ebook img

Morphological variations in relation to maturation in Pantala flavescens (Fabricius) in central Japan (Anisoptera: Libellulidae) PDF

2007·2.7 MB·English
by  IshizawaN
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Morphological variations in relation to maturation in Pantala flavescens (Fabricius) in central Japan (Anisoptera: Libellulidae)

Odonalologica36(2):147-157 JuneI.2007 Morphologicalvariations inrelation to maturationinPantala flavescens (Fabricius) incentral Japan (Anisoptera: Libellulidae) N.Ishizawa 1644-15Yamaguchi,Tokorozawa City,SaitamaPref.,359-1145,Japan [email protected] ReceivedFebruary24,2006/ RevisedandAcceptedSeptember6,2006 P.flavescens wasinvestigatedin openfields in adeciduous forest in aninland partoftheKanto regionfor 3months fromlate June2003. The sp.was estimated tobebivoltine from summertolate autumn.Thesizeoftheadults wasunchanged throughouttheseason.Thesexratioofthepopulationskewedtowards 9.Maturity degree(MD),shownasthevalue ofbodyweightdivided bythecubeofwinglength, shifted upwardsuntil thesecond halfofAugust,afterwhichitdecreased sharply. Similarlythewingloading(WL)(calculatedbydividingbodyweightbywingarea) increased until the second half of August,and decreased fromSeptember,and in early October itwas not significantlydifferent between S3 and 9$. Asthe relationship of body temperatureto ambient temperatureshowed no difference betweenmature andimmature individuals,orbetween sexes,with bothcorrelation coefficients andregressioncoefficients beinglargeforaflyertypesp., theyseemed tobeeasily affectedbythe ambienttemperature. INTRODUCTION Pantalaflavescens is the most common dragonfly onthe earthandis famous for its migratorybehaviour. InJapan it appearsinmass swarms in thehot sea- son of Bon Festivalinsummer, andis calledbon-tomboorshoryo tombo.This dragonfly flies over the ocean from the South, and its migration has beenob- served fromaweathership (ASAHINA &TSURUOKA, 1967;HASHIMOTO & ASAHINA, 1969).ISHIDA (1989) observedthespecies flying fromover the ocean atTobaCity inMiePrefecture, andWAKANA(1959) describedprecisely his observationsoflarge swarms in early autumn. Itis saidthatcoming over to the mainlandof Japan itgoes Northby repeated alterationof generations and 148 N. Ishizawa thatits eggsand larvaecannot overwinteronthemainland.Hencethey are ex- tinctthere as abreedingpopulation (ISHIDA etal., 1989). Although manyreportson migrationandforaging ofthisspecies havebeenmade, the seasonalchanges ofthe degreeof maturity,wing loading, thermoregulation and seasonalchanges of maturity in the species havescarcely receivedany at- tention.Thisstudy was madeto clarifythe morphological variationsin relation to maturationofthe species coming overtoopen fieldsin aforest far fromthe water areas. STUDY SITESAND METHODS Studiesweremade fromlate JunetoOctoberin 2003 atKunugiyamaforest(ca 152ha,35°50’N, 139°28’E,60ma.s.1.)ontheborders ofthreecities andatown,Tokorozawa,Sayama, Kawagoeand Miyoshi,which arelocatedontheMusashino lowland hillsin thewesternSaitama Prefecture.Asthe inland forestislocated morethan 1km fromthenearestwaterbodies,the specimenscollected there mightberegardedasmigrantsfromotherregions.Supplementaryinvestigationsweremadeinearly SeptemberattheKabazaka Pass (810ma.s.1.)intheOkumusashi Hills,westernSaitamaPrefecture, and in late JulyatHino (680m a.s.1.),Nagasaka-cho,Kitakoma-gun,Yamanashi Prefecture,about 90km West ofKunugiyamaforest. Size and body weightweremeasured within six hours