Morphology and reproduction of the cavefish Trichomycterus chaberti and the related epigean Trichomycterus cf. barbouri

Similar documents
PRODUCTION OF ELECTRICITY FROM A GEOTHERMAL SOURCE

Reflections on the drinking bowl 'Balance'

MECHANISMS OF ENVIRONMENTAL INCENTIVE REGULATION: WHY ECOLOGICAL POLICIES IN TRANSITION AND DEVELOPING COUNTRIES ARE NOT EFFECTIVE?

Structural Changes in Household Income Inequality in Korea: A New Decomposition Including Labor Supply Changes * Chulhee Lee

Experimental campaign on the overtopping of the seawall Malecòn Tradicional

Steady State Numerical Analysis of a Joule-Thompson Cryocooler for Cryosurgical Probe

Dimensionless Analysis for Regenerator Design

RISING FAMILY INCOME INEQUALITY IN THE UNITED STATES, : IMPACTS OF CHANGING LABOR SUPPLY, WAGES, AND FAMILY STRUCTURE.

Simultaneous Heat Integration and Batch Process Scheduling

THE BOILING OF THE REFRIGERANT R134a IN THE RECTANGULAR MICROCHANNELS OF THE CPU S COOLING SYSTEMS

Using tree-grammars for training set expansion in page classification

EVALUATION OF ALTERNATIVE CONFIGURATIONS OF A WATER-OIL HEAT EXCHANGER SYSTEM

Conventional Paper-I-2012

Overall stability of multi-span portal sheds at right-angles to the portal spans

Detection of Shallow Underground Buried Object Using Air Vibration Probe

Orthogonal Tipping in Conventional Offshore Stability Evaluations

Calculation of Theoretical Torque and Displacement in an Internal Gear Pump

Optimization Model of Oil-Volume Marking with Tilted Oil Tank

Numerical Simulation of Stresses in Thin-rimmed Spur Gears with Keyway B. Brůžek, E. Leidich

2 2D 2F. 1pc for each 20 m of wire. h (min. 45) h (min. 45) 3AC. see details J, E

Analysis of Elastic Lateral-Resistant Stiffness of Steel Plate Shear Wall

Physics Engineering PC 1431 Experiment P2 Heat Engine. Section B: Brief Theory (condensed from Serway & Jewett)

PULL-OUT AND SHEAR STRENGTH EQUATIONS FOR HEADED STUDS CONSIDERING EDGE DISTANCE

Exclusively for you. Your Perfect Vanilla Solution. Our Vanilla Powders. Our International Innovation Platform. Vanilla & Spice AND ALL THINGS NICE

234 The National Strategies Secondary Mathematics exemplification: Y7

Response of Snap Bean Genotypes to Rhizobium Inoculation and Nitrogen Fertilizer under Different Agroecological Zones of Ethiopia

Reinforcement in concrete piles embedded in sand

Planck + Einstein on -Day Fundamental Physical Constants in a Relativistic Perspective and The Design of a Black Hole Gun!

20.1 Heights and distances

To find the volume of a pyramid and of a cone

Ground Improvement Using Preloading with Prefabricated Vertical Drains

Math Practice Use a Formula

Red Green Black Trees: Extension to Red Black Trees

Dormice Glis glis activity and hazelnut consumption

Revision Topic 12: Area and Volume Area of simple shapes

The influence of freeze drying conditions on microstructural changes of food products

On the performance of gyro-mass devices for displacement

16.1 Volume of Prisms and Cylinders

Applying Trigonometric Functions. ENTERTAINMENT The circus has arrived and the roustabouts must put

青藜苑教育 Example : Find te area of te following trapezium. 7cm 4.5cm cm To find te area, you add te parallel sides 7

Russell James Department of Scientific and Industrial Research Taupo-ldairakei, New Zealand

CO-ROTATING FULLY INTERMESHING TWIN-SCREW COMPOUNDING: ADVANCEMENTS FOR IMPROVED PERFORMANCE AND PRODUCTIVITY

Fixation effects: do they exist in design problem solving?

Applications. 38 Looking for Pythagoras. Find the missing length(s).

Nursilah Ahmad 1, Mohamad Yazis 1 & Mohammad Salem Oudat 1

Energy Efficiency Retrofit of Two-Flow Heat Exchanger System

4.2 Using Similar Shapes

A Note on Software Protection and Social Welfare

Prediction of steel plate deformation due to triangle heating using the inherent strain method

Balanced Binary Trees

Delay Analysis of Sampled-Data Systems in Hard RTOS

Perimeter, Area, and Circumference

CONFIGURATION OF AN UNMANNED GROUND EFFECT VEHICLE

VALIDATION OF SEISMIC DESIGN CRITERIA FOR CONCRETE FRAMES BASED ON MONTE CARLO SIMULATION AND FULL SCALE PSEUDODYNAMIC TESTS

Tunnel-soil-pile interaction in the prediction of vibration from underground railways: validation of the sub-models

Distributed rainfall runoff analysis in a flow regulated basin having multiple multi-purpose dams

Subsidies for resident passengers in air transport markets

Participatory Evaluation of Some Tomato Genotypes for Resistance to Bacterial Wilt

A Sustainable Energy Harvesting Machine

ANALYSIS OF WORK ROLL THERMAL BEHAVIOR FOR 1450MM HOT STRIP MILL WITH GENETIC ALGORITHM

Screening and breeding for resistance to leafminer (Liriomyza langei) in lettuce and spinach

Essential Question How can you find the surface area and the volume of a cone? 3 in. π

Study of microrelief influence on optical output coefficient of GaN-based LED

Effect of user fee on patient s welfare and efficiency in a two tier health care market

Biochemical Changes of Salt-Fermented Tuna Viscera (Dayok) and Its Effect on Histamine Content During Fermentation

Progress Towards a Micromachined Heat Exchanger for a Cryosurgical Probe

Effects of the Spudcan Penetration on the Adjacent Foundations

Research regarding the setting up of the Processing Directions of Peach New Cultivars and Hybrids

p x The revenue function is 5. What is the maximum vertical distance between the line

Math GPS. 2. Art projects include structures made with straws this week.

