Above: a 3D computer model of a diatom shell in angled side-view (girdle view). Diatoms are microscopic single-celled algae, generally between 2 and 200 micrometers long (one micrometer, 1 μm = one thousandth of a mm). The shell or frustule or theca is a box of silica consisting of two halves or valves. The larger upper valve (epitheca) is the lid of the box which overlaps the smaller lower valve (hypotheca). Where the lid and basal valves overlap, there is a belt of silica called the girdle or cincture. This girdle is made up of a strip of silica on each frustule called a copula. Where the copulae of the upper and lower valves meet and overlap we have the girdle. Some diatoms have no copula on their frustules, which may overlap simply, or multiple copulae to form a girdle of multiple strips (or multiple intercalary bands, multiple girdle bands). In diatoms with several copulae in each valve, the completed box may be tall and tower-like. The model here has only a single copula in each valve. (See this link for a glossary of diatom terms: https://diatoms.org/glossary/). Silica is the chief component of glass and stone, so the diatom frustule is a kind of translucent biological stone or 'glass' (though the molecular structure may differ).
Above: the axes of symmetry of pennate diatom. Pennate diatoms, like our 3D model, have elongated frustules with two poles. Pennate diatoms may have bilateral symmetry (a left side that is the mirror-image of the right side) though the left and right sides may be asymmetric. The apical axis follows the midline (looking at a valve face) connecting the two poles (and may be curved in asymmetric forms). The transapical axis divides the diatom into front and rear halves. The pervalvular axis passes through the center and is perpendicular to the faces of the valves.
Within the diatom frustule is a single cell. A diatom is perhaps best thought of as an amoeba in a box.
The raphe is a paired longitudinal slit in one or both valve faces, but is absent in some diatoms. This feature is associated with locomotion and those diatoms with no or a very short raphe are barely motile if at all. There are many pores or punctae (singular puncta) in the diatom frustule and these are often arranged in rows called striae (singular stria). The chloroplasts are often golden-brown due to the brown pigment fucoxanthin masking the green chlorophyll. The polar and central nodules are thickened regions of silica in the frustule and are often visible as such.
Above: a 3D model of a single valve, showing the outer face (left) and the inner face (right). The areolae are indentations in the inner wall where the striae are (i.e. they are regions of thinner silica where the pores are). An extensive septum, or a short pseudoseptum may be present as a shelf of silica. Other features may be present, depending on the species.
Mechanism of Locomotion
despite being enclosed in a silica box, many diatoms are highly mobile, gliding along at considerable speeds (typically around 5 to micrometers per second) as can be observed under the light microscope.This gliding motility mainly depends on the raphe. Diatoms are most mobile when the raphe is in contact with the surface and diatoms with poorly developed raphe are less mobile and those with no raphe either immotile or very weakly so. The elongated shape of diatoms, with their elongated raphe, seems to be an adaptation to gliding motility. Some diatoms, however, can move in an upright position, gliding on one of their poles, which may use the extension of the raphe or perhaps the apical pore field. Some diatoms occasionally roll over on top one girlde during locomotion and continue gliding whilst on their side, so although rapid motility is associated with the raphe some diatoms seem capable of movement without it.
The mechanism of this gliding is difficult to analyze and is not yet fully understood. However, several working models have been proposed, one of the most recent of which we will outline here. First, it is known that gliding diatoms leave a slime trail behind themselves and that this secretion is released through the raphe groove. The diagram below illustrates the structure of the raphe (R) as seen in cross-section:
The raphe canal is often V-shaped in cross-section, with a projecting lip of silica from one side of the valve sitting inside a groove on the other side, but that the space between the two halves leaves a continuous channel between the cytoplasm inside the frustule and the outside medium. The cytoplasm, bounded by the semi-fluid plasma membrane (PM) invaginates along the raphe and here there are cables of actin filaments (AF) (actin is a filamentous protein and part of the cell cytoskeleton). Vesicles (V) containing a fibrous or mucilaginous material line either side of the raphe. The actin filaments probably act as 'rails' along which the vesicles travel. The vesicles are capable of fusing with the plasma membrane to discharge their contents (a process called exocytosis) into the raphe. This is also illustrated below:
The membrane-bound vesicles fuse with the plasma membrane of the cell, discharging their contents into the raphe. This mucilaginous substance expands on contact with water and forms the slime trail behind the locomoting diatom. This expulsion and hydration of slime could provide some propulsive force for the diatom, if the mucilage is released at one end of the raphe and travels backwards along it by capillarity. However, it is doubtful whether this would be sufficient.
