Wasps attacks.pdf

SNFE758Z04

Studies on Neotropical Fauna and Environment

2004, Vol. ••, No. •, pp. ••–••

Pompilid Wasp Interactions with Burrowing Tarantulas:

Pepsis cupripennis versus Eupalaestrus weijenberghi and

Acanthoscurria suina (Araneae, Theraphosidae)

Fernando G. Costa 1 , Fernando Pérez-Miles 1,2

and Antonio Mignone 1,2

1 Laboratory of Ethology Ecology & Evolution, Institute Estable, Montevideo, Uruguay

2 Section of Entomology, Faculty of Sciences, University of the Republic, Montevideo, Uruguay

Abstract

A female Pepsis cupripennis attacked females and large juve-

niles of Acanthoscurria suina and Eupalaestrus weijenberghi

in their own burrows, under laboratory conditions. Males

of both spider species were also attacked in open arenas

but with minor success. The wasp buried the immobilized

tarantulas in special chambers excavated from the spider’s

burrows. An egg was deposited by the wasp on the spider’s

abdomen and its development into an adult wasp was studied.

On two spiders the eggs were spoilt and the hosts completely

recovered. In the other six spiders development proceeded,

reaching the adult stage in ﬁve cases. Two female wasps were

obtained from large spiders and three males were obtained

from small hosts. The wasp apparently needed a spider six

times heavier than herself to obtain a daughter. Spider size

estimation by the wasp and ecological defensive mechanisms

of the spiders are discussed.

marginata on Cyrtopholis portoricae in Puerto Rico. He

described interactions in open laboratory arenas, including

attacks, immobilization of the spider and sucking of its

ﬂuids, burying of the spider, and placing an egg on the

spider’s abdomen. In the United States, Williams (1956)

studied the predation of some pomplilids on theraphosids

under laboratory conditions, mainly Pepsis thisbe on

Aphonopelma sp., describing interactions, burying of prey

and development of the wasp larvae until adult stage. Cazier

and Mortenson (1964) also reported a ﬁeld observation on

Pepsis formosa formosa dragging and burying a male of

Aphonopelma chalcodes . Punzo (1994a, b) reported hunting

and burying behaviors of Pepsis thisbe on Aphonopelma

echina in both ﬁeld and laboratory conditions, as well as the

larval development of the wasp.

Theraphosids show a high species richness in the

Neotropics, but paradoxically reports of pompilid predation

are limited to Puerto Rico. In Uruguay, Acanthoscurria

suina Pocock and Eupalaestrus weijenberghi (Thorell) are

the most common theraphosids and they are frequently sym-

patric, living in burrows in the meadows (Pérez-Miles et al.,

in press). Pepsis cupripennis Taschenberg is a large pompilid

(approximately 40 mm of body length) found in sympatry

with these theraphosids. In this paper, we report the preda-

tion of a single female of P. cupripennis on adult and large

juveniles of these tarantulas under laboratory conditions.

Burying behavior, larval development, and characteristics of

the wasp offspring are described.

Keywords: Burrowing tarantulas, pompilid predation,

Pepsis cupripennis , Acanthoscurria suina , Eupalaestrus

weijenberghi , Uruguay .

Introduction

Pompilid wasps are predator-parasitoids of large spiders.

Species of the genus Pepsis are characterized by their large

size and ability to capture the largest spiders, the thera-

phosids or tarantulas (Williams, 1956; Gertsch, 1979).

Petrunkevitch (1926) reported the predation of Pepsis

Received: 12 July 2002

Accepted: 17 January 2004

Correspondence : F.G. Costa, Laboratorio de Etología, Ecología y Evolución, Instituto de Investigaciones Biológicas Clemente Estable, Av.

Italia 3318, Montevideo, Uruguay. Fax: +59 82-4875548; E-mail: fgc@iibce.edu.uy

SNFE758Z04

F.G. Costa et al.

Material and methods

One female Pepsis cupripennis was collected in a meadow

on March 24, 2000 in Solymar Norte, Canelones, Uruguay

(Fig. 1). She was maintained in the laboratory in a cylindri-

cal container of 40 cm diameter and 21 cm height with soil

and leaf litter and fed with sugar solution and water. Wasp

– spider encounters involved seven individuals of E.

weijenberghi and six individuals of A. suina , mostly collected

in Canelones, Uruguay, with the exception of two females of

E. weijenberghi collected in Durazno and Flores, Uruguay.

Male spiders were housed in glass jars of 7.5 cm diameter

with soil. Female spiders and large juveniles were housed in

glass containers of 30 ¥ 15 cm base and 30 cm height with

soil (10 cm deep) where the spiders lived in burrows artiﬁ-

cially constructed against a glass wall and later adapted by

the spider. All spiders were fed mainly with cockroaches

( Blaptica dubia ) and provided with water.