1/1 FULL SIZE 3/4 QUARTER SIZE 1/2 HALF SIZE EXTRA LARGE SIZE EXTRA LONG SIZE

International Plant Protection Convention Page 1 of 10

10. Consider the following problem: A box with an open top is to. 11. A farmer wants to fence an area of 1.5 million square feet in a

OD DVOSTRUKO ZASTAKLJENOG PROZORA DO DVOSTRUKE FASADE INDIKATORI PRENOSA TOPLOTE STACIONARNOG STANJA

1/1 FULL SIZE 3/4 QUARTER SIZE 1/2 HALF SIZE EXTRA LARGE SIZE EXTRA LONG SIZE

EVALUATION OF THE CHEMICO-OSMOTIC BEHAVIOUR OF A NATURAL BENTONITE

Volumes of Pyramids. Essential Question How can you find the volume of a pyramid?

Study of Steam Export Transients in a Combined Cycle Power Plant

Farm Structure Survey 2009/2010 Survey on agricultural production methods 2009/2010

APPROXIMATION OF A CATENARY FORMED OUT OF A ROPE ACCORDING TO THE HEIGHTS OF ITS POINTS OF HOLD

Analysing the energy consumption of air handling units by Hungarian and international methods

Description of Danish Practices in Retail Trade Statistics.

MODELING OF VENTILATION AIR HEAT RECOVERY AND ITS IMPACT IN HIGH-PERFORMANCE GREEN BUILDINGS

Goal: Measure the pump curve(s)

Variance Estimation of the Design Effect

Eugene O'Neill

The beginning of wine and viticulture

Harvesting by Peel Color to Reduce Bruising of Golden Delicious Apples

Measured Adiabatic Effectiveness and Heat Transfer for Blowing From the Tip of a Turbine Blade

Cup. for cup. Top espresso quality

Influence of the mass flow ratio water-air on the volumetric mass transfer coefficient in a cooling tower

Highway Transportation Efficiency Evaluation for Beijing-Tianjin-Hebei Region Based on Advanced DEA Model

Conjoint Analysis: A Study of Canned Coffee in Taiwan

Cup. for cup. Top espresso quality

Effect of Twisted-tape Inserts on Heat Transfer in a Tube

The Cohesion Tension theory of sap ascent: current controversies

Questions. denotes answer available in Student Solutions Manual/Study Guide; O denotes objective question

Point Pollution Sources Dimensioning

INFLUENCE OF OSMOTIC TREATMENT IN THE DRYING OF SULTANINA GRAPES (Vitis vinifera L.)

APPENDIX C2: Design of Canard Aircraft

Transcription:

Journal of Fi Biology (2003) 63, 490 505 doi:10.1046/j.1095-8649.2003.00171.x, aailable online at ttp://www.blakwell-ynergy.om Morpology and reprodution of te aefi Triomyteru aberti and te related epigean Triomyteru f. barbouri M. POUILLY* AND G. MIRANDA *Intitut de Reere pour le De eloppement (IRD) / Unierite Lyon 1, Laboratoire d Eologie de Hydroyte`me Fluiaux, Bd du 11 Noembre 1918, F-69622 Villeurbanne Cedex Frane and Intituto de Eologıá, Unieridad Mayor de San Andre, Boliia, CP 10077, La Paz Boliia (Reeied 19 Mar 2002, Aepted 10 June 2003) Hypogean and epigean population of Triomyteru atfie inabit tream from different enironment (ae, eadwater, anyon and alley) in te Torotoro National Park in te Ande, Boliia. A ignifiant redution in te diameter of te eye and in te urfae area of te meenepalon wa obered in ubterranean population, along wit an inreae in te urfae area of te telenepalon. Contrary to expetation, te barbel did not appear to be longer in ypogean population. Te obered pattern of modifiation of te oter ariable (pigmentation, eye aymmetry, urfae area of te erebellum and rombenepalon, feundity and egg diameter) orreponded to a gradient of alue from alley to anyon, eadwater and ubterranean population. Ti reult argue not for a imple ditintion between epigean and ypogean population but for an adaptation to an enironmental gradient of ontraint in wi ae orrepond to an extreme ituation. # 2003 Te Fierie Soiety of te Briti Ile Key word: brain; ae life; enironmental gradient; morpology; reprodution; Triomyteru. INTRODUCTION Triomyteru i a atfi genu (Triomyteridae) inluding. 100 120 peie ommonly found in neotropial eadwater tream (Eigenmann, 1918). Ti genu preent a ig olonization potential in ontraining enironment u a ig altitude tream (>4000 m, M. Pouilly per. ob.), warm termal water (>35 C, G. Miranda per. ob.) and ubterranean tream. Romero & Paulon (2001) reported on tree troglobiti (ae-retrited) peie: Triomyteru aberti, Durand in Boliia, T. onradi (Eigenmann) in Venezuela and T. itaarambieni Trajano & de Pinna in Brazil (region of Mina Gerai). Hypogean population (not neearily ae-retrited) ae been reported in te Colombian Ande (Sket, 1988) and in te Brazilian region of Goia (Trajano & Souza, 1994), and in oter neotropial region (Trajano, 1997). Autor to wom orrepondene ould be addreed. Tel.: 33 4 72446299; fax: 33 4 72431141; email: pouilly@uni-lyon1.fr # 2003 Te Fierie Soiety of te Briti Ile 490

