the middle miocene freshwater mollusk fauna of lake gacko ......3 figure 1. geography and geological...

The Middle Miocene freshwater mollusk fauna of Lake Gacko(SE Bosnia and Herzegovina): taxonomic revisionand paleoenvironmental analysis

Thomas A. Neubauer*, Oleg Mandic and Mathias Harzhauser

Department of Geology & Paleontology, The Natural History Museum Vienna, Burgring 7, 1010 Wien, Austria.E-mails: [email protected]; [email protected]; [email protected]

Introduction

During the Early to Middle Miocene the Dinaridemountain chain harbored a series of intramontane lakes,collectively termed the Dinaride Lake System (DLS;Figs 1A–B; Krstić et al. 2003; Harzhauser & Mandic2008). These tectonically isolated lakes experienced anoutstanding endemic evolution of freshwater mollusks.Harzhauser & Mandic (2008, 2010) described the DLSfauna as entity and counted 110 gastropod and 35 bi-valve species. New investigations by Mandic et al.(2009, 2011) and De Leeuw et al. (2010, 2011) suggestthat the various basins of the DLS have to be consid-ered as separate units with different geodynamic, stra-tigraphic, and paleoenvironmental histories. Conse-

quently, new analyses of the various molluskassemblages revealed major differences between singlebasin faunas (Neubauer et al. 2011, 2012). Only duringshort phases hydrological connection between selectedbasins might have persisted to allow temporary faunalexchange. This combination of enduring isolation driv-ing endemic evolution and short-term invasions lead tocomplex distribution patterns and extensive diversifica-tions, which are hard to decipher in taxonomic terms.

Neumayr (1880) and Brusina (1897) provided thefirst taxonomic papers on the mollusk fauna of theGacko Basin. These authors partly provided descrip-tions and illustrations of few taxa, but never establisheda concise taxonomic frame of the whole basin fauna.Milojević (1966) distinguished between several mollusk

# 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Received 9 March 2012Accepted 24 September 2012Published 20 February 2013

Key Words

BivalviaGastropodaPaleoecologyStatistical analysisDinaride Lake SystemLanghian

Abstract

The early Middle Miocene Lake Gacko was part of the Dinaride Lake System (DLS)and gave rise to a poorly known freshwater mollusk fauna. This was subject to malaco-logical studies from the late 19th century onwards. Herein, we provide the first thor-ough taxonomic survey of Lake Gacko including revisions of several taxa. A totalamount of 1,077 specimens was obtained from 17 samples, comprising at least 11freshwater and 2 terrestrial gastropod species and 6 bivalve species. So far, none of thedescribed taxa has been documented outside the Dinarides and the DLS, respectively.The fauna shows variable overlap with other paleolakes of the DLS, proofing oncemore the complex biogeographic patchwork of this system. The current paleoenviron-mental reconstructions are tested and refined by application of a statistical analysis.This confirms the partition of the sedimentary history of Lake Gacko into three majordepositional phases responding to astronomically forced climate changes. The low di-verse mollusk assemblage in the initial phase, with abundant pulmonate and rissooidgastropods, signals a more arid climate with lowered lake level. During the secondinterval, the fauna becomes more diverse with common rissooid and melanopsid gastro-pods. It represents a more humid interval with enhanced precipitation and increasedlake level, entailing the installation of a perennial lacustrine environment. The finalphase is a return to arid conditions with the same elements as in the initial pulmonate-rissooid assemblage. The sphaeriid Pisidium vukovici n. sp. is introduced as new species.

Fossil Record 16 (1) 2013, 77–96 / DOI 10.1002/mmng.201300003

� Additional supporting information may be found in the online version of this article at the publisher’s web-site

* Corresponding author

units based on rough faunal differences. Krstić et al.(2009) compiled a faunal synopsis for the lower partsof the basin infill based on drill cores. However, a com-

plete survey of the fauna from Lake Gacko with taxo-nomic descriptions is still completely missing. Mandicet al. (2011) briefly documented the mollusk distribu-tion and provided a first paleoecological interpretation.By integration of more samples and by application of astatistical analysis these results will be herein testedand modified.

Geological setting

During the Middle Miocene, the Dinaride-Anatolian Is-land acted as a major paleogeographic barrier betweenthe Paratethys and the proto-Mediterranean seas(Fig. 1A). The western part of that land mass compriseda large-scale freshwater lacustrine environment, termedthe Dinaride Lake System. Today, its deposits representsedimentary infills of numerous intramontane basinsdistributed throughout the Dinaride Alps and attainingthicknesses of more than 2000 m (Hrvatović 2006;Fig. 1B).

The Gacko Basin in southern Bosnia and Herzegovi-na represents one of those basins (Mojićević & Lauše-vić 1965; Mirković et al. 1974; Fig. 1C). The architec-ture of its about 360 meters thick sedimentarysuccession can be interpreted as a single lacustrinetransgression-regression mega-cycle (Milojević 1966,1976; Mandic et al. 2011). The sedimentation startedwith detrital deposits representing the initial flooding ofthe basin. The subsequent coal building phase marksthe installation of swamp conditions, passing upwardsinto limestones originating from hard-water freshwaterlake deposition. Subsequently, dropping water levelstriggered again swamp and mire conditions in the term-inal phase of Lake Gacko.

Basin history and depositional setting have been cur-rently studied at Gračanica WNW of Gacko by Mandicet al. (2011). Additionally, a small outcrop (� 1 m thick-ness) was included from the abandoned Vrbica open-cast coal-mine NW of Avtovac, representing the bound-ary interval of the uppermost main coal unit and thelowermost Mytilopsis unit, respectively (Figs 1C, 2).The huge outcrop at the Gračanica opencast coal-mineprovided an excellent insight into the lake depositionon the basin’s paleo-margin. Results from integratedAr/Ar geochronology and magnetostratigraphy fixedthe age of the lacustrine deposition into the Langhian,implying its relation with the Middle Miocene Climatic

Neubauer, T. A. et al.: Middle Miocene freshwater mollusks from Lake Gacko78

3

Figure 1. Geography and geological setting of the studied lo-calities in the Gacko Basin. A. Paleogeographical map of theDinaride-Anatolian Island during the Langhian (after R�gl1998 and Harzhauser & Piller 2007); B. Distribution of DLSdeposits and geographic position of Lake Gacko; C. Geologicalmap of the Gacko Basin with indication of the studied sectionsGračanica and Vrbica. Numbers and column refer to the gener-al lithostratigraphic division of the Miocene deposits in theGacko Basin as established by Milojević (1966, 1976) andMandic et al. (2011).

museum-fossilrecord.wiley-vch.de # 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Optimum (Mandic et al. 2011). Furthermore, resultsfrom sediment petrography, geophysical logging andmollusk paleoecology indicated vivid changes of the re-gional water budget (Mandic et al. 2011). Indeed, thecyclostratigraphic analysis revealed the presence of twofirst order and seven second order transgression-regres-sion cycles (Fig. 2, right column). These were orbitallytuned to �400 kyr and �100 kyr eccentricity cycles re-sulting in an accurate age model for its depositionalduration from �15.8 to �15.2 Ma. The lake level high-

stands were thereby related to eccentricity maxima, ac-companied by environmental eutrophication events inconsequence to enhanced denudation and terrestrial in-put into the basin. Dry climate intervals were related to� 400 kyr eccentricity minima. In the lower part of thesuccession they resulted in iterative swamp forest ex-tension reflected by vast lignite accumulations. In theupper part they formed pedogenic and palustrine carbo-nate accumulations, distributed across the basin’s mar-gin.

Material and methods

For this study 17 samples from three different sites were investigated:10 from the Gračanica mine, 4 from the Vrbica section, and 3 fromdrill cores (near Gacko, exact locality unknown; provided by S. Ćorić,Geological Survey Vienna, and stored in the collection of theNHMW) (Fig. 1C). The samples were dissolved with diluted hydrogenperoxide and washed through a set of sieves (1 mm, 250 mm, 125 mm,63 mm). Mollusks were counted from the 1 mm fraction only, yieldinga total individual number of 1,077 specimens. In addition, a highnumber of specimens was obtained from smaller fractions comprisingmostly juvenile stages and spire fragments. In most cases, these couldnot be determined on the species level. As some taxa (e.g. planorbids,lymnaeids) are prone to fragmentation because of their thin shellwalls, they might be underrepresented in the larger fractions. Sometaxa are exclusively found in smaller fractions (< 1 mm) or in debriscollections and thus do not appear in the quantified record (Fig. 3).

The statistical analysis was compiled using Primer 6.1.10 (Clarke& Warwick 2001). The data set with percentage values was squareroot transformed to limit the contribution of over-represented taxa.Based on the Bray-Curtis distance measure an average group clusteranalysis (Fig. 4A) and a non-metric multidimensional scaling (nMDS;Fig. 4B) with minimum stress of 0.01 were generated. For the result-ing groups factors were defined for a Simper analysis (Tab. 1) to out-line which taxa were contributing to similarities and dissimilarities,respectively. Diversity indices were calculated using Past 2.06 (Ham-mer et al. 2001). Samples with � 25 individuals or those from debriscollections or drill cores (imprecise stratigraphic correlation) were ex-cluded.

Faunal composition andpaleobiogeography

The investigation yielded a total of 13 gastropod (13genera) and 6 bivalve species (3 genera). These areMelanopsis lyrata Neumayr, 1869 (Fig. 5), Bithynia juri-naci Brusina, 1884 (Fig. 6), Fossarulus moniliferusBrusina, 1876 (Fig. 6), Prososthenia neutra Brusina,1897 (Fig. 7), Bania valvatoides (Brusina, 1874)(Fig. 8), Galba sp. (Fig. 9), Radix korlevici (Brusina,1884) (Fig. 9), Gyraulus pulici (Brusina, 1897)(Fig. 10), Orygoceras dentaliforme Brusina, 1882(Fig. 10), Planorbarius sp. (Fig. 11), Ferrissia illyrica(Neumayr, 1880) (Fig. 11), Carychium sp. (Fig. 12),Vertigo sp. (Fig. 12), Mytilopsis frici (Brusina, 1904)(Fig. 13), Mytilopsis jadrovi (Brusina, 1892) (Fig. 14),Mytilopsis sp. (Fig. 14), Pisidium bellardii Brusina,1884 (Fig. 14), Pisidium vukovici n. sp. (Fig. 14), Uniorackianus Brusina, 1874 (Fig. 15). Two further species

Fossil Record 16 (1) 2013, 77–96 79

0m

10

20

30

40

50

60

70

A

B

C

D

E

F

1

2

1

2

1

2

3

Units

Melanopsis-assemblage

Fossarulus-assemblage

Fossarulus-assemblage

Mollusk

samples

Mollusk

Units

lake level- +

1

2

3

4

5

6

7

Transgression-RegressionCycles (TRC)

dark nodulessandsandmudclasttufa-lithoclastscoal lensescoal lensesrhizolithsrhizolithsbioturbationbioturbationgastropods / bivalvescharacean oogonia

ripple bedsripple beds

fossil wood

erosive surface

clay marl/limest. coal

0804/0390804/039 coal0804/039b

0804/039 debris

0907/Moll10907/Moll2

0907/Moll3

0907/Moll4, 0907/Moll4A

090713/1

0708/GA1

0708/GA14

0703/024

Gacko 1Gacko 2

Gacko 3

090713/2

Legend

Figure 2. Section Gračanica, mine pit B with position of themollusk samples. Those correlated from the Vrbica section anddrill cores are colored in gray. Bars indicate non-stratified,unit-related samples; dashed bars indicate uncertain correlationof samples from drill cores (modified after Mandic et al.2011).

