Pinnularia spec. (60x time lapse) Surirella robusta (different time lapse rates from 90x to 30x)


Identification of genus and species

The cultivation of diatoms makes it possible, to assign observations to a genus or even species. For this they must be identified. It is a difficult area with an extensive literature that is only mastered by experts. This is is the reason why the information on the videos shown here is to be questioned occasionally. Often we have intentionally limited ourselves to the assignment to the genus.

At this point only a few general hints for preparation and determination are given.

The morphological identification is based on the shape and structure of the valves and is traditionally based on a light microscopic examination. However, it is probably not possible to distinguish two different species morphologically in general. The method of DNA barcoding, which uses a molecular genetic fingerprint for taxonomic classification, is therefore considerably more sensitive (see „Revolutioniert DNA-Barcoding die Gewässergüteanalyse?“ In view of the high number of diatom species, the practical applications of the method are presumably dreams of the future.

Although the structures of the valves in living diatoms are not discernible in detail, certain genera or species of diatoms can well be identified. If features are added such as chloroplast types and arrangement in valve and girdle view, or colony formation, the possibilities of determination enlarge. This approach is followed in "Identification of Freshwater Diatoms from Live Material" (E. J. Cox).

In most cases, however, the determination is carried out using a microscopic image of a diatom frustule. It must be free from interfering cell components. In aged cultures one usually finds suitable valves of dead diatoms, so that one often gets along without preparation. If no such valves are available, preparations from living cells can be made. There are various methods for this. The use of strong acids (sulphuric acid) or oxidizing agents (hydrogen peroxide) is usually described.

A very old method, which is no longer recommended by experts, is the annealing of samples. As I do not like to deal with dangerous chemical substances and do not have a fume hood, I have taken up this very simple and fast procedure to take pictures of the valves. Durable preparations are not produced in this way and these photos do not win a beauty contest, but they do their job well. The process is described below.

First of all, drop droplets from a culture onto several clean cover slips and allows them to dry thoroughly. Then place them on a hotplate and heat them up. It has proved useful to initially increase the temperature slowly in order to prevent the bursting of cells with a high remaining water content. When approximately the hotplate begins to glow, the cover slips are removed from the hotplate bit by bit. Thus, it is more likely to find a cover slip with good results. If the heating is too low, unburnt black organic residues are found. Too strong heating leads to the destruction of fine structures. There is also the risk that the cover slips will deform considerably.

The cover slips are turned carefully, so that the diatoms are located on the underside and are placed on a microscope slide. With the upright microscope the glass thickness is obtained for which the objective lenses are optimized. The difference in refractive index between air and silicate is so large that a high-contrast image is obtained in the bright field. If immersion oil is used, the cover glass must be fixed to the slide beforehand.

As a large number of diatoms of one species can be prepared simultaneously, the probability of finding intact and well cleaned valves is high. With a bit of luck, an overview looks like this (Click to enlarge):

(Click to enlarge)

This image shows two diatoms of the genera Cymbella, which have been cleaned this way (100x oil, stacked image):

(Click to enlarge)

Now one can look at the identification literature (identification keys) or the databases on the internet.

A characteristic feature of the annealing process compared to the use of acids is that the frustules of diatoms are almost never separated. The images of the diatoms therefore appear darker in transmitted light. At high depth of field (especially at low magnifications) the structures of both valves can superimpose, which worsens the visual impression and the photographic image. If you put together an image from a stack of images at different focus levels, you have to be careful to consider only the required layers.

If the diatoms are located in a suitable spatial position, this is an advantage as it can help to gain a better spatial impression of the shape of the diatoms. However, this does not help to identify them. As an example, photographs of Cymatopleura elliptica are shown.


Cox, E.J. (1997) The Identification of Freshwater Diatoms from Live Material. Chapman & Hall, London


Cookies make it easier for me to provide you with my services. With the usage of my services you permit me to use cookies.