- Email: [email protected]
Flight and echlocation in the ecology and evolution of bats
ats (Order Chiroptera)
are
showed
th2t rmst insectivorous
bats searchiwg for airbornle prey
variations on a s
produce one or fewer echolocation pulses per wingbeatil. Gleaning ats emit several echolocation calls per wingbeat, but their calls
ay, 927species of bats are classified in two suborders: the Mega-
are oh lower intensity. and presumably of lower energetic cost.
The ~~~cove~ of the coupling of ~~~Q~~~a~~~~, respiration, and bats that feed QUI airborne liges us to reconsider the impact of ec~o~oca~~~~ and flight
prey on small vertebrates (includ-
Psnger lifespans, lower rates
of
on the ecology, behaviour and eve lution of batsl2. Here, we focus 011 the role of fnightand ec~~o~oca~~~~
M&%.x Arita is at the Center of Ecology National Univemty of Mexico (IJNAM), Apartado Postal 27-3, CP 58089, Morelia, Michoac6n, IMexico; Bock Fenton is at the Dept of Biology, York University, North York, Ontario. Canada M39 lP3.
e or, the relative and echolocation on the appearance and
istics may be linkedto flight and eeholoeatiorr, the two basic
foraging behaviour and e study of echolocation in
itional comparison
~~o~~~~~~o~o~-~~eanalysis of size and shape in plants
and animals to infe
some carnivores) have
ore developed noseleaves
correlation between diet and noseleaf morphology, hbeconto echolocation and foraging behaviolar is not clearl5.The pinnae (external ears) of some bats also reflect echolocation (Fig. 1).Acomparative study 0 onstrated three patterns of ear shape’“. In bats (BPX 3) and some free-t mechanically tuned to the soun olocation associated with puke p~~d~c~~~m, echo~ocati~~calls. In co at least in bats emitting hi intensity calls. An analysis of bats are most sensitive to lower frequency sou the scaling 06wingbeat frequency and echolocation pulse such as those associat repetition rate with body mass (using data from fieldstudies) ingli (Fig. Ic). Finally, nection
TREE vol. 1.2, no. 2 Februnry
1997
Copyright 0 1997. Elsevter Science Ltd. All rights reserved.
0169-5347/97/917.00
PII: SolG9-5347(96)10058-fi
bats that feed on small vertebrates from Ot None of these features by itself, however. clearly discriminates carnivores from other batsIR.
ol the behaviour. ecology and mar bats presents excellent 0ppor~m-1 further ccmparative studies, and the detailed phyloge~~ics avaibable Ear some groups will make it pcassible to place such studies in an evolutionary context. For example, molecular data show that the subfamily ~hy~~QstQmi~ae, which trastomid gleaners, is polye associated morphologitraits may baiie evoived the similarities between convergence rather than of parallel evolution.
The
diveasity
of echolocating Egpnan frurt oat (Rousetrus ae&pfracus) Indian flying fox (Reropus grganteus!
Bumblebee bat (Craseonycteris thong~oflgvFji) Emballonuridae awk bat (Saccolaimus peii) ~~l~~~~i~~~ Binperfa dae WI Commerson’s leaf-nosed bat (Hrpposideros commefsonI) Mcgodermatidae Ghost bat (Mocroderma @gas) Rrunolophldae
Parnell’s moiistached bal (Pteronotus pamel/rJ) Phyllostomldac Bennett’s spear-nosed bat (iW!mo:~ h+nr~et!~‘i f&-eyed bat (Chrfoderrna v~ilosurn) Greater spear-nosed bat (Phyilostomus I!astattis) Linnaeus’ f&r vamp~rc bat ~L’am~wm spectrum Lale long-eared bat (M; :ro~rycteris megalotisi Long-tongued bat (C;lossophaga soncina) fruit bat (Carollra persprcillata) Shortlionoltiea upetiamily We Molossidae European free-tailed bat (Tadanda tenrohs) Undewood’s mastiff bat (&mops un&rwoor 1) Vesperblionidae European p~plstrelle (Pipr’strellur prprstrelfus) Giant house bat (Scotophrlus nr:rita) Mexican long-eared bat (P/ecotm mexrcanus) Naked bat (Cherromeles torquatus) Western plpistrelle (Pipistreflus hesperus)
‘The rclat onship bctwecn wirrg rlKPrpboPogp aud tlict. foraging strategy an other ecological feat studied by Norberg a d her collaborators”,:S ( location and flight are parts of a single adaptive complex, 1oby and echolocation features should corree case with carnivorous batsIs. AcQm~i~at~Q~ of i~~r~~~o~Qg~~a~ characters (large body mass, low wing loading, low aspect ra d echolocation calls (short duration, low ~~~el~§~ty~ sound frequency) dist~~g~~s~es
