LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVlNOSHliyMS S3ldVdan LIBRA

D3

;d
>
po

)N^ NOIinillSNI NVINOSHillNS _S3 IBVBan_LIBRARI E S ^SMITHSONIAN “ INSTITUTION ^ NOIlfll

CO

Z </> Z </> ‘Z to ^ z

n_LIBRARIES SMITHSONIAN_INSTITUTION^NOIinj.liSNI_NVINOSHilWS S3iavaan_LIBRA

h

■oy oQ 5 5 V 2 5 ''^

)N^NOIinj.liSNl"'NVINOSHilWS S3iavaan~'LIBRARI ES^ SMITHSONIAN"" INSTITUTION ^NOlini

Z !“> Z r- Z t- z

_ m ^ ^ \,jvAsvi>' rn

± c/) ■ " f: ui ^ ^ CO

II LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIIMS S3ldVaan LIBRA
z , ^ z: > CO z: co ^ z >

to V. CO (S^ fo CO y^': CO CO to

CO

> I | ." I

)N NOlinillSNI NVIN0SHillNs‘^S3iavaan^LIBRARIEs'^SMITHS0NIAN INSTITUTION NOlini

to “HCO — CO = CO

CO

<

O ■• Q “ O

Z ‘ _J 2 _j Z — J

JT LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIIMS S3ldVaan LIBRA

m ^ m x^om^ m

CO “ CO £ — (/) ± CO

DN NOlinillSNI NVINOSHIIVMS SBIdVdail LIBRARIES SMITHSONIAN INSTITUTION NOlini

CO z CO 2:

X

z

> > -w~ S ^

11 LIBRARIES'^SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SHimS^S3 I B Vd 8 11 LIBRA
to ~ to CO

uj llJ

CO

i.\ —

A

^ < >/
cc

O \o\dC^ ~ O — x^iust^ O ~ O

DN NOIiniliSNl”‘NVINOSHillNS S3 I a Va a IT ""l I B R AR I ES^ SMITHS0NIAN""|NSTITUTI0N NOlini

NViNOSHims b'3iavdan libraries Smithsonian institution Nuiiiiii±bNi_NviNUbMxi/M;

CO ~ CO ~ CO \

CO

_ _

O _ O

-I z -J :z

SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHII/JS SBIdVaaH LIBRARIES SMITHSONlAi

m x;i£Uij5^ rn

CO ± CO _

NViNOSHiiiNS SBiavaan libraries Smithsonian institution NoiiniiiSNi nvinoshiii^'

CO Z . CO 2 V CO 2

to :jW.3W CO CO (5#^ £0 ^ (O : co

? 2 t 2

2 CO *2 (/) ^Z CO

SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIIA/S S3IBVaan LIBRARIES SMITHSONIAr

CO — 00 i: CO

2 ~J 2 '' _J 2

NVINOSHimS S3IBVaan libraries SMITHSONIAN institution NOIinillSNI__NVINOSHlBN!

> z: r- 2 r- z:

— CO ' E CO £ CO

SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHlIl^S SaiaVdaH LIBRARIES SMITHSONlAi

2 > CO 2 CO 2

o X o j: o X

CO
O
2

> S X4^osv^^ >

NViNOSHims"^ S3 1 dva a n^LI BRAR I Es"^SMITHSONlAN INSTITUTION NOlinillSNI _ NVINOSHIK^

CO n CO “ CO 2

O _ O

-j 2 _» 2

SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHimS S3IBVyai1 LIBRARIES SMilHSGMiA
“ 2

O

/ j’ *

CO

m m

CO £ — CO

NViNOSHims S3IBVBan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHlll^

CO "ZL ^ Z CO

2 '' > XlO-ciis^ >

► 2 CO *2 (/) ^ z in

SMITHSONIAN INSTITUTION NOIlOiliSNI NVINOSHillNS S3iavaan LIBRARIES SMITHSONlAi

) X CO — CO X CO

MWI MHQU I I lAlC Q "3 I W V VJ a I n I ID DAD ICC QK^ITUCOMIAM IMCTiTIITinM M r-t 1 I n I I I C M I MW I Ki.' |C; M I i

The Journal of

ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 17

SPRING 1989

NUMBER 1

THE JOURNAL OF ARACHNOLOGY

EDITOR: James E. Carico, Lynchburg College
ASSOCIATE EDITOR: Jerome S. Rovner, Ohio University

EDITORIAL BOARD: J. E. Carrel, University of Missouri; J. A.
Coddington, National Museum of Natural History, Smithsonian Institution;

J. C. Cokendolpher, Texas Tech University; F. A. Coyle, Western Carolina
University; C. D. Dondale, Agriculture Canada; W. G. Eberhard,

Universidad de Costa Rica; M. E. Galiano, Museo Argentine de Ciencias
Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; N. V. Horner,
Midwestern State University; D. T. Jennings, NEFES, Morgantown, West
Virginia; V. F. Lee, California Academy of Sciences; H. W. Levi, Harvard
University; E. A. Maury, Museo Argentine de Ciencias Naturales; M. H.
Muma, Western New Mexico University; N. 1. Platnick, American Museum
of Natural History; G. A. Polis, Vanderbilt University; S. E. Riechert,
University of Tennessee; A. L. Rypstra, Miami University, Ohio; M. H.
Robinson, U.S. National Zoological Park; W. A. Shear, Hampden-Sydney
College; G. W. Uetz, University of Cincinnati; C. E. Valerio, Universidad de
Costa Rica.

THE JOURNAL OE ARACHNOLOGY (ISSN 0161-8202) is published in
Spring, Summer, and Fall by The American Arachnological Society at Texas
Tech Press.

Individual subscriptions, which include membership in the Society, are $30.00
for regular members, $20.00 for student members. Institutional subscriptions to
The Journal of Arachnology are $50.00. Correspondence concerning subscriptions
and memberships should be addressed to the Membership Secretary (see back
inside cover). Remittances should be made payable to The American
Arachnological Society. Inquiries about availability and current prices of back
issues should be sent to Dr. Susan E. Riechert, Department of Zoology,
University of Tennessee, Knoxville, TN 37916 USA. Correspondence concerning
undelivered issues should be addressed to the Texas Tech Press, Texas Tech
University, Lubbock, Texas 79409 USA.

Change of address notices must be sent to the Membership Secretary.

Detailed instructions for the preparation of manuscripts appear in the Fall
issue of each year, and can also be obtained from the Editor and the Associate
Editor. Manuscripts that do not follow those instructions will be returned to the
author(s) without benefit of review. Manuscripts and all related correspondence
must be sent to Dr. Jerome S. Rovner, Associate Editor, Department of
Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701 USA.

P ROOTS, REPRINTS, and CHARGES: Authors will receive a reprint order
form along with their proofs. Reprints are billed at the printer’s current schedule
of costs. All authors will be billed by The American Arachnological Society for
page charges. The charge per journal page will be assessed as follows: $50. 00-
nonmembers; $35. 00-members acknowledging grant, institutional, or other
support; $25. 00-full members who do not have support; $20. 00-student members
who do not coauthor with full members and who do not acknowledge support.
Members may have page charges reduced for a limited number of pages; a
statement of justification for reduction should accompany the manuscript when it
is first sent to the Associate Editor for review.

This publication is printed on acid-free paper.

Gonzalez, A. 1989. Estudio de la evolucion de la tricobothriotaxia a traves del desarrollo
postembrionario de Theridion rufipes (Araneae, Theridiidae). J. ArachnoL, 17:1-6.

ESTUDIO DE LA EVOLUCION DE LA TRICOBOTRIOTAXIA
A TRAVES DEL DESARROLLO POSTEMBRIONARIO DE
THERIDION RUFIPES (ARANEAE, THERIDIIDAEf

Alda Gonzalez^

Facultad de Ciencias Naturales y Museo
Paseo del Bosque s/ri, 1900 La Plata, Argentina

ABSTRACT

Trichobothriotaxic evolution was studied through the postembryonic development of Theridion
rufipes. It was found that this species belongs to the piesiomorphic pattern (Lehtinen 1980) and only
presents basal growth. The number of trichobothria on the tibia of the male palp is three and on the
female palp is five. The total number of trichobothria on specimens of different sexes is similar
because the same number of molts is necessary for reaching adult stage, being six trichobothria on the
tibia of leg IV for both sexes. On the tibia of leg I the number of trichobothria is five in males and six
in females. The individual variation between right side and left or between specimens of different sexes
is not significant.

EXTRACTO

En el presente trabajo se estudio la evolucion de la tricobotriotaxia a traves del desarrollo
postembrionario de Theridion rufipes. Se demostro que esta especie pertenece al patron plesiomorfico
(Lehtinen 1980); posee unicamente crecimiento basal; el numero de tricobotrias de las tibias de los
palpos en machos y hembras es de tres y cinco respectivamente; el numero total de tricobotrias es
semejante en ambos sexos porque igual numero de mudas es necesario para alcanzar el estado aduito
(en la tibia de la pata IV es igual a seis para ambos sexos y en la tibia de la pata I es igual a cinco en
los machos y seis en las hembras); la variacion individual, entre el lado derecho e izquierdo o entre
ejempiares de distinto sexo no es significativa.

INTRODUCCION

Las unicas investigaciones existentes sobre el estudio tricobotriotaxico de
aranas de la familia Theridiidae (Gonzalez 1980, 1984) se refieren a especies del
genero Latrodectus Walckenaer 1805,

En ei presente trabajo se llevo a cabo el estudio de la evolucion de las
tricobotrias a lo largo del desarrollo postembrionario de Theridion rufipes,
determioando: (T) d momento de aparicion, (2) la distribucion en los artejos de
los apendices estudiados, (3) el numero total por artejo, (4) el modo de aparicion,
(5) la variacion individual entre los apendices del lado derecho e izquierdo de un
individuo en un estado del desarrollo y entre ejempiares de distinto sexo.

'Contribucion N° 175 del Centro de Estudios Parasitologicos y de Vectores (CEPAVE). Facultad de
Ciencias Naturales y Museo de la Universidad Nacional de La Plata.

^Miembro de la Carrera del Investigador del Consejo Nacional de Investigaciones Cientificas y
Tecnicas (CONICET).

2

THE JOURNAL OF ARACHNOLOGY

Para tal fin se efectuo: (1) el seguimiento de la evolucion de los organos
mecanoreceptores en las tibias de palpo, pata I y pata IV y en el metatarso de
pata I y pata IV durante el desarrollo en ambos sexos, (2) la comparacion de
varias series procedentes de una misma madre, (3) la comparacion de los artejos
de los apendices opuestos de un mismo par en un mismo ejemplar, (4) la
comparacion de las tablas tricobotriotaxicas de machos y hembras en un mismo
estado.

MATERIAL Y METODOS

Para realizar las investigaciones se utilizaron las exuvias sucesivas de machos y
de hembras y los ejemplares adultos de ambos sexos. Las observaciones se
efectuaron a partir de la cuarta exuvia por ser esta la primera que presenta
tricobotrias, (la cuarta exuvia es la primera que se produce fuera de la ooteca,
contando como muda uno el desprendimiento de la cuticula embrional que
arrastra consigo al diente de eclosion).

Todo el material empleado fue criado en el laboratorio a partir de los desoves
de aranas colectadas en el campo. Las condiciones de laboratorio y los metodos
de cria fueron explicados en una publicacion anterior (Gonzalez 1979). Las
observaciones se realizaron en seco, sobre lotes de 20 exuvias de cada estado.

Se empleo para la identificacion de las tricobotrias la nomenclatura espacio-
temporal propuesta por Emerit y Bonaric (1975).

RESULTADOS OBTENIDOS

Theridion rufipes, como todas las aranas de la familia Theridiidae, responde al
patron plesiomorfico o grupo 1 (Lehtinen 1980). Es decir que presenta todas las
tricobotrias en la zona dorsal del artejo con una unica tricobotria metatarsal
subdistal y dos lineas paralelas tibiales. No se observan tricobotrias en tarso y
femur.

Modo de aparicion. — En esta especie se observe un crecimiento basal. La
primera tricobotria (tricobotria promotora) de una serie surge en la zona
proximal del territorio respective, luego en mudas sucesivas, se aleja hacia la
zona distal del mismo, para dar lugar a la aparicion de nuevas tricobotrias, las
que se reconocen facilmente por ser mas cortas y de cupulas mas pequenas. No se
ha observado el crecimiento intercalar citado por Emerit y Bonaric (1975) y
observado para Latrodectus mirabilis (Holmberg 1876) (Gonzalez 1980).

Tricobotrias tibiales. — En los apendices estudiados se reconocieron dos
territories: (1) proximal (T) desde la mitad del artejo hacia la articulacion tibia-
patela, (2) distal (S) desde la mitad del artejo hacia la articulacion tibia-
metatarso,

En cada territorio se hallan comprendidos dos campos: anterior (A) y posterior
(P) teniendo en cuenta la linea media dorsal del artejo.

En la nomenclatura empleada, el mimero anterior a las letras que identifican
los territories y sus campos corresponden al estado del desarrollo en que aparece
la tricobotria y el numero posterior, a la filiacion de la misma.

Palpos: Los ejemplares adultos, machos y hembras, poseen tres y cinco
tricobotrias respectivamente. Esta caracteristica ha sido observada para las

GONZALEZ— EVOLUCION TRICOBOTHRIOTAXIA DE THERIDION RUFIPES

3

Tabla I. — Evolucion tricobotriotaxica tibial en machos.

Paipo

Pata I

Pata IV

P

A

P

A

P

A

s

4~PS1

5- PSI

6- PS2

4- PSl

5- PS2

7-ASI

T

6-PTl

5= AT I

4-PTl

4=- AT I
5~AT2

5-PTl

4- ATI
5=AT2

especies de la familia Theridiidae tratadas en trabajos anteriores (Gonzalez 1980,
1984). No obstante el orden de aparacion de los organos mecanoreceptores y su
distribucion, difieren entre Theridion rufipes y las especies del genero Latrodectus
estudiadas.

En los machos de Theridion rufipes aparece en ei estado cuatro la tricobotria
PSi, en el estado cinco la ATI y en ei estado seis la PTL

En las hembras se observo en el estado cuatro la PSI y en estado cinco la ATI,
esto concuerda con la evolucion dada en los machos. En el estado seis de las
hembras no surge ningiina tricobotria nueva. En el siguiente estado aparecen
juntas la AT2 y la PTl, completandose ei numero de cinco tricobotrias, con la
aparicioE de la PT2 en el estado adulto.

En las Tabias 1 y 2 y en la Figura 1 se halla representada la evolucion,
distribucion y numero de las tricobotrias para las tibias de los palpos de machos
y de hembras separadamente.

Pata I y Pata IV: En la pata I, en el estado cuatro, siirgen dos tricobotrias en
los machos y en las hembras, la ATI y la PTl; en el estado cinco, coincide para
ambos sexos, la aparicion de la AT2 y la PSL A partir de aqui, difiere el orden
de aparacion. En los machos se observa una nueva tricobotria en el estado seis, la
PS2, la cual aparece en las hembras en el estado ocho. La tricobotria ASl surge
en las hembras en el estado adulto, no existiendo en los machos ninguna
tricobotria en el campo anterior del territorio distal

En la pata IV, machos y hembras presentan el mismo numero de tricobotrias
(seis) y la misma distribucion, pero difieren en la secuencia de aparacion. En el
estado cuatro aparecen la ATI y la PSI en los dos sexos. En el estado cinco, en
los machos, se observa la PTl, la AT2 y la PS2, mientras que en las hembras en
este estado solo aparecen la PTl y la AT2. La PS2 surge en las hembras en el
estado siete, junto con la ASl, la cual aparece en ese mismo estado en los
machos.

Las Tabias 1 y 2 y las Figuras 2 y 3 muestran la evolucion tricobotriotaxica y
los mapas tricobotriales de m.achos y hembras para ambos apendices. Del analisis

Tabla 2. — Evolucion tricobotriotaxica tibial en hembras.

Paipo

Pata I

Pata IV

P

A

P

A

P

A

S

4™PSI

5==PS1

8-PS2

9^AS1

4.»PS1

7-PS2

7-= ASl

T

7-PTl

9^PT2

5~AT1

7=^AT2

4-PTl

4^ AT I
5-AT2

5-PTl

4^ AT I
5-AT2

4

THE JOURNAL OF ARACHNOLOGY

Qpsi

E4

AT1/b A /
E5

PALPO

atim oVti

©)PS1

E6

2

PR1

METATARSO

AT2

AT1

n

PT1

E5

PATA 1

AT1

AT2

PT1

PS2

PS1

AT2

AT1

AS1

E7

0.

°0

AT10

E4

PATA 4 aT2

AT1

E5

r\

Sj

PT1
PS1
PS 2

u

E5

PT1

PS1

.. M.

AT 2

n

AT1

U

©

PT1

AS1

0 A

PS2

U

PS1

E7

PT1 ^
PS1 X

Figuras 1-3. — Mapas tricobotriotaxicos de Theridion rufipes: I, metatarso de pata I; 2, tibia de
palpo de machos y hembras; 3, tibia de pata I y pata IV de machos y hembras. E = estados del
desarrollo.

de los mismos surge, que si bien el orden de aparacion de las tricobotrias difiere
entre machos y hembras, el numero total de las mismas presenta escasas
diferencias, lo que se debe, probablemente, a que ambos efectuan el mismo
numero de mudas para llegar al estado adulto.

GONZALEZ— EVOLUCION TRICOBOTHRIOTAXIA DE THERIDION RUFIPES

5

Tabla 3. — Variacion de la tricobotriotaxia tibial en fundon de los sexos. x promedio de
tricobotrias de palpo + pata 1 + pata IV; S == desviacion tipica; n = niimero de individuos; Ex =
estado del desarrollo.

Ex

Macho

Hembra

n

X ± 5

n

T + S

t

4

5

5.00 ± 0

5

5.00 ± 0

—

5

6

11.00 ± 0.81

5

10.00 ± 0.63

2.04

6

5

13.20 ± 0.74

7

11.00 + 1.07

3.86

7

6

13.16 + 0.69

7

13.85 ± 0.83

2.41

8

6

13.83 ± 0.69

6

14.66 ± 0.94

1.90

9

6

!4.33 ± 0.94

6

16.83 ± 0.69

4.5

Tricobotrias metatarsales. — De las observaciones efectuadas surge la existencia
de una tricobotria metatarsal solo en la pata 1. Esta se encuentra ubicada en el
campo posterior (P) del territorio proximal (R). Aparece en la cuarta exuvia y
permanece invariable a lo largo del desarrolo (Fig. 4, PRl).

Variacion tricobotriotaxica. — Emerit y Bonaric (1975) se refieren a la variacion
tricobotriotaxica individual, como a la variacion que puede existir respecto del
numero de tricobotrias entre los apendices derecho e izquierdo de! mismo par y
del mismo individuo, y a la variacion del numero de tricobotrias entre ejempiares
de distinto sexo, para un mismo estado del desarrollo. El poder utilizar la
tricobotriotaxia para delimitar ei estado del desarrollo al que pertenece un
individuo, depende de la variabilidad de estos dos factores citados.

En la Tabla 3 se demuestra que las diferencias entre el numero de tricobotrias
de machos y hembras para un mismo estado, no es significativa para ningun
porcentaje de probabilidad (test de Student). Podemos deck entonces que la
tricobotriotaxia de Theridion rufipes no presenta diferencias significativas entre
los sexos para determinar el estado del desarrollo de un ejempiar.

Tabla 4. — Variacion de la tricobotriotaxia tibial entre el lado derecho e izquierdo de un mismo
ejempiar. x = promedio de tricobotrias de palpo + pata I + pata IV; S = desviacion tipica; n =
numero de individuos; Ex = estados del desarrollo.

Derecho

Izquierdo

Ex

n y

X ± 5 (y)

n y

T + S (y)

t

MALES

4

5

5.00 ± 0

5

5.00 + 0

—

5

6

I LOO ± 0.81

6

11.83 + 0.90

1.9

6

5

13.20 ± 0.74

5

12.20 ± 0.4

2.38

7

5

13.00 ± 0.63

5

14.00 + 0.89

1.97

8

7

13.71 ± 0.70

6

14.83 ± 0.90

2.24

9

6

13.16 + 0.69

5

14.60 ± 0.80

2.94

FEMALES

4

5

5.00 ± 0

5

5.00 ± 0

—

5

5

10.00 + 0.63

6

10.83 + 0.69

2.0

6

7

11.00 ± 1.07

6

10.00 ± 0.82

1.8

7

7

13.85 ± 0.83

7

13.00 ± 0.75

2.02

8

6

14.66 ± 0.94

6

15.60 ± 0.47

2.09

9

6

16.83 ± 0.69

5

17.60 ± 0.49

1.95

6

THE JOURNAL OF ARACHNOLOGY

En la Tabla 4 se compara ios apendices opuestos de un mismo ejemplar a
traves de su desarroilo. Las diferencias observadas no son significativas, ni para
probabilidades del 95%, ni del 99% (test de Student), por lo que podemos tomar
indistintamente los valores de derecha o izquierda de un individuo para los
estudios tricobotriotaxicos y poder determinar asi el estado del desarroilo de un
individuo.

CONCLUSIONES

1 — La especie Theridion rufipes respoiide al patron plesiomorfico (Lehtinen

1980).

2 — La tricobotria mas antigua de una serie, aparece en la zona proximal del
territorio y se va alejando hacia la zona distal del mismo, dando lugar a la
aparicion de tricobotrias nuevas (crecimiento basal).

3 — Las tricobotrias nuevas se reconocen facilmente por ser mas cortas y de

cupulas mas pequenas.

4 — El numero de organos mecanoreceptores tibiales de los palpos de los
machos y de las hembras es de tres y cinco respectivamente, estableciendose esto,
como una caracteristica de la familia Theridiidae.

5 — El numero total de tricobotrias de los ejemplares adultos de distinto sexo es
semejante, por realizar machos y hembras la misma cantidad de mudas para
alcanzar el estado adulto.

6 — La variacion individual de los organos mecanoreceptores que se puede
presentar entre apendices opuestos de un individuo en un estado dado del
desarroilo o entre ejemplares de distintos sexos, no es significativa para Theridion
rufipes, pudiendose emplear la tricobotriotaxia para reconocer estadios.

AGRADECIMIENTOS

A la Profesora Maria Elena Galiano por la lectura critica del manuscrito y su
constante asesoramiento y a la Senora Maria Cristina Estivariz por el delineado
de los dibujos.

LITERATURA CITADA

Emerit, M. y J. Bonaric, 1975. Contribution a I’etude du developpment de I’appareil mecanorecepteur
des Araignees: trichobothriotaxia de Pisaura mirabilis (Araneae, Pisauridae). Zool. Jb. Anat. Bd.,
94:358-374.

Gonzalez, A. 1980. Estudio de la evolucion de los organos mecanoreceptores a traves del desarroilo
post-embrionario de Latrodectus mirabilis, Latrodectus antheratus y Latrodectus corallinus
(Araneae, Theridiidae). Rev. Soc. Entomol. Argentina, 39:207-219.

Gonzalez, A. 1984. Desarroilo post-embrionario y evolucion de los organos mecanoreceptores de
Latrodectus diaguita Carcavallo y estudio de la tricobotriotaxia de Latrodectus quartus Abalos
(Araneae, Theridiidae). Physis Secc. C, 42:1-5.

Lehtinen, P. T. 1980. Trichobothrial patterns in high level taxonomy of spiders. Proc. 8th Int.
Arachnologen Kongress Wien, 493-498.

Lucas, H. 1846. Historic naturelle des Animaux articules. Pp. 89-271, In Exploration Scientifique de
FAlgerie pendant les annees 1840, 1841, 1842, publiee par ordre du Gouvernement et avec le
concours d’une commission academique. Sciences Physiques, Zoologie, 1.

Manuscript received October 1987, revised April 1988.

Hoffman, K. M. and J. R. Brushwein. 1989. Species of spiders (Araneae) associated with the
immature stages of Mantispa pukhella (Neuroptera, Mantispidae). J. Arachnol., 17:7-14.

SPECIES OF SPIDERS (ARANEAE) ASSOCIATED WITH THE
IMMATURE STAGES OF MANTISPA PULCHELLA

(NEUROPTERA, MANTISPIDAE)

Kevin M. Hoffman

Department of Entomology
Clemson University

Clemson, South Carolina 29634-0365 USA
and

Jeffrey R. Brushwein

University of Florida, IFAS
Citrus Research and Education Center
700 Experiment Station Road
Lake Alfred, Florida 33850-2299 USA

ABSTRACT

Species of spiders associated with the immature stages of Mantispa pukhella (Banks) are presented
for the first time. Twenty species in 15 genera from the families Philodromidae, Anyphaenidae,
Oxyopidae, Thomisidae, and Salticidae harbored immature stages of M. pukhella in South Carolina.
Some characteristics of the life history of M. pukhella are analyzed and compared with those of other
North American species of Mantispinae.

The cosmopolitan family Mantispidae was recently divided into the subfamilies
Symphrasinae, Drepanicinae, Calomantispinae, and Mantispinae (Lambkin 1986).
Little is known about the life histories of the first three subfamilies. Several
species of the Symphrasinae have been associated with nests of aculeate
Hymenoptera as well as with pupae of Noctuidae and Scarabaeidae (Parker and
Stange 1965), and one species of the Drepanicinae has been associated with a
spider egg sac (Austin 1985). Natural history of the Calomantispinae remains
unknown, although MacLeod and Redborg (1982) suggested, based on laboratory
rearings, that larvae of the Symphrasinae and Calomantispinae were generalist
predators of sedentary arthropod prey. In contrast, adult Mantispinae have been
reared exclusively from spider egg sacs (Redborg and MacLeod 1985; Brushwein
1986).

Female mantispines are not known to oviposit either on spider egg sacs or on
spiders, and instead deposit clutches of approximately 200 to 2000 stalked eggs
on leaves, pieces of wood, and other objects (McKeown and Mincham 1948;
LaSalle 1986; Rice 1986b). Following eclosion, larvae locate and gain access to
spider eggs by using one or both of two general strategies. Larvae may either seek
out and penetrate previously deposited egg sacs or board female spiders and enter
egg sacs as they are being formed (Redborg and MacLeod 1985). Larvae feed on

THE JOURNAL OF ARACHNOLOGY

spider eggs by piercing the chorion and draining the contents. Pupation occurs
within the egg sac, and pharate adults exit both their own cocoons and the egg
sacs before emerging from the pupal skin.

The mantispine Mantispa pulchella (Banks) has been recorded previously from
Utah, southern Illinois, Georgia, North Carolina, and Panama (Banks 1912;
Hughes-Schrader 1969; Redborg 1976; Hoffman and Hamilton 1988). Both
Redborg and MacLeod (1985) and Brushwein (1986) reported the successful
rearing of this species during studies on other species, but no further details were
presented. The present paper records the presence of M. pulchella in South
Carolina for the first time, documents spider associations, analyzes some
characteristics of the life history of this species, and compares these characteristics
with those of other North American Mantispinae.

METHODS

The majority of the immature stages of M, pulchella and associated spiders

were collected by the authors from March 1986 through April 1988 in and
around Clemson, Pickens County, South Carolina. Spiders and spider egg sacs of
as many different taxa as could be collected visually were examined for the
presence of mantispine immatures. In addition, the spider collection within the
Clemson University Arthropod Collection (CUAC) was searched and eight first
instar M. pulchella were located. Prior to the present study, the CUAC contained
29 families, 101 genera, and 162 species of spiders from South Carolina (Gaddy
and Morse 1985). However, the collections made during the course of the present
study, that of Brushwein (1986), and the identification of previously undetermined
material encountered during examination of the CUAC have increased the
collection to 32 families, 167 genera, and 290 species.

Four adult M. pulchella were collected in 1986 in an ultraviolet light trap run
nightly from January 1984 to October 1986, and 52 additional adults were located
in the CUAC. The latter specimens had been collected from the South Carolina
counties of Barnwell, Dorchester, Florence, Oconee, Pickens, and York from 1956
to 1986.

Identity of the immature stages of M. pulchella was confirmed by the
subsequent rearing of some {n ~ 18) to adults, using procedures detailed elsewhere
(Brushwein 1986). Some juvenile spiders were also reared to maturity for species-
level identifications. Spiders were identified both by the use of selected taxonomic
references (Edwards 1958; Kaston 1973; Platnick 1974; Dondale and Redner
1976, 1978; and Roth 1985) and with the kind assistance of A. R. Brady, G. B.
Edwards, J. H. Redner, and S. H. Roach. Voucher specimens of M. pulchella
and associated spiders were deposited in the CUAC, Department of Entomology,
Clemson University.

RESULTS

Spiders were each associated with a single immature M. pulchella. First instars
of M. pulchella were found either wrapped dorsally over the pedicel and posterior
edge of the carapace of spiders (Fig. 1) or inside spider egg sacs, whereas all other
immature stages were found only inside egg sacs. Spiders with larvae on them

HOFFMAN & BRUSHWEIN— M/4yvr/N/M PULCHELLA SPIDER ASSOCIATIONS

9

1 mm

h— ^

Figure I. — Juvenile Anyphaena sp. with Mantispa pulchella larva in dorsal pedicel region, left
lateral view. Palpal and leg segments distal to coxae omitted.

usually had discolored patches present on one or both sides of the pedicel; these
patches were located ventrolaterally on the anterior portion of the pedicel or on
the membranous area just below the posterior edge of the carapace. The dark
patches were at the approximate location that larval mouthparts would come in
contact with the spider integument.

Members of the family Anyphaenidae were the most commonly collected
spiders associated with M. pulchella and comprised 65.7% {n - 134) of the
records, followed by members of the Salticidae and Clubionidae at 19.4% and
10.4%, respectively (Table 1). The families Philodromidae, Oxyopidae, and
Thomisidae combined for the remaining 4.5%. Larvae were associated with adults
as well as with penultimate and earlier instars; and of the 87 spiders whose gender
could be determined, 53 were males and 34 females. First instars of M. pulchella
were found during all months, whereas all other immature stages were found only
during May and June (Table 2). Adults were collected from May to September,
with the majority collected during July.

DISCUSSION

The presence of M. pulchella larvae on spiders confirms that this species
boards spiders to gain access to spider eggs. Five other North American species
have been studied with respect to their strategies for gaining access to eggs. First
instars of Mantispa viridis Walker located and entered previously deposited egg
sacs and fed directly on the eggs within; larvae would not board spiders and
therefore were termed obligate penetrators of egg sacs (Redborg and MacLeod
1985; Brushwein 1986). In contrast, first instars of Climaciella brunnea (Say) did
not penetrate previously constructed egg sacs and would not feed on spider eggs
unless larvae had previously been aboard spiders; larvae readily boarded spiders
and were termed obligate spider boarders (Redborg and MacLeod 1983). Recent
studies on neonate M. pulchella larvae have similarly failed to induce larvae to
feed on spider eggs unless larvae had previously boarded spiders (Brushwein and
Hoffman unpubL). First instars of Mantispa uhleri Banks are facultative spider
boarders because although they showed a strong preference for spider boarding
over direct egg sac penetration, they could still penetrate egg sacs and develop
without previously having boarded spiders (Redborg and MacLeod 1985). First

10

THE JOURNAL OF ARACHNOLOGY

Table I. — Spicier taxa associated with immature stages of Mantispa pulchella in South Carolina.
Numbers represent the number of collections of each taxon, broken down by developmental stage and
gender. Superfamiiies and families are arranged according to the taxonomic list presented by Shear
(1986). (undet. = gender undetermined).

SUPERFAMILY

Family

Species

Developmental stage and gender

Adults

Juveniles

Egg

sacs

Male

Female

Male

Female

Undet.

CLUBIONOIDEA

Clubionidae

Clubiona maritima L. Koch

0

I

0

0

0

0

Ciubiona obesa Heotz

1

0

3

2

0

I

Clubiona sp., poss. obesa

0

0

1

0

■2

0

Clubionoides excepta (L. Koch)

1

0

0

0

2

0

PHILODROMOIDEA

Philodromidae

Philodromus imbecillus Keyserling

0

0

0

0

0

i

Philodromus vulgaris (Hentz)

1

1

0

0

0

0

DICTYNOIDEA

Anyphaenidae

Anyphaena fraterna (Banks)

2

1

0

0

0

0

Anyphaena pectorosa L. Koch

0

1

I

1

0

0

Anyphaena sp., pectorosa group

0

0

21

7

23

0

Aysha gracilis (Hentz)

1

1

2

5

2

0

Teudis mordax (0. P.-Cambridge)

0

1

0

0

0

9

Wulfila albus (Hentz)

0

2

3

2

3

0

LYCOSOIDEA

Oxyopidae

Oxyopes agiossus Chamberlin

0

0

0

0

I

0

THOMISOIDEA

Thomisidae

Misumenops asperatus (Hentz)

0

0

I

1

0

0

SALTICOIDEA

Salticidae

Eris marginata (Walckenaer)

2

0

0

0

0

0

Hentzia palmarum (Hentz)

0

1

4

2

1

O'

Lyssomanes viridis (Walckenaer)

0

0

0

0

1

0

Maevia inclemens (Walckenaer)

0

0

0

!

0

0

Metacyrba undata (De Geer)

1

0

0

0

0

0

Metaphidippus exiguus (Banks)

7

3

0

0

0

0

Metaphidippus galathea (Walckenaer)

1

0

0

0

0

0

Metaphidippus peckhamorum Kaston

0

0

0

0

0

1

Metaphidippus sp.

0

0

0

1

0

0

Totals

17

12

36

22

35

12

instars of Mantispa fusckornis Banks have been found on six species of spiders
(Rice 1986b). In light of the current status of this species as a sibling species of
M. uhleri (Hughes-Schrader 1979), M. fusckornis larvae are probably also
facultative boarders. Larvae of Mantispa interrupta Say have boarded and
remained on a spider in the laboratory (Viets 1941), and recent studies indicate
that larvae of this species are facultative spider boarders (Brushweie unpubL).
The behavior of M. pulchella larvae seems to be more similar to that of C
brunnea larvae than to that of other North American mantispiee larvae, and
therefore M. pulchella larvae are probably obligate boarders as well.

HOFFMAN & BRUSHWEIN— Af/lAra/M PULCHELLA SPIDER ASSOCIATIONS

11

Table 2. — Number of Mantispa pulchella collected by month in South Carolina.

