In contrast, an absence of muscle spindles was reported in the facial muscles of rabbits, pigs, cats, guinea pigs, and rats (Smith, 1926; Cipollone, 1927). Reports on other muscles of the head and neck have been mixed. The muscles of mastication and the strap muscles in both human and sub-human species have been shown to have typical muscle spindles (Silverstein and Graham, 1978; Rokx et al., 1984; Rowlerson et al., 1988; Eriksson and Thornell, 1990; Bout and Dubbeldam, 1991; Jakubowicz et al., 1992; Eriksson et al., 1994). In extraocular muscles, typical spindles were not found in either the mouse or the monkey (Pachter et al., 1976; Ruskell and Wilson, 1983), yet they have been reported to occur in extraocular muscles of other vertebrates (Maier et al., 1975). Despite all of the indirect and reported evidence, it remains unclear whether or not human facial muscles possess either muscle spindles or some kind of spindle-like structures which have been overlooked. Alternatively, it is possible that the need for sensory feedback from muscles of facial expression is met by pre-existing peripheral cutanous mechanoreceptors (Brown, 1984). Since they were first described by Hassall in 1849, muscle spindles have been studied extensively (Barker, 1974; Boyd and Smith, 1984). Their architecture and sensory nature are now well understood. In simple terms, muscle spindles are collections of specialized "intrafusal" muscle fibers surrounded in part by a fluid-filled capsule. The intrafusal fibers are unique in that they receive both motor and sensory innervation (Barker, 1974; Landon, 1982; Boyd and Smith, 1984; Adal, 1985; Ovalle and Dow, 1985; Sahgal et al., 1985). They are highly organized, encapsulated  117  sensory receptors which lie in parallel with the extrafusal fibers making up the bulk of the muscle. As mechanoreceptors, muscle spindles respond to active and passive deformations such as muscle stretch and contractile shortening (Landon, 1982; Adal, 1985; Barker and Banks, 1986). Interestingly, Ruskell and Wilson (1983) make special reference to simple spiral endings of myelinated nerves which wrap tightly around a single muscle fiber in the extraocular muscles of rhesus monkeys. The significance of these simple spiral nerve-endings in extraocular muscles may bear some similarity to the spindle-like structure reported in this study. It is known that there are three types of intrafusal fibers within a typical mammalian spindle (Ovalle and Smith, 1972). There are nuclear-chain fibers, so named because of their small diameter and single row of centrally-located fusiform myonuclei, and there are two types of larger diameter nuclear-bag fibers, full of large, vesicular-shaped, central nuclei (Barker, 1974). The two types of bag fibers differ in ultrastructural,  histochemical, contractile and developmental  characteristics. Simply stated, the bag 1 fibers are thought to mediate the dynamic actions of the muscle and the bag 2 fibers, along with the chain fibers, are known to mediate static functions (Kennedy, 1970; Ovalle and Smith, 1972; Barker, 1974; Barker and Banks, 1986). Mammalian spindles are supplied with a primary sensory ending and one or more secondary sensory endings (Nahirney and Ovalle, 1992). Their motor innervation consists of a fusimotor (y) system for innervating intrafusal fibers, only, and a skeletofusimotor ((3) system which innervates intrafusal and extrafusal fibers (Landon, 1982; Jones, 1983; Barker and Banks, 1986; Fawcett, 1986). If the spindle-like structure observed in the orbicularis oculi (see Fig. 26d) were to function as a simple proprioceptive device, its structural and functional significance would become integral to explaining how sensory feedback works in human facial muscles. Assuming, for the  118  sake of discussion, that the spindle-like structure is a type of proprioceptive device begs the immediate question about why it appears rudimentary in form rather than appearing as other muscle spindles do elsewhere in the body (Kennedy, 1970; Adal, 1985; Ovalle and Dow, 1985; Sahgal et al., 1985). The rudimentary architecture of this spindle-like structure, hereafter termed a facial spindle, may be explained by the minimal need for stretch receptors in human facial muscles. A critical role of muscle spindles and golgi tendon organs is that of protection from either excessive stretch or contraction (Boyd et al., 1985; Hunt and Wilkinson, 1985). It has previously been postulated that, since facial muscle fibers do not undergo the same unexpected load or length changes that other spindle-containing skeletal muscles do, they do not have the same need for stretch receptors (Brown, 1984). Despite the apparent lack of need for passive stretch receptors, human facial muscles may still require a more "active" type of proprioceptor for afferent input on spatial positioning and modification of active contraction. This could possibly explain why only a rudimentary form of spindle is required, one that is spindle-like, but lacks the same complexity that is usually associated with the typical muscle spindle found in other muscles. Interestingly, the nonmammalian muscle spindle is also less complex, having one sensory ending and receiving its motor innervation from branches of axons that also innervate extrafusal muscle fibers (Barker and Banks, 1986). If the facial spindles do not require the same order of complexity that has been reported in muscle spindles elsewhere in the body (Kennedy, 1970; Adal, 1985; Ovalle and Dow, 1985; Sahgal et al., 1985), this may explain why nerve-fiber discharges could not be detected either in the facial or trigeminal nerves when human facial muscles were given intravenous succinylcholine, a drug known to excite spindle receptors (Granit et al., 1953). The possibility that this interesting spindle-like structure actually has a proprioceptive role in muscles of facial expression is merely conjecture at this point, and theorization about its function 119  must be tempered with caution. The paucity offacial spindles encountered in the present study, and the weight of the published experimental evidence still suggests that there is little need for proprioceptive mechanisms in facial muscles. However, the fact that such a structure has been observed may stimulate future studies on this topic. The possibility that the morphological and physiological nature of these facial spindles could prove important to understanding the proprioceptive character of human facial muscles certainly warrants future study.  