Jump to content

Thesis Paper

Posted by Inga, 01 March 2007 · 58 views

By Inga Köhler


Gotland, the largest of the Baltic Islands, is known for its abundance of fossils. A fossil, which is remarkable for its varied size and shape, is called “cat skulls” by the Gotlandish people (Eliason, 1999).

These fossils are stromatoporoids, which are classified among the Porifera. Many specimen show a characteristic banding inside. This so called low- and high banding is known of living Scleractinia and Sclerospongia and is being researched in sclerochronology. As the growth of marine invertebrates is dependant on external influences, sclerochronology offers an important method of exploring reactions of these organisms to environmental factors within the framework of modern climate research.

This study’s aim is to detect general patterns and hierarchies in the banding of stromatoporoids. A detailed analysis of the rhythm – counting and measuring of the individual laminae and bands – will be used to gain information about the speed of growth of reef-building organisms as well as the number of days and lunar phases in the Silurian year.

The sampled stromatoporoids originate from the outcrop Djauviksudden, located approx. 10 km north of Katthammersvik in the north- east of the island. Djauviksudden belongs to the Hemse-layers of the lower Ludlowian.


Numerous publications of the last decades focus on sclerochronology of living and fossil marine invertebrates. In 1963, John W. Wells of Cornell University, New York, states that calculation of the length of a year in the Devonian based on growth banding is possible. Wells fixes a Devonian year at ca. 400 days, a number which roughly corresponds with recent astronomic calculations (Williams, 2000).

Scrutton explores the growth rhythmics of Devonian corals in 1965. His focus is mainly on the causes for various patterns of growth. The surfaces of the specimens he examined were well preserved. Fine growth rings, about 200 rings per centimetre, can be seen there. Thirteen super ordinate bands are discernible which, according to Scrutton, develop within one Devonian year. Each of these rings itself is built up by separate finer growth rings which probably are diurnal increments. Scrutton correlates the data he gathered to the already known astronomic calculations. The diurnal rhythm may be influenced by the tides as well as by the day-night-rhythm, i.e. by light. Apart from the tides, variable food supply and reproduction cycles may be the cause for the rings (Scrutton, 1965). In transparent cuts of the samples bands become visible as well. A light and a dark band represent one year (Knutson et al., 1972). Since 1974, patterns of growth are examined more thoroughly by X-raying invertebrates. The radiographs reveal the lighter bands to have a lower density, the darker ones a higher density (Buddemeier et al., 1974; MacIntyre et al., 1974).

One of the most important studies about Silurian stromatoporoids on Gotland is the one by Young and Kershaw (2005). The authors exemplify how the specific type of banding and a dependence on the seasons can be detected and classified. This diploma thesis follows their basic assumptions, which will be explained in more detail in the chapter “ Material and Methods”.

Many of the Silurian stromatoporoids on the island of Gotland show distinct and constant bands, which are built up of laminae. The bands are interpreted, as in living Scleractinia and Sclerospongia, as annual bands (Young and Kershaw, 2005), with one year consisting of a lighter and a darker band, the so-called Low- and High Density Bands. (Figure 1)

Material and Methods

The samples of stromatoporoids were collected in October 2001 on the coast of the peninsula of Hammarudden in the outcrop “Djauviksudden 7” during an excursion based at the University of Tübingen. The outcrop is situated approx. 5,9 km ESE of Anga on Gotland’s east coast (coordinates 637535 / 167986). Along the coast- line of this region the Hemse-layers have formed in a very marginal marine area. There is a rapid alternation of most different lithologies, e.g. banks with desiccation cracks, wave-ripples, and hard- grounds.


The stromatoporoids were measured and cut in two halves along their main direction of growth and sanded. In a next step the stromatoporoids were scanned with a resolution of 300 dpi. The images were processed with Photoshop, laminae and bands were reworked with the filters “Poster Edges”. The number of laminae and the distance between them as well as the number of laminae per LD- and HD-Band were measured by means of a grid with squares 1mm x 1 mm in size.