Month

M. pulchella Developmental stage

First

Second

Third

Pupa

Adult

January

17

0

0

0

0

February

4

0

0

0

0

March

6

0

0

0

0

April

15

0

0

0

0

May

6

1

0

0

3

June

2

3

1

6

2

July

1

0

0

0

46

August

12

0

0

0

3

September

13

0

0

0

2

October

12

0

0

0

0

November

15

0

0

0

0

December

20

0

0

0

0

Totals

123

4

1

6

56

Both the position occupied while on spiders and the number of larvae on each
spider appear different for M. pulchella than for other species. Mantispa
pulchella larvae were found wrapped over the dorsal pedicel region and only one
larva was found per spider. In contrast, first instars of M. uhleri have been found
wrapped either dorsally, ventrally, or laterally around the pedicel, inside the book
lung slits, or, less commonly, on the membrane between the carapace and leg
bases, on the legs, or around the spinnerets; some spiders had two or three larvae
on them (Redborg and MacLeod 1985). Larvae of M. fuscicornis have been
found wrapped ventrally around the pedicel, inside the book lung slits, or, less
commonly, on the membrane between the carapace and leg bases or on a leg
(Rice 1986b); it was not uncommon to find 2 or 3 larvae per spider (M. Rice
pers. comm.). Climaciella brunnea larvae have only been found associated with
the carapace and the membrane between the carapace and the leg bases; the
presence of two larvae on a spider has been reported (Redborg and MacLeod
1983). The variability in the positions occupied by larvae of these other species
affords multiple resting sites on a single spider and the large size of some of the
hosts (e.g., the Lycosidae) allows for more than one larva in areas such as the
pedicel region or carapace. In contrast, the combination of a single resting site
and the relatively small size of the spiders associated with M. pulchella would
seem to preclude multiple infestations.

First instar M. pulchella may sustain themselves by feeding on the hemolymph
of the boarded spiders, thereby becoming true ectoparasites. Abdomens of larvae
commonly were distended to the point that the abdominal banding pattern
became diffuse. Similarly, first instars of M. uhleri on spiders appeared
“plumper” than neonate larvae and were found to gain weight after boarding
spiders (Redborg and MacLeod 1984). Also, the discolored patches on spiders
boarded by M. pulchella appeared to be similar to those on spiders boarded by
M. uhleri, which Redborg and MacLeod (1984) interpreted as evidence of larval
feeding damage. If this interpretation is correct, then the occurrence of patches on
both sides of a single pedicel indicates that M. pulchella larvae either occasionally
reverse their orientation while on spiders or that the spider had been infested
prior to the boarding by the current occupant.

12

THE JOURNAL OF ARACHNOLOGY

The behavioral and morphological parameters of spiders that permit successful
mantispine boardings are virtually unknown. However, the spiders associated
with M. pulchella do share a few ecological and behavioral characteristics. All
spiders were wanderers which do not build webs for prey capture and which are
commonly found on foliage (Kaston 1978). In addition, members of the
Anyphaenidae, Salticidae, and Clubionidae all construct tubular silken retreats
for resting and molting in harborages such as curled leaves and beneath bark.
Members of the Philodromidae, Oxyopidae, and Thomisidae do not. Therefore,
while M. pulchella larvae appear to board primarily wandering spiders on foliage,
spiders within retreats might be easier for larvae to locate or board than those
not in retreats.

Five other North American mantispine species have been associated with
spiders. Redborg and MacLeod (1985) found that M. uhleri larvae were
associated with a wide variety of wandering spiders and attributed the lack of
associations with web-building spiders to the inability of larvae to come into
contact with spiders suspended in webs. Similarly, all spiders associated with C.
brunnea larvae also were wanderers, namely the ground cursorial Lycosidae
(LaSalle 1986; Redborg pers. comm.). Mantispa fuscicornis larvae were associated
with both web-building and wandering spiders, but Rice (1986b) hypothesized
that the presence of larvae on web-builders was due to the tendency of those
particular species to retreat into the cracks and corners of a wooden shelter
during the day and therefore became more accessible to larvae searching on that
substrate. Larvae of M. interrupta have only been associated with three species of
wandering spiders (Redborg and MacLeod 1985; Rice 1986a). In contrast, M.
viridis larvae have been associated with a wide variety of both web-building and
wandering spiders (Brushwein 1986). The above findings on other species coupled
with the present data on M. pulchella suggest that mantispine species that use
spider boarding to gain access to spider eggs generally will be associated with
wandering spiders, whereas species whose larvae normally locate and penetrate
previously deposited egg sacs will be associated with a wider diversity of both
web-builders and wanderers.

The male to female ratio of 53 to 34 for spiders associated with M. pulchella
seems in contrast to what might be expected, given that larvae board spiders in
order to gain access to eggs. However, if the twelve infested egg sacs can be taken
as an indication of larvae having boarded females prior to their oviposition, then
the male to female ratio becomes essentially one to one (53 to 46). In any case,
the male to female ratio of M. pulchella correlated well with results obtained by
Redborg and MacLeod (1985) on M. uhleri, where the male to female ratio of
spiders was 48 to 36 and where larvae boarded spiders of both genders with equal
frequency in the laboratory.

The subsequent fate of M. uhleri larvae on male spiders was unclear because
larvae failed to transfer from males to females when mating occurred. However,
some larvae transfered from conspecific males to females when the male was
cannibalized. Redborg and MacLeod (1985) argued that transfer of larvae during
mating would require larvae to coordinate their transfer activities with the wide
variety of courtship behaviors exhibited by the various spider species. In contrast,
transfer of larvae during cannibalism required no such coordination and had the
advantage of larval transfer from any spider to any other, regardless of sex,
developmental stage, or species. The ability of larvae to transfer from one spider

HOFFMAN & BRUSHWEIN— PULCHELLA SPIDER ASSOCIATIONS

13

species to another in this manner could therefore result in occasional transfers of
larvae to spider species which for behavioral or morphological reasons would not
normally be boarded by larvae of that particular mantispine species. Therefore,
spider associations for a particular mantispine species could consist both of spider
species which are normally boarded and of species which are not normally
boarded but which acquired larvae via predation of infested spiders. Unfortu=
nately, the extent to which the spider associations of M. pulchella reflect the
latter category cannot be assessed at this time.

The presence of first instars of M. pulchella on spiders throughout the winter
indicates that this species overwinters on spiders. Furthermore, the absence of any
other developmental stages of M. pulchella during the winter months suggests
that M. pulchella overwinters in South Carolina only as first instars on spiders.
This species also may overwinter as eggs, but the lack of egg collections coupled
with reports on the larval or pupal overwintering strategies of other species
argues against this. Redborg and MacLeod (1985) concluded that M. uhleri
overwintered in southern Illinois exclusively as first instars on spiders, and
Brushwein (1986) concluded that M. viridis overwintered in northwestern South
Carolina as either larvae or pupae within spider egg sacs.

The collection of second instar M. pulchella during May and June, of third
instars and pupae during June, and of most adults during July indicates that this
species is essentially univoltine in South Carolina. However, the possibility for
more than one generation per year cannot be excluded. Additional generations
could result from adult M. pulchella produced by spiders which bred early in the
year subsequently producing larvae that board spider species destined to breed
later in the year, Redborg and MacLeod (1985) and Brushwein (1986) discussed
similar scenarios in relation to the seasonal cycles of M. uhleri and M. viridis,
respectively. The spiders associated with both M. uhleri and M. viridis included
species which bred during late summer and fall and, accordingly, an average of
two generations per year was reported for M. uhleri in Illinois and a minimum of
three generations were observed for M. viridis in South Carolina. However, no
immature M. pulchella were found either on adult spiders or inside egg sacs from
August until late December, indicating that the spider species used by M.
pulchella ail breed earlier in the year.

ACKNOWLEDGMENTS

We are grateful to the following for their assistance in identification of the
spiders: Allen R. Brady, Hope College, Holland, Michigan; G. B. Edwards,
Florida State Collection of Arthropods (FSCA), Division of Plant Industry,
Gainesville, Florida; James H. Redner, Canadian National Collection of Insects
(CNCI), Biosystematics Research Institute, Ottawa, Ontario; and Steve H. Roach,
Cotton Production Research Unit, Agricultural Research Service, U.S, Depart-
ment of Agriculture, Florence, South Carolina. Gratitude is also extended to
Peter H. Adler, Joe D. Culin, and John C. Morse of Clemson University for
reviewing an earlier draft of the manuscript, and to Marlin E. Rice (University of
Idaho) and Kurt E. Redborg for their comments. This is Technical Contribution
No. 2804 of the South Carolina Agricultural Experiment Station, Clemson
University.

14

THE JOURNAL OF ARACHNOLOGY

LITERATURE CITED

Austin, A. D. 1985. The function of spider egg sacs in relation to parasitoids and predators, with
special reference to the Australian fauna. J. Nat. Hist., 19:359-376.

Banks, N. 1912. Notes on Nearctic Mantispidae. Proc. Entomol. Soc. Washington, 14:178-179.

Brush wein, J. R. 1986. Bionomics of M antispa viridis Walker (Neuroptera: Mantispidae). Ph.D.

dissertation. Clemson University, Clemson, South Carolina, xi + 92 pp.

Dondale, C. D. and J. H. Redner. 1976. A review of the spider genus Philodromus in the Americas
(Araneida: Philodromidae). Canadian Entomol., 108:127-157.

Dondale, C. D. and J. H. Redner. 1978. The crab spiders of Canada and Alaska (Araneae:

Philodromidae and Thomisidae). Canadian Dept. Agric. PubL, 1663, 255 pp.

Edwards, R. J. 1958. The spider subfamily Clubioninae of the United States, Canada and Alaska
(Araneae: Clubionidae). Bull. Mus. Comp. Zook, 118:365-436.

Gaddy, L. L. and J. C. Morse. 1985. Common spiders of South Carolina with an annotated checklist.

South Carolina Agric. Expt. Sta. Tech. Bull., 1094, vi + 182 pp.

Hoffman, K. M. and S. W. Hamilton. 1988. First record of a mantispine larva (Neuroptera:
Mantispidae) associated with an adult caddisfly (Trichoptera: Leptoceridae). Entomol. News,
99:161-163.

Hughes-Schrader, S. 1969. Distance segregation and compound sex chromosomes in mantispids
(Neuroptera: Mantispidae). Chromosoma (Berlin), 27:109-129.

Hughes-Schrader, S. 1979. Diversity of chromosomal segregational mechanisms in mantispids
(Neuroptera: Mantispidae). Chromosoma (Berlin), 75:1-17.

Kaston, B. J. 1973. Four new species of Metaphidippus, with notes on related jumping spiders
(Araneae: Salticidae) from the eastern and central United States. Trans. American Micros. Soc.,
92:106-122.

Kaston, B. J. 1978. How to Know the Spiders. 3rd ed. Wm. C. Brown Co., Dubuque, Iowa.

Lambkin, K. J. 1986. A revision of the Australian Mantispidae (Insecta: Neuroptera) with a
contribution to the classification of the family 1. General and Drepanicinae. Australian J. Zook,
Suppk Ser., 1 16:1-142.

LaSalle, M. W. 1986. Note on the mantispid Climaciella brunnea (Neuroptera: Mantispidae) in a
coastal marsh habitat. Entomol. News, 97:7-10.

MacLeod, E. G. and K. E, Redborg. 1982. Larval platymantispine mantispids (Neuroptera:

Planipennia): Possibly a subfamily of generalist predators. Neuroptera Int., 2:37-41.

McKeown, K. C. and V. H. Mincham. 1948. The biology of an australian mantispid {Mantispa vittata
Guerin). Australian Zook, 11:207-224.

Parker, F. D. and L. A. Stange. 1965. Systematic and biological notes on the tribe Platymantispini
(Neuroptera: Mantispidae) and the description of a new species of Plega from Mexico. Canadian
Entomol., 97:604-612.

Platnick, N. 1. 1974. The spider family Anyphaenidae in America north of Mexico. Bulk Mus. Comp.
Zook, 146:205-266.

Redborg, K. E. 1976. Species of Mantispidae (Neuroptera) at Dixon Springs Agricultural Center.
Illinois Agric. Expt. Sta. DSAC, 4:17-18.

Redborg, K. E. and E. G. MacLeod. 1983. Climaciella brunnea (Neuroptera: Mantispidae): a
mantispid that obligately boards spiders. J. Nat. Hist., 17:63-73.

Redborg, K. E. and E. G. MacLeod. 1984. Maintenance feeding of first instar mantispid larvae
(Neuroptera, Mantispidae) on spider (Arachnida, Araneae) hemolymph. J. Arachnok, 11:337-341.
Redborg, K. E. and E. G. MacLeod. 1985. The developmental ecology of Mantispa uhleri Banks
(Neuroptera: Mantispidae). Illinois Biol. Monogr., 53, 130 pp.

Rice, M. E. 1986a. Spiderling survival in a Mantispa (Neuroptera, Mantispidae) infested egg sac. J.
Arachnok, 13:139-140.

Rice, M. E. 1986b. Communal oviposition by Mantispa fuscicornis (Say) (Neuroptera: Mantispidae)
and subsequent larval parasitism on spiders (Arachnida: Araneida) in south Texas. J. Kansas
Entomol. Soc., 59:121-126.

Roth, V. D. 1985. Spider Genera of North America. American Arachnok Soc., 176 pp.

Shear, W. A. 1986. Taxonomic Glossary. Pp. 403-432, In Spiders — Webs, Behavior, and Evolution (W.
A. Shear, ed.) Stanford University Press, Stanford.

Viets, D. 1941. A biological note on the Mantispidae (Neuroptera). J. Kansas Entomol, Soc., 14:70-
71.

Manuscript received January 1988, revised June 1988.

Yoshida, M. 1989. Predatory behavior of three Japanese species of Metleucauge (Araneae,
Tetragnathidae). J. Arachnol, 17:15”25.

PREDATORY BEHAVIOR OF THREE JAPANESE SPECIES
OF METLEUCAUGE (ARANEAE, TETRAGNATHIDAE)

Makoto Yostiida

Department of Biology and Geology
Faculty of Science and Engineering
Ritsumeikan University, Kyoto 603, Japan

ABSTRACT

The predatory behavior of Metleucauge kompirensis (Boes. et Str.), M. yunohamensis (Boes, et
Str.) and Metleucauge sp, was studied, and the natural prey of the former two species were collected
from their webs. These species did not immobilize their prey by wrapping, but immobilized them only
by biting. The lack of immobilization wrapping in Metleucauge can be explained by two different
hypotheses; I) Metleucauge remains in the primitive stage of the evolution of predatory behavior, or
2) the habit of immobilization wrapping has been lost, because Metleucauge captures weakly flying
insects (mainly midges and mayflies) which can be immobilized without wrapping.

INTRODUCTION

It has been suggested that there are five stages in the evolution of predatory
behavior in aratieid spiders, judging mainly from whether immobilization
wrapping is used or not (Robinson et al. 1969; Robinson 1975).

Robinson and his colleagues agreed with Eberhard (1967) that lack of
immobilization wrapping is primitive, and immobilization wrapping is advanced.
They supposed that immobilization wrapping was derived from postdmmobiiiza”
tion wrapping (that is, the wrapping after immobilization by biting) at the capture
site.

Alternatively, Levi (1985) suggested that the lack of immobilization wrapping
in Micrathena and Gasteracantha is not a primitive character, but rather a
secondary loss of immobilization wrapping.

Eberhard (1982) stated that immobilization wrapping evolved independently
along three lines, Theridiosomatidae-Anapidae, Araneidae, and Tetragnathidae-
Metinae. As examples of the spiders with immobilization wrapping in the
Tetragnathidae-Metinae line, he listed Dolichognantha spp., Leucauge spp., and
Chrysometa spp. Tetragnaiha praedonia also sometimes uses immobilization
wrapping with large prey (Yoshida 1987). The predatory behavior in the
Tetragnathidae-Metinae line has been studied only fragmeetarily, and no species
without attack wrapping has yet been found.

Levi (1980) included Tetragnathidae in the Araneidae, and divided Araneidae
into the three subfamilies (Araneinae, Metinae and Tetragnathinae). He included
several genera, such as Meta, MetelUna, Metleucauge, and Leucauge, in Metinae.
He stated that Meta and the related genera have many primitive morphological
characters, and that the genera resemble some genera of Theridiidae, a family of

16

THE JOURNAL OF ARACHNOLOGY

Araneoidea. Recently, Levi (1986) divided Araneidae again into two families,
Araneidae and Tetragnathidae, with the latter including Metinae and Tetragnathi-
dae described in the previous paper.

I report here on the predatory behavior of the three Japanese species of
Metleucauge {M. kompirensis, M. yunohamensis, and Metleucauge sp.). These
species do not use immobilization wrapping at all. I also discuss the evolution of
the predatory behavior in the Tetragnathidae-Metinae line.

Metleucauge sp. has been thought to be Meta segmentata previously by
Japanese arachnologists. But Shinkai and Takano (1984) stated that it belongs to
Metleucauge, and that they regarded it as Metleucauge segmentata. Recently,
Yaginuma (1986) said that its species name is not segmentata, though it belongs
to Metleucauge. He said that it resembles M. yunohamensis, but is a new species
of Metleucauge.

MATERIALS AND METHODS

The investigation of M. kompirensis and M. yunohamensis was done in 1986
and 1987 at two mountain streams in Kyoto city (100-200 m above sea level). M.
yunohamensis matured in April to May, and M. kompirensis matured in June to
July. The species usually made their horizontal orb-webs above the mountain
streams. I observed the predatory behavior of adults of the two species, both with
natural prey and with the fairly large prey given by me. The latter prey were
collected using a sweep-net near the study area. Less than two prey were attached
to a web by a pincette. I observed predatory behavior of many individuals of M.
kompirensis for several hundred hours, and of M. yunohamensis for several tens
of hours. The body length of both the spiders and the prey was estimated by eye.

The third species, Metleucauge sp., was investigated halfway up Mt.
Chougatake, north of Nagano Prefecture (ca 1400 m above sea level). This
species, like kompirensis and yunohamensis, also usually made its web above
mountain streams. I could observe the predatory behavior for only ten to several
hours.

RESULTS

Predatory behaviors. — The following behaviors were observed in the context of
predatory sequences:

Jerking: This consists of a rapid pulling of the radii with the first pair of legs.

Web-Shaking: Only M. yunohamensis showed this behavior. This species
sometimes shook the web slowly with legs I (and H?) as soon as the prey hit the
web. This is different from jerking, because the web was pulled perpendicularly to
its plane. The function of the low frequency vibration is unknown.

Approach to prey: Running or walking from the hub to the capture site along
the radial thread(s), pulling a drag-line from the hub. Running occurred generally
when the prey was vibrating rapidly, while walking occurred generally when the
prey was motionless (not vibrating).

Touch: The spiders touch the prey with legs I (and palps?) before most attacks
are initiated. Metleucauge species often touched the prey with legs I in
unsuccessful prey capture sequences, whereas touching was not observed at all in

YOSHIDA— PREDATORY BEHAVIOR OF METLEUCAUGE

17

successful prey capture sequences. Touching might have occurred also in the latter
case, but may have occurred too rapidly to detect without high-speed film.

Seizing in jaws: Seizing and holding the prey in the chelicerae.

Biting: Biting the prey with the chelicerae. Lubin (1980) included both seizing
and biting in biting. Certainly the jaws are used both in seizing and biting, but
there are some differences between them. Robinson and Robinson (1973) pointed
out that seizing is used mainly with very small prey, whereas biting is used mainly
with larger prey, and that the time required for seizing was very short, whereas
the time required for biting was several times longer than that for seizing. So, I
intend to distinguish seizing from biting.

Wrapping: Wrapping was used only after prey immobilization in Metleucauge
species. Wrapping occurred at three sites: at the capture site, during
transportation to the hub, and at the hub itself. At the capture site, wrapping
generally began after immobilization biting while the prey was still in the spider’s
jaws. I called this type of wrapping, also observed in Tetragnatha praedonia, as
'‘wrapping with bite” (Yoshida 1987). After “wrapping with bite”, Metleucauge
released the prey from its jaws, and then cast additional skeins of silk onto the
prey. Wrapping was often interrupted by cutting the prey from the web. Free-
wrapping (wrapping the prey beneath the web) was rarely observed. Metleucauge
was different from Argiope argentata, because it never rotated the prey around a
radius while wrapping (“rotisserie” wrapping of Eberhard [1982]).

Pulling out: Pulling the prey from the web, using the jaws.

Cutting out: Cutting the web in order to remove the prey from the web.

Carrying in jaws: Carrying the prey in the jaws to a feeding site.

Carrying on silk: Carrying the prey to the feeding site, suspended on a silk line
from the spinnerets. The silk line made by M. yunohamensis was short, so the
prey sometimes became entangled in the web. In these circumstances, the web was
destroyed, as the spider pulled the prey out by force. Other species did not show
this behavior.

Leaving the prey: The return of the spider to the hub alone, leaving the prey at

the capture site.

Predatory sequences employed. — Three predatory sequences were observed in
M. kompirensis and M. yunohamensis as follows:

Seize-Pull out: Spiders generally located prey by jerking the web, then ran to
the prey, seized it in the jaws and pulled it from the web. Spiders then carried the
prey in the jaws to the hub. In most cases, spiders returned by dropping from the
web, hanging with the dragline attached to the hub, and then climbed quickly to
the hub along the dragline (“drop and climb up” behavior). It took only 2-5
seconds for the spiders to complete seize-pull out sequences.

Bite-Pull out: This sequence resembles seize-pull out, the above sequence, but it
includes the behavior unit “biting” The time to seize the prey was very short
(perhaps less than one second), but biting took several seconds to several minutes.
Probably spiders would infuse the venomous fluid into the prey during biting.
Spiders usually bit some parts of a large prey first for a fairly long time (more
than several seconds), and then they changed the biting sites successively. As a
result, the prey was crushed gradually into a ball. Then the spider pulled it from
the viscid spirals with the jaws. The prey were carried either by drop and climb-
up mentioned above, or carrying the prey along a radial thread.

Bite-Wrap: The prey were immobilized by biting. After immobilization biting,
spiders first wrapped the prey with silk while it was still in the jaws. Then they

18

THE JOURNAL OF ARACHNOLOGY

Table 1. — Attack sequences with different types of prey in three species of Metleucauge. Frequency
of occurrence of the different sequences is shown after each sequence.

Prey type

M. kompirensis

M. yunohamensis

Metleucauge sp.

Diptera

Seize-Pull Out

54

Seize-Pull Out

4

Seize-Pull Out

0

Bite-Pull Out

2

Bite-Pull Out

1

Bite-Pull Out

1

Bite-Wrap

0

Bite-Wrap

11

Bite-Wrap

3

Ephemeroptera

Seize-Pull Out

18

Seize-Pull Out

0

Seize-Pull Out

0

Bite-Pull Out

0

Bite-Pull Out

0

Bite-Pull Out

0

Bite-Wrap

0

Bite-Wrap

5

Bite-Wrap

2

Lepidoptera

Seize-Pull Out

0

Seize-Pull Out

0

Seize-Pull Out

0

Bite-Pull Out

1

Bite-Pull Out

0

Bite-Pull Out

0

Bite-Wrap

12

Bite-Wrap

4

Bite-Wrap

1

Hymenoptera

Seize-Pull Out

1

Bite-Pull Out

0

Bite-Wrap

0

Plecoptera

Seize-Pull Out

0

Seize-Pull Out

0

Bite-Pull Out

0

Bite-Pull Out

0

Bite-Wrap

1

Bite-Wrap

I

Odonata

Seize-Pull Out

0

Bite-Pull Out

4

Bite-Wrap

12

Orthoptera

Seize-Pull Out

0

Bite-Pull Out

1

Bite-Wrap

0

Neuroptera

Seize-Pull Out

0

Bite-Pull Out

1

Bite-Wrap

0

wrapped further after releasing it from the jaws. When the prey was large,
wrapping was often interrupted by cutting the silk entangling the prey. The silk
covered only some parts of the prey, perhaps because the amount of silk was
little. Finally, spiders cut out all the entangling silk lines, in order to remove the
prey from the web. In most cases, the prey were carried to the hub suspended on
a silk line, Metleucauge sp. used bite-pull out and bite-wrap, but did not use
seize-pull out.

Frequency of sequences with different types of prey and efficiency of predation
on various types of prey. — Table 1 shows the comparison of attack sequences
with different types of prey in three species of Metleucauge. The wrap-bite
sequence, frequently used by Argiope, was not used by the spiders at all. With
Diptera and Ephemeroptera, M. kompirensis used mainly seize-pull out and did
not use bite-wrap at all, while Metleucauge sp. used mainly bite-wrap with
Diptera and Ephemeroptera, and did not use seize-pull out at all, M.
yunohamensis used only bite-wrap with Ephemeroptera, and used mainly bite-
wrap with Diptera, Lepidoptera and Odonata were attacked mainly by bite-wrap.

Table 2 shows the efficiency of predation (% of prey insects captured to the
total insects attached to the webs) on various types of prey in each species.
Diptera, Ephemeroptera, and Lepidoptera were captured efficiently by all species,
though few data are available for Metleucauge sp. Damselflies were captured
efficiently also by M. kompirensis. Hemiptera (stink bugs and leafhoppers) was
not captured at all. And Orthoptera and Hymenoptera was captured only once.

YOSHIDA— PREDATORY BEHAVIOR OF METLEUCAUGE

19

Table 2. — Efficiency of predation on various types of prey. Numerals show the number of prey

captured or not captured by each spider species. Numerals in parentheses show the percentages.

Prey type

M. kompirensis

M. yunohamensis

Metleucauge sp.

Captured

Not

Captured

Captured

Not

Captured

Captured

Not

Captured

Diptera

56 (91.6)

2(8.4)

16(100

0(0)

4(66.7)

2(33.3)

Ephemeroptera

18(100)

0(0)

5(100)

0(0)

2(100)

0(0)

Lepidoptera

12(80.0)

3(20.0)

4(80.0)

1(20.0)

1(100)

0(0)

O'donata

16(72.7)

6(23.3)

—

—

—

—

Hymenoptera

1 (50.0)

1(50.0)

0(0)

5(100)

0(0)

2(100)

Hemiptera

0(0)

5(100)

0(0)

3(100)

—

—

Orthoptera

1(10.0)

9(90.0)

—

—

—

—

Coleoptera

—

—

0(0)

3(100)

—

—

Plecoptera

1(100)

0(0)

—

—

1(100)

0(0)

Neuroptera

—

—

—

—

1(50.0)

1(50.0)

Mecoptera

—

—

—

—

0(0)

1(100)

Dermaptera

—

—

—

—

0(0)

1(100)

Lepidoptera (larva)

—

—

—

—

0(0)

2(100)

Total

105(80.1)

26(19.9)

25(67.6)

12(32.4)

9(50.0)

9(50.0)

respectively. When failing in prey-capture, the spider often touched the prey with

leg(s) L

The spiders failed to capture Hemiptera, Orthoptera, Hymenoptera and
Coleoptera 28 times (93% of these insects given), and there were several types of
failures: 1) spiders did not respond to the prey (four times), 2) the spider only
jerked its web (once), 3) prey escaped before the spiders arrived at the capture
sites (three times), 4) the spider approached its prey, but then returned to the
hub, leaving the prey at the capture site (once), 5) spiders tried to bite the prey,
but returned to the hub without biting (it is unknown why they did not bite: five
times). Of these, one spider touched the prey before attempting to bite, and two
spiders dropped the prey from the webs after attempting to bite, 6) spiders
returned to the hub after prey-touching (nine times), 7) spiders dropped their prey
after prey-touching (four times), 8) the spider bit its prey, but dropped it after
prey-touching (once).

Prey-touching was observed with 54% (15/28) of the insects not captured, very
often with Hemiptera (7/8), but was not observed at all with Coleoptera (0/3).
This behavior occurred mainly before returning to the hub alone and prey-
dropping (13/15). This prey-touching behavior was not observed at all when the

spider succeeded in capturing the prey.

Figs. 1-3 show the relation between the relative body length (prey/ spider) and
the predatory sequences employed in each species. Seize-pull out was used with
very small prey (smaller than half a spider body length or so), while bite-wrap
was used mainly with larger prey. Bite-pull out, whose frequency was low, was
used with prey of intermediate size.

Figs. 1-3 also show the relative body length of prey insects not captured by
spiders. It ranged widely. In some cases, such as Diptera, Lepidoptera,
Hymenoptera, Orthoptera (the prey of M. kompirensis), Neuroptera (the prey of
Metleucauge sp.), the insects that escaped were larger than ones captured,
suggesting that larger prey cannot be captured easily. However, the prey insects
that escaped from webs of Metleucauge sp. were smaller than ones captured. In

Number of Prey

20

THE JOURNAL OF ARACHNOLOGY

Relative Body Length (Prey/Spider)

Figure 1. — The relation between predatory sequences of Metleucauge kompirensis used to
immobilize prey and the relative body length (prey/ spider) of various kinds of prey. Dip = Diptera,
Eph = Ephemeroptera, Lep = Lepidoptera, Hem = Hemiptera, Hym = Hymenoptera, Pie = Plecoptera,
Odo = Odonata, Ort = Orthoptera. Each area shows the cases in which prey insects were captured by
the following sequence. Solid = seize-pull out, Shaded: = bite-pull out, Dotted = bite-wrap sequence.
Open area shows the case in which prey insects were not captured by spiders.

YOSHIDA— PREDATORY BEHAVIOR OF METLEUCAUGE

21

Relative Body Length (Prey/Spider)

Figure 2. — The relation between predatory sequences of Metleucauge yunohamensis used to
immobilize prey and the relative body length (prey/ spider) of various kinds of prey. Dip = Diptera,

Eph = Ephemeroptera, Lep = Lepidoptera, Hem = Hemiptera, Hym = Hymenoptera, Col =
Coleoptera. Each area shows the cases in which prey insects were captured by the following sequence.
Solid = seize-pull out, Shaded = bite-pull out, Dotted = bite-wrap sequence. Open area shows the case
in which prey insects were not captured by spiders.

2
2
2
S. 2
o 2

<D n
JQ ^

i2

z

2

2

Dip

^ . Fn

-

Eph

-

FF

Lep

r— 1

Lep(L)

1

-

Hym

1_ 1 J

-

Pie

1

-

mm 1 1

Neu

-

FF

Mec

~

— 1 ,

Der

1

1 2

3

>*

Relative Body Length (Prey/Spider)

Figure 3. — The relation between predatory sequences of Metleucauge sp. used to immobilize prey
and the relative body length (prey/ spider) of various kinds of prey. Dip = Diptera, Eph =
Ephemeroptera, Lep = Lepidoptera, Lep(L) = Lepidoptera (larva), Hym = Hymenoptera, Pie =
Plecoptera, Neu = Neuroptera, Mec = Mecoptera, Der = Dermaptera. Each area shows the cases in
which prey insects were captured by the following sequence. Solid = seize-pull out. Shaded = bite-pull
out, Dotted = bite-wrap sequence. Open area shows the case in which prey insects were not captured
by spiders.

22

THE JOURNAL OF ARACHNOLOGY

Table 3. — Prey collected in the webs of Metleucauge kompirensis.

Body length in mm

Total

Prey

0-2.0

-4.0

-6.0

-8.0

-10.0

-12.0

>12.0

(%)

Diptera

Nematocera

4023

559

24

5

0

1

0

4612(91.7)

Brachycera

7

3

1

0

0

0

0

11(0.2)

Hemiptera

85

25

0

0

0

0

0

110(2.2)

Ephemeroptera

6

103

134

22

4

2

3

274(5.4)

Plecoptera

0

3

0

3

0

0

0

6(0.1)

Hymenoptera

7

1

0

0

0

0

0

8(0.2)

Psocoptera

I

0

0

0

0

0

0

1(0.0)

Trichoptera

0

0

1

0

0

0

0

1(0.0)

Collembola

5

0

0

0

0

0

0

5(0.1)

Acarina

1

0

0

0

0

0

0

1(0.0)

Araneae

1

0

0

0

0

0

0

1(0.0)

Total

4136

694

160

30

4

3

3

5030

(%)

(82.2)

(13.8)

(3.2)

(0.5)

(0.1)

(0.1)

(0.1)

the case of Odonata given to M. kompirensis and Lepidoptera to M.
yunohamensis, there was no clear trend in relation to size. Likewise, the low
capture efficiency of Hemiptera, Orthoptera, Hymenoptera, and Coleoptera, was
not always dependent on prey size. For example, Hemiptera were not captured at
all in spite of their small size.

The prey collected in the webs of M, kompirensis and M. yunohamensis. —

Table 3 shows the prey collected from April to July in the webs of M.
kompirensis. Of 5030 prey, dipteran insects totalled 4623 (91.9%). Almost all
Diptera belonged to Nematocera. Thus, nematocerous flies are the main prey of
the spider. Aside from Diptera, Ephemeroptera (mayflies) and Hemiptera (winged
aphids) were relatively abundant. Other prey occurred rarely, such as Plecoptera,
Hymenoptera and Collembola. These prey were generally small (96% of them
were smaller than 4 mm in body length), though Ephemeroptera were relatively
larger than the other prey. Potentially dangerous insects, such as wasps and
pentatomids, were not collected at all, though tiny braconids were collected in
small numbers.

Table 4 shows the prey collected in May in the webs of M. yunohamensis.
Though the number of prey collected was smaller, it shows a similar trend as M.
kompirensis (Table 1). That is, the main prey items were nematocerous flies;
Ephemeroptera was the next abundant prey; the prey were generally small except
for Ephemeroptera.

I could not collect the prey in the webs of Metleucauge sp. because the density
of the webs was very low and I had not enough time to collect the prey.

DISCUSSION

Levi (1980) regarded Meta and the related genera as primitive in Araneidae,
judging from several morphological and behavioral characters, such as eye
placement, eye structure, male palpi, female genitalia and mating behavior.

This study showed that Metleucauge used three predatory sequences (seize-pull
out, bite-pull out, and bite-wrap). Which sequence is used is generally dependent

YOSHIDA— PREDATORY BEHAVIOR OF METLEUCAUGE

23

Table 4. — Prey collected in the webs of Metleucauge yunohamensis.

Body Length in

mm

Total

Prey

0-2.5

-5.0

-7.5

-10.0

-12.5

-15.0

>15.0

(%)

Diptera

Nematocera

529

28

0

0

0

1

0

558(84.7)

Brachycera

2

0

0

0

0

0

0

2(0.3)

Hemiptera

2

0

0

0

0

0

0

2(0.3)

Ephemeroptera

1

21

56

10

4

1

3

96(14.6)

Lepidoptera

1

0

0

0

0

0

0

1(0.2)

Total

535

49

56

10

4

2

3

659

(%)

(81.2)

(7.4)

(8.5)

(1.5)

(0.6)

(0.3)

(0.5)

on prey size, that is, seize-puil out is used for small prey, on the contrary, bite-
wrap for larger prey. Body weight, not body length, of the prey will have to be
measured in order to acertain the accurate limits for which a different sequence is
used. The spiders may decide whether the prey should be captured or not by
touching it with legs 1. The lack of attack wrapping is another remarkable
characteristic of the predatory behavior of Metleucauge.

