장음표시 사용
201쪽
4t s. and the hard wood gallery forests of the River Tisga. The forests of the B likkMis have been fui ther fragmen ted by clear- fellings whicli hec a me extensive approximate ly 200 years ago. At the fame time, an extreme reduction os gallery fore sis took place due to the regulation of the River Tis Za and iis tributaries and resulted in the complete i solation of the Sajolad forest. Our alna was io stud y the population structure and genetic variabili ty os P.
In a preliminary stud y eleven samples were collected frona the area. undanalysed by means of en Zyme electrophore sis Fig. l). The resulis stlowed that ille sumples were Clearly separated frona each ollier and thus originaled si ona different populations. Genetic dris' had a strong essect on the genetic structu re os these populations MEGLECZ et tit l99J). Considering the se findings there were two ways to continue our stud y si)the effect os genetic drisi depentis on the essective si Ze os populations. The si Zeos the populations varie d greatly in the area: three populations - Nagyoldai Ηanyahegy and Lusta valley - were sar larger than the olhers. Theres ore it was interesting to investigate the essect of population si Ze on the genetic stabili ty of the populations. si i) Although ille samples were collected in weli defined localities ineach area situ aled a sew km apart) we noti ced that these sampling sites wererat her patchy. Heiace, we were interested to examine whether the structu red nature of the habitat resulted in genetic disserentiation belween closely si tuaied habitat patChe S.
203쪽
MKR Lltitia. The Jolly-Seber method SEBER l 982. KRERX l 989) was used to estimate theaverage number os individua is present in the populations at a gi ven time intervat. The total numberos individua is during the whole night period was also estimated by summing the est imaled numberos new animais recrui ted during cach recapture intervat. in ordor to characteri se the rate os dispersion 'et con habitat patches. we compared the number os individua is captured in ut least two ope nings to the total number recaptur d. Eri me electro horexis destia. Genotype and allele freque iacies were calculated on the basis of the banding patierias. Exact test sor population disserentiation RΛYMoND and ROUssET l 995ti)was carried out to test the independe iace os the allelic compositions os the variotis populations. The genetic disserentiation os populations was also analysed by F- statisti cs EIR L COCKERHAMl 984). the total genetic variabili ty s Eli) was partitioned into with in Fis) and belween EsT) population variation. Heiace. Est indicates the amo uni os genetic disserentiation het ween the samples compared to the total genetic variance. Zero value means that there is no variation he tween the samples whild i indicates totai disserentiation. Hierarchical F- statistics s RiGHT l 978) was used to analyse the levet os disserentiationamong samples collected in disserent populations in the fame year. and si ona the fame population indisserent years. By this method the total genetic varia iace among the samples was partitioned into abet vecta population component and a bet wecn year component. Computations were carri eis out with heli' of the sol lowing sol tware: GENE POP version l. 0) RΛYMONII and ROUSSET l 995 h) to persorna exact test sor population. FSTAT version l. 2 COUDET l 995) lo compute F- statistic s. BIOSYS-l re lease l. J S OEPORD and SELANDER l98 l) to calculate hierarchical F- statisti cs and genetic distances.