wgPOadlng ( WL= WjS.where W IS the bat’s weight. and S IS the area of the wing elements) measures the force oer unit area that wmgs must suppott during flight. The aspect a&lo (AR= b/c. where II IS the wing span. ano c IS the mean width of the wmg elements) measures the shaoc of the wmgs by quantifying their relative length -8ngtip indicsrs Imeasure the shape oi the tips of the wmg; high values of the mdlces correspond to wmgs with rounded or squansh CIPS, whereas low values are typical of more-aointed wines. “_ The type of flight and toragmg habitai of bats correspond closely to the size and shape of the wmg element@. For example, free4ailed bats (Molossidae~. which are fast fliers and forage in open areas, have wings that are relatively smii (i.e. with a high wing loading) and long (i.e. with a htgh aspect ratio). Convticsely. gleaners (species that prey on insects or small vertebrates that they capture from a surface) possess wings that are large (low wing loading) and relatively short (low aspect ratio). They can fly slowly withm vegetation. carrying heavy weights.
Why are bats SO small? Bats are small mammals (Fig. 21, ranging in adult mass from 2 g (bumblebee bat) to 1500 g (Indian f1ying fox]. Winge krertebrates can attain larger sizes, as indicated by birds an jakrosaurs, showing that flight alsne does not Ii size in bats. Why are bats so small? Variation among feeding and taxQnomic groups of bats suggests that the answer rekcts present erological factors and historical evolutionary constraints, particularly factors associated with fright and echoPocation. Aerial-deeding insectivorous bats are
with most species weighing less than 3Qg ody four species (naked bat, b~~c~~~aw~bat, giant house bat and Commerson’s leaf-noses bat) weighin BOO g. The morphological constraints n r capturing flying inset a aerial insectivores. Bats with low wing a oeuvrable and, because wing loa tiona! to massi::L for similarly shaped bats, smaller animals have Power wing loading than larger ones. This relationship i~nplic~s‘IhaPsmakr bats should be more efficient at capturing airborne prey, and COUPexplain why acrid insectivorous bats and birds tend to be small’.
not small ones. Furthermor
repetition rate would impair the ability of a very large bat to
s for small size in i
small and large insects11x21.Larger bats, in contrast, depela on large insects for food. Commerson’s leaf-nosed bat, fQ ample, uses echolocation calls that are dominated by 60
in order to detect and track its prey, usually Barge (IQ-15 g) flying dung beetles”?. An exception to the rule is the at, an insectivore that p CA d frequencies higher tha its size, and that can feed on small ins large size (170g) and high flight speed23. Other large insectivorous bats can escape from energetic
sounds
location callPJ5. There are two ways to thwart this hearing-
All microchiropterans and one pteropodrd (Egvphan fruit bati cnenr bv echo locatron. and some microchrropterans also useecholocation to detect, trackand assess airborne targets. Egyptian frui? bats use tongue clicks to generate echo location sounds. while microchiropterans use (,m--’ .ti~oi&ons produced In the larynx to generate tOnal S!gmki showing structured changes tn frequency over time. Most bais use UltraSOniC (>20 kHz) echolocation calls that, because of their wave lengths, provide better resolubon of target detail. Some bats, usually larger specbes, use echolocation calls that are audible to humans (~20 kt-lm). Echolocation signals can be broadband or narrowband. Broadband or FM (fre quency modulated) calls span a range of sound frequencies (bandwidths up to I.00 kHz), while narrowband or CF (constant frequency) calls are almost oure tones, with bandwidths of 41 kHz. Microchiropterans that hunt airborne prey use a corn bination of broad- and narrowband calls. Some bats use high intensity, energeti. caky expensive echotocabon calls (>I10 dB SPL at 1Ocm): others, the ‘whispering bats’ use calls of lower inter&y (60db SPL at 10cm). The 130 species of horseshoe and Old World leaf-nosed bats and Pamelt’s moustached bat emit echolocation calls separated by brief periods of silence - a high-duty-cycle system with calls produced over 50% of the time. The remainrng echolocating microchrrop?era and Egyptian fruit bats produce echolocatron calls separated by long periods of silence - a low-duty-cycle system with calls produced 20% of the time.
