International Journal of Speleology

(ISSN 392-6672)

VOLUME 25 (1-2), 1996 Blospeleoloov

CONTENTS

DAVID PAUL SLANEY and PHILIP WEINSTEIN: Geographical variation in the tropical cave cockroach Paratemnopteryx stonei Roth (B1attel- lidae) in North Queensland, Australia .

CAMILLA BERNARDINI, CLAUDIO DI RUSSO, MAURO RAMPINI, DONATELLA CESARONI and VALERIO SBORDONI: A recent colonization of Dolichopoda cave crickets in the Poscola cave (Ort- hoptera, Rhaphidophoridae) 15

AUGUSTO VIGNA TAGLIANTI: A new genus and species oftrogIobitic Trechinae (Coleoptera. Carabidac) from southern China 33

VEZIO COTTARELLI and MARIA CRISTINA BRUNO: First record of Parastenocarfdidac (Crustacea, Copepoda, Harpacticoida) from subterranean freshwater of insular Greece and description of two new species , ' ," " 43

MARZIO ZAPPAROLI: Lithobills nilragicils n. sp., a new Lithobills from a Sardinian cave (Chilopoda, Lithobiomorpha) 59

Published quaterly by Societa Speleologica Italiana

Printed with the financial support of: Ministcro dei Bcni Culturali c Ambicntali Consiglio Nazionale delle Riccrche Musco di Speleologia «V. Rivera), L' Aquila INTERNATIONAL JOURNAL OF SPELEOLOGY Official journal of the International Union of Spcleology Achnowledged by UNESCO as a Category B Non-Governmental Organisation

U.I.S. REPRESENTATIVE: I. Paolo Forti, Dip. Scienze della Terra illS Universita di Bologna Via Zamboni 67, 1.40127 BOLOGNA, Haly Tel.: +39.51.35 45 47, Fax: +39.51.35 45 22 e-mail:[email protected]

B10SPELEOLOGY PHYSICAL SPELEOLOGY EDITOR: EDITOR: Valerio Sbordoni, Dip. Di Biologia, Ezio Burri, Dip. Di Scienze Ambicntali, Universita di Roma «Tor Vergara» Universita dell' Aquila, Via della Ricerca Scicntifica 1-67100L'AQUILA, Italy 1-00133 ROMA Haly Tel.: +39.862.43 32 22 Tel.: +39.6.72 59 51, Fax: +39.6.202 6189 Fax: +39.862.43 32 05 e-mail:[email protected] c- mail: [email protected]

EDITORIAL STAFF: EDITORIAL STAFF: Giammaria Carchini, Dip. Oi Biologia, Arrigo A. Cigna, Frazione Tuffo, Universita di Roma «Tor Vergara» 1-14023 COCCONATO (AsH), Italy Via della Riccrca Scicntifica Tel. & Fax: +39.141.907 265 1-00133 ROMA,Haly e-mail: [email protected] Tel.: +39.6.72 59 59 60, Fax: +39.6.202 61 89 Ugo Sauro, Dip. di Geografia c- mail:[email protected] Universita di Padova Marco Lucarelli, Dip. di Biologia Via del Santo 26, 1.35123 PADOVA, Italy Universila di Roma «Tor Vergata» Tel.: +39.49.827 40 93 Via della Ricerca Scientifica Fax: +39.49.827 40 99 1.00133 ROMA, Haly e-mail: [email protected] Tel.: +39.6.72 59 59 66, Fax: +39.6.20261 89 Marina Cobolli, Dip. di Scienze Ambientali ADVISORY BOARD: Universita dell' Aquila R. Bernasconi (Svwitzerland) 1-67100 L'AQUILA, Italy P. Bosak (Czech Rep.) Tel.: +39.862.43 32 08, Fax: +39.862.43 32 05 J. Choppy (France) e-mail :coboll [email protected] S. A. Craven (S. Africa) F. Cucchi (Italy) ADVISORY BOARD: A. Eraso (Spain) V. Aellen (Switzerland) C. Juberthie (France) C. A. Hill (USA) I. Andrassy (Hungary) R. W. Mitchell (USA) A. Klimchouk (Russia) R. Argano (Italy) C. N. Nath (India) A. Nunez Jimenez (Cuba) Th. C. Barr (USA) S. B. Peck (Canada) K. Pfeffer (Gennany E. Bellard Pietri (Venezuela) N. Peters (Gennany) M. Pulina (Poland) L. Botosaneanu (Nctherlands) A. Petrochilos (Greece) Y. Quinif (Belgium) J. Buresh (Bulgaria) Th. L. Poulson (USA) T. R. Shaw (UK) V. Caumartin (France) S. Rullo (Italy) I. D. Sasowsky (USA) O. Escola (Spain) B. Sket (Slovenia) H. Trimmel (Austria) E. L. Friedman (USA) G. Thines (Belgium) S. Zhang (China) J. R. Holsinger (USA) A.Vigna Taglianti (Italy) Y. Zhu (China H. Jakobi (Brazil) N. Zalesskaja (Russia)

- Manuscripts and editorial correspondence should be addressed to the respective Editors. - Correspondence concerning infonnations, subscriptions, payments and exchanges should be sent to: Marina Coholli, Dip. di Scienzc Ambientali, Universita de L' Aquila, 1-67100 L' AQUll.A, Italy - Annual subscription rates: Italy 20,000 ut, Other countries, individuals, 30.000 Ut, Institutions, 60,000 UI. Direttore Rcsponsabile: A. Lucrezi, Autori1.l. Trib. L' Aquila n. 334. Printed in Italy by TIPIGRAF s.n.c. Tivoli (Roma) Italy, June 1997 lnt. J. Speleo!., 25,1-2 (1996): 1-14. 1997

Geographical variation in the tropical cave cockroach Paratelllllopteryx stollei Roth (B1aUellidae) in North Queensland, Australia.

David Paul Slaney and Philip Weinstein *

SUMMARY Observations of cave dwelling organisms in both tropical and temperate caves often re- veal morphological modifications. which may reflect various stages of adaptation to cave life. From April 1994 to June 1995 a number of adult Paratemnopteryx stolld were collected from 7 caves in tropical North Queensland to investigate the degree of geographical variation in such troglomorphics betwen cave populations. Results of morphometric analyses showed the occurence of a morphological discontinuity between cave populations from the different geographic regions. The body dimensions particularly important in discriminating between each cave population were tegmen length (both sexes), and secondly, tegmen width and tarsus Icngth for males and females respectively. Morphological differenccs betwecn populations are discussed in relation to stages of adaptation to eave live.

INTRODUCTION Cave organisms often demostrate degrees of morphological modification (troglomorphy), for example, reduced or lost eyes, wings and bodily pigmentation, and attenualion of appendages (Christiansen, 1961; Barr, 1968; Culver, 1982; Kane and Richardson, 1985). These troglomorphic characters may reflect various stages of adaplalion to cave habilats, induced through different regimes of selective pressures, and or genetic factors such as lhe accumulation of neutral mUlalions (Poulson, 1985; Wilkens, 1992; Culver et aI., 1994). The highest morphological modification occurs in those populations which have been isolaled (ie subject to no gene flow) the longest, have experienced stronger selective pressures, or have a greater fixation rate of genetic mutations within a population. Thus, one would expect isolated cave populations to demostrate varying degrees of troglomorphy. Important factors which detennine Ihe degree of isolalion or otherwise of cave populations are the presence of geological barriers andlor interslitial

* Dept of Zoology, James Cook University, Townsville, Queensland 4811, Australia. 2 DAVID PAUL SLANEY and PHILIP WEINSTEIN

voids whithin cavernous strata (Barr, 1968; Howarth, 1982; Barr and Holsinger, 1985; Kane and Brunner, 1986). The occurence of such barriers and voids influences the dispersal ability of cave dwelling organisms, and in turn determines the geographic extent and degree of gene flow between cave populations. For example, Neaphaenops tel/kampfi, a troglobitic (obligately cave dwelling) trechine carabid beetle, in west central Kentucky contains isolated populations which have been allocated subspecies status by Barr (1979). He recognised four subspecies, based on morphological differentiation of the populations which coincides with the presence of faults and rivers between groups of morphological forms. In contrast, cave populations that are subject to intercave gene flow would have a lower degree of morphological variation. In a study of the distribution and evolution of Ptomaphaglls beetles in the southeastern United States, Peck (1984) compared 61 separate cave populations. The caves were located in a continuous limestone belt that would be expected to contain abundant subterranean avenues for dispersal. Morphological comparisons between these cave populations showed little variation, although each population was found to contain only one form of spermatheca. Peck suggested that the spermathecal forms are an indicator of species level taxonomic categories, and speciation has occured despite the presence of subterranean dispersal routes. In North Queensland tropical caves, comparison of two populations of Undarana, a cave dwelling cixiid (Hemiptera), from lava tubes 30 km apart, revealed significant differences in external and genital morphology (Hoch and Howarth, 1989). Although the populations were from distant lava tubes of separate origin, individuals may be able to disperse between caves because the lava fields are connected by other basaltic flows. Hoch and Howarth interpreted these differences as indicators of interrupted gene flow between the populations, and therefore surmised the existence of two separate biological entities. In contrast, they found populations of another cixiid, Solonaima, to be morphologically similar in the same lava systems. Early examination of cockroaches we have collected from these lava tubes and other tropical caves in North Queensland indicated that there are morpholo-gical differences between populations of Paratemnopteryx stolle; Roth (Blatto-dea: Blattellidae). This genus is distributed widely over Aus- tralia, with different species living in epigean and cave habitats (Roth, 1990). So far, three of the four cave species whithin the genus have been found only in North Queensland (P. stOlle;, P. howarthi, and P. sp4). The aim of this study was to investigate the degree of geographic variation in troglomorphies between populations of P. stolle; in North Queensland caves. GEOGRAPHICAL VARIA nON IN THE TROPICAL CAVE COCKROACH 3

MATERIALS AND METHODS

From April 1994 to June 1995 P. stone; adults were collected from 7 caves in North Queensland: Bauhinia (8 males, 5 females) and Clam (\6 males, 8 females) caves at Chillagoe, Barkers (\6 males, 25 females) and Bayliss (21 males, 27 females) lava tubes at Undara, Frig cave (3 males, 4 females) at Broken River, and Rope Ladder (\3 males, 12 females) and Bat cave (3 males, 3 females) at Fanning River (Fig. I). As cockroaches do not continue to grow once they are adults morphometric analyses were carried out only on adults, avoiding sample bias in the growth stage. In addition, both sexes were analysed separately to investigate sexual dimorphism. From these collections morphological mesaurements were made on II body dimensions using a stereo-dissecting microscope and ocular mi- crometer. The body dimensions were body length; femur, tibia, and tarsus length of hindleg; cercus length; eye length and width; tegmen length and width; and pronotum length and width.

16'

144'

Soulh Pacific 150' •Chillagoe (Kiw; & Clam Caves) Ocean

• Undua (Barkers & Bayliu Lava tubes)

Broke~River (Frig Cave)

Fannin •River 20' 01 ope I er aves

Fig. 1 - Map of North Queensland indicating the fOUf cave regions and cave study sites. 4 DA YID PAUL SLANEY and PHILIP WEINSTEIN

25 2J r 231- • I 22 1 1 E 21 •r ! .s 1 ~ 20 Cl T Z • w "- ...J 19 ! >- 0 I f T 0 18 ! lD ! ! 17 f I 15'Or • 14 co E V> Cl ~ .!!1 :;; iii ~ co "'0 lD :E >. ;r "'0 :::> ti "'"" co co co al ...J al lD•• c- o" a:

Fig. 2 - Variation in body length (mm) ::!: SE for males (circles) and females (squares) from the seven cave study sites.

The statistical analysis used to investigate geographic variatIOn between populations was Canonical Discriminant Analysis (CDA), using the SAS program package (SAS Institute, 1987). CDA is an ordination tech- nique for displaying and describing the differences between group centroids, ie for each population, by extracting the eigen vectors from the pooled within-group variance-covariance matrix. The analysis was carried out on ratio transformed data, with respect to body length, to remove the in- fluence of body size on other body characters. GEOGRAPHICAL VARIA nON IN THE TROPICAL CAVE COCKROACH 5

11 f T• T 1 10 r t ! • T • .L• .sE 9t i=- C) z w ~ -J T z • w .L :2 :f ! • C) T w ! ... • T 6l •

5' !

4 E ~ ~ Cl 1i; n; .c'" 1i; .!!! "0 tIl >. ~ "0 :E U'" "'" ::l ~ -J tIl tIl" '" tIl '" c- o" a:

Fig. 3 - Variation in tegmen length (mm) ;t SE for males (circles) and fennales (squares) from the seven cave study sites.

One-way ANOV A's were carried out on body length and the morphological characters defined by the CDA's as being particularly important in discriminating between cave populations. Based on these characters a Tukey Studentized Range (HSD) Test was carried out to define which populations were significantly different from each other (Sakal and Rohlf, 1982). 6 DAVID PAUL SLANEY and PHILIP WEINSTEIN

RESULTS

Initial exploratory data analysis of the raw data using plots of the meam of each of the measured body dimensions indicated that there were 2 main differences between males and females both between and within cave populations. These were body length and tegmen length (Figs. 2 and 3). Figure 2 shows that females had greater body length than males at each of the 7 caves, with populations from Chillagoe and Undara having greater body length than those from Broken River and Fanning River. Figure 3 indicates that females have shorter tegmina than males in populations from Chillagoe, Undara and Broken River.

.65

.60

J: •.... .55 _ eJ z w ...J ! ! >- .50 0 0 lD J: .45 T •.... ~• eJ •• z • w I ...J .40 wZ ::; • I 0 .35 • .•. •....w I • .30 ! .•.I .25 ., j;j •• E 'l: .!! .," J!1 lL'" 'tl-•• lD ~ .><- >- 'tl u 1;; •• " lD lD ..J•• •• II '" C. a:0 Fig. 4 - Radio transformed tegmen length data:t SE for males (circles) and females (squares) from the seven cave study siles. GEOGRAPHICAL VARIAnON IN THE TROPICAL CAVE COCKROACH 7

Ratio transfonning the tegmen length data produced different results to those shown in figure 3 (Fig. 4). In figure 4 females have shorter tegmina than males in populations from Chillagoe, Undara and Broken River, which is opposite to the relationship shown in figure 3. This reversal is due to the effect of body size (females are larger than males) supporting the decision to carry out further statistical analyses on ratio transfonned data. Figure 4 indicates that males had longer tegmina than females, and that sexual dimorphism in tegmen length is most pronounced in the cockroach populations from Fanning River caves. Based on tegmen length, the graph shows 3 morphological groups, these being a long wing morph at Chillagoe (males and females), a short wing morph at Undara and Broken River (males and females), and a long/short wing morph at Fanning River (males and females respectively).

8 '"•.• ~ 6

N w 4 '"-' q" ~ 2 . ~j},: ;; •• t' , , , 0 •.!c1~. . is 4 .• :J 3 Z .() 0 .2 ~ 3Q33 0 ~. 3 3 B • -4 0 • -6

-8

.10 't' I" ""~~~",""""""~II,,,I, ,I ,el .10 .8 .6 -4 .2 0 2 4 6 8 10 12. 14 CANONICAl VARIABLE 1 87.3%

Fig. 5 _ Canonical discriminate analysis plots of males with 95% confidence intervals around centroids for Bauhinia (I), Clam (2), Barkers (3), Bayliss (4), Frig (5), Rope Ladder (6), and Bat (7) caves. 8 DAVID PAUL SLANEY and PHILIP WEINSTEIN

6 , ~ • 7 <0 4 ,

-6

-8

-10 -10 -8 .6 -4 -2 0 2 4 6 8 10 CANONICAL VARIABLE 1 67.2%

Fig. 6 - Canonical discriminate analysis plots of females with 95% confidence intervals around centroids for Bauhinia (I), Clam (2), Barkers (3), I3ayliss (4), Frig (5), Rope Ladder (6), and Bat (7) caves.

