I am not convinced of this mechanism and without evidence it is nothing but speculation. A benefit from the mobility at the surface is not evident. Structures of interconnected Nitzschia sigmoidea, as described above, require a very calm water surface in addition to a high population density. Under light winds and waves, these fragile structures will certainly not be able to form or preserve themselves. I consider these patterns and their dynamics to be an artifact that only occurs in cultures, but it allows an insight into the motility of this species.
The observations of Craticula cuspidata, Cymbella spp., Rhopalodia spp. and Pinnularia spp. on the water surface differ in several aspects from those of Nitzschia sigmoidea. In the following I will restrict myself to comments on Pinnularia gentilis (Figure 1.31) from a small pond in Stuttgart-Hohenheim. At the time of observation, the diatoms were already six months in culture and had a typical length of 200 μm. When Pinnularia cultures are prepared, a fast sedimentation of inserted diatoms is usually observed. In the case of diatoms from these cultures, it is noticeable that many diatoms are stirred up when the Petri dish is carried to the microscope and settle relatively slowly on the substrate. Immediately after swirling up the diatoms, one regularly finds a few to several tens of diatoms on the water surface. Already in the first minutes many of the diatoms sink to the ground. Others remain on the surface for hours and only a few for days. In all observed cases, the sinking starts with the diatom taking a position perpendicular to the water surface. Often it remains on the water surface in this orientation for a while. Either the diatom drops to the ground in this orientation or it rotates around the transapical axis or pervalvarous axis as it sinks. It was not possible to recognize which axis it is. During one observation, a diatom returned from the vertical position back to the horizontal position on the surface, which requires an energy supply by the diatom.
Pinnularia floating horizontally on the surface are almost completely enclosed by water. A deformation of the water surface is not visible in phase contrast. There is also no formation of regular patterns due to an attractive interaction. Nevertheless, I consider a slight hydrophobicity to be possible.
Bacteria can often be found on the water. If these form a coherent turf, this probably has a significant influence on the movement of the diatoms. Also, with these observations the water surface showed only a low bacterial density, so that I consider its influence negligible.
When looking vertically at the water surface, the diatoms appear on the water surface in both valve view and girdle band view, with the valve view dominating. Occasionally there is a 90° rotation around the apical axis and thus a change between the two views. Presumably an activity of the raphe in the area of the helictoglossa is responsible for it. As on substrate, Pinnularia in valve view have a high mobility and cover longer distances, while in girdle band view back and forth movements are carried out. The movement is very similar to that on a solid substrate. Occasionally Pinnularia in valve view show rotations around the pervalvarous axis, which cannot be found on substrate. The adhesion to the substrate will probably prevent such rotations. In practice, this movement is often accompanied by a drift movement. In this context, it should be noted that the collective movement of two Pinnularia also appeared. I suspect the coupling by adhering EPS lumps.
The observations of Pinnularia gentilis, which move actively on the water surface, whereby the driving raphe is completely below the water surface, again support Bertrand’s [1.4] hypothesis of a wavelike movement of microfibrils.
1.6 Formation of Flat Colonies in Cymbella lanceolata
Cymbella species either are tube dwelling, develop stalks that often branch into tree-like structures, form colonies directly on substrates or are free-living [1.28]. The transition between free-living and colony-forming is smooth, as diatoms often leave colonies and develop new ones elsewhere. This is the topic of the observations described subsequently.
First, Cymbella lanceolata with a length of approx. 190 μm is discussed (Figure 1.32). The adhesive EPS excretions are clearly visible in phase contrast, DIC or PlasDIC (Figure 1.33).
A longer observation of accumulations reveals these processes:
1 1. Diatoms detach from a colony and move away from the colony. This typically happens at the edge of accumulations.
2 2. Diatoms move within the space between the colonies.
3 3. Diatoms meet an existing colony and remain in this cluster.
4 4. Diatoms stop their movement and attach themselves to the substrate.
5 5. Diatoms reproduce asexually inside and outside colonies.
Short-term contacts between diatoms are not mentioned here, as they are transient and of minor importance for the establishment of structures. The relevant steps for the structure formation are exemplarily illustrated in Figure 1.34. Diatoms that attach to colonies usually remain on the edge of the colony. As they themselves secrete EPS, the area produced by EPS deposits is continuously increasing. Events 1, 2 and 3 do not allow the emergence of new colonies. A colony can develop from individual adherent diatoms according to 4 by following cell divisions and attachment of diatoms.
Figure 1.32 Cymbella lanceolata.
Figure 1.33 Two small colonies photographed with PlasDIC.
Figure 1.34 Elementary steps that contribute to structure formation.
To give an impression of the movement activity of the freely moving diatoms, images were taken in artificial daylight for over one hour (one picture every 10 s) and superimposed (Figure 1.35). The movement activity of diatoms requires sufficient light intensity. With increasing intensity, the raphes become active in the observed Cymbella, regardless of whether they move freely or are in a colony. As a result, more and more diatoms detach from colonies. The driving force then exceeds the adhesion to the substrate caused by EPS. For demonstration, two small colonies were irradiated with a light intensity between 7000 lx and 9000 lx. This is significantly above the intensities used in cultivation, which were at about 500 lx. The microscope illumination with a color temperature of approx. 3000 K was used for irradiation. The light intensity at the location of the diatom under observation is essential for the magnitude of the driving force. A high degree of homogeneity of the illumination in the area under observation is not required for this observation. Figure 1.36 shows the colonies at the beginning of irradiation and after about two hours. A considerable reduction in size can be recognized.
The removal of diatoms from a colony and migration require sufficient light intensity. Correspondingly, the activity of the movement decreases when the intensity becomes low. When diatoms encounter existing colonies or deposits of EPS on the substrate at low brightness, they adhere to the colonies because they cannot overcome the adhesion (process 3). At