# 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim museum-fossilrecord.wiley-vch.de

recorded for the Gacko Basin, i.e. Stalioa parvula Neu-mayr, 1880 and Euchilus elongatus Neumayr, 1880,could not be detected in the course of this investigation.All taxonomic descriptions, except that for the newspecies, are provided as supplementary material.

All recorded freshwater taxa that were determined onspecies level are endemic to the Dinaride Lake System.Nevertheless, there are notable differences concerningthe distribution of the various taxa within the lake sys-tem. Distinct faunistic relationships of the Gacko Basinfauna on the species level are evident for the Drniš Ba-sin (57.9 %) and the Sinj Basin (52.6 %). In contrast,the roughly equally distanced Kupres Basin shares onlya single taxon accounting for 5.3 % similarity (Orygo-ceras dentaliforme). However, notable faunal overlapsare known between the Sinj/Drniš basins and theKupres Basin (39 %; Neubauer et al. 2012), which aregeographically closer to each other than to the GackoBasin. The stratigraphic age of the considered depositsis comparable, ranging around 15.8 to 15.0 Ma (Neu-bauer et al. 2012). Most probable, faunal exchange wasperformed via hydrological connections, which existedto the Sinj and Drniš basins but not (or only for shortepisodes) to the Kupres Basin. These connections are


0%

20%

40%

60%

80%

100%

0708/GA1

0907/Moll1

0907/Moll2

090713/1

090713/2

0804/039

0804/039 coal

0804/039b

0907/Moll4

0907/Moll4A

Abundance

Pisidium vukovici n. sp.

Pisidium bellardii

Mytilopsis jadrovi

Mytilopsis sp.

Carychium sp.

Ferrissia illyrica

Orygoceras dentaliforme

Planorbarius sp.

Gyraulus pulici

Radix korlevici

Bania valvatoides

Prososthenia neutra

Fossarulus moniliferus

Melanopsis lyrataSamples

Vertigo sp.

Unio rackianusCluster 2 Cluster 3 Cluster 4

Group 1 Group 2

Cluster 1

Figure 3. Abundance data of mollusk taxa with indications of groups and clusters resulting from the statistical analysis.

0804/039

100

80

60

40

0804/039coal

0804/039b

0708/GA1

090713/1

090713/2

0907/Moll1

0907/Moll2

0907/Moll4

0907/Moll4A

Similarity

2D Stress: 0.03

A

B

G1

G2

Group 2Group 1

60%

Margalef'srichness index

ephemeral

lakeconditions

0907/Moll1 090713/1090713/2

0907/Moll20708/GA1

0907/Moll4A

0907/Moll40804/039

0804/039 coal0804/039bperennial

Cluster 2 Cluster 3 Cluster 4Cluster 1

75%

3

Figure 4. Statistical analysis of quantified samples using Bray-Curtis distances of square root transformed data. A. Clusteranalysis (group average); B. Non-metric multidimensional scal-ing (minimum stress of 0.01).


likely to have occurred during the several transgressionmaxima in the history of the Gacko Basin (Mandicet al. 2011). This is emphasized by the appearance ofseveral taxa only within certain units (e.g. Bithynia juri-naci, Orygoceras dentaliforme, Mytilopsis jadrovi, Uniorackianus). However, a unit-related trend, meaning higherpercentages of invaders in specific intervals, could not bedetected.

Results of the statistical analysis

The distribution of samples within the succession isheterogeneous. The statistically analyzable sampleswere confined to three section intervals: 3.5–6.0 m(Unit A; 5 samples), 33.0–34.0 m (transition betweenunits C/D1; 3 samples), and 39.5 m (transition betweenunits D1/D2; 2 samples). In between, only few sampleswere taken because of the generally poor preservation.Samples were either too small for a significant quanti-fication or were collected from debris.

The stratigraphic grouping of the samples is reflectedin the faunal contents as proven by the statistical analy-sis (Figs 3, 4). The cluster analysis shows two maingroupings at a similarity level of 60 %, which corre-spond to the two lower intervals termed as Fossarulus-and Melanopsis-assemblages in Milojević (1966) andMandic et al. (2011). Group 1, with 71.7 % average si-milarity, is dominated by Gyraulus pulici, followed byBania valvatoides and Fossarulus moniliferus. Group 2is largely defined by Prososthenia neutra, Fossarulusmoniliferus and Melanopsis lyrata, with an average si-milarity of 70.1 % (Tab. 1). These obvious faunal differ-ences are proven by an average dissimilarity betweenthe two groups of 53.6 %.

In more detail, each of the two groups can be splitinto two clusters as suggested by the cluster analysis (ata 75 %-similarity level; Figs 3, 4). Samples from Clus-ter 1 (Unit A) are clearly dominated by pulmonate andrissooid gastropods (Fig. 3; Tab. 1), namely Gyrauluspulici, Bania valvatoides and Fossarulus moniliferus.This cluster, with an average similarity of 81.7 %, is de-

Fossil Record 16 (1) 2013, 77–96 81

Figure 5. Morphotypes of Melanopsis lyrata Neumayr, 1869. A. NHMW 2011/0138/0129, sample 0804/039 debris, morphotypeB; B. NHMW 2011/0138/0130, sample 0907/Moll4A, morphotype B; C. NHMW 2011/0138/0131, sample 0907/Moll4A, morpho-type C; D. NHMW 2011/0138/0132, sample 0907/Moll4A, morphotype C; E. NHMW 2011/0138/0133, sample 0907/Moll4A,morphotype D; F. NHMW 2011/0138/0134, sample 0907/Moll4A, morphotype E; G. NHMW 2011/0138/0135, sample 0804/039debris, morphotype E. Scale bar equals 1 cm.


fined by these 3 taxa only. Cluster 2 (also Unit A), witha comparable average similarity of 81.3 %, is definedby Bania valvatoides, Gyraulus pulici, Radix korlevici,Fossarulus moniliferus and Prososthenia neutra, ac-counting for 92.8 %. In addition, Unio rackianus (onlyrecord in the section), Pisidium bellardii and P. vukovicin. sp. appear. While the differences between the twogroups are distinct, deviations between clusters withinthe groups are in both cases comparably low. Clusters 1and 2 show an average dissimilarity of 34.8 %, resulting

from slightly varied values for Gyraulus pulici, Radixkorlevici and Bania valvatoides, and the presence ofPrososthenia neutra and Unio rackianus in Cluster 2.

The following 25 meters of limestone-coal interbed-dings were inadequate for an analysis. Few findings ofFossarulus moniliferus, Galba sp., Gyraulus pulici andFerrissia illyrica are reported from sample 0907/Moll3(Unit B1). This is the only occurrence of the lymnaeidGalba within the whole section. Samples Gacko 1 andGacko 2 from roughly correlated drill cores (Unit B2;


Figure 6. A. Fossarulus moniliferus Brusina, 1876, NHMW 2011/0138/0136, sample 090713/1, morphotype A; B. Fossarulusmoniliferus Brusina, 1876, NHMW 2011/0138/0137, sample 090713/1, morphotype A/B; C. Fossarulus moniliferus Brusina, 1876,NHMW 2011/0138/0138, sample 090713/1, morphotype B; D. Fossarulus moniliferus Brusina, 1876, NHMW 2011/0138/0139,sample 0804/039 debris, morphotype C; E. Fossarulus moniliferus Brusina, 1876, NHMW 2011/0138/0140, sample 0804/039debris, morphotype C; F. Fossarulus moniliferus Brusina, 1876, NHMW 2011/0138/0141, sample 0708/GA1, morphotype A;G. Bithynia jurinaci Brusina, 1884, NHMW 2011/0138/0143, sample 0804/039 debris; H. Bithynia jurinaci Brusina, 1884,NHMW 2011/0138/0144, sample 0708/GA14, operculum; I. Fossarulus moniliferus Brusina, 1876, NHMW 2011/0138/0142, sam-ple 090713/2, morphotype A. Scale bars equal 5 mm (A–I, K–L) and 1 mm (M), respectively.


Fig. 2) provided few remains of Fossarulus moniliferus,Radix korlevici and Gyraulus pulici.

Cluster 3 from the transition of units C/D1 with cor-related samples from the Vrbica section yielded threeanalyzable samples (Fig. 2). This cluster has an averagesimilarity of 79.2 %. The most abundant taxa are Fos-sarulus moniliferus and Prososthenia neutra followedby Radix korlevici, Melanopsis lyrata, Ferrissia illyricaand Gyraulus pulici. These 6 taxa contribute with91.5 % to the average similarity. Additional taxa are

Bithynia jurinaci (from debris only), the planorbidsPlanorbarius sp. and Orygoceras dentaliforme, the ter-restrial mollusks Carychium sp. and Vertigo sp., and thedreissenid Mytilopsis sp. Sample 0708/GA14 at the topof Unit D1, too small to be analyzed, yielded abundantBithynia opercula (68.0 %) and Ferrissia illyrica(28.0 %), followed by Gyraulus pulici (4.0 %). Thiscomposition changes rapidly c. 1 m above. The transi-tion of units D1/D2 (Cluster 4) with two samples is de-fined by 6 species accounting for 90.7 % (Melanopsis

Fossil Record 16 (1) 2013, 77–96 83

Figure 7. Prososthenia neutra Brusina, 1897. A, F. NHMW 2011/0138/0145, sample 090713/2; B. NHMW 2011/0138/0146, sam-ple 0907/Moll4A; C, H. NHMW 2011/0138/0147, sample 0907/Moll4A, distorted specimen; D, G. NHMW 2011/0138/0148,sample 0907/Moll4A; E. NHMW 2011/0138/0149, sample 0907/Moll4A. Scale bar equals 1 mm.


lyrata, Prososthenia neutra, Bania valvatoides, Gyrau-lus pulici, Fossarulus moniliferus and Mytilopsis jadro-vi). This is the only occurrence of the dreissenid bi-valve Mytilopsis jadrovi. The unity of both samples isconfirmed in the analysis by 83.6 % average similarity.Reduced dissimilarity between clusters 3 and 4(36.8 %) is rooted in similar abundance values for Fos-sarulus moniliferus and Melanopsis lyrata. The maindifferences concern the presence of Ferrissia illyrica inCluster 3 and higher percentage value of Bania valva-toides in Cluster 4. Interestingly, the dissimilarity levelsbetween clusters 1 and 4 (51.0 %) and 2 and 4(42.0 %), respectively, is lower than that between clus-ters 1 and 3 (66.6 %) and 2 and 3 (53.8 %). In bothcases, this can be explained by the recurrence espe-cially of Gyraulus pulici and Bania valvatoides in Clus-ter 4. The major dissimilarity is based on values forMelanopsis lyrata and Prososthenia neutra, which areless common or absent in clusters 1 and 2.