ofcapture.The hindwingwasused asa measureofsize,usingaslide caliper,and thebody weight(exceptinthespecimens fromNagasaka- cho)wasmeasured with anelectronicbalance to0.001 g. Degreeofmaturity(MD)wascalculatedbydividingthebodyweightbythecube ofthehindwing length,and this alsogivesthe specificgravityofaspecimen.Thevolume ofanadultisunchanged throughoutitslifeand,asitmatures, itsweightincreases. Accordingly, theweightdividedbythevol- ume,i.e. the specificgravity,mayberegardedasdegreeofmaturity.Thisvalue wasused effectively forSympetrumfrequensbyISHIZAWA(2004,2005), Wingloading(WL)wascalculated asfollows; fromthecapturedindividuals amaleandafemale, ofbodyweights340mgand431 mg,respectively wereselected asstandards.Theirwingswerephoto- copied.Usinganimagesizeof200%,thewingshapes ofthephotocopieswerecutoutandweighed. Theratio ofthe weightofeachofthese shapestothe weightofa 100cm2cut outwascalculated. This gaveavalue of 16.3cm2 forthe maleand 18.8cm2forthe female,and thesewereregardedas thestandardwingareas(STWA).Wingareasofotherspecimenswerecalculated asfollows:STWA (cm2)x thesquare(cm2)ofeach hindwinglength- thesquare(cm2)ofthestandardhindwinglength. TheneachWLwascalculated bydividingthebodyweight(mg)bythisvalue(cm2). Theabove data onWLwerecomparedwith those ofOrthetrumalbistylumspeciosum,collected at OmoripondatIrumaCity,Saitama Prefecture in2000 and2001,and werealmost thesamesize as P.flavescens. Asanindex ofmaturityI examined how fartheabdominal segmentswerepruinescentfrom the 3rd segmentbackwards, and defined the number ofthe lastpruinescentsegmentasthedegreeof pruinescense(PD).Incaseswherehalfofthelastsegmentwaspruinescent,theimmediatelyanterior segmentnumber + 0.5 wasused asthePD. Bodytemperatures(Tb) oftheindividuals weremeasured by thesamemethod appliedformeas- uringthebody temperaturesofS.frequens(ISHIZAWA,1998a);Tbweremeasured to0.1°Cwith a HoskinsF-V-K-002thermocouple(diameter0.05mm)setinahypodermicneedle(externaldiameter 0.3ram),whichwasinsulated withpaper covering, andconnected toaJohn-Fluke 55double-chan- neled digitalthermometer. Themeasurementsweremadewithin about 10secondsafternettingthe dragonflies,byinsertingtheprobethroughthemesh ofthenettothecenterofthethorax, 1mm up Morphologicalvariationsin Pantalaflavescens 149 themesothoracic spiracle.Ambient temperaturesataheightof 1m in the sun(Ta)weremeasured with anOmronHC-100thermistorsoonafterthemeasurementsofTb. Toinvestigatethedifferenceofthermoregulationamongindividuals inrelation toMD,I dividedthe dragonfliesintotwogroups (regardlessoftheseason)ineachsexbyusingthe standardsofMD4.8 in the male and MD 5.4inthe female. Those lessthan thestandards werecalled immature;those abovethem werecalledmature. Data ofcoefficients ofother Odonata in Figure4wereobtained fromAnaxjuniusfromMAY (1976),A.parthenopejulius(Ishizawa,unpublished),A.guttatus(ISH1ZAWA,1998b),Anotogaster sieboldii (ISHIZAWA,2003), patrollingCordulia aeneaamurensis (Ishizawa,unpublished),Mnais costalis oforange wingtype(ISHIZAWA,2000),Orthetrumj.japonicum(Ishizawa,unpublished), O.albistylumspeciosum(Ishizawa,unpublished)and bothsexesofS.frequensduringthereproduc- tiveperiod(ISHIZAWA,1998a).The sexormaturityofA.juniuswasunknown. Dataonall other speciesexcept S.frequenswerefrommaturemales. Data ofthesize,body weight,MDand WL weretested usingat-test;sexratios weretestedus- ingthechi-squaretest.Tb and Taamonggroups ofthesexesorimmature andmature groupswere tested usingthe ANOVA-testandt-test.Correlation coefficientsofeachgroupwerecomparedusing