For encounters with females and large juveniles, the wasp

was introduced into their containers. For encounters with

males, each male was introduced into the wasp’s container.

All behaviors of the wasp and the spiders were directly

observed, and two wasp attacks were also video-taped. If the

wasp attack did not occur within a period of 30 min the

experiment was ended. Otherwise, the observation was con-

tinued until the spider was buried, and thereafter the wasp

was returned to her container. Once the spider was buried,

we carefully excavated the chamber with the immobilized

spider, removed it and measured the chamber and funnel. The

immobilized spider was placed into an artiﬁcial observation

chamber made with soil and covered with glass to facilitate

observation. This chamber was moistened and observed

daily, and photographed twice a week. Wasp pupal cocoons

were periodically weighed. After the emergence of imagines,

sexual encounters were staged.

For the study of the sexual behavior of wasps, pairs were

placed in a glass container (47 ¥ 24 cm at base, 35 cm high)

with soil and living plants ( Tradescantia sp., Commeli-

naceae). In the experiments, each male was placed consecu-

tively with the same female during 30 min in order of male

weight, to elucidate female choice related to male size or

priority. In the ﬁrst experiment (December 7, 2000), males

were introduced in an increasing order and in the second

(December 11, 2000) in decreasing order. Adult wasps were

deposited in the Entomological Collection of Facultad de

Ciencias (EFC), Montevideo and in the Museo Argentino de

Ciencias Naturales, Buenos Aires. Rests of eaten spiders

were deposited in EFC.

Results

Wasp attacks on female spiders

In six of the seven wasp encounters with spiders in burrows,

the wasp entered the burrow attacking the spiders. One

female of A. suina was not attacked by the wasp. In this case,

the wasp had attacked and buried another spider only 24

hours before (Table 1). The other three attacks on A. suina

were successful. Of three attacks on E. weijenberghi , two

were successful and one failed to immobilize the spider

which was entrenched in the long and narrow end of her

burrow (terminal tube).

Usually the wasp found the burrow entrance and imme-

diately entered while vibrating its wings. In a case where the

burrow entrance was sealed by the spider with soil and silk,

the wasp removed the plug. The wasp walked slowly into the

burrow, waving her extended antennae. Frequently, the spider

advanced on the wasp and sometimes tried to attack her ( A.

suina ). When contacted with the spider, the wasp curved

her abdomen between her legs while intensively touching

the spider with her antennae. When the spider tried to

escape (frequently E. weijenberghi ) the wasp pursued her.

Eupalaestrus weijenberghi presented her abdomen towards

the wasp while A. suina tried to bite and/or to exit from the

burrow. The wasp tried to bite the spider’s legs to obtain

a point of support for stinging the spider repeatedly.

Stings were frequent around the sternum, and/or the coxal

joints. When the spider tried to escape from the burrow, the

wasp retained her by biting and hanging onto the spider

with her hind legs. One A. suina escaped but was recaptured

by the wasp near the burrow entrance. The spiders were

immobilized approximately 1 min after the start of the

attacks.

In one case when the terminal tube was short, the spider

( E. weijenberghi ) penetrated into the tube with the abdomen

pointing towards the wasp. The wasp tried to reach to the

spider’s cephalothorax extending her abdomen between

the spider’s body and the tube wall. Finally the wasp stung

the spider several times and later took the spider out of the

tube and immobilized her. In this case 22 min elapsed from

the beginning of the attack until the wasp moved away after

Fig. 1. Penultimate male (no. 2) of Acanthoscurria suina (left) in

recuperation from the wasp’s sting after the failure of the larva’s

development, and ﬁxed female Pepsis cupripennis (right, head–apex

= 40 mm length).

SNFE758Z04

Pompilid Interactions with Tarantulas

Table 1. Predation experiments carried out with one female Pepsis cupripennis versus Acanthoscurria suina (As) and Eupalaestrus weijen-

berghi (Ew) under laboratory conditions. Spiders no. 4 and 7 were repeatedly presented to the wasp.