The top of Unit D2 is marked by monospecific accu-mulations of Mytilopsis frici (Brusina, 1904) (Mandicet al. 2011). The last sample from Unit E1 did not pro-vide enough material for the analysis. It contains few

individuals of Fossarulus moniliferus, Bania valva-toides, Radix korlevici and Planorbarius sp. Upsection(units E2 and F), fossils are poorly preserved, mostly asmolds. Mollusk diversity is comparatively low; all de-tected specimens belong to Fossarulus moniliferus andRadix korlevici.

Mollusk paleoecology andpaleoenvironment

This investigation, which provides the first statisticalanalysis of the mollusk fauna, allows a refinement ofthe interpretation by Mandic et al. (2011) and Krstićet al. (2009). From the analysis two main groups areevident, referred to as Fossarulus-assemblage and Mel-anopsis-assemblage in literature (Milojević 1966; Man-dic et al. 2011; Fig. 2). The third group, again a Fossa-rulus-assemblage according to Milojević (1966), yieldedonly few and small samples and does not appear in ouranalysis. These assemblages, however, are not definedby the frequent occurrence of their eponymous taxa


Figure 8. Bania valvatoides (Brusina, 1874). A, G. NHMW 2011/0138/0151, sample 0907/Moll4A; B, H. NHMW 2011/0138/0152, sample 090713/2; C. NHMW 2011/0138/0181, sample 090713/2; D. NHMW 2011/0138/0182, sample 090713/2;E, I. NHMW 2011/0138/0183, sample 090713/2; F. NHMW 2011/0138/0153, sample 090713/2. Scale bars equal 1 mm.


Fossil Record 16 (1) 2013, 77–96 85

Figure 9. A. Radix korlevici (Brusina, 1884), NHMW 2011/0138/0154, sample 090713/1; B. Radix korlevici (Brusina, 1884),NHMW 2011/0138/0155, sample 090713/1; C. Galba sp., NHMW 2011/0138/0157, sample 0907/Moll3; D. Radix korlevici (Bru-sina, 1884), NHMW 2011/0138/0156, sample 0804/039 coal. Scale bars equal 5 mm (A–C, E) and 1 mm (D), respectively.

Figure 10. A, F. Gyraulus pulici (Brusina, 1897), NHMW 2011/0138/0158, sample 0804/039; B, E. Gyraulus pulici (Brusina,1897), NHMW 2011/0138/0159, sample 090713/2; C. Gyraulus pulici (Brusina, 1897), NHMW 2011/0138/0160, sample 0907/Moll4A; D. Orygoceras dentaliforme Brusina, 1882, NHMW 2011/0138/0161, sample 0804/039. Scale bars equal 1 mm.


(Fig. 3; Tab. 1). The Fossarulus-assemblage is domi-nated by the planorbid Gyraulus pulici, followed by thehydrobiid Bania valvatoides. Likewise, the Melanopsis-assemblage is characterized by frequent Prososthenianeutra and Fossarulus moniliferus rather than by Mela-nopsis lyrata itself. Nevertheless, the grouping intothree assemblages remains valid based on the largelydeviating faunal compositions (Tab. 1).

The high abundances of the pulmonates Gyraulusand Radix in the first interval (Unit A, clusters 1 and2) argue for a low-energy shallow water environmentwith dense vegetation (Gl�er 2002). Fossarulus, which

appears throughout the whole section, is thought to bea pioneer taxon (Mandic et al. 2011). In Lake Sinj re-presentatives of this fossil genus are found exclusivelyin rather calm, shallow water settings, dominated byhydrobiids (Neubauer et al. 2011). In Lake Gacko itis, in turn, relatively abundant in nearshore as well asdistal environments and thus is regarded to be a eur-yoecious taxon. The ecology of the fossil genus Baniais completely unknown. As its occurrence is oftenbound to the presence of Gyraulus, its requirementsconcerning water quality and vegetation might becomparable.


Figure 11. A–C. Planorbarius sp., apical side, NHMW 2011/0138/0163, sample 0804/039 coal; D–E. Planorbarius sp., apicalside, NHMW 2011/0138/0164, sample 0804/039b; F. Planorbarius sp., umbilical side, NHMW 2011/0138/0165, sample 0804/039b; G, I. Ferrissia illyrica (Neumayr, 1880), NHMW 2011/0138/0166, sample 0804/039b; H, J–K. Ferrissia illyrica (Neumayr,1880), NHMW 2011/0138/0167, sample 0804/039. Scale bars equal 1 mm.


The dominance of pulmonates indicates that theseconditions persisted up to Unit B. Krstić et al. (2009),however, report massive accumulation of Fossaruluswithin limestone intervals of Unit B1 in drill cores,where this unit attains about 40 m. The correspondingsection interval in the Gračanica mine pit measuresonly 2.5 m (Mandic et al. 2011). This discrepancy isexplained by the highly varying sediment infill in theGacko Basin, depending on the position in the lake: themarl units in Unit B are massive in the basin centerand thin out towards the basin margins (Milojević

1966; Mandic et al. 2011). A better correlation betweenboth sites becomes apparent in the following Unit B2with abundant pulmonates in both localities.

The second interval (Cluster 3) is marked by the firstoccurrence of Melanopsis, with smooth to weaklysculptured morphotypes, and the conspicuous increaseof Prososthenia and Fossarulus, while pulmonates de-cline. Species diversity increased significantly with 14taxa, although most of them are represented by onlyfew individuals. In contrast, 10 species were detected inthe first interval.

Fossil Record 16 (1) 2013, 77–96 87

Figure 12. Terrestrial mollusks. A. Carychium sp., NHMW 2011/0138/0168, sample 0804/039; columellar fold not visible in thisview; B, D. Carychium sp., NHMW 2011/0138/0169, sample 0804/039 coal; C, E. Vertigo sp., NHMW 2011/0138/0170, sample0804/039b. Scale bars equal 1 mm.

Figure 13. Mytilopsis frici Kochansky-Devid� & Slišković, 1978. A. RV, interior view, anteriorly most proximal part of exteriorLV with transversal ridge preserved, NHMW 2011/0138/0008, Gacko (historical collection); B. Articulated shell, postero-ventralpart of RV visible beneath antero-dorsally shifted LV, NHMW 2011/0138/0034, Gračanica Section, Mytilopsis bed (Unit D2);C. Fragment of articulated shell showing flabellate dorsal margin, NHMW 2011/0138/0032, Gračanica Section, Mytilopsis bed(Unit D2); D. Fragment of LV showing thickened margin and posterior sinus (interior view), NHMW 2011/0138/0025, GračanicaSection, Mytilopsis bed (Unit D2). Scale bar equals 1 cm.


Today, Melanopsis is known to be a generalist. It oc-curs in freshwater rivers, ponds, springs, shallow lakeswith inundated marshes, mud and gravel shores of estu-

aries, and oases, tolerating high temperatures andbrackish conditions (Brown 1994; Plaziat & Younis2005; Bandel et al. 2007). As the hydrobiid Prososthe-


Figure 14. A–C. Mytilopsis jadrovi (Brusina, 1892), NHMW 2011/0138/0171, sample 0907/Moll4A; D–F. Mytilopsis sp.,NHMW 2011/0138/0172, sample 0804/039b; G–I. Pisidium vukovici n. sp., holotype, NHMW 2011/0138/0173, RV, sample090713/2; J. Pisidium bellardii Brusina, 1884, NHMW 2011/0138/0174, RV, sample 0907/Moll4; K–L. Pisidium bellardii Brusi-na, 1884, NHMW 2011/0138/0175, sample 0907/Moll4, LV. Scale bars equal 1 mm.


nia is a fossil genus, few data on its ecologic require-ments exist. Common interpretations assign it to shal-low freshwater lakes and ponds with low-energy re-gimes (Esu & Girotti 2001; Harzhauser et al. 2012a;Neubauer et al. 2011). Gyraulus and Radix are stillpresent, also indicating shallow, low-energy settings.Additional pulmonate species are documented includ-ing Ferrissia and scarce Orygoceras and Planorbarius.Extant European Ferrissia wautieri is found in lentic tostagnant waters dwelling among leaves and twigs andis basically undemanding regarding water quality. Pla-norbarius corneus lives today in highly vegetated,standing or slowly running water bodies feeding exclu-sively on detritus (Gl�er 2002). Herein, also the onlyrecord of the terrestrial mollusks Carychium and Verti-go is documented signaling a short distance to thewaterside.

At the top of Unit D1 the faunal content is stronglydeviating from earlier/later intervals. The sample (notin analysis) consists of 3 taxa only, highly dominatedby Bithynia and Ferrissia. Extant Bithynia tentaculatais a very adaptive and undemanding r-strategist livingin stagnant and moderately running waters on detritus-rich substrates (Gl�er 2002). Hence, this extremely low

diverse composition featuring almost exclusively highlyadaptive and generally low competitive species suggestsa restricted environment with adverse conditions.

In contrast, only c. 1 m above the second part of theMelanopsis-assemblage (Cluster 4) is characterized bya well-balanced faunal composition with about uniformdistribution of abundances. Melanopsis, Prososthenia,Fossarulus, Bania, and Gyraulus all attain similar per-centage values reflecting the recovering of the faunafrom the preceding crisis. This is also documented bythe increased morphological variability in Melanopsis;for the first time, it develops all five morphotypes,which are considered in detail in the systematic chapter.This genus is known for its morphologically highlyvariable species. Often, distinct morphologies are boundto certain environmental conditions. Heller & Sivan(2002) showed strong correlation of morphotypes withflow current in Pleistocene melanopsids of the JordanValley. Smooth and slender forms represent agitated en-vironments, dwelling within or in the proximity of riv-ers. Sculptured shells, in contrast, reflect non-turbulenthabitats. Comparably, the occurrence of sculptured mor-photypes in Lake Gacko coincides with the establish-ment of perennial open-lake conditions (Mandic et al.

Fossil Record 16 (1) 2013, 77–96 89

Figure 15. Unio rackianus Brusina, 1874, all specimens from sample 090713/2. A–B, E. NHMW 2011/0138/0176, LV;C, F. NHMW 2011/0138/0177, LV; D, G. NHMW 2011/0138/0162, RV. Scale bars equal 1 cm.


2011). Consequently, the low-energy conditions mighthave promoted adaptive evolution towards sculpturedforms.

The following Unit D2 comprises grayish lime-stones signaling partial suboxic bottom conditions(Mandic et al. 2011). Coinciding with the lake-levelfall and a rising oxygen level at the top of the unit amonospecific accumulation of Mytilopsis frici occurs.This pattern is similar to developments in Lake Sinj,where the massive limestone intervals, largely barrenof gastropods, show abundant Mytilopsis aletici. Bothtaxa are interpreted to have settled in deeper, low-en-ergy settings (Mandic et al. 2009; Neubauer et al.2011).

Upsection, at the base of Unit E1, the diverse faunadisappears (base of the second Fossarulus-assemblage).The few and poorly preserved fossils could not be ana-lyzed statistically, but show a return to initial composi-tions featuring abundant pulmonates and Fossarulus.This situation persists up to the section top.