A study by Wang et al. (2013; DOI 10.1007/s00709-013-0502-2) detected small circular depressions in the slime trail of Navicula at intervals, suggestive of microscopic 'footprints' which suggested that the diatom was exerting force at localized intervals as if stepping on the slime to gain propulsion. Close to each pole of the cell they observed a concentration of actin that may be responsible for exerting such localized force, perhaps as a protoplasmic extension of the cell that passes through the raphe (though the shape of the raphe make this seem problematic) or by the cell exerting pressure from within the frustule on extracellular structures, perhaps the mucilage itself, or perhaps the cell pushes the mucilage along the raphe.
The observed footprints suggest a kind of stepping locomotion on feet positioned in the midline of the cell, perhaps two feet with one operating each raphe of a pair. Certainly diatoms do generally make jerky somewhat zigzag movements more aking to stepping than gliding. Models should look at the form of the raphe and how this might be adap[ted for slime secretion or the transmission of force from the cell to the slime trail, including a consideration of the mucilage properties (for example, is it shear thickening?). At present the exact mechanism remains largely a mystery. (The spherical structures seen at the cell poles by Wang et al. could be locomotive organelles but research should rule out other possibilities, e.g. they may be lipid vesicles).
Above: Centric diatoms (a) have radial symmetry and are either disc-shaped, cylindrical or polygonal (triangular or triradial (b), square, pentagonal, hexagonal, star-shaped), for example Actinocyclus which is disc-shaped or shortly cylindrical. Pennate diatoms (c) are elongated along the apical axis, an adaptation that facilitates motility in many forms. Biraphid pennate diatoms (d, left) have a paired raphe on each valve face and may be symmetric, e.g. Navicula, or asymmetric, e.g. Didymospheria. Monoraphid diatoms have a raphe on one face only (d, right) and are most motile when the raphe is in contact with the substrate, e.g. Karayevia. Araphid diatoms (e) have no raphe and are largely immotile, e.g. Diatoma.
Eunotoid diatoms (f) have a short raphe system and are weakly motile. The raphe extends from the valve mantle onto the valve face, e.g. Eunotia. Epithemoid diatoms (g) have a well-developed raphe but displaced from the midline towards the valve margin and enclosed within a canal, e.g. Epithemia. Nitzchioid diatoms have a raphe enclosed within a canal, again displaced towards the margin of one valve and raised onto a keel, e.g. Denticula. Surirelloid diatoms have well-developed raphe occupying the whole valve margins within a canal and often on a keel, e.g. Surirella.
Pennate diatoms
Above: a single valve in face view, showing the outer surface, of our model diatom. Note the striae, the rows of pores, which in this model are uniseriate (in single rows) and elongated parallel to the apical axis. The pair of longitudinal slits in the center-line is the raphe. Each slit is often referred to as a branch of the raphe. Sometimes only a single slit is present, but usually the raphe is divided into two branches by the central nodule (a central thickening of silica). This raphe is axial (aligned to the midline or apical axis) but in other species can be displaced to one side (along on margin) or may be circumferential, running around both margins.
Above: the inside view of a single valve.
Note the pseudospetae, the shelves of silica that extend partway from
the ends of the valve. Similar shelves may occur on the girdle bands, in
which case they are called septa, and may extend a variable distance
depending on species.
Above: an inside view of a valve on a modified version of our model. Note the apical pore fields, the lack of pseudosepta and the bending of the proximal ends (those nearer the center) of the raphe towards one side as commonly occurs in some diatom species. Raphe branches may also terminate in fissures, slits that do not penetrate the whole silica wall, but occur on the inner or outer surface only. Similarly the distal (apical) ends of the raphe may end in distal fissures, and may be curved to one side or over the apex onto the ventral surface a short distance. Some diatoms are able to glide on their narrow apical ends in an upright stance, whilst others can glide briefly on one end as they turn over and it has been suggested the distal fissure may help with this in some way.
Above: in this version of our model, pseudosepta are present and the distal raphe curve to alternate sides.
Pinnularia
Above: a frustule of Pinnularia (with
no living contents) this specimen with its asymmetric elliptic central
area closely resembles Pinnularia nobilis. The striae occur in
chambers which form distinct lines.
The raphe are slightly wavy (undulate) in the specimen above. Undulating raphe is a characteristic of Pinnularia streptoraphe.
The specimen above and below resembles Pinnularia pulchella.