echolocation calls (as measured by the characteristic freand some take much larger moths than aerial-feeders QB quencies and the ‘shape’ in a sonagram). They identified four groups of ir,secfivoroui; bats based on correMed leaequivalent size’*. tures of foraging behaviour. echolocation calls and wing PnQ~phOlO~.
is the relative role of foe
(ecological) and regional (~fstorf~a~)factors in deter the ~om~os~tfo~and structure of biotic communities. The analysis of echolocation d hology proforces that vides important clues for ape bat assemblages, vii3 ese feature5 etween species reflect different strategies of resource use. There is a close relationship between ~~~~r~hofo~and diet in some communities of insectivorous bats’, and a case of ~a~tit~o~ingof sonar frequencies, presumably associated with a partitioning of feeding resources, has been reported for an assemblage of Malay:;ianrhinolophi&P. Using bivariate plots- i~i and RolfeYn ~~e~~ffj~d distinct guilds in ~~~~~aiia~ woodland bat communities.Species
Evidence of the role of regional factors in shaping bat communities is revealed by morp ological structure, as well as by taxonomic and trophic composition, which show that local assemblages of NewWorld bats samples of the whole neotropical fauna. T the local faunas of eastern Braziland the \I exico are compared with regional poo erican bat fauna and the fauna of sout respectivePy3*133. ivariate plots of aspect ratio and wing-tip indices demonstrate more morph logical overlap within neotropical assemblages than in 01 World communitiesR4. AdditiorzDy,neotropical communities are characUerizedby
(4
wisms allowing the coexist-
adaptive complex and found
~i~~~~~~arnt correlations between wing m~rpho~o~ (as dex; srr Box 2) and type of
Ffg. i. Portraits of six New World bats showrng ear and face features mvolved m echolocation. (See Box 1 for Latin nafmS.) (a) Underwood’s mastiff bat. a fast-flying aerial msectivore with rounded ears. (b) Western pipistrelle, an aenal msectwore with the typrcal. non-ornamented face of vesperhlionids. (c) Mexrcan long-eared bat. a gleaner with very long ears. (d) Bennett’s spear-nosed bat, a partially carnivorous gleaner with long and pointed ears and noseleaf. (e) The long-tongued bat, a nectar-feeding species with short and rounded ears and noseleaf. (f) A big-eyed bat, a frugivorous species wrth large eyes and medium-sized ears and noseleaf.
~____-
--
-
-
Body 0.001
0.01
0.1
1
mass
(kg) 10
: 000
100
bats arose from gliding, notturnal insectivores that ecbo-
IQ000
Flyng birds Terrestrial
mammals
Bats aerial
insectrvores
The flight-first theory* proposes that bats evolved from primitive climbing insectivores that ate arthropods,
gleaners nectarivores Megachtroptera _I_p
Microchiroptera rrugivores
immediate developed portation
Megachlroptera Mic~ochiroptcra ____
vampir.2s erg.
. Range mbud)mds~foor varmsgroups
of vertebral~~s. NOW logmlhni~c SL~IIC
~I______..___-.-_~_tk tilat
high diversity llave
-____
.-.d ‘wlmispering
llacliated
t0 iIlclutk
knts’ (eoainly gkailiilg,
~~~~~~§t~~~~~~s~
~~~~t~V~~~~~~~ arlel
s. The situation suggests the imporhematoplaagtrens spc tance of the history of invasions of lhe New World by bats and supports the idea that the composition and structure Of bat communities are determined primarily by regional, historical factors more than by local mteractions’.
flight-first and echolocation-first Echolocation
first
of 18iL$? and o&eel, the theories, which agree that
gliding
of bats
fol- trans-
and used echoorientation. Even after the evolution of flight, lack of ~~an~e~vrabi~~~y precluded the capture of airborne prey. hnproved manoeuvrability was followed by the perfection 01 echolocation for detecting, tracking and asbration
-----
ancestors
sessing
airh~aae
targets.
in
The combination
of echolocation
and manoemvrability
new niches for bats, and
moted their adaptive
n.
ory”, use e&Jocation to detect, track a evohcd before flight with the appearanceof the stronger signals necessary to increase the effective range of echolocation. This theory proposes that gliding protobats bunted from perches, usirsg echolocation to detect airborne prey that was captured in the air. Both theories are coherent with current knowledge on echolocation and flight. ever, the mechanical co Flight frrst of flight and echolocation
The ectaolocation
could have evol a system linked to a welldeveio~ed flight mechanism.