The apparent shifts in the faunal composition, re-garding occurring taxa as well as species richness, areespecially evident in the nMDS (Fig. 4B) where thepulmonate/rissooid-dominated Fossarulus-assemblageand the Melanopsis-assemblage with abundant Pro-sosthenia and Fossarulus group in separate fields. Thevertical axis in the plot is thought to reflect ecologicalconditions. The faunal compositions of the two assem-blages/groups show an ongoing trend to higher diversi-ties and uniform distributions of abundances. The lowdiverse pulmonate-assemblage (G1) points to ephemeralconditions with a lentic to stagnant water body anddensely vegetated watersides. The following higher di-verse assemblage (G2), with taxa having broader ecolo-gical requirements, reflects the establishment of a per-ennial system with stable open-lake conditions andhabitat diversification along the shoreface. In between,local crises occurred for short episodes, as proven bythe Bithynia-Ferrissia complex in Unit D1. The thirdassemblage, not part of the statistical analysis, with acomposition similar to the initial phase, again indicatesa retreat of the lake and the return to ephemeral condi-tions at the lake margins. Taxa requiring stable water-bodies like Melanopsis and Prososthenia are comple-tely missing.

The x-axis in the plot is interpreted to mirror speciesrichness. It correlates particularly well with the Marga-lef richness index (Tab. 1), computed as species rich-ness increasing with the natural logarithm of samplesize (Margalef 1958). Samples grouping to the left con-sist of few taxa, while samples on the right have ahigher number of taxa at comparable sample size.Thus, the probability that a sample yields higher spe-cies richness at higher sample size (the “expected spe-cies richness“) increases towards the right. This, inturn, fits to the interpretation above, as ephemeral habi-tats are expected to house few but abundant taxa, whileperennial systems usually attain distinctly higher spe-cies richnesses.

Conclusions

During the Middle Miocene Climatic Optimum LakeGacko harbored at least 17 aquatic mollusk species.Within the lake, long-term climatic oscillations are ex-cellently reflected in changing faunal compositions,which are proven herein for the first time by means ofstatistical analyses. The initial as well as the final phaseboth signal a more arid climate resulting in a loweredlake level. At that time, assemblages typically are lowdiverse and consist of abundant pulmonate and rissooidgastropods. The intermediate lake high stand coincidedwith a more humid interval with enhanced precipita-tion. The mollusk fauna diversified and rissooid andmelanopsid gastropods flourished, while pulmonateswere only marginally present. This interpretation agreeswith data of Mandic et al. (2011), who provided a moredetailed classification into distinct transgression-regres-sion cycles based on sedimentary facies and geophysi-cal data.

All species are endemic to the DLS, two of them oc-cur in Lake Gacko only. Also, the fauna shows variableoverlap with those of other DLS lakes. This suggeststhat the DLS comprised a mosaic of lakes, each with adistinct species inventory with moderate overlap withneighboring lakes.

Systematic paleontology

Type material of Neumayr (1869, 1880), Brusina (1870, 1872, 1874,1876, 1882, 1884, 1897, 1902, 1904) and Kochansky-Devid� & Sli�ko-vić (1978) stored in the Natural History Museum Zagreb (NHMZ), theNatural History Museum Sarajevo (NHMS) and the Geological SurveyVienna (GBA), and material from the collection of the Natural HistoryMuseum Vienna (NHMW) have been studied. Material from the presentstudy is stored in the NHMW under the prefix 2011/0138. For furthersynonyms see Brusina (1907), Wenz (1923–1930) and Milan et al.(1974). Only the description for the new species is available in the printversion of this paper, others appear as supplementary material.

Class Bivalvia Linnaeus, 1758Superorder Heterodonta Neumayr, 1883Order Venerida Gray, 1854Family Sphaeriidae Deshayes, 1854 (1820)

Pisidium Pfeiffer, 1821

Type species. Tellina amnica M�ller, 1774, Recent, Palearctic andNorth America.

Pisidium vukovici n. sp.

Figures 14G–I

Etymology. After Dr.sc. Boško Vuković, mine geologist of RiTE“Gacko“, for his assistance and help with the geological studies in theGacko Basin.

Holotype. Right valve illustrated in Figures 14G–I, NHMW 2011/0138/0173, sample 090713/2; H ¼ 2.8 mm, L ¼ 3.0 mm, C ¼ 1.2 mm.



Paratype 1. Right valve, NHMW 2011/0138/0178, sample 090713/1.Paratype 2. Left valve, NHMW 2011/0138/0179, sample 090713/2.Paratype 3. Left valve, NHMW 2011/0138/0180, sample 090713/2.

Material. 6 single valves from samples 090713/1 (1 LV, 2 RV) and090713/2 (2 LV, 1 RV) of the type horizon.

Type locality. Open-cast coal-mine “Gračanica“ near Gacko, MinepitB.

Type horizon. Unit A of the Section Gračanica in Mandic et al.(2011), footwall coal II lithostratigraphic unit of the Gacko Basin,Langhian, Middle Miocene (~15.83 Ma).

Diagnosis. Shell subtrigonal. Hinge plate massive; prominent lateralteeth.

Description. Subtriangular shell with regularly semicir-cular dorsal margin. Umbo low and broad, measuring750 mm. In lateral view, shell roughly semilunar. Outersurface covered with faint, regular growth lines. Hingeplate prominent and broad. Ligament pit short and lan-ceolate. Cardinals subcentral, oblique, dipping ante-riorly; C 2 broadened, blunt; C 3 prominent, crescent,convex-up, sharp, surrounded by two shallow cavities;C 4 sharp and straightened. Laterals prominent andlong with bulbous, sharp cusps. On right valve anteriorteeth massive, forming distinct emargination; posteriorteeth somewhat weaker. In both cases inner teeth stron-ger than outer ones.

Remarks. This taxon can be clearly distinguished fromPisidium bellardii Brusina, 1884 by its subtrigonal shellshape and very prominent hinge plate. P. hybonatumBrusina, 1897 is somewhat broader in outline and has aless prominent hinge plate. This species, originally de-scribed from Pliocene deposits of the southern Panno-nian Basin, was additionally reported by Jovanović(1935) from the close-by Bihać Basin. Shells from thepossibly coeval Austrian Aflenz Basin, determined asPisidium cf. casertanum Poli, 1791 by Harzhauser et al.(2012b), are broader and expose a narrower, delicatehinge plate. Likewise, P. amnicum (M�ller, 1774) (=P. priscum H�rnes, 1862; non Eichwald, 1830) from theMiddle to Late Miocene of the Central and EasternParatethys and the Upper Freshwater Molasse is broad-er and has narrower and less distinct lateral teeth (cf.Schneider & Prieto 2011). The about coeval P. bako-nyensis K�kay, 2006 from Hungary is quite similar inshape but exposes an elevated umbo and smaller lateralteeth. In shape and hinge characteristics it is similar tothe Recent P. supinum Schmidt, 1851, but does not de-velop such a high umbo with the typical fold. ExtantP. tenuilineatum Stelfox, 1918 has stronger posteriorteeth and a transversally oriented C 3.

Acknowledgements

We are grateful to the following colleagues for support: Boško Vuko-vić and authorities of RiTE “Gacko“ granted access to the mine. Jo-van Olujić (Geological Survey Zvornik) introduced us to the studiedlocalities. Stjepan Ćorić (Geological Survey Vienna) kindly providedsamples from drill cores. Simon Schneider (Natural History MuseumVienna) gave helpful comments on bivalve taxonomy. Zlata Jurišić-

Polšak (Natural History Museum Zagreb), Božena Čvorović (NaturalHistory Museum Sarajevo) and Irene Zorn (Geological Survey Vien-na) made type collections accessible. Medina Mandic assisted withfossil collections in the field. Alice Schumacher (Natural History Mu-seum Vienna) provided macro-photographs. Finally, we are grateful toDaniela Esu (University of Rome) and Vitaliy V. Anistratenko(Schmalhausen Institute of Zoology of NAS, Ukraine) for constructivereviews and taxonomic discussion. The investigation contributes tothe Austrian Science Fund projects P18519-B17 and P25365.

References

Bandel, K., Sivan, N. & Heller, J. 2007. Melanopsis from Al-Qarn,Jordan Valley (Gastropoda: Cerithioidea). – Pal�ontologischeZeitschrift 81 (3): 304–315.

Brown, D. S. 1994. Freshwater Snails of Africa and their Medical Im-portance. Taylor & Francis, London.

Brusina, S. 1870. Monographie der Gattungen Emmericia und Fossa-rulus. – Verhandlungen der kaiserlichen und k�niglichen zoolo-gisch-botanischen Gesellschaft in Wien 20: 925–938.

Brusina, S. 1872. Naravoslovne crtice sa sjevero-istočne obale Ja-dranskoga mora. Dio prvi. Putopis. – Rad Jugoslavenske akade-mije znanosti i umjetnosti 19: 105–177.

Brusina, S. 1874. Fossile Binnen-Mollusken aus Dalmatien, Kroatienund Slavonien nebst einem Anhange. Actienbuchdruckerei,Agram.

Brusina, S. 1876. Decription d’esp�ces nouvelles, provenant des ter-rains tertiaires de Dalmatie. – Journal de Conchyliologie 24:109–116.

Brusina, S. 1882. Orygoceras. Eine neue Gasteropodengattung derMelanopsiden-Mergel Dalmatiens. – Beitr�ge zur Pal�ontologiesterreich-Ungarns und des Orients 2: 33–46.

Brusina, S. 1884. Die Neritodonta Dalmatiens und Slavoniens nebst al-lerlei malakologischen Bemerkungen. – Jahrb�cher der DeutschenMalakozoologischen Gesellschaft 11: 17–120.

Brusina, S. 1892. Fauna fossile terziaria di Markuševec in Croazia.Con un elenco delle Dreissensidae della Dalmazia, Croazia e Sla-vonia. – Glasnik hrvatskog naravoslovnog društa 7: 113–210.

Brusina, S. 1897. Gragja za neogensku malakološku faunu Dalmacije,Hrvatske i Slavonije uz neke vrste iz Bosne i Hercegovine iSrbije. – Djela Jugoslavenske akademije znanosti i umjetnosti 18:1–43.

Brusina, S. 1902. Iconographia Molluscorum Fossilium in tellure ter-tiaria Hungariae, Croatiae, Slavoniae, Dalmatiae, Bosniae, Herze-govinae, Serbiae and Bulgariae inventorum. Officina Soc. Typo-graphicae, Agram.

Brusina, S. 1904. Šipovo i njegova tercijarna fauna. – Glasnik Ze-maljskog muzeja Bosne i Hercegovine 16 (4): 493–498.

Brusina, S. 1907. Naravoslovne crtice sa sjevero-istočne obale Ja-dranskoga mora. Dio cetvrti i posljednji. Specijalni. – Rad Jugo-slavenske akademije znanosti i umjetnosti 169: 195–251.

Clarke, K. R. & Warwick, R. M. 2001. Changes in Marine Commu-nities: An Approach to Statistical Analysis and Interpretation. 2ndedition. Plymouth Marine Laboratory, Plymouth.

De Leeuw, A., Mandic, O., Krijgsman, W., Kuiper, K. & Hrvatović,H. 2011. A chronostratigraphy for the Dinaride Lake System de-posits of the Livno-Tomislavgrad Basin: the rise and fall of along-lived lacustrine environment. – Stratigraphy 8 (1): 29–43.

