English Deutsch Content

Grosser Mueggelsee (Großer Müggelsee)

About Grosser Mueggelsee: the perfect lake

geographical location of the urban lake

Grosser_Mueggelsee-TeubnerLake Grosser Mueggelsee, 1995:
View from the south-shore onto the shallow polymictic lake. Plenty of sailing boats are seen on this summer day.
Lake Grosser Mueggelsee (52°26’5.56''N, 13°38’6.2''E) is located in Berlin in Germany, at 45 m above sea level. It  is a small lake with a volume of 36 m3 and an almost circular area of 7 km2. The maximum depth of the lake Grosser Mueggelsee is 7.7 m only. The mixing regime of this shallow lake is hence quite frequently a year, described as polymictic. The water retention time is about 67 days, much shorter than for large deep lakes that are described for the alpine region on this website (one to seven years mean retention time, see e.g. Ammersee S, Attersee S, Mondsee S und Traunsee S). Another shallow lake in the close neighborhood of Grosser Mueggelsee, lake ‘Langer See’ (52°23’54''N, 13°38’2.96''E) , has a very elongated shape as seen on birdview photo below. This lake has further an even shorter retention time of about 4.13 days only (mean retention time of both lakes relates to the period 1992/1993; Kohl et al. 1995, Table 1 in Teubner et al. 1999 R, Table 1 in Teubner & Dokulil 2002 R). The lake-like enlargement of the river bed provides an example for a water body that is far from being a lake but it is also not just a river. Further examples of such water bodies nearby Großer Müggelsee and Langer See are Seddinsee (52°23’4.7''N, 13°40’53''E) and Flakensee (52°25’55''N, 13°45’48.7''E) having again a short mean retention time of only 13 or 29 days respectively. Lakes having a short theoretical retention time of about 3-30 (to 70) days are hence called riverine lakes (in German 'Flußseen'). Water basins having a shorter and a longer water residence time form river ecosystems and lake ecosystems, respectively, and are accordingly defined as rivers or lakes. Riverine lakes are common in the eco-region around Berlin, in Brandenburg and Mecklenburg, the floodplains e.g. of the rivers Spree, Dahme and Havel.

Langer_See-TeubnerThe 'riverine lake' Langer See, view from the tower 'Mueggelturm', 1995:
From this bird's-eye perspective, the elongated shape of the water basin is well seen. It rather looks like a river than a lake. It is actually a lake-like enlargement of the river bed.
The shorter the retention time, the greater is the impact of washing-out on plankton species. Losses by washing out have to be compensated by growth to survive in the habitat. In particular, some zooplankton species that have a life span of about 30 days, need days to weeks for the development from an egg to an mature adult stage and hence can grow well in case the water retention time is longer than 30 days. In this way - among other aspects - the flow velocity and water retention time have the potential to impact the quality and length of the food chain and thus may alter the phytoplankton composition via top-down predation effects. The riverine lakes are often conncted in the close neighorhood and hence a riverine lake with longer retention time of water or a river with retention shore areas can surve as niche or hatching zone. Rotifers and crustacean zooplankton (Cladocera and Copepoda) are common in the both mentioned riverine lakes, Grosser Mueggelsee and Langer See (Fig. 7 in Teubner et al. 1999 R). These the zooplankton species developed high abundances in the water bodies, which were mainly characterized by blooming cyanobacteria throughout growing season at that study period in 1992/1993 (cyanobacterial dominance in spring by Limnothrix redekei & Planktothrix agardhii, from summer to autumn mainly by P. agardhii or by Aphanizomenon flos-aquae & Microcystis spp., see phytoplankton further below). The phytoplankton loss rates by grazing (grazing rates) were estimated by feeding experiments with filamentous cyanobacteria (Planktothrix agardhii) in spring in 1993. The grazing loss rates were 0.17 d-1 for the strongly flushed riverine lake Langer See and almost twice high, namely 0.3 d-1 in riverine lake Grosser Mueggelsee having an about two-month water retention time (pages 334-335 in Teubner et al. 1999 R).
Phytoplankton species usually grow much faster than zooplankton. As mentioned elsewhere on this website, ‘natural’ phytoplankton cells can achieve every day (24 hours) or at least every second or third day one cell division. During seasonal periods of unfavorable growth condition, the cell division can be delayed over weeks even for planktonic algal organisms in an aquatic ecosystem. Algal cultures grown in the lab are different as they are usually adjusted to lab reference conditions that enable right one cell division a day (lab treatment within a certain range of nutrient concentrations, light intensities and temperatures).
As mentioned before, in general riverine lakes are characterized as flushed shallow lakes. These ecosystems do not leave the transient stage between a river or a lake passing the seasonal cycle a year. Other types of water basins, however, can even switch seasonally between being a river or a lake. Such an example is described by the subtropical shallow lake Poyang S in the river basin of Yangtze S in China on this website.