Pteropodids (with the exception of Egyptian fruit bats) do n0P echolocate and have simple ears and faces that Fig. 3. Evolution of echolocation and flight in the Mlcrochiroptera accofclmg to the echolxabolt-ilrst (left) and flight-first (right) hypotheses. Both theol ies consiaer an insectivorous, climbing ancestor that, throug I d series of e~!Wonary events. gave rise to a stage corresponding to the Palaeochiropterygidae. a fossil family from the Eocene. T”..: rubsequent micro clliropteran radiation gave rise to the four modem subfamilies: Emballonuroidea, Rhinolophoidea. Phyiiostomoidea. Vesper tilionoidea. Diet: INS. non-flying insects: Al, airborne insects; GL. insects or small vertebrates captured from surfaces Igleaning). Type of echolocation sound: CL, clicks; TS, tonal signals. Duty cycle: LDC, low-duty cycle: HDC. high-duty cycle. Symbols in bold indicate the ongin of new trails.
vores, megacbiro~~era~s are the Oid World e~~ivafe~~s of some phyllostomids, although they differ in appearance, echolocation ability and
stomids can hover, the flower-visiting megachiropterans tend to hang while feeding. Hovering would be too expen-
g bats are p~ero~od~ds,
phyllostomidsj, so it is not are so much smaller than
wby ~hy~f~s~o~~~frqivores pteropodid counterparts.
rather than of microchiropterans3”. e conventional view has been that mega- and ~~icrocb~~o~~~era~s are ~is~e~~~~~~s that diverged very eariy in their evolutio historyjl. Neither theory on the evolution of flight ecbolocation
orders evolved from a common ancestor, then flight has
appeared in mammals only 01.ce. In this case, according to the ecbo~oca~~o~-first theory, echolocation was lost early in the ~egac~~~~o~te~a~ line and evolved later in the genus
iderable impact on e is constrained by the enercholocation, then these spects of the ecology of rrelated with most life nt discoveries, such as the presence of folivory in bats despite the energetic conraints limiting the evolution of leaf-eating in flying anials”K, and Barclay’s”” theory of calcium as a limiting resource for the reproduction of flying animals, particularly insectivorous bais, show that our u~~ersta~di~g of how flight and echolocation affect the ecology and evolution of bats is still r~~d~~e~~a~. Future research should take into account that the lives of bats cannot be understood withnut considering the effects of flight and echolocation.
This work was partially funded by grants from the ib4Iexican Commission on Biodivers~ty (~~~A~f~~ and the National University of Mexico (UNAM). We thadr W. Zomleffer for the drawings of bat faces and 6. Ceballos, R. Medellin and two reviewers for comments on the manuscript.
hy would
evolution crewe do not know when females first evolved ore prosaic ancestor. Several hypotheses have been promasculinized
that survive lose over 6Wh of their first-born y~ung’,~?Since Aristotle. natural historians and scientists
Laurence
Frank is at the Psychology Dept,
Untversity of California. ~o~~st~ict~~~ at the base”. In place Berkeley. CA 94720, USA of the vulva is a scrotum-like sac, ([email protected]). filledvv’.thfat and connective tissue. ~--Bnterndly,however, the gross anatomy of the female reproductive eractis Otlex~eFtiQ~a~~ The Iemale mates and gives identify the rnutahionalevents responsible for this ~~~$~a~ tb~o~ghthe ~e~~f~rrnclitoris. In the ~~e~~~e~ta~ clitoris, the suite of cfnaramrs. ~r~ge~ita~canal is only slightly larger than that of the penis. Atpuberty, however, the femalecanal enlarges and becomes elastic5to allowmating,which is facilitatedby a pair of robust retractor muscles6that enable the femaleto retract the phallus upon itself, much as one pushes up a shirt sleeve, forming a Role which permits the male to achieve intrQm~ssi~~. more aggressive than males, d~m~~ati~gthem in nea Since 1877,anatomical studies have disp social situations8 such that even the lowest ranking fe belief that the spotted hyaena is hermap are able to displace the highest ranking adult males. Females
copYright @ 1997. Elsevier
Science Ltd. All rights reserved. 0169-S347/97/51i ‘JO
PII: SOl6!%34:(96)lOO63
TXEE
ool.