120  CONCLUSIONS AND FUTURE DIRECTIONS Arguably, those who are the most aware of the importance of human facial muscles are likely to be those who have lost their ability to control them. Blinking, speech, emotional expression, and eating are all incredibly unique and complex events that depend greatly, if not entirely, on proper functioning of the facial muscles. The need to understand how these skeletal muscles are organized, how they interrelate and how they function, is necessary so that physicians, clinicians and researchers can develop new and more effective ways of treating those who have lost the ability to control their mimic muscles. There is a decided lack of research on human facial muscles in comparison with the number of published studies on human skeletal muscles in other regions of the body. However, the scientific and clinical importance of compiling functional, morphological, histochemical and biochemical data on this unique set of muscles has gained considerable recognition over the past decade (Stal et al., 1987, 1990; Happak et al., 1988; McLoon and Wirtschafter, 1991; Stal et al., 1994a, b). The present study combining gross anatomical, histological, immunohistochemical, and ultrastructural techniques represents one stride forward in the effort to acquire a more complete understanding of human facial muscles. The  corrugator  supercilii and the palpebral portion of the orbicularis oculi are  commissioned with very different tasks, and their architecture,  seen at gross anatomical,  histological and ultrastructural levels, reflects their different functions and the stresses they encounter when performing their functions. Several morphological differences between these two human periorbital muscles have been elucidated in this study.  121  The orbicularis oculi muscle was found to differ from the corrugator supercilii in the following ways: 1.  The orbicularis appears to be a phasic muscle with 89% type II fibers, compared to the corrugator with only 49% type II fibers.  2.  The orbicularis is composed of smaller and rounder muscle fibers than the corrugator.  3.  The orbicularis appears to be less oxidative than the corrugator, but may have developed special mitochondrial distribution strategies to compensate, in part, for its scarcity of mitochondria.  4.  The orbicularis receives a complex and arcing motor-nerve distribution along its deep surface as compared to the corrugator which has a simpler nerve supply from its superolateral edge.  5.  Muscle fibers of the orbicularis have a complex and overlapping fascicular arrangement, in contrast to those in the corrugator which have a simple, stacked fascicular organization.  6.  The orbicularis possesses a more intricate motor end-plate distribution pattern than the corrugator, but the patterns seen in the two muscles suggest that they may both be constructed using different forms of short-fiber architecture. The results of this study represent one of several steps that must be taken to better  understand the special organization and functions of human facial muscles. Despite the fact that this study is only a preliminary investigation into human periorbital muscle architecture, there are several clinical applications which can be noted. Botulinum A toxin is an extremely potent neurotoxin which acts by irreversibly binding to the presynaptic nerve terminal, thereby preventing acetylcholine release (Jankovic et al., 1990). The  122  blockade of neurotransmitter release is believed to be the result of the toxin causing the cleavage of a synaptic membrane protein involved in vesicular exocytosis (Blasi et al., 1993). Despite its high toxicity, the use of sub-lethal local injections of botulinum toxin to induce partial paralysis is gaining considerable favor with clinicians as a method for treating a variety of conditions affecting human facial muscles (Frueh et al., 1984; Dressier and Schonle, 1991; Harris et al., 1991; Taylor et al., 1991; Elston, 1992; Poewe et al., 1992; Flanders et al., 1993). Besides its clinical application for focal dystonias, there is also a growing trend toward using botulinum toxin as a means of reducing facial wrinkling caused by tonically "hyperactive" facial muscles (Foster et al., 1996). Considering the widespread use of this neurotoxin, there is relatively little known about the cytoarchitectural alterations resulting from its use in human facial muscle. Early animal studies by Duchen (1970, 1971a, b) reported that injected botulinum toxin caused collateral and noncollateral nerve sprouting along with new motor end-plate formation. Similar results were confirmed more recently by Hassan et al. (1995) in rat gastrocnemius, and by Alderson et al. (1991) in human orbicularis oculi muscle. Nonetheless, several reports from more clinically oriented studies have supported the use of botulinum toxin as a safe and effective therapy which produced no persistent histologic changes to human muscle fibers (Jankovic et al., 1990; Harris et al., 1991; Taylor et al., 1991; Elston, 1992; Flanders et al., 1993). Harris and coworkers (1991) used Gomori trichrome and NADH-tr staining to