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 491 Permanent darkne (Langeker, 2000) and food arity (Hu ppop, 2000) are te main limiting fator in ae, and troglobiti peie ae ad to be preadapted or to profoundly ange teir pyiology, biology and eology to adapt to tee extreme ondition. Some feature, u a eye and pigmentation, loe teir funtion beaue of darkne and onequently regre (Culer & Wilken, 2000). To ompenate for te lo of iion, troglobiti organim may deelop non-opti enory truture and appendage (organ tat are reeptie to emial and meanial timuli). Te enorial information i gatered in te brain and te ize of te different brain lobe proide an indiret indiation of te leel of atiity of ea type of enorial reeption (Kotral et al., 1998). Te relatie ize of te different brain lobe may differ between epigean and troglobiti fie (Poulon & Wite, 1969). Tee adaptation are eential to improe food-finding apaity and reprodutie ontat in te dark. In addition, te limited food upply in te ae enironment may lead to a redution in te metaboli rate and to a modifiation in bot te life yle and te reprodutie trategy to inreae reitane to taration (Culer, 1982; Hüppop, 2000; Trajano, 2001). Mot atfie ow beaioural (noturnal atiity and ryptobioti abitat) and morpologial arater (preene of barbel, mall eye and emo-enory orientation) tat ould be interpreted a pre-adaptation to ypogean life (Bertin, 1958; Romero & Paulon, 2001) and tat failitate olonization of ae. Triomyteru aberti i endemi to te Torotoro National Park, Boliia, and a only been obered in te Umajalanta Cae. Population of epigean Triomyteru f. barbouri (Eigenmann) are preent in te ame area. Tey are geograpially iolated from ea oter by natural waterfall in te tream tat flow in different anyon and alley. Te iolation of te tream make it poible to ondut a omparatie tudy of pylogenetially related population tat lie independently in different enironmental ondition (ae. uperfiial tream, anyon and eadwater roky tream. alley grael-bed tream). Te aim of te tudy wa to ompare morpologial (eye, barbel, pigmentation and brain arateriti) and reprodutie (feundity and egg diameter) feature tat are known to be affeted by ypogean life, in eigt population of Triomyteru from alley, anyon, eadwater and ubterranean tream in te Torotoro National Park. Ti omparion reealed wat kind of peialization te population of T. aberti underwent after olonization of te ae enironment, and weter tee peialization are peifially linked wit a ae enironment or rater wit a gradient of ange between te different enironment found in te different type of tream. MATERIALS AND METHODS STUDY AREA Torotoro National Park i loated in te Boliian Ande (18 15 0 S; 65 45 0 W) at an altitude of between 1950 and 3850 m. Te area orrepond to a mall maif (15 by 3 km) of Cretaeou limetone araterized by karti penomena (ae and anyon). Te loal ydrograpi network ow four prinipal roky tream loated in anyon parallel to ea oter and perpendiular to te alley of te Caine Stream into wi tey drain (Fig. 1). Tey ae a ig lope (. 10%) wit a teep profile, and are # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

492 M. POUILLY AND G. MIRANDA Boliia Stream flow 65 45 65 40 18 00 Caine (1950 m) V Torotoro National Park Suuuma V 18 05 Laguna C C Torotoro 3855 m Umajalanta ytem Torotoro illage H Rodeo Upper Umajalanta H Singani S S Torotoro National Park 10 5 0 10 km 18 10 Umajalanta ae FIG. 1. Torotoro National Park. Semati map owing te loation of te eigt ampled tream. V, alley tream; C, anyon tream; H, eadwater tream; S, ubterranean tream; waterfall. araterized by a erie of waterfall (>5 m ig) tat biologially iolate tream egment by preenting te migration of fie uptream. Te Caine i an untable grael-bed tream onneted to te Amazonian ydrograpi network (Mamore - Madeira watered). Loal limati ondition are emiarid wit. 950 mm of annual preipitation, a mean annual temperature of 18 C, and are araterized by te alternation of a warm wet eaon (Otober Mar) and a old dry eaon (April September). SAMPLING LOCATION AND TECHNIQUE Six uperfiial tream were ampled in tree different kind of enironment (Fig. 1). Te Caine and Suuuma tream are loated in te alley, at an altitude of 1980 and 2020 m, repetiely. Te Laguna and Torotoro tream are loated in anyon, at an altitude of 2200 and 2300 m, repetiely. Te Rodeo and Umajalanta tream, bot tributarie of te Torotoro anyon, are loated in mall eadwater alley, at an altitude of 2740 and 2820 m, repetiely. Te Umajalanta tream a a ubterranean egment of. 3 km and flow into te Torotoro anyon by mean of a 20 m waterfall. Te Upper Umajalanta i loated in te uperfiial upper reae of te Umajalanta tream before # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 493 te tream i lot in te Umajalanta ae. Two tream, te Umajalanta and Singani, were ampled in te ubterranean enironment of te Umajalanta ae. Tee are two ditint ubterranean tream tat onerge in te ae (altitude. 2700 m). Population were ampled on two date: Noember 1998 (early wet eaon) and July 1999 (dry eaon). Due to onern for oneration, and not aing any preiou information, a maximum of 30 indiidual per population on ea ampling oaion wa aptured, or een fewer if te denity wa low. Fie were olleted wit a bakpak eletrofiing gear deliering a urrent of between 300 and 600 V (Smit-Root In., ttp://www.mit-root.om). Eletrofiing i a atifatory metod for ti kind of abitat made up of roky tream a well a for tudie of fragile or mall population. Indiidual were preered in buffered formaldeyde (4%) before tranport to te laboratory. MORPHOLOGICAL MEASUREMENTS Four attribute were eleted among toe mot ommonly ued in tudie of fi adaptation to ae life (Ginet & Deou, 1977; Culer, 1982; Weber, 2000). Tey were meaured in een to 30 indiidual from ea population. Te diameter of te eye wa meaured orizontally and expreed a a ratio between te aerage of te diameter of te rigt and left eye and te tandard lengt (L S ) of te indiidual. Te aymmetry of eye wa etimated uing te ame meaurement and orreponded to te ratio of te diameter of te rigt and left eye. Te aerage lengt of te two maxillary barbel wa meaured from inert to tip and expreed a a ratio wit L S of te indiidual. Te pigmentation wa etimated by ounting under a miroope te number of romatopore on 1 mm 2 of tree body part: te rigt pot-epali region, and te region of te lateral line at te tip of te petoral and peli fin. Eye ize and pigmentation are known to preent regreion in ae peie. An inreae in eye aymmetry i alo a arateriti of ome ae peie (Wilken, 1988). An inreae in barbel lengt i onidered to be an adaptation to obtain more emial and meanial information (Weber, 2000; Wilken, 2001). Te urfae area of te four major ditintie part of te enepalon wa etimated for fie indiidual from ea population (exept from te Upper Umajalanta): telenepalon, meenepalon, erebellum and rombenepalon (Fig. 2; Nieuwenuy, 1982; Nalbant & Linare, 1987; Trajano, 1994). Te urfae area wa etimated uing te NIH Image free oftware (ttp://rb.info.ni.go/ni-image/) on a numerial potograp of te doral iew of te dieted enepalon. Compared to epigean fie, te main modifiation tat an be obered in te brain of troglobiti fie are a redution of te opti lobe (meenepalon), and an enlargement of te telenepalon and te erebellum (Poulon, 1963; Trajano, 1994). REPRODUCTION Te trend toward more K-eleted feature of life itory (or equilibrium trategy; Winemiller, 1989) i one of te main adaptation reported for ae peie (Culer, 1982; Weber, 2000). Feundity and egg diameter are often modified and ae fi peie tend to ae fewer but larger egg (Poulon, 1963; Hu ppop & Wilken, 1991). Feundity and egg diameter were determined uing mature female (tage V; Bagenal, 1978). After te extration of te intat gonad, all te egg were ounted to etimate feundity. Finally, te diameter of 30 randomly eleted egg wa meaured wit a mirometri ale under a miroope. DATA ANALYSIS Value of morpologial and reprodutie attribute were ompared between te eigt ampled population uing ANOVA. Only a mall number of indiidual wa ued to meaure te enepalon and te reprodutie attribute (Appendix), and ti ould bia ANOVA reult. In tee ae, te ANOVA tet were onfirmed by a non-parametri # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