Like many shallow urban lakes in the world (see e.g. Old Danube S in Austria, Taihu S and Dianchi S in China on this website), the riverine lakes nearby Berlin underwent large ecosystem changes due to nutrient-enrichment by external phosphate loading from the catchment over decades. The switch point of external nutrient load in the described riverine lakes is historically linked to the year of the fall of Berlin Wall in 1989, as the economy shifted and hence trends of nutrient loading/pollution income turned around. The riverine lake descriptions on this website depict the four-year study period from January 1990 to December 1993 only and cover still the nutrient-enriched internal lake situation, quite different from less nutrient rich periods lots of decades before and of recent times.

to meet a perfect lake: seasonally balanced nutrient proportions suit perfectly the requirements of phytoplankton growth in mueggelsee

The relative quantity of nutrient elements of phytoplankton cells grown in a natural aquatic system is not by random but within a certain narrow range. The elemental composition of phytoplankton was described by Redfield stoichiometry (1958) for the ocean, called the Redfield ratio (C:N:P=106:16:1). The validity of this ratio for other aquatic habitats, other aquatic biota and other elements (N:P:Si=16:1:17 see Harris, 1986) was extensively discussed in the following years. The main nutrient elements nitrogen, phosphorus and silicon used to build-up phytoplankton biomass, are hence not utilized by phytoplankton cells by the same amount of each element (1:1:1, see Fig.14 on page 26 in Teubner 1996 R and Fig.3 in Teubner & Dokulil 2002 R) but rather close to the molar proportion of N:P:Si=16:1:17. Nutrient ratios are often assessed by x-y-plots of individual pairs of elements, as the N:P, the Si:P and the Si:N ratio. Graphs displaying together all three nutrient elements in an x-y-z plot are of the same low information and are even trickier to visualize by the three-dimensional display. A direct way for interrelated stoichiometry between the three main nutrient elements is revealed by triple ratios displayed in trigonal plots (see method and  Fig.1 in Teubner & Dokulil 2002 R). Such ratios, as N:P:Si, have the benefit of presenting multiple resource-ratio gradients and hence provide a more synoptic view than individual ratios as N:P, Si:N and Si:P. For the reason of short turnover time, ecological lake stoichiometry is commonly NOT described by the soluble reactive phosphorus fraction (this phosphorus fraction can be utilized by algae) or dissolved inorganic nitrogen (nitrate, nitrite, ammonia), but by the total pool of all fractions of phosphorus and nitrogen. In particular, in case of rapidly recycled phosphorus, common sampling methods are not really appropriate to follow the high-resolution distribution pattern of small spatial and short temporal scales of SRP in a lake. Different from P and N, in the case of silicon the physiologically relevant fraction is the dissolved fraction of soluble reactive silicon (see different turnover times for N, P and Si in the section for lake Traunsee). The triple molar ratio TN:TP:SRSi=16:1:17 can be used as a reference point for ecological stoichiometry (Teubner & Dokulil 2002 R), called the ‘optimum ratio’, in the sense of Redfield (1958) and Harris (1986) for plankton communities. Displaying the TN:TP:SRSi ratio in trigonal graphs, an axis scaling in the proportion of 16:1:17 is most appropriate and shifts the optimum point of TN:TP:SRSi=16:1:17 graphically to the triangle centre (see below the concept of the ‘balance of TN:TP:SRSi-ratios’ in lakes, Teubner & Dokulil 2002 R). Such triangular diagrams scaled in the physiological proportion of 16:1:17 aim at synoptically presenting relative nutrient availability for both diatoms and non-siliceous algae in phytoplankton communities (Fig.15 on page 28 in Teubner 1996 R, Fig.4 in Teubner & Dokulil 2002 R, for Old Danube Fig.5 E-F in Teubner et al. 2003 R, for Traunsee Fig.5 B, C, E in Teubner 2003 R).