Plots of the COA's for males and females (Figs. 5 and 6 respectively) show a clear separation of the populations with respect to geographic location, and to a lesser degree between caves. For males the canonical vari- ables I and 2 summarised 97.7% (87.3% and 10.4% respectively) of the total variation between groups, and 93.8% (67.2% and 26.6% respectively) for the females. Table I gives the coefficients for the first canonical variable, and indicates that the body dimensions which are particularly important in discriminating between cave populations are tegmen length (for both sexes), and secondly, tegmen width and tarsus length for males and females respectively. The distinctiveness between populations was confirmed by multivariate F-statistic analysis (PilIai's Trace = 2.860, F = 6.647, P = 0.0001 for males; PilIai's Trace = 2.643, F= 6.376, P = 0.0001 for females). GEOGRAPHICAL VARIA nON IN THE TROPICAL CAVE COCKROACH 9

Table I _ Coefficients for the first canonical variable from the CDA analyses carried out on males and females, ranked in order of importance in discriminating between cave populations.

MALES FEMALES Tegmen Length 0.976 Tegmen Length 0.788 0.344 Tegmen Width 0.574 Tarsus Length -0.326 Eye Width 0.511 Eye Length 0.300 Pronotum Length 0.496 Tibia Length Tibia Length -0.454 Eye Width -0.163 Femur Length -0.434 Femur Length 0.104 -0.070 Tarsus Length 0.405 Tegmen Width -0.051 Eye Length 0.353 Pronotum Width Cercus Length -0.132 Pronolum Length 0.033 -0.010 Pronotum Width -0.101 Cercus Length

Plots of tegmen width and tarsus length (Figs. 7 and 8 respectively) indicate that male cockroaches from Bat Cave at Fanning River had the wid- est tegmina and female cockroaches from Bauhinia Cave at Chillagoe had the longest tarsi. Results from the one-way ANOV A's and Tukey Studen- tized Range (HSD) Tests on body, tegmen and tarsus length, and tegmen width indicated that these characters were significantly different from each other in some populations (Results of which can be obtained from the authors).

DISCUSSION

Previous morphological work on P. stollei has been carried out by Roth (1990). He described P. stolle; from caves at Chillagoe and Undara, but did not carry out any detailed morphometric analyses. He noted that specimens taken from different caves varied in size of tegmina, and reduction in pulvilli. However, comparison of male genitalia were similar enough that he considered the specimens to be races of the same taxnn. He classi- fied specimens found in Clam cave at Chillagoe, and those from Bayliss, and Barkers caves as 3 different races. Specimens have not been reported or described from either Broken River or Fanning River caves, and our specimens therefore constitute new locality records for P. stollei. Of further in- terest was the striking sexual dimorphism in tegmen length in individuals from Fanning River caves which has not been found in any other population of P. stollei. Sexual dimorphism in wing length is not unique to individuals of P. stollei, as it has been recorded in numerous nther insects (Thayer, 1992). 10 DAVID PAUL SLANEY and PHILIP WEINSTEIN

.44

I f- .43 ~ Z UJ -' >- 42 0 0 I CD I 41 f- 0 ~ j z .40 ::;UJ I ~ UJ 39 f- I 38 I

,37 E III III ~ '2•• Ul .~ •• v U. "0•• CD :c ti ~ ~ "0 •• "•• •• CD•• ..J•• '" '" c-•• O 0::

Fig. 7 - Ratio transfonned tegmen width data :t SE for males from the seven cave study sites.

The multivariate analyses indicate that there are morphological variations between cave populations which can be resolved into 4 major groups, representing the 4 geographic regions, Chillagoe, Vndara, Broken River, and Fanning River (Figs. 5 and 6). There is a degree of variation within each population, where some individuals from caves in the same geographic region, ie Bauhinia and Clam at Chillagoe, Barkers and Bayliss at Vndara and Rope Ladder and Bat at Fanning River, fall within each others' data cloud. However, the intracave region variation does not exceed the intercave region variation observed between the four cave regions. Observations of cave dwelling organisms in both tropical and temperate caves often demonstrate troglomorphy, although the pattern and degree of morphological modification is not universal (Christiansen, 1961; Barr, GEOGRAPHICAL VARIATION IN THE TROPICAL CAVE COCKROACH II

1968; Culver, 1982; Kane and Richardson, 1985; Peck, 1990). All individuals of P. stonei, except males from Fanning River, were found to have brachypterous tegmina with females having shorter tegmina than males (Fig. 4), and all individuals had vestigial hind wings. Reduction in tegmen width was also noted between males, with both intra and intercave region variation (Fig. 7). Peck and Roth (1992) have documented tegmina and wing reduction in surface dwelling Chorisoneura cockroaches from the Gallipagos Islands. They concluded that during wing diminution. the tegmina are less affected than the hind wings, and within a population the wing is more regressed in females. Wing loss has also been found in a number of other cockroaches, with flightlessness being more frequent among females than males, and there was a higher than average frequency in species inhabiting caves (Roff, 1990). The reduction in wing and tegmen size has been explained as a response to an increased specialisation for a more homogeneous environment, such as caves.

36 f .35

.34 ,...I .33 z I '"w >--' .32 0 0 CD 31 ,...I .30 z w'" 1 .29 Ul-' OJ a:Ul .28 f ,...>( .27 1 f .26

.25 l; ;; .!!! E e .~• 'C m c .!!! • u.'" ~ :E () ~ ~ ~ CD • CD -' CD•• o•c- a:

Fig. 8 _ Ratio transformed tarsus length data:!: SE for females from the seven cave study sites. 12 DAVID PAUL SLANEY and PHILIP WEINSTEIN

Elongation of appendages is a further troglomorphic feature found in cave species. Females from Bauhinia cave at Chillagoe had longer tarsi than females from the other caves (Fig. 8). Leg attenuation in cockroaches has also been reported in species of Nocticola from caves at Undara (Stone, 1988). Peck (1986) found a tendency for appendage elongation in cavemi- colous Ptomaphagus beetles, with P. cavernicola having especially long tarsi. Attenuation of legs and antennae may provide a selective advantage by increasing the searching ability of cave organisms in an environment where there is little to no ligth (Peck, 1973). Tarsal structure in cockroaches is a highly adaplive character, showing distinctive, non phylogenetic features above the species level (Roth and Willis, 1952; Arnold, 1974). Roth (1990) noted variation in pulvillus reduction in and between populations of P. stOllei from Chillagoe. Reduction of pulvilli and increased dependance on tarsal claws is an adaptation for walking on muddy cave surfaces (Roth and Willis, 1952; Christiansen, 1961 and 1965). Inter and intraspecific variation in tarsal structure in North Queensland cave ParatemlloptelYx is currently being investigated by us in populations from caves with different substrate surfaces and humidity levels. Troglomorphic characters may reflect various stages of adaptation to cave life. The degree of which may be int1uenced by various factors. Firstly, the selective regimes operating, as body size and allometrically related morphological characters are affected by environmental factors. Thus, variability in morphology between cave populations can be a result of ecological differences between the caves. Secondly, the length of time and degree to which organisms have been isolated can affect morphological variability. Increased variability in troglomorphic characters has been proposed as resulting from intercave gene t10w and hybridisation with extant surface species (A vise and Selander, 1972). As the cave regions in this study are over 150 km from each olher it is unlikely that there is any gene flow between cave regions, but one would expect intercave connections within regions. Thus, both direct environmental induction, reflecting phenotypic plasticity, and genetic adaptation to current selection pressures may be important in contributing to the observed morphological variation (Atchley, 1983; AlIe- grucci et aI., 1992; Bilton, 1993). Thirdly, non-selectionist origins for troglomorphy have been considered, with morphological variation occuring as a result of the accumulation and fixation of neutral mutations within a cave population (Poulson, 1985; Wilkens, 1992). This also would be int1lJ- enced by the degree of intercave gene flow. Considering these points, one may expect that populations of one species showiog increased troglomor- GEOGRAPHICAL VARIAnON IN THE TROPICAL CAVE COCKROACH 13 phy, such as the males from Bauhinia Cave at Chillagoe and the females from Barkers Cave at Undara, represent populations that have experienced the greatest isolation or experienced greater selective pressure, and or a greater fixation rate of genetic mutations within a population. Alternatively morphological variability may reflect phylogenetic differences between cave populations. However, it is important not to rely on morphological characters that may be adaptive for use in inferring phy- logeny. To resolve these issues we are currently making microenvironmen- tal comparisons between caves, and molecular, morphological and behav- ioural studies at the population and species level within Paratemllopteryx. Given the similarities of the cave environment in the caves studied, it is difficult to believe that the morphological variations reflect different degrees of troglomorphy resulting from different selection pressures at each site. The morphological differences exhibited by isolated cave populations are more likely to have resulted from the fixation of characters modified by genetic drifL

ACKNOWLEDGMENTS

We thank Chris and Ray Whitney for allowing access to Fanning River caves, and Dr. David Blair, Ann-Marie Boyd, Dinah Hansman, Dwayne Kersey and Erich Volschenk for assistance in the field. Thanks also to Dr. Fred Stone. Deborah Ward and Douglas Irvin who spent many an hour showing us new collecting sites. We also whish to thank Val Speedy and every onc else at Undara Lodge for their hospitality and help, Red Dome Gold Mine for their assistance, and Queensland Nation Parks and Wildlife service for providing permits for Chillagoe and Undara caves. We also thank Drs. Ross Alford and Richard Rowe who pro- vided valuable advise in the carly stages of our statistical analyses.

REFERENCES

ALLEGRUCCt. G., F. BALDARI. D. CESARONt, R.S. THORPE, and V. SBORDONI. 1992. Morphometric analysis of interspecific and mierogeographic variation of crayfish from a Mexican cave. BioI. J. Linn. Soc. 47: 455-468. ARNOLD, J.W. 1974. Adaptive features on the tarsi of cockroaches (Insecta: Dictyoptera). Int. J. Insect Morph. Embryo!. 3: 317-334. ATCHLEY, W.R. 1983. Some genetic aspects of morphometric variation, p. 346-363. In J. Felscnstcin (cd.) Numerical Taxonomy: Proceedings of a NATO Advanced Study Insti- tute Series G. Springer-Verlag, Berlin. AVISE, lC., and R.K. SELANDER. 1972. Evolutionary genetics of cave.dwelling fishes of the genus A.Hyanax. Evolution 26: 1-19. BARR, T.C., Jr. 1968. Cave ecology and the evolution of troglobites. Evol. BioI. 2: 35-102. BARR, T.C., Jr. 1979. The taxonomy, distribution, and affinities of Neaphaenops with notes on associated species of Pseudanop"thall1lllS (Coleoptera: Carabidae). Am. Mus. Novil. 2682. BARR, T.C., Jr. and J.R. HOLSINGER. 1985. Speciation in cave faunas. Ann. Rev. Eeo\. Sy'!. 16: 313-337. 14 DAVID PAUL SLANEY and PHILIP WEINSTEIN

BILTON, D.T. 1993. A morphometric study of the diving beetle Hydroporus glabriusculus (Coleoptera, Dytiscidae) in Western Europe, including a comparison of morphological and genetic divergence patterns. Zool Anz. 231: 111-124. CHRISTIANSEN, K. 1961. Convergence and parallelism in cave Entomobryinae. Evolution 15: 288-301. CHRISTIANSEN, K. 1965. Behavior and form in the evolution of cave Collembola. Evolu. tion 19: 529-537. CULVER, D.C. 1982. Cave Life: Evolution and Ecology. Harvard University Press, Cam- bridge. CULVER, D.C., R.W.JERNIGAN, and J. O'CONNELL. 1994. The geometry of nalural se. lection in cave and spring populations of the amphipod Gammarus minus Say (Crustacea: Amphipoda). BioI. J. Linn. Soc. 52: 49-67. HOCH, H., and F.O. HOWARTH. 1989. Reductive evolutionary trends in two new cavcrnicolous species of a new Australian cixiid genus (Homoptera: Fulgoroidea). Syst. Enlo- mol. 14: 179-196. HOWARTH, F.G. 1982. Bioclimatic and geologic factors governing the evolution and distribution of Hawaiian cave insects. Enlomol. Gen. 8: 17-26. KANE, T.C, and R.C RICHARDSON. 1985. Rcgressive evolution: An historical perspec. tive. NSS Bull. 47: 71-77. KANE, T.C, and G.D. BRUNNER. 1986.Geographic variation in the cave beetle. NeaphaenopJ tellkampji (Coleoptera: Carabidae). Psychc 93: 23 I-251. PECK, S.B. 1973. A systematic revision and Ihe evolutionary biology of the Ptomaphagus (AdelopJ) beetles of North America (Coleoptera; Leiodidae; Catopinae), with emphasis on cave.inhabiting species. Bull. Mus. Compo Zool. 145: 29-162. PECK, S.B. 1984. The distribution and evolution of cavcrnicolous Ptomaphagu.t beetlcs in the southeastern United States (Coleoptera; Leiodidae; Cholevinae) wilh new species and records. Can. 1. Zool. 62: 730-740. PECK, S.B. 1986. Evolution of adult morphology and life-history characters in cavernicolous Ptomaphagus beetles. Evolution 40: 1021.1030. PECK, S.B. 1990. Eyless arthropods of the Galapagos Islands, Ecuador: Composition and origin of the crytozoic fauna of a young, tropical, oceanic archipelago. Biotropica 22: 366-381. PECK. S.B., and L.M. ROTH. 1992. Cockroaches of the Gahipagos Islands, Ecuador, with descriptions of three new spccies (Insecta: Blattodea). Can. J. Zool. 70: 2202-2217. POULSON, T.L. 1985. Evolutionary reduction by neutral mutations: Plausibility arguments and data from amblyopsid fishes and Iinyphiid spiders. NSS Bull. 47: 109.117. ROFF, D.A. 1990. The evolution of flightlessness in insects. Ecol. Monogr. 60: 389-421. ROTH, L.M. 1990. A Revision of the Australian Parcoblattini (Blauaria: Blattcllidae: Blat. tellinae). Mem. Queensl. Mus. 28: 531-596. ROTH, L.M., and E.R. WILLIS. 1952. Tarsal structure and climbing ability of cockroaches. J. Exp. Zool. 119: 483-517. SAS INSTITUTE. 1987. SASISTAT Computer package version 6.0. SAS Institute Inc., Cary, NC. SOKAL, R.R. and FJ. ROHLF. 1982. Biometry: Thc principles and practice of statistics in biological research. W.H. Freeman and Co., San Francisco. STONE, F.D. 1988. The cockroachcs of North Queensland caves and the evolution of tropi. cal troglobites. Aust. Speleo. Fed. 17th Bi. Con. Prcprints, Lake Tinaroo. Dec. 1988. THAYER, M.K. 1992. Discovery of sexual wing dimorphism in Staphylinidae (Coleoptera): Omaliul1lj7avidum, and a discussion of wing dimorphism in insects. Journal of the New York Entomological Society 100: 540-573. WILKENS, H. 1992. Neutral mutations and evolutionary progress, p. 403-422. In A.I. Camacho (cd.), The Natural History of Biospcleology. Monogr. Mus. Nac. Ciencias Nat., e.S.I.e., Madrid. Int. J. Spelea!., 25,1-2 (1996): 15-31. 1997

A recent colonization of Dolichopoda cave crickets in the Poscola cave (Orthoptera, Rhaphidophoridae).

Camilla Bernardini *, Claudio Di Russo *, Mauro Rampini **, Donatella Cesaroni * and Valerio Sbordoni *

SUMMARY

We report a series of investigations carried out on a Dolichopoda population recently discovered in the Poscola cave and in some small caves nearby (Lessini Mountains. Vicenza). This population is located north of Po river, outside the present known geographic range of this genus in Italy. Morphology of the epiphallus corroborated by chromosome and allozyme analysis indicated that this population belongs to D. laetitiae. Study of the genetic structure of population in the Poscola area revealed high gene flow levels between Poscola and the other minor caves, suggesting the occurrence of a single expanding population. This finding as well as mark-recapture data on population size. migrations, age structure and habitat type strongly suggest that the Poscola population is the result of a recent colonization due to anthropocore dispersal.

