Sexual reproduction

Cymbella culture with asexual and sexual reproduction

(2x time lapse)

Cymbella culture some time after the onset of sexual reproduction

(40x time lapse)


Sexual reproduction in Cymbella (allomixis)

Many of the observations presented on this website deal with the movement of diatoms and the formation of colonies. However, this and the following post are about the sexual reproduction of two species of the genus Cymbella.

At the end of June 2016 I gathered and cultivated a Cymbella species from a small artificial lake (Ebnisee, 48°55'25.5"N 9°36'32.8"E). From apex to apex the length was about 63 μm at that time. The picture of the valves can be seen in the upper half of the top picture of the image gallery (click to enlarge) on the left (100x-objective with oil immersion). I assume that this is Cymbella cistula.

In the middle of November large Cymbella appeared in one of the cultures. It turned out that they were the result of sexual reproduction. As many diatoms reached almost simultaneously the lower limit of the size, sexual reproduction occurs in many places in a culture. In the following weeks, this also happened in other cultures which were cultivated at about the same time or later.

The length of the small diatoms at this time was typically 54 μm, while the large Cymbella (initial cells) measured 120 μm. The ratio of lengths is 2.2. The lower half of the first picture shows the valves a few generations after sexual reproduction. One can recognize the similarities and differences between valves of extreme lengths.

Sexual reproduction begins with the formation of haploid gametes (Geitler (1954), Geitler (1957), Geitler (1967), overview in the book of Round et al. (2007)). The second picture of the image gallery shows two gametes in each of the copulating diatoms. Then fusion takes place between adjacent gametes of the two gametangia, two auxospores develop, which mature into large vegetative cells. So there is a cross-fertilisation (allomixis). In the following picture, the fusion of the gametes seems to have begun. This image combines several images of different focus levels. For this stacking the program CombineZP (open-source software by Alan Hadley) was used.

A pair of copulating diatoms and the process of the gametes fusion were sketched by Lothar Geitler (1954) (click to enlarge):

Reprinted by permission from Springer Nature: Oesterreichische Botanische Zeitschrift, Lebendbeobachtung der Gametenfusion bei Cymbella, Lothar Geitler (1954)

The top left video shows the situation after almost completed development of auxospores. On the side to the auxospores, the empty valves of the small diatoms which have copulated are visible.

The following time lapse video shows the growth of two auxospores (9000x time lapse). It is my impression that gametes and auxospores do not tolerate high light intensities over a longer period of time. The video was therefore captured at very low light intensity. This leads to high image noise. The recording lasted for 22.7 hours.

As the auxospores do not lie parallel to the bottom of the petri dish, a short image sequence is attached after the dark pause, in which the focus is moved through the sample. The structure becomes even clearer with higher

resolution and by use of an image stack, as it is shown in the picture on the left. Here one also easily recognizes the enveloping copulation jelly. The new diatoms are evidently already well developed, because one can see the structure of the valves.

In the combined image only the small valves that are close to the objective remain visible. Behind this are the corresponding other halves. They can be recognized at a lower magnification and thus a higher depth of focus, but then its structure is not dissolved. Stacking with selected images provides a moderately good image. To illustrate the spatial structure, a video was prepared in which the focus level is moved forward and backward through the sample. The upper and lower small valves are alternately visible. Like all other videos, this video can be viewed full-screen. Often the resolution of the microscopic images is insufficient for this, but in this case it is recommended.

It should be noted that an upright microscope was used for higher resolution images of living specimens. For this purpose, the objects from the petri dish were put on a microscope slide using a pipette and a 63x dry objective (Zeiss) was used for observation.

The finished cells of maximum size escape from this copulation envelope. This is probably the first benefit that the new cells get from their ability to move. The activity was recorded in a time lapse video with a speed factor of 75 (one image every 3 seconds, played with 25 fps). This is the resulting video:

The two diatoms slid out in a short time interval. In the second escaping diatom, one can see how the jelly yields elastically. The speed is considerable, but can only be estimated roughly by the frame sequence. Within three seconds (two consecutive frames)  the first escaping diatom covers a distance of about half the cell length, which makes about 20 μm per second. The second diatom begins at first slowly and then arrives at a similarly high velocity, the elastic copulation envelope probably contributing to the driving force.

The number of small diatoms in culture is reduced by copulating and dying small diatoms. The number of large diatoms increases due to sexual and asexual reproduction, so that they quickly dominate. This situation can be seen in the video at the top right.

It is worth mentioning that sexual reproduction is not always successful in cultures. It is no rare phenomenon that one sees dead auxospores. Whether this occurs in nature with similar frequency, we cannot judge. This may also be due to the high light sensitivity of gametes and auxospores.

The observation of the sexual reproduction of this Cymbella gives the possibility to determine the maximum and minimum length and to measure the age of the cell line in the aftermath in units of length. When the generation rate is assumed to be constant, the average length of the diatoms in culture decreases linearly with time. This is, of course, only a rough assumption over the period of many cultures and is valid only as long as the smallest size is not reached again. In addition, the width of the size distribution (in the ideal case binomial distribution) can be tracked over the entire time and compared with the theory.


Note (January 24, 2018):

All cells of the original culture were derived from a single diatom, thus possessing the same genome. Cultivation showed that not all diatoms formed from auxospores had good viability. This could be a consequence of inbreeding.

The viable large diatoms could be further cultivated without any problems. Eleven months after the observations described above, the length of the diatoms again was so small that sexual reproduction began for a second time. The picture below shows a sight into a culture in a region with high local density of sexual reproduction (click to enlarge).



Note (April 02, 2019)

In a second strain of Cymbella cistula the third "round" of sexual reproduction took place in spring 2019. The following video shows a look (30x time-lapse; PlasDIC) into a culture in which many sexual reproductions occur in parallel. The cells of smallest length, initial cells and the formation of pairs of auxospores are clearly visible.


Geitler, Lothar (1954) Lebendbeobachtung der Gametenfusion bei Cymbella. Oesterreichische Botanische Zeitschrift. Vol. 101(1/2). pg. 74-78.

Geitler, Lothar (1957) Die sexuelle Fortpflanzung der pennaten Diatomeen. Biological Reviews. Vol. 32. pg. 261-295.

Geitler, Lothar (1967) Paarung und Auxosporenbildung bei Cymbella. Oesterreichische Botanische Zeitschrift 114(4):484-489 · August 1967

F. E. Round; R. M. Crawford; D. G. Mann (2007), Diatoms: Biology and Morphology of the Genera, Cambridge University Press; 1 edition (2007)

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