examine the histological features of the orbicularis oculi treated with botulinum toxin. They noticed no signs of fiber necrosis, denervation, inflammation or changes in fiber diameter (Harris et al., 1991). This is in contrast to the more detailed study by Hassan et al. (1995) who evaluated the myopathic changes due to botulinum A toxin in rat skeletal muscle at light microscopic and electron microscopic levels. They observed significant sarcotubular  123  and myofibrillar disorganization, along with mitochondrial abnormalities in the muscle fibers. Subsarcolemmal localization of vacuoles, dense bodies, myeloid figures and multivesicular bodies were also noted. Ulfrastjucturally, Hassan and coworkers (1995) make reference to progressive degeneration and simplification of synaptic folding patterns as well as some nerve terminal separation from underlying muscle fibers. In light of the mixed reports on the effects of botulinum toxin on human muscle, it is clear that further study is necessary. A more detailed understanding of normal nerve branching patterns and motor end-plate distributions, coupled with a clearer knowledge of the ultrastructural characteristics of facial muscle motor end-plates will be necessary if pathologic alterations are to be fully appreciated. This study may provide some baseline information for future researchers and clinicians. Perhaps the results of this study may be used by clinicians who inject botulinum A toxin into these regions so that more accurate and lower dose injections can be met with better and safer results. Furthermore, a more detailed understanding of the functional significance of motor end-plate architecture and distribution in facial muscles might allow clinicians and researchers to better understand the possible dangers which may be associated with disrupting the motor end-plate organization by repeated botulinum toxin injections. Data emanating from this study may be used by plastic and reconstructive surgeons when considering the best possible muscle match when selecting donor muscles for facial reariimation and transplantation surgeries. The results of this study may also be of use to experimental pathologists interested in morphological changes to these periorbital muscles, as well as to those studying the normal morphology of other human facial muscles and other skeletal muscles which develop from the branchiomeric mesoderm. 124  There are many avenues in the investigation of human facial muscles that require further and more detailed study. There is need for further research on the fiber organization of all human facial muscles to determine whether they are constructed of either short-fiber or long-fiber arrangements. Such a study would involve the use of several histological staining methods applied to whole, fresh cadaver muscle samples. A n interesting physiological study detailing motor unit distribution and firing patterns could be combined with high-speed photography to determine if there is any functional significance to the arcing motor-nerve supply found in the palpebral portion of the orbicularis oculi. Finally, an intensive study using a systematic light and electron microscopic search for the sensory innervation and feedback control system used by human facial muscles would be extremely helpful. Such a study would either corifirm the presence of specialized, sensory facial spindles within the substance of some of the human facial muscles or, alternatively, it may show that facial muscles use pre-existing cutaneous sensory receptors, as has been suggested others (Brown, 1984).  125  REFERENCES Adal, M.N. 1985 Variation of the T-system in different types of intrafusal muscle fibre of cat spindles. In: The muscle spindle. L A . Boyd and M . H . Gladden, eds. Stockton Press, New York, pp. 35-38. Adams, W.M. 1946 The use of the masseter, temporalis and frontalis muscles in the correction of facial paralysis. Plast. Reconst. Surg. 7:216 Alderson, K., J.B. Holds, and R.L. Anderson 1991 Botulinum-induced alteration of nerve-muscle interactions in the human orbicularis oculi following treatment for blepharospasm. Neurol. 47:1800-1805. Ammirati, M . , A. Spallone, J. Ma, M . Cheatham, and D. Becker 1993 An anatomicosurgical study of the temporal branch of the facial nerve. Neurosurg. 35:1038-1044. Anderl, H. 1973 Reconstruction of the face through cross face nerve transplantation in facial paralysis. Chir. Plast. 2:17-46. Anderl, H. 1979 Cross-face nerve transplant. Clinics Plast. Surg. 5:433-449. Andrew, R.J. and H. Ernest 1972 Evolution of facial expression: two accounts. Arno Press, New York, pp. 1-234. Ballance, C. and A.B. Duel 1932 The operative treatment of facial palsy by inroduction of nerve grafts into the fallopian canal and other infratemporal methods. Arch. Otolaryng. 75:1-70. Bardosi, A., H. Goebel, and E. Stennert 1987 The ultrastructure of normal and denervated human facial muscle. Plast. Reconst. Surg. 79:171-176. Barker, D. 1974 The morphology of muscle receptors. In: Muscle Receptors. D. Barker, C C . Hunt and A.K. Mclntyre, eds. Springer-Verlag, Berlin, pp. 1-190. Barker, D. and R.W. Banks 1986 The muscle spindle. In: Myology: basic and clinical. A . G . Engel and B.Q. Banker, eds. McGraw-Hill, New York, pp. 309-341. Baum, J. 1899 Beitrage zur kenntnis der muskelspindeln. Anat. Hefte. 73:249-305. Beatty, C H . and R.M. bocek 1970 Biochemistry of the red and white muscle. In: Physiology and biochemistry of muscle as a food. E.J. Briskey, R.G. Cassens and B.B. Marsh, eds. Univ. of Wisconsin Press, Madison, pp. 155-192. Belal, A., Jr. 1982 Structure of human muscle in facial paralysis. J. Laryng. Oto. P6:325-334.  