12. no. 2 Fe’ebrumy 1997
are
showed
th2t rmst insectivorous
bats searchiwg for airbornle prey
variations on a s
produce one or fewer echolocation pulses per wingbeatil. Gleaning ats emit several echolocation calls per wingbeat, but their calls
ay, 927species of bats are classified in two suborders: the Mega-
are oh lower intensity. and presumably of lower energetic cost.
The ~~~cove~ of the coupling of ~~~Q~~~a~~~~, respiration, and bats that feed QUI airborne liges us to reconsider the impact of ec~o~oca~~~~ and flight
prey on small vertebrates (includ-
Psnger lifespans, lower rates
of
on the ecology, behaviour and eve lution of batsl2. Here, we focus 011 the role of fnightand ec~~o~oca~~~~
M&%.x Arita is at the Center of Ecology National Univemty of Mexico (IJNAM), Apartado Postal 27-3, CP 58089, Morelia, Michoac6n, IMexico; Bock Fenton is at the Dept of Biology, York University, North York, Ontario. Canada M39 lP3.
e or, the relative and echolocation on the appearance and
istics may be linkedto flight and eeholoeatiorr, the two basic
foraging behaviour and e study of echolocation in
itional comparison
~~o~~~~~~o~o~-~~eanalysis of size and shape in plants
and animals to infe
some carnivores) have
ore developed noseleaves
correlation between diet and noseleaf morphology, hbeconto echolocation and foraging behaviolar is not clearl5.The pinnae (external ears) of some bats also reflect echolocation (Fig. 1).Acomparative study 0 onstrated three patterns of ear shape’“. In bats (BPX 3) and some free-t mechanically tuned to the soun olocation associated with puke p~~d~c~~~m, echo~ocati~~calls. In co at least in bats emitting hi intensity calls. An analysis of bats are most sensitive to lower frequency sou the scaling 06wingbeat frequency and echolocation pulse such as those associat repetition rate with body mass (using data from fieldstudies) ingli (Fig. Ic). Finally, nection
TREE vol. 1.2, no. 2 Februnry
1997
Copyright 0 1997. Elsevter Science Ltd. All rights reserved.
0169-5347/97/917.00
PII: SolG9-5347(96)10058-fi
bats that feed on small vertebrates from Ot None of these features by itself, however. clearly discriminates carnivores from other batsIR.
ol the behaviour. ecology and mar bats presents excellent 0ppor~m-1 further ccmparative studies, and the detailed phyloge~~ics avaibable Ear some groups will make it pcassible to place such studies in an evolutionary context. For example, molecular data show that the subfamily ~hy~~QstQmi~ae, which trastomid gleaners, is polye associated morphologitraits may baiie evoived the similarities between convergence rather than of parallel evolution.
The
diveasity
of echolocating Egpnan frurt oat (Rousetrus ae&pfracus) Indian flying fox (Reropus grganteus!
Bumblebee bat (Craseonycteris thong~oflgvFji) Emballonuridae awk bat (Saccolaimus peii) ~~l~~~~i~~~ Binperfa dae WI Commerson’s leaf-nosed bat (Hrpposideros commefsonI) Mcgodermatidae Ghost bat (Mocroderma @gas) Rrunolophldae
Parnell’s moiistached bal (Pteronotus pamel/rJ) Phyllostomldac Bennett’s spear-nosed bat (iW!mo:~ h+nr~et!~‘i f&-eyed bat (Chrfoderrna v~ilosurn) Greater spear-nosed bat (Phyilostomus I!astattis) Linnaeus’ f&r vamp~rc bat ~L’am~wm spectrum Lale long-eared bat (M; :ro~rycteris megalotisi Long-tongued bat (C;lossophaga soncina) fruit bat (Carollra persprcillata) Shortlionoltiea upetiamily We Molossidae European free-tailed bat (Tadanda tenrohs) Undewood’s mastiff bat (&mops un&rwoor 1) Vesperblionidae European p~plstrelle (Pipr’strellur prprstrelfus) Giant house bat (Scotophrlus nr:rita) Mexican long-eared bat (P/ecotm mexrcanus) Naked bat (Cherromeles torquatus) Western plpistrelle (Pipistreflus hesperus)
‘The rclat onship bctwecn wirrg rlKPrpboPogp aud tlict. foraging strategy an other ecological feat studied by Norberg a d her collaborators”,:S ( location and flight are parts of a single adaptive complex, 1oby and echolocation features should corree case with carnivorous batsIs. AcQm~i~at~Q~ of i~~r~~~o~Qg~~a~ characters (large body mass, low wing loading, low aspect ra d echolocation calls (short duration, low ~~~el~§~ty~ sound frequency) dist~~g~~s~es
wgPOadlng ( WL= WjS.where W IS the bat’s weight. and S IS the area of the wing elements) measures the force oer unit area that wmgs must suppott during flight. The aspect a&lo (AR= b/c. where II IS the wing span. ano c IS the mean width of the wmg elements) measures the shaoc of the wmgs by quantifying their relative length -8ngtip indicsrs Imeasure the shape oi the tips of the wmg; high values of the mdlces correspond to wmgs with rounded or squansh CIPS, whereas low values are typical of more-aointed wines. “_ The type of flight and toragmg habitai of bats correspond closely to the size and shape of the wmg element@. For example, free4ailed bats (Molossidae~. which are fast fliers and forage in open areas, have wings that are relatively smii (i.e. with a high wing loading) and long (i.e. with a htgh aspect ratio). Convticsely. gleaners (species that prey on insects or small vertebrates that they capture from a surface) possess wings that are large (low wing loading) and relatively short (low aspect ratio). They can fly slowly withm vegetation. carrying heavy weights.