thez-test. RESULTS Thefirstrecordintheforestwas on26thJune2003, whensix individualswere sighted flyingaroundover anopenfield; thelastrecordwas on20thOctober. Figure 1showsthechange ofthesex ratioofcapturedspecimens. Throughout theinvestigation periodthesexratiowassignificantly skewedtowardsthefemales (67.3%, P <0.05)except thatofthe specimens captured in lateJuly at Nagasa- ka-cho, which were skewed towardsthe males(73.3%, P <0.05,n= 11). Hindwinglength wasca40 mmandnotnoticeablydiffer- entbetweensexesthroughout theseason;40.2±1.2mm,n= 53in the maleand 40.6±1.5 mm, n = 109 in the female, respectively (P > 0.05; Tab. I).Comparing theearly sea- son (before September) and thelateseason(fromSeptem- beronwards), thedifference in length was not significant (P=0.46); inthemalesit was 40.4±1.1mm(n= 26) in the early season and 39.9±1.2 mm(n=27)inthelateseason Fig. 1.Sexratios ofPantalaflavescenscapturedat surveyed (P>0.05),although malesdo areas. 150 N.Ishizawa have apropensity to become smaller. Inthe females the size was 40.6±1,2mm (n = 60) in theearly seasonand40.7±1.8mm(n =49) inthelateseasonandthe differencewas not significant, althoughthe s.d. tendedtoincreaseinthe latter. Body weight ofthe malesincreasedfromthesecondhalfof July(332.6±51.1 mg,n= 19),andafterreaching apeak (401.0±18.0 mg,n=2)in thesecondhalf of August, it reached itslowestlevelin October(Tab. I).In late September one ratherheavy individual(398 mg)was captured, though mostof thoserecorded in thelateseason werelight(ca 200mg).InearlySeptember oneimmatureindi- vidualof 159mginweightwas captured. Thedifferenceinweight betweenboth seasonswas significant (P <0.001). Inthe females, body weight was sustainedatalevelof ca400mg untilthesec- ondhalfof August, whenit reached apeak of 446.8±53.8mg (n = 6).Itthen decreasedrapidly by 100 mg,reaching its lowestlevelof ca200mgin October. Thedifferencein thebody weight between the seasons was significant as inthe males(P <0.001). Theheaviestfemalespecimen (537 mg)was captured on July 16andonecaptured onSeptember28 weighed 533mg.Thesecondwas theonly onethatexceeded 500mgin thelate season. Thebody weights ofboth sexes in the second halfof October (228.0±37.5 mg, n= 3 in the maleand 246.7±37.0 mg,n= 15inthe female)weresimilartothosefromtheKabazakaPasscaptured inearly September. Table I MorphometricdataofPantalaflavescenscapturedatKunugiyamaforestand KabazakaPass- [Hind- wing:mm, Bodyweight:mg, Wingloading(WL): mg/(cm-cm). - Abbreviations: s: second,f: firsthalf,MD; maturity(thedivision ofthebodyweightbythe cubeofthehindwinglength),PD: pruinescens(thelastpruinescentsegmentnumber,and incaseofhalfofthe lastsegmentwaspru- inescent,the immediatelyanteriorsegmentNo.+0.5 was definedasPD). Firstline ofvalues males, secondfemales. Meanvalueswith standard deviation] Julys. Aug.f. Aug.s. Sept.f. Sept.s. Oct.f. KabazakaPass Number 19 4 2 13 n 3 4 36 17 6 23 12 15 7 Hindwing 40.4+1.1 40.511.0 39.410.8 39.910.8 40.411.1 38.311.6 40.210.6 40.5±1.1 40.611.3 40.811.1 40.811.1 40.213.1 40.911.2 40.911.0 Body weight 332.6±51.1 358.8120.5 401.0118.0 263.8151.5 321.8145.6 228.0137.5 230.8157.5 395.0±84,5 437.2147.5 446.8153.8 323.7175.4 339.91110.7 246.7137.0 262.4133.7 MD 5.0±0.7 5.410.2 6.610.7 4.210.9 4.911.0 4.111.0 3.610.8 6.011.2 6.510.4 6.610.8 4.811.1 5,211.5 3.610.5 3.810.5 WL 19.812.7 21.310.8 25.412.1 16.213.3 19.413.3 15.313.1 13.913.4 23.014.7 25.011.8 25.613.0 18.514.3 19.615.7 14.011.8 14.912.0 (range) (25.9-12.0) (22.3-20.5) (27.5-23.3) (20.9-9.4) (27.4-15.4) (18.9-11.4) (18.7-9.4) (29.4-13.7) (28.7-21.6) (30.5-21.0) (25.8-10.7) (28.0-9.1) (17.3-11.3) (17.2-10.9) PD 5.111.3 5.310.7 5.810.4 5.810.5 (n=19) (n=16) (n=6) (n=13) Morphologicalvariations inPantalaflavescens 151 As the hindwing length was unchanged throughout theseasonand the body weightdifferedbeforeand afterthe beginning ofSeptember, MDsynchronized naturallywiththevariationsinthebody weight. ThemeanMDwas4.8±1.0(range 2.4-7.4) inthemalesand5.4±1.4(range 2.4-8.0) inthefemales, respectively. The MDofbothsexes increased untilthe secondhalfof August, butdecreasedsig- nificantlyfromSeptember onwards,although a smallpeak appeared inthesec- ondhalfofSeptember (Tab. I, 6: P= 0.00753 9: P= 2.664E-10). Therangeof MDwas 2.4-7.4 inthemales and2.4-8.0inthefemales, respectively. Regarding theindividualsofwhichMDexceededthemeanvalueas sexually mature adults, thesharerate oftheseintheearlyseasonandinthelateseasonwas80.8%:14.8% inthe malesand 82.8%:28.6%in thefemales, respectively. Onlyonemalewithared abdominaldorsumwas captured during theseason, and ithada MDof7.3. One femalewithamature ovary (found by dissection) hadaMDof4.7,allotherswithamature ovaryfallingintherange5.4-7.4.Prob- ablytheindividualwithaMD of4.7mighthavepreviously oviposited. Theaveragewing loading(WL)ofthe maleswas 18.8±3.6mg/cm2, (range of 9.4-27.5mg/cm2 n = 53), andthatofthe females was 20.8±5.5mg/cm2 (range , , 9.1-30.5mg/cm2-n= 107)foralltheseasons. ThedifferenceofWLbetweenthe Fig.23Dgraphofwingloadingsby periodin Pantalaflavescensfemales capturedatKunugiyama forestfromthesecondhalfof Julytothefirst halfof Octoberin2003. 152 N. Ishizawa sexes was significant (p <0.01). TheaverageWLofthemaleswas20.4±2.9; mg/ cm2(n =26) intheearlyseason, and 17.413.6; mg/cm2(n =27),inthelateseason, asignificant difference(P < 0.001). ThedifferenceofWLbetween the seasons was also significant inthefemales(23.614.3; mg/cm2,n=58,intheearly season and 17.414.7;mg/cm2 n= 49,inthelateseason) (P<0.001). Inthe secondhalf , of July morethan two thirdsof femalesexceeded aWLof20 mg/cm2 andWL was sustained untilthe second halfof August. Figure 2 shows a 3D histogram oftheWLofthefemalescaptured at Kunigi- yama forest.Itindicates thatmature individualswith high WLexceeded imma- ture ones intheearly season, andas the season advanced, immatureindividuals outnumberedthe mature ones. As femalesmature,most of them pruinosed on the ventral sideof theirab- domensfromthe 3rdsegment backwards. The degreeofpruinescens (PD)rose from5.111.3in thesecondhalfofJuly,toreacha peak of5.810.4inthesecond halfof August. Itremained atthis level(5.810.5) into the first halfof Septem- ber, butthereafternopruinescent individualwascaptured. However,therewere examples whereanindividualof533mginweight wasnot pruinescent, whereas oneofonly330mginweight was pruinescent,(PD 5.5).Therewasnosignificant correlationbetweenPDandbody weight(r= 0.140,P>0.3) orbetweenPDand MD(r=0.094, P> 0.5). Figures 3Aand3B showtherelationships ofbody temperature(Tb)to theam- bienttemperature(Ta) ineach sex ofmature andimmatureindividuals.Thecor- relationcoefficientandthe regression coefficientofimmaturemales were0.906 Fig. 3.Thecorrelation betweenbody temperatureand ambienttemperature inthe male(A) and in the female(B),respectively. Asolid andopentriangleshow animmature male and amaturemale, respectively, and asolid and opencircle showanimmature female andmaturefemale,respectively. Allofthedata werecollected atKunugiyama