the strands were originally synthesized”; (b) hydration is assumed to occur along the whole length of each mucopolysaccharide fibril while it is still within the raphe; (c) the fibrils are assumed to swell and elongate as they come out of the crystalloid bodies; (d) a mechanism is needed by which “the mucilage strands are broken free from the plasmalemma on reaching the apical raphe ending [trailing pore].” The microfilament bundle is presumed to provide the motive force. Reversal of the direction of motion is suggested to occur as follows: “Since... there are two bundles of filaments..., perhaps the actin filaments in each bundle are oriented in one direction, and the polarities in the two bundles of filaments are opposite. If this were so, bidirectional movement could be envisaged as occurring as...the raphe adhesive is moved by some controlled activation of or coupling with the actin bundles alternately.” Only one microfilament bundle is shown. The other, oppositely oriented, would be placed behind this one. Thus, two rows of fibrils are possibly present in a raphe simultaneously. The second row would be behind the one shown” [14.117] (reprinted with permission of Elsevier).Figure 14.30 Diatom adhesion as a sliding toilet plunger (1966). Left: “Suction cup with handle, used to study suction-seal-substratum relationships” [14.79] (reprinted with permission of Springer Nature). Right: Ryan W. Drum in 2003.Figure 14.31 Diatom as a monorail that lays its own track (1967). Top: In 1967: “Either of the mechanisms indicated…could keep the trail against the raphe and explain adhesion forces. In the…osmotic… hypothesis…(a), the trail T is held in position by low pressure in the water W in the raphe slit. The pressure would be kept low by osmosis across the cell membrane. Alternatively,…the interfacial tension hypothesis… (b) shows the raphe filled with a liquid L immiscible with water. Interfacial tension at I would then keep the trail in place…. Schematic raphe cross-sections. Substrate stippled. I, water-liquid interface; L, liquid; T, trail; W, water at low pressure being sucked into the diatom by osmosis” [14.142] (reprinted with permission of Taylor & Francis). Middle right: A monorail train gripping its track [14.309] (public domain image). If diatoms do indeed lay down their own track, we have much to learn from them [14.430]. Bottom left: Margaret A. Harper and John F. Harper, wedding photo taken at Willesden Parish Church, NW London, UK, 29 August 1964. Bottom right: Margaret A. Harper and John F. Harper, at present, with their permission.Figure 14.32 “Four locomotion theories [as of 1977], represented on schematic sections of pennate diatoms in their apical planes. Trail precursor is shown by coarse stipple, trail by fine stipple. (a) Jarosch (1962) [14.177]. Mucilage secreted at pores a and c being driven by undulating actin filaments connected to the protoplast. (b) Hopkins & Drum (1966) [14.164]. Trail secreted through entire lengths of raphes ab and cd and expanding on leaving the raphe system. (c) Harper & Harper (1967) [14.142]. Motion due to trail secretion from pore a. Secretion from upper pores e and h forming lump at l and particle p being carried forward. (d) Gordon & Drum (1970) [14.122]. Capillary flow of a liquid along raphes ab and cd, conversion to trail at pores b and d” [14.141] (reprinted with permission of John Wiley and Sons). Note that the central nodule is lacking in diatoms such as Bacillaria [14.439], which either reduces the complexity from four pathways, as in Liebsch (1929) [14.218], or indicates that the central nodule does not play the role suggested here.Figure 14.33 The diatom as a “compressed air” Coanda Effect gliding vehicle (1967). A wing profile assisted by compressed air coming out of the plenum chambers, with the air flowing close to the surface due to the Coanda effect [14.22] (reprinted with kind permission of Stephen D. Prior, superimposed by diatom Cymbella cistula [14.244] with permission of the Muséum national d’Histoire naturelle under a Creative Commons license). Of course, the raphes should be rotated 90° for a proper match.Figure 14.34 The electrokinetic diatom (1974). Depiction of ionic currents that break the symmetry of the spherical Fucus egg [14.172] (reprinted with permission of Elsevier). Such currents, if they occur in pennate diatoms, could bias the flow of charged molecules in the raphe or the cell membrane just within the raphe or the microfilaments adjacent to the raphe. Reversal of the field would then be predicted when the diatom reverses direction.Figure 14.35 Discovery of the “crystalloid bodies” and their relationship to the raphe. Left: “1. A three-dimensional view of the silica structure of a diatom. The raphe system (RS) is composed of a central pore (CP) and a terminal pore (TP) with an outer groove of the raphe fissure (R). Continuation grooves (CG) and a probable anterior pore (AP) are shown. 