Wavelet analysis was carried out with these data. In the graph, the x-axis shows the distance, the y-axis the number of laminae. The coloured Wavelet-card depicts the frequency spectrum of a specific section of laminae. The altitude of the peaks is shown in different colours. The x-axis shows the number of laminae, the y-axis shows the frequency. Statistically relevant areas are bordered in black. These are areas projecting the mean noise with a probability of 95%. Only peaks crossing the dashed line are usable.

Analysis of Banding

Analysis of banding follows Young’s and Kershaw’s (2005) basic approach. Young and Kershaw use X-ray examinations to distinguish between HD and LD bands by visualizing the space between skeletal elements, bulges and additional patterns. A weakening of the organism can cause discontinuity of growth, which results in cracks or sediment inclusions in the sponge. Pressure can cause breaks in bands or diagenetic modifications in stromatoporoids, which on first sight can be interpreted as bands. Besides, several corals and stromatoporoids have only weak or no banding. Specific taxonomic groups have no bandings at all. Usually banded groups, however, can comprise individuals without banding.


148 to 811 laminae were found per stromatoporoid with an average space of 0,23 to 0,25 mm between them. A wavelet analysis resulted in a periodicity of 3 in all stromatoporoids. In some specimens, periodicities of 50 and 100 were found. LD-bands contained 6 to 35 laminae with an interspace of 0,26 to 0,5 mm. HD-bands consisted of 3 to 13 laminae with an interspace of 0,3 to 0,5 mm. A majority of specimens contained a total number of 13 laminae per LD and HD band, with a variable sum of laminae per LD and HD band.


Djauviksudden contains numerous large and dome-shaped stromatoporoids. The samples have substantial similarities in size and shape as well as in and their bandings: 17 of 20 stromatoporoids are larger than 10 cm and have broad LD banding. Great quantity and size of specimens point to advantageous conditions necessary for growth, which result in constant development. Broad LD banding is an indicator for sound growth as well. According to Young and Kershaw, LD bands represent periods of sound growth, HD bands of poor growth. Thus, stromatoporoids with broad LD and narrow HD bands had good conditions for growth. The majority of large samples have broad LD bands. HD bands occur in equal distribution in all samples regardless of their size.

In order to built up a skeleton strong enough for their size, stromatoporoids integrate laminae as prop elements. Kershaw assumes that this process is genetically controlled (Kershaw, unpublished material). If integration of laminae is genetically controlled, the process is assumed to proceed in constant intervals. Thus the interspace between laminae indicates the speed of growth in a specific span of time.

Analyzed in the light of this assumption, the samples reveal variations in their speed of growth, the average growth rate being similar in almost all samples. Most large stromatoporoids have high, small stromatoporoids low growth rates.

A Wavelet analysis of the data produced a periodicity of 3. This rhythm may be caused by periodically repeated environmental changes. In contrast to other periodicities of 200 or 250, a periodicity of 3 is statistically less relevant. A closer scrutiny of number and space between laminae in LD and HD bands was made based on the assumption that laminae are built up in genetically fixed phases with one LD and HD band representing one year. The sum of laminae amounts to 40 in LD and to 13 in HD bands (figure 2). In large stromatoporoids, the number of laminae in LD and HD bands averaged 40. According to astronomical calculations, there were 400 days in a Silurian year (Wells, 1963) which points to a ten days’ cycle with a growth rate of 2 cm a year. Another explanation may be disruptions in growth. As formation of HD bands is presumably caused by unfavourable environmental conditions, growth may have been disrupted during those periods. This may result in complete “omission” of HD bands, which makes HD bands look broader. Furthermore, there is a chance that there were no inhibitions to growth in particular years, so that no HD band was formed. With regard to the relation between size of the stromatoporoids and number of LD bands, another theory seems more plausible. Here, too, a kind of “omissions” of HD bands is responsible for large LD sectors. Due to advantages of location, negative environmental changes may have been less distinctive for some specimens. Stromatoporoids with a favourable location were not influenced by environmental changes, which initiated HD banding in others. Only under extremely bad environmental influences these specimens may have formed HD bands as well (figure 3). Conversely, weak stromatoporoids or such with an disadvantageous location may have incorporated additional “stress bands” (figure 4). Due to this “omission” and/or additional formation of HD bands it cannot be stated with certainty that one LD and HD band together represent one year. Several of the specimens showed a number of laminae about 13. Given that the Silurian year had about 13 months (Scrutton, 1965), this could indicate to the stromatoporoids forming one lamina per month with a growth rate of 0,5 mm per month and 2 to 10 mm per year. This rate corresponds to the average growth speed of 3 to 7 mm per year identified by Young and Kershaw (2005). In these cases one LD and HD band together would represent one year (figure 5). Taken this as a basis, the results of wavelet analysis point to a quarterly cycle in case of a periodicity of 3, periodicities of 200 and 250 point to a cycle of 15 and 20 years respectively.