204쪽
Tahle l. Subpopulation disserentation at two sitos. k number os subpopulations. P α cxaci proh
Genetic disserentiation with in the populations was analysed by comparing the resulis of the electrophoretic studies in ille samples collected frona the three clearings at Banyahegy and in two samples originaling si ona Nagyoldai. Exactprobabilities sor population disserentiation and Usi values were calculated for thesamples in both populations s Table l). None os ille analyses showed disserentia-lion at Banyahegy, while ille Usi value was smali but significantly higher ilian Zero at Nagyoldat. li implied that ille buttersites frona ille three smali clearingsbe longed io the fame population at Banyahegy, while there was a stighi disserentialion het ween the two samples at Nagyoldai. The resulis os the MRR and electrophoretic studies are in good agreement at Banyahegy in spite os ille structured habitat ille genetic composition os ille population seems to he homogeneotis. This sincling can be explained by the sty in ghelia vi our of the buttersites. Males fearch sor se males frona moria ing to the early asteria oon hours in sunshine. Their Vrandona walkV co vers severat hundred meters. Shrub- like vegetation and sparse sol est do not forna an essective barrier sordispersing. Thus. gene flow can equalise ille locat disserentiation heiween sub- population Ν. Dense forests, however, serve as essective barrier even in s hori distance e. g. at Sajolad and Nagyoldal). loci os P multition Siro. in the MRR siti ly. we esii maled the populationsi Ze at two si tes: Banyahegy and Sajolad. As we diu not si nil genetic disserentia lion belween the samples collected in disserent habitat patches at Banyahegy weestimated the population si Ze sor the three clearings logether. Ii is rem arkable thalthe population si Ze at Banyahe gy was abolii ten times larger than at Sajolad Table 2). it appears that the assumption concerning the si Ze of the investigate ld ahlo 2. Resulis os the naui k-rcleasc-recapture VRR) study. N α average num her os individual spresent in the population at a glucia lime. Nim total number os individuals in ille population during
205쪽
Tahle 3. Resulis os hierarchical F-statisti cs sor large Banyahegy, Lusta valley. Nagyoldal) an is mali si ollosteto. Gyei lyan valley. Bid sZentlas Zio. Menes valley. Sajolad) populations separalely Comparison Large populations S mali populations
populations proved to be true namely, three of the populations - Banyahegy Na-gyoldai and Lusta valley - were much largor than ille othei S. Theres ore we analysed the variation hetween the samples collected in disse ferent years frona the fame population with the hel p of hierarchical F- statis ii cs sorthe large and the smali populations separate ly Table 3). In the large populations a s mali portion os the totai variation was only found belween the two years while in the smali populations abo ut 62 in os ille totai variance was due to the bet weenyear variation and only 38 to the bet ween- population Variation. Thus we Concluded that population si Ze had a significant effect on the genetic stabili ty of the se population S. Kee ping in minit that in the pre liminary studies genetic drisi proved to bethe main evolutio nary force acting in the populations and at the fame time considering that the genetic stabili ty of the populations depended on their si Ze it was interesting to see the relative significance of disserent evolutio nary forces affeci-ing these populations willi Variotis si Zes. Heiace we calculated exact probabilitiessor population differentiation and Usi values for three different data seis. In thesii si data set. we included ali the samples frona the three large populations whichwere situ aled in t vo regions Fig. l: Aggtelek-karst - Nagyoldai No: l 904.
l99 β). In the seconii data set the samples frona the large populations of the B likkMts were only include l. In the third data set. we included ali the samples frona the smali populations of the Bijkk mountains. When we analysed the sit si data set we found clear differentiation at ali loci among the samples of the large populations frona the two differeni region s Table d). In contrast we observed low levet os differentiation compared with thes amples frona the large B likk populations s i. e. using the seconii data se t). As thes amples frona the large B likk populations were hard ly differentiateis; the significant in homogenei ly in the sit si data set could only come frona the disserence bet ween the two regions. This finding was also supported by the resulis of hierarchi cal P- statisti cs among large populations the bet ween region component Fui 0. 0J2) of the totai variance among samples Upim 3. 086) was much larger
206쪽
than the willi in region component Ui,st . l ). This indicates a certa in geographic patieria in the genetic structu re os the large populations due to restricte clgene flow be tween the two region X. At the sanae time we soti nil strong disserentiation be tween ille samples even illain a single region when we compared the smali populations os the B likk Mis third data se t). ll thus appears that in the smali populations genetic drist is themost important evolutionary force. Although the fores ted area os the B likk hec a me fragmen ted only recent ly in the geological sense, there was enough time sorthese sinali populations io be disserentialed due to genetic drist.