126  Bell, C. 1821 On the nerves: Giving an account of some experiments on their structure and function which leads to a new arrangement of the systems. Phil. Trans. R. Soc. Lond. Biol. 5:398424. Blasi, J., E.R. Chapman, E. Link, T. Binz, S. Yamasaki, and P. De Camilli 1993 Botulinum neurotoxin A selectively cleaves the synaptic protein SNAP-25. Nature, 555:160-3 and 104-5. Blevins, C E . 1964 Studies on the innervation of the stapedius muscle of the cat. Anat. Rec. 149:151-112. Borodic, G.E., D. Cozzolino, R. Ferrante, A.W. Wiegner, and R.R. Young 1991 Innervation zone of orbicularis oculi muscle and implications for botulinum A toxin therapy. Ophthal. Plast. Reconst. Surg. 7:54-60. Bout, R.G. and J.L. Dubbeldam 1991 Functional morphological interpretation of the distribution of muscle spindles in the jaw muscles of the mallard (Anas platyrhynchos). J. Morph. 270:215-226. Boyd, LA. and R.S. Smith 1984 The muscle spindle. In: Peripheral neuropathy. P.J. Dyck, P.K. Thomas, E.H. Lombert, and R. Bunge, eds. W.B. Saunders, Philadelphia, Vol. 1, pp. 171-202. Boyd, I. A., P.R. Murphy, and C. Mann 1985 The effect of chain fibre 'driving' on the length sensitivity of primary sensory endings in the tenuissiums, peroneus tertius and soleus muscles. In: The muscle spindle. LA. Boyd and M . H . Gladden, eds. Stockton Press, New York, pp. 195-200. Bradley, M . M . , P.J. Lang, and B.N. Cuthbert 1993 Emotion, novelty, and the startle reflex: habituation in humans. Behav. Neurosci. 707:970-980. Broniatowski, M . , P. Lavertu, and H.M. Tucker 1990 The deep temporal nerve-orbicularis oculi muscle pedicle as a possible means for reanimation of the upper face. Ear Nose Throat J. 69:140142,144. Broniatowski, M . , S. Brundfest-Broniatowski, C R . Davies, G.B. Jacobs, H . M . Tucker, and Y. Nose 1991 Dynamic rehabilitation of the paralyzed face: balanced coupling of oral and ocular musculature from the intact side in the canine. Otolaryng-Head & Neck Surg. 105:121-133. Brooke, M.H. and W.K. Engel 1969 The histographic analysis of human muscle biopsies with respect to fiber types. 1. adult male and female. Neurol. 79:221-228. Brown, W.F. 1984 The physiological and technical basis of electromyography. Butterworth, Boston, pp. 232-454. Buchthal, F. and H. Schmalbruch 1980 Motor units of mammalian muscle. Physiol. Rev. 60:90142.  127  Buller, A.J. 1976 The motor end-plate: function. In: The peripheral nerve. D.N. Landon, ed. Chapman and Hall, Salisbury, England, pp. 495-511. Burke, R.E., D.N. Levine, M . Salcman, and P. Tsairis 1974 Motor units in cat soleus muscle: physiological, histochemical and morphological characteristics. J. Physiol. 255:503-514. Calabresi, P., P.S. Schein, and S.A. Rosenberg 1985 Medical oncology basic principles and clinical management of cancer. Macmillan, New York, pp. 331-336. Camougis, G. 1970 Nerves, Muscles, & Electricity: A n introductory manual of electrophysiology. Appelton-Century-Crofts, New York, pp. 1-48. Carlson, F.D. and D.R. Wilkie 1974 Muscle physiology. Prentice-Hall, Englewood Cliffs, pp. 2651. Christie, K . N . and P.J. Steward 1977 A cytochemical study of acid phosphatases in dystrophic hamster muscle. J. Ultrastruct Res., 55:219-226. Cipollone, L.T. 1927 Ricerche sull' anatomia normale e patologica delle termanzione nervose nei muscoli striati. Suppl. Annuali di Medicina Navale, 5:282. Clemente, C D . 1987 Anatomy: a regional atlas of the human body. Urban Schwarzenberg, Baltimore, pp. 344-346. Craig, R. 1986 The structure of the contractile filaments. In: Myology: basic and clinical. A . G . Engel and B.Q. Banker, eds. McGraw-Hill, New York, pp. 73-123. Cross, P . C and K.L. Mercer 1993 Cell and tissue ultrastructure: a functional perspective. W. H . Freeman, New York, pp. 74-155. Crouch, J.E. 1978 Functional human anatomy. Henry Kimpton, London, pp. 89-200. Cruveilhier, J. 1844 The anatomy of the human body. Harper and Brothers, New York, pp. 100334. Cullen, M.F. and F.L. Mastaglia 1982 Pathological reactions of skeletal muscle. In: Skeletal muscle pathology. F.L. Mastalglia and J. Walton, eds. Churchill Livingstone, London, pp. 88-139. Cuthbertson, A.R. 1990 Mechanism de la physionomie humaine ou analyse eiectro-physiologique de l'expresson des passions. Cambridge University Press, Cambridge, pp. 1-195. Deschenes, M.R., J. Covault, W.J. Kraemer, and C M . Maresh 1994 The neuromuscular junction: muscle fibre type differences, plasticity and adaptability to increased and decreased activity. Sports Med. 77:358-372.  128  Dittert, D.D. and A. Bardosi 1989 Fiber types in whole cross sections of human platysma. Acta Anal. 134:206-211. Doane, M . G . 1980 Interactions of eyelids and tears in corneal wetting and the dynamics of the normal human eyeblink. Am. J. Ophthal. #9:507-516. Dressier, D. and P.W. Schonle 1991 Hyperkinesias after hypoglossofacial nerve anastomosistreatment with botulinum toxin. Eur. Neurol. 57:44-46. Dubowitz, V . and M . H . Brooke 1973 Muscle biopsy: a modern approach. W.B. Saunders, London, pp. 20-74 and 383-442. Duchen, L.W. 1970 Changes in motor innervation and cholinesterase localization induced by botulinum toxin in skeletal muscle of the mouse: differences between fast and slow muscles. J. Neurol. Neurosurg. Psych. 55:40-54. Duchen, L.W. 1971a An electron microscopic study of the changes induced by botulinum toxin in the motor end-plates of slow and fast skeletal muscle fibres of the mouse. J. Neurol. Sci. 14:47-60. Duchen, L.W. 1971b Changes in the electron microscopic structure of slow and fast skeletal muscle fibres of the mouse after the local injection of botulinum toxin. J. Neurol. Sci. 74:61-74. Duchenne, G.B. 1876 Meconisme de la physionomie humaine. J.B. Bailliere et Fils, Paris, pp. 1200. Edstrom, J. and K. Torlegard 1969 Area estimation of transversely sectioned muscle fibers. Z Wiss Mikrosk, 69:166-175. Eisenberg, B.R. 1983 Quantitative ultrastructure of mammalian skeletal muscle. In: Handbook of physiology. L.D. Peachey, R.H. Adrian and S.R. Geiger, eds. Am. Physiolo. Soc, Maryland, pp. 23-72. Elston, J.S. 1992 The management of blepharospasm and hemifacial spasm. J. Neurol. 