De Leeuw, A., Mandic, O., Vranjković, A., Pavelić, D., Harzhauser,M., Krijgsman, W. & Kuiper, K. F. 2010. Chronology and inte-grated stratigraphy of the Miocene Sinj Basin (Dinaride Lake Sys-tem, Croatia). – Palaeogeography, Palaeoclimatology, Palaeoecol-ogy 292: 155–167.

Deshayes, G. P. 1854. Catalogue of the Conchifera or bivalve shells inthe collection of the British Museum, 1–2. British Museum, Lon-don.

Fossil Record 16 (1) 2013, 77–96 91


Eichwald, E. 1830. Naturhistorische Skizze von Lithauen, Volhynienund Podolien in geognostisch-mineralogischer, botanischer undzoologischer Hinsicht. Leipzig (published by author).

Esu, D. & Girotti, O. 2001. The genus Prososthenia Neumayr in Italy(Gastropoda: Prosobranchia: Hydrobiidae). – Bollettino della So-ciet Paleontologica Italiana 40 (2): 179–184.

Gl�er, P. 2002. Die Tierwelt Deutschlands, 73. Teil: Die S�ßwasser-gastropoden Nord- und Mitteleuropas. Bestimmungsschl�ssel, Le-bensweise, Verbreitung. ConchBooks, Hackenheim.

Gray, J. E. 1854. A revision of the arrangement of the families ofbivalve shells (Conchifera). – Annals and Magazine of NaturalHistory (series 2) 13 (77): 408–418.

Hammer, Ø., Harper, D. A. T. & Ryan, P. D. 2001. Past: Paleontologi-cal Statistics Software Package for Education and Data Analysis.– Palaeontologia Electronica 4: 1–9.

Harzhauser, M. & Mandic, O. 2008. Neogene lake systems of Centraland South-Eastern Europe: Faunal diversity, gradients and interre-lations. – Palaeogeography, Palaeoclimatology, Palaeoecology260: 417–434.

Harzhauser, M. & Mandic, O. 2010. Neogene dreissenids in CentralEurope: evolutionary shifts and diversity changes. In van derVelde, G., Rajagopal, S. & bij de Vaate, A. (eds). The Zebra Mus-sel in Europe. Backhuys Publishers, Leiden/Margraf Publishers,Weikersheim: pp. 11–29.

Harzhauser, M., Mandic, O., Latal, C. & Kern, A. 2012a. Stable iso-tope composition of the Miocene Dinaride Lake System deducedfrom its endemic mollusc fauna. – Hydrobiologia 682: 27–46.

Harzhauser, M., Neubauer, T. A., Mandic, O., Zuschin, M. & Ćorić,S. 2012b. A Middle Miocene endemic freshwater mollusc assem-blage from an intramontane Alpine lake (Aflenz Basin, EasternAlps, Austria). – Pal�ontologische Zeitschrift 86 (1): 23–41.

Harzhauser, M. & Piller, W. E. 2007. Benchmark data of a changingsea. – Palaeogeography, Palaeobiogeography and Events in theCentral Paratethys during the Miocene. – Palaeogeography, Pa-laeoclimatology, Palaeoecology 253: 8–31.

Heller, J. & Sivan, N. 2002. Melanopsis from the Pleistocene site of’Ubeidiya, Jordan Valley: direct evidence of early hybridization(Gastropoda: Cerithioidea). – Biological Journal of the LinneanSociety 75: 39–57.

H�rnes, M. 1859–1867. Die fossilen Mollusken des Terti�r-Beckensvon Wien. II. Bivalven. – Abhandlungen der Geologischen Reichs-anstalt 4: 1–430 [1–116 (1859); 117–214 (1862); 215–342(1965); 343–430 (1867)].

Hrvatović, H. 2006. Geological Guidebook through Bosnia and Her-zegovina. Geological Survey of Federation BiH, Sarajevo.

Jovanović, R. N. 1935. Prilog fauni Bihackog tercijarnog bazena. –Vesnik geoloskog instituta kraljevine Jugoslavije 4: 157–176.

Kochansky-Devid�, V. & Slišković, T. 1978. Miocenske kongerijeHrvatske, Bosne i Hercegovine. – Palaeontologia jugoslavica 19:1–98.

Krstić, N., Olujić, J., �ajić, S., �ord–ević-Milutinović, D. & Tanasković,L. 2009. Fossils from the drill hole GS-1 near Gacko, SE DinaricAlps. – Bulletin of the Natural History Museum 2: 35–61.

Krstić, N., Savić, L., Jovanović, G. & Bodor, E. 2003. Lower Miocenelakes of the Balkan Land. – Acta Geologica Hungarica 46: 291–299.

K�kay, J. 2006. Nonmarine mollusc fauna from the Lower and MiddleMiocene, Bakony Mts., W Hungary. – Geologica Hungarica, Ser-ies Palaeontologica 56: 1–196.

Linnaeus, C. 1758. Systema naturae per regna tria naturae, secundumclasses, ordines, genera, species, cum characteribus, differentiis,synonymis locis. Tomus 1. Laurentius Salvius, Holmiae.

Mandic, O., De Leeuw, A., Vuković, B., Krijgsman, W., Harzhauser,M. & Kuiper, K. F. 2011. Palaeoenvironmental evolution of Lake

Gacko (Southern Bosnia and Herzegovina): Impact of the MiddleMiocene Climatic Optimum on the Dinaride Lake System. – Pa-laeogeography, Palaeoclimatology, Palaeoecology 299: 475–492.

Mandic, O., Pavelić, D., Harzhauser, M., Zupanič, J., Reischenbacher,D., Sachsenhofer, R. F., Tadej, N. & Vranjković, A. 2009. Deposi-tional history of the Miocene Lake Sinj (Dinaride Lake System,Croatia): a long-lived hard-water lake in a pull-apart tectonic set-ting. – Journal of Paleolimnology 41: 431–452.

Margalef, R. 1958. Information theory in ecology. – General systema-tics 3: 1–10.

Milan, A., Sakač, K. & Žagar-Sakač, A. 1974. Katalog originala tipo-va vrsta pohranjenih u Geološko-paleontološkom muzeju u Zagre-bu. Geološko-paleontološki muzej u Zagrebu, Zagreb.

Milojević, R. 1966. Korelacioni i neki ekonomsko-geološki elementirazvoja ugljenonosne serije Gatačkog bazena. – Geološki glasnik11: 355–370.

Milojević, R. 1976. Mineralne sirovine Bosne i Hercegovine. Prvitom. Ležišta uglja. Ležišta nemetala. Geoinžinjering, Sarajevo.

Mirković, M., Kalezić, M., Pajović, M., Rašković, S., Cepić, M. &Vujisić, P. 1974. Gacko. Osnovna geološka karta 1:100 000. Sa-vezni geološki zavod, Beograd.

Mojićević, M. & Laušević, M. 1965. Nevesinje. Osnovna geološkakarta 1:100 000. Savezni geološki zavod, Beograd.

M�ller, O. F. 1773–1774. Vermium terrestrium et fluviatilium histor-ia, seu animalium Infusoriorum, Helminthicorum et Testaceorumnon marinorum succincta historia. Heineck & Faber, Havniae etLipsiae.

Neubauer, T. A., Mandic, O. & Harzhauser, M. 2011. Middle MioceneFreshwater Mollusks from Lake Sinj (Dinaride Lake System, SECroatia; Langhian). – Archiv f�r Molluskenkunde 140 (2): 201–237.

Neubauer, T. A., Mandic, O., Harzhauser, M. & Hrvatović, H. 2012 inpress. A new Miocene lacustrine mollusc fauna of the DinarideLake System and its palaeobiogeographic, palaeoecologic, andtaxonomic implications. – Palaeontology.

Neumayr, M. 1869. II. Beitr�ge zur Kenntniss fossiler Binnenfaunen.– Jahrbuch der kaiserlichen und k�niglichen geologischen Reichs-anstalt 19: 355–382.

Neumayr, M. 1880. V. Terti�re Binnenmollusken aus Bosnien und derHercegovina. – Jahrbuch der kaiserlichen und k�niglichen geolo-gischen Reichsanstalt 30 (2): 463–486.

Pfeiffer, C. 1821. Naturgeschichte deutscher Land- und S�sswasser-Mollusken. Erste Abtheilung. Landes-Industrie-Comptoir, Weimar.

Plaziat, J.-C. & Younis, W. R. 2005. The modern environments ofMolluscs in southern Mesopotamia, Iraq: A guide to paleogeogra-phical reconstructions of Quaternary fluvial, palustrine and mar-ine deposits. – Carnets de G�ologie/Notebooks on Geology, Arti-cle 2005/01 (CG2005_A01).

Poli, J. X. 1791. Testacea utriusque Siciliae eorumque historia et ana-tome tabulis aeneis illustrata. Tomus primus. Regio Typographeio,Parmae.

R�gl, F. 1998. Paleogeographic considerations for Mediterranean andParatethys seaways (Oligocene to Miocene). – Annalen des Na-turhistorischen Museums in Wien 99A: 279–310.

Schmidt, A. 1851. Malakologische Mittheilungen. – Zeitschrift f�rMalakozoologie 7 (8): 113–120.

Schneider, S. & Prieto, J. 2011. First record of an autochthonous com-munity of fluviatile freshwater molluscs from the Middle/LateMiocene Upper Freshwater Molasse (southern Germany). – Ar-chiv f�r Molluskenkunde 140 (1): 1–18.

Stelfox, A. W. 1918. The Pisidium fauna of the Grand Junction Canalin Herts. and Bucks. – Journal of Conchology 15 (10): 289–304.

Wenz, W. 1923–1930. Fossilium Catalogus I: Animalia. Gastropodaextramarina tertiaria. W. Junk, Berlin.



Appendix

Fossil Record 16 (1) 2013, 77–96 93

Table 1. Simper analysis, using Bray-Curtis distances and a cut-off level for low contributions of 90%. Additionally, data onMargalef ’s richness index are included.