Commonly, a one element (i.e. TN or TP or SRSi) is seasonally invariant relative to the remaining two elements in a lake. These are lakes with ‘imbalanced nutrient ratios’ (Teubner & Dokulil 2002 R). Lakes where TN:TP:SRSi ratios fluctuated evenly around the ecological reference point of TN:TP:SRSi=16:1:17, in a cyclic pattern within a given year, are the exception rather than the rule (lakes with ‘balanced nutrient ratios’). Assuming that the optimum ratio 16:1:17 indicates average requirements of algae in the plankton communities, it is not surprising, that the lakes with balanced nutrient ratios yield the highest algal biomass in comparison with other lakes of the same trophy (see hyperthrophic lakes LANS and MUES and its inflow MUEZ: the triple nutrient ratios are shown in Fig.4A and the TP:chlorophyll-a -response in Fig.2A in Teubner & Dokulil 2002 R). The shallow lake Müggelsee with high annual phytoplankton biomass for the nutrient-rich period 1990-1993, provides an example for a lake with balanced nutrient proportions. In that study period of the early nineties, the three nutrient elements in lake Mueggelsee had a stoichiometry that suited perfectly the requirements of phytoplankton growth.

The Redfield Ocean is seen as the ‘perfect sea’ due to a balanced flow of C, N and P in and out of the biota. In the context of stoichiometric ecology, lake Grosser Mueggelsee stands for the ‘perfect lake’ (see text on page 6 in Teubner 2004) for three reasons: (i) the nutrient-resource situation, described by TN:TP:SRSi ratios (1990-1993), shifts evenly around the stoichiometric optimum of 16:1:17 within a year and (ii) the elemental ratio of biota (stoichiometry of particulate organic matter, POM, POC:PON:POP) is very close to C:N:P=106:16:1. An overlay of both seasonal patterns, TN:TP and PON:POP, mirrors the complementary relationship between external and internal stoichiometry of plankton in Grosser Mueggelsee. Such stoichiometric shift towards the limiting element seems to be a common phenomenon of individual adaptation of producer organisms and can be even recognised on an ecosystem level (more details see Teubner & Dokulil 2002 R; and  Fig.5F and text on page 1147 in Teubner et al. 2003 R).

seasonal plankton dynamics: how accurate can a ‘phytoplankton forecast’ be for lake mueggelsee?

Summer phytoplankton in nutrient-rich lake Grosser Mueggelsee was commonly dominated by the cyanobacteria Aphanizomenon flos-aquae and Microcystis spp., while alternatively in a neighboring shallow lake Langer See the cyanobacterium Planktothrix agadhii was mainly developed (study period 1990-1993). A sensitive moment for the differentiation of the plankton development to the one or the other cyanobacterial summer bloom was the time in the year (Julian day), when the total nitrogen to total phosphorus ratio, the TN:TP ratio, dropped below the critical threshold value of 16:1 (Figs. 39-40 on page 110-111 in Teubner 1996 R, Figs. 1-2 in Teubner et al. 1999 R). In addition, the phytoplankton composition at this critical moment was of decisive importance. Rapid growth of the N2-fixing A. flos-aquae was favoured at TN:TP<16:1 in both lakes, when the timing of the critical TN:TP ratio and low biomass of P. agardhii due to the clear water phase coincided. In all four years studied the lake Mueggelsee, the rapid growth of the heterocyst-forming cyanobacterium A. flos-aquae started at the time when TN:TP was equal to 16:1, even in those years when this critical ratio was delayed by several weeks. In some years, however, the spring biovolume of P. agardhii was already quite high that early in the year. In such years, P. agardhii exceeded already the biovolume of 6 mm3 L-1 at the time when the critical TN:TP ratio was reached. The mass development of this cyanobacterium was then further continued, the summer into autumn, whereas A. flos-aquae was only present in traces during the growing season. This alternative blooming of these two cyanobacterial regimes was further associated with different planktonic diatom assemblages. A summer-autumn plankton dominated by A. flos-aquae was associated with filamentous diatoms Aulacoseira spp, while Planktothrix agardhii with Stephanodiscus hantzschii, Cyclostephanos dubius and Actinocyclus normanii.