259:5-8. Engel, A . G . 1986 Quantitative morphological studies of muscle. In: Myology: basic and clinical. A . G . Engel and B.Q. Banker, eds. McGraw-Hill, New York, pp. 1045-1080. Engel, A . G . and B.Q. Banker 1986 Ultrastructural changes in diseased muscle. In: Myology: basic and clinical. A . G . Engel and B.Q. Banker, eds. McGraw Hill, New York, pp. 909-1044. Eriksson, P.O., A . Eriksson, M . Ringqvist, and L.E. Thornell 1980 The reliability of histochemical fibre typing of human necropsy muscles. Histochem. 65:193-205. Eriksson, P. and L.E. Thornell 1990 Variation in histochemical enzyme profile and diameter along human masseter intrafusal muscle fibers. Anat. Rec. 226:168-176. 129  Eriksson, P.O., G.S. Butler-Browne, and L.E. Thornell 1994 Immunohistochemical characterization of human masseter muscle spindles. Muscle. Nerve, 77:31-41. Fardeau, M . and F.M.S. Tome 1994 Ultrastructure of muscle and neuromuscular junction: an historical survery of the early French contributions. Biol. Cell, 50:1 15-117. Fawcett, D.W. 1986 Bloom and Fawcett: a textbook of histology. W. B. Saunders, Philadelphia, pp. 265-300 and 381-390. Feher, J. 1977 Myofibre abnormalities of orbicular muscle in malposition of the eyelid. Acta Morphol. Acad. Sci. Hung. 25:205-218. Flanagan, J.C. 1984 The surgical correction of essential blepharospasm. Trans. Penn. Acad. Ophthal. Otolaryng. 57:34-38. Flanders, M . , D. Chin, and D. Boghen 1993 Botulinum toxin: preferred treatment for hemifacial spasm. Eur. Neurol. 55:316-319. Flowers, R.S., G.G. Caputy, and S.S. Flowers 1993 The biomechanics of brow and frontalis function and its effect on blepharoplasty. Clinics Plast. Surg. 20:255-268. Foster, J.A., D. Barnhorst, F. Papay, P.M. Oh, and A.E. Wulc 1996 The use of botulinum A toxin to ameliorate facial kinetic frown lines. Ophthal. 705:618-622. Freilinger, G., H. Gruber, W. Happak, and U. Pechmann 1987 Surgical anatomy of the mimic muscle system and the facial nerve: importance for reconstructive and aesthetic surgery. Plast. Reconst. Surg. 50:686-690. Freilinger, G., W. Happak, G. Burggasser, and H. Gruber 1990 Histochemical mapping and fiber size analysis of mimic muscles. Plast. Reconst. Surg. 56:422-427. Frueh, B.R., D.P. Felt, T.H. Wojno, and D.C. Musch 1984 Treatment of blepharospasm with botulinum toxin. Arch. Ophthal. 702:1464-1466. Fujii, M . and N . Goto 1989 Nerve fiber analysis of the facial nerve. Ann. Oto. Rhino. Laryng. 95:732-736. Furnas, D.W. 1981 The orbicularis oculi muscle. Management in blepharoplasty. Clinics Plast. Surg. 5:687-715. Gans, C. and A.S. Gaunt 1991 Muscle architecture in relation to function. Biomech. 24:53-65. Gauthier, G.F. 1988 The muscular tissues. In: Cell and tissue biology: a textbook of histology. L. Weiss, ed. Urban Schwarzenberg, Baltimore, pp. 255-276. 130  Girlanda, P., G. Vita, C. Nicolosi, S. Milone, and C. Messina 1992 Botulinum toxin therapy: distant effects on neuromuscular transmission and autonomic nervous system. J. Neurol. Neurosurg. Psych. 55:844-845. Goldberg, B.D. and M . Rabinovitch 1988 Connective tissue. In: Cell and tissue biology: a textbook of histology. L. Weiss, ed. Urban Schwarzenberg, Baltimore, pp. 155-188. Granit, R., S. Skoglund, and S. Thesleff 1953 Activation of muscle spindles by succinylcholine and decamethionium. The effects of curare. Acta Physiol. Scand. 28:134-151. Hale, R. and B. Terence 1979 Albinus on anatomy. Watson-Guptill, New York, pp. 1-97. Hamming, N. 1983 Anatomy and embryology of the eyelids: a review with special reference to the development of divided nevi. Ped. Derm. 7:51-58. Happak, W., G. Burggasser, and H. Gruber 1988 Histochemical characteristics of human mimic muscles. J. Neurol. Sci. #5:25-35. Harii, K., K. Ohmori, and S. Torii 1976 Free gracilis muscle transplantation with microneurovascular anastomoses for the treatment of facial paralysis. Plast. Reconst. Surg. 57:133143. Harris, C P . , K. Alderson, J. Nebeker, J.B. Holds, and R.L. Anderson 1991 Histologic features of human orbicularis oculi treated with botulinum A toxin. Arch. Ophthal. 709:393-395. Hassall, A . H . 1849 The microscopic anatomy of the human body. Samuel Highley, London, pp. 120. Hassan, S.M., G.I. Jennekens, and H. Veldman 1995 Botulinum toxin-induced myopathy in the rat. Brain, 775:533-545. Hazelton, R.D. 1969 Origin and migration of facial muscle primordia and their significance in development of the cranial nervous system. J. Can. Dent. Assoc. 55:271. Hermsen, V . M . and R.F. Dreyer 1982 Ophthalmic anatomy. Primary Care: Clinics in Office Practice, 9:627-645. Hijikata, T., H. Wakisaka, and T. Yohro 1992 Architectural design, fiber-type composition, and innervation of the rat rectus abdominis muscle. Anat. Rec. 254:500-512. Hines, M . 1927 Nerve and muscle. Q. Rev. Biol. 2:149-180. Hoffman, W.Y. 1992 Reanimation of the paralyzed face. Otolaryng. Clin. N . Am. 25:649-667.  131  Hoh, J.F. 1992 Muscle fiber types and function. Curr. Opin. Rheumat. 4:801-808. Hojo, T. and K. Shinoda 1982 Anatomical variation of the m. corrugator posticus Santorini in Kyushu-Japanese. Okajimas Folia Anat. Jap., 55:391-398. Holly, F.J. 1980 Tear film physiology. Am. J. Opt. Physiol. Optics, 57:252-257. Holt, S.J. and R.M. Hicks 1966 The importance of osmiophilia in the production of stable azoindoxyl complexes of high contrast for combined enzyme cytochemistry and electron microscopy. J. Cell Biol. 77:208-212. Hudlicka, O., M . Brown, and S. Egginton 1992 Angiogenesis in skeletal and cardiac muscle. Physiol. Rev. 72:369-411. Hung, G., F. Hsu, and L. Stark 1977 Dynamics of the human eyeblink. Am. J. Opt. Physiol. Optics, 54:678-690. Hunt, C C . and R.S. Wilkinson 1985 Passive and active properties of mammalian intrafusal fibres. In: The muscle spindle. L A . Boyd and M.H. Gladden, eds. Stockton Press, New York, pp. 151-152. Hunyor, A.P. 1994 Reflexes and the eye. Austral. New Zeal. J. Ophthal. 