forestfromJulytoOctober 2003.Equationsare:im- maturemale,Tb= 1.085*Ta+ 3.676,r=0.906(n=7);maturemale,Tb=0.757*Ta + 14.021,r= 0.885 (n= 15);immature female,Tb=0.906*Ta + 9.608,r= 0.906(n=21);maturefemale,Tb= 0.827*Ta + 12.420,r=0.938 (n=32). Morphologicalvariations inPantalaflavescens 153 and 1.085,respectively, andinmature males:0.885,0.757, respectively, inimma- turefemales:0.906,0.906andinmaturefemales:0.938,0.827, respectively. There was no significant differenceofboth coefficientsbetween thesexes orbetween immatureandmature individualsofbothsexes (P>0.05, two-side, Z-test). Themean Tb was 32.9±3.5°Cin immaturemalesand 34.9±3.4°Cin mature males; 32.8±3.9°Cin immaturefemalesand36.2±4.5°Cinmaturefemales(Tab. II).TheTbamongthesegroupswas significantly different(p =0.042,d.f.= 3,F = 2.874, ANOVA-test)butonly the Tbof mature femalesdiffered significantly fromothers (P< 0.05, t-test). DISCUSSION Pantalaflavescens fliesover to the Japanese mainlandinspring, and thefirst recordisearlier(inearly andmidApril) atthecoastalareasof thePacific Ocean such assouthernKyushu, Shikoku, andtheKantoregion andwestwardthanat otherareas (Musashino Satoyama Research Group etal, 2004). Themeanfirst recordfor23 yearsbetween 1973and 1995at Amakusa, KumamotoPrefecture was May 17th, and the earliest datewas April 20th(TANAKA & HIGASHI, 2003). Thelastrecordwas October25thatChichibuCity(ARAI, 1995).Therecords from variousareas are mostly early and midOctober; earlierinHokkaido and theTohokuregion, theNortheastof Japan,andlateratthecoastalareas ofthe Pacific Oceanof theTokai region andwestward.The latestrecord, November 23rd, was reported by FUKUI (2005). TherecordfromKunugiyama seems to bemoderate. It deserves special mentionthat the sex ratioskewed towards femalesexcept forthespecimens fromNagasaka. Eveninthespecimens collectedrandomly for DNA sequencefromvarious areas inAsiait skewed towardsfemales, 57.0%, n = 241 and43.0%, n= 182 inthemale(p <0.01, chi-square test)(HAYASHI & ARAI,2004). However,according toKUWADA (1972) thesex ratioskewed to- wardsmales(60%). Ambienttemperature,timezone, period orlocality mayin- fluencethe sex ratio. TANAKA & HIGASHI (2003) notedthat P.flavescens exhibitedtwo popu- Table II Thermal statistics ofPantalaflavescensatKunugiyamaforest(°C±s.d.) Male Female Immature Mature Immature Mature Number 7 15 21 34 Bodytemperature (32.9±3.5) (34.9±3.4) (32,8±3.9) (36.2±4.5) Ambienttemperature (26.9±2.9) (27.6±4.0) (25.6±4.0) (28.9±5.2) 154 N.Ishizawa lationpeaks in KumamotoPrefecture, attheend of Juneandat theend ofAu- gust,andARAI(1995) suspected thatthisspecies mighthavetwogenerations at ChichibuCityaftercoming over there.Judging fromthetwo population peaks at Kunugiyama, one between July and August and the otherfrom September onward, Iagree withARAI (1995)on theassumption thatP.flavescens maybe bivoltineat settlementson themainlandof Japan (except thenortherndistricts) aftercomingfromoverseas. Thefactthatadultsizeremainedfairlyconstantthroughout theseasonandthat two population peaks wereobserved atKunugiyama (though size did decrease alittleinthe lateseason) maysuggestno differenceoflarval durationbetween each generation. TheindoorlarvaldurationofP.flavescens wasextremely short, 34-55days,andoutdoorsatapondoutsideitssouthernrangeinAustraliait was 51 days (HAWKING & INGRAM. 