To summarize, it seemed to be a simple story to predict in late spring what cyanobacteria in summer would grow in such eutrophied shallow lakes of short retention time. Actually, even just described for this four-year study period in the nineties, this rule of alternative blooming of cyanobacterial regimes (Teubner et al. 1999 R) was seen also for other years in these two lakes. It was somewhat of a scientific gamble in spring, to project how phytoplankton situation will evolve in the very next days with the on-set of summer. The weather forecast is common, but it seems that under such a certain circumstance, some phytoplankton forecast s work as well?! In view of aquatic science, the timing of events, e.g. the date in the year passing a certain threshold of light availability, water temperature or nutrient concentration, is commonly studied for lake ‘phenology’. Such aspects are most relevant to the study of the climate response of lakes (see Mondsee S and Ammersee S).

seasonal phytoplankton structure: the only two principal periods a year

ParsteinerSee-TeubnerParsteiner See in the Biosphere Reserve Schorfheide-Chorin, in the north of Berlin, 1990:
The annual mean of Secchi depth was 4.4 m for this deep mesotrophic lake. The lake was one of 11 sites that had been examined in a limnological study describing the phytoplankton dynamics in north Germany. The mean Secchi depth of the both riverine lakes Grosser Mueggelsee and Langer See was much lower during this study, only 0.9 and 1.6 m, respectively (investigation period 1990-1993).
Beside Grosser Mueggelsee and Langer See, nine other mainly shallow lakes were studied in the nineties in the Berlin-Brandenburg region (see map in Fig.1 in Teubner 1996 R). The majority of these temperate lakes were shallow and covered trophic states from mesotrohic to hypertrophic. In the vicinity of urban area of Berlin, lake Grosser Mueggelsee and lake Flakensee and its both inflows, lake Langer See, the groundwater-seepage lake Kiessee (52°39’9.4''N, 13°22’59''E) and the dystrophic lake Krumme Lake (52°25’5.2''N, 13°41’17.39''E) were studied. Furthermore, three mesotrophic lakes in the north of Berlin, the Biosphere Reserve Schorfheide-Chorin, were included. These were two dimictic lakes Parsteiner See (52°55’48.6''N, 13°59’7.7''E) and Rosinsee (52°53’28.2''N, 13°58’27''E) with a maximum depth of 27 and 9 m respectively and one shallow, slightly dystrophic lake, Grosser Plagesee (52°53’16.8''N, 13°56’16.7'E; Table 2 in Teubner 1996 R, Table 1 in Teubner & Dokulil 2002 R, Table 1 in Teubner 1997 R).

The taxa found in the 11 water bodies, referred mainly to the cyanobacteria, diatoms (Teubner 1995 R, Teubner 1997 R) and chlorophytes. Common species during that study are illustrated by microscopical photographs (pages 57-67 in Teubner 1996 R, diatoms only on  pages  238-247 Teubner 1997 R). The individual sites were studied over 3 to 4 years from 1990-1993, which accounts for ‘34 lake-years’ (page 7 in Teubner 1996 R). The two main results of phytoplankton seasonality found for these sites are described in the following paragraphs.