22:153-159. Jakubowicz, M . , A. Radziemski, and A. Kedzia 1992 Histological and quantitative studies of the muscle spindles in human fetal infrahyoid muscles. Folia Morph. 57:55-59. Jankovic, J. and S. Fahn 1988 Dystonic syndromes. In: Parkinson's disease and movement disorders. J. Jankovic and E. Tolosa, eds. Urban Schwarzenberg, Baltimore, pp. 283-314. Jankovic, J., K. Schwartz, and D.T. Donovan 1990 Botulinum toxin treatment of cranial-cervical dystonia, spasmodic dysphonia, other focal dystonias and hemifacial spasm. J. Neurol. Neurosurg. Psych 55:633-639. Jones, E.G. 1983 The nervous tissue. In: Cell and tissue biology: a textbook of histology. L. Weiss, ed. Urban Schwarzenberg, Baltimore, pp. 277-352. Kadanoff, D.v. 1956 Die sensiblen nervenendigungen in der mimischen muskulatur des menschen. Z. Mikrosk. Anat. Forsch. (52:1-15. Karnovsky, M.J. 1965 A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J. Cell Biol. 27:137A-138A. Karnovsky, M.J. 1967 The ultrastructural basis of capillary permeability studied with peroxidase as a tracer. J. Cell Biol. 55:213-236. Karson, C N . 1989 Blinking. Bull, de la Soc. Beige d Ophtal., 257:443-457. 132  Katz, A.D. and P. Catalano 1987 The clinical significance of the various anastomotic branches of the facial nerve. Report of 100 patients. Arch. Otolaryn. Head Neck Surg. 7/5:959-962. Kennedy, W.R. 1970 Innervation of normal human muscle spindles. Neurol. 20:463-475. Kubota, K. and T. Masegi 1972 Muscle spindle distribution in snout musculature of the Japanese shrew-mole. Anat. Rec. 772:703-709. Kubota, K., T. Masegi, and Y. Sato 1975a Muscle spindle supply to the snout musculature in moles (Talpidae). Bull. Tokyo Med. Dent. Univ. 22:165-173. Kubota, K., T. Masegi, and Y. Sato 1975b Location of proprioceptive neurons innervating the muscle spindles of the snout muscles in the talpoides. Exp. Neurol. 45:142-151. Kubota, K., Y. Sato, T. Masegi, M . Kobayashi, and Y. Shishido 1978 Electron microscopic study of the snout muscle spindles of the mole following denervation. Anatomischer Anzeiger, 744:8196. Landon, D.N. 1982 Skeletal muscle - normal morphology, development and innervation. In: Skeletal muscle pathology. F.L. Mastalglia and J. Walton, eds. Churchill Livingstone, London, pp. 1-87. Laskawi, R., W. Damenz, P. Roggenkamper, M . Schroder, and J. Brauneis 1970 The effects of botulinum toxin on hemifacial spasm: an electrophysiologic investigation. Ear Nose Throat J. (59:704-705. Leibgott, B. 1982 The anatomical basis of dentistry. Saunders, Philadelphia, pp. 241-412. Lev-tov, A., C A . Pratt, and R.E. Burke 1988 The motor-unit population of the cat tenuissimus muscle. J. Neurophys. 59:1128-1142. Lind, L.R. 1949 The epitome of Andreas Vesalius. The MacMillan Company, New York, Loeb, G.E., C A . Pratt, and C M . Chanaud 1987 Distribution and innervation of short, interdigitated muscle fibers in parallel-fibered muscles of the cat hindlimb. J. Morph. 797:1-15. Lovell, M . , D. Sutton, and R.C. Lindeman 1977 Muscle spindles in nonhuman primate extrinsic auricular muscles. Anat. Rec. 759:519-523. Mackinnon, S.E. and A . L . Dellon 1988 Technical considerations of the latissimus dorsi muscle flap, a segmentally innervated muscle transfer for facial reanimation. Microsurg. 9:36-45. Maier, A., M . DeSantis, and E. Eldred 1975 The occurrence of muscle spindles in extraocular muscles of various vertebrates. J. Morph. 745:408-414. 133  Malone, B. and R.H. Maisel 1988 Facial nerve manual: anatomy of the facial nerve. Am. J. Oto. 9:494-504. Manning, K.A., C. Evinger, and P.A. Sibony 1990 Eyelid movements before and after botulinum therapy in patients with lid spasm. Ann. Neurol. 25:653-660. McLoon, L.K. and J. Wirtschafter 1988 Doxorubicin chemomyectomy: injection of monkey orbicularis oculi results in selective muscle injury. Invest. Ophthal. Visual Sci. 29:1854-1859. McLoon, L.K., G. Bauer, and J. Wirtschafter 1991a Quantification of muscle loss in the doxorubicin-treated orbicularis oculi of the monkey. Invest. Ophthal. Visual Sci. 52:1667-1673. McLoon, L.K. and J. Wirtschafter 1991b Regional differences in the orbicularis oculi muscle: conservation between species. J. Neurol. Sci. 104:197-202. Monkhouse, W.S. 1990 The anatomy of the facial nerve. Ear Nose Throat J. 69:677-687. Murata, F. and T. Ogata 1969 The ultrastructure of neuromuscular junctions of human red, white and intermediate striated muscle fibers. Tohoku J. Exp. Med. 99:289-301. Myint, K., A . L . Azian, and F.A. Khairul 1992 The clinical significance of the branching pattern of the facial nerve in Malaysian subjects. Med. J. Malays. 47:114-121. Nahirney, P.C. and W.K. Ovalle 1992 Immunocytochemical localization of dystrophin and neurofilament protein in muscle spindles of normal and dystrophic mice. Anat. Rec. 252:64A-65A. Nahirney, P.C. and W.K. Ovalle 1993 Distribution of dystrophin and neurofilament protein in muscle spindles of normal and mdx-dystrophic mice: an immunocytochemical study. Anat. Rec. 255:501-510. Nahirney, P . C , P.R. Dow, and W.K. Ovalle 1996 Quantitative morphology of mast cells in skeletal muscle of normal and genetically dystrophic mice. Anat. Rec. (In Press) Niccoli Filho, W. and J.T. Varandas 1988 Surgical anatomy of the facial nerve and the parotid gland. Rev. de Odonto. Da Univ. de Sao Paulo, 2:48-50. Novikoff, A.B., W. Shin, and J. Drucker 1961 Mitochondrial localization of oxidative enzymes: staining results with two-tetrazolium salts. J. Biophys. Biochem. Cytol. 9:47-61. O'Brien, B . M . 1987 Microvascular reconstructive microsurgery. Churchill Livingstone, Edinburgh, pp. 301-305.  