using bungee cords moving on pairs of motorized pulleys to test the possibility that motile diatoms can swim. The outside portions of the bungee cords lie in grooves, allowing them to make contact with the water outside. Particles in the water could show up in any induced water flows.Figure 14.41 Surface treadmilling, swimming and snorkeling diatoms (2007). A snorkeling pennate diatom.Figure 14.42 Acoustic streaming: the diatom as vibrator or jack hammer (2010). Perhaps a vibrating diatom can smash its way through sand, like a jack hammer removes concrete [14.306] (reprinted with kind permission of Robert Bosch Tool Corporation). The diatom is Lyrella esul [14.338] (reprinted with kind permission of David A. Siqueiros Beltrones).Figure 14.43 Propulsion of diatoms via many small explosions (2020). Nuclear propulsion via hundreds of “small” nuclear bombs: Project Orion (conceived by Stanislaw Ulam, 1946), an analog of diatom propulsion via explosive hydration of raphe fibrils. Top left: artist’s conception [14.420]. Top right: Principle [14.315]: The “Pulse unit injection” would correspond to a raphe carrying raphe fibrils that explosively hydrate on being exposed to water. Middle right: A space faring diatom [14.119] [14.124] Lyrella esul propelled by explosions [14.134]; Lyrella esul [14.338] with kind permission of David A. Siqueiros Beltrones. Bottom: Project Orion configuration [14.420], all others from NASA, in public domain).Figure 14.44 Diatoms walk like geckos (2019). Was the gecko inspired by the diatom, or vice versa? (Gecko [14.206] . Pseudoraphid diatom Podocystis adriatica [14.139] with permission of Paul Hargraves and the New England Botanical Club).Figure 14.45 Intraprotein pores in hyaluronan synthase and cellulose synthase. Top left: “A model for interaction of the HAS [hyaluronan synthase] tandem-motif region with HA [hyaluronan, the #s refer to membrane domains]. The… growing HA-UDP chain, containing alternating GlcNAcβ1,4 (blue squares) and GlcUAβ1,3 (blue-white diamonds) attached to UDP [uridine diphosphate] (red inverted triangle) at the reducing end. Preliminary results indicate that the nonreducing end contains a chitin oligomer cap with four GlcNAc residues, which is the primer on which HA synthesis is initiated” [14.18] (reprinted with permission of Oxford University Press). Top right: AFM experiment suggesting that raphe fibrils are multistranded [14.153] (like cellulose) (reprinted with permission of Elsevier). Bottom: “Updated models of plant CESAs [cellulose synthases] and CSCs [cellulose synthesis complexes]. (a) A computational model of a plant CESA catalytic domain with P-CR and CSR regions (light grey). The glucan chain (purple) is from the homologous Rhodobacter structure. The location of the transmembrane helices (TMH) is represented with grey boxes. (b) Three CESAs, encoded by three different genes, may interact to form a trimeric particle, which in turn may assemble into a hexameric rosette, depicted in (c). The glucan chains are represented in red. (Image (a) is adapted from [14.339] and is courtesy of Jonathan Davis)…. the intra-protein tunnel [pore]… provides a low-energy pathway for translocating the growing glucan chain to the external membrane surface from the cytoplasmic side, where the catalytic site transfers a glucose residue from UDP-glucose to the reducing end of the glucan” [14.62] (reprinted with permission of Elsevier).Figure 14.46 Left: “Stages in secretion at the raphe (hypothetical). (a) Polysaccharide fibrils are discharged from a vesicle into the raphe at the central pore. The preferred secretory sites are at both central and apical (not shown) raphe endings; (b) Secreted fibrils experience hydration and begin to swell and elongate. A second vesicle is docked at the plasmalemma and primed for exocytosis. (c) After considerable swelling the mucilaginous strands project through the raphe; their proximal ends are attached to the plasmalemma and their distal ends are free to make contact with a substratum; (d) Strands, which were produced at the raphe ending, have moved along the raphe but still occupy the slit and maintain attachment to the plasmalemma. Microfilaments act as a barrier preventing vesicle discharge here (not evident in all cells)” [14.93] (reprinted with permission of Elsevier). Note that the hydration of the raphe fibrils does not likely occur within the raphe, as shown, because the walls of the raphe are hydrophobic [14.92] [14.93] [14.122]. On the contrary, hydration at the external tip would generate a force pulling the fibrils out of the raphe. Right: The silica lining the raphe differs from that of the valve [14.66] (reprinted with permission of John Wiley & Sons).Figure 14.47 A general model for noncellulosic polysaccharide biosynthesis [14.362] (reprinted with permission of Oxford University Press).Figure 14.48 “TEM of thin sections of Amphora veneta. 