2. A raphe plane section of a diatom moving upon a plane surface (GS) showing the four pores as in #1. The cytoplasm contains a fibrillar bundle (FB) and crystalloid bodies (C) containing minute fibrils (F). These are found in the diatom raphe system and in the diatom trail. The point of locomotor-adhesion contact (LAC) is indicated. 3. A raphe plane section of a diatom moving upon particles (PS). In this case, locomotor-adhesion contact (LAC) can involve any point along the raphe system” [14.164] (reprinted with permission of Taylor & Francis). Right: “Transverse cross-section of a raphe fissure and adjacent cytoplasm of a diatom moving over a particulate substratum; the locomotor adhesion seal (arrow), firmly attached to particles, is being pushed along raphe by streaming directed against that seal; the fibrillar bundle (F) lies next to the raphe (R); longitudinal and transverse sections of crystalloid bodies (CB) are also shown” [14.79] (reprinted with permission of Springer Nature). The fibrillar bundles were later identified as similar to smooth muscle and presumed to contract, pushing the crystalloid bodies and moving the diatom [14.173].Figure 14.36 The diatom clothes line or railroad track (1980). Left: Clothespin line model for diatom motility, amalgamated from [14.11] [14.427] (public domain image). Top to bottom: microfilament bundle, myosins, clothespins are membrane bound raphan synthase carrying raphan inside a raphe, where raphan is a polysaccharide raphe fibril. Top right: “Model for the organization of the motor apparatus in diatoms. Adhesive mucilage secreted by the diatom adheres to the substrate and binds to as yet undefined transmembrane components. The cytoplasmic domain of the membrane-associated complex is linked to a diatom myosin which actively translocates the membrane complex and attached mucilage rearward along a track of cortical actin filaments, leading to forward gliding of the diatom” [14.146] (reprinted with permission of Elsevier). This is the generally assumed model of Lesley Edgar and Jeremy Pickett-Heaps, though they also allowed for an indirect coupling of the myosin to the raphe fibrils [14.93]. Bottom Right: Similar model of Rick Wetherbee et al. (1998): “Diagram suggesting how the adhesion complex would look in a raphid diatom. Note that only the actin [microfilament] has been described for certain. There has been no attempt to illustrate where components responsible for motility (e.g., a motor) might be located” [14.401] (reprinted with permission of John Wiley & Sons). The actin-associated proteins would presumably be myosins, and the transmembrane proteins would be raphan synthase.Figure 14.37 Diatom ion cyclotron resonance (1987). A take on diatoms and cyclotrons [14.207] (public domain image). Accelerating diatoms are Cymbella cistula [14.244] (reprinted with permission of the Muséum national d’Histoire naturelle under a Creative Commons license).Figure 14.38 Diatoms do internal treadmilling (1998). Left: In treadmilling of microfilaments, actin monomers are added at one end and removed at the other end, resulting in a constant length, with motion [14.45] (reprinted with permission of Springer Nature). Upper right: Molecular motors may be involved in regulating the length of a treadmilling microfilament: “A treadmilling filament is described by a lattice of dynamic length. Motors are represented as particles that occupy the sites. At one end sites are added at a rate of α. At the opposing end, empty (occupied) sites are removed at a rate of β . Particles attach to empty sites at a rate of ω and detach from the lattice at a rate of . Particles hop to adjacent free sites in the direction of the shrinking end at a rate of γ” [14.100] (reprinted with permission of the American Physical Society Publishing). Lower right: “A model showing how myosin driven actin sliding with the combination of tethering proteins can potentially drive ER [endoplasmic reticulum] and Golgi mobility. Myosins are shown linking actin filaments within a bundle and are responsible for filament sliding” [14.234] (reprinted with kind permission of Joseph F. McKenna). Similar sliding might be occurring in the microfilament bundles adjacent to raphes.Figure 14.39 Flow along a line, such as a raphe, induces a circulating flow of the adjacent fluid, by analogy to the early neural plate in a chicken [14.402] (reprinted with permission of Springer Nature).Figure 14.40 Rough schematic for a diatom robot (diatombot), a neutrally buoyant, remotely controlled