Purification of uterine myosin and synthetic filament formation

Rabbit uterine myosin was purified by DEAE-Sephadex column chromatography. The purified myosin was shown to be free from contamination with actin or other impurities by sodium dodecyl sulfate gel electrophoresis. It was also shown to have two light chains with molecular weights of 22,000 and 17,000. The C protein normally found in crude, rabbit skeletal muscle myosin was not found in crude, rabbit uterine myosin.On reducing the ionic strength of a solution of rabbit uterine myosin, the myosin molecules first aggregated to form short, tapered bipolar filaments with bare zones. These were observed in the presence or absence of 10 mm-magnesium. These filaments ranged in length from 0.3 μm to 0.6 μm. On reaching 0.1 m-KCl, and only if 10 mm-magnesium was present, the filaments grew in length by linear overlap of the tapered bipolar filaments and obliteration of the bare zone regions. These filaments ranged in length from 0.7 μm to 1.2 μm.     

Human platelet myosin. II. In vitro assembly and structure of myosin filaments

We have used electron microscopy and solubility measurements to investigate the assembly and structure of purified human platelet myosin and myosin rod into filaments. In buffers with ionic strengths of less than 0.3 M, platelet myosin forms filaments which are remarkable for their small size, being only 320 nm long and 10-11 nm wide in the center of the bare zone. The dimensions of these filaments are not affected greatly by variation of the pH between 7 and 8, variation of the ionic strength between 0.05 and 0.2 M, the presence or absence of 1 mM Mg++ or ATP, or variation of the myosin concentration between 0.05 and 0.7 mg/ml. In 1 mM Ca++ and at pH 6.5 the filaments grow slightly larger. More than 90% of purified platelet myosin molecules assemble into filaments in 0.1 M KC1 at pH 7. Purified preparations of the tail fragment of platelet myosin also form filaments. These filaments are slightly larger than myosin filaments formed under the same conditions, indicating that the size of the myosin filaments may be influenced by some interaction between the head and tail portions of myosin molecules. Calculations based on the size and shape of the myosin filaments, the dimensions of the myosin molecule and analysis of the bare zone reveal that the synthetic platelet myosin filaments consists of 28 myosin molecules arranged in a bipolar array with the heads of two myosin molecules projecting from the backbone of the filament at 14-15 nm intervals. The heads appear to be loosely attached to the backbone by a flexible portion of the myosin tail. Given the concentration of myosin in platelets and the number of myosin molecules per filament, very few of these thin myosin filaments should be present in a thin section of a platelet, even if all of the myosin molecules are aggregated into filaments.     

General model of myosin filament structure. 3. Molecular packing arrangements in myosin filaments

Following the original proposals about myosin filament structure put forward as part of a general myosin filament model (Squire, 1971, 1972) it is here shown what the most likely molecular packing arrangements within the backbones of certain myosin filaments would be assuming that the model is correct. That this is so is already indicated by recently published experimental results which have confirmed several predictions of the model (Bullard and Reedy, 1972; Reedy et al., 1972; Tregear and Squire, 1973).The starting point in the analysis of the myosin packing arrangements is the model for the myosin ribbons in vertebrate smooth muscle proposed by Small &; Squire (1972). It is shown that there is only one reasonable type of packing arrangement for the rod portions of the myosin molecules which will account for the known structure of the ribbons and which is consistent with the known properties of myosin molecules. The dominant interactions in this packing scheme are between parallel myosin molecules which are related by axial shifts of 430 Å and 720 Å. In this analysis the myosin rods are treated as uniform rods of electron density and only the general features of two-strand coiled-coil molecules are considered.Since the general myosin filament model is based on the assumption that the structures of different types of myosin filament must be closely related, the packing scheme derived for the myosin ribbons is used to deduce the structures of the main parts (excluding the bare zones) of the myosin filaments in a variety of muscles. It is shown in each case that there is only one packing scheme consistent with all the available data on these filaments and that in each filament type exactly the same interactions between myosin rods are involved. In other words the myosin-myosin interactions involved in filament formation are specific, they involve molecular shifts of either 430 Å or 720 Å, and are virtually identical in all the different myosin filaments which have been considered. Apart from the myosin ribbons, these are the filaments in vertebrate skeletal muscle, insect flight muscle and certain molluscan muscles.In the case of the thick filaments in vertebrate skeletal muscle the form of the myosin packing arrangement in the bare zone is considered and a packing scheme proposed which involves antiparallel overlaps between myosin rods of 1300 Å and 430 Å. It is shown that this scheme readily explains the triangular profiles of the myosin filaments in the bare zone (Pepe, 1967, 1971) and many other observations on the form of these myosin filaments.Finally it is shown that the cores of several different myosin filaments, assuming they contain protein, may consist of different arrangements of one or other of two types of core subfilament.     