Why are bats SO small? Bats are small mammals (Fig. 21, ranging in adult mass from 2 g (bumblebee bat) to 1500 g (Indian f1ying fox]. Winge krertebrates can attain larger sizes, as indicated by birds an jakrosaurs, showing that flight alsne does not Ii size in bats. Why are bats so small? Variation among feeding and taxQnomic groups of bats suggests that the answer rekcts present erological factors and historical evolutionary constraints, particularly factors associated with fright and echoPocation. Aerial-deeding insectivorous bats are
with most species weighing less than 3Qg ody four species (naked bat, b~~c~~~aw~bat, giant house bat and Commerson’s leaf-noses bat) weighin BOO g. The morphological constraints n r capturing flying inset a aerial insectivores. Bats with low wing a oeuvrable and, because wing loa tiona! to massi::L for similarly shaped bats, smaller animals have Power wing loading than larger ones. This relationship i~nplic~s‘IhaPsmakr bats should be more efficient at capturing airborne prey, and COUPexplain why acrid insectivorous bats and birds tend to be small’.
not small ones. Furthermor
repetition rate would impair the ability of a very large bat to
s for small size in i
small and large insects11x21.Larger bats, in contrast, depela on large insects for food. Commerson’s leaf-nosed bat, fQ ample, uses echolocation calls that are dominated by 60
in order to detect and track its prey, usually Barge (IQ-15 g) flying dung beetles”?. An exception to the rule is the at, an insectivore that p CA d frequencies higher tha its size, and that can feed on small ins large size (170g) and high flight speed23. Other large insectivorous bats can escape from energetic
sounds
location callPJ5. There are two ways to thwart this hearing-
All microchiropterans and one pteropodrd (Egvphan fruit bati cnenr bv echo locatron. and some microchrropterans also useecholocation to detect, trackand assess airborne targets. Egyptian frui? bats use tongue clicks to generate echo location sounds. while microchiropterans use (,m--’ .ti~oi&ons produced In the larynx to generate tOnal S!gmki showing structured changes tn frequency over time. Most bais use UltraSOniC (>20 kHz) echolocation calls that, because of their wave lengths, provide better resolubon of target detail. Some bats, usually larger specbes, use echolocation calls that are audible to humans (~20 kt-lm). Echolocation signals can be broadband or narrowband. Broadband or FM (fre quency modulated) calls span a range of sound frequencies (bandwidths up to I.00 kHz), while narrowband or CF (constant frequency) calls are almost oure tones, with bandwidths of 41 kHz. Microchiropterans that hunt airborne prey use a corn bination of broad- and narrowband calls. Some bats use high intensity, energeti. caky expensive echotocabon calls (>I10 dB SPL at 1Ocm): others, the ‘whispering bats’ use calls of lower inter&y (60db SPL at 10cm). The 130 species of horseshoe and Old World leaf-nosed bats and Pamelt’s moustached bat emit echolocation calls separated by brief periods of silence - a high-duty-cycle system with calls produced over 50% of the time. The remainrng echolocating microchrrop?era and Egyptian fruit bats produce echolocatron calls separated by long periods of silence - a low-duty-cycle system with calls produced 20% of the time.