Pinnularia are often large with coarse (broad) striae and so are clearly observed under the light microscope and are good specimens to illustrate the basic features of a pennate diatom.
Above: the central and polar nodules and raphe are clearly visible in this diatom. This diatom has the form of Pinnularia. Both valves have a raphae (the diatom is biraphid).
Pinnularia
Pinnularia is also biraphid. The frustule is described in more detail below.
Boat-shaped diatoms, e.g. Navicula, Pinnularia. The one below (and possibly also the ones above) looks like Craticula. This one was found on the bottom of a runnel in a salt marsh near to a tidal estuary.
Sigmoid or S-shaped Diatoms
e.g. Pleurosigma, Gyrosigma, Hasea nipkowii
The diatom above looks like Gyrosigma, although some species have Haslea also have a similar shape. This shape is described as sigmoid (S-shaped) or rhomboidal sigmoid. This is possibly Gyrosigma littorale. This diatom is biraphid.
The one above looks like Gyrosigma fasciola.
Diploneis
This diatom has a pair of longitudinal canals in the frustule, one each side of the raphe. These canals open to the outside by a series of pores but do have no internal openings. These canals have an unknown function.
The frustule below is likely from another species of Diploneis. Note the complex nature of the areolae pores. This genus is biraphid.
The diatom below resembles such forms as Diploneis krammeri or Diploneis mollenhaueri and maybe a similar type.
Studying living diatoms is clearly important, but the images below are of dead diatoms, enabling the detailed architecture of the frustules to be clearly seen.
Eunotioid diatoms, e.g. Eunotia
Above and below: fossilized frustules of Eunotia tetraodon (with no living contents). The frustule is curved in valve view, with a convex dorsal side and a concave ventral side. Note the uniformity of shape between specimens: there are consistantly 4 'humps' on the dorsal valve edge. Eunotia may be found attached to rocks and plants by a secretion of mucus from their ventral side.
The reduced raphe can also be seen at the ventral ends of the frustule. Eunotia is found in freshwater and is mainly benthic (living on the substrate) or on submerged plants / algae. Some forms are planktonic. Some forms are colonial. Eunotia asterionelloides forms stellate /radiant colonies or zig-zag chains.
In pennate diatoms, cell division is as follows. The cell swells, increasing its volume and forcing the epitheca and hyptheca apart, but not enough to expose the naked protoplast within. This is followed by mitosis, a copy division, in which the cell splits into two daughter cells. One daughter is about the same size as the parent cell, the other one smaller. Each cell inherits one of the parental valves which becomes the new epitheca (lid) and then secretes the smaller hypotheca (base), then the two cells separate.
In this way, one cell remains about the same size as the parent, but one has become smaller and can not grow larger due to it occupying a smaller box. Repeated divisions thus reduces the average size of the population. At this point the small diatoms will form auxosopores, usually following sexual union. The auxosopore expands, restoring cell size. Cells that do not form auxosopores will continue dividing and shrinking until they die. A few forms can prevent shrinkage with division due to the presence of an elastic, stretchable girdle, allowing the cell to expand.
Above: Eunotia tetraodon with a diatom possessing asymmetry about the transapical axis, which is characteristic of such genera as Gomphonema, Gomphonella, Gomphoneis, Gomphonsinica and Gomphosenia.
Cell division in pennate diatoms has been well-studied in Eunotia. In Eunotia these cell divisions reduce the cell size and the small cells will continue dividing, however, once they reach a minimum size at which point the cells become sexually receptive. If such a cell encounters a partner then the partners will move together and enclose themselves in a common mucilage envelope. (Otherwise, the cell will continue dividing, getting smaller until it dies). The chloroplast on the hypotheca side swells, whilst the other on the epitheca side shrinks. The nucleus becomes displaced slightly towards the epitheca.
The partner cells then undergo meiosis, a reduction division in which the normal diploid nucleus divides first into two then into 4 haploid nuclei. The first stage results in the formation of two daughter cells within each parental theca. The parental hypotheca houses the larger daughter cell (with the larger chloroplast) and the epitheca the smaller daughter cell (with the smaller chloroplast). The smaller protoplast takes no further part and aborts / degenerates. Meanwhile, a gap opens between the valves at one end on the ventral side through which a copulation papilla begins to grow, one growing from each box towards its partner.
The second stage sees the nucleus of each daughter cell divide again, so that each contains 2 haploid nuclei, one of which aborts and degenerates. Thus, two haploid cells result, still within the parental theca.