Spider

Wasp

Spider

Larval

number

Species

Sex

Date

Attack

Arena

Immobilized

Buried

Fate

development

female

24 March

yes

burrow without tube

yes

eaten

complete

juvenile male

27 March

yes

burrow

yes

recuperation

small larva dead

female

29 March

yes

burrow, short tube

yes

eaten

complete

female

30 March

burrow

–

female

31 March

yes

burrow

yes

eaten

complete

female

05 April

yes

burrow, long tube

–

juvenile male

05 April

yes

burrow

yes

eaten

complete

male

07 April

open ﬁeld

–

male

07 April

yes

open ﬁeld

yes

recuperation

small larva dead

male

14 April

open ﬁeld

–

male

14 April

yes

open ﬁeld

yes

eaten

pupa dead

male

24 April

open ﬁeld

–

male

25 April

open ﬁeld

–

male

25 April

open ﬁeld

–

male

25 April

yes

open ﬁeld

yes

–

male

28 April

yes

open ﬁeld

yes

eaten

complete

drinking ﬂuids. In another case a female E. weijenberghi

moved deeply into a long and narrow tube where the wasp

could not reach her. The wasp spent 27 min trying to sting

the spider but then abandoned the burrow, leaving the spider

unharmed.

Males also used their long legs to maintain distance from the

wasp. Of four successful attacks on males only one ‘ﬁnal

sting’ was observed.

Spider immobilization

Once a spider was immobilized, the wasp inserted her sting

deeply into the spider’s mouthparts while biting the coxae of

the anterior legs or the palps. After this ﬁnal sting, the wasp

cleaned her antennae, wings, and legs, and rubbed her body

against the soil. The wasp alternated these grooming behav-

iors with drinking ﬂuids from the sting sites and from the

spider’s mouthparts. If the spider was immobilized outside

the burrow, the wasp transported her back inside the burrow.

In other cases, the wasp removed the spider from the burrow

to facilitate the excavation of a nest chamber, returning

the spider after that. In these cases, the wasp periodically

alternated excavation with contacts with the spider. Males

of A. suina were buried by the wasp, while the male of E.

weijenberghi was abandoned.

Wasp attacks on male spiders

Of nine wasp encounters with male spiders in the open arena,

only four resulted in successful attacks: all of three encoun-

ters with A. suina and one (of six) with E. weijenberghi . Both

spider species showed different defensive behaviors in the

presence of the wasp. Eupalaestrus weijenberghi elevated the

body and raised the abdomen somewhat, constantly orient-

ing it towards the wasp. Usually E. weijenberghi males did

not try to escape from the wasp. The wasp investigated the

spider with the extended antennae, performed few wing

vibrations and curved her body moderately but she did not

attack except in one case when she stung the male that died

and was abandoned. In the other cases, the spider and the

wasp ignored one another after the encounter.

Males of A. suina usually threatened the wasp by raising

of the forelegs and opening of the chelicerae. Wasp attacks

were similar as described above and spiders tried to bite the

wasp and/or escape. Male defensive actions in the open arena

led the wasp to spend more time and effort than females in

burrows. The wasp spent 11.6 min (±0.8, four cases) from

start to ﬁnal of attacks on males whereas attacks on females

lasted only 5.3 min (±2.3, three cases, with the female in the

short terminal tube eliminated). Males frequently escaped

and were pursued by the wasp which performed short ﬂights.

Burying and oviposition

Those spiders that were captured and immobilized in their

burrows (ﬁve cases) were also buried in them. The wasp con-

structed a horizontal funnel which started in the bowl of the

spider burrow and ended in a chamber. The chamber was

shaped like a prism, with rounded corners, and its walls were

thoroughly compacted as a result of pressure applied by the

apex of the wasp’s abdomen. Dimensions of funnel and

chamber are given in Table 2. A signiﬁcantly positive

SNFE758Z04

F.G. Costa et al.

Table 2.

Dimensions of brood chamber, funnel and egg size of Pepsis cupripennis.

Spider

Chamber

Funnel

Number

Species

Sex

length (cm)

width (cm)

height (cm)

length (cm)

Egg size (cm)

female

6.0

4.0

2.5

–

0.6 ¥ 0.2

juvenile male

6.2

3.5

3.0

1.5

0.6 ¥ 0.2

female

6.1

3.0

2.9

5.8

0.6 ¥ 0.2

female

5.5

2.7

2.2

7.5

–

juvenile male

7.5

3.0

2.2

6.5

–

male

7.0

3.0

2.5

12.0

–

male

6.0

4.0

2.5

10.0

–

male

7.0

3.5

2.8

–

0.5 ¥ 0.2

Wasp development

The development of the wasps from egg to adult emergence

took a mean of 237.40 days (±15.63 SD), with an egg-larval

period of 18.67 ± 2.25, and a pupal phase of 218.60 ± 17.78

days. The egg–larval growth sequence is given in Figs. 3, 4.