494 M. POUILLY AND G. MIRANDA a b d FIG. 2. Doral iew of te dieted enepalon of a Triomyteru aberti from Umajalanta ae (120 mm L S ). Meaured brain part are indiated: a, telenepalon, b, meenepalon,, erebellum and d, rombenepalon. Krukal Walli tet. Wen population were ignifiantly different, a Tukey pair-wie tet wa performed to ee wi population differed from one anoter. All te tet were performed wit SYSTAT oftware (Wilkinon et al., 1992). A ritial leel of 5% (P < 005) wa ued in determining te ignifiane of te tet. MORPHOLOGY RESULTS Eye Signifiant differene exited between eye diameter in te eigt population [ANOVA, d.f. ¼ 7 and 169, P < 0001, Fig. 3(a)]. Pair-wie Tukey tet allowed tree group of population to be ditinguied: te population liing in te Umajalanta ytem (Umajalanta ae, Singani ae and Upper Umajalanta) ad a maller eye diameter; fie belonging to te Laguna population ad te larget eye diameter and te oter population repreented intermediate alue. Redution in eye diameter wa aymmetri witin population and, depending on te population, affeted one eye or te oter [Fig. 3(b)]. Fie from te Singani owed a redution in te diameter of te rigt eye, werea fie from te Umajalanta owed a redution in te ize of te left eye. In anyon and eadwater enironment, fie from te Upper Umajalanta, Laguna and Torotoro owed a redution in te ize of te left eye wile fie from te Rodeo owed a mall but ontant redution in te ize of te rigt eye. # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 495 Relatie maxillary barbel lengt Eye diameter : L S 4. 5 (a) 4. 0 3. 5 3. 0 1. 25 1. 20 1. 15 1. 10 2. 5 1. 05 1. 00 2. 0 0. 95 1. 5 0. 90 1. 0 0. 85 0. 20 0. 18 0. 16 0. 14 0. 12 () Umajalanta Singani Upper Umajalanta Rodeo Laguna Torotoro Suuuma Caine Site Number of romatopore mm 2 Rigt : left eye diameter 100 90 80 70 60 50 (b) (d) Umajalanta Singani Upper Umajalanta Rodeo Laguna Torotoro Suuuma Caine FIG. 3. Mean S.D. of (a) eye diameter : tandard lengt ratio, (b) eye diameter aymmetry (orizontal line repreent ymmetry), () relatie maxillary barbel lengt and (d) pigmentation for eigt Triomyteru pp. population from Torotoro National Park., ubterranean;, eadwater;, anyon;, alley population. See Appendix for te number of indiidual meaured. Population liing in alley tream diplayed ymmetri eye. Te leel of ariation appeared to be iger in ae population tan in epigean population (S.D. of032 for te Singani ae and 028 for te Umajalanta ae) and lower in alley population (S.D. of 008 for Suuuma and 009 for Caine). Canyon and eadwater population owed intermediate alue (ranging from 011 to 015). Barbel Tere wa a ignifiant differene in te lengt of maxillary barbel in te eigt population [ANOVA, d.f. ¼ 7 and 169, P < 0001). Tee differene, oweer, did not orrepond to an enironmental or patial pattern. Pair-wie Tukey tet allowed tree group of population to be ditinguied [Fig. 3()]. Te Laguna and Suuuma population ad te longet barbel and tee were # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