Although the predatory behavior in Metinae has not been studied in detail,
Eberhard (1982) found that Leucauge spp. and Chrysometa spp. used attack
(^immobilization) wrapping. Leucauge magnifica also often uses attack wrapping
(Yoshida, unpub. data). The lack of attack wrapping of three Japanese species of
Metleucauge therefore may suggest that Metinae represent the primitive
evolutionary stages of predatory behavior (Robinson et al. 1969). This behavioral
primitiveness is consistent with the primitiveness of other morphological and
behavioral characters (Levi 1980). The lack of attack wrapping can be also
explained in another manner, that is, the habit may have been lost secondarily
(Levi 1985).

Robinson (1975) listed two merits of attack wrapping: one is the economy of
the time absent from the hub, and another is the ability to attack large and/or
potentially dangerous prey without the dangers involved in the intimate contact
of a biting attack. Furthermore, he said that attack wrapping is never lost
because its merits were too large. By which manner can the lack of attack
wrapping in Metleucauge be explained?

The three Japanese species of Metleucauge usually make their webs above
mountain streams, as shown earlier, and the fourth species, M. eldorado, also
“makes an orb — between rocks near streams” (Levi 1980). Many insects (mayflies,
midges, stoneflies, caddisflies, damselflies, dragonflies) fly above streams. As
shown in the results, the main prey of M. kompirensis and M. yunohamensis are
mayflies and midges. Eberhard (personal communication) pointed out that the
prey left in the webs may not be the main prey if they are not attacked by the
spiders. However, M. kompirensis actually feeds mainly on midges and mayflies
based on my many observations. These insects are weak fliers, and can be
immobilized easily by biting without danger. Metleucauge species seem to be
accustomed to handling these insects, since the spiders ran to the prey rapidly and
without hesitating when mayflies or midges were caught in their webs. Other
insects flying above streams are rarely captured by the webs. For example, there
are many agriid damselflies in the habitat of M. kompirensis, but I saw them
captured only a few times during several years of investigation.

24

THE JOURNAL OF ARACHNOLOGY

When presented with an unusual prey, not normally trapped under natural
conditions (such as pentatomids or grasshoppers), the spiders approached slowly
and hesitantly to the prey, often touching the prey with legs L This suggests that
Metleucauge species are not accustomed to such types of prey.

Another merit of attack wrapping, economy of time away from the hub, seems
to be a lesser difficulty for these species. Because mayflies and midges are
frequently captured, and are firmly restrained by the webs, the spiders need not
handle the prey in a hurry in order to capture the next prey. Given these
conditions, the ability to attack wrap may have been lost. This discussion suggests
that if the ancestor of Metleucauge had the ability to attack wrap, and if
potentially dangerous prey were included in the prey, the ability may have not
been lost. In this respect, it is important to acertain whether the species of Meta
and MetelUna can attack wrap prey, and also, what kinds of prey they eat in
nature. During 1987 I conducted a preliminary investigation of the predatory
behavior of Meta reticuloides and found that some spiders attempted to
immobilize an ant by wrapping. It appears that species related to Metleucauge
may have the ability to attack wrap and that this may result from a difference in
prey items. The predatory behavior of Meta and MetelUna must be studied and
be compared with that of Metleucauge.

If attack wrapping was fully developed (as seen, for example, in Argiope) in the
ancestor of Metleucauge, this ability may not have been lost. But, if the ability
was incomplete in its development, as occurs in Leucauge magnifica (Yoshida,
unpub. data), and if incomplete ability does not have much advantage, it may
have been easily lost. Other spiders such as Mastophora and Dichrostichus may
also have lost attack wrapping because all the prey are non-dangerous moths and
must be captured rapidly (Eberhard, pers. comm.).

I can not decide now whether the lack of attack wrapping is a primitive
character or has been lost secondarily. Thorough study of the predatory behavior
of genera related to Metleucauge {Meta, MetelUna, Chrysometa and Homa-
lometa) may solve this question.

ACKNOWLEDGMENTS

Dr. Y. Lubin and Dr. W. G. Eberhard kindly corrected many unsuitable
expressions of my English, and gave me many usefull suggestions. Mr. A. Shinkai
of Kunitachi Music High School, read my manuscript and gave me many
valuable suggestions, mainly on the terminology of predatory behaviors in
spiders. Mr. Y. Chikuni kindly helped my investigation on Metleucauge sp. I wish
to express my hearty thanks to them.

LITERATURE CITED

Eberhard, W. G. 1967. Attack behavior of diguetid spiders and the origin of prey wrapping in spiders.

Psyche, 74:173-181.

Eberhard, W. G. 1982. Behavioral characters for the higher classification of orb-weaving spiders.

Evolution, 36:1067-1095.

Levi, H. W. 1980. The orb-weaver genus Mecynogea, the subfamily Metinae and the genera

Pachygnatha, Glenognatha and Azilia of the subfamily Tetragnathidae north of Mexico. Bull.

Mus. Comp. ZooL, 149:1-74.

YOSHIDA— PREDATORY BEHAVIOR OF METLEUCAUGE

25

Levi, H. W. 1985. The spiny orb-weaver genera Micraihena and Chetacis (Araneae: Araneidae). Bull.
Mus. Comp. ZooL, 150:429-618.

Levi, H. W. 1986. The neotropical orb-weaver genera Chrysometa and Homalometa. Bull. Mus,
Comp. ZooL, 151:91-215.

Lubin, Y, D. 1980. The predatory behavior of Cyriophora (Araneae: Araneidae). J. Arachnol., 8:159-
185.

Robinson, M. H. 1975. The evolution of predatory behavior in araneid spiders. Pp. 292-312, In
Function and Evolution in behaviour. Essays in honour of Professor Niko Tinbergen. (G.
Caerends, C. Baer and A. Manning, eds.). Clarendon Press, Oxford.

Robinson, M. H., H. Mirick and O. Turner. 1969. The predatory behavior of some araneid spiders
and the origin of immobilization wrapping. Psyche, 76:487-501.

Robinson, M. H. and B. Robinson. 1973. The ecology and behavior of the giant wood-spider Nephila
maculata (Fabricius) in New Guinea. Smithsonian Contrib. ZooL, 149:1-76.

Shinkai, E. and S. Takano. 1984. A Field Guide to the Spiders of Japan. Tokai University Press,
Tokyo. 205 pp (In Japanese).

Yaginuma, T. 1986. Taxonomic notes on Japanese spiders (II) Araneus, Neoscona, Metellma, Cispius,
Heptathela. Fac. Lett. Rev. Otemon Gakuin Univ., 20:187-200 (In Japanese).

Yoshida, M. 1987. Predatory behavior of Tetragnatha praedonia (Araneae: Tetragnathidae). Acta
Arachnol., 35:57-75.

Manuscript received March 1988, revised June 1988.

i-»f ’.s

-!(l‘ :!■

- " ^ r>v. *■

■ .,v' ‘ '"V <i ^• ■ '■^'■ t'i.*,h'^Vrhdr . ' '

. ..... ,, '. •> _. ;;: ; av ., ' iV ,' '

_ \>1
'}'"’ ,'jir^<li *

' * . * '■’ • ■ 't i’’*' -fi^'T’*'/'/ •' '■ ' . ^ ■**/*■■ ".Tkttr't."'-' -i"^' •' ■ i vi't j •■

/ - * ^ ^ Jv ^'>l' t

vV j*'

v•'i<F^^■ • -^■i^:^^ v' .1' ♦.

:f. ,

'■>» A . ■ ’

'.- ‘■•‘.'iM Wsr^- . ’ " i-J'»Vtfc''.i -si.'V

■ r s> ' y ■ ■■ • " . .

4. . • / M * ; */^„V ^ t /

.>V. ■■ ^V.v,^; .f ,r,

■-' ' ' \’'*:'’i kj .’i '. ■ i‘

^■,'. .f,. - A*'

r/1’ -

>^-^v

^,^-. '-- '■ .'.; , .!,0 <•'.;• •tVV

,. ■. Wow-. ri'-r-f

■ ■‘‘' *‘5.v- \- ‘^.>^^ '

. ' ' • '■ »’ ■- * . .

» I

i .‘^u<^.-' ■■ - ■'J', tlx/- *j,Cr'-''’ rV".^.w,|^‘r

-"■> ’■ >y y '^' -wk '.'^. .'iE‘'

' A ■■;'Lc,-..:’f«

■ A, W

i'fti ... >, . .-. r-sj'f i.’v.'t t^f

^ ., -''.'it-" ' ' ;;‘.w ,:'■'’■ •

• < T^^:- j.4’ ; ■"• ' *

— . . .'r - .X . ■ - -, •■

* 4 . - ' J: ■•■ '. >r^ ! .*■ • -.' vt.ir^'^ -■ ♦ ^

Young, O. R, T. C. Lockley and G. B. Edwards. 1989. Spiders of Washington County, Mississippi. J.

Arachnol., 17:27-41.

SPIDERS OF WASHINGTON COUNTY, MISSISSIPPI

Orrey P. Young

Southern Field Crop Insect Management Laboratory
USDA=ARS, P.O. Box 346
Stoneville, Mississippi 38776 USA

Timothy C. Lockley

Imported Fire Ant Station
USDA-APHIS-^PPQ
3505 25th Avenue
Gulfport, Mississippi 39501 USA

and

G. B* Edwards

Florida State Collection of Arthropods
Division of Plant Industry
Florida Dept. Agric. & Cons. Serv.

P.O. Box 1269

Gainesville, Florida 32602 USA

ABSTRACT

Over a seven-year period, approximately 35,000 spiders representing 26 families, 133 genera, and
234 species were captured in Washington County, Mississippi, by pitfall, sweepnet, vacuum, bag, and
hand. Specimens were collected in 10 different habitat types and in four vegetational strata. Old-field
habitats yielded the most species (152) and residential lawns the fewest (14). Considering all habitats
sampled, the ground layer produced 111 species, the herbaceous strata 133, the shrub layer 49, and the
tree strata 30 species. The sweepnet method of capture obtained 128 species, pitfall 95, hand 61,
vacuum 53, and bagging 19 species. The largest number of species were obtained in spring and early
summer (maximum of 125 in May), with the fewest in mid-winter (Jan. = 24). Twenty-one species
were considered abundant, 51 common, 67 uncommon, and 95 rare. Additions to the state list of
Dorris (1972) number 102 species, for a new state total of 364 species.

A comparison with the North American fauna and with other surveys indicates that Washington
County is underrepresented both in cursorial forms active on the soil surface and web-spinning forms
typical of undisturbed habitats. The high incidence of disturbed habitats associated with intensive
agricultural activities in Washington County seems to have produced a depauperate spider fauna, but
spider populations of certain species characteristic of disturbed habitats are of sufficiently high density
and broad distribution to have a potential affect on crop insect pests.

INTRODUCTION

Stoneville, Mississippi is the site of the Delta Branch Experiment Station,
Mississippi State University, and the Delta States Research Center, U.S.
Department of Agriculture. For more than 50 years, scientists have been

28

THE JOURNAL OF ARACHNOLOGY

observing, collecting, and experimenting with the arthropods associated with
crops on site and in the surrounding Washington County. In the last 25 years,
much of that effort has been directed toward the arthropods of cotton and
adjacent habitats (e.g., Pfrimmer 1964; Stadelbacher 1981). Although spiders have
been indicated as potentially important predators in cotton, they have usually not
been identified to the genus or species level (e.g., Pfrimmer 1964; Smith and
Stadelbacher 1978). Over the last 10 years, the improved status of spider
taxonomy and a broad awareness of spiders as biological control agents has
changed the research environment concerning field studies of spiders in
agricultural situations (Riechert and Lockley 1984).

Beginning in 1981, field collections have attempted to delineate the structure
and composition of spider populations in the Stoneville environs. Since 1984, we
have focused on the spiders in habitats adjacent to cotton, particularly those
species that could be determined to be predators of the tarnished plant bug,
Lygus Uneolaris (Palisot) (Heteroptera: Miridae), These have included Oxyopes
salticus Hentz (Oxyopidae), Phidippus audax (Hentz) (Salticidae), and Pisaurina
mira (Walckenaer) (Pisauridae) (Lockley and Young 1986a, b; Lockley et al. 1989;
Welbourn and Young 1988; Young 1989a, b, c, d; Young and Lockley 1985, 1986,
1988, 1989a, b). The purpose of this report is to present the results of seven years
of sampling for spiders in Washington County. These data are compared with
other studies, and the potential role of this assemblage of spiders as agents for
crop pest suppression is discussed.

METHODS AND MATERIALS

Washington County, Mississippi, is in the west-central portion of the state,
adjacent to the Mississippi River, and in the approximate center of the Yazoo-
Mississippi Delta. This delta began formation about 18,000 years ago at the end
of the last ice age and is ideally suited for intensive agriculture (Fisk 1944). Deep
alluvial deposits, a flat terrain, ample moisture, hot and humid summers, and
mild winters combine to facilitate the growth of plants, and their associated
arthropods. Washington County contains ca 200,000 ha, of which ca 122,000 ha
(61%) are under cultivation in such crops as cotton, rice, milo, and soybean.
Timbered areas comprise ca 44,000 ha (22%) and include several state and federal
parks and wildlife refuges in addition to areas located outside the levees. There
are 1,365 km of roads in the county which, assuming an average width of 12 m,
occupy 18,000 ha (9%). The remaining 16,000 ha (8%) are composed of residential
and business areas, lakes, waterways, standing water, and marshes (Gunn et al.
1980). This pattern of land use provides a high percentage of “disturbed” habitats.
All crop fields are routinely plowed, cultivated, sprayed with herbicides and
insecticides, and otherwise made inhospitable for arthropods. “Edge” habitats —
edge of road, edge of ditch, edge of crop field, edge of forest island, edge of wet
area, etc. — also are typically disturbed areas that are mowed, sprayed with
herbicides, cultivated, or otherwise intruded upon at irregular intervals. These
disturbed habitats, combined with residential and business lawns and gardens,
probably comprise over 75% of the county area.

Beginning in 1981, systematic sampling of a variety of habitats, both disturbed
and undisturbed, was conducted utilizing five collection methods (Table 1).

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY

29

Table 1. — Sampling effort for spiders in Washington County, Mississippi, 1981-1987.

Habitat or host plant

Sampling

period

Freq. of
collect.

Method

of coll.

No. of
samples

No. of
spiders

Soybeans

VII-IX-81

Weekly

Sweep

40

1816

Forest

VI-82-VI-83

Biweekly

Pit

25

227

Erigeron spp.

IV-IX-84

We

Swp

136

909

Forest

IV-VI-84

We

Pit

45

278

Old-field

V-XlI-84

We

Vac

103

332

Pasture

VI-VIII-84

We

Vac

9

1800

Erigeron spp.

IV-IX-85

We

Swp

104

782

Cotton margins

II-XII-85

Bi

Vac

140

1867

Old-field

III-85-V-86

Bi

Pit

117

1674

Margins, pasture, old-field

X-85-X-86

Bi

Swp, Vac

487

8994

Erigeron spp.

III-IX-86

We

Swp

502

3471

Roadside grass

IV-VI-86

We

Vac

336

7459

Margin flowers

V-86-IV-87

Monthly

Swp

34

317

Spanish Moss

V.86-V-87

Mo

Bag

15

613

Forest margins

IX-XI-86

We

Swp

54

1393

Misc. habitats

1981-87

—

Hand

—

3109

Totals

2147

35041

Ground-dwelling spiders were sampled with several types of pitfall traps, some
with covers and some with interception barriers. Vegetation above the soil surface
was sampled with a dense muslin-mesh sweepnet, diameter 39 cm. A motorized
suction device (D-vac®) with a 34 cm diameter opening and a nozzle speed of ca
100 km/h was used to sample all strata, as was the technique of capturing
specimens by hand. Terminal portions of tree branches containing Spanish moss
also were bagged and removed. Over the seven-year period, samples were
obtained during every week of the year and every hour of the diel. Samples were
brought into the laboratory and frozen at — 20°C until they could be examined
and then thawed, sorted, identified, counted, and recorded. Voucher specimens
and unidentified material were stored in alcohol for later processing.

Representatives of every spider species were examined by G. B. Edwards, A. R.
Brady, Hope College, Holland, Michigan, or D. B. Richman, New Mexico State
University, Las Cruces, New Mexico. Voucher specimens are deposited at the
Mississippi Entomological Museum, Miss. State University, Starkville, in the
personal collection of T, C. Lockley, and in the Florida State Collection of
Arthropods, Division of Plant Industry, Gainesville.

RESULTS AND DISCUSSION

Within-county comparisons. — At least 234 species of spiders in 133 genera and
26 families were identified from ca 35,000 specimens collected over a seven-year
period (Appendix 1). This assemblage occurred primarily (203 spp.) in ecotonal
areas such as the margins of roads, fields, forests, and water, and in early-
successional habitats such as old-fields and pastures (Table 2). Old-field habitats,
2-5 years post-cultivation and abandonment, contained the highest diversity of
spiders (152 spp.), but represented one of the rarest habitat-types in a county
under intense agricultural management. Road and crop-field margins contained
the second-highest diversity of spiders — 98 spp. — and represented considerably

30

THE JOURNAL OF ARACHNOLOGY

Table 2. — Number of spider species distributed among various parameters, Washington County,
Mississippi.

Strata

Method of capture

Abundance

Ground

111

Pitfall. . .

............ 95

Rare

.. 95

Herb

133

Sweep . . .

.128

Uncommon

...67

Shrub. ..............

49

Vacuum.

............. 53

Uncommon. ........

...51

Tree

30

Hand...

61

Abundant. .........

...21

Bag

19

Habitat

Forest. ..............

......... 33

Lawn, resid. . , . . .

.. 14

Transitional (T)

......... 49

Building

.. 17

Crop, Field

......... 43

Spanish moss. . . .

.. 17

Meadow, Grassland (M)

......... 23

Old-field (O)..........

.........152

Combine, M,R. .

..101

Water margin (W)

16

Combine, T,0,W.

..169

Road, Field margin (R)

98

Combine, M,R,T,0,W.

203

more acreage than old-fields. Cotton and soybean fields, though representing over
50% of the county surface area, contained only 43 species. Thus only 19% of the
spider species available in the county for predation on crop pests actually
occurred on crops. Fortunately for crop pest control, most local pests (e.g.,
Anthonomus spp., Heliothis spp., Lygus spp.) also occupy habitats adjacent to
crops at some time in their life cycle. Because no spider species was found
exclusively on crops, potentially as many as 234 species may prey on crop pests in
these adjacent habitats.

The diversity of spiders obtained by our collection methods was low in the
winter months, with a minimum of 24 species collected in January (Fig. 1).
Spider populations dramatically increased in April, and by May 125 species were
active. The number of species captured each month gradually declined through
the summer and fall, with 62 species still active in October. Twenty-one species
were considered “abundant” in the habitats in which they occurred (Table 2).
These included 1 1 species that were abundant in cotton, as well as in adjacent
habitats. Given the large amount of acreage devoted to cotton, these 11 species
were probably the most abundant spiders in Washington County and may have a
major impact on cotton insect pests. They were Neoscona arabesca (Walck.),
Tetragnatha laboriosa Hentz, Ceraticelus emertoni (O. P.-Camb.), Lycosa lenta
group, Lycosa rabida Walck., Pardosa milvina (Hentz), Oxyopes salticus Hentz,
Pisaurina mira (Walck.), Metaphidippus galathea (Walck.), Phidippus audax
(Hentz), and P. clams Keys. Several of these species are important predators in
Washington County on the tarnished plant bug, Lygus lineolaris (Young 1989a, b,
c, d). They may also have an affect on the cotton bollworms, Heliothis spp.
(Stadelbacher and Lockley 1983), and the sterile bollworm hybrids and braconid
parasites currently under consideration as control agents of the bollworms. The
most abundant spider species in forested areas were Agelenopsis naevia (Walck.)
and Gladicosa gulosa (Walck.); in Spanish moss, Methaphidippus tillandsiae
Kaston; in roadside and field margins, Oxyopes salticus and Ceraticelus emertoni;
in old-fields, Pardosa milvina, Schizocosa spp., Xysticus ferox (Hentz), and
Tetragnatha laboriosa; and on Erigeron spp. (Compositae), Metaphidippus
galathea and Misumenops asperatus (Hentz).

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY

31

No. species

Month

Figure 1. — Maximum number of spider species captured each month during 1981-1987 in
Washington Co., Mississippi.

The sweepnet method of spider capture obtained 55% (128 spp.) of the fauna in
Washington County (Table 2), and was the most frequently used capture
technique. The vacuum method obtained the least numbci of species (53), but did
produce five species not captured by other means. This collection technique
probably could have been eliminated, with the resultant savings in time and effort
more profitably directed toward other methods of collection.

The highly disturbed nature of most of the soil in Washington County suggests
that typical soil spiders should be under-represented. Ground-dwelling spiders,
obtained primarily by pitfall traps, represented 47% (111 spp.) of all species
(Table 2), whereas foliage-dwelling spiders represented 70% of the species total
(some species occupied both ground and foliage strata). For comparison, a three-
year study conducted in frequently-disturbed soybean fields in Delaware indicated
that 70% of the species occurred on foliage and only 32% occurred on the ground
(Culin and Rust 1980) Conversely, a study that included sampling of deep leaf-
litter in undisturhcii Kansas woodland indicated that 75% of the species were
ground-dwelling and 25% foliage-dwelling (Fitch 1963). Most of the forested area
in Washington County is covered with water at some time each year, due either
to flooding of the Mississippi' River or to slow run-off after heavy rains. Leaf-
litter depth is typically shallow or non-existent in these areas and, combined with
the inundation, probably supports a very depauperate spider fauna (e.g., Uetz et
al. 1979). Thus it is not surprising that, considering all habitats sampled, foliage-
dwelling spiders are relatively very well-represented in Washington County,

Comparison with the North American fauna. — In 1985, V. D. Roth published a
compilation of the families and genera of spiders known to occur in North
America. He also included an estimate of the number of species in each genus.
This information now permits a comparison of limited-area surveys with the
entire North American fauna. In Washington County, specimens were obtained
from 26 families and 133 genera (Table 3). This represents 54% of the 48 families
and 28% of the 469 genera (Roth 1985) occurring in North America. The 234
species from Washington County represent only 7% of the 3311 North American

32

THE JOURNAL OF ARACHNOLOGY

Table 3. — Proportion of genera and species of the North American spider fauna (Araneomorphae)
that occur in Washington County, Mississippi.

Family

Genera

Species

N.A.

Wash. Co.

%

N.A.

Wash. Co.

%

Agelenidae

25

2

8

252

3

1

Amaurobiidae

8

—

—

82

—

—

Anapidae

1

—

—

i

—

—

Anyphaenidae

5

4

80

37

6

16

Aphantochilidae

1

—

—

1

—

—

Araneidae

42

23

55

192

39

20

Caponiidae

2

—

—

3

—

—

Ciubionidae

20

8

32

193

13

7

Ctenidae

3

—

—

5

—

—

Desidae

1

—

—

1

—

—

Dictynidae

9

1

11

159

2

I

Diguetidae

1

—

—

6

—

—

Dinopidae

1

—

—

1

—

—

Dysderidae

3

1

33

7

1

14

Filistatidae

3

1

33

13

1

8

Gnaphosidae

24

12

50

248

25

10

Hahniidae

3

2

67

19

3

16

Hersiliidae

1

—

—

2

—

—

Homalonychidae

I

—

—

2

—

—

Hypochilidae

1

—

—

4

—

—

Leptonetidae

2

—

—

34

—

—

Linyphiidae

152

11

7

845

17

2

Loxoscelidae

I

1

100

13

2

15

Lycosidae

15

10

67

234

32

14

Mimetidae

2

1

50

13

I

8

Mysmenidae

3

—

—

6

—

—

Nesticidae

3

I

33

31

1

3

Ochyroceratidae

1

—

—

1

—

—

Oecobiidae

2

1

50

7

1

14

Oonopidae

8

1

13

24

1

4

Oxyopidae

3

2

67

20

2

10

Philodromidae

5

3

60

95

7

8

Pholcidae

10

3

30

31

3

10

Pisauridae

4

3

75

14

4

29

Plectreuridae

2

—

—

15

—

—

Salticidae

45

25

55

288

47

16

Scytodidae

I

1

100

9

1

11

Selenopidae

1

—

—

5

—

—

Sparassidae

3

—

—

8

—

—

Symphytognathidae

I

—

—

I

—

—

Telemidae

1

—

—

3

—

—

Tengellidae

I

—

—

5

—

—

Theridiidae

27

6

22

231

7

3

Theridiosomatidae

1

1

100

2

1

50

Thomisidae

10

7

70

128

11

9

Uioboridae

7

2

29

15

2

13

Zodariidae

2

—

—

4

—

—

Zoridae

1

—

—

!

—

—

Totals

469

133

28.4

3311

234

7.1

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY

33

Table 4. — Comparison of spider guilds. North America and Washington County, Mississippi. Each
family assigned to a guild based on data from Roth (1985), Kaston (1981), Gertsch (1979), and
Comstock (1940),

Web-spinning

%

Wandering

%

N.A. Fauna

No. Families

25

52

23

48

No. Genera

307

65

162

35

No. Species

1955

59

1356

41

Wash. Co.

No. Families

11

42

15

58

No. Genera

51

38

82

62

No. Species

78

33

156

67

species. There is little doubt that areas of similar size to Washington County that
had a more diversified range of habitats would have substantially more species in
a larger set of genera and families.

It is also possible to compare certain functional aspects of the North American
and Washington County faunas. By the use of such sources as Roth (1985),
Kaston (1981), Gertsch (1979), and Comstock (1940), each spider family can be
designated as composed primarily of either web-spinning or wandering species.
The North American fauna at the species level is thus estimated to be 59% web-
spinners and 41% wanderers (Table 4). The Washington County fauna, however,
is estimated to include 33% web-spinners and 67% wanderers. The considerable
differences between these estimates probably are due to the preponderance of
disturbed habitats in Washington County and to the negative effect of habitat
disturbance (destruction, loss) on web-spinning spider populations.

Comparisofi with other faunal surveys. — Spider faunal surveys were reviewed
to compare with our efforts in Washington County. Spider faunal lists can be
classified in the following categories: a) specific plant association, e.g., peppermint
(Mclver and Belnavis 1986), daisy (Judd 1965); b) specific habitat, e.g., tree-bark
(Bower and Snetsinger 1985), salt-marsh (LaSalle and Cruz 1985); c) general
habitat type, e.g., old-field communities (Berry 1970), broomsedge communities
(Barnes and Barnes 1955); d) multi-habitat natural area, e.g., Itasca St, Park
(Heimer et al. 1984), Univ. Kansas Natiir. Hist. Res. (Fitch 1963); e) restricted
geographic area such as a town (Brown 1974) or island (Drew 1967); f) county
(Dorris 1968); g) multi-county (Branson and Batch, 1970); h) state or providence,
e.g., Wisconsin (Levi and Field 1954), British Columbia (West et al. 1984); i)
multi-state, e.g., Georgia area (Chamberlin and Ivie 1944).

An examination of this literature showed few previous surveys in common with
our county-wide study. The one survey that covered a single county was merely a
checklist of the species, wdth no additional data (Dorris 1968). The 108 species in
the Dorris study were collected in one year by sweepnet, sifting of litter, and
hand-picking, ail in unspecified habitats. Two multi-county studies, from
northwest Iowa (Abraham 1987) and northern Kentucky (Branson and Batch
1970), were of limited comparative value. The Iowa study listed only the genera,
but claimed 154 species. The Kentucky study listed 85 species, but was based on
only 503 specimens obtained by limited collecting. An attempt to extract county
data from state lists was not productive. Most state lists contained county
records, but very little information on habitats, seasonality, abundance, or

34

THE JOURNAL OF ARACHNOLOGY

sampling methods [e.g., Maryland (Muma 1945), Nebraska (Worley and Pickwell
1927), Oklahoma (Banks et al. 1932), Texas (Vogel 1970), Washington (Worley
1932)]. Kaston’s magnus opus on the spiders of Connecticut (1981) is certainly an
exception to that statement, but unfortunately his data are not in a format that
allows ready comparison with other faunal surveys.

Perhaps the only studies remotely comparable to the Washington County data
involve multi-habitat natural areas and restricted geographic areas. The Itasca
State Park (Minnesota) study of Heimer et al. (1984) listed 124 species, but did
not indicate the size of the area sampled, the amount of sampling effort through
time, the number of specimens examined, or detailed habitat information. A
study from the University of Oklahoma Biological Station (Branson 1966) listed
83 species identified from ca 1000 specimens collected during four summers, but
did not indicate the area sampled, contained little ecological data or analysis, and
was essentially a key to the genera of Oklahoma spiders. The most thorough
study of a natural area was that of Fitch (1963) at the 300 ha University of
Kansas Natural History Reservation. This study was conducted over a 13 year
period utilizing most sampling techniques, during all months of the year, and in a
variety of microhabitats within the tail-grass prairie and deciduous woodland
habitats. Of the 192 species listed, 119 (62%) were considered to be characteristic
of a deciduous forest habitat and 56 (29%) were associated with grasslands.
Within the woodlands, 85 species (71%) were obtained from leaf-litter and 29
(24%) from arboreal situation. This compares with 47% of the Washington
County species collected from the ground strata and 70% from above-ground sites
(all habitats combined, some species occurring in several strata).

Several studies that involve restricted geographic areas may be appropriate for
comparison. Brown (1974) reported 147 species of spiders collected over a six-
month period from Nacogdoches, Texas, and vicinity. This check-list contained
no information on the area sampled and no analysis, but did indicate that the
families Salticidae (30 spp.) and Araneidae (29 spp.) represented most of the
species. In Washington County, these two families also were highly represented
(Salticidae — 47 spp., Araneidae— 39 spp.). Perhaps the most defined geographic
area that has been examined for spiders is Beaver Island in Lake Michigan (Drew
1967). This 15,000 ha island is 24 km from the nearest mainland and has a well-
documented flora and fauna. Spiders were collected day and night in a variety of
habitats over a four-year period by sweepnet, sifting, beating, and hand. Of the
211 species obtained, 54% were web-spinners and 46% were wanderers. In
Washington County, 33% of 234 species were web-spinners and 67% were
wanderers. Web-spinners are characteristic of undisturbed sites containing
adequate web supports (Duffey 1978), and their comparatively low incidence in
Washington County could be due to the overwhelming predominance of disturbed
habitats.

Comparison of various characteristics of the Washington County spider fauna
with data from other faunal surveys has clearly indicated that the Washington
County fauna is disproportionately well-represented by species typical of
disturbed habitats. Some species, such as Oxyopes salticus, Tetragnatha laboriosa,
and Pardosa milvina, may develop high population densities in habitats adjacent
to crop fields. Crop insect pests occupying these habitats are probably exposed to
considerable predation by spiders. Management of these habitats to conserve and
enhance spiders and other predators could have a significant effect on crop pest
populations.

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY

35

ACKNOWLEDGMENTS

We appreciate the manuscript review provided by C. D. Dondale, M, J.
Gaylor, D, T Jennings, D. B. Richman and W. L. Sterling.

LITERATURE CITED

Abraham, B, J. 1987. Keys to the spiders of Northwestern Iowa. Proc. Iowa Acad. Sci., 94:37-50.

Banks, N., N. M. Newport and R. D. Bird. 1932. Oklahoma spiders. Publ. Univ. Oklahoma Biol
Surv., 4:7-49.

Barnes, R. D. and B. M. Barnes. 1955. The spider population of the abstract broomsedge community
of the southeastern Piedmont. Ecology, 36:658-666.

Berry, J. W. 1970. Spiders of the North Carolina Piedmont old-field communities. J. Elisha Mitchell
Sci. Soc., 86:97405.

Bower, L. F, and R. Snetsinger. 1985. Tree bark as an overwintering niche for spiders (Araneae) in an
oak-maple forest edge. Melsheimer Entomol. Ser., 35:1-5.

Branson, B. A. 1966. Spiders of the University of Oklahoma Biological Station, Marshall County,
Oklahoma, with observations on species used by mud-daubers as larval food, and a review of the
species known from Oklahoma, Southw. Natur., 11:338-371.

Branson, B. A. and D. L. Batch. 1970. Spiders (Arachnida: Araneida) from northern Kentucky, with
notes on phalangids and some other localities. Trans. Kentucky Acad. Sci., 31:84-98.

Brown, K. M. 1974. A preliminary checklist of spiders of Nacogdoches, Texas, J. ArachnoL, 1:229-
240.

Chamberlin, R. V. and W. Ivie. 1944. Spiders of the Georgia region of North America. Bull. Univ.
Utah, 35:1-267.

Comstock, J. H, 1940. The Spider Book (revised and edited by W. J. Gertsch). Cornell Univ. Press,
Ithaca, New York.

Culin, J. D. and R. W. Rust. 1980. Comparison of the ground surface and foliage dwelling spider
communities in a soybean habitat. Environ. Entomol, 9:577-582.

Dorris, P. R. 1968. A preliminary study of the spiders of Clark County Arkansas compared with a
five year study of Mississippi spiders. Arkansas Acad, Sci. Proc,, 22:33-37.

Dorris, P. R. 1972, A list of spiders collected in Mississippi. Notes Arachnol. Southwest, 3:1-4,

Drew, L. C, 1967. Spiders of Beaver Island, Michigan. Pubi, Mus. Michigan St. Univ. Biol Ser.,
3:153-208.

Duffey, E. 1978. Ecological strategies in spiders including some characteristics of species in pioneer
and mature habitats. Symp. ZooL Soc. London, 42:109-123.

Fisk, H. N. 1944. Geological Investigations of the Alluvial Valley of the Lower Mississippi River.
Mississippi River Commission, Vicksburg.

Fitch, H. S. 1963. Spiders of the University of Kansas Natural History Reservation and Rockefeller
Experimental Tract. Univ. Kansas Mus. Nat. Hist. Misc. PubL, 33:1-202.

Gertsch, W. J. 1979. American Spiders (2nd ed.). Van Nostrand Reinhold, New York.

Gunn, C. R., T. M. Pullen, E. A. Stadelbacher, J. M. Chandler and J. Barnes. 1980. Vascular Flora
of Washington County, Mississippi, and Environs. USDA, SEA, AR(SR), 150 pp.

Heimer, S., W. Nentwig and B. Cutler. 1984. The spider fauna of the Itasca State Park (Minnesota,
USA). Faun. Abh. Mus. Tierk. Dresden, 11:119-124.