INTRODUCTION

Colonization, i.e. the successful invasion of a new area (or habitat) by a species, is a process with important evolutionary consequences. Particularly, the study of colonization events of isolated habitat, such as caves or islands, because of their simplified ecology, represents a good opportunity to investigate any detectable evolutionary changes in animal and plant populations both from the ecological and the genetic point of view. However, for the cave habitat, only few experimental transplantations have been reported in the last forty years. These experiments include the transplantation of Amphipod Crustacean Niphargus virei in the Balme cave in French (Ginet, 1965), of the Cholevid beetles Speollomus IOllgicomis, S. diecki, S. stygius and Bathysciola derosasi introduced respectively in the

* Dipartimento di Biologia. Universita di Roma "Tor Vergata". Via della Ricerca Scientifica, 00133 Roma-Italy. ** Dipartimenta di Biologia Animale e dell'Uomo, Universita di Rama "La Sapienza". Viale dell'Universita 32, 00185 Rama-Italy. 16 C. BERNARDINI. C. DI RUSSO. M. RAMP INI. D. CESARONI and V. SBORDONI

Ramioul cave in Belgium (Bouillon & Hubart. 1982; Tercafs & Brouwir. 1991) and in the Patrizi cave in Italy (Patrizi, 1956) and of Dolichopoda /inderi populations in two caves of southern France (Oi Russo, 1993). Do/ichopoda cave crickets were recorded in the Poscola cave for the first time in 1991 by one of us (M. Rampini). Previous records of cave crickets from this cave included Trog/ophilus cavico/a and T. neg/ectus only (Bartolomei, 1957). The location of this cave, which is largely outside of the presently known range of Dolichopoda (Fig. I), and the low vagility of these crickets, usually limited to hypogean habitats, led us to hypothesize that the population presently inhabiting Poscola cave is the result of a recent colonization. In order to test this hypothesis and to detail the biology of this population, genetic and ecological studies were carried out throughout periodical sampling.

Poscola cave

~ • Dligustica () o

Fig. 1 - Geographic location of Pascola cave and distribution range of D.ligustica and D.llleritiae. A RECENT COLONIZATION OF DOLICHOPODACAVE CRICKETS 17

MATERIAL AND METHODS

Geographical location and cave description.

Poscola cave (cave register number: 136 VNI) belongs 10 a Karst system located at 275 m asl on the eastern slope of the Faedo-Casaron mountain near Priabona village (Lessini Mountains, Vicenza). The cave develops horizontally for 325 m, showing two entrances. The main entrance opens near the garden of the Priabona church, while the smaller. one is surrounded by a thick wood. Furthermore a stream springs near the secondary entrance and flows through the cave (Fig. 2).

-25m

.. ,'.

',' .

Fig. 2 - Map of the Pascola cave: A, the main entrance; B, secondary entrance; 1,2,3. and 4, the different sectors for samplings. 18 C. BERNARDINI, C. DI RUSSO, M. RAMPINI, D. CESARONI and V. SBOROONl

The chek list of the fauna of this cave, compiled by Fabiani (1904) and later completed by Bartolomei (1957), is here reported: Gastropoda (Oxy- chi/us sp.); Copepoda (Speocyclops sp.); Isopoda (Androniscus dentiger, Armadillidium sp., Monolistra berica); Amphipoda (Niphargus stigius costozzae, Gammarus balcanicus); Oecapoda (Astacus jluviatilis); Chilopo- da (Lithobius sp.); Pseudoscorpionida (Chthonius sp.); Opiliones (Gyas sp.); Orthoptera (Troglophi/us cavicola, T. neglectus); Coleoptera (Neoba- thyscia fabianii, Antisphodrus sp.); Lepidoptera (Scoliopteryx libatrix); Oiptera (Culex sp.); Chiroptera and Rodentia. Following our first record of Dolichopoda in 1991, several other small caves near the main Poscola cave were periodically checked for Doli- chopoda occurrence. Out of 12 only 6 caves were inhabited by Dolicho- poda; Poscola I (PSC I), Poscola II (PSC 2), Poscola III (PSC 3), Poscola IV (PSC 4, nOl528 VNi), Marchiori cave (n° 579 VNi), Buco sopra la cava (n02254 VNi) (Fig. 3).

Epiphallus morphology

Morphology of the epiphallus is the taxonomic character usually utilized to identify species in Dolichopoda (Baccetti & Capra, 1959, 1970). Samples of D. laeti/iae and D. ligustica were therefore examined for comparative analysis. These two species were considered because of their geographic distribution and their overall similarity. Study samples were from Bunker near Muratone pass (LIG), Pigna- Liguria (D. ligustica) and Zigolo cave (ZIG), Castel dei Britti-Emilia; Drago cave (ORA), Apecchio-Marche; Bella cave (FSS I), Genga-Marche; Tana Tennini cave (TER), San Marcello Pistoiese-Toscana; Nebbia cave (FUN), Fungaia-Toscana; Etruscan Mitreo (SUT), Sutri-Lazio (D. laetitiae). After dissection the epiphallus was treated with KOH solution (0.5N) and dehydrated in graded ethanol series. These structures, mounted on stubs, were coated with a thin layer of gold in a sputtering unit (MEOOIO sputter coater, Balzer Union). Observations were carried out with a Stereo Scan Cambridge 200 scanning electron microscope operating at 15 kV. Micrographs were taken on 120 Agfapan 25.

Chromosome analysis

Mitotic chromosomes were obtained from male nymph crickets. After dissection in a proper solution (NaCI 0.7%, CaCI2 0.02%), testes were iso- A RECENT COLONIZATION OF DOLICHOPODA CAVE CRICKETS 19

N 1

1 km

Fig. 3 - Detailed map of Pascola area. The main cave and the six small visited caves in the neighbourhood are indicated.

lated, fixed in methanol-acetic acid 3:I for 30 sec., then incubated in a drop of acetic acid for I min. Tissues were squashed under a siliconized coverslip which was removed with a razor blade after immersing the cytological preparation in liquid nitrogen. Chromosomes were stained with. nucleophilic Giemsa solution (5% in deionized water for 20 min,) and 20 C. BERNARDINI, C. DI RUSSO, M. RAMPINI, D. CESARONI and V. SBORDONI observed wiIh a Leitz Laborlux 12 microscope equipped with a IDO/SOW mercury lamp.

Allozyme analysis

The degree of geneIic similarity between Poscola cave population and other populations of D.laetitiae was evaluated by means of allozyme poly- morphisms. Samples of two D.laetitiae populations, from Pitigliano cave (PIT) in Tuscany and S. Giacomo cave (GIA) in Umbria, were collected and analysed for comparative purpose. On a narrower geographic scale, samples from some small caves near Ihe Poscola cave (PSCI, PSC3 and PSC4) were also considered to investigate population genetic structure (Fig. 3). All samples were assayed electrophoretically for genetic variation at 19 gene loci: acidphosphatase (Acp), aldolase (Aida), diaphorase (Dia), este- rase (Est-I, Est-2), glutamate-oxaloacetatetransaminase (Got), glucose- phosphateisomerase (Gpi), isocitratedehydrogenase (ldh-I, Jdh-2), leucino- aminopeptidase (Lap-I, Lap-2), malatedehydrogenase (Mdh-l, Mdh-2), mannose-phosphateisomerase (Mpi), peptidase (Pep), phosphoglucomutase (Pgm), pyruvate-kinase (Pk), proteins (Pt). The methodology essentially followed technical procedures reported in Allegrucci et al. (1992) with minor modifications. Allele frequencies, observed and expected heterozygosities, and genetic distances were calculated using the BIOSYS-I program of Swofford and Selander (1981). Genetic distance indices were calculated using the Nei's (1978) algorithm. F-statistics (Wright, 1951) was used to investigate the genetic structuring of Dolichopoda samples from the Poscola area, by using the procedure of Weir and Cockerham (1984) and the FSTAT program (verI.2) developed by Jerome Goudet, University of Losanna, Switzerland. The extent of gene flow was evaluated by means of Nm values, i.e. the average number of migrants exchanged per generation, estimated from the relationship: FST = 1/ (4Nm+I).

Estimation of population size by seasonal samples

Poscola cave was divided in four sectors (Fig. 2) to relate the spatial distribution of Dolichopoda to thermal condition of the cave and trophic resources availability. Air temperature of the four sectors was recorded seasonally by a digital thermometer (:1:0.1). A RECENT COLONIZAnON OF DOLICHOPODA CAVB CRICKETS 21

Population size was estimated by mark-recapture method (Lincoln, 1930) modified for small population by Bailey (1951). This method was previously tested in several Dolichopoda populations (Carchini et aI., 1983; Rampini et aI., 1983). According to Bailey (1951), population size N was calculated by [N = M(T+l)/(R+I)], where M is the number of marked individuals, T is the total number of recaptured individuals and R is the total number of recaptured individuals previously marked. Variance of the estimates was calculated as follow: [V(N) = M\T+I)(T-R)/(R+I)2 (R+2)] and confidence limits (C.f.) are given by the relation [C.f. = :l: t *" V(N)] with t = 1.96 for a = 95%. Samplings were carried out every three months starting from March '94 to June '95. Additional samplings, conducted on December '92, June '93 and March '96, were also available. In order to record displacements inside the cave and migration outside the cave, crickets were marked with different colours in different sectors. Age structure of each sample was studied by means of individual metatibiallength recorded with a vernier caliper.

RESULTS

Taxonomic assessment of Pascola cave population

The inspection of the epiphallus. of Poscola specimens showed substantial differences from D. ligllstica, because of its deep incision at the basal part of the structure, while its shape appeared to fall within the variation of D. laetitiae (Fig.5). Morphological similarity to D. laetitiae was also revealed by other structures as male X uroterga, subgenital plate and ovipositor in the female. However, because of a wide range of variation in these morphological structures within D. laetitiae (Sbordoni et aI., 1979), chromosome and allozyme analyses were carried out to support taxonomic assignment. Figure 4 illustrates mitotic chromosomes obtained from a nymph of Poscola population. A diploid number equal to 31 was established, with sex determination XX, XO. As found in other Dolichopoda species (A.P. Bianchi, pers. com.) a supernumerary chromosome occurred. Number and morphology of chromosomes appeared to match to D.laetitiae (Baceetti, 1958). On the contrary the Poseola karyotype differed from that of the geographically closest species, D.ligllstica, which presents a diploid number of 30 chromosomes (Baccetti, 1982). 22 C. BERNARDlNt, c. DI RUSSO, M. RAMPtNI, D. CESARONt and V. SBORDONt

Furthermore allozyme polymorph isms analysis showed a low level of genetic divergence between the Poscola population (PSC) and the two D.laetitiae populations (PIT and GIA), as indicated by D values (Nei, 1978) ranging from 0.031 to 0.054. These values in Dolichopoda are usually found between conspecific populations (Sbordoni et aI., 1985; Allegrucci et aI., 1992). On the basis of these analyses wc can safely assign the Poscola population to Dolichopoda laetitiae.

Fig. 4 - Mitotic plate from a male nymph of Poscola cave. The X-chromosome is indicated.

Population size

Table I lists the number of cave crickets in Poscola cave estimated by mark-recapture method since December 1992 until March 1996. Estimates ranged between a minimum of 496 individuals (c,1.:t 127) in June 1995 and a maximum of 1988 individuals (c.1. :t 588) in October 1994. Population size showed strong variation apparently not related to any seasonal trends. A RECENT COLONIZATION OF DOLICHOPODA CAVE CRICKETS 23

Fig. 5 _ Scanning electron microscope photographs of the epiphallus: 1 ;::D. ligustica (LIO); 2=Dolichopoda from Poscola cave; 3-8=D. laetitiae, (3=ZIa. 4=DRA, 5;;::FSSI, 6=TER, 7=FUN. 8=SUT; for population symbols see text). 24 C. BERNARDINI, C. DI RUSSO, M. RAMPINI, D. CESARONI and Y. SBORDONI

Table I - Population size (N) estimated in Poscola cave by mark-recapture method. M :: marked individuals; T :: total individuals recaptured; R :: marked individuals re. captured; V :: variance; C.1. :: confidence limits

M T R N V c'1,(+/-) Dec-92 256 283 79 909 7330 166 Juo-93 188 283 32 1618 67520 509 Mar-94 183 163 37 790 12252 215 Juo-94 182 252 40 1123 25070 310 Oct-94 259 283 36 1988 90154 588 J30-95 513 360 98 1870 25405 311 Apr-95 527 418 116 1887 21667 288 Juo-95 173 105 36 496 4209 127 I\lar-96 167 144 22 1053 7644 171

Age structure and sex-ratio

Figure 6 reports the frequency distributions of metatibia length obtained from individual measures in seasonal samples. Individual growth showed a clear seasonal trend, resulting from the occurrence of different age groups in different periods of the year. Adults start to emerge at the begining of summer and show a frequency peak in autumn, while in winter they disappear. During the winter, the population is composed only by nymphs of different age which stop growing because of dormancy. In autumn, two distinct choorts of individuals are clearly identifiable, A first one is mainly formed by adults, and a second by youngs and nymphs. The sex ratio for each of the samples was not statistically different from the ratio 1: I,

Spatial distribution and migratioll

Temperature variation in space and time is summarized in Table 2. The sector I and 4 appeared rather variable with seasonal 6.T ranging from 7.7°C to 9°C, On the contrary sectors 2 and 3 appeared relatively more stable, with a 6.T ranging between 4.6°C and 5.5°C, Sector I showed a mean temperature relatively low (I a,2°C), while sector 4, with a mean temperature of 13.5°C, was close the optimum temperature (15°C) of Dolichopoda life cycle (Di Russo & Juberthie, 1995), According to these data sector I was seldom inhabited by Dolichopoda, most of which occured in the sector 4 in every season. Only in winter the spatial distribution of crickets appeared more heterogeneous as a consequence of displacement of Dolichopoda inside of the cave, particularly in the sectors 2 and 3. Mark- recapture data also confirmed these observations. In fact, several crickets, A RECENT COLONIZATION OF DOLICHOPODA CAVE CRICKETS 25

:1_.- L. •

.,.....•..•..• • ••••p.lI """,

...... " zo •• mel.hbi.l I.ng'h (",m) me'.Uhl.1 le"g.h. (mm)

Fig. 6 _ Comparison of age structures in periodical samples from Pascola cave population. Frequency distributions are based on meta tibia length. marked in autumn and winter and coming from the sectors close to the entrances (l and 4), were recaptured in the sectors 2 and 3. On the contrary, 10 individuals, marked in winter in the two inner sectors, were found in the sector 4 on spring and summer. Dolichopoda were also discovered in the small caves near Poscola cave. Migration between these caves can occur, as established by recapture in PSCI of one individual previously marked in the main cave. Apparently only one out of these small caves (PSC3) showed a stable population. This was estimated by periodical counts ranging between 200 and 400 individuals. In the summertime, occasional sights of Dolichopoda, were also recorded outside caves.

Table 2 - Summary of thermal condition in the four sectors of Pascola cave mince max °C mean l>T (0C) sector 1 5 14 10.2 9 sector 2 7.7 t3.2 t 1.5 5.5 sector 3 tt t5.6 12.8 4.6 sector 4 t 1.6 t9.3 13.5 7.7 26 C. BERNARDINI. c. DI RUSSO. M. RAMPINI. D. CESARONI and V. SBORDONI

Genetic structure Allele frequencies at 19 loci calculated for the four population samples (PSC, PSCI, PSC3 and PSC4) are reported in Table 3. Different samples showed allele frequencies quite similar at most loci, and genetic distance values were very low in any pairwise comparison, ranging from 0.000 to 0.003. The analysis of the genetic structuring for these nearby populations was performed by using 8 (Weir and Cockeram, 1984) as FST estimator. The global 8 value was 0.009, suggesting absence of population heterogeneity in the study area, and the deriving Nm product was 27.52. Accordingly, Nm values for single pairwise comparisons between samples appear rather high (all of the six values> 14), suggesting that several migrants are exchanged per generation between these four population units. In Table 3 estimates of genetic variability at 19 gene loci in all study samples are also reported. On the whole, expected heterozygosity (He) ranged from 0.111 to 0.174.