Groups Clusters Margalef index

0708/GA1 1 1 0.456

0907/Moll2 1 1 0.858

0907/Moll1 1 2 1.077

090713/1 1 2 1.507

090713/2 1 2 1.447

0804–039 2 3 1.719

0804–039 coal 2 3 1.853

0804–039b 2 3 2.212

0907/Moll4 2 4 1.580

0907/Moll4A 2 4 1.514

Groups

Group 1 –– Average similarity: 71.69%

Species Av. Abund Av. Sim Sim/SD Contrib % Cum. %

Gyraulus pulici 6.66 27.28 2.80 38.05 38.05

Bania valvatoides 4.57 18.28 6.08 25.50 63.55

Fossarulus moniliferus 3.60 14.49 6.17 20.21 83.76

Radix korlevici 2.92 8.04 1.14 11.21 94.97

Group 2 – Average similarity: 70.05%


Prososthenia neutra 5.27 18.77 14.82 26.80 26.80


Melanopsis lyrata 3.87 11.76 2.92 16.79 66.69

Gyraulus pulici 2.65 7.36 3.16 10.50 77.19

Radix korlevici 2.07 5.99 2.33 8.55 85.74


Groups 1 & 2 – Average dissimilarity: 53.56%

Group 1 Group 2

Species Av. Abund Av. Abund Av. Diss Diss/SD Contrib % Cum. %

Prososthenia neutra 1.06 5.27 9.23 3.00 17.23 17.23

Gyraulus pulici 6.66 2.65 9.00 1.68 16.81 34.04

Melanopsis lyrata 0.00 3.87 8.39 2.96 15.66 49.70

Bania valvatoides 4.57 2.04 5.81 1.51 10.85 60.55

Fossarulus moniliferus 3.60 5.00 4.03 1.66 7.53 68.08

Radix korlevici 2.92 2.07 3.70 1.63 6.92 75.00

Ferrissia illyrica 0.00 1.34 2.85 1.06 5.33 80.32

Unio rackianus 0.92 0.00 1.79 0.80 3.34 83.67

Mytilopsis jadrovi 0.00 0.71 1.56 0.79 2.91 86.58

Orygoceras dentaliforme 0.00 0.72 1.53 1.17 2.86 89.44

Pisidium bellardii 0.45 0.91 1.53 1.26 2.85 92.29



Table 1. Continued

Clusters

Cluster 1 – Average similarity: 81.67%


Gyraulus pulici 8.60 48.49 *** 59.38 59.38

Bania valvatoides 3.43 17.98 *** 22.02 81.39

Fossarulus moniliferus 3.20 15.20 *** 18.61 100.00




Gyraulus pulici 5.38 20.16 15.42 24.80 50.26

Radix korlevici 4.05 14.96 20.91 18.40 68.66







Radix korlevici 2.66 9.55 10.86 12.06 64.26

Melanopsis lyrata 3.06 9.34 2.75 11.80 76.06

Ferrissia illyrica 2.23 6.42 21.22 8.11 84.16

Gyraulus pulici 1.96 5.80 15.25 7.33 91.49



Melanopsis lyrata 5.10 18.70 *** 22.37 22.37

Prososthenia neutra 4.73 18.44 *** 22.06 44.43

Bania valvatoides 3.80 11.64 *** 13.92 58.35

Gyraulus pulici 3.69 10.97 *** 13.12 71.47

Fossarulus moniliferus 3.78 9.86 *** 11.79 83.26

Mytilopsis jadrovi 1.79 6.23 *** 7.46 90.72


Fossil Record 16 (1) 2013, 77–96 95

Table 1. Continued

Clusters 1 & 2 – Average dissimilarity: 34.78%

Cluster 1 Cluster 2


Gyraulus pulici 8.60 5.38 8.04 2.78 23.11 23.11

Radix korlevici 1.23 4.05 7.17 1.73 20.62 43.72



Unio rackianus 0.00 1.53 3.68 1.28 10.57 81.10


Pisidium vukovici n. sp. 0.00 0.88 2.12 1.28 6.10 94.84


Cluster 1 Cluster 3


Gyraulus pulici 8.60 1.96 15.62 5.61 23.47 23.47






Radix korlevici 1.23 2.66 3.52 1.07 5.29 85.45


Mytilopsis sp. 0.00 0.89 2.10 1.20 3.16 92.81


Cluster 2 Cluster 3




Gyraulus pulici 5.38 1.96 6.96 2.65 12.95 44.24




Unio rackianus 1.53 0.00 2.97 1.33 5.52 76.93

Radix korlevici 4.05 2.66 2.76 1.79 5.13 82.06


Mytilopsis sp. 0.00 0.89 1.80 1.23 3.36 89.87

Pisidium vukovici n. sp. 0.88 0.00 1.71 1.33 3.18 93.06



Table 1. Continued


Cluster 1 Cluster 4



Gyraulus pulici 8.60 3.69 11.82 3.39 23.17 47.06




Radix korlevici 1.23 1.18 2.93 5.50 5.75 89.31

Pisidium bellardii 0.00 1.06 2.52 13.20 4.95 94.26


Cluster 2 Cluster 4




Radix korlevici 4.05 1.18 5.82 3.69 13.88 53.16


Gyraulus pulici 5.38 3.69 3.56 1.18 8.49 70.36


Unio rackianus 1.53 0.00 3.01 1.29 7.18 85.50



Cluster 3 Cluster 4







Gyraulus pulici 1.96 3.69 3.32 1.51 9.04 66.76

Radix korlevici 2.66 1.18 2.85 4.63 7.76 74.52



Mytilopsis sp. 0.89 0.00 1.73 1.20 4.71 90.23


Supplementary online material

The Middle Miocene freshwater mollusk fauna of Lake Gacko (SE Bosnia and Herzegovina): taxonomic revision and paleoenvironmental analysis Thomas A. Neubauer*, Oleg Mandic and Mathias Harzhauser Department of Geology & Paleontology, The Natural History Museum Vienna, Burgring 7, 1010 Wien, Austria; e-mails: [email protected]; [email protected]; [email protected] *corresponding author Systematic paleontology Type material of Neumayr (1869, 1880) and Brusina (1870, 1872, 1874, 1876, 1882, 1884, 1897, 1902, 1904) stored in the Natural History Museum Zagreb (NHMZ), the Natural History Museum Sarajevo (NHMS) and the Geological Survey Vienna (GBA), and material from the collection of the Natural History Museum Vienna (NHMW) have been studied. Material from the present study is stored in the NHMW under the prefix 2011/0138. For further synonyms see Brusina (1907), Wenz (1923–1930) and Milan et al. (1974). Class Gastropoda Cuvier, 1797 Subclass Caenogastropoda Cox, 1959 Order Cerithiimorpha Golikov & Starobogatov, 1975 Superfamily Cerithioidea Fleming, 1822 Family Melanopsidae H. & A. Adams, 1854 Melanopsis Férussac, 1807 Type species. Buccinum praemorsum Linnaeus, 1758, Recent, Spain. Diagnosis. Ovate to conical shell, highly variable in outline. Protoconch smooth, comprising a little more than one whorl, not demarcated from the teleoconch. Acute oval aperture with siphonal canal and usually with prominent callus pad. Axial or spiral sculpture might be present. No umbilicus developed (modified after Bandel 2000). Melanopsis lyrata Neumayr, 1869 Figures 5A–G 1869 Melanopsis (Canthidomus) lyrata Neumayr, p. 358, pl. 11, fig. 8. 2011 Melanopsis lyrata. – Mandic et al., figs 5.1–4.

1

2011 Melanopsis lyrata. – Neubauer et al., p. 207, pl. 1, figs 1–9, 22 [cum syn.]. Holotype. GBA 1869/01/6. Type locality. Ribarić N Maljkovo, Sinj Basin, SE Croatia; Middle Miocene. Material. 98 specimens from samples 0804/039, 0804/039 coal, 0804/039b, 0804/039 debris, 0907/Moll4 and 0907/Moll4A; mostly fragmented. Description. Slender to rather bulky conical shell with smooth protoconch, measuring around 1 whorl. Usually up to 15 mm in height and 6 mm in width; occasionally forms occur with height > 20 mm. Shape highly variable, ranging from broad and rather sturdy forms with inflated last whorl and coeloconoid outline up to highly slender ones with straight outline. Shell with 8–9 whorls that are moderately convex to straight-sided. Last whorl attaining somewhat more than half of shell height. Aperture rather narrow, exposing short siphonal canal. In broad morphotypes weak callus pad and/or columellar swelling may be developed. Coloring preserved in some individuals in the form of yellow/brownish-white chess or zigzag pattern. Morphotypes also differing in sculpture. Early teleoconch always entirely smooth. On last 1–2 whorls sculpture may occur in the form of weak to distinct axial ribs and/or two rows of nodes below upper suture. In most specimens from Vrbica lower row stronger expressed (Fig. 5G). Nodes forming well-rounded knobs or sometimes produce rather distinct spikes. Expression of ribs and nodes may vary even on same individual. Thus, a classification into discrete morphotypes is difficult. Five forms can be roughly distinguished: smooth (morphotype A; not illustrated); intentions of ribs on last whorl (morphotype B; Figs 5A–B); weak ribs on last two whorls (morphotype C; Figs C–D); more distinct ribs with two rows of weak nodules (morphotype D; Fig. 5E); ribs with strong nodes, occasionally ribs reduced (morphotype E; Figs 5F–G). Remarks. The variability of Melanopsis lyrata was shown by Olujić (1999), Mandic et al. (2011), and Neubauer et al. (2011). The specimens found in Vrbica with distinct ribs and nodes (morphotype E) correspond to the Lake Sinj costata-morphotype in Olujić (1999) and morph C in Neubauer et al. (2011), respectively. However, the prominence of the nodes in the lower row is uncommon in Lake Sinj morphotypes. Those recorded from Gračanica mine (Fig. 5A–E; Mandic et al. 2011) are usually weaker sculptured with blunt riblets and/or regularly rounded small knobs. The slender Melanopsis visianiana Brusina, 1874 can be distinguished by the distinct angle between flank of the last whorl and the concave base. Melanopsis sostarici Brusina, 1897 from Dugo Selo (Glina subdepression, southern Pannonian Basin) has fewer and relatively higher whorls, has no callus pad and develops two fascioles on the neck. The bulky forms are reminiscent of M. camptogramma Brusina, 1876 and M. lanzaeana Brusina, 1874, which usually develop a more bulky body whorl. The overall similar shape might indicate a close relationship. Occurrence. Gacko Basin (Vrbica, Gračanica). Sinj Basin (Lučane W Sinj, Ribarić N Maljkovo), Drniš Basin (Miočić) (Neumayr 1869; Brusina 1874). Order Littorinimorpha Golikov & Starobogatov, 1975 Superfamily Rissooidea Gray, 1847 Family Bithyniidae Gray, 1857

2

Bithynia Leach in Abel, 1818 Type species. Helix tentaculata Linnaeus, 1758, Recent, Europe. Diagnosis. Conical to ovoid shell with highly convex to stepped whorls. Calcified operculum with concentrical growth lines (modified after Glöer 2002). Bithynia jurinaci Brusina, 1884 Figures 6G–H 1869 Bythinia [sic] tentaculata. – Neumayr, p. 363, pl. 12, fig. 8 [non Helix tentaculata Linnaeus, 1758]. 1884 Bythinia [sic] Jurinaci Brusina, p. 53. non 2004 Bithynia jurinaci. – Harzhauser & Binder, p. 7, pl. 2, figs 8–11. Neotype. NHMZ 3853-1493 (syntypes lost). Type locality. Miočić, Drniš Basin, SE Croatia; Middle Miocene. Material. 1 complete specimen and 1 operculum from 0804/039 debris, 17 opercula from 0708/GA14. Description. Broad conical shell, measuring 8 mm in height and 5 mm in width, consisting of 5 convex whorls. Protoconch insufficiently preserved. Last whorl particularly broad, sometimes even developing weak shoulder, attaining 65–75% of shell height. In late ontogeny last whorl displaying weakly allometric growth, reflected by stronger anteriorly-oriented growth. Base straight to slightly convex. Aperture only rudimentarily preserved but known to be roughly ovoid, exposing typical notch at posterior part of inner lip, resulting in posterior tip. Umbilicus narrow, slit-like. In lateral view, aperture inclined to the axis with c. 10°; lips parallel. Directly behind peristome thin, shallow furrow for operculum occurring inside shell. Calcified operculum showing fine concentrical growth lines. Remarks. This species is the only documented Bithynia from the DLS. It is morphologically closely related to the extant B. tentaculata (Linnaeus, 1758). Harzhauser & Binder (2004) identified specimens from the Late Miocene of the Vienna Basin as B. jurinaci. Despite the similar morphology the large stratigraphic gap of more than 5 Ma raises doubts on that identification and the Pannonian species might need a reevaluation. Occurrence. Gacko Basin (Vrbica, Gračanica). Sinj Basin (Ruduša [= Goručica] and Župića potok in Sinj, Trnovača, Turiake), Drniš Basin (Miočić) (Brusina 1884). Family Emmericiidae Brusina, 1870 Fossarulus Neumayr, 1869 Type species. Fossarulus stachei Neumayr, 1869, Miocene, Dalmatia/SE Croatia. Diagnosis. Shell bulky to slender conical, often with spiral sculpture. Protoconch smooth, large and immersed. Umbilicus narrow. Aperture broad and ovoid. Lips