Structure of the cross-striated adductor muscle of the scallop

The structure of the cross-striated adductor muscle of the scallop has been studied by electron microscopy and X-ray diffraction using living relaxed, glycerol-extracted (rigor), fixed and dried muscles. The thick filaments are arranged in a hexagonal lattice whose size varies with sarcomere length so as to maintain a constant lattice volume. In the overlap region there are approximately 12 thin filaments about each thick filament and these are arranged in a partially disordered lattice similar to that found in other invertebrate muscles, giving a thin-to-thick filament ratio in this region of 6:1.The thin filaments, which contain actin and tropomyosin, are about 1 μm long and the actin subunits are arranged on a helix of pitch 2 × 38.5 nm. The thick filaments, which contain myosin and paramyosin, are about 1.76 μm long and have a backbone diameter of about 21 nm. We propose that these filaments have a core of paramyosin about 6 nm in diameter, around which the myosin molecules pack. In living relaxed muscle, the projecting myosin heads are symmetrically arranged. The data are consistent with a six-stranded helix, each strand having a pitch of 290 nm. The projections along the strands each correspond to the heads of one or two myosin molecules and occur at alternating intervals of 13 and 16 nm. In rigor muscle these projections move away from the backbone and attach to the thin filaments.In both living and dried muscle, alternate planes of thick filaments are staggered longitudinally relative to each other by about 7.2 nm. This gives rise to a body-centred orthorhombic lattice with a unit cell twice the volume of the basic filament lattice.     

Three-Dimensional Structure of the M-region (Bare Zone) of Vertebrate Striated Muscle Myosin Filaments by Single-Particle Analysis

The rods of anti-parallel myosin molecules overlap at the centre of bipolar myosin filaments to produce an M-region (bare zone) that is free of myosin heads. Beyond the M-region edges, myosin molecules aggregate in a parallel fashion to yield the bridge regions of the myosin filaments. Adjacent myosin filaments in striated muscle A-bands are cross-linked by the M-band. Vertebrate striated muscle myosin filaments have a 3-fold rotational symmetry around their long axes. In addition, at the centre of the M-region, there are three 2-fold axes perpendicular to the filament long axis, giving the whole filament dihedral 32-point group symmetry. Here we describe the three-dimensional structure obtained by a single-particle analysis of the M-region of myosin filaments from goldfish skeletal muscle under relaxing conditions and as viewed in negative stain. This is the first single-particle reconstruction of isolated M-regions. The resulting three-dimensional reconstruction reveals details to about 55 Å resolution of the density distribution in the five main nonmyosin densities in the M-band (M6′, M4′, M1, M4 and M6) and in the myosin head crowns (P1, P2 and P3) at the M-region edges. The outermost crowns in the reconstruction were identified specifically by their close similarity to the corresponding crown levels in our previously published bridge region reconstructions. The packing of myosin molecules into the M-region structure is discussed, and some unidentified densities are highlighted.     

Control of filament length by the regulatory light chains in skeletal and cardiac myosins

The possible role of the regulatory light chains (LC2) in in vitro assembly of rabbit skeletal and dog cardiac myosins was examined by formation of minifilaments and synthetic thick filaments. After LC2 was removed, the resulting myosin preparations exhibited little aggregation in 0.5 M KCl and 0.05 M potassium phosphate (pH 6.5). Minifilaments migrated as a single, hypersharp peak during sedimentation velocity, but electron microscopic analysis revealed a more destabilized structure for LC2-deficient minifilaments. Thick filaments were formed in buffers containing 0.15 M KCl and the following: 20 mM imidazole; 20 mM imidazole, 5 mM ATP; or 20 mM imidazole, 5 mM ATP, and 5 mM MgCl2, all at pH 7.0. Skeletal and cardiac myosin filaments formed in imidazole buffer alone were bipolar, tapered at both ends, and about 1.6 micron long. Removal of LC2 resulted in the formation of shorter thick filaments (1.2 micron long). This effect could be reversed by reassociation with LC2. Inclusion of ATP in the buffer disrupted the filament structure, resulting in irregular, short filaments (less than 0.6 micron); addition of both ATP and MgCl2 largely reversed the effects of ATP alone. In cardiac myosin filaments, the bare zone diameter increased from 16 nm as measured in control and LC2-recombined samples to 20 nm in LC2-deficient myosin assemblies. These results implicate LC2 in an active role in controlling synthetic thick filament length in both skeletal and cardiac muscles.     