134  O'Brien, B.M., J.D. Franklin, and W.A. Morrison 1980 Cross-facial nerve grafts and microneurovascular free muscle transfer for long established facial palsy. Brit. J. Plast. Surg. 35:202-215. Ogata, T. and M . Mori 1964 Histochemical study of oxidative enzymes in vertebrate muscles. J. Histochem. Cytochem. 72:171-182. Ogata, T. 1965 A histochemical study on the sturctural differences of motor endplate in the red, white and intermediate muscle fibers of mouse limb muscle. Acta Med. Okayama, 79:149-153. Ogata, T. and F. Murata 1969 Cytological features of three fiber types in human striated muscle. Tohoku J. Exp. Med. 99:225-245. Ogata, T. and Y. Yamasaki 1985 Scanning electron-microscopic studies on the three-dimensional structure of sarcoplasmic reticulum in the mammalian red, white and intermediate muscle fibers. Cel Tiss.Res. 242:461-467. Ogata, T. 1988a Structure of motor endplates in the different fiber types of vertebrate skeletal muscle. Arch. Histol. Cytol. 57:385-424. Ogata, T. 1988b Morphological and cytochemical features of fiber types in vertebrate skeletal muscle. Crit. Rev. Anat. Cell Biol. 7:229-275. Olson, C.B. and C.P.J. Swett 1966 A functional and histochemical characterization of motor units in a heterogeneous muscle (flexor digitorum longus) of the cat. J. Comp. Neurol. 725:475-498. Ovalle, W.K. and R.S. Smith 1972 Histochemical identification of three types of intrafusal muscle fibres in the cat and monkey based on the myosin ATPase reaction. Can. J. of Physiol. Pharm., 50:195-202. Ovalle, W.K. and P.R. Dow 1983 Comparative ulfrastructure of the inner capsule of the muscle spindle and the tendon organ. Am. J. Anat. 766:343-357. Ovalle, W.K. and P.R. Dow 1985 Morphological aspects of the muscle spindle capsule and its functional significance. In: The muscle spindle. LA. Boyd and M . H . Gladden, eds. Stockton Press, New York, pp. 23-28. Ovalle, W.K. and P.R. Dow 1986 Alterations in muscle spindle morphology in advanced stages of murine muscular dystrophy. Anat. Rec. 276:111-126. Ovalle, W.K. 1987 The human muscle-tendon junction: a morphological study during normal growth and at maturity. Anat. Embryol. 7 76:281 -294. Ozcan, G., S. Shenaq, B. Mirabi, and M . Spira 1993 Nerve regeneration in a bony bed: vascularized versus nonvascularized nerve grafts. Plast. Reconst. Surg. 97:1322-1331. 135  Pachter, B.R., J. Davidowitz, and G.M. Breinin 1976 Light and electron microscopic serial analysis of mouse extraocular muscle: morphology innervation and topographical organization of component fiber populations. Tissue Cell, 5:547-560. Padykula, H.A. 1988 Histochemistry and cytochemistry. In: Cell and tissue biology: a textbook of histology. L. Weiss, ed. Urban Schwarzenberg, Baltimore, pp. 93-112. Page, S.G. and H.E. Huxley 1963 Filament lengths in striated muscle. J. Cell Biol. 79:369-376. Park-Matsumoto, Y.C., V . Askanas, and W.K. Engel 1992 The influence of muscle contractile activity versus neural factors on morphologic properties of innervated cultured human muscle. J. Neurocyt. 27:329-340. Patrinely, J.R. and R.L. Anderson 1988 Anatomy of the orbicularis oculi and other facial muscles. Adv. Neurol. 49:15-23. Patten, B . M . 1968 Human embryology. McGraw-Hill, New York, pp. 250-305. Pepe, F.A. 1971 Structure of the myosin filament of striated muscle. Prog. Biophys. Molec. Biol. 22:77-96. Pitanguy, I. 1979 Section of the frontalis-prOcerus-corrugator aponeurosis in the correction of frontal and glabellar wrinkles. Ann. Plast. Surg. 2:422-427. Podvinec, M . and C R . Pfaltz 1976 Studies on the anatomy of the facial nerve. Acta Oto-laryng. 57:173-177. Poewe, W., L. Schelosky, B. Kleedorfer, F. Heinen, M . Wagner, and G. Deuschl 1992 Treatment of spasmodic torticollis with local injections of botulinum toxin. J. Neurol. 239:21-25. Proctor, B. 1991 The anatomy of the facial nerve. Otolaryng. Clin. N . Am. 24:479-504. Reynolds, E.S. 1963 The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell Biol. 77:208-212. Roden, R.L., S.P. Donahue, G.A. Schwartz, J.G. Wood, and A.W. English 1991 200 kD neurofilament protein and synapse elimination in the rat soleus muscle. Synapse, 9:239-243. Rokx, J.T.M., J.D. Van Willigen, and H.W.B. Jansen 1984 Muscle fibre types and muscle spindles in the jaw musculature of the rat. Arch. Oral Biol. 29:25-31. Rosenwasser, R.H., E. Liebman, D.F. Jimenez, W.A. Buchheit, and D.W. Andrews 1991 Facial reanimation after facial nerve injury. Neurosurg. 29:568-574.  136  Rowlerson, A., F. Mascarello, D. Barker, and H. Saed 1988 Muscle-spindle distribution in relation to the fibre-type composition of masseter in mammals. J. Anat. 161:37-60. Rowlerson, A . 1994 An outline of fibre types in vertebrate skeletal muscle: Histochemical identification and myosin isoforms. Basic App. Myol. 4:333-352. Ruskell, G.L. and J. Wilson 1983 Spiral nerve endings and dapple motor end plates in monkey extra-ocular muscles. J. Anat. 756:85-95. Ruskell, G.L. 1985 Facial nerve distribution to the eye. Am. J. Opt. Physiol. Optics, 62:793-798. Safran, A.B. 1989 Eyelid resting position and movements. Bull, de la Soc. Beige d Ophtal., 257:459-484. Sahgal, V., V . Subramani, S. Sahgal, and H. Kochar 1985 Morphology and morphometry of human muscle spindles. In: The muscle spindle. L A . Boyd and M . H . Gladden, eds. Stockton Press, New York, pp. 107-114. Salmons, S., D.R. Gale, and F.A. Sreter 1978 Ultrastructural aspects of the transformation of muscle fibre type by long term stimulation: changes in Z discs and mitochondria. J. Anat. 727:1731. Sarnat, H.B. 1983 Muscle Pathology & Histochemistry. Amer. Soc. Clin. Pathol. Press, Chicago,pp. 26-119. Sataloff, R.T. 1990 Embryology of the facial nerve and its clinical applications. Laryngoscope, 700:969-984. Saunders, J.B. and C D . O'Malley 1950 The illustrations from the work of Andreas Vesalius Of Brussels. World Publishing Co., Cleveland Ohio, pp. 1-57. Scaramella, L.F. 1971 L'anastomosi trai due nervi faciali. Arch. Otolaryng. 