Judd, W. W. 1965. Spiders and their insect prey from heads of ox-eye daisy, Chrysanthemum
kucanthemum L., in southwestern Ontario. Proc. Entomol. Soc. Ontario, 95:137-139.

Kastoe, B. J. 1981. Spiders of Connecticut (rev. ed.). Connecticut St. GeoL Nat. Hist. Surv. Bull,
70:1-1020.

LaSalle, M. W. and A. A, de la Cruz. 1985. Seasonal abundance and diversity of spiders in two
intertidal marsh plant communities. Estuaries, 8:381-393.

Levi, H. W. and H. M. Field. 1954. The spiders of Wisconsin. American Midi, Natur., 51:440-467,

Lockley, T. C. and O. P. Young. 1986a. Phidippus audax (Araneae, Salticidae) predation upon a
cicada {Tibicen sp.) (Homoptera, Cicadidae), J. Arachnol., 14:393-394.

Lockley, T. C. and O. P. Young. 1986b. Prey of the striped lynx spider, Oxyopes salticus (Araneae,
Oxyopidae), on cotton in the Delta area of Mississippi. J. Arachnol, 14:395-397.

36

THE JOURNAL OF ARACHNOLOGY

Lockley, T. C., O. P. Young and J. L. Hayes. 1989. Nocturnal predation by a crab spider, Misumena
vatia (Araneae, Thomisidae). J. Arachnol. (submitted)

Mclver, J. D. and D. L. Belnavis. 1986. A list of the spiders of peppermint in western and central
Oregon. Proc. Entomol. Soc. Washington, 88:595-598.

Muma, M. H. 1945. An annotated list of the spiders of Maryland. Univ. Maryland Agric. Exp. Stn.
Tech. Bull., A38:L65.

Pfrimmer, T. R. 1964. Populations of certain insects and spiders on cotton plants following insecticide
applications. J. Econ. Entomol., 57:640-644.

Riechert, S. E. and T. C. Lockley. 1984. Spiders as biological control agents. Annu. Rev. Entomol.,
29:299-320.

Roth, V. D. 1985. Spider genera of North America. American Arachnol. Soc., 176 pp.

Smith, J. W. and E. A. Stadelbacher. 1978. Predatory arthropods: Seasonal rise and decline of
populations in cotton fields in the Mississippi Delta. Environ. Entomol., 7:367-371.

Stadelbacher, E. A. 1981. Role of early-season wild and naturalized host plants in the buildup of the
El generation of Heliothis zea and H. virescens in the Delta of Mississippi. Environ. Entomol.,
10:766-770.

Stadelbacher, E. A. and T. C. Lockley. 1983. The spiders of Geranium dissectum Linneaus in
Washington County, Mississippi. J. Georgia Entomol. Soc., 18:398-402.

Uetz, G. W., K. L. Van der Laan, G. F. Summers, P. A. K. Gibson and L. L. Getz. 1979. The effects
of flooding on floodplain arthropod distribution, abundance and community structure. American
Midi. Natur., 101:286-299.

Vogel, B. R. 1970. Bibliography of Texas spiders. Armadillo Papers, 2:1-36.

Welbourn, W. C. and O. P. Young. 1988. Mites parasitic on spiders, with a description of a new
species of Eutrombidium (Acari, Eutrombidiidae). J. Arachnol, 16:373-385.

West, R., C. D. Dondale and R. A. Ring. 1984. A revised checklist of the spiders (Araneae) of British
Columbia. J. Entomol. Soc. British Columbia, 81:80-98.

Worley, L. G. 1932. The spiders of Washington. Univ. Washington Publ. Biol, 1:1-63.

Worley, L. G. and G. B. Pickwell. 1927. The spiders of Nebraska. Univ. Nebraska Studies, 27:1-129.
Young, O. P. 1989a. Predators of the tarnished plant bug, Lygus lineolaris (Heteroptera: Miridae):

Laboratory Evaluations. J. Entomol. Sci., 24:174-179.

Young, O. P. 1989b. Relationships between Aster pilosus (Compositae), Misumenops spp. (Araneae:

Thomisidae), and Lygus lineolaris (Heteroptera: Miridae). J. Entomol Sci., 24:250-255.

Young, O. P. 1989c. Field observations of predation by Phidippus audax (Araneae: Salticidae) on
arthropods associated with cotton. J. Entomol. Sci., 24:264-271.

Young, O. P. 1989d. Predation by Pisaurina Mira (Araneae, Pisauridae) on Lygus lineolaris
(Heteroptera, Miridae) and other arthropods. J. Arachnol 17:43-48.

Young, O. P. and T. C. Lockley. 1985. The striped lynx spider, Oxyopes salticus (Araneae:

Oxyopidae), in agroecosystems. Entomophaga, 30:329-346.

Young, O. P. and T. C. Lockley. 1986. Predation of striped lynx spider, Oxyopes salticus (Araneae:
Oxyopidae), on tarnished plant bug, Lygus lineolaris (Heteroptera: Miridae): A laboratory

evaluation. Ann. Entomol. Soc. America, 79:879-883.

Young, O. P. and T. C. Lockley. 1988. Dragonfly predation upon Phidippus audax (Araneae,

Salticidae). J. Arachnol, 16:121-122.

Young, O. P. and T. C. Lockley. 1989a. Spiders of Spanish Moss in the Delta of Mississippi. J.

Arachnol, 17: (in press)

Young, O. P. and T. C. Lockley. 1989b. Interactions between the predators Phidippus audax

(Araneae: Salticidae) and Hippodamia convergens (Coleoptera: Coccinellidae) in cotton and in the
laboratory. Entomol. News, 100:43-47.

Manuscript received June 1988, revised August 1988.

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY

37

APPENDIX

SPIDERS OF WASHINGTON COUNTY, MISSISSIPPI, 1981-87.

Explanation of symbols. — Month Collected: Each month is designated by its first letter and is listed
in chronological sequence. A dash between letters indicates that all intervening months not listed
contained specimens of that species. A blank space between letters indicates that no specimens were
found in the unlisted months. When certain months are listed alone, a second letter is added to avoid
confusion: Ja = January, Je = June, Jy = July, Mr = March, My = May, Ap = April, Au = August.

Habitat: F = forest, T = transitional area between forest and field or road, C = crop field, M =
meadow or grassland, O = old-field in early successional state 2-5 years after plowing, W = water
margin such as edge of pond, stream, or drainage ditch, R = road or field margin, L = lawn in
residential area, B = building, S = Spanish moss hanging from trees 2-5 m above ground.

Strata: G = ground, H = herbs and grass 0. 5-2.0 m above ground, S = shrubs 1-4 m above ground,
T = tree 3-5 m above ground.

Abundance (Ab): A = abundant, C = common, U = uncommon, R = rare. Each species was assigned
an abundance designation after a review of all collection records from the seven-year period and,
although quite subjective, is our best estimate in the absence of quantitative data.

Capture method: P = pitfall trap, S = sweepnet, V = vacuum device, H = hand.

State Record (St Rc): Asterisk indicates addition to state list of Dorris (1972).

Taxon

Month coll.

Habitat

Strata

Ab

Capture

St Rc

AGELENIDAE

Agelenopsis naevia (Walck.)

JJASO

TMOR

GHS

A

SVH

Agelenopsis utahana (Chamb. & Ivie)

Jy

0

H

R

P

*

Coras medicinalis (Hentz)

Ja D

OR

G

C

PV

ANYPHAENIDAE

Anyphaena celer (Hentz)

JFMA SOND

FO

GH

U

PV

Anyphaena maculata (Banks)

FMAM SOND

S

T

A

HB

*

Anyphaena sp. A

My

OR

H

R

PS

Aysha gracilis (Hentz)

My

R

H

R

S

Teudis mordax (O.P.-Cambridge)

MJ

R

H

R

SB

*

Wulfila sp.

AM S

O

GH

R

SV

*

ARANEIDAE

Acacesia hamata (Hentz)

Au

O

S

R

S

*

Acanthepeira stellata (Walck.)

M JJASON

FTORC

HST

C

SVH

Alpaida calix (Walck.)

MJ

O

H

R

V

Araneus cingulatus (Walck.)

Jy

o

H

R

V

*

Araneus marmoreus Clerck

s

T

T

U

H

Araneus miniatus (Walck.)

Je S

T

S

u

H

Araneus pratensis (Emerton)

Au

T

T

u

H

Araneus sp. A

Jy

TC

H

R

S

Argiope aurantia Lucas

ASON

TCOR

HST

c

SH

Argiope trifasciata (Forskal)

AS

O

H

R

H

Colphepeira catawba (Banks)

J— D

O

H

U

SV

*

Cyclosa conica (Pallas)

Je

R

H

R

S

Cyclosa turbinata (Walck.)

My

R

H

R

s

Eustala cepina (Walck.)

Ja AM

OSM

HT

R

VB

*

Eustala sp. A

My

0

H

R

s

*

Gasteracantha cancriformis (L.)

Ja JASOND

TO

ST

C

SH

Gea heptagon (Hentz)

AMJJA

CMOR

GH

c

SV

*

Glenognatha foxi (McCook)

AMJJAS

CMORL

GH

c

PSV

*

Hypsosinga sp.

Ap

O

H

R

s

*

Leucauge venusta (Walck.)

MJJ

FTCR

HS

c

SH

Mangora sp.

0

O

H

R

s

Mastophora phrynosoma Gertsch

O

T

T

R

H

*

Metazygia wittfeldae (McCook)

Je

R

H

R

S

*

Micrathena gracilis (Walck.)

JJ

F

ST

c

H

38 THE JOURNAL OF ARACHNOLOGY

Micrathena mil rata (Hentz)

Jy

F

ST

U

H

*

Micraihena sagittata (Walck.)

Jy

F

ST

U

H

Neoscona arabesca (Walck.)

MJJAS

CMORW

HS

A

SVH

Neoscona domiciliorum (Hentz)

My J

TCRS

HST

U

SB

Neoscona hentzii (Keys.)

AS

TCWRS

HST

c

SB

*

Neoscona nautica (L. Koch)

SO

T

T

u

H

*

Neoscona pratensis (Hentz)

AS

TC

T

u

S

*

Nuctenea sp.

O

O

H

R

s

*

Pachygnatha sp.

A JJA

CO

GH

c

PS

Tetragnatha elongata Walck.

AMJJAS

COR

HS

c

sv

*

Tetragnatha iaboriosa Hentz

AMJJASO

TCOWR

HS

A

SVH

Tetragnatha straminea Emerton

AMJJAS

COWR

HS

u

SV

Tetragnatha versicolor Walck.

AMJJASO

TCOR

HS

c

sv

Verrucosa arenata (Walck.)

Au

F

ST

u

H

Wixia sp.

AM J

COR

H

R

SV

ATYPIDAE

Sphodros bicolor (Lucas)

A Je

F

G

R

p

*

CLUBIONIDAE

Agroeca pratensis Emerton

Jy

L

G

R

p

Castianeira gertschi Kaston

s

O

GH

R

sv

Castianeira longipalpus (Hentz)

Ap N

O

GH

U

sv

Chiracanthium inclusum (Hentz)

JJA

CO

H

R

s

Clubiona abbotii L. Koch

AMJ ON

O

H

C

sv

Clubiona obesa Hentz

AMJ

O

H

u

sv

Clubionoides sp.

F AMJJAS

TCRS

HT

u

SB

*

Phrurotimpus sp.

A J AS

FOR

GH

c

PV

*

Scotinelia sp.

Je

R

H

R

s

*

Trachelas deceptus (Banks)

JA

B

G

c

H

*

Tracheias similis F.O.P.-Cambridge

Jy

R

H

R

s

*

Trachelas tranquillus (Hentz)

Au

O

G

R

p

Trachelas sp. A

Jy

R

H

U

s

DICTYNIDAE

Dictyna hentzi Kaston

AMJ

OR

GH

u

PSV

*

Dictyna sp. A

AM

OR

GH

u

PSV

DYSDERIDAE

Dysdera croc at a C. L. Koch

Ap

F

G

R

p

FILISTATIDAE

Filistata hibernalis Hentz

J— N

B

G

c

H

GNAPHOSIDAE

Calliiepis imbecilla Keys.

My S

OR

H

U

PS

Cesonia bilineata (Keys.)

Ap

S

T

R

B

Drassodes gosiutus Chamb.

Au

o

G

R

P

*

Drassyllus aprilinus (Banks)

My

o

G

R

P

Drassyllus covensis Exline

My

o

G

R

P

*

Drassyllus creolus Chamb. & Gertsch

My

o

G

R

P

*

Drassyllus dixinus Chamb.

My

o

G

R

P

*

Drassyllus ellipes Chamb. & Gertsch

Je

o

G

R

P

*

Drassyllus gynosaphes Chamb.

AMJ A

FTO

G

C

P

*

Drassyllus lepidus (Banks)

Jy

O

G

R

P

*

Drassyllus novus (Banks)

My

O

G

R

P

*

Gnaphosa fontinalis Keys.

JJA

OR

GH

C

PS

*

Gnaphosa sericata (L. Koch)

Au

O

G

u

p

Haplodrassus signifer (C. L. Koch)

My

R

H

R

s

Herpyllus ecclesiasticus Hentz

AM A 0

B

G

C

H

*

Micaria delicatula Bryant

AM ON

FO

GH

u

PV

*

Nodocion floridanus (Banks)

MAMJ D

S

T

C

B

*

Sergiolus capulatus (Walck.)

Je

o

G

R

P

*

Sergiolus minuius (Banks)

My

o

G

R

P

*

Sergiolus ocellatus (Walck.)

Je O

o

G

U

PV

*

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY

Urozelotes rusticus (L. Koch)

Ap

o

G

R

P

Zelotes aiken Platnick & Shadab

My

T

G

R

P

Zelotes duplex Chamb.

AM

TO

G

U

P

Zelotes hentzi Barrows

MAM

OR

GH

U

PS

Zelotes laccus (Barrows)

AM

R

G

c

P

HAHNIIDAE

Hahnia cinerea Emerton

J MAM D

FO

G

c

PV

Hahnia flavkeps Emerton

AM

0

G

R

P

Neoantistea agilis (Keys.)

N

o

G

u

V

LINYPHIIDAE

Ceraticelus emertoni (O.P.-Camb.)

J— D

TCMORL

GH

A

PSVB

Ceraticelus sp. A

My

0

G

U

P

Eperigone sp. A

AM

R

G

U

P

Eperigone sp. B

AM

R

G

U

P

Eperigone sp. C

Ap

R

G

R

P

Erigone sp. A

AM

R

G

U

P

Erigone sp. B

My

R

G

R

P

Floricomus sp. A

AM

R

G

U

P

Floricomus sp. B

Ap

R

G

R

P

Florinda sp.

AM

R

GH

U

PS

Frontinella pyramitela (Walck.)

J M Je D

O

H

c

s

Grammonota sp. A

AM

R

GH

c

PS

Grammonota sp. B

Ap

R

G

R

p

Linyphia sp.

AMJ

OR

H

u

s

Meioneta sp.

AM

R

G

R

p

Neriene radiata (Walck.)

JJA

COR

H

C

sv

Pityohyphantes sp.

My

R

G

R

p

LOXOSCELIDAE

Loxosceles reclusa Gert. & Mui.

J JJA ON

B

G

U

H

Loxosceles rufescens (Dufour)

F AS

B

G

R

H

LYCOSIDAE

Allocosa absoluta (Gertsch)

AMJ

R

G

C

P

Allocosa funerea (Hentz)

F AMJJ

ORE

G

c

P

Allocosa sp. A

Ap

R

G

R

P

Arctosa littoralis (Hentz)

Je A

OR

G

U

PS

Gladicosa bellamyi (Gert. & Wall.)

AMJJA

T

G

c

p

Gladicosa gulosa (Walck.)

J— N

FTO

G

A

PH

Gladicosa pulchra (Keys.)

JFMA S ND

FL

G

R

P

Lycosa acompa Chamberlin

FM MJ N

OL

G

C

PH

Lycosa annexa Chamb. & Ivie

J-J A-D

FTORLB

G

A

PH

Lycosa aspersa Hentz

MJ

TO

G

R

P

Lycosa antelucana Montgomery

A J A

OL

H

R

PH

Lycosa baltimoriana (Keys.)

S

0

G

R

P

Lycosa carolinensis Walck.

o

O

G

R

P

Lycosa georgicola Walck.

J AMJJASON

FTO

G

U

PH

Lycosa helluo Walck.

MJJ

FT

G

C

PH

Lycosa helluo group

MJJAS

CR

GH

c

H

Lycosa lento (Hentz)

J— D

COR

G

c

PH

Lycosa lento group

MJJAS

CR

GH

A

H

Lycosa punctulata Hentz

AM SON

TRO

GH

u

PH

Lycosa rabida Walck.

MJJAS N

CMORL

HS

A

SH

Pardosa atlantica Emerton

AM

R

GH

U

PS

Pardosa milvina (Hentz)

J— D

FTCMOWRL

GHS

A

PSVH

Pardosa saxatilis (Hentz)

MJ A

MOR

GH

C

PS

Pirata insularis Emerton

AMJJA

MOR

GH

c

PS

Pirata minutus Emerton

J AMJ N

OR

GH

u

PSV

Pirata sp. A

JJ

O

G

R

p

Schizocosa avida (Walck.)

AMJJASOND

FTMORL

GH

A

PVH

Schizocosa humilis (Banks)

S

O

G

R

P

40

THE JOURNAL OF ARACHNOLOGY

Schizocosa ocreata (Hentz)

AMJJASOND

FTOL

GH

C

PH

Sosippus mimus Chamb.

F

F

G

R

H

*

Traheops aurantiaca (Emerton)

F AMJJAS

FTOR

GH

U

PS

*

Trochosa avara Keys.

F— D

FTO

G

u

PS

MIMETIDAE

Mimetus sp.

Ap D

RS

T

R

HB

NESTICIDAE

Nesticus sp.

MJ

O

H

R

P

*

OECOBIIDAE

Oecobius sp.

My JA

B

G

R

H

*

OONOPIDAE

Orchestina saltitans Banks

MJ

B

G

R

H

*

OXYOPIDAE

Oxyopes salticus Hentz

AMJJASON

TCMOWRL

HS

A

PSVH

Peucetia viridans (Hentz)

Jy

F

S

R

S

PHILODROMIDAE

Ebo latithorax Keys.

MAMJ AS ND

FTOR

H

U

sv

Philodromus keyserlingi Marx

AM

O

G

R

VB

*

Phiiodromus marxi Keys.

AMJJASO

TORS

GHST

C

PSV

Philodromus placidus Banks

MJ

T

S

R

s

*

Philodromus vulgaris (Hentz)

My

S

T

R

B

*

Tibellus duttoni (Hentz)

AMJJAS

TOR

HS

A

SVH

*

Tibellus oblongus (Walck.)

Jy

R

H

R

S

PHOLCIDAE

Pholcus phalangioides (Fues.)
Psilochorus sp.

J D

Jy

B

CR

H

C

R

H

S

*

Spermophora meridionalis Hentz

AM

B

R

H

PISAURIDAE

Dolomedes triton (Walck.)

MAMJJASON

TMOWR

GH

A

PSVH

Pisaurina mira (Walck.)

AMJJA ON

TCOWRL

GHS

A

PSVH

Pisaurina undulata (Keys.)

J— D

FMOR

HS

A

SVH

Tinus peregrinus (Bishop)

O

0

H

R

s

SALTICIDAE

Agassa cyanea (Hentz)

Mr— D

TMOR

GH

C

PSV

Baiius cinctipes (Banks)

Je

0

H

R

S

*

Ballus sp. A

AS

OW

H

R

sv

Corythalia canosa (Walck.)

Je

B

R

H

*

Coryihalia latipes C. L. Koch

Ap

O

G

R

P

*

Eris aurantia (Lucas)

Jy

0

HS

U

s

Eris militaris (Hentz)

M MJJASO

TCMORS

HST

c

SB

Euophrys sp.

Jy

0

H

R

s

*

Evarcha hoyi (Peckhams)

AMJJASO

OR

H

C

sv

Habrocestum pulex (Hentz)

Je S

FO

GH

u

PS

Habronattus agilis (Banks)

MJJASO

MOR

HS

c

PSV

Habronattus calcaratus (Banks)

Jy

0

S

R

s

*

Habronattus coecatus (Hentz)

AMJJASO

MOWR

GH

A

PSV

Habronattus decorus (Blackwall)

Au

cw

H

U

H

Hentzia mitrata (Hentz)

JF Au

RS

HT

R

SB

Hentzia palmarum (Hentz)

JJ

CR

H

U

s

Maevia inclemens (Walck.)

MJ S

OWB

H

c

SH

Marpissa bina (Hentz)

Je

O

HS

u

S

Marpissa formosa (Banks)

AM

0

H

u

s

*

Marpissa lineata (C. L. Koch)

Mr

0

H

R

s

*

Marpissa pikei Peckhams

S

O

H

u

s

Metacyrba taeniola (Hentz)

A JJ

O

HS

u

HS

Metaphidippus galathea (Walck.)

F Ap— N

TCMOWR

HS

A

S

Metaphidippus protervus (Walck.)

AMJJASO

OR

HS

c

s

Metaphidippus tillandsiae Kaston

JFMAM OND

S

T

c

B

*

Neon neliii Peckhams

ASO

TOR

H

R

s

YOUNG ET AL.— SPIDERS OF WASHINGTON COUNTY 41

Neonella vinnula Gertsch

Au

R

H

R

S

*

Phidippus audax (Hentz)

Ap— D

COW

HST

A

SH

Phidippus dams Keys.

MJJASON

CO

HS

A

SH

Phidippus insignarius C. L. Koch

S

0

HS

U

S

Phidippus otiosus (Heotz)

MJ S

O

S

R

S

Phidippus princeps Peckhams

Jy

0

S

R

S

Phidippus purpuratus Keys.

Je

O

S

U

S

Phidippus puinami (Peckhams)

S D

S

T

R

B

*

Phiegra fasdata (Hahn)

Au

O

S

R

S

Platycryptus undatus (DeGeer)

JJAS

WBS

HT

U

SHB

Piexippus paykuHi (Audouin)

SO

O

H

R

S

*

Sarinda sp.

Jy

o

H

R

S

Sassacus papenhoei Peckhams

N

0

H

R

S

Siiticus cursor Barrows

AMJ

0

HS

R

S

*

Thiodina puerpera (Hentz)

MJ O

FO

H

C

S

Thiodina sylvana (Hentz)

MJJASO

CO

HS

U

S

Tutelina eiegans (Hentz)

JJASO

CO

HS

U

S

Tutelina similis (Banks)

My

R

H

u

S

Zygoballus nervosus (Peckhams)

My D

OS

HT

u

HB

Zygoballus rufipes Peckhams

AMJJ SOND

MO

GHS

c

SV

Zygoballus sexpunctatus (Hentz)

AMJ ASO D

OW

GH

u

V

SCYTODIDAE

Scytodes thoradca (Latr.)

J— D

B

G

u

H

THERIDIIDAE

Achaearanea globosa (Hentz)

JJ SO

B

G

R

H

Achaearanea tepidariorum

J— D

B

G

A

H

(C. L. Koch)

Euryopis furiebris (Heotz)

My

S

T

R

B

Latrodectus mac tans (F.)

J— D

CMOB

GH

U

H

Steatoda sp.

JJA

OR

H

u

S

*

Theridion frondeum Hentz

MJJASO

TMOR

GH

u

PSV

Theridula opuknta (Walck.)

MJJAS

MOR

GH

u

PSV

*

THERIDIOSOMATIDAE

Theridiosoma gemmosum (L. Koch)

Mr AS D

COR

GH

R

PSV

THOMISIDAE

Coriarachne sp.

AM

OR

H

R

SV

Misumena vatia (Clerck)

MJJAS

TCOR

HS

C

SH

Misumenoides formosipes (Walck.)

JJASON

TCOR

HS

c

SH

Misumenops asperatus (Hentz)

JJAS

TCOR

HS

u

S

Misumenops celer (Hentz)

JJA

COR

H

u

S

Misumenops oblongus (Keys.)

MJJ S

OR

H

u

S

Oxyptila monroensis Keys.

AMJ OND

FO

G

c

PS

Synaema parvuia (Hentz)

JA

CO

H

u

S

Xysticus ferox (Hentz)

AMJJAS

FOR

GH

c

PS

Xysiicus fraternus Banks)

AMJJAS

FO

G

c

PS

*

Xysticus triguttatus Keys.

Je

F

G

R

P

ULOBORIDAE

Uioborus sp.

MA JJAS

TO

HS

u

SV

Zosis geniculatus (Olivier)

Jy

0

S

R

s

te'-"' *' ;iv '''

ft ' aO •; .,

'3- /^'1&'Q'’

/!’k -5^,7

f { ■' :^..a‘ i-' . Va^ V\

di**/ »/■•■ ; / »,s > -T' .

, TV - 4 ^ itvi’ -v^ V ', '

,,ij;T,;-TJ»i,','m:f.-,

"‘t;-

^7Vv/>Cv‘^ /■: ’,:

- -Tij- J - '•' ) ’■' .V'--fc.--> ■' ■■ - :■' i- v'^ ■ ''‘'*m^a'-'‘'*V''^>'*4^'-- ■- "

'"4

Young, O. P. 1889. Predation by Pisaurina mira (Araneae, Pisauridae) on Lygus iineolaris
(Heteroptera, Miridae) and other arthropods. J. ArachnoL, 17;43“48.

PREDATION MY PISAURINA MIRA (ARANEAE, PISAURIDAE)
ON LYGUS LINEOLARIS (HETEROPTERA, MIRIDAE)
AND OTHER ARTHROPODS

Orrey P. Young

Southern Field Crop Insect Management Laboratory
ARS^USDA, RO. Box 346
Stoneviile, Mississippi 38776 USA

ABSTRACT

In the Delta area of Mississippi, a 13-morith sampling program in old-field habitats adjacent to
cotton fields demonstrated high densities of Pisaurina mira in June, which was coincident with high
Lygus Iineolaris populations. In July, E mira populations were lower by 50% and L. iineolaris by
almost 100%, From August through October, P. mira populations continued to decline while L.
Iineolaris increased. Field observations in the same old-field habitats indicated L. Iineolaris to be the
most frequently captured prey item of P. mira. In laboratory feeding experiments over a 3-year period,
all P. mira individuals consumed L, Iineolaris and 22 of 35 other species of co-occurring potential
arthropod prey. These data suggest that the predator R mira may affect L. Iineolaris populations and
can survive on other prey when L. Iineolaris is less abundant.

INTRODUCTION

Members of the Pisauridae in North America are wandering spiders that do
not build snares and are typically found on vegetation or at water margins
throughout North America (Carico 1972). Pisaurina mira (Walckeeaer) is one of
the most common spiders in the eastern United States and occurs in woods, old-
fields, and meadows, but is especially abundant in the ecotonal areas between
woods and fields (Carico 1972). This species can be an important predator in row
crops and has been recorded from rice (Woods and Harrel 1976), sugarcane
(Negm et al. 1969), peanuts (Ageew et ai. 1985), soybeans (LeSar and Uezicker
1978), cotton (Whitcomb and Bell 1964), and alfalfa (Wheeler 1973). R mira has
been documented recently in the laboratory as a voracious predator on the
tarnished plant bug (TPB), Lygus iineolaris (Palisot) (Heteroptera, Miridae)
(Young 1989). The TPB is an economically important pest on many crops in the
United States, feeds on over 350 species of plants, and is abundant in habitats
adjacent to row crops (Young 1986). Since these are the same habitats in which P.
mira is abundant, it is possible that this spider could have a significant impact on
TPB populations. The purpose of this study was to delineate the seasonal
populations of P mira in field edge habitats, and to determine which arthropods
were present in those same habitats which R mira could feed upon.

44

THE JOURNAL OF ARACHNOLOGY

Figure 1. — Average number per sample of
Pisaurina mira adults and immatures for each of
13 consecutive months in old-field habitats adja-
cent to cotton fields in Washington County,
Mississippi.

METHODS AND MATERIALS

Field. — Arthropods were collected by sweepnet and/or vacuum (D-vac®), 10
sweeps or 10 row-feet per sample, at 15 undisturbed early-successional sites
adjacent to cotton fields in Washington County, Mississippi. Sampling was
conducted on alternate weeks during the period October 1985 through October
1986. All P. mira and potential prey were removed from the samples, counted,
and determined as to immature or adult. On an irregular basis throughout the 13-
month period, observations of P. mira predation on various prey at these same
sites were recorded.

Laboratory. — Individuals of P. mira, obtained from the above-mentioned and

other similar sites during 1984-86, were placed in plastic cups (4 by 10 cm) with a
cloth-mesh cover and maintained in an environmental chamber at 25° C, 80%
RH, and a 14:10 (L:D) photoperiod. The feeding protocol involved starving
(water available) each spider for the seven days after capture, placing an
individual live prey in each container, recording the prey status after 24 hours,
and removing the unconsumed material. Spiders not consuming prey were offered
one adult TPB, which was always accepted. The starvation — feeding — recording
cycle was then repeated several times for each spider using different potential prey
in each cycle. These feeding trials were conducted during the late summer and fall
of 1984-1986, with both P. mira and potential prey collected from the same sites.
Each potential prey species was offered to 3-10 different spiders, depending upon
availability of the prey.

RESULTS AND DISCUSSION

Seasonal occurrence. — During the 13-month sampling period, 44 adult P mira
were captured, with 98% occurring in October and November (Fig. 1). Three-
hundred and seventy immatures were also captured, with the largest number
(40%) occurring in June. These patterns of occurrence indicate that adults and
late-instar immatures overwinter, that eggs hatch in May with dispersal of
immatures from nursery webs in June, and that P mira is univoltine. Published
information from other collection sites indicate a similar pattern. Adults of P
mira are most abundant during June in Illinois (Jones 1940), Kansas (Fitch 1963),
and Tennessee (Gibson 1947), and during May in North Carolina (Berry 1971)
and Arkansas (Peck et al. 1971). Immatures were most abundant during

YOUNG PREDATION BY PISAURINA MIRA

45

Table 1. — Field observations of prey capture by Pisaurina mira.

Prey taxon

Prey stage

No. observations

Heteroptera: Miridae

Lygiis lineolaris (Palisot)

Adult

2

L. lineolaris

Immature

3

Polymerus hasalis (Reuter)

Adult

1

Homoptera: Cicadellidae, Undet. sp.

Immature

2

Coleoptera: Chrysomelidae

Diahrotica undecimpunctata howardi Barber

Adult

2

Diptera: Muscidae, Undet. sp.

Adult

1

Hymenoptera: Apidae

Apis sp.

Adult

1

September in Ohio (Elliott 1930) and Tennessee (Gibson 1947), during August in

North Carolina (Berry 1971) and Kansas (Fitch 1963), and abundant throughout
the June to October period in Illinois (Jones 1940) and Arkansas (Peck et al.
1971). Mississippi populations appear to develop earlier than those further north,
but there is no evidence for a second generation.

Field observations of predation. — A survey of the spider literature revealed an
absence of records involving prey of R mira. Considering the dense and low
vegetation in which this species is most abundant, perhaps the lack of prey
records is not surprising. During several hundred hours of field work over a 13-
month period, only 12 P. mira with prey were recorded (Table 1). The species
captured were among the most abundant species present at the time of the
observations. Five of the 12 prey records (43%) involved the TPB, and since the
TPB comprised considerably less than 43% of the arthropod population (Young
unpubl. data), it is possible that R mira demonstrated some specificity for the
TPB.

Laboratory observations. — Over a 3-year period, 41 individuals of R mira were
offered 36 species of prey representing 22 families of insects in eight orders and
three families of spiders (Table 2). Not all of these potential prey were captured
and consumed, as 77 of the 179 specimens were rejected (43%). Thirteen species
(56 individuals) were completely rejected by R mira. Situations in which the
potential prey was as large or larger than the spider, or considerably smaller,
usually resulted in an absence of prey capture, as did the presence of probable

Table 2. — Laboratory observations of prey capture by Pisaurina mira.

Prey taxon

Prey
life state

Prey mean
body length
(mm)

P. mira mean
body length
(mm)

Prey consumed

Yes No

COLEOPTERA

Carabidae

Lebia viridis (Say)

Ad

5

5

6

Coccinellidae

Cycloneda munda (Say)

Ad

5

6

4

Hippodamia convergens Guerin

Ad

6

7

8

Chrysomelidae

Diabrotica undecimpunctata
howardi Barber

Ad

7

9

2 3

46

THE JOURNAL

OF ARACHNOLOGY

DIPTERA

Syrphidae, Undetermined species

Ad

6

6

4

Calliphoridae, Uedet. sp.
HEMIPTERA

Ad

6

7

3

Coreidae, Undet. sp.

Lygaeidae

Imm

14

10

4

Blissus sp.

Ad

3

7

2

5

Geocoris punctipes Say

Ad

4

5

6

2

Oncopeitis sp.

Ad

15

8

3

Oncopeltis sp.

Miridae

Imm

8

9

2

Lygus lineolaris (Palisot)

Ad

5

7

10

Taylorilygus pallidulus Blanchard
Nabidae

Ad

4

6

5

Reduviolus roseipennis (Reuter)
Tropiconabis capsiformis

Ad

7

8

4

(Gemar)

Pentatomidae

Ad

8

7

1

2

Stiretrus anchorago (E)

Reduviidae

Ad

8

7

1

2

Sinea diode ma (F.)

Ad

13

10

3

Zelus sp.

HOMOPTERA

Cicadellidae

Imm

12

13

6

Chlorotettix sp.

Ad

5

7

6

Gyponana sp.

Ad

5

9

3

Fulgoridae, Undet. sp.

Membracidae

Ad

8

9

2

Spissistilus festinus (Say)
HYMENOPTERA

Halictidae

Ad

5

8

6

Augochlor sp.

Ad

5

7

6

Undet. sp.

LEPIDOPTERA

Ad

8

13

3

Geometridae, Undet. sp.

Noctuidae

Imm

11

10

4

Heiiothis sp.

Imm

7

7

6

Spodoptera sp.

Syntomidae

Imm

7

7

6

Ctenucha sp.

Yponomeutidae

Ad

9

9

3

Atteva sp.

NEUROPTERA

Chrysopidae

Ad

10

8

4

Chrysopa sp.

ORTHOPTERA

Imm

6

6

5

Acrididae, Undet. sp.

Tettigoniidae

Ad

8

8

3

Neoconocephalus sp.

ARANEAE

Oxyopidae

Imm

5

7

5

Oxyopes salticus Heetz

Salticidae

Imm

4

8

8

Metaphidippus sp.

