FECUNDITY ASSESSMENT OF VENDACE, COREGONUS ALBULA L. FROM SIX LAKES IN POLISH WESTERN POMERANIA

Vendace acquired for the fecundity analysis originated from six west Pomeranian lakes: Pile (97 specimens), Ko­ morze (41), Drawsko (62), Pelcz (89), Morynskie (91), and Lesne (70). The fish were caught during commercial catches using anchored gillnets, 24-mm mesh size. Substantial differ­ ences were stated in individual biological parameters between fish representing different lakes, even if the dominant age group everywhere were 2+ vendace. The fish of Drawsko Lake showed the highest individual weight and the total length. The highest fecundity was stated in fish from lakes Komorze and Drawsko (7.21-16.85·103 and 6.9-22.23·103, respectively), whereas in Pelcz Lake this parameter reached the value as low as 1.61-4.12·103 • All vendace exhibited a high, statistically significant correlation between the absolute fecundity and their total length.


INTRODUCTION
Fecundity of fishes is one of the basic biological features enabling survival and con tinuation of species and may also have a divisive effect on their population size. Because of the diversity of aquatic environment, the differences in amounts of deposited eggs are ob served not only in individual species but also in populations from various bodies of water. Determination of fish fecundity is particularly significant in species of short life cycle, which are very vulnerable to overfishing related to bad fisheries practice. V endace, attain ing its market size within 2-3 years is currently one of the most economically important items of Polish ichthyofauna. It has been known for its fast growth rate, relatively high adaptive potential to different water bodies, and also shows schooling behaviour, which in turn facilitates catch and regulation of population structure. The above-mentioned advan-tages and highly priced meat of this fish have contributed to the fact that many fisheries enterprises switched to a "vendace model" of lake management. Rational fisheries man agement implemented in "vendace-type" lakes should consider the values of basic biologi cal parameters offish from different bodies of water, including their reproductive potential.
The majority of studies assessing fecundity of vendace were (was) carried out in Masuria (Ciepielewski 1974;Demska-Zak�s and Dlugosz 1995) and Wielkopolska (Budych and Iwaszkiewicz 1964;Mastynski 1978). Very few data on values of this pa rameter are available on fish representing lakes of Polish W estem Pomerania. Some ran domly-taken and already outdated information on the quantities of roe deposited by ven dace in lakes Kalensko, Bucerz, and Krzemien was provided only by a single author (Walczak 1953).
The aim of the present work was to assess the absolute-and relative fecundity of vendace collected in autumn from a number of commercially exploited by fishermen lakes of Polish Western Pomerania and comparing them with literature data.

MATERIAL AND METHODS
The material for the present study constituted vendace acquired in autumn of 1999 and 2000 from six west Pomeranian lakes: Pile, Komorze, Drawsko, Pelcz, Morynskie, and Lesne ( Fig. l)

���������������
The above-mentioned fish were caught with anchored gillnets of the mesh size of 24 mm. Subsequently, after transporting to the laboratory of the Department of Fisheries Management of Inland Waters, Agricultural University of Szczecin the were individually weighted to the nearest 0.1 g (on an electronic balance "Axis") and measured (total length)-with an electronic slide calliper coupled with a computer. Also their condition status was determined according to commonly accepted Fulton formula. Fish age was de termined from scales collected from a defined area located above the lateral line, between dorsal-and anal fin (Bernatowicz 1952). The scales were analysed based on an image formed by a microfilm projector (l 7x magnification). The first more-regularly suboval sclerite, appearing after a series of "horseshoe sclerites" characteristic for juvenile coregonids (Hogman 1968), was assumed as the first annual sclerite. Each fish was aged based on independent readings of three scales.
Absolute-and relative (related to 100 g of fish weight) fecundity of fish from indi vidual lakes were determined by "wet" weight method. After determining the total weight of a gonad, two samples of 0.2-03 g were cut with a scalpel from a central part of this or gan. The samples were subsequently weighed to the nearest 0.0001 g and placed in glass vials and flooded with formalin (4-5%). After 2 minutes of intensive shaking the parts of membranes surrounding the gonad were separated from eggs. Eggs in each sample were counted automatically with the aid of digital camera, microcomputer, and "Multiscan" software. In addition to the counting function, this system allowed a focus control and im provement of picture quality, which translated into enhancement of counting accuracy. Ab solute fecundity offish was presented in a table, separately for different age groups. Vari ance analysis (ANOV A) was used for demonstrating and interpretation of differences, similarly as hierarchical agglomeration analysis where distance between analysed objects is Euclid's distance. Value of the latter parameter translates into a geometric distance in a multi-dimensional space. At the same time a regression analysis (Sokal and Rohlf 1998) was used to determine relation between the absolute fecundity (Fa) and individual charac ters of fish (length and age). In addition a gonadosomatic index ( GS/) was determined, de scribing the percentage share of gonads in the total weight of a fish.

RESULTS
1bree-year-old fish dominated in the vendace samples of all ( except Drawsko Lake) lakes. They constituted between 51.4% (Lesne Lake) and 97.6% (Komorze Lake) of ac quired vendace females. Only in Drawsko Lake fish aged 2+ constituted a majority of 41.9%. It can be assumed that the observed age structure resulted from employment of an chored gillnets of the same mesh size (24 mm) in all lakes. Despite considerable selectivity of this fishing gear, some differences were apparent in abundance of the remaining age groups. For instance in Lesne Lake the share of2-year-old fish was high and it amounted to 4 7.1 %, whereas the samples from the remaining bodies of water it was much lower and it did not exceed 8%. Also the number of fish aged 3+ was relatively low and in the analysed material only the sample form Drawsko Lake contained 41.9% of vendace representing this age group (Table 2). Table 2 Total length (mm), individual weight (g), and age structure ofvendace studied  Table 2 contains also the remaining data on the total length, individual weight, and condition factor of vendace from respective lakes. The values of the above-mentioned pa rameters show substantial differences between the fish collected. The highest level of dif ferences was observed between vendace of lakes Pelcz and Drawsko. The total length of fish of the former body of water, compared to that of fish of Drawsko Lake was by 26 per centage points shorter and the Pelcz lake vendace had almost 3 times lower individual weight. The reason behind the above-mentioned differences were different environmental conditions of these bodies of water. Highly eutrophic Pelcz Lake, known for oxygen defi-ciencies below than 7 m (in 2000, according to Czerniejewski and Filipiak 2002) the ven dace in catches is small of low condition factor (1.16). In the remaining lakes ( except Ko morze Lake) the vendace acquired were characterised by a relatively good Fulton coeffi cient (> 1.20).
Data on the absolute fecundity along with the results of variance analysis (ANOV A) are shown in Table 3. In all lalces where vandace were caught, these fish attained sexual maturity as early as at the age of l +. Despite the lack of specimens of this age group in the analysed sample of lakes Komorze and Drawsko a similar conclusion can be drawn based on earlier published records as well as on a personal information obtained from the lessee of those bodies of water. Because as many as 76. l % of acquired vendace constituted fish aged 2+, the present work concentrated on describing results pertaining to this age group as the most reliable. The mean absolute fecundity of fish of this age from lakes of Western Pomerania varies in wide range: from 2936 eggs (Pelcz Lake) up to 8516 (Komorze Lake). It must be concluded from the results of LSD test (p < 0.05) that vendace from lakes (Ko morze and Drawsko constitute a uniform group which shows higher fecundity and this dif ference is statistically significant. It is possible that the above difference has been influ enced by environmental factors rather that geographical ones. The above statement may be confirmed on one hand, by the fact that Pile Lake situated also in the Drawskie Lake Dis trict has not been assigned to lakes of highest fecundity of vendace. On the other hand sta tistically significant differences were found in absolute fecundity between lakes Lesne and Morynskie located nearby Pelcz Lake (Fig. 2). A similar variability of absolute fecundity of vendace can be found also in fish aged 3+ (Table 3). In all lakes the vendace fecundity in creases with fish age. The correlation, however, between those parameters is week and only for the fish from Drawsko Lake, the correlation coefficient is high and it amounts to 0.88. Analyses of relation between absolute fecundity (Fa) and the total length (TL) were performed employing a linear function and a power one (Table 4). High correlation coeffi cients, from 0.85 (Pekz Lake) to 0.94 (lakes Komorze and Drawsko) were obtained for the linear function, which suggests the existence of a strong linear relation. The strength of this relation was further confirmed by high determination coefficients, which also give evi dence about a good fitting of the regression function to the empirical values. For example, in the case of fishes of lakes Drawsko and Komorze only 12% of information on the fecun dity were not explained (determined) by a variable-total length of fish. This linear vari able in vendace of all bodies of water was statistically significant (p < 0.05). Compared to the linear function, the power curve only in the case of vendace of lakes Drawsko and Morytiskie, better describes the existing relation. Values of determination coefficients (D) in the above lakes amounted to 0.93 and 0.78, respectively.

'fable 3
Gonadosomatic index (GS/), mean absolute fecundity and results of variance analysis (ANOVA) and linear regression between absolute fecundity (Fa)  *Data in columns, marked with the same letter in superscript do not differ statistically ,(p < 0.05), As mentioned earlier, the relative fecundity of vendace was defined as the number of eggs related to 100 g of body weight of a female. Results of variance analysis (ANOV A) shown in Table 5, indicate existence of statistically significant differences in individual age groups, between relative fecundity of fish from individual lakes. The distinctly lowest value of this parameter characterised vendace of Pekz Lake. For example relative fecun dity in 3-year-old fish (the most representative group) from the above-mentioned lake is by 17.5 percentage points higher from fish acquired from Morytiskie Lake and by as many as 32.9 percentage points from fish of Komorze Lake (Table 5, Fig. 3). Another conclusion from the analysis of relative fecundity was that fish from the Drawskie Lake District (lakes:

DISCUSSION
The vendace represents fishes of relatively low fecundity, although it compensates it with early attainment of sexual maturity. Depending on environmental conditions in lakes, females of this species are able to breed, most often, in their second year of life, whereas the males-in third (Bernatowicz et al. 1975). A number ofresearchers stated that in some cold bodies of water of north-eastern Europe the gonads of vendace mature as late as in the third year of their lives (Koskova 1977) and in the case of Siberian vendace-at the age of 4+ (Vysegorodcev 1977). In Polish lakes, because of the thermal conditions of waters, the vendace is ready for reproduction as early as at the age of 1 + (Budych and Iwaszkiewicz 1964;Ciepielewski 1974;Demska-Zakys and Dlugosz 1995). It has been also confirmed in the course of the present study, carried out in lakes of Western Pomerania. On the spawn ing ground, however, dominate 3-year-old fish of much higher fecundity than that of 1 + vendace (Bernatowicz et al. 1975).
The size of gonads, at the time of reproduction ( described in literature using gona dosomatic index, GSI) is dependant on environmental conditions, age, length, and weight of fish. For example in vendace females occurring in brackish waters of the Gulf of Both nia it fits within 10-12% (Friman 1972), whereas in neighbouring lakes Keitele and Pyha jarvi it is 23,9% and 19.3%, respectively (Lehtonen 1981;Sarvala et al. 1992). In Polish bodies of water the values of GS! are also very variable. In Lakes of West Pomerania (Bucerz, Krzemien, and Kaleitsko) the size of gonads constituted between 14.63 and 23.16% of the body weight of vendace (Walczak 1953). Even wider range (9.5-33.2%) was determined for Masurian lakes Dargin, Dobskie, and Kisajno by Bernatowicz (1963). The mean values of the gonadosomatic index of vendace from lakes of West Pomerania fitted into the above range, however, few specimens from lakes Pile and Pelcz showed low values of gonadosomatic index (GS/< 10%), whereas in a single vendace female from Moryitskie Lake the value of the above parameter amounted to more tan 30% of the fish body weight.