82:209. Schiaffino, S., E. Severin, M . Cautini, and S. Sartore 1977 Tubular aggregates induced by anoxia in isolated rat skeletal muscle. Lab. Invest. 57:223-232. Schmidtke, K. and J.A. Buttner-Ennever 1992 Nervous control of eyelid function. A review of clinical, experimental and pathological data. Brain, 775 7:227-247. Schwarting, M . , E. Schroder, H.H. Stennert, and H.H. Goebel 1982 Enzyme histochemical and histographic data on normal human facial muscles. J. Oto-Rhino-Laryng. Rel. Spec. 44:51-59. Scott, A.B., A . L . Rosenbaum, and C C . Collins 1973 Pharmacologic weakening of extraocular muscles. Invest. Ophthal. 72:924-927.  137  Scott, A.B. 1980 Botulinum toxin injection into extraocular muscles as an alternative to strabismus surgery. Ophthal. 70:1044-1049. Scott, A.B. 1981 Botulinum toxin injection of the eye muscles to correct strabismus. Trans. Am. Ophthal. Soc. 79:734-770. Scott, S.H., D.B. Thomson, F.J.R. Richmond, and G.E. Loeb 1992 Neuromuscular organization of feline anterior sartorius: intramusclular length changes and complex length-tension relationships during stimulation of individual nerve branches. J. Morph. 273:171-183. Segal, R.L. 1992 Neuromuscular compartments in the human biceps brachii muscle. Neurosci. Lett. 740:98-102. Siegel, R. 1984 Surgical anatomy of the upper eyelid fascia. Ann. Plast. Surg. 73:263-273. Siegel, R.J. 1993 Essential anatomy for contemporary upper lid blepharoplasty. Clin. Plast. Surg. 20:209-212. Silverstein, L.D. and F.R. Graham 1978 Eyeblink E M G : a miniature eyelid electrode for recording from orbicularis oculi. Psychophysiol. 75:377-379. Simionescu, N . and M . Simionescu 1988 Cardiovascular system. In: Cell and tissue biology: a textbook of histology. L. Weiss, ed. Urban Schwarzenberg, Baltimore, pp. 353-400. Slater, C.R., P.R. Lyons, T.J. Walls, and P.R.W. Fawcett 1992 Structure and function of neuromuscular junctions in the vastus lateralis of man. Brain, 775:451-478. Smith, J.J. and C S . Breathnach 1990 Functions of the seventh cranial nerve. Ear Nose Throat J. 69:688-695. Smith, O . C 1926 Private communication to M . Hines. (1927) Sperber, H.G. 1981 Craniofacial embryology. Wright, Bristol,England, pp. 35-87. Stal, P., P.O. Eriksson, A . Eriksson, and L.E. Thornell 1987 Enzyme-histochemical differences in fibre-type between the human major and minor zygomatic and the first dorsal interosseus muscles. Arch. Oral Biol. 32:833-841. Stal, P., P.O. Eriksson, A. Eriksson, and L.E. Thornell 1990 Enzyme-histochemical and morphological characteristics of muscle fibre types in the human buccinator and orbicularis oris. Arch. Oral Biol. 35:449-458. Stal, P. 1994a Characterization of human oro-facial and masticatory muscles with respect to fibre types, myosins and capillaries. Morphological, enzyme-histochemical, immuno-histochemical and biochemical investigations. Swed. Dent. J. Supp. 95:1-55. 138  Stal, P., P.O. Eriksson, S. Schiaffino, G.S. Butler-Browne, and L.E. Thornell 1994b Differences in myosin composition between human oro-facial, masticatory and limb muscles: enzyme-, immunohisto- and biochemical studies. J. Mus. Res. Cell Mot. 75:517-534. Stennert, E., H.H. Goebel, M . Schwarting, and E. Schroder 1985 Morphology of human mimic musculature. In: Symposium on the facial nerve. M . Portmann, ed. Masson, Paris, pp. 177-180. Strum, J.M. and E.C.B. Hall-Craggs 1982 A method demonstrating motor endplates for light and electron microscopy. J. Neurosci. Meth. 6:305-309. Stuart, D.G., C.G. Mosher, and R.L. Gerlach 1972 Properties and central connections of golgi tendon organs with special reference to locolmotion. In: Research in muscle development and the muscle spindle. B.Q. Banker, R.J. Przybylski, J.P. Van Der Meulen and M . Victor, eds. Excerpta Medica, London, pp. 437-464. Tallmadge, M . 1968 Galen on the usefulness of the parts of the body. Cornell University Press, Ithica, New York, pp. 1-321. Taylor, J.D.N., S.P. Kraft, M.S. Kazdan, M . Flanders, W. Cadera, and R.B. Orton 1991 Treatment of blepharospasm and hemifacial spasm with bomlinum A toxin: a Canadian multicentre study. Can. J. Ophthal. 26:133-138. Thomson, D.B., S.H. Scott, and F.J.R. Richmond 1991 Neuromuscular organization of feline anterior sartorius: asymmetric distribution of motor units. J. Morph. 270:147-162. Trotter, J.A. 1990 Interfiber tension transmission in series-fibered muscles of the cat hindlimb. J. Morph. 206:351-361. Trotter, J.A. 1991 Dynamic shape of tapered skeletal muscle fibers. J. Morph. 207:211-223. Trotter, J.A. 1993 Functional morphology of force transmission in skeletal muscle. Acta Anat. 746:205-222. Uehara, Y., G.R. Campbell, and G. Burnstock 1976 Muscle and its innervation: an atlas of fine structure. Edward Arnold, London, England, pp. 55-442. Ushiro, K., M . Yanagida, T. Kumazawa, Y. Yamanouchi, and T. Katoh 1993 M R imaging of vascular compression in hemifacial spasm. Acta Oto-laryng. Suppl. 500:54-57. Vestal, K.P., J.E. Rathbun, and S.R. Seiff 1994 Anatomy of the terminal nerves in the upper eyelid. Ophthal. Plast. Reconst. Surg. 70:1-5. Vidic, B. 1978 The anatomy and development of the facial nerve. Ear Nose Throat J. 57:236-242.  139  Virchow, H. 1908 Gesichtsmuskeln und Gesichtsausdruck. Arch. Anat. Physiol. 27:321-330. Williams, P.R., R. Warwick, M . Dyson, and L.H. Bannister 1989 Gray's Anatomy. Churchill Livingstone, New York, pp. 564-1228. Wilson-Pauwels, L., E.J. Akesson, and P.A. Stewart 1988 Cranial nerves - anatomy and clinical comments. Decker, Toronto, pp. 49-114. Wissig, S.L. 1964 The transport by vesicles of protein across the endothelium of muscle capillaries. Anat. Rec. 745:411-417. Zide, B . M . 1981 Anatomy of the eyelids. Clinics Plast. Surg. 5:623-634.  140  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items