207쪽
208쪽
Di oppings On dry sheep-runs in two Trans lanubi an sites Hungary) ando ne si te in the Kiskunsag Nationat Park C Hungary) were sui veyed in ordor todeteci seatures os communi ty coloni sation os coleopteran and dipteran populations PAPH l99β). we collected data for each stes' os communi ly organi sation os these pioneer assem blages and attempted to identi sy the so urce os specimens coloni sing di oppings. In this paper sield resulis will he elaborate i. The numbers os species detected in the fietis samples os siles are analysed in comparison wilhsonae known methods os estimation os species richness and a new simulation method. Estimation os species richness is an interesting and permanent ly evolvingsield os quantitative ecology. The prese iace of the individuals or populations os large-hodie l. hight y abundant or exceed ingly active species is comparative lyeasy to ascertain. Is ille individua is os a species are snaall-bodie l. or is ii is dissi- Cult to Sample them for any other reason. the presen ce os rare species is dis icti li
209쪽
to demonstrate. Among in secis, where the rare and poorly detectabie species cans orna a major part of species richness, this practical difficulty hecomes a theorelicat problem or even a problem os epistemiology. For a si illed ento mologist it is obviolis that the number of insect species populations present at a gi ven pol ni osti me in an area selected for any practical reason cannot be precisely determine l. e must rely on estimations obtained by sonae kind of approximation in otii
case, extrapolation S). Consequently, a large proportion Os rare Species always rema in undetected. Is rare species heco me extinct in a gi ven area at an in determinate but neveriheless exciting rate, and they recoloni Ze again. we attempt to G
ProXimiate EXI res Olcite) the quiaSi-equilibrium Uecies number. Ecology has been confronted wi ila these problenas for a long time for a bibliography, See e. g. SOUTHwOOD l9J8). The species number estimation methods were main ly elaborated during the l9J0s and l 980s. resulting in a plethora of current techniques. The de Velopment of the numero us procedures raises a question are the dubio usspecies numbers applicabie for theoretical considerations , Or, are the theories applicabie and valvabie is they are based on such unstabie data samong others wewould recommen d for a re view ROSENT E lG's sl 995) excellent book fuit os wii and fair criticis m). This paper is not an overview of the procedures and methods avat labie forestimating species number, instead we recommend the receni and good re vie w of COL EI L and CODDINGTON sl 994) on extrapolation method s. Only a limited number os procedures known in the vasi literature is selected for analysis os otii data and for comparison with the resulis obtained by the new method proposed. e would like to stress that two approaches are presented in this paper sortite number of species: a) we had severat hut individualty in adequale samples fori he estimation of the total number of species in the site: h) we set out frona a definite frequency structure to estimate the undetected number of species at an actualti me and plaCC. The paper consists of the folio ing paris
Hodkinsons ' method. originalty developed for estimat ing the total numberos species in unexplored faunas sor. unknown, undescribed species), is modis ecl. It is adopted to siluations when there are subgroups willi significantly disserent capture probabilities. In such cases we may apply an estimation for the Probable- ClOSer IO iactucit - tollit Humber of UecieX by assuming the total number os species is ali were caught by the probabili ty of the best collectibi e subgrou p. Ac cord ing to COL ELL & CODDINGTON's sl 904) classi fication. i his is a 'taxonratio ' method of estimation and extrupolution . Applying the expected species-individual diversi ty 'rarefaction ) method our fletu data on the observed number os species were compared to the m m
210쪽
Eliati Vc Ghuri tinc e PectorX. The parameter m equat s sample si Ze in ali cases. In this case we actualty do not aim at estimation number os species. The compari sonos ille estimat ed and observed species numbers was used to evaluate whether ille relative abundance vectorfp tot os agi ven sample and that os the 'communi ty V are congruent wilh the known relative abundance vector we Chosen us basis. Finalty we propose a new simulation method hased on urn mode is relaled tollie truncateil lognormal distribution. To test the new method. the estimations os the species numbers generated by simulation are compared to the observed num-
Based Ora our experien Ce we can say that there are no disserences in ille species composition in tale spring. early summer anu carly autumia. as sar as the important dipterous species are con- cerned ΡΛΡΡ l 9853. This is why we caia regarit ali the sumples collected at the sanae place and yearas samples si ona ille sanae pool populations. we a re aware os the limiis os this assumption. hul we