Distribution of mass within native thick filaments of vertebrate skeletal muscle

The distribution of mass within the vertebrate skeletal thick filament has been determined by scanning transmission electron microscopy. Thick and thin filaments from fresh rabbit muscle were mixed with tobacco mosaic virus (TMV), fixed with formaldehyde, dried onto thin carbon films and viewed in a computer-linked microscope. Electron scattering data from both TMV and thick filaments were analysed with reference to the long axis of the particles so that the distribution of mass within the particles could be determined. While TMV appeared to be a uniform rod at the resolution employed (4.3 nm), the thick filament was clearly differentiated along its length. M-line remnants at the centre of the filament were flanked by regions of low mass per unit length, corresponding to the bare zone of the filament, and then by the more massive cross-bridge regions. The mass per unit length was approximately constant through most of the cross-bridge zone and declined at the filament tips, in a manner consistent with a constant number of myosin molecules per 14.3 nm interval (crown) throughout the cross-bridge zone. Fourier analysis of the data failed to detect the expected 43 nm periodicity of C-protein. The total mass of the thick filament was 184 Mdalton (s.e.m., 1.6 X 10(6); n = 70). The mass of adhering M-line proteins was highly variable but, on average, was about 4 Mdalton. The total mass of the filament and the mass distribution in the cross-bridge zone are consistent with three myosin molecules per crown.     

The influence of pressure on the self-assembly of the thick filament from the myosin of vertebrate skeletal muscle.

The thick-filament-monomeric-myosin equilibrium was prepared from pure myosin at pH 8.1. The application of hydrostatic pressure to the self-assembly equilibrium resulted in a biphasic dissociation curve in which a linear decrease in turbidity (a measure of weight added to or lost from the filament) was followed by a transition to a second pressure-insensitive phase. The first phase represents the effect of hydrostatic pressure on the growth or propagation phase of filament assembly. Here is was shown that hydrostatic pressure served to shorten the filaments in concert towards the bare zone whilst maintaining the narrow length distribution seen at atmospheric pressure; the filament concentration remained constant during the experiment. A more precise definition of the delta-v for the assembly of monomer into filament was obtained than had hitherto been possible. The positioning of the bare zone at the centre of the filament seems to be one of the more obvious functions of the length-regulation mechanism. It also appears that all the basic structural elements of the native thick filament are potentially present in the pH 8.1 homopolymer; its length can be increased by physiological concentrations of MgCl2 and decreased by pressure. The monodisperse native filament could then be formed by a fine tuning of the basic length-regulation mechanism of the homopolymer by the co-polymerization of the other thick-filament proteins.     

The myosin filament: V. Intermediate voltage electron microscopy and optical diffraction studies of the substructure

Using a 200 kV electron microscope (JEM 200 A), thick (up to 0.4 μm) crosssections of the myosin filaments of vertebrate striated muscle were studied. It was found that: (a) with increasing section thickness the cross-sectional profiles of the shaft of the filament were increasingly more triangular and in sections 0.4 μm thick each apex of the triangle was clearly blunted. This unique cross-sectional profile is predicted by the model proposed by Pepe (1966,1967) in which 12 parallel structural units are packed to form a triangular profile with a structural unit missing at each apex of the triangle. (b) With increasing section thickness the substructure of the myosin filament was enhanced, with the best substructure visible in sections 0.2 μm to 0.3 μm thick. This strongly supports parallel alignment of structural units in the shaft of the filament as proposed by Pepe (1966,1967). (c) The substructure spacing, determined by optical diffraction from electron micrographs of cross-sections of individual myosin filaments or groups of filaments is about 4 nm. (d) The different optical diffraction patterns observed from individual myosin filaments can be explained if the projection of each structural unit in the plane of the section has an elongated profile. With a substructure spacing of 4 nm an elongated cross-sectional profile could be produced by having two myosin molecules per structural unit. Models drawn with two myosin molecules per structural unit in the model proposed by Pepe (1966,1967) gave optical diffraction patterns similar to those observed from individual filaments. (e) The different optical diffraction patterns observed from individual myosin filaments can be explained if the elongated profiles in each structural unit are similarly oriented but with the orientation changing along the length of the filament. The change in orientation per unit length of the filament must be small enough to maintain an elongated profile for the projection of the structural unit in the plane of the sections 0.3 μm thick. All of these observations and conclusions strongly support the model for the myosin filament proposed by Pepe (1966,1967).     