Imm

3

9

3

3

M. galathea (Walck.)

Ad

4

6

5

4

Phidippus sp.

Thomisidae

Imm

6

9

4

Misumenoides formosipes
(Walck.)

Imm

2

6

3

YOUNG— PREDATION BY PISAURINA MIRA

47

Figure 2. — Average number per sample of Lygus
lineoiaris adults and nymphs for each of 13
consecutive months in old-field habitats adjacent
to cotton fields in Washington County, Mississippi.

distasteful prey (coccinellids and coreids). When the offered prey was also a
predator, such as nabids or other spiders, prey capture sometimes did not occur.
The size range of P. mira adults and juveniles utilized in these feeding trials was
5-13 mm body length and that of accepted prey was 4-14 mm. Twenty-three
species were accepted as prey under these laboratory conditions, and four
additional species were observed as prey in the field. These 27 prey species were
among the most abundant arthropods at the sampled sites, suggesting that P.
mira can capture most arthropod species on vegetation within a particular size
range. It must be remembered, however, that all laboratory spiders were offered,
and accepted, adult TPB at some time during their captivity.

Potential impact on TPB populations. — The seasonal distribution of TPB at
the same sites from which P. mira was collected is presented in Fig. 2. Population
peaks of TPB occur in October and June, with the June peak followed in July by
a veritable absence of TPB. The population peak of P. mira was also in June
(Fig. 1). This correspondence of high population levels in June indicates that the
TPB was abundantly available as prey for P. mira at that time. Subsequent
population decline of TPB in July (Fig. 2) may have been due to emmigration
out of field margins into cotton and/or mortality due to predators such as P.
mira. Field and laboratory observations also suggest that when TPB is abundant,
P. mira will be a frequent predator of this prey, and when TPB is relatively
scarce, P mira can readily feed on other prey.

ACKNOWLEDGMENTS

The technical assistance of T C. Lockley is appreciated, as is the manuscript
review provided by G. L. Bernon, J. E. Carico, G. B. Edwards, S. J. Fleischer, J.
L. Hayes, T. J. Kring, and S. H. Roach.

LITERATURE CITED

Agnew, C. W., D. A. Dean and J. W. Smith. 1985. Spiders collected from peanuts and non-
agricultural habitats in the Texas west crosstimbers. Southw. Natur., 30:1-12.

Berry, J. W. 1971. Seasonable distribution of common spiders in the North Carolina Piedmont.
American Midi. Natur., 85:526-531.

Carico, J. E. 1972. The nearctic spider genus Pisaurina (Pisauridae). Psyche, 79:295-310.

Elliott, F. R. 1930. An ecological study of the spiders of the beech-maple forest. Ohio J. Sci., 30:1-22.
Fitch, H. S. 1963. Spiders of the University of Kansas Natural History Reservation and Rockefeller
Experimental Tract. Univ. Kansas Mus. Nat. Hist., Misc. Publ., 33:1-202.

48

THE JOURNAL OF ARACHNOLOGY

Gibson, W. W. 1947. An ecological study of the spiders of a river-terrace forest in Western Tennessee.
Ohio J. Sci., 47:38-44.

Jones, S. E. 1940. An annotated list of the spiders of an East Central Illinois forest (Wm. Trelease
Woods, Univ. Illinois). Trans. Illinois State Acad. Sci., 33:216-220.

LeSar, C. D. and J. D. Unzicker. 1978. Soybean spiders: Species composition, population densities,
and vertical distribution. Illinois Nat. Hist. Surv. Biol. Notes, 107:1-14.

Negm, A. A., S. D. Hensley and L. R. Roddy. 1969. A list of spiders in sugarcane fields in Louisiana.
Proc. Louisiana Acad. Sci., 32:50-52.

Peck, W. B., L. O. Warren and 1. L. Brown. 1971. Spider fauna of shortleaf and loblolly pines in
Arkansas. J. Georgia Entomol. Soc., 6:87-94.

Wheeler, A. G., Jr. 1973. Studies on the arthropod fauna of alfalfa. V. Spiders (Araneida). Canadian
Entomol., 105:425-432.

Whitcomb, W. H. and K. Bell. 1964. Predaceous insects, spiders, and mites of Arkansas cotton fields.
Univ. Arkansas Agric. Exp. Stn. Bull., 690:1-84.

Woods, M. W. and R. C. Harrel. 1976. Spider populations of a Southeast Texas rice field. Southw.
Natur., 21:37-48.

Young, O. P. 1986. Host plants of the tarnished plant bug, Lygus lineolaris (Heteroptera: Miridae).
Ann. Entomol. Soc. America, 79:747-762.

Young, O. P. 1989. Predators of the tarnished plant bug, Lygus lineolaris (Heteroptera: Miridae):
Laboratory evaluations. J. Entomol. Sci., 24:174-179.

Manuscript received August 1988, revised September 1988.

Morse, D. H. 1989. Nest acceptance by the crab spider Misumena vatia (Araneae, Thomisidae). J.
ArachnoL, 17:49-57.

NEST ACCEPTANCE BY THE CRAB SPIDER
MISUMENA VATIA (ARANEAE, THOMISIDAE)

Douglass H. Morse

Graduate Program in Ecology and Evolutionary Biology
Division of Biology and Medicine
Brown University

Providence, Rhode Island 02912 USA

ABSTRACT

I tested whether female crab spiders Misumena vatia responded differently to their own nests and to
other spider nests or parts of nests. They accepted nests of conspecifics as readily as their own,
although initially showing more activity on others’ nests than their own. They accepted intact M. vatia
nests more often than nest silk placed on leaves, turned-under leaves, or artificial nests; and unaltered
leaves less frequently than any of the latter. They accepted nests of other spider species less frequently
than conspecific nests, probably in response to the differences between these nests and conspecific
nests. Some crab spiders about to lay their eggs accepted conspecific nests, but hunting adult females
did not accept them more than predicted by chance.

INTRODUCTION

Parental care has recently engendered considerable interest because of its
importance to theories of parental investment (Trivers 1972), parent-young
conflict (Trivers 1974) and life histories (Stearns 1976, 1977). Considerably less
attention has been simultaneously paid to the proximal cues used by parents to
identify their young or their nest-sites. Nevertheless, these cues are crucial to the
success of parental strategies and are therefore demanding of attention. In the
past such cues have been the subject of considerable attention in their own right
as sign stimuli or releasers (e. g., Tinbergen 1951).

Parental care ranges from guarding eggs to caring for them and the resulting
offspring until they approach reproductive maturity. The extent of this
commitment depends in part on the ability of parents to perform a variety of
acts, ranging from site tenacity to recognizing their own nests, and their
predisposition to care for eggs or young (see Trivers 1974). Sometimes it is
unclear whether parental care results from site tenacity or recognition, especially
among invertebrates, including ones whose behavioral patterns might help to
clarify the evolution of parental care and other types of sociality (Wilson 1975),
Understanding the conditions, cues, and mechanisms associated with parental
care in a broad range of animals is thus particularly desirable.

Here I provide the results of studies on nest recognition by the semelparous
crab spider Misumena vatia (Clerck) (Thomisidae). Female crab spiders lay their
single egg mass in a nest that they construct from a large ovate leaf. They build
this nest by bending under the distal tip of the leaf and securing it to the under
side of this leaf (mid-section). Eggs are laid on silk within the cavity produced.

50

THE JOURNAL OF ARACHNOLOGY

and the lateral sides are then drawn tight (Morse 1985). They subsequently guard
this nest, often until young have emerged from the nest nearly four weeks later
and sometimes until all the young have dispersed from the nest, as much as 12
days after that time (Morse 1987). Guarding behavior enhances survival of the
affected offspring (Morse in press).

These spiders seldom wander far from the nests during the guarding phase, and
they usually are connected to their sites by a silken line, which would facilitate
return. However, they occasionally become separated from their nests. These
individuals do not appear to have abandoned the sites “intentionally”, for they
may assume guarding positions in similar locations on nearby plants {N = 4).

Misumena vatia can be readily removed from their nests for weighing at any
time during the guarding period and then returned, without affecting their
desertion rate (Morse 1987). This result suggested that they could be transferred
to other selected substrates, most obviously nests of conspecifics, in order to
compare their performance with those on their own nest sites.

In this paper I present the results of experiments designed to establish whether
female crab spiders respond differently to their own nests and to other nests. Also
I investigated several cues that may facilitate nest recognition. Specifically, I ask:
1) can crab spiders recognize their own nests, or 2) nests of their own species? 3)
If so, what characteristics are crucial in their identifications? If not, to what cues
do they respond that their nests have in common with other features of the
environment?

METHODS

All of the post-laying Misumena used in this study constructed their nests on
milkweed {Asclepias syriaca L.) leaves in a field in Bremen, Lincoln Co., Maine
(see Morse 1985). These spiders were removed from their natural nest sites within
a week of laying and placed on a variety of sites for 1-hour periods, and their
movements recorded during this time. Spiders were characterized as active if they
moved on the nest or changed orientation five or more times per hour, or if they
left the nest during this time. A random sample of spiders brooding their own
eggs moved 3.3 ± 3.8 times per h ( x ± SD) {N = 34) (Morse 1987).

Tests included spiders removed from and then returned to their own nests and
ones placed on other M. vatia nests; as well as individuals placed on superficially
similar nests of other thomisid, salticid, and theridiid spiders; on unaltered
milkweed leaves; on leaves turned under by spiders or by the investigator; and on
leaves with M. vatia nest silk placed on them. Controls were also run to
determine that one-hour periods insured a representative bout of behavior. After
the experiments the spiders were returned to their own nest sites. For
comparison, similar transplantation experiments were run on two other groups of
spiders: “broody” spiders, which had turned under leaves preliminary to egg-
laying; and large, actively-feeding spiders. Both were placed on completed
Misumena nests. All of these presentations were sequential, in keeping with the
natural patterns observed in the field. Thirty spiders were used in each
experiment, none of the individuals in more than one experiment.

Another group of spiders was periodically tested for recognition of their own
nests over the entire period that they normally would guard nests. I removed
these individuals from their nests immediately after they laid their eggs and

MORSE— CRAB SPIDER NEST ACCEPTANCE

51

Table 1. — Responses and activity levels of spiders placed on different substrates. Only the first 15-16
replicates were scores for activity in Experiments 2-11. N = 30 in each experiment, a = All at
guarding stage except for Experiments 10 and 11. b = Had turned under leaf preparatory to laying.
Laid 1-5 days later (x ± SD = 2.5 ± 0.7 days), c = Actively hunting and showing no indication of
nest building. Laid 4-12 days later (x ± SD = 8,2 + 3.9 days.

Experiment

Treatment'"

Remain
on site

Leave site

Stationary

Active

1

Monitor, remove & return

Before removal

30

0

25

5

After return

30

0

22

8

2

On other nest and return

On other nest

29

1

12

18

After return

30

0

20

10

3

Unaltered leaf

4

26

1

14

4

Leaf turned under by spider

14

16

4

11

5

Nest silk on leaf

15

15

6

9

6

Artificial nest

19

11

8

7

7

Xysticus nest

18

12

4

11

8

Metaphidippus nest

15

15

7

8

9

Enoplognatha nest

10

20

4

11

10

Misumena nest*"

13

17

8

7

11

Misumena nest*'

5

25

3

13

completed their nests. They were placed inside large bags of nylon tricot that
covered plants similar to the ones upon which they laid and subsequently kept in
these bags, except when they were run in experiments on their own nests, either
once or every several days. These manipulations permitted an assessment of
whether time away from the nest affected the tendency to guard a nest.

RESULTS

Experiment 1. Spiders removed from their own nest and returned. — In an
initial test, several spiders were monitored for an hour to record activity, then
removed from their nests, moved about their home milkweed clone in a small
shell vial concealed from any possible visual cues, and returned to their sites and
monitored for another hour. Only a small percentage of the individuals exhibited
regular activity prior to removal (Table 1), and this number did not differ
significantly from those exhibiting activity after their return {G = 0.89, P > 0.3).
Thus, removal in itself does not measurably affect the behavior of these species
and should not be the basis for any differences reported in Experiments 2-9
presented below. This pattern of low activity characterizes guarding spiders (see
Methods).

Experiment 2. Spiders switched to other recently-occupied Misumena vatia
nests. — Spiders laying at similar times were removed from their nests and

immediately placed on each others’ nests for an hour. During this period one of

the 30 individuals left the nest after 5 min. It was replaced and remained the next
55 min until removed. All other individuals remained on their foster nests for the
full 1-hour period. Thus, no significant difference occurred between the tendency
to remain on their own nests (Experiment 1) and those of other individuals at a
similar stage of development {G = 1.40, df = 1, P> 0.2). However, the activity

52

THE JOURNAL OF ARACHNOLOGY

of these displaced spiders on different nests was greater than that of similarly
displaced ones returned to their own nests (Experiment 1) {G = 6.93, P < 0.01).
The spiders made several lines that secured the nest to nearby leaves or other
structures. The spiders became less active when they were returned to their own
nests after one hour, (G = 4.34, P < 0.05) (Table 1), but some activity still
remained. The number of active individuals did not significantly exceed that of
Experiment 1 (G = 0.16, P> 0.5).

Experiment 3. Bare milkweed leaves. — I placed spiders on leaves of similar size
and location to those used for nests, which had not been manipulated by spiders.
Only four of the 30 M. vatia placed on these leaves remained for an hour, which
differs significantly from the behavior of spiders returned to their own nests (G =
58.55, P < 0.001) or put on other M. vatia nests (G = 50.25, P < 0.001). Clearly
spiders do not respond to unmodified leaves in the way that they respond to their
own nests or those of conspecifics. This experiment indicates that the spiders are
active enough to leave their sites within an hour. Failure of the spiders in
Experiments 1 and 2 to move from the nests is thus not a consequence of a low
rate of activity.

Experiment 4. Leaves with tips turned under by other Misumena vatia. — One
of the distinctive features of a M. vatia nest is that the leaf used is turned down
at the tip and folded under by the parent spider (Morse 1985: fig. 1). The first
step in making a nest is to turn a leaf under and secure it with strands of silk
between the distal tip and the ventral side of the leaf’s midrib. After performing
this act spiders typically remain in the resulting shelter for 1-3 days before laying
their eggs and drawing the sides of the leaf tightly together with silk to complete
the nest. This position differs from the one assumed by spiders that have already
laid their eggs. Several spiders were collected for experiments at the leaf-turning
stage, resulting in a ready source of such leaves for this experiment. Only 14 of 30
brooding spiders placed on these leaves remained there for the full test period.
This response is significantly weaker than from the response to both their own
nests (G = 78.13, P < 0.001) (Experiment 1) and to the nests of other M. vatia
(G = 21.30, P < 0.001) (Experiment 2), but significantly stronger than the
response to unmodified leaves (Experiment 3) (G = 8.29, P < 0.01). Thus,
turned-under leaves appear to be a stimulus for remaining at a site, although not
as strong a stimulus as a completed nest.

Experiment 5. Leaves with Misumena vatia silk applied to their distal ends. —
Another potentially important feature promoting site tenacity might be the
presence of silk. Misumena vatia place considerable amounts of silk about their
nests, noticeably stiffening the distal part of the leaf. This silk can be readily
removed as a sheet from the surface of the nests. In Experiment 5, sheet-like silk
was removed from other nests and placed on the distal 1 cm of both upper and
lower surfaces of otherwise undisturbed leaves similar to those usually used as
nest sites. Except for the silk added, these leaves were similar to the leaves used
in Experiment 3.

Half of the spiders remained on these sites during the experiment. Their
frequency of staying is significantly lower than that of spiders placed on the nests
of other conspecifics (G = 19.23, P < 0.001) (Experiment 2) or returned to their
own nests (G = 25.89, P < 0.001) (Experiment 1). This response was stronger
than the response to plain leaves (Experiment 3) (G = 9.77, P < 0.01) but not the
response to turned-under leaves (Experiment 4) (G = 0.07, P > 0.7). Thus, the

MORSE— CRAB SPIDER NEST ACCEPTANCE

53

presence of silk also appears to play a role in determining what constitutes a nest,
but it is not as strong as a completed nest.

Experiment 6. Artificial nests secured by thread. — It is possible that the results
in Experiment 4 were a consequence of the distal and medial parts of the leaves
being only partly apposed to themselves, rather than due to a lack of silk. To test
this possibility, I sewed leaves into the form of nests with fine white thread,
placing segments of dried timothy grass {Phleum pratense L.) inflorescences inside
to approximate the shape of a nest. Spiders in this experiment could not crawl
inside the nest, which distinguishes this manipulation from Experiment 4.

A majority of the spiders remained on these artificial nests for the entire period
of the experiments. This result did not differ from that of Experiment 4 (leaves
turned under by spiders), in which the spiders had access to leaves partially
turned under by spiders {G = 1.69, P > 0.1). It was weaker than the response to
the nests of other M. vatia (Experiments 1 and 2), however (G = 17.74 and 11.85,
P < 0.001, < 0.001). Thus, turned-under, closed leaves may provide important
cues for spiders, but further experiments would be required to establish whether
they differ from simple turned-under leaves.

Experiment 7. Nests of other species of crab spiders emertoni. — The
brown crab spider Xysticus emertoni (Keyserling) sometimes places its egg sacs in
positions similar to those of M. vatia. Misumena vatia placed on X. emertoni
nests remained significantly more frequently than they did on unaltered milkweed
leaves (Experiments 1 and 2) {G = 19.67, 14.97, P < 0.001, < 0.001). However,
their response did not differ from those to turned-under leaves {G = 1.07, P >
0.2), silked leaves (Experiment 5) (G = 0.61, P > 0.3) or artificial nests
(Experiment 6) (G = 0.07, P > 0.7).

Experiment 8. Jumping spider nests. — The small jumping spider Metaphidippus
insignis (Banks), abundant in the study area, builds nests on milkweed leaves that
are similar in location and general characteristics to M. vatia nests and guarded
from within. Half of the M. vatia placed on M. insignis nests remained on them
for the entire hour. Thus they responded more strongly to them than they do to
bare leaves (Experiment 3) {G = 9.77, P < 0.01) and more weakly than to their
own nests (Experiment 1) {G = 25.89, P < 0.001). However, their response does
not differ significantly from the response to the artificial nests (Experiment 6) (G
= 1.09, P>02).

Experiment 9. Theridiid spider nests. — A theridiid spider Enoplognatha ovata
(Clerck) occurs in the study area in small numbers. It also builds nests by turning
under leaves (Wise and Reillo 1985), often milkweed leaves, but these nests are
less stiff than M. vatia nests, probably because Enoplognatha ovata do not use
large amounts of sheet-like silk in construction. They also guard their nests from
inside. Nests built on milkweed leaves were chosen for a set of experiments.

Only one-third of the M. vatia remained on E. ovata nests for an hour,
significantly fewer than the number that remained on their own nests (Experiment
1) (G = 38.19, P < 0.001), or artificial nests (Experiment 6) (G = 5.49, P < 0.02).
However, this result does not differ significantly from the number remaining on
plain leaves (Experiment 3) (G = 3.44, E > 0.1) or M. insignis nests (G = 1.72, P
>0.1).

Experiment 10. Pre-laying Misumena vatia placed on nests. — Spiders that had

already turned under leaves in apparent preparation for egg laying (“broody”)
were placed on completed nests, as in Experiment 2. Significantly fewer

54

THE JOURNAL OF ARACHNOLOGY

individuals remained on the nests for an hour than did post-laying individuals on
other nests (Experiments 1 and 2) (G = 30.47 and 23.48, P < 0.001, < 0.001).
Although relatively active at this time, spiders did exhibit a significantly greater
tendency for site tenacity prior to laying their own eggs than did post-laying
individuals on a plain leaf (Experiment 3) {G = 6.91, P < 0.01).

Experiment 11. Active Misumem vatia placed on nests. — A group of spiders
within the size range of egg-laying individuals, but which had not shown any
signs of broodiness (i.e., ones that had not turned under a leaf, and were still
actively hunting), were also tested. These spiders showed relatively little tendency
to remain on nests, significantly fewer remaining than for the broody group
(Experiment 10) (G = 5.22, P < 0.05), or individuals that had already laid
(Experiments 1 and 2) (G = 54.47 and 46.31, P < 0.001, < 0.001). These
individuals did not differ from post-laying spiders that were placed on leaves
randomly (Experiment 3) {G = 0.13, P> 0.5).

Experiment 12. The effect of time away from nest on response. — Individuals in
Experiments 1-11 were tested within a few days of laying their eggs. The tendency
to remain could thus be a consequence of whether a spider was already guarding
its nest. If spiders maintain strong site tenacity, contact with the nest might
suffice to retain a predisposition to guard. However, if somehow separated from
its nest, the spider might sometimes reoccupy it in the process of random
movement by contacting the lines of silk in the vicinity of the nest that are
normally laid wherever an individual goes. In the process of this study two
spiders, both of known history, that had earlier left their nests established
residence at abandoned nests of other conspecifics. These observations are
consistent with individuals becoming separated from their own nests, but
retaining an affinity to nest cues. These two individuals had been away from their
own nests only one and three days, however, which raised the question of how
long they would retain this predisposition in the absence of regular contact with
nests.

Spiders removed from their nests after laying and placed in nylon tricot bags
on similar plants exhibited a strong correlation between time of separation from
the nest and tendency to guard when returned (Fig. 1). This ranged from an 88%
tendency to remain when away from the nest for 1-5 days to a 20% tendency to
remain when removed from 26-30 days, roughly the period between laying and
emergence of young (Morse 1987). The pattern of decrease in tendency to remain
appears to be somewhat stepped, but relatively slow. The number of confined
individuals remaining on their former nest sites for the 1-hour testing period was
significantly lower than that of unconfined, free-living individuals at each period
{G = 4.40, P < 0.05 for one five-day period to G = 9.82, P < 0.01 for the 6-10
day period), suggesting that absence from the nests affects the probability of
remaining on them.

This test measures the effect of absence from the nest; however, since each
period away from the nest was about 5 days, it does not directly measure true
time away from nests. Each of these individual tests might represent a mere five-
day absence to the spider, whether five or 30 days from the initial removal.
Individuals away from the nests 15 and 30 days from the initial removal were
used to compare these effects (Fig. 1). They revealed no clear difference, falling
very close to the periodic removals at 11-15 days (G = 0.03, P > 0.8) and not

MORSE— CRAB SPIDER NEST ACCEPTANCE

55

° 1-5 6-10 11-15 16-20 21-25 26-30

DAYS AFTER LAYING

Figure 1. — Percentage of spiders remaining on their nests for one hour after being separated from
these nests for five-day intervals (filled circles), percentage remaining for one hour if separated for 11-
15 or 21-25 days (x’s). Numbers to right of symbols are A’s. Percentage of 37 unscreened nesting
individuals remaining on nest at the same time in the study area (open circles).

significantly different at 21-25 days, at a time that the tendency to remain at the
nest was dropping rapidly (G = 0.70, P > 0.3).

Experiment 13. Tenure times of natural and cross-fostered spiders. — The cross-
fostered individuals remained on their new nests for a period similar to that of
undisturbed spiders (x ± SD = 29.3 ± 11.7 vs 27.6 ± 13.3 days, N = 40, 30; P
> 0.5 in a two-tailed Mann- Whitney f/-Test). This result further supports the
conclusions from Experiment 2. Thus, both Experiments 1 and 2 are probably
typical of the results that would be obtained if they were to be run for more than
one hour.

DISCUSSION

The response of Misumena vatia to these stimuli does not appear to be
individual-specific, or even species-specific. It appears related to shape and tactile
characteristics. The difference in activity between spiders returned to their own
nest or placed on a conspecific’s nest appeared to be related to the presence or
absence of silken lines to adjacent leaves. Active individuals spent much of their
time spinning these lines. The response of the spiders to a variety of different
stimuli suggests that multiple factors play a role in nest retention. Leaves turned
under by other spiders, leaves with spider nest silk affixed to them, and

56

THE JOURNAL OF ARACHNOLOGY

artificially constructed nests (Experiments 4-6) all elicited a much more frequent
response than did unaltered leaves (Experiment 3), although not as strong a
response as to complete nests (Experiments 1 and 2). Resemblance of other
spiders’ nests to one or both cues was probably the key factor in determining how
M. vatia responded to them (Experiments 7-9). The frequency of the response
could be a consequence of either the characteristics of the silk, or of its volume,
Enoplognatha ovata nests are extremely pliable in comparison to the other nests,
especially the M. vatia nests, probably because members of this species apply less
silk to the outside of their nests than do the others. If M. vatia respond primarily
to tactile cues, they might perceive E. ovata nests as quite different from their
own. Alternatively, the spiders may have responded primarily to the shape of the
nests. This explanation is supported by their relatively strong response to the
artificial nests.

Does the ability to reclaim a lost nest site have any selective significance, given
that individuals normally occupy their nests continually, so that site tenacity itself
might normally suffice to insure continuous occupation? Several factors may
result in short-term disappearance, and, with the advantage that guarded nests
produce more offspring than unguarded ones (Morse in press), reoccupation
should be favored. I have observed several instances of temporary disappearance
of M. vatia females from their nests in response to potential predators of either
the spider or the egg mass. In 1983, several egg masses were attacked by an
unknown predator. Two females that disappeared from their nest plants at this
time returned to their nests within a day and resealed them tightly with silk.
Spiderlings subsequently emerged from remaining uneaten eggs in both nests.
Another spider was driven from its nest by ants {Formica sp.) shortly after
completing it. It left the plant at this time, but reoccupied its nest within a day.
Deer {Odocoileus virginianus) occasionally feed on milkweed in small amounts
(Wilbur 1976), and I have had one spider nest inadvertantly consumed in this
way. Another nest plant was badly damaged at the same time, and the spider
disappeared in the process. However, it reappeared the following day and secured
its badly damaged nest in a way that protected it through the rest of the nesting
period.

Individuals that have left their nests occasionally occupy other individuals’
nests, thereby confirming the experimental results demonstrating that site tenacity
was not confined to individuals’ own nests. I have recorded two instances in
which marked individuals have occupied nests earlier abandoned by other
individuals. I also have one record of a displaced M. vatia occupying a X.
emertoni nest.

These observations support the experiments, suggesting that displaced spiders
sometimes find their way back to their nests, but that any such ability results
from a general response that may produce an inappropriate relocation. Since the
spiders are usually sedentary at this time and do not nest at high densities, most
reoccupations are likely to occur on their own nest.

The decrease in acceptance of nests over time by the bagged individuals closely
matched the disappearance of undisturbed spiders from their nests (Morse 1987).
This result suggests that the removed individuals’ behavioral patterns are relevant
to those of undisturbed guarding spiders. The similarity in responses of
individuals periodically returned to their nests and those only returned a single

MORSE— CRAB SPIDER NEST ACCEPTANCE

57

time suggests that the tendency to reoccupy a nest is not related to time away
from a nest, but to the actual condition of the spider.

ACKNOWLEDGMENTS

I thank J. P. Mailman, C. S. Hieber, R. A. Jackson, G. Rosenqvist, and J. K.
Waage for comments on the manuscript. G, B. Edwards kindly identified the
jumping spider. E. B. Noyce and the National Audubon Society generously
permitted use of their properties. S. Chang, J. M. Cleaveland, K. Hayford, A.
Kahn, M. Memmott, E. K. Morse, J. R. Morse, W. P. Morse, S. Schrag, and M.
Spitzer assisted with the experiments. This work was supported by the National
Science Foundation (BSR8L18105-AO1 and BSR85-16279).

LITERATURE CITED

Morse, D. H. 1985. Nests and nest-site selection of the crab spider Misumena vatia (Araneae,
Thomisidae) on milkweed. J. Arachnol., 13:383-390.

Morse. D. H. 1987. Attendance patterns, prey capture, changes in mass, and survival of crab spiders
Misumena vatia (Araneae, Thomisidae) guarding their nests. J. Arachnol., 15:193-204.

Morse, D. H. Nesting success of the crab spider Misumena vatia and its relationship to earlier patch-
choice decisions. Ecology (in press).

Stearns, S. C. 1976. Life history tactics: a review of the ideas. Quart. Rev. Biol., 51:3-47.

Sterns, S. C. 1977. The evolution of life-history traits: a critique of the theory and a review of the
data. Annu. Rev. Ecol. Syst., 8:145-171.

Tinbergen, N. 1951. The Study of Instinct. Clarendon Press, Oxford.

Trivers, R. L. 1972. Parental investment and sexual selection. Pp. 136-179, In Sexual selection and the
descent of man, 1871-1971. (B. Campbell, ed.). Aldine, Chicago.

Trivers, R. L. 1974. Parent-offspring conflict. American Zook, 14:249-264.

Wilbur, H. M. 1976. Life history evolution in seven milkweeds of the genus Asclepias. J. EcoL,
64:223-240.

Wilson, E. O. 1975. Sociobiology. Harvard University Press, Cambridge, MA.

Wise, D. H. and P. R. Reillo, 1985. Frequencies of color morphs in four populations of Enoplognatha
ovata (Araneae: Theridiidae) in eastern North America. Psyche, 92:135-144.

Manuscript received June 1988, revised September 1988.

/>;l' i' ■' =

■ . '•»'• ■■ *

' r

-. :.V-

■ ■/ '■ ■

:,4

/. -f} ;3f>.a!{!r- ■■ - ■■■'

‘;.</ ■

i-A

: ,.*i

H/'

- -^.’v if '

■k.

' ' ^ • .••* ■ • :. fy ’

-iU .A . . . ■

•/ ' ;■ r ~,*'t;} >v ^ K* .

• ^ • 'v ' Vi^ r ' y '.V . . ■ . /

^ '' ' ;'■ ; . •■ ’■ :.

. . y4'.-^;*V,^ j ,:/v' A-: ‘■' •; '

>'..> vV' '-’i, .;t^-* '-i;.''', ■*;'•> • • '

.■■■ »•

'/''•• '"it.;

■v '

' 'V i? :v. ^ ■'. ,-^n’ » u^ ■

■ K -i

f-^' 'Hy?.

■ ■■

&‘:

‘:x>

^ ;

4 'it;

' ‘v

■t''-. 'Ov.fifc?

.-'M-

r

'■'■'>-'y:' ,,i .3-^ _ -'.^.‘v ^ v '■

;• t tr" ^v* 1 fc/iAj* .}' .

A'.' ^ ;.

' ..j. - i ■ '■ <’i<i ‘‘

'J'i M.i

Coyle, F. A. and T. E. Meigs. 1989. Two new species of kleptoparasitic Mysmenopsis (Araneae,
Mysmenidae) from Jamaica. J. ArachnoL, 17:59-70.

TWO NEW SPECIES OF KLEPTOPARASITIC MYSMENOPSIS
(ARANEAE, MYSMENIDAE) FROM JAMAICA

Frederick A. Coyle and Thomas E, Meigs

Department of Biology
Western Carolina University
Cullowhee, North Carolina 28723 USA

ABSTRACT

Two new species of kleptoparasitic Mysmenopsis spiders {M. monticola and M. furtiva) from
Jamaica are described; diagnoses and natural history data are also provided. These sister species
appear to have coevolved with their respective host populations of Ischnothele, also each other’s
closest relatives. The probable close relationship of this Mysmenopsis species pair to M. tibialis
(Bryant) is discussed.

INTRODUCTION

As defined by Platnick and Shadab (1978), Mysmenopsis includes 18 known
species of tiny tropical and subtropical American mysmenid spiders with cusps
attached distally to the male palpal tibia. These spiders are of special interest
because at least some of them live as kleptoparasites (food stealers) in the
funnelwebs of diplurid spiders (Platnick and Shadab 1978; Vollrath 1978). In this
paper we describe two new Mysmenopsis species, which the first author, in the
course of research on the systematics of ischnotheline diplurids, recently
discovered living in Ischnothele webs in Jamaica. We also discuss the
relationships of these new species and include observations on their natural
history.

RELATIONSHIPS

The large number of character states shared by these two Jamaican
Mysmenopsis species, including many similarities in male and female genital
characters, clearly indicates that these are sister species. Several female character
states, in particular, are probably synapomorphic for these two species: 1) a pair
of spherical spermathecal chambers on each side of the epigynum (Figs. 16, 31),
2) comma-shaped posterior spermathecal lobes (Figs. 16, 31), 3) epigynal apex
somewhat keeLshaped dorsally (Figs. 16, 31), 4) a rough retrolateral protuberance
at distal end of femur I (Figs. 11, 27, 28), and 5) the absence of spines on
metatarsus L It is important to point out that each of these species was found
living with a different member of a pair of undescribed allopatric Ischnothele
morphs which also appear to be sister species (differing in color, habitat, and very
few morphological traits) (Coyle in preparation). This appears to be the first clear

60

THE JOURNAL OF ARACHNOLOGY

evidence for the kind of host-symbiont cospeciation process which Platnick and
Shadab (1978) suggested might play a role in Mysmenopsis evolution.

Of all the previously described species of Mysmenopsis, M. tibialis (Bryant) is
most similar to, and consequently seems most closely related to, these Jamaican
species; however we should point out that our failure to examine specimens of
any Mysmenopsis species other than M. tibialis [we relied upon Muller (1987),
Platnick and Shadab (1978), and earlier descriptions] and the fact that males of
some Mysmenopsis species are unknown make this hypothesis especially tentative.
The following character states appear to be synapomorphies uniting the Jamaican
species and M. tibialis [the unknown males of M. wygodzinskyi and M.
schlingeri, which in Platnick and Shadab ’s (1978) cladogram form a trichotomy
with M. tibialis, may be found to share some of these apomorphies]: 1) epigynum
on end of pleated lobe (Figs. 5, 6, 17, 18, 32, 33), 2) embolar base sharply
delimited from bulb (Figs. 9, 25), 3) male palpal patella with distal retrolateral
keel (Figs. 9, 10, 25, 26), 4) sperm duct of palpal organ with a distinctive pattern
of loops (Figs. 9, 25), and 5) cusps on male palpal tibia greatly reduced or lost.

The examination of M. tibialis types revealed that the palpal sperm duct is
looped as in the Jamaican species (Figs. 9, 25), not as shown in Platnick and
Shadab’s (1978) fig. 59 of a paratype palp. Also, we could not see (using the
dissecting microscope at 100 and 200X magnification) the small cusps illustrated
by Platnick and Shadab (1978, fig. 59) on the inside of the tibial ledge. We
suggest that the absence of tibial cusps in the Jamaican species is not
plesiomorphic but the result of a secondary loss, and that the very small cusps of
M. tibialis (if present) represent a stage in that loss process. We have redrawn the
epigynum of M. tibialis (Figs. 5-7) to reveal more clearly than in the illustrations
of Bryant (1940) and Platnick and Shadab (1978) some of the character states
which suggest a close relationship to the Jamaican species.