The one outcome relates to the seasonal change in the size structure of phytoplankton assemblages. After spring overturn of the water body and therefore, at the time of the replenishment of nutrients from deeper water into the surface layer, mainly small short-lived forms dominate the assemblage. At this time, the so-called ‘bottom up effects’ control the phytoplankton development, that mainly small fast growing phytoplankton species become predominant. According to allometric rule (i.e. here that cell physiology depends on cell size), the fraction of small-sized cells of phytoplankton can achieve a higher photosynthetic efficiency than that of large-sized cells (see 14C measurements on phytoplankton from the alpine region: Lake Lucerne, Traunsee and Mondsee, table 2 in Teubner et al. 2001 R). The small cells are hence adjusted to low underwater light intensities. They benefit from low incoming radiation as typically found in spring. This situation early in the year usually coincides with the nutrient replenishment by overturn (mixing of the water body by wind in spring) or by external nutrient load from the catchment. The advantage of being small was found to be in accordance with their cellular pigment ratio, of having relatively high concentrations of light-harvesting chlorophyll-a but low of light-protective ß-carotene (Fig.8 in Teubner et al. 2001 R). The opposite applied for the large cells of phytoplankton assemblages. They accomplished a lower photosynthetic efficiency which was associated with a lower pigment ratio of chlorophyll-a to light-protective ß-carote. They hence indicated an adjustment to high under water light intensity. Large cell forms or colonial forms with a longer life span are rather common in summer, in particular, at the growth period immediately after a clear-water phase. This period relates therefore, primarily to a ‘top down control’, i.e. the effects by selective grazing pressure of zooplankton on phytoplankton. The dynamic of changing size structure with seasons could be illustrated by the annual time-course of the surface to volume ratio of phytoplankton (pages 79-86 in Teubner 1996 RTeubner & Dokulil 2000 R, see also seasonal phytoplankton develpoment discussed for see Bergknappweiher S). This ratio increased from winter to spring, reaching often even an annual peak before it was when abruptly declining within few weeks (Fig.22 A-C on page 79 shows examples for lake Grosser Müggelsee and its inflow 'Müggelsee-Zufluß', and lake Langer See, in Teubner 1996 R). With the exception of the pico-phytoplankton size fraction, which is defined by a cell size smaller than 2µm and not studied here, higher ranked taxa as the Ulotrichales (needle shaped green algae), Oscillatoriales (non-colony forming trichomes of some cyanobacteria) and Pennales (needle-shaped diatoms) have exceptional high surface to biovolume ratios (Fig.23 on page 82 in Teubner 1996 R). Examples of taxa of low surface to biovolume ratios are the dinoflagellates. Large differences also can be found with a phytoplankton group. The thin trichomes of the common cyanobacteria Planktolynbya limnetica and Limnothrix redekei have a much higher surface to volume proportion than the cyanobacterial trichomes of Anabeana taxa (Fig.24 on page 83 in Teubner 1996 R) . Further within the diatoms, the pennate Nitzschia acicularis or the small centric diatoms of Cyclotella atomus or Stepahonodiscus pseudostelligera, C. parvus and C. minutulus indicate much higher surface to volume ratios than the large cells of unicellular centric diatom Actinocyclus normanii and the filamentous forms of centric diatoms, Melosira spp. (Fig.25 on page 84 in Teubner 1996 R). The annual mean values of surface to volume ratio varied among the 11 sites. These values, however, were statistically NOT significant different while the nutrient state varied largely among the water bodies (Fig.26 on page 85 in Teubner 1996 R, Table 1 in Teubner & Dokulil 2000 R). It can therefore be concluded that the surface to volume ratios of phytoplankton was not linked to the trophic state but mirrors the general pattern of intra-annual phytoplankton succession as mentioned before for the annual time courses in this paragraph.

The second pattern of phytoplankton seasonality refers to the timing of the compositional shifts within the year. This study focused on cyanobacteria and diatoms, as these phytoplankton taxa were common in the 11 studied water bodies. It could be found for the ’34 lake-years’ that the composition of winter and spring phytoplankton, on the one hand, and of summer and autumn phytoplankton on the other were statistically quite similar. Further, the winter and spring phytoplankton was statistically far different composed from those in the summer-autumn period. Therefore, significant compositional changes for both algal classes occurred concurrently two times a year only, i.e. during the transition from spring to summer and from autumn to spring (Figs.36&56 on pages 104 & 136 in Teubner 1996 R, DCA-plots of Figs.4&5 in Teubner 2000 R, see also seasonal phytoplankton develpoment discussed for see Bergknappweiher S). This reduction of seasonality from four to just two principal phytoplankton assemblages a year coincided with the seasonal pattern of the TN:TP-ratio, while those of SRSi:TN and SRSi:TP proportions varied among sites dependent from individual lake basin morphometry and the geological background (TN = total nitrogen, TP = total phosphorus, SRSi = soluble reactive silicon). The interpretation of the nutrient status by the dissolved fraction as for silicon on the on side and by the total pool as for nitrogen and phosphorus on the other, refers mainly to the different turnover time of these three nutrient elements and is in greater detail discussed for the lakes Mondsee S, Traunsee S and Old Danube S on this website.

citations for mueggelsee & related lakes

Dokulil, M., K. Donabaum Teubner, K. 2007. Modifications in phytoplankton size structure by environmental constraints induced by regime shifts in an urban lake. Hydrobiologia, 578: 59-63. doi:10.1007/s10750-006-0433-4 Abstract OpenAccess

Teubner, K. 2004. More or less? Smaller or bigger? How relevant are relative changes in aquatic ecosystems? Habilitation thesis on Ecological Stoichiometry, Fac. of Sciences and Mathematics, Institute of Ecology and Conservation Biology University Vienna: 188 pp.