Equilibrium of the actin-tropomyosin interaction

The actin-tropornyosin interaction was studied by means of light-scattering. The experimental data were analysed on the basis of the model of co-operative binding of large ligands to a one-dimensional lattice with overlapping binding sites. The affinity of tropomyosin for actin filaments was dependent on the magnesium concentration. A fivefold increase of the magnesium concentration (from 0·5 mm to 2·5 mm) enhanced the equilibrium constant twofold (from 700 to 1600 m^?1) for the isolated binding of tropomyosin molecules to actin filaments. At low magnesium concentrations (0·5 mm), tropomyosin molecules were bound to isolated binding sites on an actin filament about 600 times more weakly than to contiguous binding sites. At increased magnesium concentrations (2·5 mm), the tendency of tropomyosin to bind contiguously increased twofold. Due to the co-operative nature of the actin-tropomyosin interaction, a small change in the magnesium concentration may cause a great change of the structural organisation of the complex. A small enhancement of the magnesium concentration (from 1 mm to 1·5 mm) caused bare filaments to be covered almost completely with tropomyosin. The length of tropomyosin clusters and the number of gaps on actin filaments depended strongly on the magnesium concentration. From the values of the experimentally determined equilibrium constants, it was concluded that the end-to-end interaction of tropomyosin was not strong enough to bring about all-or-none behaviour, where actin filaments of physiological length (~1000 nm) are either completely covered with or completely free of tropomyosin.     

Cold-sensitive regulatory mutants of simian virus 40

A preparation of short synthetic myosin filaments (minifilaments) in the absence of other myosin forms is reported. Myosin minifilaments have been prepared by dialysing myosin from vertebrate striated muscle into 10 mm-citrate/Tris buffer (pH 8.0 at 4 °C) containing no other salt. These polymers of myosin are very stable and show little tendency to aggregate or dissociate in the original solvent. Sedimentation velocity, diffusion and viscosity measurements indicate that the minifilaments are composed of 16 to 18 molecules. Examination of electron micrographs reveals that the bare central region of minifilaments extends over 1600 to 1800 Å and the entire particles are about 3000 Å long with a diameter of 80 Å across the smooth region. They have the appearance of short bipolar filaments (Huxley, 1963). In solution the minifilaments are homogeneous in terms of size distribution and exhibit normal MgATPase and CaATPase activities. When examined in the ultracentrifuge, the minifilaments sediment in the form of a hypersharp peak (or bar) with a sedimentation coefficient independent of rotor speed. The minifilaments can be dissociated by ATP, hardly by MgATP; whereas KCl (between 0.04 and 0.2 m) induces further polymerization. It is suggested that the minifilaments are an intermediate in the assembly of myosin filaments.     

Characterization of a non-indexible equatorial x-ray reflection from frog sartorius muscle.

X-ray equatorial reflections from frog sartorius muscle were studied using a position sensitive detector. A weak reflection appeared between the 10 and 11 peaks which did not index on the hexagonal filament lattice. This reflection, first reported by Elliott et al. (1967), was further characterized. The spacing of the reflection varied in direct proportion to that of the 10 peaks for sarcomere lengths between 2·0 μm and 3·0 μm. Its intensity appeared relatively insensitive to length changes. Optical diffraction patterns from electron micrographs of oblique sections through muscle gave ratios for the spacings of the myosin filaments and the Z-disc lattice that correlated very closely with the X-ray results. It is suggested that the Z-disc structure is the major source of this nonindexible reflection.     