496 M. POUILLY AND G. MIRANDA ignifiantly different (Tukey tet probabilitie P < 005) from te Torotoro, Umajalanta ae and Caine population. Population from te Rodeo, te Singani ae and te Upper Umajalanta formed an intermediate group, wi only owed a ignifiant differene from Torotoro population. Pigmentation All eigt population diplayed ignifiant differene in te denity of romatopore [ANOVA, d.f. ¼ 7 and 169, P < 0001). Exepting te Singani ae population, te alue of pigmentation followed a gradient from ae to eadwater, anyon and alley population [Fig. 3(d)]. A pair-wie Tukey tet allowed tree group of tream to be ditinguied. Te Umajalanta ae and Rodeo population owed ignifiantly lower denitie of romatopore tan te tree population from te Caine, Suuuma and te Singani ae. Te oter population owed intermediate and igly ariable denitie of romatopore. Te iget denity of romatopore wa obered in te Singani ae population. In addition, unoloured indiidual were obered during ampling, a mentioned in te original deription of te T. aberti (Durand, 1968), but ti wa reerible if te indiidual were expoed to unligt for a few minute. In ti ae, ligtening eem to be produed by te onentration of melanin in te entral part of romati ell. Enepalon Tere wa no differene between te total urfae area of te enepalon in te population (ANOVA, d.f. ¼ 6 and 28, P ¼ 0397; Krukal Walli tet, KW, d.f. ¼ 6, P ¼ 0511). Te telenepalon (ANOVA, d.f. ¼ 6 and 28, P < 0001; KW, d.f. ¼ 6, P ¼ 0001), meenepalon (ANOVA, d.f. ¼ 6 and 28, P < 0001; KW, d.f. ¼ 6, P ¼ 0004) and erebellum (ANOVA, d.f. ¼ 6 and 28, P ¼ 0005; KW, d.f. ¼ 6, P ¼ 0006) preented different relatie urfae area between population. Tukey tet owed tat tere wa a ignifiant differene in te telenepalon and meenepalon between te ae population (Umajalanta and Singani) and te epigean population [Fig. 4(a), (b)]. Te Tukey tet performed on te erebellum did not reeal ignifiant differene between te population, but a trend of delining alue wa obered from ypogean to eadwater, anyon and alley population [Fig. 4()]. Tere wa no ignifiant differene in te urfae area of te rombenepalon between te population (ANOVA, d.f. ¼ 6 and 28, P ¼ 0063; KW, d.f. ¼ 6, P ¼ 0065), but, a in te ae of te erebellum, a trend of inreaing alue wa obered toward te epigean population from te alley [Fig. 4(d)]. REPRODUCTION Due to te mall number of indiidual aptured and te limited number of ampling oaion, te tudy failed to proide alid alue for reprodution ariable u a eaonality, ex ratio and ize at te onet of maturity. Te minimum ize reorded for a female tat ad tarted a maturity yle (tage II, Bagenal, 1978), ranged from 33 to 47 mm L S in bot epigean and ypogean population. Exept for te Singani and Suuuma population, te perentage of mature female appeared iger during te dry eaon (from 30 to 100% in # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 497 Telenepalon relatie urfae area Cerebellum relatie urfae area 0. 40 0. 35 0. 30 0. 25 0. 20 0. 40 0. 35 0. 30 0. 25 0. 20 (a) () Umajalanta Singani Rodeo Laguna Torotoro Suuuma Caine Site Meenepalon relatie urfae area Rombenepalon relatie urfae area 0. 30 0. 25 0. 20 0. 15 0. 10 0. 30 0. 25 0. 20 0. 15 0. 10 (b) (d) Umajalanta Singani Rodeo Laguna Torotoro Suuuma Caine FIG. 4. Mean S.D. relatie urfae area of (a) telenepalon, (b) meenepalon, () erebellum and (d) rombenepalon for een Triomyteru pp. population from Torotoro National Park., ubterranean;, eadwater;, anyon;, alley population. See Appendix for te number of indiidual meaured. July 1999) tan at te beginning of te wet eaon (from 0 to 66% in Noember 1998). Ti may orrepond to an inreae in reprodutie atiity during te dry eaon (July). It wa impoible to undertake analyi of mature female in te Torotoro and Umajalanta ae population, beaue of te low number of mature indiidual (two and one, repetiely). Feundity (ANOVA, d.f. ¼ 5 and 32, P ¼ 0027; KW, d.f. ¼ 5, P ¼ 0011) and egg diameter (ANOVA, d.f. ¼ 5 and 32, P < 0001; KW, d.f. ¼ 5, P < 0001) differed ignifiantly in te ix population. Feundity alue owed a trend to an inreae in te number of egg from ubterranean (Singani) to eadwater (Rodeo, Upper Umajalanta), to anyon (Laguna) and to alley population (Caine, Suuuma) [Fig. 5(a)]. Conerely, a redution in egg ize wa obered from Singani and Upper Umajalanta population to Laguna population, to Rodeo and to alley population (Caine, # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