Except the cuticle, the entire spider including legs and palps

was eaten by the wasp larva. The larva reached narrow parts

of the spider body due to its small head. Once the spider was

eaten, the larva spun an opaque pale brown cocoon with

mouth silk and moulted to a pupa, remaining in this state

until the warm season. Immediately after cocooning, a wet

zone appeared at the narrow pole of the cocoon, probably due

to the ﬁrst defecation (‘meconium’). Cocoon size averaged

4.29 ± 0.41 cm in length and 1.63 ± 0.19 cm in diameter. The

changes of pupal weight are given in Fig. 5, with the initial

pronounced loss of weight coinciding with defecation. After

that and until wasp emergence, cocoon weight changed only

slightly.

Fig. 2. Spider weight in relation to the volume of the brood

chamber. A signiﬁcant correlation ( r = 0.93, p < 0.05) was found

between females’ and large juveniles’ weights (circles) and chamber

volume. Adult males (triangles) do not seem to follow this trend.

Wasp emergence

correlation was found between female or juvenile spider

weight and estimated chamber volume (Fig. 2). The three

male spiders captured at open arenas were transported to and

buried in burrows excavated by the wasp. The funnel started

at the soil surface followed by an inclined portion, turned

horizontally next to the container ﬂoor and ended in an ovoid

chamber. Male chamber dimensions seem to be a different

data population of major values with respect to previous ﬁve

cases (Fig. 2).

All spiders were found in the chamber with their legs

pointing towards the funnel; probably the spiders were pulled

by the wasp by their spinnerets. In all buried spiders, the

funnel was completely ﬁlled with compacted soil making the

entrance indistinguishable from the surrounding soil or bowl

wall. The spiders were lying on their dorsum or side. A single

egg was always deposited on the most elevated part of the

abdomen. The egg was usually placed near the posterior book

lungs, under the epigastric furrow or near the caudal extreme

of the abdomen. The female wasp died on 4 May, 2000, 41

days after capture.

Of the six cocoons obtained, ﬁve wasps emerged successfully

while the sixth (a male) died due to fungal infestation during

the pupal stage. Wasps emerged by cutting a cap into the

widest pole of the cocoon (Fig. 6). Three males emerged

simultaneously on November 26, one female emerged on

November 30, and another female emerged on December 3.

Female wasps were obtained from heavy spiders ( E.

weijenberghi females) while males were obtained from

lighter ones (Table 3). The entire yield of the spider by the

wasp was very low, averaging 15.96 ± 5.55%.

Sexual encounters of wasps

Experimental copulations among adult wasps were tried

on December 7 and 11. Each male was introduced into the

female’s container for 30 minutes. In the ﬁrst trial, males

were introduced one at a time, in an order of decreasing

weight. The ﬁrst (=largest) male copulated 9.2 min. He

mounted on the female’s back, and then remained hanging

SNFE758Z04

Pompilid Interactions with Tarantulas

Figs. 3, 4. Development of the wasp Pepsis cupripennis on a male of Acanthoscurria suina under laboratory conditions . (3) Egg. (4) Almost

mature larva (four days before pupation).

Fig. 6. Open pupal cocoon of the same wasp as in Figs. 3 and 4

after emergence of an adult male.

second male mated twice, during 9 and 13 s respectively, in

the mounted position. The third male copulated for 60 s in

the mounted position.

Fig. 5. Progressive loss of weight from pupation to adult wasp

emergence, through time. Initial abrupt decrease corresponds to ﬁrst

defecation. Terminal right points correspond to emerged adult

wasps: empty triangles and empty squares are females, the asterisks

correspond to a male that did not emerge due to fungal infestation,

and the other are adult males.

Recuperation of two spiders

In two cases, the eggs fell off the spider’s abdomen and died

due to the manipulation. Both spiders gradually recovered

their motility and coordination. After eight months they fed

and showed threat displays. One of the spiders, collected as

an adult male, showed a longer life span (10 months since

the death of the egg) than the other males reared in the lab-

oratory (4.14 ± 1.92 months, n = 20). The other spider

moulted to an adult male (ten months after sting) and had a

normal life-span (ﬁve months as an adult).

from their copulatory organs. The second male copulated for

20 s in the mounting position while the third male did not

copulate. In the second trial, males were introduced in an

order of increasing weight. The ﬁrst (=smallest) male copu-

lated 5.5 min in both the mounting and hanging position. The

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: