SPATIAL STRUCTURE OF A GASTROPOD COMMUNITY IN THE LITTER OF A BEECH FOREST OF DENTARIO GLANDULOSAE-FAGETUM IN THE TATRA MOUNTAINS

The grid method was used in the present study of the spatial structure of a gastropod community. 2,474 specimens representing 29 species were collected from an area of 9 sq m. The numbers of species were higher in squares with higher densities of snails (rNS = 0.9). Species diversity varied between the squares, ranging from about 2.6 to 4.1, and was not very closely correlated with increase in species number (rSH = 0.6). Snail density was not correlated with average dry litter weight in a square (rNM = –0.03). The values of the indices of Lexis and of Morista show that the distribution of snails in the studied forest patches was aggregated. The material suggests an obvious relation between snails and plants. key WordS: Gastropoda, mountain malacofauna, snail community structure, species composition, abundance, dominance, ecological parameters Folia Malacologica 3/1989 was originally published as No. 1216 of Scientific Bulletins of University of Mining and Metallurgy, Cracow. This digitalised version was prepared by the Association of Polish Malacologists and first published on-line on December 30th, 2016. http://dx.doi.org/10.12657/folmal.003.004 Folia Malacol. 3: 55–72


INTRODUCTION
The knowledge of the spatial structure is basic for the recognition of the factors that influence animal populations and communities.It seems especially important iA the case of the gastropod fauna (Huflejt and Karwowski 1961).Stuoies concerning molluscs are relatively very few.In the Polish literature several papers on Helix poeatia Linnaeus, 1758 can be mentioned (lomnicki, Wasilewski and Kosior 1964, lomnicki 1971, Woycie-Chowski 1980).Also some biocenological studies deal with similar problems (Urba~ski 1939, Orozdowski 1961, Dzi~czkowski ' 1972, Dyduch 1980 and Oyduch--Falniowska and'Fyda 1986).
In the paper the au~ors have attempted at a description of the spatial structure of a gastropod community in a beech forest of Oentario-glandulosae Fagetum .Additionally, the spatial structure of populations of the most abundant species of the studied communities has been considered.

MATERIAL AND METHODS
In the present study the grid method was used (Kwiatkowska and Symonides 1980).The material was collected using the method proposed by Diem and modified by Oekland (DziQczkowski 1972) for mollusc studies.It consists in taking samples with a square frame.The size of the frame used was 2D x 20 cm.The frame surface is referred to as "plot" in the text.
In each of three patches of a beech forest three sites were chosen, 25 samples to be taken from each site.The plots of each site adjoined each other forming a 100 x 100 cm square (Figs 2 -4) .The resulting "big squares" were numbered from I to IX.The frame was stuck in the ground 2 -3 cm deep, so that each sample contained litter and the superficial soil layer.All samples were hand-sorted, because this most time-consuming method is necessary in both faunistic and quantitative studies (Dziqczkowski 1972, Umi~ski 1973, Oyduch 1980), and gastropods fixed in 75~ ethanol.
2,474 specimens representing 29 species (Tab .1) were collected altogether.Plants from all plots were determined and counted, the litter collected was dried at 25 -30"C and weighted.The material was analysed using the following indices:  The valley of the Spadowiec stream lies on the Spadowiec nappe, and is a strict nature reserve.The localities differed between each other in humidity, altitude, bedground and plant cover (Fig. 1).This eastward exposed local i ty was situated at 950 m a. s. 1 ., on a humid flattening of a ridge.It was covered with a quite thick forest of beech and fir 60 -lOO years old.The cover of trees was about 70\ resulting in a prevalence of semi-darkness.The herb layer of 70 -80% of cover comprised 24 species out of which the most abundant were Oxalis acetosella L. , Dentaria glandulosa W. K., Dryopteris filix-mas (L .) Schott, and ~io nemorensis L. The thickness of the leaf litter of the forest was 5 -B cm.The soil and litter were moderately humid.The bedground consisted of sandy sediments and loams.
The litter dry weight also differed between the squares.In square I it ranged from 35 to 135 g depending on plot, the average being 68.8 g.In square II the average was slightly lower: 65.9 g, the dry weight ranging from 28.9 to 94.3 g.In square Ill the dry weight ranged from 26 to 108.2 g, its average being the highest (69.9 g) of the nine squares.
Locality 8 (squares •Iv, V, VI) The locality was situated 915 m a. s. 1., about 400 m from the path called Droga pod Reglami, on the right (square IV) as well as on the left (squares V and VI) bank of the stream Spadowiec, in a patch of fir trees With an admixture of beech and spruce.The cover of trees reached about 50\.Rubys idaeus L. occurred abundantly apart from fl~ beech and spruce seedlings constituting the shrub layer .The much diversified herb layer (49 species) reaching 100\ of cover consisted of the following vascular Plant species: Petasites albus (L.) Gaertn., Stellaria nemorum L., ELUruLls elatior (L.) Grufb., Senecio nemorensis L. subsp.nemorensis L., Caltha lgeta Sch.N. K. , as well as of some species of mosses, lichenes and liverworts.The soil and litter humidity was high .The leaf litter was moderately (4 -5 cm) thick .Shales, greenish sandstone and organodetritic limestone formed the bedground .
Out of the nine.squares examined square IV was the one most densely covered with vegetation.The plant species recorded there were: Petasites The litter dry weight in square IV ranged from 14.4 to 65 . 2 g, the average being 43.6 g, while i n squares V and VI it ranged from 20 to 92.6 g and from 16.3 to 50.5 g respectively, the corresponding averages being 53.7 and 34.2 g.

Locality C (squares VII, VIII, IX)
The locality was situated at about 1025 m a. s. 1., lOO m from the path Droga pod Reglami, slightly below a ridge.It was covered with a dry beech forest with an admixture of fir (4\) and spruce (1\).The cove r of trees was 80 -85\.Ther e was no shrub layer.Oxal is acetosella L. and Cardamine llifglia L. were the most abundant plant species in the herb layer.The so11 was rather dry, stony , penetrated with roots of Qxalis acetosella.The litter layer was varied in thickness (i. e .thicker in depressions , thinner elsewhere) but in general moderately thick.The bedrock of the locality was formed of dolomites and limestones.
In square VII the following herb species were recorded: Petasites albus (L.) Gaertn.The litter dry weight in square VII ranged from 10.8 to 89.4 g per plot, the average being 45 g, while in square VIII: from 6.9 to 29 .5 g~ the average being 19.08 g that is the lowest from among the squares.In square IX the dry weight varied between 6.1 and 45.7 g, the average amounting to 22.5 g.

RESULTS
29 gastropod species were found occurring in the litter of• the studied patches of a beech forest (Tab.1).Figs 1 -3 illustrate the gastropod spatial distribution in each square.

DISCUSSION
In the literature on the subject no papers can be found whose results are closely comparable to those of the present study.There are, indeed, not less than two approaches to the problem thi s paper deals with.The ~irst is to study the structure of a litter snail community, that is, its species composition (5), species diversity (H'), equitability (J'), density (N), etc., which has been done by several malacologists, particu1ary

rx
Another approach is to study the spatial structure of the community, that is, distribution interrelations and dispersion characters of particular species, the age structure of dominants, etc.This approach is not common in the malacological literature.lomnicki (1971) and Woyciechowski (19BO) in their studies on Helix pomatia have applied some methods that render their results slightly similar to this type of analys i s.
Yet another possible approach is to study factors influencing the microhabitat dispersion of a snail fauna . in such studies it is necessary to examine not only the snail fauna but also plants and litter of the studied Plots.All the mentioned approaches are considered here in a preliminary lll'lalysis, which is to provide a basis for a study on interspecific interrelations within a Qastropod community as well as on factors influencing the community structure and function.The sampling method used in our study should allow to trace relationships overlooked in stud1es completed using other methods.
In the material collected 29 species were recorded, the species number varying from 11 to 28 per square.In squares I, II, VIII and IX (Tab.In our study the species numbers were more strongly correlated with snail densities in the squares (r 5 N = 0.9) than with the values of H (r 5 ~ = 0.6) that ranged from 2.6 to 4.1.The values of the indices and their relations do not depart from those calculated for other localities in the Tatra Mts.Neither the species number nor snail density were correlated with the average dry litter weight in a square (r 5 M = 0:01, rNM = = -0.03).
The two methods combined for the purposes of this study (the grid method and square frame method) have provided information on the distribution of particular snail species within the community (Figs 2 -4) as well as on the dispersion of the populations of the most abundant species.The Lexis and Morista indices have been used here to describe the gastropod r

OOL.
D1JaS04a~~D~"O:-:-I:-O-s=-a=-.A-:-o=a:-:-1 ---------+----------J In further studies of this kind it would be necessary to apply a com-Parative analysis of theoretic and empiric distribution patterns.This Concerns not only the spatial dist~ibution of species in the community but also the distribution within populations of the most abundant species: ~trea diaphana, V. transsylvanica and Carychium tridentatum.The values of Morista and Lexis indices calculated for these populations also indicate their aggregated spatial distribution (Tab.3).
Correlation coefficient values for snail and plant densities indicate some snail-plant interrelations.The gastropod densities were strongly Positively correlated with the densities of Petasites albus, but negatively correlated with the densities of .Oxalis acetosella (r = O.B4, and r ~ = -o.B4 respectively: Fig. 5).o. acetosella takes advantage of local increases in soil acidity, since it can grow in habitats of an increased acidity, avoided by other herbs (5.Michalik personal communication).The negative correlation indicates that such habitats are also avoided by gastropods.

( 1 ) 2 •I
H' = N(N log N-E ni log n 1 ) 3.321928, H' -species diversity of the community according to the Shannon-Wiener formula, H~ax = log 2 s, J' -equitability (evenness), Nnumber of individuals in the community, ni -number of individuals of ith species, 5 -number of species in the community); (3) indices of dispersion according to Lexis: 6 2 , and to Morista (lA), after Hutlejt and Karwowski (1981): i2 + j2 -i -(marith•etic mean of the density of individuals on a plot, 6 2 -varstudied were situated in the valley of the Spadowiec stream in the western part of the Tetra Mts., between two valleys: the valley of the Bialy stream and tne valley called Dolina ku Dziurze.

Fig. 1 .
Fig. 1.Location of Spadowiec Valley (TPN -Tatra National Park)Locality A (squares I, 11, Ill) This eastward exposed local i ty was situated at 950 m a. s. 1 ., on a humid flattening of a ridge.It was covered with a quite thick forest of beech and fir 60 -lOO years old.The cover of trees was about 70\ resulting in a prevalence of semi-darkness.The herb layer of 70 -80% of cover comprised 24 species out of which the most abundant were Oxalis acetosella L. , Dentaria glandulosa W. K., Dryopteris filix-mas (L .) Schott, and ~io nemorensis L. The thickness of the leaf litter of the forest was 5 -B cm.The soil and litter were moderately humid.The bedground consisted of sandy sediments and loams.The species composition of the herb layer was slightly differentiated between the squares, the following plant species having been recorded: square I: Oxalis acetosella L. (426 specimens), Asarum europaeum L. (l specimen), and S~io nemorensis L. subsp.nemorensis L. (1 specimen); square 11 : a1most entirely covered with 0. acetosella (644 specimens); square Ill: 0. acetosella (453 specimens), and Cardamine trifolia L. (4 specimens).

Fig• 3 .
Fig• 3. Distribution of gastropods in litter of fir forest with admixture ~I beech and spruce (locality 8), IV -square IV , V -square V, VI -square • For symbol descriptions see Fig. 2.
Lexis index was applied in a population analysis of Helix Pomatia (lomnicki 1971).Both indices are based on variance, and then depend o~ sample size.Since in our study the samples taken were of a similar snail number, this source of error can be neglected.The values of the two indices (Tab.3) show that t~ distribution of snails in the studied forest patches was aggregated <-; 1, 1 4 > 1).The data, however, are not sufficient for a more detailed characteris~ics.For instance, Huflejt and Karwowski (1981) observed that the indices are not adequate for parameters of an ecological significance.The present results obtained by means of the two indices differ bet~een the patches.In one case (squares IV -VI) the highest value of~ accompanied the lowest of I, this Proportion being not maintained in the other cases (squares I -III and vrr -IX, Tab. 3).