3

continuous and thickened, outer lip sinusoidal in lateral view (modified after Neumayr 1869). Fossarulus moniliferus Brusina, 1876 Figures 6A–F, I 1876 Fossarulus moniliferus Brusina, p. 111. 1882 Fossarulus fuchsi Brusina, p. 38 [nomen nudum]. 1897 Fossarulus Bulići Brusina, p. 22, pl. 8, figs 19–20. 1897 Fossarulus Buzolići Brusina, p. 22, pl. 8, figs 14, 17–18. 1897 Fossarulus Buzolići complanatus Brusina, p. 22, pl. 8, figs 15–16. 1897 Fossarulus fuchsi. – Brusina, p. 21, pl. 7, figs 27–28. 1897 Fossarulus moniliferus. – Brusina, p. 21, pl. 8, figs 3–6. 1902 Fossarulus Bulići. – Brusina, pl. 11, figs 40–43. 2009 Fossarulus tricarinatus. – Krstić et al., p. 42, pl. 1, figs 1–2 [non Fossarulus tricarinatus Brusina, 1870]. 2011 Fossarulus bulici. – Mandic et al., fig. 5.6. 2011 Fossarulus buzolici. – Mandic et al., fig. 5.7. 2011 Fossarulus fuchsi. – Mandic et al., fig. 5.8. 2011 Fossarulus fuchsi. – Neubauer et al., p. 212, pl. 5, figs 3–4, 9, 13. Holotype. NHMZ 3024-670. Type locality. Ribarić N Maljkovo, Sinj Basin, SE Croatia; Middle Miocene. Material. 228 specimens from samples 0804/039, 0804/039 coal, 0804/039b, 0804/039 debris, 0907/Moll1, 0907/Moll2, 0907/Moll3, 0907/Moll4, 0907/Moll4A, 090713/1, 090713/2, 0708/GA1, 0703/024, Gacko 1; mostly fragmented. Description. Conical shell with 4–4.5 moderately to strongly convex whorls and maximum height of 9.5 mm. Protoconch smooth, strongly immersed, with less than 1 whorl, grading continuously into teleoconch; measures 650–700 µm. Depending on highly variable sculpture three morphotypes can be distinguished: smooth (morphotype A); keeled (B); another form with nodes on the keels (C). Especially between morphs A and B transitions are observed in the form of faint keels (Fig. 6B). If present, 3 keels are usually expressed; on convex base additional weak subjacent keels may emerge. Short sutural ramp developed above upper keel. Aperture ovoid with typical posterior notch on inner lip. Sometimes posterior canal developed. Outer lip thickened and occasionally everted; in profile exposing posterior and wide anterior emarginations. Remarks. Within the DLS a great number of Fossarulus species and subspecies were introduced. The conspicuous morphological overlap of several of these taxa suggests that several of the taxa are superfluous. All forms mentioned in the synonymy list correspond largely regarding shape, size and whorl convexity but display some variability concerning sculpture. These are mere morphotypes rather than distinct species: Fossarulus buzolici complanatus represents the smooth morphotype A; F. buzolici displays a transitional form with faint keels (morphotype A/B); F. moniliferus and F. bulici comprise finely keeled specimens

4

(morphotype B); F. fuchsi is the strongly sculptured with distinct nodes on the keels (morphotype C) exclusively found in Vrbica. Maybe also the more bulky F. stachei Neumayr, 1869, F. armillatus Brusina, 1876, F. auritus Brusina, 1897, and F. hoernesi Brusina, 1897 are only extreme morphs of a single polymorphic species. More detailed studies of type material will have to be performed to elucidate these relations. Fossarulus tricarinatus Brusina, 1870 and F. eginae Brusina, 1897 are more slender, have less convex whorls and a relatively broad last whorl. Specimens from drill cores determined as F. tricarinatus by Krstić et al. (2009) fully correspond to the present description. Occurrence. Gacko Basin (Vrbica, Gračanica). Sinj Basin (Potravlje, Ribarić N Maljkovo), Drniš Basin (Miočić) (Brusina 1876, 1882, 1897). Family Hydrobiidae Stimpson, 1865 Subfamily Pyrgulinae Brusina, 1881 Prososthenia Neumayr, 1869 Type species. Prososthenia schwartzi Neumayr, 1869, Miocene, Dalmatia/SE Croatia. Diagnosis. Small shell, conical to ovoid. Protoconch weakly granular. Last whorl often narrow and allometric. Aperture oblique and ovoid. Lips continuous and thickened, with outer lip protruding centrally in lateral view (modified after Neumayr 1869). Prososthenia neutra Brusina, 1897 Figures 7A–H 1897 Prososthenia? neutra Brusina, p. 19, pl. 9, figs 3–4, 7–8. 2011 Prososthenia neutra. – Mandic et al., figs 5.9–10. 2011 Prososthenia neutra. – Neubauer et al., p. 209, pl. 4, figs 3–4, 6, 12. Holotype. NHMZ 3041-687. Type locality. Miočić, Drniš Basin, SE Croatia; Middle Miocene. Material. 160 specimens from samples 0804/039, 0804/039 coal, 0804/039b, 0804/039 debris, 0907/Moll1, 0907/Moll4, 0907/Moll4A, 090713/1 and 090713/2. Description. Drop-shaped, smooth shell with up to 6 moderately convex whorls. Protoconch granular, consisting of about 1 whorl; continuous transition to teleoconch. This taxon comprises individuals with a variety of shapes, ranging from fairly broad to extremely slender. Whorls moderately convex. Last whorl making up c. 60% of shell height, grading into slightly convex base. Aperture including posterior tip poorly thickened and not detached; ranging from ovoid to semilunar in slender elongated specimens. Sometimes thin umbilicus developed. Laterally, quite weak posterior indentation occurring at outer lip. Apart from fine prosocline growth lines faint spiral lines covering whorls. In some specimens traces of weak subsutural band may be developed.

5

Remarks. Its apertural characteristics classify it as true Prososthenia. It is broader and has rounder whorls than P. eburnea Brusina, 1884. Occurrence. Gacko Basin (Vrbica, Gračanica). Sinj Basin (Lučane W Sinj), Drniš Basin (Miočić) (Brusina 1897; Neubauer et al. 2011). Subfamily unknown Bania Brusina, 1896 Type species. Bania prototypica Brusina, 1872, Middle Miocene, Dalmatia/SE Croatia; by monotypy (ICZN 2001). Diagnosis. Small and sturdy to broad ovoid shell. Protoconch reticulate. Whorls convex to step-like. Aperture ovoid, simple or thickened, occasionally with opercular ridge (compiled after Brusina 1874, 1897; Neubauer et al. 2012). Bania valvatoides (Brusina, 1874) Figures 8A–I 1872 Stalioa valvatoides Brusina, p. 144 [nomen nudum]. 1874 Stalioa valvatoides. – Brusina, p. 61, pl. 4, figs 9–10. 1897 Pseudoamnicola? Stošićiana crassa Brusina, p. 22, pl. 12, figs 17–18. 1896 Bania valvatoides. – Brusina, p. 130. 2009 Pseudoamnicola (Staja) šoštarićiana [sic]. – Krstić et al., pl. 1, fig. 5. 2011 Bania prototypica. – Mandic et al., fig. 5.14 [non Stalioa prototypica Brusina, 1874]. Neotype. NHMZ 3925-1565/1-3 (holotype lost). Type locality. Miočić, Drniš Basin, SE Croatia (Holotype from Goručica in Sinj, Sinj Basin, SE Croatia); Middle Miocene. Material. 139 specimens from samples 0804/039, 0907/Moll1, 0907/Moll2, 0907/Moll4, 0907/Moll4A, 090713/1, 090713/2, 0708/GA1 and 0703/024. Further occurrences (< 1 mm) are recorded from 0804/039 coal, 0804/039b and 0708/GA14. Description. Small, shiny and bulky shell with up to 4 whorls. Protoconch consisting of 1.1 whorls; initially covered with mesh of densely arranged wrinkles, fading out towards transition to teleoconch and forming pattern of irregular spiral threads. Spire shape variable, ranging from broad and bulky to more slender and drop-shaped in some cases. Whorls strongly convex with moderately incised sutures, usually exposing a bluntly step-like profile. Last whorl attaining up to 80% of total height. Aperture elliptical to ovoid, only touching base of preceding whorl in its posterior part, forming fairly wide umbilicus. In lateral view, outer lip slightly protruding below umbilicus, forming broad basal emargination. Peristome sharply edged in earlier ontogeny; becomes distinctly everted and sometimes slightly thickened in latest stages. Opercular ridge developed in fully-grown specimens, ranging from weak riblet to strong varix-like structure (as in type material). Distinct prosocline growth lines and faint spiral grooves covering whorl surface.

6

Remarks. The morphological variability of this species is enormous. Shell shape ranges from low valvatoid, as is the case for the (lost) holotype and the syntypes, to higher coiled, drop-shaped forms, including all kind of intermediates. Likewise, the expression of the opercular ridge and the thickening of the peristome vary greatly, partly depending on ontogenetic stage. Concerning shell shape, this species is similar to 'Lithoglyphus' tripaloi Brusina, 1884 (Bania tripaloi after Neubauer et al. 2012). This form has less convex whorls and no stepped spire, lacks the apertural thickening, the opercular ridge and the basal emargination. Some specimens from units A and D, erroneously determined as Bania prototypica (Brusina, 1874) by Mandic et al. (2011: fig. 5.14), fully correspond to the present description of B. valvatoides. B. prototypica, as well as 'Amnicola' torbariana (Brusina, 1874) and 'Amnicola' stosiciana (Brusina, 1874), differ from B. valvatoides in the higher spire and the more distinctly stepped whorl outline with highly incised sutures. All forms share the characteristic protoconch sculpture and are therefore assigned to the genus Bania (Neubauer et al. 2012). The subspecies B. stosiciana crassa Brusina, 1897, originally described from Gacko, fully corresponds to the slender specimens studied here and is thus considered synonymous. Krstić et al. (2009) identified specimens from drill cores as "Pseudoamnicola (Staja) šoštarićiana Brusina, 1874". With regard to the denoted authority, this attribution obviously is a wrong spelling of Pseudamnicola [= Bania] stosiciana (Brusina, 1874). The specimens fully match with the present description. Detailed discussions of this genus and attended nomenclatorial problems as well as its phylogenetic relations are provided by Kadolsky (1993, 1998) and Neubauer et al. (2012). Occurrence. Gacko Basin (Vrbica, Gračanica). Sinj Basin (Ruduša in Sinj), Drniš Basin (Miočić) (Brusina 1884). Clade Panpulmonata Jörger et al., 2010 Order Hygrophila Férussac, 1822 Suborder Branchiopulmonata Morton, 1955 Family Lymnaeidae Rafinesque, 1815 Galba Schrank, 1803 Type species. Buccinum truncatulum Müller, 1774, Recent, Europe. Diagnosis. Small, slender, elongated ovoid shell. Protoconch conical and smooth. Whorls well rounded to stepped, slowly increasing in dimensions; separated by deep sutures. Last whorl large, but not inflated. Aperture narrow, egg-shaped, with posterior angulation. Columella fairly straight and everted over the slit-like umbilicus (compiled from Zilch 1959–1960; Neubert 1998). Galba sp. Figure 9C Material. 3 spire fragments from sample 0907/Moll3.