Fecundity of vendace and also other fish species underwent different fluctuations, depending on environmental conditions and population aspects (Zawisza and Backiel 1970). Bernatowicz et al. (1975) believe that vendace in lakes of Western Poland shows higher fecundity than this species occurring eastern parts of the country. Cluster analysis was employed to compare and visualise literature data on absolute fecundity of vendace from lakes of Wielkopolska, Masuria, and Pomerania with the results of the present work (Fig. 4). The above statistical method demonstrated that fecundity of this species may be similar-which is evidenced by Euclid' s distances-even in lakes distant to each other geographically. For example the closest similarity of this parameter was stated in vendace of west Pomeranian Pile Lake and of Masurian Dargin Lake (Euclid's distance between those bodies of water was only 1.6), On the other hand, in the case of two Masurian lakes (Mar6z and Mamry) a high value ofEuclid's distance was determined (11.7). The dendro gram (Fig. 4) did not include vendace of Sremskie Lake, showing the highest fecundity. Budych and Iwaszkiewicz (1964) believed that the mean absolute fecundity of vendace from this lake was 18.2 (2+) and as many as 62.6 thousand eggs at the age of 4+. following lake names denote respective publications: 1, Ciepielewski (1974); 2, Mastynski (1978); 3, Demska-Zak�s and Dlugosz (1995); 4, Budych and Iwaszkiewicz (1964); 5, Bernatowicz et al. (1975);6, present data In the literature, fecundity of fishes is subjected to analyses of correlation and regres sion, to determine the nature of relations between variables ( e.g. between absolute fecun dity and age or length, or weight of a fish). The above relation is usually formulated as a power function (Healey and Nicol 1975;Backiel and Zawisza 1988) or a linear function (Svala et al. 1992;Anwand 1998). The results of the present work suggest that the relation between the absolute fecundity and the total length of vendace can be formulated in both ways: as a power function or as a linear one. In both cases, high determination coefficients were obtained with small, lower than 3%, differences in the above parameters (except ven dace from lakes Drawsko and Lesne ). It must be emphasised that in all vendace acquired from 6 lakes the absolute fecundity was much closer correlated (higher correlation and de termination coefficients) with the total length of fish than with their age. It is possible that the above-mentioned difference was caused by the differences in the growth rate of ven dace from those lakes. Any further, detailed regression analysis between those parameters is not possible, however, because of the lack of literature data pertaining to the growth rate of length and of weight of fish from those bodies of water. A similar regularity in vendace was stated by Demska-Zak�s and Dlugosz (1995), who believed that fish showing faster growth rate, at the same time show higher fecundity and those parameters depend, in large extent, on the environmental conditions of each lake. CONCLUSIONS 1. Vendace in lakes of West Pomerania spawn at the age of l+. On the other hand the dominant age group on spawning ground are 3-year-old females, which constitute 51.4-97.6%. Only in the case of Drawsko Lake this age group was less abundant (41.9%). 2. The values of absolute fecundity in vendace from different lakes of western Pomerania show substantial variability. V endace of lakes Komorze and Drawsko show the highest absolute fecundity (7.21-16.85· 10 3 and 6.9-22.23· 10 3 of eggs), whereas the lowest fe cundity can be attributed to vendace of Pelcz Lake (2.61-4.12· 10 3 ) 3. Very high correlation between the absolute fecundity and the total length of fish was determined in vendace collected from 6 lakes. The correlation was similarly high in linear function and in power function. Results of regression analysis indicate that the absolute fecundity is better determined by the total length than the fish age.
4. Geographical location of lakes does not have a crucial effect on the absolute fecundity of vendace. Vendace acquired from lakes located a high distance from each other may show similar fecundity.