The copulation papillae meet to form a conjugation canal or bridge. the protoplasts, each now an amoeboid gamete, migrate into this canal, meeting and fusing in the middle to form a binucleate zygote. The zygote swells into an auxospore which secretes a temporary silicon wall called a perizonium around itself, which is enlarged as the auxospore swells. The two haploid nuclei fuse.
The auxospore then undergoes mitosis with nuclear division, producing two nuclei, one of which degenerates and deposits the larger epitheca. It then undergoes mitosis once more to form two nuclei, one of which degenerates again and secretes the hypotheca. The thecae are laid down within the perizonium which is shed. Thus, the cell has its size restored and incorporates genetic material from both parents. This cell will divide as normal, producing a population of smaller and smaller cells until the cycle repeats.
Above: Eunotia tetraodon with another Gomphonema type diatom (and a large fragment of what is probably Pinnularia). If you look very closely (click on image for full size) the diatom on the left appears to have two stigma (stigmoids, isolated pores) on the leftmost side in the central area, suggesting this may be Gomphonema duplipunctatum, though the head end (at the bottom in the image) appears a little too pointed.
Mitosis in diatoms is of the open type, meaning the cell nucleus disperses its envelope during division, though the telophase spindle persists. During interphase, a small darkly staining body embedded in the surface of the nucleus becomes the primary microtubule-organizing center (MTOC). The mitotic spindle forms near this MTOC between two darkly-staining plates, one at each pole.
The specimens below strongly resemble Eunotia tenella. Note the asymmetry but this time there is a single convex dorsum with only a single hump.
Gomphonema
This shape is characteristic of Gomphonema acuminatum, which is typically between 19 and 77 micrometers in length. one end, the tail-pole or foot-pole, is narrow and the broadest point occurs at the head-pole. In life, the tail pole may be attached to a secreted mucilaginous stalk. Though as a biraphid these diatoms can presumably also glide well.
If you look carefully at the image above (click images for full size) then you may notice that the pair of striae in the very middle of the swollen mid-section are short but that one of them is accompanied by an isolated pore in the center, on the right-hand side. This single stigma (stigmoid), as it is called, may or may not be present but is a characteristic of Gomphonema and certain other genera. Some genera of pennate diatoms have several such stigmata. Sometimes stigmata are also called stigmoids or isolated punctata (though some authors consider punctata to be pores with a different structure with a punctatum having the same structure as an areola pore in the striae, whereas a stigma has a slit-like or more complicated internal opening). The diatom to the upper right of the Gomphonema looks like Tetramphora. A few Tabellaria are also present.
Above: A Gomphonema frustule together with frustules of Tabellaria.
Above and below: Gomphonema in girdle-view (side-view) showing the characteristic wedge-shape with the narrow foot-pole.
Above and below: the 'bow-tie' shaped central region of these diatoms, and their overall shape, suggest they are Gomphonema truncatum. These specimens were on the small side.
Above: three Gomphonema frustules. The top one looks like Gomphonema truncatum, the middle one like Gomphonema brebissonii. The bottom one is in girdle view and this makes it hard to determine the species.
Cymbella
Cymbella typically secretes a
mucilaginous stalk from the apical pore field at one end by which it
attaches to the substrate. Colonies are often produced, with branching
stalks bearing a diatom cell at the end of each branch. These colonies
can be several centimeters across and resemble dirty balls of
cotton-wool.
Stauroneis
This genera has a stauros, a thickening of silica in the middle of the valve face - a transverse strip, giving the whole a cross-like appearance. This is a characteristic feature of the genus Stauroneis.
Tabellaria
This diatom is araphid, it has no raphid. In girdle view it is
box-shaped. The striae are very feint but can be seen under the light
microscope. Some Pinnularia are a similar shape, but the surface
of the valves are more textured in the latter.
Tabellaria posses a number of narrow girdle bands (copulae) that often detach in preparations. These copulae often have a septa at one or both ends, a horizontal shelf of silica which can be seen by the transverse lines at the end of the septum.
In nature Tabellaria is colonial and forms zig-zag chains and small plates of cells connected in zig-zag chains in which the cells are connected end-to-end by mucilage pads secreted through their apical pore-fields.
Cocconeis (monoraphid)
Above and below: Cocconeis, possibly Cocconeis grovei on account of the small central area and hyaline ring around the margin. Cocconeis is monoraphid, it has a raphid on the flatter ventral valves only with which it can glide over the substrate. They occur on algae (such as Cladophora) and whales (such as on Antarctic Minke Whales) where they may occur in high density, almost completely covering patches of the substrate in a monolayer of their cells.