Our hypothesis that M. tibialis is the sister species of the two Jamaican species
is in conflict with synapomorphies 6 and 7 in Platnick and Shadab’s (1978)
cladogram. If their interpretation of those two character states is valid and if our
hypothesis is correct, it is necessary to postulate that the anterior spermathecal
ducts and female metatarsus I spines were reduced or lost in the ancestor of the
Jamaican species.

Platnick and Shadab (1978) suggested that M. palpalis, which they placed in a
clade with M. tibialis and two other species, may be misplaced and belong instead
to a clade of six species which includes M. cidrelicola, because, of the five
Mysmenopsis species known to be kleptoparasites, M. palpalis was the only one
not in the M. cidrelicola clade. Given the evidence that the two new
kleptoparasitic species are probably members of the M. tibialis clade, it is no
longer necessary to suggest on the basis of kleptoparasitic habits that M. palpalis
is misplaced (although it may eventually prove to be misplaced on the basis of
other characters). Based upon the kleptoparasitic nature of the Jamaican species,
we predict that M. tibialis, and perhaps M. schlingeri and M. wygodzinskyi, will
prove to be kleptoparasites.

METHODS

The quantitative characters used in this study are abbreviated and defined as
follows: BL, total body length; CL, carapace length; CW, carapace width; SL,
sternum length; SW, sternum width; IFL, ITL, IML, and ITarL, lengths of leg I

COYLE & MEIGS— TWO NEW MYSMENOPSIS FROM JAMAICA

61

articles; ITX, distance along longitudinal axis of male tibia I from perpendicular
line through distal edge of tibial spur tubercle base to perpendicular line through
proximal point of articulation; IFT, maximum diameter of male femur I; ITT,
diameter of male tibia I at distal edge of tibial spur tubercle base; ITS, distance
from tip of male tibial spur to distal edge of its tubercle base; PTW, maximum
width of male palpal tibia perpendicular to longitudinal axis of cymbium; EBW,
maximum width of embolus base; DTA, distance along longitudinal axis of
female femur I from perpendicular line through ventral tubercle apex to proximal
end of femur; FDT, maximum diameter of female femur I including ventral
tubercle; HFT, height of female femur I tubercle (distance from tubercle tip to
point of most abrupt slope change at junction of tubercle and edge of femur); EL,
distance from posterior (distal) tip of epigynum to anterior edge of spermatheca
chamber (Fig. 33); EPL, distance from anterior edge of spermatheca chamber to
junction of dorsoposterior surface of epigynum with ventral surface of abdomen
(Fig. 33); EW, width of epigynum at anterior (proximal) end of sclerotized plate
(Fig. 17).

BL was measured in side view. All carapace and sternum measurements were
performed from a ventral view with the lateral borders of the sternum in the
horizontal plane. CW endpoints were at the intersections of the carapace edge
and the retrolateral surface of legs II. The length of each leg I article was
measured in retrolateral view and equals the distance from the proximal point of
articulation to the most distal point of the article. IFT and PTW were measured
from a retrolateral view. ITX was measured from an approximately retrolateral
view with the tibial spur in the horizontal plane. ITT and ITS were measured
from an approximately prolateral view with the tibial spur in the horizontal
plane. DTA, FDT, and HFT were measured from a view between ventral and
retrolateral with the ventral tubercle in the horizontal plane so that the HFT
value was the maximum possible. IFT, ITT, FDT, and HFT were measured
perpendicular to the longitudinal axes of their respective articles. EL and EPL
were measured from a side view (left) of the abdomen; EW was measured from
an approximately ventral view with the epigynum in the horizontal plane. All
appendage measurements were recorded from the left appendage, unless it was
damaged, missing, or not fully regenerated (in which case the right appendage
was measured).

Measurements were performed with a Wild M-5 stereomicroscope with 20X
ocular lenses and an eyepiece micrometer scale. BL measurements were performed
at SOX and are accurate to 0.018 mm, and all other measurements were
performed at lOOX and are accurate to 0.009 mm. All measurements are given in
millimeters.

Spermathecae were cleared in 85 percent lactic acid, viewed at 400X through a
compound light microscope, and drawn with the aid of a drawing tube.

Each species description is a composite of all the adult specimens examined;
these sample sizes are given in Tables 1 and 2. The quantitative character values
recorded in these tables are an integral part of each description.

Mysmenopsis monticola, new species

Figs. L3, 8-23, 39-42

Types. — Male holotype and six male and 13 female paratypes collected in
Ischnothele webs on roadbanks in humid mountain forest along road between

62

THE JOURNAL OF ARACHNOLOGY

Figures 1-7. — Mysmenopsis spp.: 1-3, M. monticola paratypes; 1, female carapace, dorsal; 2, female
body, lateral; 3, male abdomen, dorsal; 4, M. furtiva paratype male abdomen, dorsal; 5-7, M. tibialis
paratype epigynum; 5, ventral; 6, lateral; 7, cleared, dorsal. Scale lines: 0.5 for Figs. 1-4; 0.25 for Figs.
5, 6; 0.1 for Fig. 7.

Newcastle (3800 ft elevation) and Hardwar Gap (4200 ft elevation), St. Andrews
Parish, Jamaica (8 April 1988; F. Coyle, R. Bennett, and A. Robinson), deposited
in the American Museum of Natural History.

Etymology. — The specific name refers to the montane habitat of this species.
Diagnosis. — Males of M. monticola can be distinguished from those of M.
furtiva by the following differences: 1) The embolus base is wider (Fig. 9) [EBW

COYLE & MEIGS— TWO NEW MYSMENOPSIS FROM JAMAICA

63

= 0.056; EBW(100)/PTW = 25-30, 26.8 ± 1.60] than in M. furtiva (Fig. 25)
[EBW == 0.037; EBW (100/ PTW = 22-24, 22.7]. 2) The embolus tapers more
abruptly above its base in retrolateral view (Fig, 9) and is more evenly curved in
ventral view (Fig. 10) than in M. furtiva (Figs. 25, 26). 3) The ventral ledge on
the palpal tibia forms a proportionally larger lobe (Figs. 9, 10) than in M. furtiva
(Figs. 25, 26). 4) The tibia I clasping spur (including the tubercle from which it
arises) is longer (Figs. 8, 41) [ITS = 0.194-0.241, 0.213 + 0.015; ITS(100)/ITT =
169-192, 180.9 ± 7.59] and more distal [ITX(IOO) / ITL = 84-89, 85.8 ± 1.59]
than in M. furtiva (Figs. 24, 41) [ITS = 0.157-0.176, 0.167; ITS(IOO)/ ITT = 150-
158, 154.3; ITX(100)/ITL = 80-83, 81.0].

Females of M. monticola can be distinguished from those of M. furtiva by the
following differences: 1) The femur I ventral tubercle (Figs. 11-14, 39, 40) is larger
[HFT = 0.046-0.083, 0.055 + 0.012; HFT (100)/ IFL = 5. 1-8.0, 6.38 + LOO], its
proximal slope is much less steep, and it is more prolateral and distal [DTA(IOO)/
IFL = 84-90, 86.8 ± 1.61] than in M. furtiva (Figs. 27-30, 39, 40) [HFT = 0.019-
0.037, 0.026 ± 0.009; HFT(100)/IFL = 2.7-4.4, 3.42 ± 0.75; DTA(100)/IFL =
75-77, 75.9 + 0.81]. 2) The rough swelling on the retrolateral side of the distal
end of femur I (Fig. 11) is much less prominent than the homologous tubercle
(Figs. 27, 28) in M. furtiva. 3) The epigynum is at the end of a pleated lobe which
is usually broader and shorter (Figs. 17, 18) (EPL = 0.009-0.074, 0.031 ± 0.023)
than in M. furtiva (Figs. 32, 33) (EPL = 0.065-0.278, 0.176 + 0.071). 4) The
epigynum has a pronounced distoventral expansion (Figs. 18-23) which is lacking
in M. furtiva (Figs. 33-38). 5) The epigynal hairs extend beyond the end of the
epigynum more than one-third of the length of the epigynum (Figs. 16-18) instead
of much less than one-third the epigynal length (Figs. 31-33). 6) The lip of the
epigynal opening is broad and well-sclerotized (Fig. 16), not U-shaped and very
weakly sclerotized as in M. furtiva (Fig. 31).

Both sexes of M. monticola exhibit on the abdominal dorsum (Fig. 3) a light
grey H-shaped pigment pattern that is different from the pattern of M. furtiva
(Fig. 4) (see description).

The following characters best separate M. monticola and M., furtiva from their

close relative, the Cuban species, M. tibialis, (see Platnick and Shadab 1978, figs.
59-64): 1) No apophysis arises from the embolus base (Figs. 9, 25) as in M.
tibialis. 2) The ventral lobe-like ledge on the male palpal tibia (Figs. 9, 10, 25, 26)
is not present in M. tibialis. 3) The male palpal patellar keel is broader, thinner,
and sharper (Figs. 9, 25) than in M. tibialis. 4) The male tibia I spur is more
sharply bent near its distal end and its tubercle is shorter (Figs. 8, 24) than in M.
tibialis. 5) Female tibia I (Fig. 15) lacks the ventral tubercle and female
metatarsus I (Fig. 15) lacks the spine row present in M. tibialis. 6) The female
femur I ventral tubercle and distal retrolateral protuberance (Figs. 11, 27) are not
present in M. tibialis. 1) Several differences exist in the form of the epigynum and
spermathecae (Figs. 5-7, 16-18, 31-33).

Males. — Table 1. Figs. 3, 8-10, 41. Palpal patella with sharp distal keel on
retrolateral surface. Palpal tibia with ventral ledge produced into rather large
lobe; no cusps associated with this ledge. Embolus with relatively broad base;
narrows abruptly above base; smoothly curved in S-shape, ending at distal edge
of cymbium. Tibia I clasping spur attached to moderately short tubercle and bent
distally. Legs with dark patches on pale tan background; coxae mostly
unpigmented; femur I mostly dark, rest of femora with dark spot in middle and

64

THE JOURNAL OF ARACHNOLOGY

Figures 8-23. — Mysmenopsis moniicoia: 8-10, holotype male; 8, tibia and metatarsus I, approx.

prolateral; 9, 10, palp; 9, retrolateral; 10, ventral; 11-23, paratype females; 11, femur I, ventral; 12-14,
femur I ventral tubercle of three specimens, view between ventral and retrolateral; 15, tibia and
metatarsus I, prolateral; 16-23, epigyna; 16, cleared, dorsal; 17, ventral; 18-23, lateral view of six
specimens. Scale lines: 0.2 for Figs. 9, 10, 12-14; 0.5 for Figs. 8, II, 15; 0.1 for Fig. 16; 0.25 for Figs.
17-23.

at both ends; patellae with dark spot; tibiae with dark spot in middle and distally;
metatarsi with distal dark spot; tarsi unpigmented. Carapace and sternum dark
grey. Abdomen with many areas of dark grey or black separated by areas of light

COYLE & MEIGS— TWO NEW MYSMENOPSIS FROM JAMAICA

65

Table I. — Quantitative character values for Mysmenopsis males. Character abbreviations are defined
in the methods section of the text. All measurements given in millimeters. Range, mean, and standard
deviation given for samples larger than four.

monticola

(A=7)

furtiva

(A-3)

monticola

holotype

furtiva

holotype

BL

1.39-1.72 (1.53+0.12)

1.33-1.55 (1.43)

1.61

1.41

CL

0.71-0.81 (0.748+0.038)

0.66-0.73 (0.688)

0.79

0.68

CW

0.59-0.70 (0.643+0.041)

0.56-0.62 (0.586)

0.68

0.58

SL

0.46-0.53 (0.486+0.029)

0.43-0.46 (0.441)

0.52

0.43

SW

0.44-0.51 (0.468+0.029)

0.42-0.45 (0.432)

0.51

0.43

IFL

0.81-1.06 (0.926+0.087)

0.76-0.84 (0.799)

1.06

0.80

ITL

0.69-0.88 (0.766+0.070)

0.64-0.70 (0.666)

0.88

0.66

IML

0.38-0.44 (0.404+0.021)

0.33-0.35 (0.342)

0.43

0.33

ITarL

0.42-0.45 (0.432+0.015)

0.34-0.35 (0.348)

0.45

0.34

ITX

0.57-0.75 (0.658+0.063)

0.51-0.56 (0.540)

0.75

0.55

I FT

0.185-0.305 (0.251+0.041)

0.222-0.259 (0.237)

0.296

0.231

ITT

0.111-0.129 (0.118+0.007)

0.102-0.111 (0.108)

0.130

0.111

ITS

0.194-0.241 (0.213+0.015)

0.157-0.176 (0.167)

0.241

0.167

PTW

0.185-0.222 (0.207+0.012)

0.157-0.167 (0.163)

0.213

0.167

EBW

0.056 (0.056+0.0)

0.037 (0.037)

0.055

0.037

ITX(100)/ITL

84-89 (85.8+1.5)

80-83 (81.0)

85

83

ITS(I00)/ITT

169-192 (180.9+7.5)

150-158 (154.3)

186

150

EBW(I00)/PTW

25-30 (26.8+1.6)

22-24 (22.7)

26

22

grey with white spots; dorsally, two prominent, longitudinal, winding light grey
bands separated by dark central area except for one pair of broad median lobes
of light grey that almost connect at median line to produce light grey “H”
pattern.

Females. — Table 2. Figs. 1, 2, 11-23, 39, 40. Femur I with prominent tubercle
on prolateral aspect of ventral surface near distal end; tubercle with gently
sloping proximal face, steep distal face, and rounded apex; rough-surfaced
retrolateral swelling at distal end of femur. Tibia I cylindrical. Metatarsus I
without spines. Epigynum at end of short, broad, unsclerotized, pleated lobe; tip
with rough rounded distoventral prominence and rather sharp and keel-like
distodorsal prominence; hairs extend well beyond tip; spermathecae consist of
pair of oval chambers with curved posterior median lobe on each side, and broad,
well-sclerotized lip bordering opening on ventral surface of epigynum. Pigmenta-
tion similar to that of males.

Variation. — We found relatively little variation in any of the characters in this
population sample except for EPL (Table 2). The extremely wide range of
epigynum lobe lengths may be due to pronounced developmental plasticity and /
or use-induced changes during mating or oviposition. The pleated nature of the
unsclerotized lobe suggests that it may be lengthened during these reproductive
activities. Despite the wide range of EPL variation within both the M. monticola
and M. furtiva samples, there is surprisingly little overlap between these samples.
Variation in epigynum shape and femur I tubercle shape is illustrated by Figs. 18-
23 and 12-14.

Natural history. — Ischnothele host webs were concentrated in or near humid
forest on roadbanks ranging from low pebbly soil banks to tall rock outcrops and
from heavily shaded and moist to more exposed and drier. The host population
was quite dense; in two spots there were approximately 50 webs in a 6 m long by

66

THE JOURNAL OF ARACHNOLOGY

Table 2. — Qualitative character values for Mysmenopsis females. Character abbreviations are
defined in the methods section of the text. All measurements given in millimeters. Range, mean, and
standard deviation given.

monticola

(A=13)

furtiva

(iV=6)

BL

1.50-2.07 (I.67±0.13)

1.41-1.65 (1.51+0.08)

CL

0.73-0.87 (0.798+0.044)

0.65-0.79 (0.709+0.062)

CW

0.60-0.75 (0.660+0.048)

0.52-0.63 (0.569+0.047)

SL

0.47-0.57 (0.517+0.032)

0.41-0.50 (0.449+0.038)

sw

0.41-0.54 (0.467+0.039)

0.38-0.47 (0.419+0.037)

IFL

0.73-1.04 (0.867+0.090

0.65-0.89 (0.749+0.100)

ITL

0.55-0.78 (0.645+0.068)

0.49-0.68 (0.567+0.079)

IML

0.41-0.55 (0.472+0.039)

0.37-0.47 (0.409+0.044)

ITarL

0.42-0.47 (0.438+0.017)

0.34-0.40 (0.367+0.026)

DTA

0.63-0.92 (0.754+0.084)

0.49-0.68 (0.569+0.077)

FDT

0.194-0.287 (0.236+0.030)

0.167-0.231 (0.197+0.031)

HFT

0.046-0.083 (0.055+0.012)

0.019-0.037 (0.026+0.009)

EL

0.102-0.139 (0.117+0.012)

0.093-0.129 (0.108+0.014)

EPL

0.009-0.074 (0.031+0.023)

0.065-0.278 (0.176+0.071)

EW

0.111-0.148 (0.133+0.011)

0.102-0.120 (0.109+0.007)

DTA(100)/IFL

84-90 (86.8+1.6)

75-77 (75.9+0.8)

HFT(I00)/IFL

5. 1-8.0 (6.38+1.00)

2.7^4.4 (3.42+0.75)

1.5 m tall section of roadbank. The retreat tubes penetrated into rock crevises,
soil cavities, moss, and leaf litter. The retreats opened out onto exposed capture
webs composed of one or two roughly horizontal sheets and other non-horizontal
sheets and strands anchored to surrounding substrate and plant surfaces. The
average capture web area was about 400 square cm, but the largest webs covered
about 1000 square cm.

M. monticola were observed in many of the larger Ischnothele webs. Typically
three to six mysmenids were seen per inhabited web, but as many as twelve were
counted in one web. They were always observed in the capture portion of the
host web near the retreat mouth. Adults of both sexes were found together in
some of the host webs. Several females were carrying (it was not clear which
appendages were being used) very thin-walled (nearly transparent) and roughly
spherical egg sacs about equal to their body volume. Three of these egg sacs were
collected and examined. One was 1.18 by 1.54 mm in size and contained nine
eggs ranging in size from 0.41-0.46 mm minimum diameter to 0.51-0.55 mm
maximum diameter. The other two egg sacs each contained four unpigmented
pre-emergent spiderlings, but these sacs had been torn open (probably during or
shortly after capture) and therefore may not have contained their full complement
of spiderlings.

Distribution. — Known only from the type locality in the mountains of eastern
Jamaica (Fig. 42).

Material examined. — Only the type specimens.

Mysmenopsis furtiva, new species
Figs. 4, 24-42

Types. — Male holotype and one female and two male paratypes collected in
Ischnothele webs on hillside in dry limestone forest (approximately 300 m

COYLE & MEIGS— TWO NEW MYSMENOPSfS FROM JAMAICA

67

Figures 24-38. — Mysmenopsis furtiva: 24-26, holotype male; 24, tibia and metatarsus I, approx,
prolateral; 25, 26, palp; 25, retrolateral; 26, ventral; 27-38, females; 27-29, 14 mi E Kingston; 27, 28,
femur I, ventral; 29, 30, femur I ventral tubercle, view between ventral and retrolateral; 30, paratype;
31-38, epigyna; 31-33, paratype; 31, cleared, dorsal; 32, ventral; 33, lateral; 34-38, lateral view of five
specimens, 14 mi E Kingston. Scale lines: 0.2 for Figs. 25, 26, 28-30; 0.5 for Figs. 24, 27; 0,1 for Fig.
31; 0.25 for Figs. 32-38.

elevation) 14-15 miles east of Kingston, St. Thomas Parish, Jamaica (10 April
1988; F. Coyle, R. Bennett, B. Freeman, and A. Robinson), deposited in the
American Museum of Natural History.

Etymology. — The specific name refers to the stealthy, furtive nature by which
these kleptoparasites are thought to procure food.

Diagnosis. — To distinguish M. furtiva from its close relatives, M. monticola
and Af. tibialis, refer to the diagnosis for M monticola.

68

THE JOURNAL OF ARACHNOLOGY

.08

•

.9-

•

.06 ■

X .04 ■

^ »■ "•> • • •

1^

via

V

.02 •

.5-

5

39

.8

IFL

1.0

6

CO .20

41

.8

IFL

10

40

Figures 39-41. — Scattergrams for Mysmenopsis monticola (circles) and M. furtiva (squares) with
regression lines (values in mm); 39, 40 females; 39, HFT vs. IFL {M. monticola regression: y = 0.089x
— 0.022, r — 0.673; M. furtiva regression: y = 0.089x — 0.041, r = 0.979); 40, DTA vs. IFL (M.
monticola regression: y = 0.923x — 0.047, r = 0.989; M. furtiva regression: y = 0.767x — 0.006, r =
0.997); 41, males, ITS vs. ITT {M. monticola regression: y = I.75x + 0.007, r = 0.810; M. furtiva
regression: y = L5x + 0.005, r = 0.866). Fig. 42. — Distribution of Jamaican Mysmenopsis species.

Males. — Table 1. Figs. 4, 24-26, 41. Palpal patella with sharp distal keel on
retrolateral surface. Palpal tibia with ventral ledge produced into small lobe; no
cusps associated with this ledge. Embolus with relatively narrow base; tapers
gradually above base; unevenly curved (crooked) in ventral view; ending at distal
edge of cymbium. Tibia I clasping spur attached to moderately short tubercle and
bent distally. Legs with moderately dark patches on pale tan background; coxae
mostly unpigmented; femur I with faint dark patches at proximal end, middle,

COYLE & MEIGS— TWO NEW MYSMENOPSIS FROM JAMAICA

69

and near distal end; rest of femora with darker patches at each end and in
middle; patellae with distal dark spot; tibiae with distal dark area and small dark
spot in middle; metatarsi with distal dark spot; tarsi unpigmented. Overall, leg I
lighter than other legs. Carapace and sternum moderately dark grey-brown.
Abdomen with many areas of dark grey or black separated by areas of light grey
with white spots; dorsally, two prominent, longitudinal, winding, light grey bands
separated by dark central area except for two pairs of narrow median lobes of
light grey which come moderately close to connecting at median line.

Females. — Table 2. Figs. 27-40. Distal half of femur I with small yet prominent
ventral tubercle with steeply sloping proximal and distal faces, and angular apex;
small, rough-surfaced tubercle on retrolateral surface at distal end of femur. Tibia
I cylindrical. Metatarsus I without spines. Epigynum at end of long, narrow,
unsclerotized, pleated lobe; tip with gentle distoventral slope and blunt, keel-like
distodorsal prominence; hairs extend only slightly beyond tip; spermathecae
consist of pair of oval chambers with a curved posterior median lobe on each
side, and U-shaped, weakly sclerotized lip bordering opening on ventral surface of
epigynum. Pigmentation similar to that of males.

Variation. — See the discussion of variation in M. monticola regarding the
especially wide range of EPL variation in M. furtiva. Variation in epigynum
shape and femur I tubercle shape is illustrated by Figs. 33-38 and 28-30.

Natural history. — The M. furtiva populations were found in much drier
habitats than that occupied by M. monticola and its host. The Fort Clarence —
Hellshire Hills population lived in a hot, dry, cactus thorn scrub community on
limestone substrate with little soil. Ischnothele host webs at this site were usually
found at the bases of rocks where leaf litter accumulates under the scattered small
trees, and they were similar in size and shape to those occupied by M. monticola.
The other M. furtiva population inhabited a dry forest community on a rocky
(limestone) hillside. Here the host webs were most often found at the bases of
rocks and exposed roots, their retreats penetrated the loose limestone pebble
substrate, and they were much more abundant than in the cactus thorn scrub. At
both sites individuals of M. furtiva were observed in several webs, always in the
capture web near the retreat mouth. One female was observed carrying an egg
sac.

Distribution. — Known from two localities along the south coast of eastern
Jamaica (Fig. 42).

Material examined. — The type specimens and the following: JAMAICA: ST CATHERINE
PARISH; Fort Clarence and Hellshire Hills near Seafort, 10-50 m elevation, cactus thorn scrub, 9
April 1988 (F. Coyle, R. Bennett, B. Freeman, and A. Robinson), 5 females, several juvs. (AMNH).

ACKNOWLEDGMENTS

We are grateful to Mr. Robert Bennett, Mr. Abraham Robinson, and Dr.
Brian Freeman for their help in collecting Jamaican Mysmenopsis, to Dr. Herbert
Levi and Ms. Laura Liebensperger for the loan of M. tibialis types and a copy of
Bryant’s description, and to Drs. Norman Platnick and Charles Griswold for
reviewing a draft of this paper. This study was supported by National Science
Foundation Grant BSR-8700298.

70

THE JOURNAL OF ARACHNOLOGY

LITERATURE CITED

Bryant, E. B. 1940. Cuban spiders in the Museum of Comparative Zoology. Bull. Mus. Comp. Zool.,
86:350-353.

Muller, H.-G. 1987. Spiders from Colombia V. A new Mysmenopsis from the Cienaga Grande de
Santa Marta, northern Colombia (Araneae, Mysmenidae). Bull. British Arachnol. Soc., 7(6): 185.
Platnick, N. I. and M. U. Shadab. 1978. A review of the spider genus Mysmenopsis (Araneae,
Mysmenidae). American Mus. Novitates, (2661): 1-22.

Vollrath, F. 1978. A close relationship between two spiders (Arachnida, Araneidae): Curimagua
bayano synecious on a Diplura species. Psyche, 85:347-353.

Manuscript received August 1988, revised September 1988.

Coddington, J. A. 1989. Spinneret silk spigot morphology: evidence for the monophyly of orbweaving
spiders, Cyrtophorinae (Araneae), and the group Theridiidae plus Nesticidae. J. ArachnoL, 17:71-

95.

SPINNERET SILK SPIGOT MORPHOLOGY:

EVIDENCE FOR THE MONOPHYLY OF ORBWEAVING SPIDERS,
CYRTOPHORINAE (ARANEIDAE),

AND THE GROUP THERIDIIDAE PLUS NESTICIDAE

Jonathan A. Coddington

Department of Entomology
National Museum of Natural History
Smithsonian Institution

Washington, DC 20560 USA

ABSTRACT

Labelled scanning electron micrographs comprehensively illustrating the spinning fields of Deinopis
(Deinopidae), Octonoba (Uloboridae), Araneus (Araneidae), Leucauge (Tetragnathidae), Latrodectus
and Theridula (Theridiidae), Gaucelmus (Nesticidae), and Frontinella (Linyphiidae) are presented for
the first time. Evidence from scanning electron micrographs supports the monophyly of orbweavers
(Orbiculariae = Uloboridae, Deinopidae, Araneoidea), the araneid subfamily Cyrtophorinae and the
close relationship between Nesticidae and Theridiidae. This evidence is presented in the context of
guidelines for the logical taxonomic interpretation of spider spigot morphology. This morphological
system, including gross and detailed morphology, location and number of spigots, and serial
homology relationships, may be one of the most complex, yet under-utilized, taxonomic character
systems in spiders.

INTRODUCTION

Spinnerets both define and epitomize spiders. The history of studies on spider
silk glands has been reviewed by Kovoor (1977b). Despite her elegant studies
(e.g., Kovoor 1972, 1977a, 1977c, 1988; Kovoor and Lopez 1982; Kovoor and
Peters 1988), filled with interesting facts about peculiar spinneret morphologies,
systematists rarely report on this diverse character system in their phylogenetic
studies. Several explanations for this neglect suggest themselves, such as belief
that histological evidence on glands is essential for interpretation of the
morphology, or that one must fix living material to obtain satisfactory results.
Perhaps the chief difficulty is that a detailed histological study is so time
consuming that a large diversity of species within a taxon cannot easily be
surveyed. Variation thus remains undocumented, and taxonomists remain
uncertain about the constancy of a feature, and the pattern of its distribution.
However, on close inspection, none of these objections are insuperable. This
paper outlines how spigot morphology can be studied using ordinary museum
material. It presents some of the more interesting results so far, and summarizes a
conceptual framework for the interpretation of the data that, I hope, will
encourage the use of spigot morphology in spider systematics.

72

THE JOURNAL OF ARACHNOLOGY

In this study I present preliminary results from a larger study on silk spigot
diversity in spiders in general. The taxa included herein have been chosen to
illustrate how silk spigot characters may bear on particular systematic problems,
and to illustrate the diversity of silk spigot morphology. The specific systematic
problems discussed are the monophyly of orbweavers, the monophyly of
Cyrtophorinae (Cyrtophora and Mecynogea, at least), and the monophyly of a
group including Theridiidae and Nesticidae. Although the evidence in all three
cases seems positive, one must be cautious until further taxa are surveyed.

I have previously reviewed the evidence for the monophyly of Araneoidea and
Deinopoidea (=Uloboridae and Deinopidae; Coddington 1986a, b). The evidence
formerly believed to support the polyphyly of orbweavers was largely due to a
confusion between symplesiomorphy and synapomorphy — the misinterpretation
of primitive features that “defined” the symplesiomorphic group “Cribellatae.”
When the features that once defined the Cribellatae are recognized as primitive,
there are no credible synapomorphies that place Deinopoidea with the rest of the
cribellates, rather than with the other orbweavers (Araneoidea). On the contrary,
most of the evidence implicates them as the sister taxon of the ecribellate
Araneoidea. However, contrasting points of view have subsequently been
expressed by Eberhard (1987), Shear (1986), Kovoor and Peters (1988) and
Tyshchenko (1986).

The monophyly of the Cyrtophorinae, on the other hand, although never
explicitly justified, has rarely been doubted (Levi 1980, 1983; Levi and
Coddington 1983). This paper offers evidence independent of genitalic morphol-
ogy and web architecture to confirm cyrtophorine monophyly.

The composition of the Theridiidae has frequently been questioned, as well as
its relationship to Nesticidae. For example, Lehtinen and Saaristo (1980) placed
the latter two families in different superfamilies, mainly because of genitalic
differences. In particular they suggested that the fourth tarsal comb of serrated
bristles common to both families was “purely adaptive.” Palp and epigynal
morphology among theridiids and nesticids obviously is diverse, but in any case
offers no evidence to ally either group more closely with other araneoid taxa than
with each other. However, the fourth tarsal comb is part of a behavior-
morphology complex that enables nesticids and theridiids to subdue their prey
with viscid sticky silk (Whitehouse 1987). This attack behavior is unique among
spiders and stands as a strong synapomorphy of theridiids and nesticids
(Coddington 1986a). This paper offers additional morphological evidence
concordant with the fourth tarsal comb and the attack behavior itself. It is
therefore additional evidence in favor of the monophyly of Nesticidae and
Theridiidae.

This study also outlines the analytical methods one can use to deduce
homologies among spigots, and thereby to use the data in phylogenetic analysis.
Basically, Remane’s criteria of homology (position, special similarity, ontogeny)
seem entirely adequate, and thus the comparative study of silk spigots can
proceed to some extent independently of other lines of evidence, such as histology
or histochemistry. This is not to say that the spigot evidence is superior to
histological evidence, but only that the morphology and exact location of
araneomorph spigots seems complex enough and consistent enough to support
generalizations. Histology is not required in order to infer spigot homologies.
Spigot morphology may even help to decide questions of homology when the

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

73

histological evidence is equivocal Indeed, when patterns in histological and
histochemical data (e.g., Kovoor 1987) are compared with well-corroborated
groups in spider phytogeny, many details of histochemical reactions, gland
ultrastructure, and gland cell type are apparently homoplasious (see Discussion).

MATERIALS AND METHODS

Spigot morphology is extremely difficult to see with light microscopy. If one
must use the light microscope, the best results are obtained with epi-illumination,
but even then distinctive details are easy to miss (e.g., Mikulska 1966, 1967;
Wasowska 1966, 1970). No doubt that helps to explain the neglect of spigot
morphology as a character system in the past. However, the increasing
availability of scanning electron microscopes (SEM) puts study of spigot
morphologies well within the grasp of most spider taxonomists.

The best preparations have the spinnerets widely spread, are clean, and can be
scanned from all angles. Several workers with whom I have spoken have had
difficulties in getting good results, usually because the spinnerets are contracted
upon themselves so that the distal articles cannot be seen, or else the spigots are
covered with debris. In this study, these obstacles were overcome with good
success by the following techniques.

Selection of material. — Obviously fresh material is best, although I have
successfully prepared 50-100 year old specimens. If live material is available, kill
the animals by direct, sudden immersion in boiling water or fixative. Under these
conditions, the spinnerets are widely spread. In the case of ordinary museum
material, the only real requirements are an intact set of spinnerets and a flexible
abdomen. The latter is important in case the spinneret tips need to be spread.
However, even hardened material can be used if one digests the spinneret group
in a trypsin solution prior to mounting. All the material here is from the USNM
collection, and voucher specimens are deposited there. Except in one case, only
adult females were used.

Forceps squeeze. — As long as the abdomen of the specimen is still flexible, the
spinnerets can be spread with forceps. This technique has been modified from one
originally suggested by Dr. J. Kovoor. One must use cross-action forceps or
otherwise be able to lock them in a closed position. One can use a rubber band
around the forceps blades, or a wire collet that can slide forward to lock the tips.
By experimenting with various angles, one can usually squeeze the abdomen
immediately above (dorsal or anterior to) the spinnerets so that all six spinnerets
spread widely. At this point, lock the forceps shut, sever the spinnerets from the
abdomen with a razor, and run the spinnerets through an alcohol series up to
anhydrous ethanol while they are still grasped by the forceps. Dehydration
stiffens the spinnerets so that one can remove the forceps and the spinnerets will
stay in a spread position. At this point, they are ready for cleaning.

If the forceps technique does not separate the spinnerets, sever the spinneret
group and digest it completely in trypsin solution. This removes all muscle tissue
and leaves only the cuticle of the spinneret group, so that it will be flexible
enough to spread the spinnerets during the mounting process.

Ultrasonic cleaning. — Immersing the spinnerets (still gripped in the forceps) in
an ultrasonic cleaner for I-IO minutes will remove debris. If the specimen is

74

THE JOURNAL OF ARACHNOLOGY

extremely fragile (e.g., Ochyroceratidae, Nesticidae, Symphytognathidae),
decrease the time. Clean spigots are distinctly visible at lOOX. Special care must
be taken to clean trypsin-digested spinnerets. Brief soaking in 10% KOH to
remove enzymatic protein is often helpful

Stub mounting. — The cleaned, spread spinneret group is now ready to be
mounted. One must use a stub so that the spinneret group can be rotated into
nearly any angle for viewing. Standard, disc-shaped SEM stubs do not work well
because the stub edge obscures the view or ruins the background contrast. I use
1/8 in. diameter steel rivets. The hemispherical head permits an unobstructed
view of the spinnerets from any angle, and one can tip the stub to a 90° position
so that the SEM background will be completely black. Spinnerets prepared by
the forceps technique can be mounted with the usual adhesives, such as silver
paint.

Trypsin-digested spinnerets that require spreading should be mounted with a
stickier adhesive, such as the gum from double stick tape. In this case, the cured
surface of the rivet anchors the outer edge of the cuticle, and by careful
additional pushing and denting of the cuticle surface, one can spread the
spinneret tips at this point.