Electron microscopy and X-ray diffraction of a hexagonal net of light meromyosin

Light meromyosin, prepared by brief digestion of rabbit myosin, forms at low ionic strength tactoids with a 43 nm periodicity and open nets. These nets, when negatively stained, show strands intersecting at intervals of ~ 60 nm and at an angle of 120 ° to form hexagonal arrays (Huxley, 1963).By slow dialysis of light meromyosin from 0.35 to 0.1 m-KCl we have obtained large, highly ordered hexagonal nets, which we have subjected to structural analysis by electron microscopy of both negatively stained and sectioned material, and by X-ray diffraction. The net is a three-dimensional crystalline array whose overall shape is that of an oblate ellipsoid. Viewed down the short axis, a hexagonal appearance is seen. Analysis of other views of the net suggests that it has a simple layered structure, each layer consisting of a set of parallel strands of diameter about 10 nm. Each strand crosses over those in neighbouring layers at intervals of 64.4 nm and at an angle of 120 °, so that in the whole structure there is a 3-fold screw axis through each node of the net. A model for a strand is described in which light meromyosin molecules, ~ 100 nm in length, are arranged in an anti-parallel manner, each molecule having one end at a node of the lattice. If this end corresponds to the free end of the myosin tail, one of the interactions is similar to that found in type 1 segments of myosin rod (Harrison et al., 1971). The molecular packing within strands may be related to the packing of myosin tails in the bare zone of muscle thick filaments.     

Substructure and accessory proteins in scallop myosin filaments

Native myosin filaments from scallop striated muscle fray into subfilaments of approximately 100-A diameter when exposed to solutions of low ionic strength. The number of subfilaments appears to be five to seven (close to the sevenfold rotational symmetry of the native filament), and the subfilaments probably coil around one another. Synthetic filaments assembled from purified scallop myosin at roughly physiological ionic strength have diameters similar to those of native filaments, but are much longer. They too can be frayed into subfilaments at low ionic strength. Synthetic filaments share what may be an important regulatory property with native filaments: an order-disorder transition in the helical arrangement of myosin cross-bridges that is induced on activation by calcium, removal of nucleotide, or modification of a myosin head sulfhydryl. Some native filaments from scallop striated muscle carry short "end filaments" protruding from their tips, comparable to the structures associated with vertebrate striated muscle myosin filaments. Gell electrophoresis of scallop muscle homogenates reveals the presence of high molecular weight proteins that may include the invertebrate counterpart of titin, a component of the vertebrate end filament. Although the myosin molecule itself may contain much of the information required to direct its assembly, other factors acting in vivo, including interactions with accessory proteins, probably contribute to the assembly of a precisely defined thick filament during myofibrillogenesis.     

Intermolecular versus intramolecular interactions of Dictyostelium myosin: possible regulation by heavy chain phosphorylation

Dictyostelium myosin has been examined under conditions that reveal intramolecular and intermolecular interactions that may be important in the process of assembly and its regulation. Rotary shadowed myosin molecules exhibit primarily two configurations under these conditions: straight parallel dimers and folded monomers. All of the monomers bend in a specific region of the 1860-A-long tail that is 1200 A from the head-tail junction. Molecules in parallel dimers are staggered by 140 A, which is a periodicity in the packing of myosin molecules originally observed in native thick filaments of muscle. The most common region for interaction in the dimers is a segment of the tail about 200-A-long, extending from 900 to 1100 A from the head-tail junction. Parallel dimers form tetramers by way of antiparallel interactions in their tail regions with overlaps in multiples of 140 A. The folded configuration of the myosin molecules is promoted by phosphorylation of the heavy chain by Dictyostelium myosin heavy chain kinase. It appears that the bent monomers are excluded from filaments formed upon addition of salt while the dimeric molecules assemble. These results may provide the structural basis for primary steps in myosin filament assembly and its regulation by heavy chain phosphorylation.     

Phosphorylation of tropomyosin extends cooperative binding of myosin beyond a single regulatory unit

Tropomyosin (Tm) is one of the major phosphoproteins comprising the thin filament of muscle. However, the specific role of Tm phosphorylation in modulating the mechanics of actomyosin interaction has not been determined. Here we show that Tm phosphorylation is necessary for long-range cooperative activation of myosin binding. We used a novel optical trapping assay to measure the isometric stall force of an ensemble of myosin molecules moving actin filaments reconstituted with either natively phosphorylated or dephosphorylated Tm. The data show that the thin filament is cooperatively activated by myosin across regulatory units when Tm is phosphorylated. When Tm is dephosphorylated, this "long-range" cooperative activation is lost and the filament behaves identically to bare actin filaments. However, these effects are not due to dissociation of dephosphorylated Tm from the reconstituted thin filament. The data suggest that end-to-end interactions of adjacent Tm molecules are strengthened when Tm is phosphorylated, and that phosphorylation is thus essential for long range cooperative activation along the thin filament.     