Four specimens are quite large in spite of a low growth rate (figure 6). It can be assumed that these stromatoporoids grew in detached areas of the outcrop. Food and light supply were limited so that growth was inhibited. However, a detached habitat protects against fluctuation in environmental conditions, which had positive effects on the span of life of these specimens. Broad LD areas can be taken as an indicator for uninhibited growth. Where fluctuations in environmental conditions lead to the formation of HD bands in samples taken from exposed patches, the protected stromatoporoids remained widely unaffected. The almost constant number of HD bands suggests that environmental conditions lasted for about the same time every year.

The differences in density are partly caused by the smaller space between laminae, as the scans and acetate peels show (figure 7 and 8). Besides, there is a noticeable change in colour to the darker in HD bands. This can – besides incorporation of organic material, iron, or vanadium – result from strengthening of other skeletal elements, such as dissepiments or pilae, which, however, could not be found in the samples. Such a strengthening could be the result of increased separation of carbonate, caused by an increased solar radiation or an increased pH-value within the skeleton. The diminished space between laminae in HD bands points to a slowed linear growth of the stromatoporoids in this period. With broader space between laminae, LD bands represent a time of sound growth.

As Young and Kershaw (2005) explored in their study, the relation between LD and HD bands could give information about the seasonal formation of bands. Constant results in stromatoporoids of the same species point to formation of laminae dependent on the seasons. Except for the samples of Parallelostroma typicum/”Stromatopora” bekkeri the so-called L/H-relations are constant in all species, which is indicative of a seasonal cause for bandings in general. Vertical skeletal elements continually grown beyond borders of LD and HD bands provide additional evidence for seasonality.

In this context, the average number of 5 laminae per HD band is striking. Provided that o0ne lamina was built per month, this would mean a five-months’ disturbance in growth. Which seasonal factors were active in the formation of bands in stromatoporoids on Gotland in the Silurian can only be surmised. Endogenous factors such as reproduction, parasites or diseases can have effects on growth as well. Small variations of salinity, sediment inflow, concentration of nutrients, water temperature, and light intensity are to be considered as exogenous factors. Young and Kershaw explain that differences in light supply were responsible for bandings and that the relation between LD and HD bands can be used as an indicator for the quantity of light available. As the stromatoporoids analyzed here did not show any sign of zooxanthelles, this theory is quite conceivable, but cannot be proved. It is certain, however, that the relation between LD and HD bands in stromatoporoids give information about the circumstances of growth. High results in L/H relations indicate sound growth, low results indicate poor growth. Which of the factors affected the growth of the stromatoporoids, can only be verified by logic interpretation of evidence.

The data gathered in this diploma thesis together with astronomically proven 13 months in Silurian year indicate that one LD and HD band together can but do not inevitably have to represent one year. The data verify that stromatoporoids can form several HD bands per year or none at all. This inhomogeneity in growth indicates that in spite of the assumed genetically control of the incorporation of laminae, growth of stromatoporoids is essentially influenced by specific factors which depend on the specimen and/or is location as wells as on environmental factors.

October 2017

151617 18 192021

Recent Entries

Recent Comments

My Picture

0 user(s) viewing

0 members, 0 guests, 0 anonymous users


    Search My Blog