RESULTS

Figure 1 diagrams the typical distribution of spigots in an araneoid spider.
Figures 2-41 illustrate the diversity of spigots in Deinopis (Deinopidae), Octonoba
(Uloboridae), Araneus (Araneidae), Leucauge (Tetragnathidae), Cyrtophora and
Mecynogea (Araneidae: Cyrtophorinae), Latrodectus and Theridula (Theridiidae),
Gaucelmus (Nesticidae), and Frontinella (Linyphiidae). Spigots in all figures are
labelled in accordance with Table 1. Each plate of figures is laid out the same
way. The upper left micrograph shows the left three spinnerets. Anterior is always
at the top. The upper right micrograph shows the left AES (anterior lateral
spinneret) tip; the lower left micrograph the left PMS (posterior median
spinneret) tip; and the lower right micrograph the left PLS (posterior lateral
spinneret) tip. In the three close-up micrographs anterior is usually at the top or
left, but if not, the orientation can be figured out by referring to the upper left
micrograph. Abbreviations of spinneret and spigot terminology are given in the
legend to Figs. 2-5. Spigots in Figs. 2-9 are labelled with arrows for precision,
thereafter the labels are adjacent to the spigots.

Deinopoidea. — The cribellate Deinopis and Octonoba illustrate the orb-weaver
ground plan. The piriform spinning field is more or less uniformly distributed
across the AES tip (Figs. 3, 7; PI). The major ampullate spigot(s) is (are) on the
mesal margin of the ALS (Figs. 3, 7; MAP). Like the Araneoidea, adult female
Octonoba have one functional major ampullate spigot, bordered posteriorly by a
vestigial nubbin (also diagrammed in Fig. 1), presumably the remainder of the
second MAP present in juveniles. Deinopis have multiple MAPs, a fairly rare
feature among araneomorph spiders. The cuticle sculpturing is grooved and fluted
(see also Kovoor and Peters 1988), apparently a synapomorphy for Neocribella-
tae.

The posterior median spinneret of Octonoba shows four classes of spigots.
There are three cylindrical spigots and one posterior minor ampullate spigot (Fig.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

75

PMS

Figure 1. — Generalized araneoid spinneret and spigot location.

8; CY, mAP). Around and between them is a group of many (>30) small
aciniform spigots (Fig. 8; AC), which extend forward on to the anterior wall of
the PMS spinneret. Finally, there is an anterior brush of many elongate,
annulated, paracribeliar spigots (Fig. 8; PA). The condition in Deinopis is similar,
although they seem to have many more cylindrical gland spigots than uloborids
(Fig. 4;CY).^

The posterior lateral spinneret is also complex. On the mesal basal margin in
Octonoba are six cylindrical spigots (Fig. 9; CY). On the anterolateral margin is
the pseudoflagelliform spigot (Fig. 9; PF). Distributed across the face of the PLS
are a second group of aciniform spigots (Fig. 9; AC). Figure 5 shows a Deinopis
immature female — thus she lacks all cylindrical spigots, and has only the
pseudoflagelliform and aciniform spigots (Fig. 5; PF, AC,). Adult female
Deinopis, like other deinopoids, have multiple, basal, cylindrical gland spigots on
the PLS.

Araneoidea. — The spinning fields of Araneus illustrate the rather conservative
and consistent araneoid ground plan. The cuticle sculpturing is lenticular or
squamate, rather than fluted or grooved. That feature, of course, is found in
other spider taxa than Araneoidea. As in Deinopoidea, the piriform spinning field
is uniform across the ALS tip (e.g., Figs. 11, 23, 27, 31, 35, 39; PI). This
condition has been distinctively modified in Mecynogea and Cyrtophora,
however, whose piriform field has been restricted posteriorly to a ribbon (Figs.

76

THE JOURNAL OF ARACHNOLOGY

Table 1. — Orbweaver silk spigots categorized by number (singular or multiple), the glands they

presumably serve, and position in the spinneret field.

Spigot type

Position/ Description

SINGULAR

1.

Major ampullate

A single spigot on the mesal border of the ALS. Present in most if not
all araneomorphs.

2.

Minor ampullate

A single spigot on the posterior margin of the PMS. Present in most if

not all araneomorphs.

3.

Cylindrical

In Araneoidea, 3 spigots: 2 on basal margin of PLS tip and 1 on ante-
rior margin of PMS. Number varies in other groups, but apparently

always on PMS and PLS.

4.

Flagelliform

In Araneoidea, a unique spigot between the aggregate spigots of the
PLS.

5.

Pseudoflagelliform

One spigot on the antero-lateral margin of the PLS in some cribellate

spiders; homologue of the flagelliform in araneoids?

6.

Aggregate

In Araneoidea, two similar spigots near the flagelliform spigot.

MULTIPLE

7.

Piriform

A group of small apiculate spigots on the ALS. Present in most if not
all araneomorphs; morphology but not position variable.

8.

Aciniform

The most numerous gland type; small spigots present in multiples on
PMS and PLS. Present in most if not all araneomorphs; morphology
variable.

9.

Paracribellar

A group of long, thin, often annulated spigots on the anterior PMS
margin in some cribellate spiders.

15, 19; PI). The major ampullate spigot with its vestigial partner is on the mesal
margin of the ALS in all the araneoids (Figs. 11, 15, 19, 23, etc.; MAP).

The posterior median spinneret shows three classes of spigots: the single
anterior cylindrical spigot, the single posterior minor ampullate spigot, and the
surrounding and/or intervening group of many small aciniform spigots (Figs. 12,

16, 20, 24, 28, 32, 36, 40; CY, mAP, AC, respectively). In Araneus (Fig. 12),
Mecynogea (Fig. 16), and Cyrtophora (Fig. 20), as in cribellate orbweavers, the
aciniform spigot field extends forward on the anterior wall of the spinneret. The
restriction of the PMS aciniform field to the spinneret tip, or at least its absence
from the anterior face in Leucauge (Fig. 24), and other derived araneoids such as
theridiids (Figs. 28, 32), nesticids (Fig. 36), and linyphiids (Fig. 40), is a derived
condition. Leucauge, Latrodectus, and Gaucelmus have just three aciniform
spigots, Theriduia and Frontinella (as well as other linyphiids) have only two.

In Araneoidea, the posterior lateral spinneret is the most complex. On its mesal
basal margin are two cylindrical spigots (e.g., Figs. 13, 25; CY), although their
position sometimes shifts (e.g., Frontinella, Fig. 41). On the anterolateral margin
is the flagelliform, and the two aggregate glue gland spigots (e.g.. Figs. 13, 25, 41;
FL, AG). Distributed across the PLS tip is a second group of aciniform spigots
(e.g.. Figs. 17, 21; AC). The PLS flagelliform-aggregate complex in Cyrtophora is
entirely absent, whereas it is merely reduced in Mecynogea (compare Figs. 17, 21
with “normal” spigots in Araneus, Fig. 13, or Frontinella, Fig. 41).

The theridiid and nesticid complements are also modified, but in a different
way. On the theridiid PLS, the two aggregate glands are relatively enormous, the
ectal larger than the mesal (e.g., Latrodectus, Theriduia Figs. 29, 33; AG). The
same situation occurs in nesticids, although the difference between the two

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

77

Figures 2-5. — Deinopis spinosus Marx spinnerets: 2, left spinneret group; 3, anterior lateral
spinneret, closeup; 4, posterior median spinneret, closeup (subadult female); 5, posterior lateral
spinneret, closeup. Abbreviations for Fig. 1-41: AC = aciniform; AG = aggregate; ALS = anterior
lateral spinneret; FL = flagelliform; MAP = major ampullate; mAP = minor ampullate; CY =
Cylindrical; PA = paracribellar; PF = pseudoflagelliform; PI = piriform; PMS = posterior median
spinneret; PLS = posterior lateral spinneret. Scale bars = 50 pm.

78

THE JOURNAL OF ARACHNOLOGY

Figures 6-9. — Octonoba octonarius (Muma) spinnerets: 6, left spinneret group; 7, anterior lateral
spinneret, closeup; 8, posterior median spinneret, closeup; 9, posterior lateral spinneret, closeup.
Abbreviations as in Figs. 2-5. Scale bars = 50 jxxn.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

79

Figures 10-13. — Araneus diadematus Clerck spinnerets: 10, left spinneret group; 11, anterior lateral
spinneret, closeup; 12, posterior median spinneret, closeup; 13, posterior lateral spinneret, closeup.
Abbreviations as in Figs. 2-5. Scale bars = 50 jum.

80

THE JOURNAL OF ARACHNOLOGY

Figures 14-17. — Mecynogea lemniscata (Walckenaer) spinnerets: 14, left spinneret group; 15,
anterior lateral spinneret, closeup; 16, posterior median spinneret, closeup; 17, posterior lateral
spinneret, closeup. Abbreviations as in Figs. 2-5. Scale bars = 50 ixm.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

81

aggregate spigots is not as pronounced {Gaucelmus, Fig. 37; Nesticus and
Eidmanella, not figured). In addition, the aciniform spinning fields on the PMS
are much reduced in comparison with araneids, and are limited to the posterior
and apical surface of the spinnerets.

DISCUSSION

Spigot homologies. — The above results argue that by judicious use of spigot
number, placement, appearance, and known ontogenetic patterns, one can often
work out spigot homologies without reference to histological data. A basic rule of
inference is the difference between morphological singulars and multiples, or
homologues and homonoms (Riedl 1979). Morphological singulars are unique
and can be exactly specified, such as ‘'the left third tarsus”, or, in this case, “the
basal cylindrical gland of the PLS.” On the other hand, morphological
“multiples” are present in many copies and are not individually specifiable, such
as “abdominal setae,” or, in this case, the “aciniform spigots of the PMS.” One
can make exactly specifiable homology statements about morphological singulars,
but usually one can only homologize groups or sets of morphological multiples.
In the case of spigots one has both singulars, such as major ampullates, minor
ampullates, cylindrical, pseudoflagelliforms, aggregates and flagelliforms, and
“multiples,” such as paracribellars, piriforms, and aciniforms. When multiple
spigots are reduced in number, they may be consistent enough in their
distributions to support hypotheses of individual homology (compare aciniform
spigots in Figs. 4, 8, 12 vs. Figs. 28, 32, 36, 40). In combination with their
placement, the distinction between singulars and multiples offers a way to tell the
different kinds of spigots apart.

By trading off similarities between singular spigots, groups of spigots, and
where they occur, one can devise a set of rules to guide homology statements
about spigots, at least within the orbweavers. They can be summarized as follows:

MULTIPLES

1. If the spigots are unique to the ALS and are small, numerous, and occupy
most of the ectal spinning field, they are piriform gland spigots. They are
used to attach draglines together or to the substrate. They should occur in
ail instars and both sexes.

2. If they occur both on the PMS and PLS and are small, but slightly more
elongate that the ALS piriforms, they are probably aciniform gland spigots.
They have several uses: prey wrapping, retreats, and egg sacs. They should
occur in all instars and both sexes.

3. If they occur only on the PMS and are thin, very long, and with annulate
shafts, they are probably paracribellar spigots. Their function is unclear,
although they contribute a component to the sticky thread. They should
occur in all instars and both sexes of all cribellate orbweavers, at least.

SINGULARS

4. If a set of three spigots occur as two on the araneoid PLS and one on the
anterior portion of the araneoid PMS, they are cylindrical gland spigots.
They are apparently used to produce specialized silk for egg sacs. They

82

THE JOURNAL OF ARACHNOLOGY

Figures \^-2\ . -Cyrtophora citricola (Forskal) spinnerets: 18, left spinneret group; 19, anterior
lateral spinneret, closeup; 20, posterior median spinneret, closeup; 21, posterior lateral spinneret,
closeup. Abbreviations as in Figs. 2-5. Scale bars = 50 ;um.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

83

Figures 22-25. — Leucauge venusta (Walckenaer) spinnerets: 22, left spinneret group; 23, anterior
lateral spinneret, closeup; 24, posterior median spinneret, closeup; 25, posterior lateral spinneret,
closeup. Abbreviations as in Figs. 2-5. Scale bars = 50 /am.

84

THE JOURNAL OF ARACHNOLOGY

should only occur in adult females. The situation in cribellate orbweavers is
more complex because more cylindricals are present, but their distribution is
as in araneoids.

5. If a pair (or group, Deinopidae) of spigots occur as one on the mesal
margin of the ALS and one on the posterior margin of the PMS, they are
ampullate gland spigots, usually one major and one minor. They are used
for components of the dragline and major structural lines. They should
occur in all instars and both sexes.

6. If a pair of large spigots with wide openings is juxtaposed on the
anterolateral margin of the PLS, they are aggregate gland spigots, used to
make the viscid glue of the sticky line. They should occur in juveniles and
adult females, but are supposed to be absent in all adult males (exceptions
may occur).

7. If a single large spigot is unique to the PLS anterolateral or apical margin,
it is the flagelliform or pseudoflagelliform gland spigot, used to make the
base fibers of the sticky line. Flagelliform spigots are usually between the
two aggregate spigots, but not always. Like the aggregate gland spigots,
flagelliform and pseudoflagelliform spigots should occur in juveniles and
adult females, and should be absent in adult males.

These rules were used to identify the various spigots present in the genera
illustrated here. Of these, only Araneus, Latrodectus, Cyrtophora and Mecynogea
have been studied histologically (Kovoor 1972, 1977a, 1988; Kovoor and Lopez
1982, 1988), but no SEM illustrations of them have been published. As far as I

know, only the spinnerets of Nephila among the araneoids have been

comprehensively illustrated (Kovoor 1986), although details of spigots of several
uloborids have been published (Peters and Kovoor 1980; Kovoor and Peters,
1988).

As an example one can use the above set of rules to correlate the large spigots
on Leucauge PMS with known gland types. Because they are few in number, and
are individually recognizable, they are probably examples of morphological
singulars. Cylindrical glands exit on the PMS and PLS (#4, above). Araneoid

ampullate glands exit only on the ALS and the PMS (#5, above). In Figs. 24-25,

Leucauge has three large spigots with broad bases, sharply tapering, fluted, blunt
shafts, and wide tips. The junction between the base and the shaft has a narrow
indistinct rim. One of these spigots exits on the PMS (Fig. 24, CY), and two on
the PLS (Fig. 25, CY). Therefore, they probably serve cylindrical glands.

Likewise, the major ampullate gland on the Leucauge ALS is similar to the
spigot labelled as the minor ampullate gland on the PMS (Fig. 24, mAP). One
can thus homologize spigots on orbweaving spiders, for example between
Leucauge, a tetragnathid, and Frontinella, a linyphiid, even without histological
evidence in these particular cases.

As an example of morphological multiples, piriform spigots are the only set of
multiple spigots on the araneoid ALS tip. They are easy to recognize. However,
in more distantly related groups, such as Deinopidae, two sets of multiples occur
on the ALS. Only one set is on the mesal margin, set apart from the rest of the
spinning field, and it is less numerous, and the spigots are larger. Because they
occur in the same place as major ampullate glands, one can guess that they are
indeed ampullate glands, and that Deinopidae is derived in having multiple ALS

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

85

Figures 26-29. — Latrodectus variolus (Walckenaer) spinnerets: 26, left spinneret group; 27, anterior
lateral spinneret, closeup; 28, posterior median spinneret, closeup; 29, posterior lateral spinneret,
closeup. Abbreviations as in Figs. 2-5. Scale bars = 50 nm.

86

THE JOURNAL OF ARACHNOLOGY

Figures 30-33. — Theridula opulenta (Walckenaer) spinnerets: 30, left spinneret group; 31, anterior
lateral spinneret, closeup; 32, posterior median spinneret, closeup; 33, posterior lateral spinneret,
closeup. Abbreviations as in Figs. 2-5. Scale bars = 20 pm.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

87

Figures 34-37. — Gaucelmus angustinus Keyserling spinnerets: 34, left spinneret group; 35, anterior
lateral spinneret, closeup; 36, posterior median spinneret, closeup; 37, posterior lateral spinneret,
closeup. Abbreviations as in Figs. 2-5. Scale bars = 50 jxm.

THE JOURNAL OF ARACHNOLOGY

Figures Frontinella pyramitela (Walckenaer) spinnerets: 38, left spinneret group; 39,

anterior lateral spinneret, closeup; 40, posterior median spinneret, closeup; 41, posterior lateral
spinneret, closeup. Abbreviations as in Figs. 2-5. Scale bars = 20 ^m.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

89

ampullate glands. This may be a synapomorphy for the family (it also occurs in
Menneus, pers. obs.). By position and morphology, the other group of smaller
multiples on the deinopid ALS are therefore the piriforms. By similar logic, one
can also guess that deinopids are strange in having up to 50 pairs of cylindrical
gland spigots (Fig. 4, CY). Although the histology of deinopid glands is
unknown, one may predict the following peculiarities: 5-10 pairs of major
ampullate glands, numerous cylindrical glands, and a single pair of pseudoflagelli-
form glands.

As a final example of spigot identification, the PMS of Araneus have three

kinds of spigots: one kind of multiples and two kinds of singulars. The multiple
spigots (Fig. 12; AC) are recognizably more like the multiple spigots of the PLS
(Fig. 13; AC) than they are like the multiple kind on the ALS (Fig. 11; PI).
Probably they are aciniforms. Histological data (Kovoor 1972) suggests that only
one sort of gland present in numerous copies exits on both the PMS and PLS:
the aciniform glands (#2, above). In this case histology and morphology are again
concordant. Similarly, the single posterior spigot on the PMS (Fig. 12; mAP) is
more like the single ALS spigot (Fig. 11; MAP), than it is like any single spigot
on the PLS. The single anterior spigot on the PMS is more like the two basal
spigots on the PLS than it is like other PLS or ALS spigots. Histological data
confirms that Araneoidea have a single cylindrical spigot on the PMS and two on
the PLS, and only a single minor ampullate gland on the PMS.

Interpretation of patterns. — Once over the hurdle of identifying spigots, one
can look for interesting differences among taxa. For example, deinopoid piriform
spigots have raised bases with rounded shoulders (Figs. 3, 7; PI), but araneoid
piriform spigots have sunken bases and sharp rims. I have found that deinopoids
resemble other cribellate araneomorphs, and thus the deinopoid condition is
primitive, and the araneoid condition is derived, probably a synapomorphy for
the superfamily.

Aciniform spigots usually are small, have longer bases, distinct rims at the
base-shaft junction, and elongate, slow-tapering shafts with a fine tip. The PLS
aciniform field in Leucauge is obviously modified by being narrowed and focused
into more regular, elongate rows. I have found the same feature in Tetragnatha,
Pachygnatha, and to a lesser extent in Meta, all “metine” genera. It also occurs in
at least some linyphiids (e.g., Frontinella, Fig. 41). Perhaps it is a synapomorphy
for the same group of derived araneoids defined by use of an inside first leg
forward tap during sticky segment localization behavior (Coddington 1986a). The
distribution of PMS aciniform spigots is also intriguing. In Araneus, Mecynogea,
and Cyrtophora a brush of aciniform spigots on the anterior face of the PMS is
present (Figs. 12, 16, 20; AC). In the same position, cribellate orbweavers such as
Deinopis and Octonoba (Figs. 4, 8) have paracribellar spigots, but they also have
an extensive anterior brush of aciniform spigots. This is also true, for example, of
Micrathena and Cyclosa among the araneids, various other deinopids and
uloborids, and potential cribellate outgroups to orbweavers such as amaurobiids.
An extensive anterior PMS aciniform brush is probably a plesiomorphic feature.
More derived araneoids, such as Leucauge, Latrodectus, Theridula, Gaucelmus,
and Frontinella illustrated here, have no similar anterior aciniform brush. Nephila
also lacks aciniform spigots in the same area (Kovoor 1986; pers. obs.). The trait
is also present in Anapidae, Theridiosomatidae, Mysmenidae, and other
theridiids, nesticids, tetragnathids and linyphiids not illustrated here. Perhaps it is

90

THE JOURNAL OF ARACHNOLOGY

related to increasing specialization of the spinning apparatus. Kovoor and Peters
(1988) noted that histologically two classes of aciniform glands exist, aciniform A
and aciniform B. I find that distinguishing their spigots with SEM is difficult,
but, based on the uloborid Polenecia, they suggested that aciniform B spigots
were somewhat larger. Interestingly, they also point out that linyphiids, metines,
and theridiids (at least) among the araneoids lack aciniform A glands. Very
possibly the reduced PMS aciniform field which can be seen with the SEM is the
external morphological evidence for the lack of aciniform A glands among
derived araneoids. If so, a reduced complement of PMS glands and spigots
becomes an additional synapomorphy of derived araneoids (see Coddington
1986a for others), and is the first evidence for a more exact placement of
theridiid-nesticids among the araneoids.

The reduction and focusing of aciniform spigots on both the PMS and PLS in
derived araneoids correlates well with the absence (or reduction) of prey-wrapping
behavior. Aciniform spigots probably are mostly responsible for the threads used
in prey-wrapping (Table 1). As intimated previously (Coddington 1986a), prey-
wrapping and in particular attack-wrapping seems to be a plesiomorphic feature
of orbweavers that has been lost in the more derived lineages. This interpretation,
which follows directly from cladistic reasoning and outgroup comparison,
contradicts previous hypotheses about the evolutionary history of prey-wrapping
that were based on the assumed adaptive value of the trait (Robinson 1975;
Eberhard 1982). It also illustrates how adaptive hypotheses formed in the absence
of cladistic information can mislead (Coddington 1988).

Some patterns are harder to explain. For example, Araneus, Mecynogea,
Cyrtophora, Leucauge, and Gaucelmus all have what appears to be a vestigial
spigot on the PMS, posterior to the mAP spigot (Figs. 12, 16, 20, 24, 36). It is
probably the vestigial remnant of a minor ampullate gland spigot which is lost in
the adult instar (like the vestigial ALS major ampullate, e.g.. Figs. 1, 15). This
nubbin is absent in Theridula, Latrodectus, Frontinella, and the cribellates. With
this distribution, the feature might be another araneoid synapomorphy, uniquely
lost in theridiids, or a synapomorphy of theridiids plus linyphiids, but one would
need more evidence to say.

The data presented here also bear on a question in the spinneret histology
literature. Based on histological evidence, Kovoor (1977c, 1978) was unsure
whether the uloborid pseudoflagelliform gland was homologous to the araneoid
flagelliform gland. They are apparently similar in shape, anatomy, and chemistry.
Micrographs show that orb weavers all have a distinctive PLS spigot on the
anterior margin. The morphology and placement of the spigot strongly suggest an
interpretation of homology.

Nevertheless, Kovoor and Peters (1988) recently denied homology between
flagelliform and pseudoflagelliform glands (and presumably their spigots), and
also the monophyly of orbweavers. However, their arguments, when closely
analyzed, misconstrue accepted rules of phylogenetic inference. They assert only
that araneoids possess features not found in Deinopoidea (aggregate glands — a
autapomorphy of Araneoidea); that Deinopoidea possess some features not found
in araneoids (the cribellum, calamistrum, and paracribellum — plesiomorphies
found in many cribellate taxa); and they imply that the pseudoflagelliform and
flagelliform glands and spigots “cannot be regarded as homologous.” The first
two assertions are irrelevant to the problem at hand because they refer to an

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

91

autapomorphy and a plesiomorphy, respectively, and they unfortunately do not
detail their evidence for the last assertion. On the other hand, they admit the
many behavioral and web-architectural similarities between cribellate and
ecribellate orbweavers, and presumably acknowledge the additional morphologi-
cal similarities (Coddington 1986a). Given this suggestion of monophyly, and the
lack of any evidence that links deinopoids or araneoids to a non-orbweaving
group, most phylogeneticists would accept Hennig‘s principle that features should
be regarded as homologues unless contradictory evidence overrules that inference.
Put another way, one accepts the homology of bird wings and mammal forelimbs
not because the differences between the structures are small or large, but because
we have no evidence to contest the inclusion of birds and mammals in tetrapod
amniotes. Exactly the same situation obtains in the case of the Orbiculariae (=
Deinopoidea plus Araneoidea).

If both spigot morphology and gland histology agree, as in the issues of
identification discussed above, then homology statements are doubly strong. If
one source of evidence is suggestive but ambiguous, the other may resolve the
issue, as for pseudoflageiliform and flagelliform glands. If actual conflicts in
synapomorphy schemes exist, however, it would be difficult to decide with
complete objectivity which source to accept, especially since we do not have much
experience in evaluating for phylogeny either gross spigot morphology, or
histochemistry and gland ultrastructure.

Comparison with histological and behavioral data. — It is interesting that a
surprisingly high number of histochemical and ultrastructural facts have not been
concordant with other comparative data on spiders, and thus with inferred
phylogenies. Some strange examples are: an S-shaped major ampuilate gland is
characteristic of Hersiliidae and Nephilinae (Kovoor 1987); the proximal part of
the ampuilate gland is reduced to a collar of cells in, e.g., Hypochilus, Fiiistata,
Dictyna, Amaurobius, Telema, Pholcus, IJroctea, and Linyphia (but not in
Oecobius or other araneoids); three secretory regions are only present in the
ampuilate glands of Cyrtophora, Cydosa, and Gasteracanlha (not a moeophyletic
group), but only two in remaining araneids; pyriform glands are tripartite in
Leucauge and Oecobius, but unipartite in Uroctea and, presumably, other
metines; Hersiiia, but not Oecobius or Uroctea, has tripartite acieiform glands;
amino terminal groups are present in acieiform glands of theridiids and
iinyphiids, which apparently correspond to the aciniform B glands of Araneidae,
Hersiliidae, and Polenecia uniquely among the uioborids (most examples from
summary in Kovoor 1987). Any biologist familiar with corroborated phylogenies
of spiders would be puzzled, to say the least, by the above groupings.

I am not sure why this is so. It may be because histochemical analysis often
focuses on the chemical behavior of molecules, and not on their informational
structure. The same lack of concordance with other systematic information was
apparent in the early biochemical analysis of enzymes that gave, for example,
percentages of specific amino acids, pH data, or molecular weights. Only later did
biochemists discover that the informational content of enzymes was in the
sequence of amino acids, rather than their relative abundances or other such
summary features. Likewise, chemical characterizations of glandular products, for
example as “acidophilic”, “tyrosine-rich”, “carboxyl-rich”, or “rich in reducing
groups” simply may not identify conservative phylogenetic features. While
realizing that the phylogenetic analysis of these ultrastructural characters is still

92

THE JOURNAL OF ARACHNOLOGY

young, at this point it is clear that observed points of similarity in some cases
contradict massively corroborated phylogenetic groupings. Therefore homology
arguments based on histochemistry and ultrastructure apparently need careful
evaluation.

Whether the same difficulties of interpretation will characterize the study of
gross morphology and distribution of spigots remains to be seen. Thus far the
best known groups are the araneoids, and their spigot distributions are apparently
more or less concordant with other phylogeneticaliy useful character systems.

Spinning behavior obviously must depend to some extent on spinnerets and
spigots. Therefore it is appropriate to comment also on recent behavioral
research. Eberhard (1987) recently studied aspects of cribellate web-building
behavior in what are apparently primitively non-orbweaving groups. He
concluded that the tendency to spin sticky silk centripetally in cribellate and
ecribellate orbweavers was widespread and probably plesiomorphic. That is, other
non-orbweaving cribellate taxa such as fiiistatids, eresids, psechrids, and dictynids
also start the spinning of sticky silk at the edge of their webs, and finish at the
center or at the retreat. A centripetal tendency in the spinning of cribellate silk
outside the true orbweavers was one specific prediction of the monophyly
hypothesis (Coddington 1986a: 362). The fact that it occurs in the entire range of
cribellate taxa is disappointing because so broad a distribution offers no evidence
as to which of these taxa is the sister group of orbweavers. On the other hand, it
partially overlaps the distribution of the pseudoflagelliform and paracribellar
spigots, and thus ail of these features when analyzed in tandem may elucidate
araneomorph phylogeny.

In contrast, Eberhard (1987) argued that this widespread behavioral trait made
it more plausible that behaviors characteristic of orb-weaving had evolved at least
twice. In contrast, I still see no evidence that these behaviors are convergent. As
pointed out above, such a conclusion would be logical only if synapomorphic
features were discovered that linked only a portion of the orbweavers with a
primitively non-orbweaving group.

Of course orbweavers, whether cribellate or not, still exhibit many unique and
detailed behaviors, such as laying sticky silk in a continuous spiral, shifting
combing legs (cribellates) or plucking-snubbing legs (araneoids) halfway through
construction of a single sticky segment, sticky spiral localization, frame behavior,
exploration behavior, non-sticky spiral construction, and the over-all algorithm of
web construction. Orbweavers are also distinctive among all other major groups
of spiders because complete and typical webs are produced in a single behavioral
bout, usually lasting a few hours or less. Other web-spinning spiders typically
take several days, and several bouts of behavior, to complete the architecture
typical of their taxon. The behaviors unique to orbweavers are similar not only in
gross aspect and function, but also in the details of movements of individual legs.
Many of these appear to be true behavioral synapomorphies for orbweavers. The
tendency to lay sticky silk centripetally is all very well, and it may even define a
monophyletic group (however huge, if it includes everything from Filistatidae to
Araneoidea). Indeed it seems possible that the centripetal tendency in sticky silk
spinning may even be a primitive trait of Araneomorphae.

Interestingly, the same sort of conclusion apply to the pseudoflagelliform. gland.
Kovoor (1987) mentions glands “identified as pseudoflagelliform” for eresids,
amaurobiids, psechrids, and zoropsid spiders. She also mentions that such glands

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

93

are not found in Filistatidae or, surprisingly, in Dictynidae. Histological
observations on Hypochilidae, and Austrochilidae have not been published, but I
have found several distinctly shaped spigots on the PLS of filistatids, hypochilids,
and austrochilids, as well as amaurobiids and eresids. Given the position of
flagelliform gland spigots in araeeoids and pseudoflageliiform spigots among
deinopoids, one would expect to find homologues of these glands uniquely on the
PLS. On the other hand, I have not yet found any specialized morphologies in
eresids or amaurobiids, and thus something of a contradiction is developing
between histological and morphological pattern. Regardless of how this smaller
controversy is resolved, if the histological research accurately identifies
homologues of the uioborid pseudoflageliiform gland, we may have to conclude
that primitive pseudoflageliiform glands and spigots evolved soon after the origin
of araneomorph spiders — at least soon after the origin of the Araneoclada (all
araneomorphs exclusive of Hypochilidae, Austrochilidae, and Gradungulidae).

Like a “centripetal tendency” in sticky silk spinning, thee, the pseudoflagellL
form gland would .become part of the primitive ground plan for most true
spiders. Arguments for the unique homology of the deieopoid and araneoid
condition would then depend on further special similarities such as that the single
pair of glands opens only on the PLS, and that they provide the sole pair of base
fibers of the sticky line in both cases.

Conclusions. — Taken together, the spigot evidence thus far corroborates the
hypothesis of orbweaver monophyly, and certainly does not dispute it. The
spinning fields of cribellate orbweavers are more similar to those of ecribellate
orbweavers than they are to nomorbweaving cribellate groups. Among araneoids,
the Araneidae still exhibit fairly primitive spinneret morphologies. The same can
be said for the details of web construction (Coddington 1986a, b). As always,
being primitive in one feature does not make a taxon primitive in all respects.
Thus araoeid genitalia may be relatively more derived than those of metines or
the symphytognathoid taxa, or at least far from the orbweaver ground plan.

Ail of the spiders considered here are orbweavers, or are descended from
orbweavers, based on other evidence. Although this character system is obviously
useful within the orbweavers, it would be interesting to know how well this
system of logic will work for more distantly related, and less derived cribellate
groups which may be the sister taxon to orbweavers, such as Hypochiloidea,
Amaurobioidea, or Dictynoidea. Even though among distantly related and little
studied groups the use of the conventional names for araneoid glands becomes
increasingly risky and less justifiable, the basic method of comparing between
singulars and multiples, between sexes and instars, and from one spinneret to
another should be a primary tool for deciphering spigot homology.

ACKNOWLEDGMENTS

I would like to thank Scott Larcher for help with scan preps and darkroom
work. Walter Brown, Suzanne Braden, and Brian Kahn of the USNM Scanning
Electron Microscope Facility were always helpful. I thank Prof. Dr. Hans Peters
and Jacqueline Palmer for many stimulating conversations and sharing their ideas
and results in advance of publication. The comments of Herbert W. Levi and
Norman 1. Platnick greatly improved the manuscript. However, my greatest debt

94

THE JOURNAL OF ARACHNOLOGY

is to Jacqueline Kovoor, whose steady investigation of spider spinning structures
first suggested this study, and provided the context that made it possible.

LITERATURE CITED

Coddington, J. A. 1986a. The monophyletic origin of the orb web. Pp. 319-363, In Spiders: Webs,
Behavior, and Evolution. (W. A. Shear, ed.). Stanford Univ. Press, Stanford, CA.

Coddington, J. A. 1986b. Orb webs in non-orb-weaving ogre-faced spiders (Araneae: Deinopidae): a
question of genealogy. Cladistics, 2(l):53-67.

Coddington, J. A. 1988. Cladistic tests of adaptational hypotheses. Cladistics, 4(1): 1-22.

Eberhard, W. G. 1982. Behavioural characters for the higher classification of orb-weaving spiders.
Evolution, 36(5): 1067-1095.

Eberhard, W. G. 1987. Construction behavior of non-orb weaving cribellate spiders and the
evolutionary origin of orb webs. Bull. Brit. Arachnol. Soc., 7(6): 175-178.

Kovoor, J. 1972. Etude histochimique et cytologique des glandes sericigenes de quelques Argiopidae.

Ann. Sci. nat. Zool., 14:1-40.

Kovoor, J. 1977a. Donnees histochimiques sur les glandes sericigenes de la veuve noire Latrodectus
mactam Fabr. (Araneae: Theridiidae). Ann. Sci. nat. Zool., 19:63-87.

Kovoor, J. 1977b. La soie et les glandes sericigenes des Arachnides. Ann. Biol., 16:97-171.

Kovoor, J. 1977c. L‘appareil sericigene dans le genre Uloborus Latr. (Araneae: Uloboridae). Rev.
Arachnol., 1:89-102.

Kovoor, J. 1978. L‘appareil sericigene dans le genre Uloborus Latreille (Araneae: Uloboridae), II.

Donnees histochimiques. Ann. Sci. nat. Zool., 20:3-25.

Kovoor, J. 1986. L‘appareil sericigene dans les genres Nephila Leach et Nephilengys Koch; anatomic
microscopique, histochemie, affinites avec d‘autres Araneidae. Rev. Arachnol., 7(1): 15-34.