The striae are fine in this species and hard
to discern with light microscopy, with the areolae pores merging in the
images due to the insufficient resolution, to appear as longitudinal or
wavy lines.
Naviculoid diatoms
Naviculoid diatoms are boat-shaped in contour when seen in valve view.
The one above and below look like Navicula. The genera Kurtkrammeria
and Encyonopsis are similar but have slight asymmetry around the
apicular axis (their left and right halves differ noticeably). Some Gomphonema
are different but they have some asymmetry around the transapical axis
(their front and back halves differ noticeably). Some other genera have
naviculoid forms and the classification of Navicula itself is
problematic. The shape is probably an adaptation to fast locomotion.
Fragilaria
Fragilaria is araphid (lackes raphe on both valves) and is usually colonial, the individual cells connecting at their centers to form a ribbon
Above and below: the shape of these long thin frustules and the fact that the central region appears to be mainly on one side, the rounded (capitate) ends and that the striae are in opposite pairs suggests that this is likely to be Fragilaria synegrotesca.
Above: two cells joined together as part of a Fragilaria ribbon.
More Diatoms
The diatoms below are hard to identify since
they are in girdle view, but they illustrate the hirdle bands and the
box-like nature of the frustule.
Centric diatoms
Centric diatoms have no raphe and are radially symmetric with pores arranged around the center; many are circular or elliptical in outline and form short discs or tall cylinders. Lacking raphe they are less motile though some have been reported to shift their position. Others are polygonal (triangular, squarish, pentagonal, hexagonal) or star-shaped (with various numbers of rays depending on species).
Click on the images to view full size. Above and below: Cyclostephanos. The pores are arranged in sectors called fascicles (lit. bundles), separated by thicker ribs (costae) of silica in this species. Spines are often present around the rim and visible as tiny projections pointing away from the face.
Above: the central annulus from which the fascicles radiate can be seen. Centric diatoms are a distinctly separate, but related, group from the pennate diatoms with key biological differences. Their division is similar and has been well studied in Stephanopyxis turris. This diatom exists in pairs of cells, connected by their overlapping girdles. The pairs connect to other pairs via spines on the valve surface, forming chains of cell pairs.
In this species division commences after daybreak, following the night-time rest period, and begins as the cylindrical cells increase in height (length) by swelling, pushing the valves apart slightly in a telescopic fashion. (It is for this reason that cells reduce more in diameter when they divide than they do in height).
The nucleus, which normally sits in the lower half within the hypovalve, moves to the cell equator and divides by mitosis. The cell protoplast itself divides following nuclear division.
The daughter protoplasts swell a bit and secrete a new hypovalve each, forming two new cells slightly smaller in diameter than the parent cell.
Repeated cell division results in smaller cells and when cells are reduced to 20 to 40% of their original diameter they become sexually receptive and will mate to produce an auxosopore, but if they fail to do this they will keep dividing and getting smaller until they die.
The smaller cells generally become male, the larger cells female. The male cell will undergo 2 or 3 rapid cell divisions without swelling to produce a chain of 8 or 16 small cells with reduced silica walls. These cells swell, creating a gap between the valves through which the male gametes or spermatozoa will escape. Each spermatozoid has a single tinsel flagellum for swimming (a tinsel flagellum or pleuronematic flagellum has hairs on it that assist thrust generation by pulling the spermatozoid through the water). Each male cell undergoes meiosis to produce 4 spermatozoids (and a residual protoplast containing the chloroplasts that dgenerate). The 4 colorless spermatozoids then escape to locate a female egg.
The female cells swell and the nucleus moves to the cell equator. After the first meiotic division, one of the two nuclei produced degenerates, then the remaining nucleus undergoes the second meiotic division t o produce two haploid nuclei, one of which degenerates, leaving one nucleus to form the egg. The egg swells, pushing the valves apart slightly and bends in the middle where a papilla receptive to spermatozoids emerges. When a spermatozoid finds the egg it docks with the papilla and injects its nucleus. The male and female haploid nuclei fuse to form a zygote which becomes an auxospore. The auxospore expands, made possible by its stretchable wall which is covered in silica scales. Thus, cell size is restored.
Above: Cyclostephanos; some spines are visible projecting towards the viewer from the cell margin.
Above: this diatom looks like Cyclotella or Melosira. The centric diatoms in girdle view below, form colonial chains. Melosira will do this and also has little ornamentation on the frustules as do these.