498 M. POUILLY AND G. MIRANDA Feundity (mean total number of egg per female) 300 200 100 0 (a) 1. 8 (b) 1. 6 Mean egg diameter (mm) 1. 4 1. 2 1. 0 0. 8 0. 6 Singani Upper Umajalanta Rodeo Laguna Suuuma Caine Site FIG. 5. Mean S.D. (a) feundity and (b) egg diameter of 38 mature female from ix Triomyteru pp. population from Torotoro National Park., ubterranean;, eadwater;, anyon;, alley population. See Appendix for te number of indiidual meaured. Suuuma). Te Tukey tet owed a ignifiant differene between two extreme group: te Singani and Upper Umajalanta population ad te larget egg wile Rodeo, Suuuma and Caine ad te mallet [Fig. 5(b)]. DISCUSSION A redution in eye diameter i te mot obiou morpologial feature of ae-retrited animal (Eigenmann, 1909; Culer, 1982; Weber, 2000). Te eye # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 499 regreion proe begin by a redution in te ize of te opti apparatu (eye and opti lobe in te brain) tat reult in partial or total blindne. Depending on te lengt of iolation of te population, te regreion proe an be een to eole to total lo of te opti apparatu (anoptalmy) and, in te ae of fie, to profound modifiation in te truture of te kull (Breder, 1944; Wilken et al., 1989). Eye of ypogean Triomyteru from Umajalanta and Singani tream were till preent but owed a ignifiant redution in diameter in omparion to epigean population. Moreoer, ti redution in eye diameter wa aoiated wit a redution in te urfae area of te meenepalon, learly indiating te firt tep in a regreion proe a a reult of te dereae in optial atiity. In addition, rigt : left aymmetry wa iger and more ariable in ypogean tan in epigean Triomyteru. Wilken (2001) obered te ame pattern for te blind Pimelodidae, Ramdia reddelli Miller, and noted tat ti penomenon i orrelated wit te proe of ontogeneti regreion. Symmetry i likely to be an important feature for te effiient funtioning of a paired organ, tu aymmetry may be onidered a an indiation of te beginning of a regreion proe due to lo of funtion. In anoter way, te flutuating aymmetry ypotei (FA) aume tat te differene in te expreion of a arater between te rigt and left ide of a ymmetri organim proide a meaure of deelopmental intability (Palmer & Strobek, 1986; Paron, 1992). Ti meaure ould be indiatie of a tre tat an reult from internal geneti or external enironmental fator. In te preent tudy, te obered pattern of eye diameter owed inreaed aymmetry from alley population to anyon and eadwater population and finally to ypogean population. Aording to te FA, it an be aumed tat te aymmetry of te eye i not only due to te ontrat between epigean and ypogean enironment but tat it i alo te onequene of a gradient of enironmental ontraint. A ompenation for te lo of iion, ae animal may enane teir emial and meani enory pereption. Triomyteru aberti did not ow any eidene of inreaed barbel lengt a an adaptation to ae life, a a been obered in two peie of blind Pimelodidae, Ramdia reddelli Miller and te blind form of Ramdia latiauda (Kner) (Wilken, 2001). Triomyteridae, like te majority of iluriform fie, poe well deeloped barbel, wi i onidered to be a pre-adaptation to ae life (Bertin, 1958; Weber, 2000). One reaon for te lak of pattern in barbel lengt ould be tat in te ae of te Triomyteru, tee organ are already uffiiently long to utain an inreae in atiity. Te relatie ize of a periperal organ i poitiely orrelated wit te ize of te related brain entre, and an gie an indiation of te relatie importane of a partiular enory apaity (Kotral et al., 1998). Boliian Triomyteru owed imilar brain truture to Triomyteru guianeni (Eigenmann) from Venezuela (Nalbant & Linare, 1987) exept for te meenepalon wi appeared to be redued in te preent oberation. Beide te redution in te ize of te opti lobe (meenepalon), T. aberti owed a marked enlargement of te teleenpalon in omparion to te epigean T. f. barbouri. Similar reult ae been obered for eeral troglobiti fie and interpreted a an improement in olfation, topograpial memory and oial beaiour # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