1994),so thatthe larval durationofeach generation may beroughly estimated attwo months(TANAKA & HIGASHI, 2003). This is differentfromsomezygopteran species such as Ischnuraasiatica (ISHIDAetal., 1989)andAciagrionmigratum(UDONO etal., 1997),wherelar- valdurationandadultsize are differentbetween 1st voltineand2ndvoltine,and theadultsoccurring lateinthe season are smaller.Theuniformity inthe sizeof P.flavescens maybe dueto the shortlarvaldurationbetweenvoltines. P.flavescens issaidtostartitsmigrationintheimmaturestage(CORBET, 1984, 1999). Accordingto ASAHINA&TSURUOKA (1967) allindividualscollected onaweathershipattheTango weatherobservationspotonthePacificOceanin August weremature;threemaleswerereddenedontheirabdominaldorsumand the ovariesof femaleswere mature. I donot know the maturityofadultscom- ing over to the mainlandin spring. However, as the above factshows, manyof the migrating individualsin summer maybemature. Thisagreeswiththeshifts of body weight and MDofthe individualsof Kunugiyama forest.Themature individualsatKunugiyama inthe early season mightrepresentamixtureofthe maturedescendantsofthefirstcomers tothemainlandinspring andthemature migrantadultsfromsouthernoverseas. Inautumn MD decreasedsharply. Asonly afewmature individualswereob- servedinthis seasonandindividualsofseemingly soonafteremergencewerein- cluded in the groupsat Kunugiyama, most ofthe individualsmight havebeen thesecond generation thatresulted fromthemature adultsofthe early season. As theshifts ofMDandPD show,itisassumed thatthematurityofthe second generation maynot advance.MD of individualsof whichabdomenswerepru- inescentexceeded4.6andtherewasnot sucha highcorrelationbetweenMDand PD. Therefore, PD cannot beadecisiveindexof maturation. WLofevenmaturefemaleswas significantlylowerthanthatofmaturefemales of S. frequens (28.6±2.0 mg/cm2 n= 41), whichisamigratory species likePfla- , vescens (P <0.05) (Ishizawa, unpublished). Inthe malesP.flavescens, themean WL for theseason was significantly lower than thatof the malesof Orthetrum MorphologicalvariationsinPantalaflavescens 155 albistylum speciosum ofthe samesize, and also itwas thesamewhen compared withthatofO. cancellatummalesfromEurope (26.4±2.54 mg/cm2)(GRABOW &RUPPELL, 1995).Sincemostofmigratingfemalescaptured ontheocean were mature(ASAHINA&TSURUOKA,1967;HASHIMOTO&ASAHINA,1969), inP.flavescens, the lowWLmaybeadvantageous for itsmigratory dispersal. FromSeptember onwardindividualsofWLlowerthan20 mg/cm2weremost frequent, andinOctober mostindividualsofbothsexes weighed ca 15mg/cm2. TheWLis as lowas thoseof someZygoptera (RUPPELL & HILFERT, 1993). ARAI(2003) notedthatalthough manyP.flavescens swarms were sighted, their ovipositionwas seenrarelyinsummer. P.flavescens isatropical species, hencelow- eringofTamaysuppressmaturationandincreaseofbody weight, orWLmaybe suppressed. This is different fromthecaseof S.frequens, inwhichbody weight increas- es inearly and midSeptem- beras Ta falls (ISHIZAWA, 2004, 2005). AccordingtoMAY(1981), wingbeatfrequency ofP.fla- vescens was 22.9 Hz (25°C), thelowest among species of approximately thesame WL of 20 mg/cm2. Therequired power for fluttering flight is proportional to the prod- uctof the 5thpowerof wing length (I) and the cubeof wingbeatfrequency(/),while theavailablepowerispropor- Fig. 4. Relation of regressioncoefficient (REG) tocorrela- tionalto the product of the tioncoefficient(COR) in thepercher-typespecies andinthe body mass(m)andthewing- flyer-type species. Data ofcoefficients of otherOdonata: beatfrequency(f)(AZUMA, Anaxjunius(MAY, 1976), A.parthenopejulius (Ishizawa, 1987). Ignoring coefficients unpubl,),A. guttatus(ISHIZAWA,1998b),Anotogastersie- of