Teubner K, Crosbie N, Donabaum K, Kabas W, Kirschner A, Pfister G, Salbrechter M, Dokulil MT (2003) Enhanced phosphorus accumulation efficiency by the pelagic community at reduced phosphorus supply: a lake experiment from bacteria to metazoan zooplankton. Limnol Oceanogr, 48(3):1141–1149 Look-Inside OpenAccess

Teubner, K. & M. T. Dokulil. 2002. Ecological stoichiometry of TN:TP:SRSi in freshwaters: nutrient ratios and seasonal shifts in phytoplankton assemblages. Archiv für Hydrobiologie (now: Fundamental and Applied Limnology), 154 (84): 625-46. Look-Inside   FurtherLink 

Teubner, K. 2000. Synchronised changes of planktonic cyanobacterial and diatom assemblages in North German waters reduce seasonality to two principal periods. Arch Hydrobiol, Spec Iss Adv Limnol  55: 564-80. Look-Inside FurtherLink 

Teubner, K. M. T. Dokulil. 2000. Seasonal dynamic of surface:volume-ratio of phytoplankton assemblages. Verh int Ver Limnol, 27, 2977-78. Look-Inside  

Teubner, K., Th. Teubner & M. T. Dokulil. 2000. Use of triangular TN:TP:SRSi-diagrams to evaluate nutrient ratio dynamics structuring phytoplankton assemblages.Verh int Ver Limnol, 27, 2948. Look-Inside  

Dokulil, M. & K. Teubner. 2000. Cyanobacterial dominance in lakes. Hydrobiologia 438: 1-12. Abstract FurtherLink 

Teubner, K., R. Feyerabend, M. Henning, A. Nicklisch, P. Woitke & J.-G. Kohl. 1999. Alternative blooming of Aphanizomenon flos-aquae or Planktothrix agardhii induced by the timing of the critical nitrogen-phosphorus-ratio in hypertrophic riverine lakes. Arch Hydrobiol, Spec Iss Adv Limnol , 54: 325-344. Look-Inside FurtherLink 

Teubner, K.1997. Merkmalsvariabilität bei planktischen Diatomeen in Berlin-Brandenburger Gewässern. Nova Hedwigia, 65 (1-4): 233-50. Look-Inside FurtherLink 

Teubner, K. 1996. Struktur und Dynamik des Phytoplanktons in Beziehung zur Hydrochemie und Hydrophysik der Gewässer: Eine multivariate statistische Analyse an ausgewählten Gewässern der Region Berlin-Brandenburg. Ph.D thesis, Dept. Ecophysiology, Humboldt University Berlin: 232 pp. Look-Inside FurtherLink 

Woitke, P., T. Schiewitz, K. Teubner & J.-G. Kohl. 1996. Annual profiles of photosynthetic pigments in four freshwater lakes in relation to phytoplankton counts as well as to nutrient data. Arch Hydrobiol 137: 363-84. Look-Inside FurtherLink 

Teubner, K. 1995. A light microscopical investigation and multivariate statistical analyses of heterovalvar cells of Cyclotella-species (Bacillariophyceae) from lakes of the Berlin-Brandenburg region. Diatom Res, 10 (1): 191-105. Look-Inside FurtherLink 

Kohl, J.-G., A. Nicklisch, G. Dudel, M. Henning, H. Kühl, P. Woitke, K. Luck, K. Teubner, T. Schiewitz, R. Feyerabend, H. Haake & T. Rohrlack. 1995. Ökologischer Zustand und Stabilität von Flußseen von Spree und Dahme und ihre Reaktionen auf Belastungsän-derungen. Abschlußbericht zum Forschungsvorhaben im Auftrag des Bundesministeriums für Forschung und Technologie mit dem Kennzeichen BEO 339400A, Berlin .

Harris, G. P. 1986. Phytoplankton ecology. Structure function and fluctuation. Chapman and Hall.

Redfield, A. C. 1958. The biological control of chemical factors in the environment. Amer Sci, 46: 205-222..