7

Description. Extremely slender lymnaeid. Most complete specimen comprising 4 whorls, exposing highly convex whorl outline producing stepped spire. Protoconch smooth, passing into teleoconch without distinct rim. Aperture unknown. Strong growth lines covering shell. Remarks. The distinctly stepped whorl outline and the slender shape suggest a classification as Galba. The slim habitus reminds of Omphiscola Rafinesque, 1819, which has convex but not stepped whorls. No comparable taxon is recorded from the DLS, but the insufficient material and preservation, do not allow establishing a new species. Radix Montfort, 1810 Type species. Helix auricularia Linnaeus, 1758, Recent, Europe. Diagnosis. Globular ovoid, thin-walled shell with short spire and acute apex. Whorls increase fast in dimensions. Last whorl large, bulgy, and inflated. Aperture round, wide, often slightly everted, and sharply terminated. Columella slightly twisted, with broad margin (compiled from Montfort 1810; Zilch 1959–1960). Radix korlevici (Brusina, 1884) Figures 9A–B, D 1884 Limnaea [sic] Korlevići Brusina, p. 56. 1897 Limnaea [sic] Korlevići. – Brusina, p. 3, pl. 2, figs 6–7. 1902 Lymnaea hyaloleuca Brusina, pl. 1, figs 36–39. 2009 Lymnaea (Omphiscola) aff. servica. – Krstić et al., pl. 1, fig. 3. 2009 Succinea sp. ex gr. S. drnišana juv. – Krstić et al., pl. 1, fig. 4. 2011 Radix hyaloleuca. – Mandic et al., fig. 5.5. Holotype. NHMZ 2945-591. Type locality. Miočić, Drniš Basin, SE Croatia; Middle Miocene. Material. 76 largely fragmented specimens from samples 0804/039, 0804/039 coal, 0804/039b, 0907/Moll1, 0907/Moll2, 0907/Moll4, 0907/Moll4A, 090713/1, 090713/2, 0703/024 and Gacko 3; some remains < 1mm are found in samples Gacko 1 and Gacko 2. Description. Glossy, smooth, thin-walled and slender shell with blunt apex and 4 whorls. Precise measurements cannot be performed as the largest specimens were fragmented; estimated height: 15–20 mm. Protoconch smooth with indistinct transition to teleoconch. Whorls moderately convex. Last whorl (including aperture) attaining c. 85% of total shell height. Aperture ovoid, slightly everted and posteriorly elongated. Inner lip thin and attached to base of preceding whorl, leaving no umbilicus; columella rarely with weak fold. Weak prosocline growth lines covering surface. Remarks. The generic affiliation is based on the wide and elongated aperture. ‘Lymnaea’ hyaloleuca Brusina, 1902 is represented by two syntypes, which slightly deviate in the relative size of the last whorl. While the first one is slightly broader and exposes an angle next to the umbilicus, the second one (Brusina 1902: pl. 1, figs 38–39)

8

fully corresponds to the holotype of Radix korlevici. In respect to intraspecific variability both forms are regarded as synonyms of Radix korlevici (Brusina, 1884). Lymnaea klaici Brusina, 1884 from Miočić is much larger and more slender. Krstić et al. (2009) misidentified this species and even attributed some fragmented apices to the terrestrial gastropod Succinea. Occurrence. Gacko Basin (Vrbica, Gračanica). Drniš Basin (Miočić) (Brusina 1884, 1902). Family Planorbidae Rafinesque, 1815 Subfamily Planorbinae Rafinesque, 1815 Gyraulus Charpentier, 1837 Type species. Planorbis albus Müller, 1774, Recent, Europe; by subsequent designation (Dall 1870). Diagnosis. Small discoidal planorbid taxon. Protoconch with distinct spiral striae. Umbilical side almost flat or immersed. Few rounded, angular or carinate whorls, increasing rapidly in diameter. Aperture wide with expanded basal margin (modified after Zilch 1959–1960; Brown 1994). Gyraulus pulici (Brusina, 1897) Figures 10A–C, E–F 1897 Planorbis Pulići Brusina, p. 6, pl. 2, figs 8–10. 1902 Planorbis Pulići. – Brusina, pl. 3, figs 22–24. 2009 Segmentina sp. – Krstić et al., pl. 1, fig. 6. 2009 Planorbis pulići. – Krstić et al., pl. 1, figs 7–8. 2011 Gyraulus pulici. – Mandic et al., figs 5.11, 5.15. Holotype. NHMZ 2946-592. Type locality. Gacko-Avtovac, Gacko Basin, SE Bosnia and Herzegovina; Middle Miocene. Material. 254 largely fragmented shells from samples 0804/039, 0804/039 coal, 0804/039b, 0907/Moll1, 0907/Moll2, 0907/Moll3, 0907/Moll4, 0907/Moll4A, 090713/1, 090713/2, 0708/GA1, 0708/GA14, Gacko 1 and Gacko 3; some fragments < 1 mm are found in Gacko 2. Description. Broad planorbid shell with up to 3 whorls. Protoconch consisting of c. 0.7 whorls bearing 7 spiral striae that fade out after about half a whorl. Transition to teleoconch marked by initiation of growth lines, which are quite prominent in some specimens. Whorls overgrowing up to 50% of preceding ones and attaining up to more than double height. Last whorl making up c. 40% of shell diameter. No point of maximum convexity in early ontogeny; in later stages sometimes stronger convexity emerging in lower half resulting in roughly trapezoid outline. Peristome sharply edged and with c. 30° inclined to the axis.

9

Remarks. Gyraulus geminus (Brusina, 1897) and G. dalmaticus (Brusina, 1884) from Sinj and Drniš basins both are flatter with a more distinct angulation in lateral view. The smaller Gyraulus nedici (Brusina, 1902) from Dugo Selo (Glina subdepression, southern Pannonian Basin) has more convex whorls with a fairly equilateral profile. The likewise small Gyraulus oncostomus (Brusina, 1902) from Dugo Selo has a more distinctly angled profile with a deltoid shaped aperture and is relatively broader. Some fragments of earliest whorls were erroneously identified as Segmentina sp. by Krstić et al. (2009). Occurrence. Endemic to the Gacko Basin (Vrbica, Gračanica) (Brusina 1897). Orygoceras Brusina, 1882 Type species. Orygoceras dentaliforme Brusina, 1882, Miocene, Dalmatia/SE Croatia. Diagnosis. Shell small, dentaliform, asymmetrical and more or less curved. Protoconch planispiral bearing faint spiral striae. Teleoconch surface with faint growth lines, occasionally with ring-like sculpture or spiral lines. Aperture oblique and elliptical to semilunar (modified after Brusina 1882, 1902). Orygoceras dentaliforme Brusina, 1882 Figure 10D 1882 Orygoceras dentaliforme Brusina, p. 42, pl. 11, figs 9–15. 2011 Orygoceras dentaliforme. – Mandic et al., fig. 5.17. 2013 Orygoceras dentaliforme. – Neubauer et al., p. 146, figs 6D–E [cum syn.]. Syntypes. NHMZ 3574-1214/1a, NHMZ 3576-1216/1a. Type locality. Ribarić N Maljkovo, Sinj Basin, SE Croatia; Middle Miocene. Material. 4 fragments from samples 0804/039, 0804/039 coal and 0804/039b. Description. Shell uncoiled and dentaliform. Protoconch consisting of about half a whorl bearing faint striae. Teleoconch mainly smooth apart from fine growth lines. In some specimens distinct rings may occur in various ontogenetic stages. Shell elliptical to semilunar in cross-section. Aperture not preserved in any specimen. Remarks. The development of sculpture is assumed to range within intraspecific variability as no clear separation of distinct morphotype-groups can be performed. Discussion on the validity of the many species introduced by Brusina (1882, 1902) is provided by Neubauer et al. (2012). Occurrence. Gacko Basin (Vrbica, Gračanica). Sinj Basin (Ribarić N Maljkovo, Lučane W Sinj, Župića potok in Sinj, Trnovača), Drniš Basin (Miočić, Parčić), Kupres Basin (Fatelj), Džepi Basin, southern Pannonian Basin (Dugo Selo / Lasinja) (Brusina 1882; Neubauer et al. 2012).

10

Planorbarius Duméril, 1806 Type species. Helix cornea Linnaeus, 1758, Recent, Europe; by subsequent designation (Froriep 1806). Diagnosis. Shell moderately to very large, thick-shelled, disciform. Shell on both sides immersed, more on apical side. Whorls high, evenly rounded, increasing regularly in diameter, covered with growth lines and spiral striae. Aperture round or elliptical, basal margin protruding (compiled from Zilch 1959–1960; Brown 1994). Planorbarius sp. Figures 11A–F 2011 Planorbarius sp. – Mandic et al., fig. 5.12. Material. 5 spire fragments from samples 0804/039b, 0907/Moll4 and 0703/024; additional fragments < 1 mm found in 0804/039 coal. Description. Spire completely immersed with highly convex initial whorls and strongly incised sutures. Protoconch consisting of c. 1 whorl and measuring 550 µm. Initial cap making up c. 50 µm; covered with short and faint irregular threads. Termination indicated by onset of spiral rows of distinct circular pits (c. 5 µm). Between them fragile irregular ridges may be formed in axial direction. Onset of teleoconch indicated by formation of growth lines. Whorls exposing distinct shoulder to moderately curved flank. Strong irregularly shaped, spiral grooves covering whole teleoconch surface. They are confined to bundles of 4–5 grooves each, separated by slightly raised smooth bands (Fig. 11F). Where they cross more pronounced growth lines a weakly undulated pattern is formed. Aperture only marginally preserved; seemingly not conspicuously thickened or everted. Remarks. The identification of the fragments is based on the characteristic protoconch features and the spiral teleoconch sculpture, which are typical elements for Planorbarius. A determination on species level is not possible from the available material. Subfamily

the middle miocene freshwater mollusk fauna of lake gacko ......3 figure 1. geography and geological...

Documents