both equations; l5xfl = boldii (ISHIZAWA, 2003),patrollingCordulia aeneaamu- rensis (Ishizawa, unpubl.),Mnais costalis oforange wing m~xfP'X-fl =m. type(ISHIZAWA,2000),Orthetrumj.japonicum(Ishizawa, In P. flavescens the wing unpubl.),O. albistylum speciosum(Ishizawa,unpubl.)and length was unchanged as the both sexes ofSympetrumfrequensatreproductive period season advanced. However, (ISHIZAWA, 1998a).The sex ormaturity ofA.juniuswere body weight, as defined by unknown. Otherdata except forS.frequenswerefromma- turemales. Least squares linear regressions are:thepercher WL, decreased. Hence, the type:adottedline,REG= 1.457*COR-0.099,r=0.766,p= wingbeat frequency may de- 0.131;theflyertype:asolidline,REG=0.354*COR+0.046, crease inthe autumn. As Tb r=0.907,p=0.034. 156 N.Ishizawa andthe wingbeat frequency are positivelycorrelated, decrease of wingbeat fre- quencybringsloweringofmetabolicheatproduction(MAY, 1981), so thatTb is naturally lowered. P.flavescens flieswithlessflutteringonthewindandlowerWLisadvantageous forits flight. Thelatter may be disadvantageous for thermoregulation andtem- peraturedependency inevitablybecomeslarge forthe dragonfly.Actually, between bothsexes orbetweenmatureindividualsandimmatureindividualstherewas no significant differenceofthe regression coefficientsofTbto Ta. They appeared tobeaffectedbyTa andthiswas shownin thelargeregression coefficients.This species belongs to theflyer type(CORBET, 1962) likeAeshnidaeandCordulii- dae.Regression coefficientsofthe flyer type dragonflies tend tobe lower than thoseof the percher type (Fig. 4). Theformerare mostly endothermicwhereas thesecond are, in general,ectothermic(MAY, 1976, 1987).However, thoseofP. flavescens approximate thoseofthepercher type. Pflavescens isliableto swarmeveninhotdaytimeinmidsummer.ARAI(1995) reported thatsomeofthoseswarmingwereseenflyingwiththeirabdomensbent downward.ThiswasoftenobservedwhenTaexceeded30°Candthey flewagainst thewindwiththisposture.Therefore, itisassumedthattheythermoregulate their Tbbyexposing theirabdomensto thewind.CORBET(1999) describedthatthis is aradiatoreffect.Inthis caselowWL mightbeprofitable forthem, onlyflying driftinglyin the windwithless fluttering.Thehigh dependency on temperature maysuppress maturationofthespecies inthecool autumn. Thus, the swarm migrationsuch as observedby WAKANA(1959) mayprob- ably beabehaviourforadaptation to suitableambienttemperatures. ACKNOWLEDGEMENTS Iwishtoexpressmycordial thankstoMr H.NARAOKA forprovidingliterature andforhiscrit- ical readingofmy manuscriptand also toDrs K. HIGASHIand I.WAKANA foraccesstotheir literature,and toananonymousreferee forhiscritical readingand correction ofmymanuscript. REFERENCES ASAHINA.S.&Y.TSURUOKA, 1967.Records oftheinsects visitingaweathershiplocatedatthe OceanWeather Station ‘Tango’onthePacific.Kontyu35(4):353-360. [Jap.] ARAI,Y., 1995. Somebiological observations ofPantalaflavescens inSaitama Prefecture. Gracile 53: 19-22.[Jap.] ARAI,Y,2003. ProblemsonPantalaflavescensandSympetrumfrequens.In:Y,Arai,[Ed.],Acoun- trywide survey ofred dragonfliesin2003, pp. 13-18,Musashino Satoyama Res.Gr. & Inst. Agric. andNat. Environ.,Yorii-cho,Saitama. [Jap.] AZUMA,A.,1987. Those magnificentmotionsoflivingcreatures.Kyontsu Shuppan,Tokyo.[Jap.] CORBET,P.S.,1962. Abiologyofdragonflies.Witherby,London. CORBET,P.S., 1984. Orientation andreproductivecondition ofmigratingdragonflies(Anisoptera). Odonalologica13(1):81-88.

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.