26: T. S. [thin section] through a perinuclear dictyosome [Golgi]. The dictyosome is composed of 8 cisternae (ct) which are blebbing fibrous (v1) and smaller granular (v2) vesicles (× 44,000). 27: Oblique L. S. through the central region. Irregularly shaped fibrous (v1) and smaller granular vesicles (v2) are apparent (× 32,400). 28: T. S. part of cell to show frustule-cell-membrane interface (fcm). Large numbers of vesicles (v4) are present outside the cell membrane (cm) but within the silica wall (sw) (× 33,000). 29: T. S. showing fibrous material (fm) extruding from a raphe fissure (rf) (× 22,400)” [14.70] (reprinted with permission of Springer Nature). Note: “The largest [vesicles] (27), were derived from the mature dictyosome [Golgi] cisternae, and were often irregular in outline whilst containing fibrillar material. These vesicles were widely distributed throughout the cytoplasm and the contents were similar in morphology to the extracellular mucilaginous material (29).”Figure 14.49 “Four possible configurations of the microfilament bundles, assuming that they are capable of sliding a short distance relative to the silica raphe. Only the ends of the raphe are shown here, represented by two pores. The view is from the inside of the cell, looking out. The placement of the raphe pores on opposite sides of the apical axis [14.34] is a common feature of raphid diatoms. In this scheme, the microfilament bundles block access of the crystalloid bodies to the raphe along their whole length. In the capillarity model, such blockage would explain the direction and reversal of motion, as indicated on the right” [14.117]. (Reprinted with permission of Elsevier).Figure 14.50 Top: Swirling, draining patterns on a bubble film draining downwards [14.250] (reprinted with kind permission of Michael Reese Much FRMS). Bottom: One frame from a movie showing how dynamic these patterns are [14.288]. The interference colors are due to varying thickness. Note that the flows organize themselves along narrow lines which can draw in material from a wide area. This could be analogous to a diatom’s raphe drawing the liquid cell membrane towards it, bearing proteins (raphan synthase) ready to secrete their polysaccharide raphe fibrils (raphan) upon reaching the raphe.Figure 14.51 Surfing diatoms, Achnanthes and Podocystis [14.95].Figure 14.52 Membrane surfing: A new working hypothesis for the diatom motor (2020). Surfing working model for diatom motility. Raphan synthase is a hypothesized membrane protein that is functional either in the cell membrane, or more likely, in vesicles (crystalloid bodies) into which it deposits raphan, the presumed polysaccharide constituting the raphe fibrils seen in raphes. The whole cell membrane flows, bringing the raphan synthases or the crystalloid bodies, represented as miniatures of Figure 14.51, to the raphe. Myosin motor molecules move along the microfilament bundles (only one shown) hydrodynamically inducing flow of the cell membrane by carrying the vesicles over the raphe. As the vesicles fuse with the cell membrane, they dump their hydrophobic contents into the raphe, where the hydrophobic raphan is shown as short, vertical lines. As the raphe lining is hydrophobic, they fill the raphe via capillarity. Those that come in contact with a substrate, such as the chondrite [14.125] shown here, hydrate, swell, and stick to it. As these hydrated raphan molecules exit, more hydrophobic raphan molecules fill in the raphe, causing a net flow of the anhydrous raphan within the raphe, shown by the small arrow. The result is that the diatom moves relative to the chondrite in the opposite direction, shown by the large arrow. On a flat substrate, the swollen raphan fibrils are left behind as the diatom trail. The flow of the cell membrane determines the direction in which the diatom moves relative to its substrate. Capillarity provides the tremendous force. “On Earth, capillary forces have to fight gravity. But in space, the only resistance is the viscosity of the liquid, which slows the flow but cannot stop it” [14.240]. At the size scale of the diatom raphe, gravity has negligible effect, so the slogan applies.Figure A14.1 Could squeezing out of raphe fluid cause diatom motility? Photo of caulking gun [A.14.4] superimposed with a Cymatopleura diatom [A.14.1] under the Creative Commons Attribution License (CC BY 4.0).