Table 3 - Allele frequencies, observed (Ha) and expected (He) heterozygosities for the 4 population samples from Poscola area; n refers to the number of assayed individuals.

Locus PSC PSCI I'SC3 PSC4 Locus PSC PSCI PSC3 PSC4 Acp n. 16 24 33 28 Lap-} n. 14 21 34 27 A 0.969 0.917 0.955 1.000 C 0.321 0.452 0.309 0.296 C 0.031 0.083 0.045 0.000 D 0.536 0.524 0.618 0.611 A!do n. 16 24 34 29 E 0.143 0.024 0.074 0.093 A 1.000 1.000 1.000 1.000 Lap-2 n. 14 12 27 24 Dia n. 16 23 31 26 A 0.821 0.958 0.963 0.812 A 0.156 0.152 0.226 0.173 B 0.143 0.000 0.037 0.167 B 0.719 0.848 0.694 0.808 C 0.036 0.042 0.000 0.021 C 0.125 0.000 0.081 0.019 Mdh-l n. 16 24 34 25 E.~(-I n. 16 22 33 28 A 0.187 0.000 0.029 0.t40 B 0.906 0.977 0.924 0.875 B 0.812 0.958 0.926 0.840 C 0.094 0.023 0.076 0.125 D 0.000 0.042 0.044 0.020 Est.2 n. 16 24 34 29 Mdh-2 n. 16 24 34 29 A 0.062 0.125 0.279 0.103 A 0.937 1.000 0.985 1.000 B 0.750 0.792 0.603 0.776 D 0.062 0.000 0.DI5 0.000 C 0.187 0.083 0.118 0.121 Mpi n. 16 24 34 28 Got n. 14 23 32 23 B 1.000 1.000 1.000 1.000 A 0.036 0.022 0.016 0.022 Pep-I n. 16 24 33 29 B 0.964 0.978 0.984 0.978 A 1.000 1.000 1.000 1.000 Gpi n. 16 24 34 29 Pgm n. 16 23 34 27 A 0.094 0.000 0.044 0.000 A 0.062 0.022 0.DI5 0.019 B 0.406 0.479 0.426 0.569 B 0.906 0.978 0.985 0.981 C 0.500 0.521 0.529 0.431 C 0.031 0.000 0.000 0.000 ldh-J n. 16 24 34 29 Pk-2 n. 16 23 34 25 C 1.000 1.000 1.000 1.000 A 1.000 1.000 1.000 1.000 Idh-2 n. 16 24 34 29 Pt-4 n. 16 24 34 29 B 1.000 1.000 1.000 1.000 A 1.000 1.000 1.000 1.000 Pr-5 n. 16 24 34 29 A 1.000 1.000 1.000 1.000 110 0.181 0.098 0.126 0.124 He 0.174 0.111 0.138 0.139 A RECENT COLONIZATION OF DOLICHOPODA CAVB CRICKETS 27

DISCUSSION AND CONCLUSIONS

The present genetic and morphological investigations carried out on Dolichopoda from the Poscola area allowed us to assign them to D. laetitiae, a species whose northern limit is represented by Po river (Baccetti, 1982). The occurrence of this species to the north of Po river was never reported in the literature. In addition, the absence of Dolichopoda from Poscola cave was witnessed by earlier investigations (Bartolomei, 1957). As a consequence, a quite recent colonization of the Poscola cave seems highly realistic. Concerning the mechanism of colonization, the most plau- sible hypothesis should imply an anthropocore dispersion. Unintentional transplantation of eggs, nymphs or even adults is a likely event in Dolicho- poda. This mechanism has been hypothized to explain the occurrence of Dolichopoda schiavazzii in cellars and other artificial caves in the Argen- tario Promontory, where D. baccettii diverged and speciated in insular allopatric condition. Since the present spread of D. schiavazzii in this area is limited to the neighbouring of the Passionisti Abbey in the inner part of the Promontory, the occurrence of this species in this area has been explained with anthropocore dispersion (Allegrucci et aI., 1982). Alterna- tive hypotheses implying active colonization by Dolichopoda appear unlikely, because of lack of records within a wide area north and south of Po river, which apparently represents a geographic barrier for these crickets. Previous studies on geographic variation of molecular and morphological traits proved that both geographic distribution and speciation processes in Dolichopoda have been strongly affected by such river barriers (Sbordoni et aI., 1991; Cesaroni et aI., 1996). If only few individuals, or even a single female, founded the Poscola cave population, an expected consequence could be a reduction in genetic variability of the present population. The allozyme survey showed that heterozygosity in PSC (He= 0.174) is comparable to the levels of genetic variability reported for other Dolichopoda populations (Sbordoni et aI., 1985, 1991). However, these findings do not necessarily conflict with the hypothesis of a small propagule founding the Poscola population. In fact, several other factors can counterbalance the loss of genetic diversity due to founder effect. Particularly, Nei et al. (1975) developed a general model concerning the effects of bottlenecks on the heterozygosity in respect to the intrinsic rate of growth (r) of the populations. According to this model, for any given mutation rate, a r value ~ 0.1 should correspond to a high decrease of genetic variability in the population. On the contrary for r values ~ 1 the average heterozygosity appears weakly affected by 28 C. BERNARDINI. C. DI RUSSO. M. RAMPINI. D. CESARONI and V. SBORDONI

bouleneck. Being estimated r value ranging between 0.7 and I in Dolichopoda spp. including D. /aetitiae (Di Russo, 1993), it is reasonable to argue that founder effect, whichever number of founders was, did not played a significant role in changing the level of genetic variability of the Poscola cave population. Mark-recapture experiments proved that present Dolichopoda population is abundant and wealthy, confirming the successful achievement of the cave colonization. Population size reached a maximum value of 2000 individuals. This estimated size is probably close to the carrying capacity of the cave. However, relevant variations in population size, observed in various periods of the year, were probably due to either migration outside the cave or winter grouping. In fact crickets were commonly found in suitable habitats (either small caves or wood floor) surrounding the main cave. These migrations, chiefly occurring to search food outside the cave, can represent the main reason for Dolichopoda dispersal across other small caves in the area. The very low FST estimate (8 = 0.009) calculated for the four population samples indicates compleIe lack of genetic structuring of population units which actively exchange individuals and genes. The degree of genetic structuring between samples from Poscola area is comparable to that found between subpopulations of D./aetitiae within a single cave (Cappellani, unpublished data) or in adjacent artificial hypogea (Sansotta, 1993). Since we expect that FST is positively related to the age of the populations and to the occurrence of extrinsec barriers to migration, the value obtained for the Poscola populations also supports a recent colonization and present active dispersion of crickets in this area. Short distance active dispersion of Dolichopoda in other areas of foothills of the Alps appears to be favoured by the prevailing mesophilic bioclimatic conditions. This is clearly showed by recent records of new population units of D. ligllstica septentrionalis near Bergamo whose settlement may have not occurred before 1980 (Comotti, 1982; pers. com.). Population age structure showed a typical seasonal trend as often found in Dolichopoda populations inhabiting artificial caves with trophic resources largely represented by vegetable matters seasonally available outside caves. This kind of age structure might be unexpected in a natural limestone cave as Poscola, relatively rich in bat guano deposits. In similar situations a costantly heterogeneous age structure is generally found as an adaptation to exploit resources inside cave and to reduce intraspecific competition (Di Russo et aI., 1987; Carchini et aI., 1991; De Pasquale et aI., 1995). However, preliminary data on feeding habits of these crickets indicated a diet mainly based on vegetable maUers (L. De Pasquale, A RECENT COLONIZATION OF DOLICHOPODA CAVE CRICKETS 29 pers.com.). Even the spalial distribution of crickets inside the cave supports dependence upon external resources. In fact, in spring and summer, most of crickets were located near the secondary entrance surrounded by mesophilic wood, a suitable environment to move and search food. These findings suggest that the seasonal age structure revealed in this population might be the consequence of both recent colonization of the Poscola area and bioclimatic conditions that favour exploitation of trophic resources outside the cave. Since Dolichopoda cave crickets show a typical semivoltine life cycle (Boudou Saltet, 1971; Di Russo et aI., 1987), it could be expected that a single colonization event might determine lack of mobile individuals during 6-9 months, corresponding to the embryonic development, and emergence of adults every two years. On the contrary, observed phenology clearly showed the occurrence of two choorts formed by individuals of different age overlapping each other in autumn and the emergence of adults every summer. This finding suggests that choort overlap can arise in a few generations, as a possible consequence of variability in the duration of embryonic diapause, nymphal dormancy and asincrony in laying eggs. Di Russo (1993) described a similar situation in two experimentally transplanted populations of Dolichopoda linderi in France, which, starting from a little number of founders, showed choort overlap arised in less than 10 generations.

ACKNOWLEDGEMENTS

We thank Enninio Piva, Isabel Ferrari. Giuseppe Perctto of the Club Speleologico Proteo (Vicenza) and Claudia Tedeschi for the useful collaboration in all the steps of field study. We also thank Niccolb Falchi for drawing pictures.

REFERENCES

ALLEGRUCCI, G., A. CACCONE, D. CESARONI, M. COBOLLl SBORDONI, E. DE MATIHAEIS. and V. SBORDONI. 1982. Natura! and experimental interspecific hy- bridization between populations of Dolichopoda cave crickets. Experientia 38: 96.98. ALLEGRUCCI, G., A. CACCONE, D.CESARONI, and V. SBORDONI. 1992. Evolutionary divergence in Dolichopoda cave crickets: a comparison of single copy DNA hybridiza- tion data with allozymes and morphometric distances. J. Eva!. BioI. 5: 121-148. BACCETTI, B. 1958. Notulae Orthopterologicae.lX. Osservazioni cariologiche sulle Doli. chopoda italiane. Redia 43: 315-327. BACCETTI, B. 1982. Ortotteri cavernicoli italiani (Notulac Orthoptcrologicae.XXXVI.). Lav. Soc.ltal. Biogeogr. Verona. 1978 (n.s.); 6: 195-200. BACCETTI, B. and F. CAPRA. 1959. Notulae Orthopterologicae. XII. Revisione delle specie italiane del genere Dolichopoda. Bol. (Orth. Rhaph.). Redia, 44: 165-217. 30 C. BERNARDINI. C. DI RUSSO. M. RAMPINI. D. CESARONI and Y. SBORDONI

BACCEITI, B. and F. CAPRA. 1970. Notulae Orthopterologicae. XXVII. Nuove osserva. zinni sistematiche su akune Dolichopoda italiane esaminate anehe a1 microscopic clet- tronieD a scansione. (Orth. Rhaph.). Mem. Soc. Eot. Ital., 48: 351-365. BAILEY, N.T.I. 1951. On estimating the size of mobile population from recapture data. Biometrika, 38: 293-306. BARTOLOMEI, G. 1957. La groua della Poscola. Rassegna Speleologica Italiana, 9: 51-59. BOUILLON, M. and I.M.HUBART. 1982. Premiers resultats d'unc experience de transplantation de cavernicoles pyreneens dans une grotte de Belgique. Bull. Cherch. Wall. 25: 97-106. BOUDOU-SALTET. P. 1971. Observations du cycle vital, en elevage, d'un Orlhoptt~re ca. vernicole Dolichopoda linderi Duf. Bul!. Soc. Hist. Natur. Toulouse. 107: 301-305. CARCHINI. G.• M.RAMPINI. C.SEYERINI. and Y.SBORDONI. 1983. Population size estimates of four species of Doliehopoda in natural and artificial caves of Central Italy (Orthoptera, Rhaphidophoridae). Mem. Biospeol. 10: 341-347. CARCHINI, G., C.DI RUSSO, and V.SBORDONI. 1991. Contrasting age structures in cave cricket populations: patterns and significance. Ecol. Entomo!. 16: 305-314. CESARONI. D.• P.MATARAZZO. G.ALLEGRUCCI. and Y.SBORDONI. 1996. Comparing patterns of geographic variation in cave crickets by combining geostatistic methods and Mantel tests. J.Biogeogr.: in press. COMOTfI, G. 1982. Fauna cavernicola orobica - 1°. Presenza di Dolichopoda sp. nel ber- gamasco (osservazioni ecologiche e biologiche). OL. BUS - Speleo Club Orobico _ CAl. - BG. 5 (5): 21-39. DE PASQUALE. L.. D. CESARONI. C. DI RUSSO. and Y. SBORDONI. 1995. Trophic ni- che, age structure and seasonality in Dolichopoda cave crickets. Ecography 18: 217-224. 01 RUSSO, C. 1993. Adaptations et divergence evolutive des populations des Dolichopoda des Pyrenees Orientales (Groupe D.linderi). These de doctorat. Universite P. Sabatier Toulouse III. DI RUSSO. C.• A. YELLEr. G. CARCHINI. and Y. SBORDONI. 1987. Life cycle age structure of Dolichopoda populations (Onhoptera, Rhaphidophoridae) from natural and artificial cave habitats. Boll. Zoo!., 54: 337-340. 01 RUSSO C. and C. JUBERTHIE. 1995. Effetto della temperatura sullo sviluppo embrio- nale in Dolichopoda (Orthoptera, Rhaphidophoridae). Fragm. Entomo!. 27 (I): 51-59. FABIANI, R. 1904. Contributo alia conoscenza della fauna delle grotte di Monte di Malo, Priabona e Cereda. Riv. It. Speleologia 2 (I). GINET, R. 1965. Experience de colonisation souterrain aquatique par Niphargus (Crust. Amphipodae); premiers resultats biologiques. Bul!. Soc. Zool. Fr. 90: 581-588. LINCOLN, F.e. 1930. Calculating waterfowl abundance on the basis of banding returns. Circ. U.S. Dep. Agric .• nx r 18. NEI, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics, 89: 583-590. NEI, M.• T. MARAUYAMA. and R. CHAKRABORTY. 1975. The hottelneek effect and genetic variability in populations. Evolution, 29: 1-10. PATRIZI, S. 1956. Introduzione ed acclimatazione del Coleottero Catopide Bathysciola de- m,wsi Dod. in una groua laziale. Le Groue d'Italia, 1:303. RAMPINI. M.. G.CARCHINI, and Y.SBORDONI. 1983. Ecologia di una popolazione di Dolichopoda geniculata (Costa) in una cavita artificiale del sottosuolo di Roma. (Orthoptera, Raphidophoridae). Fragm. Entomol., 17: 85-99. SBORDONI. Y.. G. ALLEGRUCCI. D. CESARONI. and G. SAMMURI. 1978. Sulla posi- zione sistematica e Ie affinita di Dolichopoda "etrusca" in base a dati elettroforetici. Fragm. Entomo!., 15: 67-78. SBORDONI. Y.• G. ALLEGRUCCr. D.CESARONI. M. COBOLLI SBORDONI, and E. DE MATfHAEIS. 1985. Genetic structure of populations and species of Dolichopoda cave crickets: evidences of peripatric divergence. In V.Sbordoni (cd.), Genetics and Ecology in Contact Zones of Populations, Boll. Zoo!., 52: 95-114. SBORDONI, Y.• G. ALLEGRUCCI. and D. CESARONI. 1991. A mullidimensional appro- ach to the evolution and systematics of Dolichopoda cave crickets, pp.171-179. In A RECENT COLONIZATION OF DOLlCHOPODA CAVB CRICKETS 31

a.Hewitt. A.W.BJohnston and J.P.W.Young (005.), Molecular Techniques in Taxonomy, NATO Asi Series, Vol. H57. Springer, Berlin, Heidelberg. SANSOTf A, A. 1993. Grilli cavernicoli e torobe elrusche: steuttura genetica di una popolazione in capporta all'eterogeneita dell'habitat. Tesi di Donaralo di Ricerca in Biologia Evoluzionistica. Universita di Rama "Tor Vergata". SWOFFORD, D. and R.B. SELANDER. 1981. BIOSYS-I. A computer for the analysis of allelic variation in genetics. University of Illinois. Urbana, Illinois. TERCAFS, R. and C. BROUWIR. 1991. Population size of Pyrenean tfoglobiont coleopters (Speonomus species) in a cave in Belgium. Int. J. Speleol. 20: 23-35. WEIR, B.S. and C.C. COCKERHAM. 1984. Estimating F-statistics for the analysis of population structure. Evolution, 38: 1358-1370. WRIGHT, S. 1951. The genetical structure of populations. Ann. Hum. Genet., 15: 323-354.