The myosin filament: XI. Filament assembly

The critical parameters required for the assembly of myosin filaments with a length distribution comparable to that for native myosin filaments were examined. It was found that: Two steps are required in the dilution of a myosin solution from 0.6M KCl to 0.15M KCl. In Step I the KCl concentration is reduced from 0.6 to 0.3M KCl and in Step II from 0.3 to 0.15M KCl. The rate of change of KCl required for Step I is different than that required for Step II. Increasing the total time of dilution in either Step I or II alone leads to an increase in length and a broadening of the length distribution. In Step I assembly of myosin molecules into nonsedimentable units occurs. These may be the basic units from which the filaments are assembled in Step II. Rapid dilution in Step I alone has no effect on the length distribution obtained at 0.15M KCl, but rapid dilution in Step II alone leads to short filaments (about 0.6 micron). Increasing the time of dilution in Step II alone to 3 hrs or 6 hrs gives a bimodal distribution in lengths with one peak at about 0.8 micron and the other at about 2.2 microns. The length distribution obtained at 0.15M KCl is not critically dependent on information contained in the portion of the filament previously assembled in Step II, but is critically dependent on the rate of change of KCl concentration during the assembly of the rest of the filament.     

Thermal activation energy for bidirectional movement of actin along bipolar tracks of myosin filaments

Previous in vitro motility assays using bipolar myosin thick filaments demonstrated that actin filaments were capable of moving in both directions along the myosin filament tracks. The movements; however, were slower in the direction leading away from the central bare zone than towards it. To understand the mechanism underlying these different direction-dependent motilities, we have examined the effects of temperature on the velocities of the bidirectional movements along reconstituted myosin filaments. Activation energies of the movements were determined by Arrhenius plots at high and low concentrations of ATP. As a result, the thermal activation energy of the movement away from the central bare zone was significantly higher than that of the movement toward the zone. Given that the backward movement away from the central bare zone would cause the myosin heads to be constrained and the stiffness of the cross-bridges to increase, these results suggest that elastic energy required for the cross-bridge transition is supplied by thermal fluctuations.     

Effect of adenosine triphosphate, inorganic phosphate and divalent cations on the size and structure of synthetic myosin filaments. An electron microscope study

We have studied the influence of ATP, inorganic monophosphate, MgCl₂ and CaCl₂, alone or in combination, on the formation of synthetic myosin filaments by means of electron microscopy. Both crude and column-purified rat skeletal myosin were studied systematically, and in some instances parallel experiments were carried out using crude rabbit skeletal myosin.The behaviour of filaments formed by a standard polymerization procedure at pH 6.8, in the absence of ATP or inorganic phosphate, is not influenced by MgCl₂, CaCl₂ or ethylenediaminetetraacetic acid at concentrations up to 5 mm. Such filaments are homogeneously 30 to 50 nm wide and 5 to 15 μm. long, i.e. larger than physiological size. They do not systematically display tapered ends. Filaments are built up from 2 to 3-nm wide threadlike subunits, arranged roughly parallel to the long axes. Sometimes filamentous projections up to 60 nm long, not associated with accidentical filament bending, are seen. In heavily contrasted preparations these projections are replaced by irregular globular structures. This transformation of the projections is accompanied by a slight decrease in the diameter of the filament shaft.When polymerization is carried out in the presence of millimolar amounts of ATP or inorganic phosphate, but in the absence of divalent cations, regular filaments are not formed. Instead long branched structures, or small twig-like filaments are obtained, which are 10 to 15 nm wide and 0.2 to 0.4 μm long. The addition of ATP or inorganic phosphate to preformed regular filaments does not bring about this structural disorder.The presence of millimolar amounts of MgCl₂ or CaCl₂ in the polymerization medium efficiently counteracts the disorganizing effect of ATP and inorganic phosphate. The size of filaments formed under these circumstances critically depends upon the nature of the divalent cation. CaCl₂ induces the formation of filaments similar in every respect to those already described. In the presence of MgCl₂ distinctly thinner ones, with close to physiological diameters (15 to 17 nm), are obtained. In both cases the lengths are still 5 to 15 μm.The filaments with physiological diameters consistently display tapered ends. They are built up from the same threadlike subunits as wider ones, and most display filamentous projections up to 60 nm long, systematically pointing in the direction of the filament ends. Central zones of polarity reversal are easily identified. In heavily contrasted preparations these filamentous projections are again replaced by irregular globular structures and the apparent diameter of the shaft is slightly diminished.In no case did we notice any significant influence of the origin, or degree of purity, of the myosin on filament behaviour.