echolocation calls (as measured by the characteristic freand some take much larger moths than aerial-feeders QB quencies and the ‘shape’ in a sonagram). They identified four groups of ir,secfivoroui; bats based on correMed leaequivalent size’*. tures of foraging behaviour. echolocation calls and wing PnQ~phOlO~.
is the relative role of foe
(ecological) and regional (~fstorf~a~)factors in deter the ~om~os~tfo~and structure of biotic communities. The analysis of echolocation d hology proforces that vides important clues for ape bat assemblages, vii3 ese feature5 etween species reflect different strategies of resource use. There is a close relationship between ~~~~r~hofo~and diet in some communities of insectivorous bats’, and a case of ~a~tit~o~ingof sonar frequencies, presumably associated with a partitioning of feeding resources, has been reported for an assemblage of Malay:;ianrhinolophi&P. Using bivariate plots- i~i and RolfeYn ~~e~~ffj~d distinct guilds in ~~~~~aiia~ woodland bat communities.Species
Evidence of the role of regional factors in shaping bat communities is revealed by morp ological structure, as well as by taxonomic and trophic composition, which show that local assemblages of NewWorld bats samples of the whole neotropical fauna. T the local faunas of eastern Braziland the \I exico are compared with regional poo erican bat fauna and the fauna of sout respectivePy3*133. ivariate plots of aspect ratio and wing-tip indices demonstrate more morph logical overlap within neotropical assemblages than in 01 World communitiesR4. AdditiorzDy,neotropical communities are characUerizedby
(4
wisms allowing the coexist-
adaptive complex and found
~i~~~~~~arnt correlations between wing m~rpho~o~ (as dex; srr Box 2) and type of
Ffg. i. Portraits of six New World bats showrng ear and face features mvolved m echolocation. (See Box 1 for Latin nafmS.) (a) Underwood’s mastiff bat. a fast-flying aerial msectivore with rounded ears. (b) Western pipistrelle, an aenal msectwore with the typrcal. non-ornamented face of vesperhlionids. (c) Mexrcan long-eared bat. a gleaner with very long ears. (d) Bennett’s spear-nosed bat, a partially carnivorous gleaner with long and pointed ears and noseleaf. (e) The long-tongued bat, a nectar-feeding species with short and rounded ears and noseleaf. (f) A big-eyed bat, a frugivorous species wrth large eyes and medium-sized ears and noseleaf.
~____-
--
-
-
Body 0.001
0.01
0.1
1
mass
(kg) 10
: 000
100
bats arose from gliding, notturnal insectivores that ecbo-
IQ000
Flyng birds Terrestrial
mammals
Bats aerial
insectrvores
The flight-first theory* proposes that bats evolved from primitive climbing insectivores that ate arthropods,
gleaners nectarivores Megachtroptera _I_p
Microchiroptera rrugivores
immediate developed portation
Megachlroptera Mic~ochiroptcra ____
vampir.2s erg.
. Range mbud)mds~foor varmsgroups
of vertebral~~s. NOW logmlhni~c SL~IIC
~I______..___-.-_~_tk tilat
high diversity llave
-____
.-.d ‘wlmispering
llacliated
t0 iIlclutk
knts’ (eoainly gkailiilg,
~~~~~~§t~~~~~~s~
~~~~t~V~~~~~~~ arlel
s. The situation suggests the imporhematoplaagtrens spc tance of the history of invasions of lhe New World by bats and supports the idea that the composition and structure Of bat communities are determined primarily by regional, historical factors more than by local mteractions’.
flight-first and echolocation-first Echolocation
first
of 18iL$? and o&eel, the theories, which agree that
gliding
of bats
fol- trans-
and used echoorientation. Even after the evolution of flight, lack of ~~an~e~vrabi~~~y precluded the capture of airborne prey. hnproved manoeuvrability was followed by the perfection 01 echolocation for detecting, tracking and asbration
-----
ancestors
sessing
airh~aae
targets.
in
The combination
of echolocation
and manoemvrability
new niches for bats, and
moted their adaptive
n.
ory”, use e&Jocation to detect, track a evohcd before flight with the appearanceof the stronger signals necessary to increase the effective range of echolocation. This theory proposes that gliding protobats bunted from perches, usirsg echolocation to detect airborne prey that was captured in the air. Both theories are coherent with current knowledge on echolocation and flight. ever, the mechanical co Flight frrst of flight and echolocation
The ectaolocation
could have evol a system linked to a welldeveio~ed flight mechanism.