500 M. POUILLY AND G. MIRANDA (Trajano, 1994). In te preent reult, tere wa a marked differene in te meenepalon and te telenepalon between epigean and ypogean population, altoug tere wa only a gradual modifiation in te rombenepalon and erebellum along te enironmental gradient. Tu a diret enory ompenation for lo of iion by te ene aoiated wit te telenepalon (mainly olfation and topograpi memory) may be ypoteized in te ae of T. aberti, ompleted by a gradual enlargement of te erebellum (omplex integratie entre for equilibrium and meano-enitiity; Poulon, 1963). Altoug not ignifiant, te dereaing ize of te rombenepalon toward te ypogean population, aoiated wit te obered tability in barbel lengt may indiate tat tate (one of te ene aoiated wit ti organ; Nieuwenuy, 1982) wa not improed in te ae of T. aberti. Te eond mot obiou morpologial feature aoiated wit ae animal i te redution in melani pigmentation (Ginet & Deou, 1977; Culer, 1982; Weber, 2000). In Triomyteru population from Torotoro, alue for te denity of romatopore follow a gradient of dereaing alue in fie from alley tream to anyon, eadwater and ubterranean tream. Beaue of ti gradient, and beaue of te ig leel of pigmentation of te ypogean population of T. aberti inabiting te Singani tream [Fig. 3(d)], te redution in pigmentation annot only be explained by te ontrat between ypogean and uperfiial enironment. Crypti olouration in fie i interpreted a a protetion againt iually orientated predation (Endler, 1980). Triomyteru population of alley tream are part of a multipeifi aemblage, and ould tu be ubjet to predation by aquati, aerial and terretrial predator. Triomyteru population in mountain tream (anyon and eadwater) are in a monopeifi ituation and are a priori only ubjet to predation by aerial and terretrial predator. Triomyteru aberti population are in a monopeifi ituation wit no iual predator. Tu, a lo of pigmentation aoiated wit a dereae in predation may be ypoteized at te ame time a a onequene of te abene of ligt. For logiti reaon, te tudy of reprodution ould not be onduted troug a omplete eaonal yle, limiting te interpretation of temporal ariation in reprodutie atiity. In ontrat wit oberation for te ae peie T. itaarambieni, woe reprodutie atiity our during te rainy eaon (Trajano, 1997), in Torotoro population, more mature female were obered during te dry eaon. Cange in te alue of feundity and egg ize reulted in a ange in te reprodutie trategy of te population, wi ould be linked to te enironmental ondition of te tream. Te population of T. aberti in te Umajalanta ae laid larger but fewer egg tan epigean T. f. barbouri population. A imilar differene in egg ize wa found between epigean and ypogean population of amblyopid fie (Poulon, 1963). Ti trend araterie a K-trategy in te r-k ontinuum. A imilar trend toward K-eleted feature a been deribed for te troglobiti Amblyopidae (Poulon, 1963), Pimelodella kronei (Miranda-Ribeiro), (Trajano, 1991) and T. itaarambieni from Brazil (Trajano, 1997) baed on growt, longeity and frequeny of reprodution. In teir analyi of 216 Nort Amerian fi peie, Winemiller & Roe (1992) onidered tat te ae fie Amblyopi roae (Eigenmann) and Amplyopi # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 501 pelaea DeKay ontitute te extreme form of te equilibrium trategy, equialent to te K-trategy (Winemiller, 1989). Te trend toward K-trategy feature of reprodution in te troglobiti peie ould be interpreted a an adaptation to food arity in a ypogean enironment (Culer, 1982; Hu ppop, 2000). Te inreae in egg ize ould proide better autonomy at te beginning of te laral tage and ould ubequently be faourable for te deelopment of te indiidual in tream wit a low food upply by proiding iger reitane to taration and iger mobility for food foraging (Poulon & Wite, 1969). In te preent tudy, eye diameter and brain truture preented alue tat oppoed epigean and ypogean population. Eye regreion learly appear a a ditintie arater of troglobiti organim tat may inole ompenation by oter enorial ytem (emial and meanial). Te obered pattern of modifiation of te oter ariable (pigmentation, eye aymmetry, feundity and egg diameter), oweer, orreponded to a gradient of alue from alley to anyon, eadwater and ubterranean tream. Speialization of T. aberti to ae life appear to be te reult of two kind of enironmental ontraint, a ditint one between epigean and ypogean abitat (ligt and darkne) and a gradual one tat may reult from te ombination of multiple fator (e.g. predation and food upply), in wi ae orrepond to an extreme ituation. Ti work wa part of te BIOBAB projet (aquati biodierity in te Boliian Amazon bain) deeloped by IRD, te Intituto de Eologı a from La Paz Unierity (Unieridad Mayor de San Andre ) and te Centro de Inetigaio n de lo Reuro Auátio of Trinidad Unierity (Unieridad Te nia del Beni). We tank tee intitution for teir upport. G. Miranda wa a tudent at La Paz Unierity and wa upported by a grant from te Intituto de Eologı a of La Paz Unierity. L. Arteaga, G. Alarez, D. Kopp, M. Flore and C. Iban ez from La Paz Unierity elped wit logiti, fieldwork and te meaurement of peimen. We would alo like to tank M. Jaldin and R. Beerra de la Roa from te Torotoro National Park. We tank B. Hugueny (IRD), F. Herant and J. Matieu (Unierité Lyon 1) for teir reiew of te manuript. Referene Bagenal, T. B. (1978). Apet of fi feundity. In Metod for te Aement of Frewater Fi Prodution (Gerking, S. D., ed.), pp. 75 110. Oxford: Blakwell. Bertin, L. (1958). Eologie. Poion aerniole. In Traite de Zoologie, Vol. 13 (Grae, P. P., ed.), pp. 1918 1923, 2660 2662. Pari: Maon. Breder, C. M. J. (1944). Oular Anatomy and ligt enitiity tudie on te blind fi from Cuea de lo Sabino, Mexio. Zoologia 29, 131 143. Culer, D. C. (1982). Cae Life: Eolution and Eology. Cambridge, MA: Harard Unierity Pre. Culer, D. C. & Wilken, H. (2000). Critial reiew of te releant teorie of te eolution of ubterranean animal. In Subterranean Eoytem (Wilken, H., Culer, D. C. & Humprey, W. F., ed), pp. 381 398. Amterdam: Eleier. Durand, J. P. (1968). Etude de poion re olte dan la grotte de Umayalanta (Boliie), Triomyteru aberti p. n. Annale de Spe le ologie 23, 343 353. Eigenmann, C. H. (1909). Cae ertebrate of Ameria, a tudy in degeneratie eolution. Carnegie Intitute of Waington Publiation 104, 1 241. # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