Int. J. Speleol.. 25. 1-2 (1996): 33-41. 1997

A new genus and species of troglobitic Trechinae (Coleoptera, Carabidae) from southern China

Augusto Vigna Taglianti *

SUMMARY

Guizhaphaenops zorzini n.gen,n.sp. is described from Anjia Yan Cave, Shuicheng County, Guizhou (China). This highly specialized troglobite species is easily recognizable from the other cave dwelling Trechini from China for the main morphological external characters, but its true relationships remain uncertain. the male being still unknown. Similar in habitus 10 Cathaiaphaenops and 10 Sinotroglodytes. the new taxon is much more related 10 the latter. being dorsally glabrous and having the mentum fused with the submentum, with a deep oval fovea. but it differs in its elongated head. with incomplete frontal furrows and without posterior frontal setae.

Only few years ago an extraordinary new genus and species. Sinaphaenops mirabilissimus Ueno & Wang, 1991, was discovered in Guizhou (Libo County), in Southeast China; it was the first record of an anophthalmic trechine beetle from mainland China (Ueno & Wang, 1991). None of the eyeless trechine species, quite abundant in the temperate zones, especially in Northern Hemisphere and also in Eastern Asia, with many genera and species described from Taiwan, Korea and especially Japan (see also Vigna Taglianti & Casale, 1996), was previously known from China. Anolher new genus and species. Dongodytes fowleri Deuve, 1993, was described two years later from Guangxi, collected by an expedition organized by Chinese and British speleologists. Recently, in a paper by Deuve (1996), three other anophthalmic cave- dwelling trechines were described from Northwest Hunan, collected in Ihe karst of Longshan in August 1995, by French and Chinese speleologists of the expedition "Xiangxi 95"; they were ascribed to Iwo new genera (Cathaiaphaenops and Sinotroglodytes) and to one new subgenus (Cimmerirodes) of the Japanese genus Coroblemus Ueno, 1970. One year before, in August 1994, also "China Caves '94", the Ihird spe- leological Italian expedition to China (Zorzin & Melotti, 1995; see also Vigna Taglianti, 1995, for the "Sichuan '93 Project"). visiled some karslic areas in Guangxi and Guizhou provinces. This expedition was organized by dr Roberlo

* Dipartimento di Biologia Animale e dell'Uarna (Zoologia), Universila di Rama "L, Sapien. za", Viale dell'Universita.. 32. I 00185 Roma, Haly. 34 AUGUSTO VIGNA TAGLIANTI

Zorzin, geologist at the Musco Civico di Staria Naturale di Verona, Department of Geology and Paleontology (for the scientific aspects), and dr Silvana Melotti, of the Centro Ricerche Naturalistiche Monti Lessini, Verona (for the speleolo- gical aspects), and carried out in cooperation wilh the Guizhou Normal Univer- sity. Among the few zoological samples collected in Chinese caves, and given to prof. Sandro Ruffo, former Director of the Museo Civico di Storia Naturale di Verona, for sorting and studying, there was also one carabid beetle, that was subsequently kindly and friendly forwarded 10 me. This specimen belongs to a new genus and species of highly specialized eyeless Trechinae, whose description is the object of the present note.

Guizhaphaenops n.gen. Diagnosis

An highly specialized, but not aphaenopsoid, trechine genus, with very elongated antennae and legs, round pronotum and large ovate elytra, with serrulate shoulders. Similar to the genus Sinotroglodytes Deuve, 1996, in the general shape of the body and appendages, and in having the mentum fused with submentum, with a deep oval fovea, the new taxon clearly differs in elongated head, with incomplete frontal furrows, and without posterior frontal setae, in the right mandible strongly bidentate, in the labial tooth clearly bifid, in the outer prebasilar setae longer and stronger, in the presence of the posterior lateral setae on pronotum, in the clearly keeled protibiae and in the pubescent ventral surface. Known only on a female specimen, the true relationships of the new genus arc still uncertain, the male remaining unknown.

Type species

Guizhaphaenops zorzini Vigna Taglianti, n.sp.

Description

A trechine genus, showing a high morphological adaptive modification to the cave environment. Medium sized, with very elongated antennae and legs, round pronotum and large ovate elytra (Fig. I); surface completely glabrous on dorsum, except the thiny pubescent genae, but well pubescent on abdominal sternites. Inner wings totally reduced. Colour reddish brown, relucent and shiny. Head elongated, longer than prothorax, narrow, with a clear neck constriction; frontal furrows incomplete, divergent and obsolete posteriorly; eyes completely vanished; genae parallel, with very short, thin and sparse hairs; only one A NEW GENUS AND SPECIES OF TROGLOBITlC TRECHINAE 35

1 mm

Fig. 1 _ Guiz.haphaenops zonini R.gen. n,sp., from Anjia Van cave in Guizhou, holotype: habitus.

pair of fronlal sclae (Ihe anterior ones), behind Ihe level of ocular oblique sutu- ra; two pairs of short setae above the antennal base; c1ypeus with right margin, 36 AUGUSTO VIGNA TAGLlANTt

and with four setae, the outers being much longer and stronger; labrum genIly emarginate at the apex, with six setae, the Duters much longer, the inner ones very short and thin. Mandibles slender, feebly arcuate, acute at apices; the right mandible strongly bidentate (Fig. 2), the left one with straight cutting edge. La- bium (Fig. 3) completely fused, without trace of labial suture; mentum with a porrect tooth in apical emargination, half as long as lateral lobes, bifide at the tip; in the middle, the mentum shows a depressed round-oval fovea, deep, behind the apical setae (two, lined with the deeper emarginations) and before the semicircular series of nine prebasilar setae. Of these setae, the inner (five, at the apex of submentum) and the apical ones (two, at the basal sides of mentum) are similar in length and size, the outcr (at the side angles of submentum) are much more longer and stronger. Ligula triangularly produced, octosetose (the apical two setae are longer and stronger); paraglossae moderately arcuate, extending much beyond the ligula; labial pal pus long and slender, penultimate segment slighIly curved and dilated at the apex, with two setae at the apex and one, longer, at half the length, on the anterior side; apical segment about four-fifths as long as the penultimate. Maxillae slender, with lacinia gently curved and provi- ded with recurved spines and hairs on the inner margin; maxillary palpus (Fig. 3) long and slender, glabrous, with only three short setae at the apex of penultimate segment; penultimate segment long, subcylindrical in the basal half and si- nuately dilated towards the apex; apical segment about eight-ninths as long as the penultimate. Antennae long and slender, reaching the four-fifths of the elytrallength; scape is the thickest of all segments, and the less setose, while the penultimate is the shortest and segment 3 the longest. Prothorax short; pronotum round-ovate, as long as head (without labrum) and slighIly wider than long, with the maximum width at the half of the lenght; sides gently arcuate, morc gradually narrowing towards the base than towards the apex; front angles obtusely rounded, hind angles completely rounded, continuous with the base; lateral finc borders continuing onto apex and base, gently and gradually enlarged behind the basal third; basal foveae indistinct, superfi- cial, extending anteriorly with the side borders. Marginal setae present, the apical one being the longest; the apical ones situated before the first-fifth of the length, the basal ones at the posterior tenth, before the hind angle. Scutellum distinct. Elytra ovate, much wider than prothorax, with maximum width at basal third, genIly narrowing towards the apex; slighly convex, with flattened basal and humeral area; oblique prehumeral borders, distinct shoulders, narrowly rounded apices; side borders complete from basal peduncle to apex, serrulate at shoulders, with short setae at the apex of each denticulation (10-12, in the basal fourth); striae shallow, vanishing at the sides, at the base and at the apex; the inner four striae more complete and deeper; basal pore present; two setiferous A NEW GENUS AND SPECIES OF TROGLOBITIC TRECHINAE 37 dorsal pores present on slria 3, the anterior one before the fore sixth, the posterior one clearly before Ihe middle; preapical pore absent; two apical pores present, slightly internal to the recurrent stria; marginal series of umbilicate pores not aggregate, as in Fig. I: Ihe first pore a little distant from marginal gutter, the second (at the level of Ihe fore discal pore) adjoining Ihe marginal gutter, Ihe Ihird a little distant from gutter, likc the first one; all three anlerior pores very close to each other, equidistant; the fourlh widely distant (Ihree time the dislan- ce between thc previous), displaced from the gutter; fifth and sixth also very close to each other, the sixth adjoining Ihe gutter; seventh and eighth at a similar distance to each olher, at the apical sixth of elytron, the eighth adjoining marginal gutter. Ventral surface with sternites 4-7 bearing a relatively thick and long pubescence; 4-6 with onc pair of median setae along the posterior margin; in female, slernite 7 with two pair of similar setae, the outer being a little longer Ihan inner ones. Legs very long and slender; fore legs with clearly sulcate libiae, wilh thick apical pubescence on inner side; tibiae of median and hind legs thiny sulcate on the entire inner side, with pubescence on inner face, thicker towards the apices; tarsi slender, pubescent on dorsal side. Female genital appendages (Figs 4-5) without peculiar features; penultimate segment of gonostyli wilh two subapical spines on dorsal surface; apical segment elongate, gently curved, with two dorsal (inner and outer) spines and ventral sensorial fovea, with two thin hairs.

Derivatio nominis

The generic name is composed by "Guizhou" (the Province toponym) and "aphaenops", to recall Ihe name of the classic, specialized, speciose and wide- spread genus of cave-dwelling Trechini, Aphaenops, meaning "without visible eyes".

Guizhaphaenops zafzin; n.sp.

Diagnosis A medium sized trechine, included in Guizhaphaenops n. gen., wilh elongated head, neck constriction clearly visible, shorlened frontal furrows, frontal posterior seta lacking, right mandible strongly bidenlate, labial tooth clearly bifid, outer prebasilar selae longer and stronger, hind laleral setae of pronotum present, clearly keeled protibiae and pubescent ventral surface. 38 AUGUSTO VIGNA TAGLIANTI

E E E E E E d '" d '"d

Figs 2.5 - Guizhophaellops wrzilli n.gen. n.sp., holotypc: right mandible (2), labium and maxillary palpi (3), genital appendages (4) and left gonostylus. at major magnification (5).

Typ~material Holotype: femal~, China, Guizhou Province, Suicheng County, Show Ga, Anjia Yan.cave, 20.VIII.l994, R. Zorzin leg., in ColI. Vigna Taglianti, at the Zoological Museum of .~;LaSapienza'~University, Rome. ; A NEW GENUS AND SPECIES DF TROGLOBtTtC TRECHtNAE 39

Description

Complete description as for the genus. Total length 6.12 mm from apical margin of clypeus to apices of elytra. Head a little longer than wide: length (from apical margin of clypeus to apical margin of pronotum) 1.25 mm, greatest width (behind level of the frontal setae) 0.9 nun. Antennae 5.1 mm long. Microsculpture reticulate, in fine polygonal meshes. Pronotum round-oval; length (along the mid-line) 1.18 mm, greatest width (at mid length) 1.18 mm. Microsculpture reticulate, with fine meshes transver- sely stretched. Elytra ovate, narrowing towards apices; greatest length 3.67 mm, greatest width (behind level of the 4 marginal umbilicate pore) 2.24 mm. Microsculture clearly transverse. Legs slender, elongated; length of hind tibia 2.55 mm.

Derivatio nominis It is a pleasure to dedicate this new species. the sixth specialized cave dwelling trechine so far known from China, to its collector, dr Roberto Zorzin, a good friend of the author and organizer of the "China Caves '94". Italian spe- leological expedition.

Ecological notes The type locality, Anjia Yan cave, whose name is from a nearby little 'village in the Shuicheng County (Guizhou, China), lies in the locality Show Ga, about 50 km northwestwards from Shuicheng, 2160 m above sea level. This cave, open inside limestone aged between Devonian and Triassic, of fossil type, with only few pools with dropping water, is about 300 m long, with 65 m of difference in heigth from enirance to the end. The cave is shortly described by Zorzin & Meloni (1995), and the map and section were made by the "China Caves '94" expedition (not yet published). The cave has a wide entrance, lenghtened orizontally, with an enormous rock slab above large masses of slides. The vault develops sub-horizontally and has a rather regular configura- tion, while the floor is covered by slide masses. Many concretions are present, occasionally very large. A narrow meander (7-8 m deep on average, at some points 20 m deep) runs along the entire length, with pools of dropping water on the bed. The cave, from the ecological point of view, is of the temperate type: the temperature of the air into the cave was 1O.8'C (oiItdoor l8.6'C), of the water 1O.6°C. 40 AUGUSTO VIGNA TAGLIANTI

The unique specimen of Guizhaphaenops zarzini D.gen. n.sp.was found in one of the meander pools, without water, on a rotten wood.

Remarks

At the present state of knowledge on the cave-dwelling Trechini from Chi- na, it is nearly impossible to interpret correctly the status and affinities of the new taxon and to make some hypotheses about the phylogenetic relationships. Examining the previously known taxa separately, Sinaphaenops mirabilissimus Veno & Wang, 1991, from a cave in Libo County, Guizhou Province, is un- doubtfully the most specialized leech inc from China and one "showing the highest morphological modification adaptive to subterranean existence" in the world; its peculiar morphological features are quite different from the new taxon, nevertheless a certain "family likeness" (i.c. enlarged humeri, oval convex elytca, arrangement of marginal setae of umbilicate pores of elytra) would be not so impossible to detect between them. Dongodytes fowled Deuve, 1993, from a cave in Guangxi, the other "aphaenopsoid" species from China, is totally different, and much more similar to the highly specialized European trechines, with vanishing shoulders and flat elongate elytra. The other three known taxa, all from caves in Longshan area, in the northwest Hunan Province, show a much less specialized habitus: Gotoblemus (Cimmeritodes) huangi Deuve, 1996, a little pubescent species, belonging to a subterranean genus previously known from Japan, is clearly unrelated to the new taxon, while Cathaiaphaenops del- prati Deuve, 1996, and Sinotroglodytes bedosae Deuve, 1996, arc more similar in habitus to Guizhaphaenops ZOfzini n.sp. The new taxon herein described is easily distiguishable from both in the frontal furrows incomplete and in lacking the frontal posterior setae; from Cathaiaphaenops also by lacking upper pubescence and preapical discal setae, and by the labial fusion and the bidentate mandible; from Sinotroglodytes also by the bidentate mandible and the presence of hind pro notal seta. In the key presented by Deuve (I 996), the new genus Guizhaphaenops would have an intermediate position between the first group ("morphologie "aphaenopsienne" hyperspecialisee ... siIlons frontaux incomplets, reduits", i.e. Sinaphaenops and Dongodytes) and the second one ("morphologie generale mains specialisee ... sillons frontaux complets", i.e. Golob/emus (Cimme- ritodes), Cathaiaphaenops, Sinotroglodytes), because of the elongated head, with marked neck, the incomplete frontal furrows and the lack of posterior frontal setae. In my opinion, Guizhaphaenops ZOfzini would rather be a more specialized taxon of the same line of (and perhaps congeric with) Sinotroglodytes bedosae, and Sinaphaenops mirabilissimus would be possibly the highest specialized taxon of the same line. A NEW GENUS AND SPECIES OF TROGLOBITIC TRECHINAE 41

Only the knowledge of Ihe male of Guizhaphaenops and of much more taxa of cave dwelling Trechini from the unexplored karstic areas in China would allow to clarify Ihe systematic relationships between these different taxa (now provisionally ranked at the level of different genera), and their biogeo- graphic histories.

ACKNOWLEDGEMENTS

I am indebted to the organizers of the expedition "China Caves '94", namely de Roberto Zor- zin, geologist, fonnerly member of the Department of Geology and Paleontology of the Museo Ci~ vieD di Staria Naturale di Verona, and de Silvana Melotti, of the Centro Ricerche Naturalistiche di Verona; of course. I am also indebted to all other speleogists of C.R.N. and of Unione Speleologica Veronese. for collecting the biological samples during the expedition. But wann acknowledgements are due 10 Roberto Zorzin. not only as collector of the new carabid. but also for giving me the relevant geographical and ecological informations, and to my teacher and friend, prof. Sandro Ruffo, formerly director of the Museo civico di Storia naturale di Verona, for having forwarded to me the specimen of Guizhaphaenops. Many thanks also to prof. Achille Casale, for the useful discussions and informations, to dr Thierry Deuve and dr Shun-Ichi Veno, for sending me their papers on the Trechini fauna from China, to dr Marco Oliverio, for the help in the revision of the manuscript and to dr NiccolI) Falchi, for inking the illustrations.

REFERENCES

DEUVE, T. 1993. Description de Dongodytes fowleri n.gen., n.sp., Coleoptere troglobie des karst du Guangxi, Chine (Adephaga, Trechidae). Bulletin de la Societe entomologique de France 98 (3): 291-296. DEUVE, T. 1996. Descriptions de trois Trechinae anophtalmes cavernicoles dans un karst du Hu- nan, Chine (Coleoptera, Trechidae). Revue fran~aise d'Entomologie (n.s.) 18 (2): 41-48. VENO, S.-I. & F.WANG. 1991. Discovery of a highly specialized cave Trechine (Coleoptera, Tre- chinae) in Southeast China. Elytra, Tokyo 19 (1): 127-135. VIGNA TAGLIANTI, A. 1995. A new Jujiroa from Sichuan, China (Coleoptera, Carabidae). In- ternational Journal of Speleology 23 (3-4) (1994): 179-189. VIGNA TAGLIANTI, A. & A. CASALE. 1996. CarJ.bid beetles and subterranean environments. XX International Congress of Entomology, Firenze, Italy, August 25-31, 1996, Proceedings: 355. ZORZIN, R. & S. MELOTTI. 1995. China Caves '94: Guangxi and Guizhou provinces. The Inter- national Caver Magazine 13: 17-25.

Int.!. Speleo!.. 25.1-2 (1996): 43-57.1997

First record of Parastenocarididae (Crustacea, Copepoda, Harpacticoida) from subterranean freshwater of insular Greece and description of two new species.

Vezio Cottarelli and Maria Cristina Bruno *

SUMMARY

The genus Parastenocaris. new fOf Greece, has been discovered in the hyporheic habitat of Kos and Kythira Island with two new species. that are described and discussed in this work. Para.~tenocarjs ae.~culapii n. sp. shares some characters wilh P. nolli from Germany and P. italica from Italy, Macedonia and Turkey. Parastenocaris aphroditis n. sp belongs. according to the Authors, to a group of species living exclusively in estuarine interstitial bu. bitats. which are all characterised by peculiar morphology and ecology, that are as well considered and interpreted.

INTRODUCTION

The large genus Parastellocaris Kessler. 1913. is almost exclusive of subterranean waters from all over the world; according to Dussart & Defaye (1990), it is widely distributed in Southern Europe except, strange enough. in Greece. For this area there is only one dubious report on the presence of Parastenocaris (see above), even if subterranean habitats have been recently invesligated (Bruno & Cottarelli. in press; Forniz et al.. 1990; Pesce, 1981, 1985; Pesce & Maggi. 1983). This gap has now been filled with the record of two new species from the Greek Islands of Kos and Kythira. that are described and discussed here. Some remarks on relationship. biogeo- graphy and ecology of these Parastenocaris, will also be considered.

MATERIAL AND METHODS

Samples were collecled using the Karaman-Chappuis method, fixed in 5% buffered Formalin. Specimens were mounted in permanent slides with Faore's medium and drawn at 1250X, with an oil immersion lens, using a

* Dipartimento di Scienze Ambientali. Univcrsita della Tuscia. Via San Camillo de Lellis

_ 01100 Viterbo. Italy. ,t 44 VEZIO COITARELLI and MARIA CRISTINA BRUNO

drawing tube mounted on a Zeiss Axioskop phase contrast microscope. The following abbreviations are used, when required, throughout the text and figures: AI= antennula; A2= antenna; Bsp= basipodite; Enp= endopodite; Exp= exopodite; Fu= furcal rami; Ga= genital field; Gsg= genital somite; Md= mandible; Mx= maxilla; Mxl= maxillule; Mxp= maxilliped; Op= anal operculum; PI-P5= thoracic appendages; R= rostrum. The nomenclature and descriptive terminology follow Dussart & De- faye (1995) and Huys & Boxshall (1991). For both species, the material of the typical series is temporarily deposited at the "Dipartimento di Scienze Ambientali, Universita della Tuscia", Viterbo (senior author's collection) except for the holotype and one female paratype, deposited in the collection of the "Museo Civico di Storia Natu- rale A. Doria", Genoa (Italy).

Fam. PARASTENOCARIDIDAE Chappuis, 1933 Gen. Parastenocaris Kessler, 1913 Parastenocaris aesculapii n. sp.

Type material Holotype: I male, V. Cottarelli leg., 5-04-1992, hyporheic water of a small stream near the Asklepeion, in Kos, Dodecanese Islands, dissected and mounted on a slide labelled: Kos - Parastenocaris aesculapii hI. Para- types: I male and 5 females from the same station, V. Cottarelli leg., 23-03- 1989; 24-02-1990, 5-04-1992, also mounted on slides numbered from I to 6, labelled: Kos - Parastenocaris aesculapii pI.

Description of the holotype Body cylindrical and elongate, unpigmented, eyeless; length, from the rostrum to the distal apex of the furcal rami: 0.387 mm. Anal operculum (Fig. 2): rounded, with a row of spinules near the distal margin. Furcal rami (Fig. 2): shorter than the last abdominal somite; length/ width ratio: 3.5. Armature consisting of a group of three setae of different length at about 2/3 of the lateral outer margin, a composite long dorsal seta, a long and strong main apical seta flanked by a short inner and a somewhat longer outer setae. Rostrum and antennule (Fig. 3): rostrum as in figure. Antennule: first segment bare, second segment with five distal setae, one is plumose; third segment with four setae. Fourth segment enlarged, has a seta and a tipped FlRST RECORD OFPARASTENOCARIDIDAE FROM SUBlERRANEAN FRESHW A1ER 45 apophysis on Ihe proximal margin, a distal tubercle with two setae and an aesthetasc that reaches in length the end of the last segment. Fifth segment partially merged with the previous one, without armature, as the sixth segment, which is lengthened in a strong curved tip. Seventh segment with eight setae and an aesthetasc. Antenna (Fig. 4): one-segmented exopodite with a slender apical seta; allobasipodite with a row of cuticular spinules at about half of the inner lateral margin. Endopodite bearing distally three geniculated setae, three one- side barbed setae. Remaining ornamentation as in figure. Mandible (Fig. 7): one-segmented palp, with two distal setae. Maxillule (Fig. 8): arthrite of the praecoxopodite with three distal curved spines and some spinules; a curved seta is inserted at about half length of the arthrite. Coxopodite with a distal seta, basipodite with two apical setae. Maxilla (Fig. 9): syncoxopodite with two endites, one bears a normal and a leaf-like setae. The other one with a barbed seta. Basipodite prolonged into an apical curved and spinulose tip with a lateral normal seta. En- dopodite reduced to a small tubercle with two setae. Maxilliped (Fig. 10): slim and elongate; basipodite unarmed, first segment with some small distal spinules; second segment unarmed, with a long, curved and pennate apical spine. PI (Fig. II): exopodite three-segmented. almost as long as the endopodite. Third segment with two geniculate apical setae, an apical and a subapical spines. Endopodite two-segmented; second segment with a long and geniculate and a barbed setae on the apex. P2 (Fig. 12): exopodite three-segmented, armature shown in figure. Endopodite shorter than 1/3 of the first segment of the corresponding exopodite, represented by a little cylindrical segment, with some apical spinules and one seta. P3 (Fig. 16): squat, basipodite with a row of spinules and a long seta; a row of very small spinules near the lateral inner margin. Endopodite is missing. First segment of the exopodite ending in a strong and long apophysis with a group of four cuticular spinules inserted on the outer margin. Second segment represented by a leaf-like appendix, almost as long as the apophysis, with undulating margins. P4 (Fig. 15): Exopodite three-segmented, chaetotaxy as shown in figure. Endopodite as long as half of the first segment of the corresponding exopodite, outer margin with a row of spinules decreasing in length proxi- mally to distally. On the basipodite, near the insertion of the endopodite, a short barbed seta. 46 VEZIO corr ARELLI and MARIA CRISTINA BRUNO

1 O.025mm I I

1 1

------

Figs. 1-6 - Parastenocaris aesculapii n, sp. 2.3,4: holotype male. 1,5,6: paratype female. I, Gsg and Ga; 2, Op and Fu; 3, R and AI; 4, A2; 5, Fu; 6, A I.

P5 (Fig. 13): represented by a sharpened plate of approximately triangular shape, prolonged in a strong tip. About half way along the distal margin are one long and two short setae.

Description of the paratype (female)

Length: 0.396 mm. Rostrum, antenna, oral appendages, maxilliped, PI, P2 and P4 exopodite, anal operculum, as in the male. Genital somite provi- FlRST RECORD OF PARASlENOCARIDIDAE FROM SUBlERRANEAN FRESHW AlER 47 ded ventrally with an undulated row of spinules, below the genital field (Fig. 1). Furcal rami (Fig. 5): with the same armature as in the male, but shorter. Length/width ratio: 2.6.

17 f~ IO.025mm

Figs. 7-19 - Parastenocaris aesculapii n. sp. 7,8,9,10,11,12,13,15,16: holotype male. 14.17,18,19: paratype female. 7, Md; 8, Mxl; 9, Mx; 10, Mxp; II, PI; 12, P2; 13, PS; 14, Enp P4; 15, P4; 16, P3; 17, PS; 18, Enp P2; 19, P3. 48 VEZIO COlT ARELLI and MARIA CRISTINA BRUNO

Antennule (Fig. 6): seven-segmented, second segment with a plumose composite seta and four distal setae; third segment with five distal setae. Fourth segment with a distal tubercle bearing two long setae and an aesthetasc. Seventh segment with an aesthetasc on the apex and seven marginal setae. P2 endopodite (Fig. 18): similar to the male's one but longer, with the same ornamentation. P3 (Fig. 19): endopodite represented by a sharp barbed spine, longer than 2/3 of the corresponding two-segmented exopodite, which is armed as in figure. P4 endopodite (Fig. 14): a little shorter than the first segment of the corresponding exopodite, with sharp apex and some apical spinules. P5 (Fig. 17): similar to the male's one, but with longer setae.

Derivatio nominis The species name has been chosen since the species has been collected near the Askleipeion, the famous temple dedicated to Aesculapius, the Greek God of Medicine.

Variability The features described above appear to be constant in all the specimens of the typical series, except for some small variations in the total length values.

Relationships

Parastenocaris aesculapii n. sp. is well characterised by the morphology and ornamentation of P3 and P4 endopodite in males and of furcal rami in both sexes, but its arrangement of features makes difficult to identify the affinities. The simple structure of P4 endopodite in males, accompanied by a single seta near its insertion, is not so frequently occurring at least between the European Parastenocaris. The same features are shared, for example, by P. budapestiensis Torok, 1935, from Budapest and by P. nolli Kiefer, 1938 from Germany. The male's P3 of P. aesculapii also resembles to the same appendix of P. nolli and P. italica Chappuis, 1953, a species distributed not only in Italy, but also in Macedonian and Turkey. Moreover, another species, still in study, frum Turkey shows some morphological affinities with P. aesculapii. FIRST RECORD OF PARASlENOCARIDIDAE FROM SUBTERRANEAN FRESHWATER 49

Parastenoearis aplzroditis n. sp.

Type material

Holotype: I male, M.e. Bruno and V. CottareJli legg., 3-04-1996, hyporheic water of a small stream mouth in Kastraki beach, near Paleopolis, in Kythira Island, dissected and mounted on a slide labelled: Kythira - Pa- rastenoearis aplzroditis ht. Paratypes: 21 males and 32 females from the same station, also mounted on slides numbered from I to 53, labelled: Kythira - Parastenoearis aplzroditis pt.

Description of tlze Izolotype

Body cylindrical and elongate, unpigmented, eyeless; length, from the rostrum to the distal apex of the furcal rami: 0.354 mm. Anal operculum (Fig. 25): with smooth, rounded margin. Furcal rami (Fig. 20): subcylindrical, shorter than the last abdominal segment; length/width ratio: 2.2. Armature consisting of a subapical composite dorsal seta, a long main apical seta flanked by two shorter ones of different length, two short subapical outer setae. Rostrum and antennule (Fig. 21): rostrum as in figure. Antennule: first segment bare, second segment with four normal and a plumose seta; third segment with five apical setae. Fourth segment enlarged, brings distally one seta near the origin of a tubercle bearing a seta and an aesthetasc that reaches in length the end of the last segment. Fifth segment partially merged with the previous one, with no armature, as the sixth segment which is lengthened in a strong curved tip. Seventh segment with seven setae and an aesthetasc. Antenna (Fig. 30): one-segmented exopodite with a slender apical seta; allobasipodite unarmed. Endopodite bearing distally two geniculate and three barbed setae. Remaining ornamentation as in figure. Mandible (Fig. 26): one-segmented palp, with two distal setae. Maxillule (Fig. 27): arthrite of the praecoxopodite with four distal curved spines; a long and curved seta is inserted at about half length of the arthrite. Coxopodite with a long distal seta, basipodite with two apical setae. Maxilla (Fig. 28): syncoxopodite with two endites, one bears a normal and a geniculate setae, the other endite bears one normal seta. Basipodite prolonged into an apical curved and distally spinulose tip, with a lateral seta. Endopodite represented by a small tubercle with two setae. 50 VEZIO corr ARELLI and MARIA CRISTINA BRUNO

Figs. 20-25 - Parastenocaris aphroditis n. sp. 20,21,23,25: holotype male. 22,24: paratype female. 20, Fu; 21, R and AI; 22, AI; 23, P3; 24, Fu; 25. Op.

Maxilliped (Fig. 29); slim and elongate; basipodite and first segment unarmed, second segment with an apical long, curved and one-side barbed spine. P! (Fig. 3 I); basipodite with a hook near the endopodite insertion. Exopodite three-segmented, almost as long as the endopodite. Third segment with two geniculate, an apical and a subapical spines. Endopodite FIRST RECORD OFPARASlENOCARIDIDAE FROM SUBTERRANEAN FRESHWATER 51 two-segmented, second segment with a long and geniculate and a barbed setae on the apex. P2 (Fig. 32): exopodite three-segmented, armature shown in figure. Endopodite almost as long as half of the first segment of the corresponding exopodite, represented by a cylindrical segment with some apical spinules and one seta. P3 (Fig. 23): quite squat, basipodite with a row of spinules and a long seta; endopodite represented by a seta. First segment of the exopodite with a proximal spine near the lateral outer margin and a group of four spines on the same margin, ending in a leaf-like apophysis with undulating margin and a spinule on the apex. Second segment longer than the end of the apophysis, represented by a thin, elongate appendix. P4 (Fig. 34): exopodite three-segmented, chaetotaxy as shown in figure. Endopodite as long as the second segment of the corresponding exopodite, with the first third smooth and a roundish lateral apophysis. The remaining tenninal part of the endopodite is leaf-like, sharpening to the top, with small lateral tuber- cles and spinules. Near the insertion of the endopodite are inserted a thick spine with curved tip, almost as long as 1/3 of the endopodite itself and a shorter and thinner spine. P5 (Fig. 33): represented by a plate of approximately rectangular shape, prolonged in a distal inner tip. Four setae of different length along the distal margin.

Description of the paratype (female)

Length: 0.363 mm. Furcal rami, rostrum, antenna, oral appendages, maxilliped, PI, exopodite P2 and P4, anal operculum, as in the male. Furcal rami (Fig. 24). Antennule (Fig. 22): seven-segmented, second segment with three normal and a plumose setae; third segment with five distal setae. Fourth segment with a distal tubercle bearing two setae and a long aesthetasc. Sixth segment with a distal short seta; seventh segment with an aesthetasc on the apex and eighth terminal setae. P2 endopodite (Fig. 37): similar to the male's one, with the same ornamentation. P3 (Fig. 35): endopodite transformed in a strong spine with curved tip, a little longer than the first segment of the corresponding two-segmented exopodite, which is armed as in figure. P4' endopodite (Fig. 38): a little longer than the first segment of the corresponding exopodite, ending in a tip, with some small distal spinules. 52 VEZIO COlT ARELLI and MARIA CRISTINA BRUNO

P5 (Fig. 36): with the same ornamentation as in the male, but slimmer and longer.

Derivatio nominis

The species name has been chosen since Kythira is the island were the cult of Aphrodite, in the ancient time, has starIed.

Variability

The features described above appear to be constant in all the specimens of the typical series, except for some small variations of the Iotal length values in the paratypes.

Relationships Parastenoearis aphroditis n. sp. belongs to the minuta group Lang 1948, and shows the closest affinities with some of the species of this group: P. andalusiea Enckell, 1965, P. hera Cottarelli, 1969, P. stellae Cottarelli, Saporito & Puccetti, 1981, P. rivi Cottarelli & Bruno, 1994, P. oligoalina Cottarelli, Bruno & Venanzetti, 1995, P. etrusea Cottarelli, Bru- no & Venanzetti, 1995. All of these species, as well as P. aphroditis n. sp., have been collected in the estuarine hyporheos of streams and rivers and in the Mediterranean area; all of them share a similar ecology and a peculiar habitus, expressed mainly in male's P3, in furcal rami, in endopodite P2 and P5 in both sexes. However, the new species is easily distinguishable by the ornamentation of Ihe first segment of male's P3, by the presence of bOIh the long and hooked and the short and straight spines near the insertion of endopodite P4 in males, by the remarkable length of this endopodite, by the shape and size of endopodite P4 in females. In particular, the morphology of endopodite P4 in males of P. aphroditis n. sp., P. andalusiea and P. etrusea is similar (Cottarelli et aI., 1995b; Rouch, 1990). As regards male's P3, only in P. stellae the spines on the first segment of the exopodite are arranged in two groups (one group is made of two spines at about 1/3 of the length, the other one of a disIal row of five spines); in all the Parastenoea- ris mentioned before there is only a distal raw composed by a variable number of spines. The morphology and ornamentation of P5 is quite constant in all the Parastenoearis so far discussed: in P. hera, P. stellae, P. oligoalina, P. etrusea it is an approximately rectangular plate, prolonged distally in an inner tip with four marginal setae of different length, in P. rivi the inner tip is FIRSTRECORDOF PARASlENOCARIDIDAEFROMSUB1ERRANEANFRESHWA1ER 53

I O.025mm I

Figs. 26-38 - ParaJtenocaris aphroditis n. sp. 26,27,28,29,30,31,32,33,34: holotype male. 35.36,37,38: paratype female. 26. Md; 27, MxI; 28, Mx; 29, Mxp; 30, A2; 31, PI; 32, P2; 33, P5; 34, P4; 35. P3; 36, P5; 37, Enp P2; 38, P4. substituted by a spine. P. andalusica differs because P5 has a triangular shape and bears only three setae. In previous works Cotlarelli & Bruno (1994). Cotlarelli et al. (1995b) had already noticed that it can be useful, in order to distinguish species with a very close morphology, as in Ihis case, to use features such as maxillules and maxillae, that are not usually considered in the taxonomy of this genus. Accordingly, we summarise in the following list the data regarding all the Parastenoear;s of Mediterranean rivermouths. 54 VEZIO COlT ARELLI and MARIA CRISTINA BRUNO

Maxillules Maxillae

P. andallisica Not described Not described

P. aphroditis Arthrite: 4 distal spines Endopoditc: 2 setae I lateral seta Enditc I: 2 setae Coxa: I seta Enditc 2: I seta Basipoditc: 2 setae

P. etrusca Arthritc: 3 distal spines Endopodite: 2 setae I lateral seta Endite I: 2 setae Coxa: I seta Endite 2: 1 seta Basipodite: 2 setae

P. hera Arthrite: 2 distal spines Endopodite: 1 seta I lateral seta Endite I: 2 setae Coxa: I seta Enditc 2: missing Basipodite: 2 setae

P. oligoalina Arthrite: 3 distal spines Endopoditc: 2 setae 1 lateral seta Endite I: 2 setae Coxa: I seta Enditc 2: missing Basipodite: 3 setae

P. rivi Arthritc: 3 distal spines Endopodite: 2 setae Coxa: I seta Endite I: 2 setae Basipodite: 2 setae Endite 2: I seta

P. stellae Not described Endopodite: 1 seta Endite I: 1 seta Endite 2: 2 setae

DISCUSSION

The discovery of these Greek Parastellocaris fills a gap in the distribution of Parastellocaris in Ihe Mediterranean area. In parlicular the already mentioned relaIionships between P. aesculapii n. sp. and a Turkish con- generic species can be explained by Ihe occurrence of past links between Kos Island and Turkey. Kos is nowadays only five kilometres dislant from Bodrum, on the Turkish Coasl and the two lands were connected during Tortonian-Messinian, (Sondaar et aI., 1986; Dermitzakis, 1990) and even for a long lime in Pleistocene, (Audisio & De Biase, 1990). The discovery of P. aphroditis n. sp. allows 10 better trace Ihe dislribu- tion of the species group milluta, confinning that Ihe group is widely represented in Southern Europe and in the Mediterranean Islands. It has been FlRST REmRD OF PARASTENOCARIDlDAE FROM SUBlERRANEAN FRESHWATER 55 collected not only on large islands such as Sicily (Pesce et a!., 1988) and Sardinia (Cottarelli et a!., 1995a), but also on little islands such as Kos, Kythira, San Pietro (Sardinia) (Cottarelli & Bruno, 1994); in Montecristo (Tuscany Archipelago) have been collected (Cottarelli & Torrisi, 1976) only female specimens of a species probably belonging to the same group. As regards continental Greece, there are not data concerning the presence of Parastenoearis: Pesce & Maggi (1983), report of a Parastenoearis sp. from Northern Greece that has never been mentioned in further papers. Anyway, it will be probably possible to demonstrate a wide distribution for this genus in continental Greece too, if some peculiar habitats, such as hyporheos and lacustrine psammon, will be sufficiently investigated. The researches on the interstitial meiofauna adapted to the estuarine ecotone that we have been carrying out for several years, allow us to identify the group of six Parastenoearis in the minuta species group, (P. andalu- s;ea, P. etrusea, P. hera, P. otigoatina, P. r;v;, P. stella e) that share peculiar morphological affinities. These species are the only ones tipically living in Perimediterranean Europe in estuarine habitats, where the ecological factors have a wide variability, to which the Parastenoear;s discussed here seem to be the only adapted species. In fact P. hera, distributed in Central and Southern Italy, has been found only in sandbanks from the mouth of Sele River to 450 meters upriver, where this species is replaced by other Parastenoearis. P. hera shows the same spatial distribution in Volturno and Garigliano Rivers; the same pattern has been detected for P. etrusea and P. stellae, collected from Fiora River (Central Italy) and Cedrino River (Sardinia) respectively, that are replaced upriver by other species of Para- stenoear;s. When the streams are short and small, as those where P. otigoatina (from a small stream in Sardinia), P. rivi (from a small stream in S. Pietro Island,. near the Sardinian Western Coast) and P. aphroditis (from a small stream mouth) live, only these species, typical of estuarine habitats, have been collected. In any case, all the species mentioned here have been often found together with harpacticoids of the genus Psammopsyllus Ni- cholls, 1945 and lchnusella Cottarelli, 1971. P. andalusiea has been collected in Rio Fuengirola, Malaga, Spain, in "sandbank 300 meters from the river mouth. Freshwater interstitial habitat, hardly affected by sea waters" (Enckell, 1965). Two hypotheses can be proposed about the origin of these related taxa: according to the first one, some ancestors already adapted to life in subterranean freshwater would have from this habitat "re-colonised" the estuaries. In this case, the observed morphological and ecological similarities could reflect strong evolutionary convergence in a discrete number of species, di- 56 VEZIO COIT ARELLI and MARIA CRISTINA BRUNO

stributed on a wide area. According to the second hypothesis, that seems to us more convincing, it can be assumed a common origin of these Paraste- nocar;s from an ancestor adapted to life in the river mouths around the Me- diterranean Sea, which would have colonized continental hyporheic habitats afterwards. The ancestor should have already had those morphological characters, probably in a plaesiomorphic state, that are still shared by this group of Parastenocar;s.

ACKNOWLEDGEMENTS

These researches have been supported by grants from M.U.R.S.T (40%) and C.N.R; some samples have been collected (1989) during researches carried out with the Oceanographic Ship Bannock, belonging to the Italian "Consiglio Nazionale delle Ricerche".

REFERENCES

AUDISIO, P. e A. DE BlASE. 1990. Materiali per una biogeografia delle Hydraena e Hae. nidra della Grecia e delle isole egee. I Contributo (Coleoptera, Hydraenidae). Fragm. Entomol. 22 (2): 381-398. BRUNO, M.e. and. V. eDIT ARELLI. New records of Harpacticoids (Crustacea, Copepoda) from subterranean fresh and brackish waters of Greece. BioI. Gallo-Hellenica. in press. COTTARELLJ, V. 1969. Nuove Parastenocaris (Copepoda, Harpacticoida), dell'Italia centro.meridionale. Riv. Idrobioi. 8 (3): 1-8. COIT ARELLI, V. e M.R. TORRISI. 1976. Ciclopoidi e Arpacticoidi (Crustacea, Copepoda) di acque sotterraneee dell'Isola di Montecristo (Arcipelago Toscano). Lav. Soc. It. Bio- geogr. n. s. 5: 357.370. COITARELLI, V., P.E. SAPORITO e A. PUCCETL 1981. Parastenocaris stellae n. sp. della falda iporreica del fiume Cedrino (Sardegna) (Crustacea, Copepoda, Harpacticoi- da). Fragm. Enlomol. 16 (1): 1-7. COIT ARELLI, V. e M.C. BRUNO. 1994. Parastenocaris rivi n. sp. di acque iporreiche dell'isola di S. Pietro (Sardegna). Ann. Mus. Civ. St. nat. G. Doria, Genova. 90: 509-519. COIT ARELLl, V., M.C. BRUNO e C. FORNIZ. 1995a. Copepodi Arpaetieoidi e Sinearidi (Crustacea) di acque sotterranee delle isole circumsarde. Lav. Soc. Itai. Biogeogr., in press. COTIARELLI, V., M.e. BRUNO and F. VENANZETII. 1995b. Two new species of Para. stenocaris from the interstitial waters of rivennouths in Latium and Sardinia (Crustacea, Copepoda, Harpacticoida). Fragm. Entomo1. 26 (2): 229-247. DERMITZAKIS, M.D. 1990. Paleogeography, Geodynamic processes and Event Strati- graphy during the Late Cenozoic of the Aegean Area, pp. 263-288. In: Biogeograhical aspects of Insularity. Atti Convegno Lincei. 85. DUSSART, R. et D. DEFA YEo 1990. Repertoire mondial des Crustaces Copepodes des eaux interieurs. III. Harpacticoi'des. E. J. Brill (ed.), Leiden, The Netherlands. DUSSART, R. and D. DEFA YEo 1995. Introduction to Copepoda. SPB Academic Pu. blishing, Amsterdam, The Netherlands. ENCKELL, P.H. 1965. New Harpacticoids of Spain. Acta Univ. Lund. 2 (19): 1-9. FORNIZ, e., F. VENANZEITI and V. COIT ARELLI. 1990. First record of phreatic harpacticoids from some Aegean islands (Abstract). 5th International Congress on the Zoo- geography and Ecology of Greece and adjacent Regions. Iraklion, Crete; 75. HUYS, R. and G. BOXSHALL. 1991. Copepod evolution. The Ray Soc., London. LANG, K. 1948. Monographie dcr Harpacticiden. Nord. Bokhand., Stockholm. ARSTRECORDOFPARASTENOCARIDIDAEFROMSUBTERRANEANFRESHWATER 57

PESCE, G.L. 1981. Some remarks on the stygofauna of Greece. BioI. Gallo-hellenica. 10: 103-112. PESCE, G.L. 1985. Stygobiological researches in subterranean waters of Lesbos (Greece) and description of Stygonitocrella petkovskii n.sp. (Crustacea, Copepoda: Ameiridae). Fragm. baleanica. 12 (273): 125-139. PESCE, G.L. e D. MAGGI. 1983. Ricerche faunistiche in acque sotterranee della Orecia me. ridionale ed insulare e stato attuale delle conoscenze sulla stigofauna di Orceia. Natura, Milano. 74 (1-2): 15-73. PESCE, a.L. D.P. GALASSI and V: COTIARELLI. 1988. First representative of the family t Parastenocarididae from Sicily (Italy), and description of two new species of Parasteno- caris Kessler (Crustacea, Copepoda: Harpacticoida). Bull. ZOO!. Mus. Amsterdam. 11: 137-142. ROUCH, R. 1990. Deux nouvelles Parastenocaris (Copepodes, Harpacticoides) des Py- renees. Annis Limno!. 26 (I): 19-28. SOONDAR. P.Y., J. DE VOS and M.D. DERMITZAKIS. 1986. Late Cenozoic Faunal Evo- lution and Paleogeography of the South Aegean Islands. Arc. Modern Geology. 10: 249- 259.

Int. J. Speleol.. 25.1-2 (1996): 59-66.1997 lithobius nuragicus n. sp., a new litllObius from a Sardinian cave (Chilopoda Lithobiomorpha)

Marzio Zapparoli *

SUMMARY

A new species of Lithobius from a Sardinian cave (Cagliari. Santadi, loco Su Benatzu. Grotta "Pirosu", 631 Sa/Ca, m 270) is described. Uthobius nuragicus n. sp. belongs to the subgenus Lithobius s. str. and is related to Uthobius variegatus Leach, 1814, occurring in the British Isles, Brittany, Channel Isles, Iberian Peninsula, Maghreb, Sicily and Southern Italy. This new species is differentiated from L variegatus by the number of prosternal teeth (3+3), the number and arrangement of ocelli (I +3; little, depigmentcd, not contiguous to each other. in the center of a depigmented area, posterosuperior ocellus larger than the other ocelli), the size of the organ of Tomosvary (larger), the number of antennal articles (79-86), the number of dorso-lateral and dorso-median setae and the shape and size of the claw of the female gonopods (4.5; 10-11; short, with a small lateral denticle on the internal side).

INTRODUCTION

The species of troglobitic centipedes known to date are very few and most of them belong to the order Lithobiomorpha (Negrea & Minelli, 1994). One of the most important world areas in terms of number of species is the Mediterranean basin, and Sardinia, W Liguria and NE Alps are the most interesting and more characterized Italian regions (Minelli, 1982). The most specialized centipedes in Italy occur in Sardinia, such as the two endemics Lithobius sbordonii Matic, 1967. a troglobitic species whose affinities are uncertain, and Lithobius doderoi Silvestri, 1907, with a lower degree of cave adaptations than L. sbordonii and apparently close to some cavernicolous Lithobius from the Pyrenees. In Sardinian caves we can also find Lithobius cerii Verhoeff, 1942-43, also found in a cave in Capri Island (Thyrrhenian Sea); the affinities of this species are unclear but certainly it is not close to the two species mentioned before. About twenty centipedes are to date known in Sardinian caves (Puddu & Pirodda, 1974; Minelli. 1985; Grafitti & Zapparoli, 1992). In the present article a description of a new species of cave Lithobius, collected in the

* Dipartimenlo di Prolezione delle Piante, Universita della Tuscia, Via S. Camillo de Lellis, 01100 Viterbo, Italy. 60 MARZIO ZAPPAROLI

Sulcis area in the framework of the biospeological campaigns carried out in Sardinia by the Gruppo Speleo-Archeologico "G. Spano", from Cagliari in 1994, is given. The new species, which shows clear morphological adaptations to cave habitat, is not close to other Sardinian cave Lithobius already known, but it seems close to some forms occurring in W-Europe and in the Maghreb.

Diagnosis

A large sized Lithobius s. str. (20.0 mm long), morphologically close to Lithobius variegatus Leach, 1814; antennae of 79-86 articles, 1+3 ocelli on each side, organ of Tomosvary larger than the ocelli, prostemum with 3+3 conical teeth, porodont spiniform; tergites 9, II and 13 with broad triangular posterior projections, increasingly larger from T. 9 to T. 13; coxal pores 6-8, VaC absent, 15th legs without accessory apical claw; female gonopods with 2+2 short, cylindro-conical spur and a short claw with a small lateral denticle on the internal side. Male unknown. The new species differs from L. variegatus especially for the lower number of prosternal teeth, and for the shape of the apical claw of the female gonopods; female gonopods also shows a lower number of dorsolateral setae on the first article and a lower number of dorsomedian setae on the second article and on the apical claw.

Material examined Holotypus: I 9, Sardegna, Cagliari, Santadi, loc. Su Benatzu, Grotta "Pirosu", 631 Sa/Ca, m 270, 25.V1.1994, M. Mucedda leg., at the Museo di Zoologia dell'Universita di Roma "La Sapienza".

Description

Female. Body 20.0 mm long, 2.5 mrn broad at T. 10, tergites pale chestnut, head dark chestnut, legs slender and long, spines long. Head feebly wrinkled, broader than long, as broad as T. I, posterior border straight, median thickening of marginal ridge very distinct; antennae 14.5- 15.0 mm long, with 79-86 articles as broad as long, terminal article about two times longer than broad; ocelli (Fig. I) I+3 on each side, little, not contiguous each other, in the middle of a depigmented area, posterosuperior ocellus larger than the other ocelli, organ of Tomosvary (Fig. I) larger than ocelli; prosternum (Fig. 2) with 3+3 conical teeth, porodont spiniform, shoulders asim- metrical, absent on the right side, very feebly on the left side. A NEW LlTHOBIUS FROM A SARDINIAN CAVE 61

Uthobius nuragicus o. Sp.

(Figs. 1-4)

0.5 mm

Figs. 1-4 _ Lithobius nuragicus n. Sp .• HoJotypus. Head, lateral (1); prosternum, ventral (2); left female ganoped. ventral (3), dorsal (4). 62 MARZIO ZAPPAROLI

Tergites feebly wrinkled and long; T.I trapeziform, larger than T.3, with posterior angles rounded and posterior border feebly concave; lateral borders parallel in IT. 3, 5 and 7, feebly convergent to the posterior end and almost straight in IT. 8 and 10, feebly convergent to the posterior end and feebly ar- cuated in IT. 12 and 14; posterior angles rounded in IT. 3 and 5, angulated in IT. 7, 8, 10, 12 and 14; posterior border straight in T. 3, feebly concave in T. 5, concave in IT. 8 and 10, feebly emarginate in IT. 7, 12 and 14; intermediate tergites with lateral borders almost arcuate in IT. 2, 4 and 6, parallel in IT. 9, II and 13; posterior angles blunt in T. 2, angulated in IT. 4 and 6, with broad, short and increasingly larger triangular projections in IT. 9, II and 13, the triangular projections in T. 13 are longer and sharper than in T. 9 and T. 11; posterior border straight in all the intermediate tergites. Coxal pores 6, 7, 8, 7, the proximal pores are circular and little, the distal pores are oblong. Legs 14th and 15th are 9.5 mm and 13.0 mm long respectively, with femur, tibia and tarso with numerous glandular pores on the internal side, 15th legs without accessory apical claw. VaC absent, see Tab. 1 for spinulation. Female gonopods (Figs. 3-4) with 2+2 short, cylindro-conical spurs especially those on the external side, claw short, with a little denticle on the internal side; first article with a group of 10-I I dorsomedian setae proximal to the insertions of the spurs; second article with 4-5 dorsolateral setae in a line and one dorsomedian seta; apical claw with 4-5 shorter dorsolateral setae in a line.

Table I -lirhobius nuragicus n. sp. Spinulation of legs. C ::::coxa, t::::: trochanter, p::::prcfemur, F::::femur, T::::tibia, a::::anterior spine, m::::median spine, p::::posterior spine.

Ventral Dorsal legs C P P T C P F I T amp am am mp ap a 2 amp amp am amp ap ap 3 amp amp am amp ap ap 4 amp amp am amp ap ap 5 amp amp am amp ap ap 6 amp amp am amp ap ap 7 amp amp am amp ap ap 8 amp amp am amp ap ap 9 amp amp am amp ap ap 10 amp amp am amp ap ap II amp amp am amp 'p ap 12 m amp amp am a amp ap ap 13 m amp amp am a amp ap ap 14 m amp am am a amp ap ap IS m amp am a amp ap A NEW L1THOBIUS FROM A SARDINIAN CAVE 63

Derivatio nominis

After the culture of nuraghi, characteristic of Sardinian prehistory.

Affinities The new species is morphologically close to Lithobius variegatus s.l., an epigeic species distinct in two subspecies, L. v. variegatus Leach, 1814, known from S Scotland, England, Galles, Ireland, Channel Islands (introduced ?), Brittany, NW Spain and N Portugal, and L. v. rubriceps Newport, 1845, recorded in Central and Southern Iberian Peninsula, Mo- rocco, Tunisia, Sicily and Calabria (Eason, 1964; Eason & Minelli, 1976; Eason & Serra, 1986). The British populations of L. v. variegatus show pre- ference especially for mesophilous woodlands and seem not to tolerate low winter temperatures (Eason, 1964; Eason e Serra, 1986); the habitat prefe- rences of other populations of L. variegatus s.1. are unknown. L. variegatus s.1. and L. nuragicus n. sp. share the following morphological characters: the large general size, especially as regards the body lenght, between 16-24 mm in L. v. variegatus and up to 40 mm in L. v. rubriceps; the number of prosternal teeth, more than 2+2; the shape of T. I, trapezioidal; the presence of triangular projections in the posterior angles of TT. 9, II and 13; the absence of VaC; the absence of accessory apical claw on 15th legs; the general shape of the female gonopods with 2+2 spurs and no tridentate claw; the general arrangement of spinulation. The new species differs from L. variegatus s.1. by the presence of 3+3 prosternal teeth instead of 6+6 or 7+7; by the claw of the female gonopods with a distinct lateral denticle, instead of quite simple without denticula- tions as in L. v. rubriceps, or with small and irregular lateral denticle as in L. v. variegatus. L. nuragicu; n. sp. and L. variegatus s.1. also differ from one another in the arrangement of the setae of the female gonopods: in L. nuragicus n. sp. dorsolateral setae of the second article and those of the apical claw are arranged on a longitudinal line of 4-5 setae, whereas in L. variegatus S.1.dorsolateral setae of the same articles are arranged on a band of 2-3 lines and their number is very higher than in L. nuragicus n. sp. (about 20 on the second article); L. Iluragicus n. sp. also shows about 10-11 dorsomedian setae arranged in a small area of the first article proximal to the insertion of the spurs, instead of more numerous setae arranged on a larger area as in L. variegatus s. I. L. nuragicus n. sp. also differs from L. variegatus s.1. in characters, such as the lower number and different arrangement of ocelli, I+3 in the center of a depigmented area insetad of 13-20 in 4-5 irregular rows, and in 64 MARZIO ZAPPAROLI

the higher number of antennal articles, 79-86 instead of 35-52, obviusly as a result of morphological adaptations of the new species to the cave habitat. For differential characters of Lithobius variegatus Leach, 1814 s.l. and Lithobius nuragicus n. sp. see Tab. 2.

Table 2 - Differential characters of Lithobius variegatus Leach, 1814 s.l. and Lithobius nuragicus n. sp.

L v. varie~atus L v. rnbriceps L nUffl!?icus n. sp. body lenghl (mm) 16-24 (British Isles) 40-42 20 24-30 (fuerian Peninsula) body colour paJebrown unifonn dark bro'M1 unifonn pale brown with variegated markings (in preserved holotype) (in live specimens) n° ocelli 1+13-18 1+15-20 1+3 nO oroslema1 teeth 6+6,7+7 3+3 n° antenna.! articles 35-46 48-52 79-86 triangular projections small but distinct very sm.-tilor absent absent onT.7 glandular pores on 15th sparse or absent very numerous very numerous legs ISlh legslbody Icnghl m- ea.1> ca,l/3 ca. 1/2 tio n° coxal PJres 5.7 (British Isles) 7-10. occa<;ionally3-5 6-8 7-10 (Iberia Peninsula) claw of female gonopods with small or irregular simple, without 1aternl with a laternl denticle lateml denticles denticles dorsomedian setae ca. 15 ca. 10 Isl article of female gononOOs dorsolatcml setae ca 20 on a band of2-3 lines 4-5 on a line 2nJ article of female e:onopods dorsolateral setae ca. 10 on a band of2-3lines 4-5 on a line claws of fermle gonopods

Distribution and ecology

The new species is only know from the type locality, the Grotta "Pirosu", 631 Sa/Ca, Su Benatzu, Santadi (Sardinia, Cagliari province), in the Sulcis. The cave develops in Cambrian limestones and dolomias, it has a lenght of about 475 m and a difference of level of - 52 m. The cave has three entrances leading to a large main hall, from which some short galleri- es start; for a map of the cave see Maxia (1972). "Pirosu" cave was used during the nuragic age as a holy place and today, because of important ar- A NEW LlTHOBIUS FROM A SARDINIAN CAVE 65 chaeological findings, is one of the most important Sardinian prehistoric sites (Todde, 1972). Due to this, entrance to the cave is limited. The holotypus of L. nuragicus n. sp. has been collected among organic debris on a stalagmite (Grafitti pers. com.). In the cave a rich fauna is present, including also troglobitic species, with some Sardinian endemics (see Argano & Rampini, 1973; Puddu & Pirodda, 1974; Strasser, 1974; Mahnert, 1976; Gardini, 1980): Oxychilus sp. oppressus (Fisher & Studer, 1878) (Gastropoda Sylommatophora Zonitidae), troglophilic; Chthonius si- culus Beier, 1961 (Arachnida Pseudoscorpionida Chthoniidae), troglophi- lie; Roncus puddui Mahnert, 1976 (Arachnida Pseudoscorpionida Neobisii- dae), troglobitic, Sardinian endemic; Catalauniscus puddui Argano, 1973 (Crustacea Isopoda Trichoniscidae), troglobitic, Sardinian endemic; Bla- niulus eulophus Silvestri, 1903 (Diplopoda lulida Blaniulidae), troglophilic, Sardinian endemic; Paranchus albipes (Fabricius, 1796) (Insecta Coleopte- ra Carabidae), trogloxene; Catopomorphus orientalis Aube, 1850 (Insecta Coleoptera Cholevidae), troglophilic; Aglenus brunneus (Gyllenhal, 1813) (Insecta Coleoptera Othniidae); Nycteribia vexata Westwood, 1835 (Insecta Diptera Nycteribiidae), ectoparasite on Chiroptera.

ACKNOWLEDGMENTS

I wish to express my thanks to Giuseppe Grafitti, of the Gruppo Spcleologico Sassarese, for providing infonnations on the type locality of the new species described; Alessandro Mi- nelli, of the Dipartimento di Biologia dell'Univcrsita, Padova, for the useful suggestions and discussion; M. Mucedda, of the Gruppo Spe1eo-Archeologico "G. Spano", Cagliari, for collecting the holotype of the new species.

REFERENCES

ARGANO. R. & M. RAMPINI. 1973. Note sulla distribuzione dei Trichoniscidae in Sarde- gna. Int. J. Spelcol.5: 311-317. EASON, E.H. & A. MINELLI. 1976. The identity of the species of Lithobiidae described by F. Fanzago and G. Fedrizzi from 1874 to 1881 (Chilopoda, Lithobiomorpha). Fragm. entomol. 12: 183-205. EASON, E.H. & A. SERRA. 1986. On the geographical distribution of Lithobius variegatus Leach, 1814, and the identity of Lithobius rubriceps Newport, 1845 (Chilopoda: Litho- biomorpha).1. nat. Hisl. 20: 23-29. EASON, E.H. 1964. Centipedes of the British Isles. Warne, London, X+294 pp. GARDlNI, G. 1980. Catalogo degli Pseudoseorpioni cavernieoli italiani. Mem. Soc. entomol. ita!.. 58 (1979): 95-140. GRAFITII, G. & M. ZAPPAROLI. 1992. Note su alcune specie di Chilopodi eavernieoli di Sardegna (Chilopoda). Not. Cire. Speleol. Romano, n.s. 6-7: 121-130. MAHNERT, V. 1976. Pseudoscorpions des grottes de la Sardaigne. Fragm. entomo!. 12: 309-316. MAXIA, C. 1972. La grona di Sanradi, primo tempio nuragico scoperto in Sardegna. Speleol. Sarda 2: 3-9. 66 MARZIO ZAPPAROLI

MINELLI, A. 1982. Chilopodi e Diplopodi cavernicoli italiani. Lav. Soc. ital. biogeogr .. N.S. 7 (1978): 93.110. MINELLI, A. 1985. Catalogo dei Diplopodi e dei Chilopodi cavernicoli italiani. Mem. Mus. civ. St. nat. Verona (II serie), sez. biologica 4: I-50. NEGREA, S. & A. MINELLI. 1994. Chilopoda. In: Juberthie C., V. Decu (cds.), Encyclo. paedia biospeologica, Societe de Biospeologie. Mulis (CNRS), Bucarest (Academie Roumaine) 1: 249-254. PUDDU, S. & G. PIRODDA. 1974. Catalago sistematico rngionato della fauna cavernicola della Sardegna. Rend. Sem. Fac. Sci. Univ. Cagliari 43 (1973): 151-205. STRASSER, C. 1974. I Diplopodi Chilognati della Sardegna. Fragrn. entornol. 10: 231.293. TODDE. F. 1972. La scoperta della Grona Pirosu a Santadi. Speleol. Sarda 4: 18-31.

NOTICE TO CONTRIBUTIONS I. Papers should be submitted on 3.5" diskette plus two copies of the complete text of each article. Any word processor commonly used for Macintosh and PC is admitted. For long tables Excel should preferably be used. Submission of a paper will be taken to imply that is unpublished and is nol considered for publication elsewhere. 2. Papers should be written preferably in English. Other allowed languages are French. German. Italian and Spanish. Authors using a language not their own are urgently requested to have their manuscripts checked for linguistic correctness before submission. SI system should be used. Dates should be in the fonn «5 February 1975». 3. Papers should be headed by a tille, the namc(s) in full of authors{s) and an exact description of the post held and business address of the author(s). If more than one author, please underline the name of the person to whom correspondence and proofs should be sent. All papers should contain at least an English summary giving a synopsis of the paper with sufficient detailed informations. The English translation of the title must always be reported. 4. Each paper will be subject to editorial review by one or morc referees. The Editor reserve the right to refuse any manuscript submitted, whether on invitation or otherwise, and to make suggestions and modifications before publication. Submitted papers should be in a final form ready for publication. Correction to proofs should be restricted to preinter's and editorial errors only. Other that these, very substantial alterations may be charged to the author. 5. Bibliographical references should be listed in alphabetical order at the end of the paper. References should be in the following forms:

Article: GOURBAULT N. 1976. Recent karyological research on cave Planarians. Int. J. Speleol. 8: 69-74.

Book: JAKUS L. 1977. Morphogenetics of karst regions. Akademiai Kiadb, Budapest.

Chapter: NEI M.1976. Mathematical models of speciation and genetic distance: 726-765. In: KARLIN S. & NEVO E. (Eds.) - Population genetics and ccology. Academic Press, New York.

6. References should be cited in the text in parentheses, c.g. «(Jones, 1961») except when the author's name is part of the sentence, e.g. <