Pteropodids (with the exception of Egyptian fruit bats) do n0P echolocate and have simple ears and faces that Fig. 3. Evolution of echolocation and flight in the Mlcrochiroptera accofclmg to the echolxabolt-ilrst (left) and flight-first (right) hypotheses. Both theol ies consiaer an insectivorous, climbing ancestor that, throug I d series of e~!Wonary events. gave rise to a stage corresponding to the Palaeochiropterygidae. a fossil family from the Eocene. T”..: rubsequent micro clliropteran radiation gave rise to the four modem subfamilies: Emballonuroidea, Rhinolophoidea. Phyiiostomoidea. Vesper tilionoidea. Diet: INS. non-flying insects: Al, airborne insects; GL. insects or small vertebrates captured from surfaces Igleaning). Type of echolocation sound: CL, clicks; TS, tonal signals. Duty cycle: LDC, low-duty cycle: HDC. high-duty cycle. Symbols in bold indicate the ongin of new trails.
vores, megacbiro~~era~s are the Oid World e~~ivafe~~s of some phyllostomids, although they differ in appearance, echolocation ability and
stomids can hover, the flower-visiting megachiropterans tend to hang while feeding. Hovering would be too expen-
g bats are p~ero~od~ds,
phyllostomidsj, so it is not are so much smaller than
wby ~hy~f~s~o~~~frqivores pteropodid counterparts.
rather than of microchiropterans3”. e conventional view has been that mega- and ~~icrocb~~o~~~era~s are ~is~e~~~~~~s that diverged very eariy in their evolutio historyjl. Neither theory on the evolution of flight ecbolocation
orders evolved from a common ancestor, then flight has
appeared in mammals only 01.ce. In this case, according to the ecbo~oca~~o~-first theory, echolocation was lost early in the ~egac~~~~o~te~a~ line and evolved later in the genus
iderable impact on e is constrained by the enercholocation, then these spects of the ecology of rrelated with most life nt discoveries, such as the presence of folivory in bats despite the energetic conraints limiting the evolution of leaf-eating in flying anials”K, and Barclay’s”” theory of calcium as a limiting resource for the reproduction of flying animals, particularly insectivorous bais, show that our u~~ersta~di~g of how flight and echolocation affect the ecology and evolution of bats is still r~~d~~e~~a~. Future research should take into account that the lives of bats cannot be understood withnut considering the effects of flight and echolocation.
This work was partially funded by grants from the ib4Iexican Commission on Biodivers~ty (~~~A~f~~ and the National University of Mexico (UNAM). We thadr W. Zomleffer for the drawings of bat faces and 6. Ceballos, R. Medellin and two reviewers for comments on the manuscript.
hy would
evolution crewe do not know when females first evolved ore prosaic ancestor. Several hypotheses have been promasculinized
that survive lose over 6Wh of their first-born y~ung’,~?Since Aristotle. natural historians and scientists
Laurence
Frank is at the Psychology Dept,
Untversity of California. ~o~~st~ict~~~ at the base”. In place Berkeley. CA 94720, USA of the vulva is a scrotum-like sac, ([email protected]). filledvv’.thfat and connective tissue. ~--Bnterndly,however, the gross anatomy of the female reproductive eractis Otlex~eFtiQ~a~~ The Iemale mates and gives identify the rnutahionalevents responsible for this ~~~$~a~ tb~o~ghthe ~e~~f~rrnclitoris. In the ~~e~~~e~ta~ clitoris, the suite of cfnaramrs. ~r~ge~ita~canal is only slightly larger than that of the penis. Atpuberty, however, the femalecanal enlarges and becomes elastic5to allowmating,which is facilitatedby a pair of robust retractor muscles6that enable the femaleto retract the phallus upon itself, much as one pushes up a shirt sleeve, forming a Role which permits the male to achieve intrQm~ssi~~. more aggressive than males, d~m~~ati~gthem in nea Since 1877,anatomical studies have disp social situations8 such that even the lowest ranking fe belief that the spotted hyaena is hermap are able to displace the highest ranking adult males. Females
copYright @ 1997. Elsevier
Science Ltd. All rights reserved. 0169-S347/97/51i ‘JO
PII: SOl6!%34:(96)lOO63
TXEE
ool.
12. no. 2 Fe’ebrumy 1997