502 M. POUILLY AND G. MIRANDA Eigenmann, C. H. (1918). Te Pygiidae, a family of Sout Amerian atfi. Memoir of te Carnegie Mueum 7, 259 398. Endler, J. A. (1980). Natural eletion on olour pattern in Poeilia retiulata. Eolution 34, 76 91. Ginet, R. & Deou, V. (1977). Initiation a` la biologie et a` l e ologie outerraine. Pari: Edition Unieritaire Jean-Pierre Delarge. Hu ppop, K. (2000). How do ae animal ope wit te food arity in ae? In Subterranean Eoytem. Eoytem of te World (Wilken, H., Culer, D. C. & Humprey, W. F., ed), pp. 159 184. Amterdam: Eleier. Hu ppop, K. & Wilken, H. (1991). Bigger egg in ubterranean Atyanax faiatu (Caraidae, Pie). Zeitrift Fur Zoologie Sytematik und Eolutionforung 29, 280 288. Kotral, K., Van Staaden, M. J., Huber, R. (1998). Fi brain: eolution and enironmental relationip. Reiew in Fi Biology and Fierie 8, 373 408. Langeker, T. G. (2000). Te effet of ontinuou darkne on ae eology and aerniolou eolution. In Subterranean Eoytem (Wilken, H., Culer, D. C. & Humprey, W. F., ed), pp. 135 158. Amterdam: Eleier. Nalbant, T. T. & Linare, O. (1987). A tudy of a ubterranean population of Triomyteru guianee (Eigenmann, 1909) from Venezuela (Pie, Siluriforme, Triomyteridae). In Fauna ipogea y emieda fia de Venezuela y de otro paie de Ame ria del Sur 1 (Deu, V., Orgidan, T., Danau, D., Bordon, C., Linare, O., Urbani, F., Trononi, J. & Boque, C., ed), pp. 211 217. Buureti: Editura Aademiei Republiii Soialite Romaˆnia. Nieuwenuy, R. (1982). An oeriew of te organization of te brain of Atinopterygian fie. Amerian Zoologit 22, 287 310. Palmer, A. R. & Strobek, C. (1986). Flutuating aymmetry: meaurement, analyi, pattern. Annual Reiew of Eology and Sytemati 17, 391 421. Paron, P. A. (1992). Flutuating aymmetry: a biologial monitor of enironmental and genomi tre. Heredity 68, 361 364. Poulon, T. L. (1963). Cae adaptation in amblyopid fie. Amerian Midland Naturalit 70, 257 290. Poulon, T. L. & Wite, W. B. (1969). Te ae enironment. Siene 165, 971 981. Romero, A. & Paulon, K. M. (2001). It a wonderful ypogean life: a guide to te troglomorpi fie of te world. Enironmental Biology of Fie 62, 13 41. Sket, B. (1988). Speleobiologial inetigation in te Colombian Ande 1984. Bioloky Vetrik 36, 53 61. Trajano, E. (1991). Populational eology of Pimelodella kronei, troglobiti atfi from outeatern Brazil (Siluriforme, Pimelodidae). Enironmental Biology of Fie 30, 407 421. Trajano, E. (1994). Comparatie tudy of te brain and olfatory organ of te troglobiti atfi Pimelodella kronei (Ribeiro 1907), and it putatie anetor, P. tranitoria (Ribeiro 1912) (Siluriforme, Pimelodidae). Tropial Zoology 7, 145 160. Trajano, E. (1997). Food and reprodution of Triomyteru itaarambieni, ae atfi from out-eatern Brazil. Journal of Fi Biology 51, 53 63. Trajano, E. (2001). Eology of ubterranean fie: an oeriew. Enironmental Biology of Fie 62, 133 160. Trajano, E. & Souza, A. M. (1994). Te beaior of Anitru ryptoptalmu, an armored blind atfi from ae of entral Brazil, wit note on yntopi Triomyteru p. (Siluriforme, Loriariidae, Triomyteridae). Me moire de Biope ologie 21, 151 159. Weber, A. (2000). Fi and ampibia. In Subterranean Eoytem (Wilken, H., Culer, D. C. & Humprey, W. F., ed), pp. 109 132. Amterdam: Eleier. Wilken, H. (1988). Eolution and geneti of epigean and ae Atyanax faiatu (Caraidae, Pie). Support for te neutral mutation teory. In Eolutionary Biology, Vol. 23 (Het, M. K. & Wallae, B., ed), pp. 271 367. New York: Plenum. Wilken, H. (2001). Conergent adaptation to ae life in te Ramdia latiauda atfi group (Pimelodidae, Teleotei). Enironmental Biology of Fie 62, 251 261. # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 503 Wilken, H., Streker, U., Yager, J. (1989). Eye redution and pylogeneti age in opidiiform ae fi. Zeitrift Fur Zoologie Sytematik und Eolutionforung 27, 126 134. Wilkinon, L., Hill, M. A., Vang, E. (1992). Sytat Software: Statiti, Verion 5.2. Eanton, IL: Sytat In. Winemiller, K. O. (1989). Pattern of ariation in life itory among Sout Amerian fie in eaonal enironment. Oeologia 81, 225 241. Winemiller, K. O. & Roe, K. A. (1992). Pattern of life itory dierifiation in Nort Amerian fie: impliation for population regulation. Canadian Journal of Fierie and Aquati Siene 49, 2196 2218. # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

504 M. POUILLY AND G. MIRANDA APPENDIX. Summary of data of number of indiidual, tandard lengt range and mean S.D. of morpologial, brain and reprodutie attribute for eigt population of Triomyteru pp. from Torotoro National Park, Boliia (ee Fig. 1). Morpology Enironment Stream n L S range (mm) Eye diameter : L S Eye aymmetry 1 Barbel lengt : L S Pigmentation 2 Valley Caine 30 339 682 234 032 099 010 015 002 8281 3248 Valley Suuuma 30 343 659 240 030 099 008 018 003 8462 3276 Canyon Laguna 20 302 816 382 162 107 014 018 002 7480 4450 Canyon Torotoro 30 311 816 214 032 108 015 013 003 7698 2508 Headwater Rodeo 30 326 754 239 035 097 011 016 004 5484 2910 Headwater Upper Umajalanta 7 680 875 159 022 107 014 017 001 5853 512 Cae Umajalanta Cae 15 415 1037 127 050 115 028 015 002 4655 1396 Cae Singani 15 417 828 161 066 096 032 017 003 8823 3671 1 Rigt eye diameter : left eye diameter. 2 Number of romatopore mm 2. Brain Relatie urfae (%) Enironment Stream n L S range (mm) Telenepalon Meenepalon Cerebellum Hombenepalon Valley Caine 5 339 682 031 002 022 002 029 002 018 001 Valey Suuuma 5 414 659 032 002 020 002 028 001 020 001 Canyon Laguna 5 453 660 030 004 024 002 029 003 017 002 Canyon Torotoro 5 437 816 031 003 021 002 031 002 017 003 Headwater Rodeo 5 326 600 029 002 024 003 030 003 017 002 Cae Umajalanta Cae 5 415 939 036 001 014 002 033 001 017 001 Cae Singani 5 417 828 038 003 013 002 033 002 015 001 # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505

EPIGEAN AND HYPOGEAN TRICHOMYCTERUS 505 Reprodution Enironment Stream n 1 LS range (mm) Feundity 2 Egg diameter 3 Valley Caine 7 610 709 23129 19276 084 023 Valley Suuuma 3 447 617 19933 17473 080 044 Canyon Laguna 4 587 778 14750 3679 125 031 Headwater Rodeo 9 373 471 7900 3524 087 012 Headwater Upper Umajalanta 11 539 976 11164 4169 149 027 Cae Umajalanta Cae 4 440 688 4225 1837 152 026 1 Mature female of tage V (Bagenal, 1978). 2 Total number of egg per female. 3 Etimation from 30 egg per female. # 2003 Te Fierie Soiety of te Briti Ile, Journal of Fi Biology 2003, 63, 490 505