Kovoor, J. 1987. Comparative structure and histochemistry of silk-producing organs in arachnids. Pp.

160-186, In Ecophysiology of Arachnids. (W. Nentwig, ed.). Springer- Verlag, Berlin.

Kovoor, J. 1988. L‘appareil sericigene des Mecynogea Simon (Araneae: Araneidae). Rev. Arachnol.,
7(5):205-213.

Kovoor, J., and H. Peters, 1988. The spinning apparatus of Polenecia producta (Araneae,
Uloboridae): structure and histochemistry. Zoomorphology, 108:47-59.

Kovoor, J. and A. Lopez. 1982. Anatomic et histologic des glandes sericigenes des Cyrtophora
(Araneae, Araneidae): affinites et correlations avec la structure et la composition de la toile. Revue
Arachnol., 4:1-21.

Kovoor, J. and A. Lopez. 1988. L‘appareil sericigene des Mecynogea Simon (Araneae, Araneidae).
Revue Arachnol., 7(5):205-212.

Lehtinen, P. T. and M. I. Saaristo. 1980. Spiders of the Oriental-Australian regions, 11. Nesticidae.
Ann. Zool. Fenn. 17:47-66.

Levi, H. W. 1980. The orb-weaver genus Mecynogea, the subfamily Metinae and the genera
Pachygnatha, Glenognatha and Azilia of the subfamily Tetragnathinae north of Mexico (Araneae:
Araneidae). Bull. Mus. Comp. Zool., 149(1): 1-75.

Levi, H. W. 1983. The orb-weaver genera Argiope, Gea and Neogea from the western Pacific region

(Araneae: Argiopinae, Araneidae). Bull. Mus. Comp. Zool., 150(5):247-338.

Levi, H. W. and J. A. Coddington. 1983. Progress report on the phylogeny of the orb-weaving family
Araneidae and the superfamily Araneoidea. Verb, naturwiss. Ver. Hamburg, NF(26): 151-154.
Mikulska, 1. 1966. The spinning structures on the spinnerets (thelae) of Nephila clavipes (L.).
Zoologica Poloniae, 16(3-4):209-222.

Mikulska, I. 1967. The external spinning structures on the thelae of the spider Argiope aurantia

Lucas. Zoologica Poloniae, 17(4):358"365.

Peters, H. M. and J. Kovoor. 1980. Un complement a Tappareil sericigene des Uloboridae (Araneae):

le paracribellum et ses glandes. Zoomorphology, 96(1-2):91-102.

Riedl, R. 1979. Order in Living Organisms. John Wiley & Sons, New York.

Robinson, M. H. 1975. The evolution of predatory behavior in araneid spiders. Pp. 292-312, In
Function and Evolution in Behavior (G. Baerends, C. Beer and A. Manning, eds.). Clarendon
Press, Oxford.

CODDINGTON— SPINNERET SPIGOT MORPHOLOGY

95

Shear, W. A. 1986. The evolution of web-building behavior in spiders: a third generation of
hypotheses. Pp. 364-400, In Spiders: Webs, Behavior, and Evolution. (W. A. Shear, ed.). Stanford
Univ. Press, Stanford, CA.

Tyshchenko, V. P. 1986. New confirmation of the convergent origin of orb webs in cribellate and
ecribellate spiders. Doklady Akad. Nauk SSSR, 287(5): 1270-1273.

Wasowska, S. 1966. Comparative morphology of the spinning fields in females of some spider species.

Zoologica Poloniae, 16(l):9-30.

Wasowska, S. 1970. Structures fileuses exterieures sur les filieres (thelae) de Taraignee Argiope
bruennichi (Scopoli). Zoologica Poloniae, 20(2):257-268.

Whitehouse, M. E. A. 1987. “Spider eat spider”: the predatory behavior of Rhomphaea sp. from New
Zealand. J. Arachnol., 15:355-362.

Manuscript received June 1988, revised October 1988.

■»r>

. ■■ ' I ' ;,

= ' r ''”V ■;■:■■ *' ■'*- .v '.

^ „ V ’■ ■ '■;,

;■* ■■ ■■ : r,'i' tS J '. 7^ ViifviLt^Ni' k1^'^'Sz,.i-.i«yF^!:.-..5:^ . <•: '• A ; . » , M.r Z.,^;it/5im1U2iiiii3HTKZU.*v

’ -I pm

V

>:. ^;|i; Ji

®. : ' ■*‘t

. 'li ■.■T »■ ■*;ii'A.. Wi -f;

>: I ■■'J>. t'''^ :<&; ,.(■ !(. i-; 4

fe?

■I '

.-*^^1 ,i4A, > ‘-■* ,v,'' V t ■)H. JBII|H?*|

'!(>-■ fm.. if ^ ui-t^ ■ Z'V?*i^- .' ■ v

.Va

■Jt «

7 #•'

i.C: <.SiC#r '^f^M:,!

* M, ■'■'»’■ .'* fr f ■ I .*«y ■4'^r^if' 'Y;.j i?;;

lit!* ''v'*P*f'^.’ 'is;;

■»• JL-.'''‘ <> T ^JaI :!% *' ‘*’''' ■ ■'*r ' '/'; !• ' ' / -*~* ' -

-: ... . . , - , - .'-I , ,"

■,. <^-% ' r

■■■ '- •'-‘’z

Corey, D. T. and W. K. Tayior. 1989. Foliage-dwelling spiders in three central Florida plant
communities. J. Arachnol., 17:97-106.

FOLIAGE-DWELLING SPIDERS IN THREE
CENTRAL FLORIDA PLANT COMMUNITIES

David T. Corey * and Walter K. Taylor

Department of Biological Sciences
University of Central Florida
Orlando, Florida 32816 USA

ABSTRACT

Foliage-dwelling spiders were collected using sweep nets in pond pine, sand pine scrub, and
flatwoods plant communities on the University of Central Florida campus near Orlando. Collections
were made bimonthly from May, 1983 through March, 1984.

A total of 4,022 spiders was collected; 2,076 in pond pine, 1,258 in sand pine scrub, and 688 in
flatwoods. Spider diversity was greatest in pond pine, followed by sand pine scrub and then flatwoods
community. Similarity in spider species was greatest between pond pine and flatwoods.

Salticids represented 40.2% of the combined populations. Misumenops celer (Hentz) was found in
all three plant communities and was abundant.

INTRODUCTION

The southeastern United States has a rich spider fauna. Investigators have
studied the spider fauna in North Carolina (Barnes 1953; Barnes and Barnes
1955; Berry 1970, 1977). In Florida, Muma (1973), Rey and McCoy (1983), and
Lowrie (1963, 1971) have sampled the spider fauna.

The foliage-dwelling spiders of the pond pine, sand pine scrub, and flatwoods
communities have not been described. The scorpion, pseudoscorpion, opilionid,
and ground surface spider faunas in these communities have been described
(Corey and Taylor 1987).

This paper compares the foliage-dwelling spider faunas in the pond pine, sand
pine scrub, and flatwoods communities. In addition we show seasonal differnces
in the three communities.

MATERIALS AND METHODS

Study sites. — The three plant communities where our study occurred lie in the
eastern part of the University of Central Florida campus, located approximately
17 km east of Orlando in Orange County (SIO R31E T22S). The three plant
communities studied were pond pine, sand pine scrub and flatwoods. For a
description of the plant communities, see Corey (1987) and Corey and Taylor
(1987).

Methods. — The sweep net consisted of a 9L4-cm handle, 40.6-cm ring, and
collecting bag made of white canvas. A single sweep consisted of; 1) first stroke

'Present address: Department of Zoology, Southern Illinois University, Carbondale, IL 62901 USA.

98

THE JOURNAL OF ARACHNOLOGY

Figure 1. — Mean number of spiders caught on
foliage using sweep nets in pond pine (•), sand
pine scrub (A), and flatwoods (■).

M J S N J IVf

of the net started on the left and moved toward the right forming a 180 degree
arc, 2) the second stroke covered the same area as the first stroke, but the net was
moved in the opposite direction, 3) after completing the two strokes, one step
forward was taken and the two-stroke method was repeated (for 100 steps), 4)
then the 100 steps were retraced using the two-stroke method. Each sweep
consisted of 1,200 strokes.

Thirty sets of sweeps per collecting month per plant community were made
beginning in May 1983 and ending in March 1984 for a total of 540 sweeps.
Sweeps were made on three consecutive days, one day for each community. All
materials netted were placed into plastic bags and returned to the laboratory.
Spiders were separated from the debris and placed into baby food jars containing
70% ethanol.

Identification. — Spiders were identified using a dissecting microscope. Difficult
specimens were identified or verified by Jonathan Reiskind, University of Florida;
James H. Redner, Biosystematics Research Institute; Norman I. Platnick,
American Museum of Natural History; G. B. Edwards, Florida State Collection
of Arthropods; and Jonathan Coddington, Smithsonian Institution.

All spiders were identified to lowest possible taxon. Many immatures were
identified only to family level. Some spiders collected in poor condition could not
be identified to family; these specimens are reported as undetermined (See Table
3).

RESULTS AND DISCUSSION

Four-thousand and twenty-two spiders from 18 families and 89 species were
collected using sweep nets; 2,076 individuals, 14 families and 58 species from
pond pine, 1,258 individuals, 15 families and 53 species from sand pine scrub, and
688 individuals, 13 families and 54 species from flatwoods. See appendix for a
complete list of the spider species. An average of 7.45 spiders per sweep was
observed. Figure 1 shows the mean number of individuals collected per sweep for
the six collecting periods. Sixty-nine percent more spiders were found in pond
pine than in flatwoods, 39% more in pond pine than in sand pine scrub, and 45%
more in sand pine scrub than in flatwoods.

Analysis of guild composition shows differences between communities (Fig. 2).
Guilds were patterned after Gertsch (1979). Guilds are (1) jumping spiders;
Salticidae, (2) crab spiders; Thomisidae and Philodromidae, (3) aerial web
spinners; Theridiidae, Linyphiidae, and Araneidae, (4) hunting spiders; Pisauri-

COREY & TAYLOR— FOLIAGE-DWELLING SPIDERS IN FLORIDA

99

O

Q

QQ

Others

Running spiders
Hunting spiders

Aerial web
spinners

Crab spiders

- Jumping spiders

Figure 2. — Guild composi-
tion of individual spiders for
the three study sites, PP =
pond pine, SPS = sand pine
scrub, and FW = flatwoods.

dae, Lycosidae, and Oxyopidae, (5) running spiders; Gnaphosidae, Clubionidae,
and Anyphaenidae, and (6) others; remainder of the spiders. Relative abundance
of jumping spiders declined in pond pine compared to that of sand pine scrub
and flatwoods. Crab spiders increased substantially in pond pine.

Simpson’s Index of Diversity was calculated for the three communities
(Simpson 1949). Pond pine had a value of 0.84, sand pine scrub of 0.88, and
flatwoods of 0.89. These low values may be due to the high species richness found
in each community and the small number of dominant species.

Sorensen’s Index of Similarity (Krebs 1978) was used to determine the
similarities of spider species composition among communities. Species composi-
tion was more similar between pond pine and flatwoods (0.68), followed by sand
pine scrub and flatwoods (0.63). Pond pine and sand pine scrub (0.57) were least
similar. It was expected that habitats with similar vertical plant structure would
be similar in spider species. This relationship, however, was not true; pond pine
was less similar to sand pine scrub than to flatwoods as to species composition.
Height and vertical structure of the vegetation swept may have allowed a greater
spider abundance, but not a greater similarity in species. Pond pine had the
highest understory swept (3 m) and the largest number of spiders, followed by
sand pine scrub (averaged 2 m). Flatwoods with a low understory (averaged 1.5
m) had the smallest number of spiders.

Pond pine community had the most complex vegetation layer and this may
have attributed to its having a greater spider abundance. Few herbaceous plants
occurred in flatwoods as compared to pond pine and sand pine scrub. Flatwoods
had a smaller surface area for spiders and this may account for the low
abundance.

Table 1 shows the mean number of individual spiders occurring in the three
communities. For each monthly mean, 95% confidence intervals were calculated
SLS X + t (SE) (Simpson et al. 1960). Pond pine in November was significantly
different from the other communities in mean number of spider individuals.

100

THE JOURNAL OF ARACHNOLOGY

Table 1 . — Mean number of individual spiders occurring on foliage in the study sites by collection month.

Collection month

Community

Pond pine

T ± (SE)

Sand pine scrub

T ± (SE)

Flatwoods

T ± (SE)

May

81.33 (13.40)

64.33 (2.89)

34.33 (1.77)

July

70.67 (9.85)

90.00 (5.69)

26.00 (4.59)

September

123.33 (26.97)

66.00 (4.01)

10.41 (6.02)

November

186.00 (5.04)

73.00 (11.15)

10.26 (5.93)

January

81.67 (19.17)

53.33 (10.94)

8.14 (4.71)

March

147.67 (29.79)

98.33 (26.69)

8.14 (4.71)

Mean numbers of individuals captured in flatwoods in May, July, and September
were less than and significantly different from sand pine scrub, but not from
pond pine, which had large standard errors associated with the means. In
contrast, mean number of spider species captured in flatwoods in July were
significantly different {p = 0.05) from the other two communities (Table 2). Pond
pine in November was significantly different from flatwoods in mean number of
species. The lack of significant differences may be due to the large variance in
number of individuals and species found among the communities.

Spider families, represented by individuals collected on foliage, are listed in
Table 3. The three most common families for all communities were salticids,
thomisids, and linyphiids; these represent 76.5% of all spiders captured in sweeps.
In pond pine, thomisids, salticids, and linyphiids represented 82.4% of that
community’s total spider assemblage. In sand pine scrub, salticids, linyphiids, and
thomisids represented 74.8% of the total spider assemblage. In flatwoods,
salticids, oxyopids, and araneids represented 71.6% of the total spider assemblage.
Figure 3 shows seasonal abundance of three common families occurring on
foliage.

Table 4 shows the 10 most common species collected by frequency of
occurrence. Abundant species were Thiodina sylvana (Hentz) and Misumenops
celer (Hentz).

Twenty-seven species occurred in all communities (Table 5). No single species
was common in all three communities. Nine species of spiders were collected from
only one of the three communities: Uloborus qlomosus (Walck.), Pholcomma
hirsutum Emerton, Neriene radiata (Walck.), Ceratinopsis sp., Tibellus oblongus
(Walck.) and Zyqoballus rufipes (Peckham & Peckham) in pond pine; Hyptiotes

Table 2. — Mean number of spider species occurring on foliage in the study sites by collection month.

Collection month

Community

Pond pine

T ± (SE)

Sand pine scrub

T ± (SE)

Flatwoods

T ± (SE)

May

17.33 (0.88)

16.00 (3.61)

12.33 (1.46)

July

14.33 (0.88)

17.33 (1.46)

8.00 (0.58)

September

13.67 (1.46)

11.00 (0.58)

13.33 (2.89)

November

18.33 (0.88)

15.33 (1.77)

12.00 (0.58)

January

10.67 (0.34)

11.67 (0.88)

8.33 (0.88)

March

17.67 (1.86)

15.67 (1.77)

13.00 (1.53)

COREY & TAYLOR-FOLIAGE-DWELLING SPIDERS IN FLORIDA

101

Table 3. — Number of individuals and percent of spiders by family for the three communities collected
with sweep nets.

Pond

pine

Sand pine

Flatwoods

Total

#

%

#

%

#

%

#

%

Uloboridae

6

0.3

5

0.4

0

0.0

11

0.3

Dinopidae

1

0.1

0

0.0

0

0.0

I

0.02

Dictynidae

0

0.0

1

0.1

0

0.0

1

0.02

Theridiidae

95

4.6

52

4.1

20

2.9

167

4.2

Linyphiidae

168

8.1

206

16.4

35

5.1

409

10.2

Linyphiinae

93

4.5

6

0.5

20

2.9

119

3.0

Erigoninae

75

3.6

200

15.9

15

2.2

290

7.2

Araneidae

115

5.5

101

8.0

79

11.5

295

7.3

Tetragnathidae

10

0.5

2

0.2

16

2.3

28

0.7

Agelenidae

0

0.0

0

0.0

7

1.0

7

0.2

Hahniidae

0

0.0

4

0.3

0

0.0

4

0.1

Pisauridae

0

0.0

0

0.0

2

0.3

2

0.05

Lycosidae

3

0.1

1

0.1

0

0.0

4

0.1

Oxyopidae

27

1.3

43

3.4

92

13.4

162

4.0

Gnaphosidae

3

0.1

1

0.1

3

0.4

7

0.2

Clubionidae

35

1.7

30

2.4

6

0.9

71

1.8

Anyphaenidae

19

0.9

34

2.7

11

1.6

64

1.6

Thomisidae

836

40.3

146

11.6

66

9.6

1048

26.1

Philodromidae

9

0.4

3

0.2

3

0.4

15

0.4

Salticidae

706

34.0

589

46.8

321

46.7

1616

40.2

Undetermined

43

2.1

40

3.2

27

3.9

no

2.7

Figure 3. — Seasonal distribution of the most
common families of foliage-dwelling spiders caught
with sweep nets in pond pine (•), sand pine scrub
(A), and flatwoods (■). Salticidae (upper), Thomi-
sidae (middle), and Linyphiidae (lower).

M J S N J M

102

THE JOURNAL OF ARACHNOLOGY

Table 4. — Ten most common spider species ranked by frequency of occurrence within each plant
community. NR = no rank assigned.

Species

Pond pine

Sand pine scrub

Flatwoods

Synema viridans

1

NR

9

Thiodina sylvana

2

5

5

Misumenops celer

3

4

3

Lyssomanes viridis

4

3

NR

Frontinella pyramitela

5

NR

8

Theridula sp.

6

NR

NR

Hentzia palmarum

7

1

1

Phidippus pulcherrimus

7

NR

NR

Grammonota sp. #1

9

2

NR

Leucuage venusta

10

NR

NR

Grammonota sp. #2

NR

6

NR

Hypsosinga pygmaea

NR

7

NR

Aysha gracilis

NR

8

NR

Marpissa pikei

NR

9

4

Tmarus floridensis

NR

10

NR

Peucetia viridans

NR

NR

2

Phidippus sp.

NR

NR

6

Acacesia hamata

NR

NR

7

Oxyopes salticus

NR

NR

9

Table 5. — Spiders occurring in all three plant communities. R = rare (less than 1% of total population
for that community), P = present (1% to 4.9% of total population), and C = common (5% or more of the
total population).

Species

Pond pine

Sand pine scrub

Flatwoods

Theridula sp.

P

P

R

Thymoites unimaculatum

R

p

R

Frontinella pyramitela

P

R

P

Grammonota sp. #1

P

C

P

Leucauge venusta

P

R

R

Argiope aurantia

R

R

P

Acacesia hamata

R

P

P

Neoscona domiciliorum

R

R

R

Hypsosinga pygmaea

R

P

R

Wagneriana taurkornis

R

R

R

Peucetia viridans

R

P

C

Oxyopes salticus

R

R

P

Aysha gracilis

R

P

P

A. velox

R

R

R

Tmarus floridensis

R

P

R

Misumenops celer

C

C

P

Synema viridans

c

R

P

Paramaevia michelsoni

R

R

R

Marpissa pikei

R

P

P

Hentzia palmarum

P

C

C

Zygoballus rufipes

R

R

R

Thiodina sylvana

C

P

P

Lyssomanes viridis

c

C

R

Phidippus sp.

R

R

P

Phidippus pulcherrimus

P

R

R

COREY & TAYLOR— FOLIAGE-DWELLING SPIDERS IN FLORIDA

103

cavatus (Walck») in sand pine scrub; and Agelenopsis naevia (Walck.) and
Coriarachne sp. in flatwoods.

Peucetia viridans (Hentz) and Lyssomanes viridis (Walck.) overwintered as
juveniles. Berry (1971) found several spider species to overwinter as juveniles in
the North Carolina Piedmont. He also found that adults and juveniles of some
species appeared in large numbers at the same time of the year after a period of
time when no or very few adults and juveniles were found. The following spiders
exhibited this behavior in our study: Grammonota sp. #1 and Misumenops celer
in pond pine, Thymoites unimaculatum (Emerton) and Hypsosinqa pyqmaea
(Sundevall) in sand pine scrub, and Hentzia palmarum (Hentz) in flatwoods. This
behavior may be a result of rapid maturation.

Lowrie (1963, 1971) studied oxyopids in the vegetation strata in northern
Florida. Of five species he collected, Oxyopes salticus (Hentz) and Peucetia
viridans were represented in our study.

Rey and McCoy (1983) studied arachnids for fifteen months in tidal marshes of
northwest Florida and collected 47 species and 14 families. The following eight
species were represented in their study and in our study: Nephila clavipes (L.),
Oxyopes salticus, Peucetia viridans, Tetraqnatha laboriosa Hentz, Florinda
coccinea (Hentz), Metaphidippus galathea (Walck.), Marpissa pikei (G. & E.
Peckham), and Synemosyna petrunkevitchi (Chapin).

ACKNOWLEDGMENTS

We thank Jonathan Reiskind, Jonathan Coddington, G. B. Edwards, Norman
L Platnick, and James Redner for examining specimens. In addition, we thank I.
Jack Stout, David H. Vickers, Henry O. Whittier, A1 Fliss, and Michael D. Albig
for their assistance throughout this study. Jonathan Reiskind and G. B. Edwards
made helpful comments to improve an earlier draft of the manuscript. This work
was supported by a grant to David T. Corey from the Exline-Frizzell Fund for
Arachnological Research, grant No. 8.

LITERATURE CITED

Barnes, R. D. 1953. The ecological distribution of spiders in non-forest maritime communities at
Beaufort, North Carolina. Ecol. Monog., 23:315-337.

Barnes, R. D. and B. M. Barnes. 1955. The spider population of the abstract broomsedge community
of the southeastern Piedmont. Ecology, 36: 658-666.

Berry, J. W. 1970. Spiders of the North Carolina Piedmont old-field communities. J. Elisha Mitchell
Sci. Soc., 86:97-105.

Berry, J. W. 1971. Seasonal distribution of common spiders in the North Carolina Piedmont.
American Mid. Nat., 85:526-531.

Corey, D. T. 1987. Arachnid fauna in three central Florida plant communities. M.S. thesis. University
of Central Florida, Orlando, FL. 290 pp.

Corey, D. T. and W. K. Taylor. 1987. Scorpion, pseudoscorpion, and opilionid faunas in three central
Florida plant communities. Florida Scient., 50:162-167.

Corey, D. T. and W. K. Taylor. Ground surface spiders in three central Florida plant communities. J.
Arachnol., 16:213-221.

Gertsch, W. J. 1979. American Spiders. Van Nostrand Reinhold, New York.

Krebs, C. J. 1978. Ecology: The Experimental Analysis of Distribution and Abundance. Harper &
Row. New York.

104

THE JOURNAL OF ARACHNOLOGY

Lowrie, D. C. 1963. The effects of grazing and intensive collecting on a population of the green lynx
spider. Ecology, 44:777-781.

Lowrie, D. C. 1971. Effects of time of day and weather on spider catches with a sweep net. Ecology,
52:348-351.

Muma, M. H. 1973. Comparison of ground surface spiders in four Central Florida ecosystems.
Florida EntomoL, 56:173-196.

Rey J. R. and E. D. McCoy. 1983. Terrestrial arthropods of Northwest Florida salt marshes: Araneae
and Pseudoscorpiones (Arachnida). Florida EntomoL, 66:497-503.

Simpson, E. H. 1949. Measurement of diversity. Nature, 163:688.

Simpson, G. G., A. Roe and R. C. Lewontin. 1960. Quantitative Zoology. Harcourt, Brace and Co.,
New York.

Manuscript received July 1988, revised October 1988.

APPENDIX

SPIDERS IN THREE CENTRAL FLORIDA PLANT COMMUNITIES

Species of spiders collected with sweep nets are listed by family and community: Pond pine (PP), sand
pine scrub (SPS), and flatwoods (FW).

Family

Species

PP

SPS

FW

Dinopidae

1. Dinopis spinosa Marx

1

TOTALS

I

0

0

Uloboridae

2. Uioborus glomosus (Walck.)

6

3. Hyptiotes cavatus (Hentz)

5

TOTALS

6

5

0

Dictynidae

4. Dictyna sp.

1

TOTALS

0

1

0

Theridiidae

13

17

9

5. Pholcomma hirsutum Emerton

7

6. Spintharus flavidus Hentz

1

1

7. Theridula sp.

66

17

4

8. T. unimaculatus (Emerton)

2

17

3

9. Theridion flavonotatum Becker

1

2

10. Steatoda triangulosa (Walck.)

1

1

1 1. Stemmops bicolor (OPC)

1

12. Argyrodes elevatus (Walck.)

3

1

TOTALS

95

52

20

Linyphiidae

Linyphiinae

3

3

1

13. Frontinella pyramitela (Hentz)

68

3

14

14. Neriene radiata (Walck.)

7

15. Florinda coccinea (Hentz)

15

5

Totals

93

6

20

Erigoninae

16

16

5

16. Grammonota sp. #I

48

142

7

17. Grammonota sp. #2

3

36

18. Grammonota sp. #3

4

19. Ceratinopsis sp.

6

20. Species #1

1

21. Species #2

1

22. Species #3

2

3

75 200 15

TOTALS

COREY & TAYLOR— FOLIAGE-DWELLING SPIDERS IN FLORIDA

Araneidae

17

10

25

23. Gasteracantha elipsoides (Walck.)

1

24. Micraihena gracilis (Walck.)

1

2

25, M. sagittata (Walck.)

1

26. Leucauge venusta (Walck.)

43

4

2

27. Nephila clavipes (Linnaeus)

I

28. Argiope sp.

1

29. A. aurantia Lucas

5

12

8

30. M angora placida (Hentz)

9

6

31. Acanthepeira stellata (Marx)

2

10

32. Acacesia hamata (Hentz)

6

24

15

33. Neoscona arabesca (Walck.)

I

34. N. domiciliorum (Hentz)

3

1

1

35. Araneus minatus (Walck.)

1

I

36. Hyposinga rubens (Hentz)

13

1

37. H. pygmaea (Sundevall)

1

28

3

38. Wagneriana tauricornis (OPC)

15

5

1

39. Scoloderis cordatus (Tacz.)

1

1

40. Species #1

9

4

TOTALS

115

101

79

Tetragnathidae

7

1

8

41. Tetragnatha sp.

2

3

42. T. laboriosa Hentz

1

2

43. T. straminea Emerton

1

3

TOTALS

10

2

16

Agelenidae

2

44. Agelenopsis naevia (Walck.)

5

TOTALS

0

0

7

Hahniidae

45. Hahnia cinerea Emerton

4

TOTALS

0

4

0

Lycosidae

3

1

0

Pisauridae

0

0

2

Oxyopidae

5

22

35

46. Peucetia viridans (Hentz)

18

15

45

47. Oxyopes salticus (Hentz)

4

6

11

48. Hamataliwa grisea Key.

1

TOTALS

27

43

92

Gnaphosidae

3

3

49. Poecilochroa decorata (Kaston)

1

TOTALS

3

1

3

Clubionidae

14

25

5

50. Trachelas deceptus (Banks)

1

51. r similis (FOP Cambridge)

1

52. Chiracanthium inclusum (Hentz)

1

2

53. Clubionoides sp.

20

1

54. C. gertschi Kaston

1

TOTALS

35

30

6

Anyphaenidae

2

1

55. Aysha gracilis (Hentz)

10

27

7

56. A. velox (Becker)

3

7

3

57. Wulfila alba (Hentz)

4

TOTALS

19

34

11

Thomisidae

43

28

14

58. Tmarus sp.

4

2

59. T floridensis Key.

4

25

3

60. Misumenops celer (Hentz)

190

75

26

61. Misumenops asperatus (Hentz)

4

105

106

THE JOURNAL OF ARACHNOLOGY

62. M. oblongus (Key.)

63. Misumenoides formosipes (Walck.)

64. Synema viridans (Banks)

65. Coriarachne sp.

66. Xysticus variabilis Key.

TOTALS

Philodromidae

67. Ebo contrastus (RJS & NIP)

68. Philodromus placidus Banks

69. P. rufus Walck.

70. P. formosipes (Walck.)

71. Tibellus oblongus (Walck.)

TOTALS

Salticidae

72. Synemosyna formica (Hentz)

13. S. petrunkevitchi (Chapin)

74. Peckhamia picata (Hentz)

75. Paramaevia michelsoni (Barnes)

76. Marpissa pikei (G & E Peckham)

77. Agassa cerulea (Walck.)

78. Hentz ia palmar um (Hentz)

79. Zygoballus rufipes P&P

80. Z. sexpunctatus (Hentz)

81. Thiodina sylvana (Hentz)

82. T peurpera (Hentz)

83. Phidippus audax (Hentz)

84. P. pulcherrimus Key.

85. Phidippus sp.

86. Metaphidippus galathea (Walck.)

87. M. tillandsiae Kaston

88. Admestine tibialis (CL Koch)

89. Lyssomanes viridis (Walck.)

TOTAL

Unplaced

1

1 4

598 9 11

5

1

836 146 66

1

1

2 3

3
5

9 3 3

106 132 106

2

1

3

4 6 6

11 26 24

1

52 265 127

11 1 1

1 2

299 71 19

1 1

3

52 1 6

13 6 16

1 1

2 2

4

156 67 6

706 589 321

43 40 27

1989. The Journal of Arachnology 17:107

RESEARCH NOTES

SCYTODES POENITENS CHAMBERLIN, NOT PNOEITENS
(ARANEAE, SCYTODIDAE)

In his descriptions of new species from “the shores and islands of the Gulf of
Baja California,” Chamberlin (1924) showed a great propensity for scientific
names with religious or philosophical meanings. For instance, he used such
religious terms as theologus, dogmaticus, agnosticus, reformans, protestans,
calvanisticus (sic), catholicus, and even scepticus, and redempta. Names with a
philosophical reference are: positivus, rationalis, empiricus, syntheticus, eclecticus,
pragmaticus, reads ticus, scholastica, philosopha. Sometimes these names were
used several times in different genera.

Among these was a new species, Scytodes poenitens, correctly spelled, and
again misspelled in the same paper. Subsequent cataloguers perpetuated the
misspelling (Bonnet 1958: 3988), or produced a new misspelling (Roewer 1942:
330). We correct these errors, and present our reasoning.

Scytodes poenitens Chamberlin

Scytodes poenitens Chamberlin 1924:572 (in a list of species of San Marcos Island).

Scytodes pnoeitem Chamberlin 1924:592 (description of the new species); Bonnet 1958: 3988 (printer’s
error).

Scytodes pnocitens Roewer 1942:330 (lapsus calami or printer’s error).

The fact that Chamberlin’s 1924 paper has two different spellings is sufficient
to justify the correction, since there is in the original publication itself clear
evidence of an iiiadvertant error (International Commission on Zoological
Nomenclature 1985: Art. 32c). The form pnoeitens of Chamberlin is clearly an
error of the printer, who simply got the n in the wrong place. Roewer’s (1942:
330) pnocitens is the result of a further error in which the printer, working from
Roewer’s handwriting, read the e as a c.

Chamberlin started with the correct poenitens (1924: 272) meaning “penitent”,
correctly formed as the present participle of the Latin verb poeniteo. Subsequent
references are all misspelled.

LITERATURE CITED

Bonnet, P. 1958. Bibliographia Araneorum, 2: U5058.

Chamberlin, R. V. 1924. The spider fauna of the shores and islands of the Gulf of California. Proc.
California Acad. Sci. 4th Series, 12(28):56I"694.

International Commission on Zoological Nomenclature. 1985. International Code of Zoological
Nomenclature, 3rd ed. University of California Press, Berkeley and Los Angeles. 338 pp.

Roewer, C. E 1942. Katalog der Araoeae. 1-1040.

1989. The Journal of Arachnology 17:108

Vincent D. Roth, Box 136, Portal, Arizona 85632 USA and H. D. Cameron,
Department of Classical Studies, University of Michigan, Ann Arbor, Michigan
48109 USA.

Manuscript received June 1988.

DO FEMALE MIAGRAMMOPES ANIMOTUS
(ARANEAE, ULOBORIDAE)

SPIN COLOR-COORDINATED EGG SACS?

Like other members of their pantropical genus, Miagrammopes animotus
Chickering females spin capture webs consisting of only a few threads and
monitor them from one of their attachment points (Lubin 1986; Lubin et al.
1978). In this position, the spiders’ cylindrical abdomens, long, slender legs, and
linear web-monitoring postures (Opell 1987) contribute to their twig-like
protective resemblance. Unlike most members of the family Uloboridae that
produce stellate or lenticular egg sacs (Opell 1984), Miagrammopes females
construct cylindrical egg sacs consisting of two columns of eggs wrapped tightly
with tubuliform (cylindrical) gland silk (Foelix 1982; Lubin et al. 1978; Opell
1984). The color of both M. animotus females and their egg-sac-wrapping silk
ranges from light tan to brownish gray to dark, rusty red. During the day,
females hang contiguously with their egg sacs, maintaining fourth leg contact with
the egg sac and first leg contact with a support to which their web is attached
(Fig. 1). This alignment of females with their egg sacs enhances the twig-like
appearance of both and may reduce threats to the spiders from visually hunting
predators such as lizards, birds, wasps, and other spiders, and to their egg sacs
from these predators and egg parasitoids. If a female’s proximity to her egg sac
serves primarily to permit her to chase away egg parasitoids, the pair’s stick-like
appearance places her in less jeopardy while she tends her egg sac.

Unless the colors of a female and her egg sac are similar, each would appear
more distinct and the crypsis of both would be compromised. This study tests the
hypothesis that the colors of M. animotus females and their egg sacs are linked
by determining if the colors of females and sacs are significantly correlated. It
was conducted from 20 February to 10 March, 1987 at the Center for Energy and
Environmental Research’s El Verde field station, located in the Luquillo National
Forest of Puerto Rico.

During day and night field observations, I collected a total of 94 M. animotus
females and their egg sacs. After accumulating 7-21 female-egg sac pairs, I
separated each female from her egg sac and placed them in separate vials with
matching numbers. To quantify color, I removed each spider and egg sac from its
vial, placed it directly onto the paint chips of a Naturalist’s Color Guide (NCG)
(Smithe 1975), and recorded the best color match. Specifically, I used the 1981
color dilution series 1 and 2 of sepia and raw umber pigments and series 1 of
Vandyke brown pigment. This provided 25 possible colors, 13 of which were
matched by females and egg sacs. I scored egg sac and spider color in separate
observational series conducted from 1400-1700 under natural light. If the dorsal

1989. The Journal of Arachnology 17:109

Figure 1. — A female Miagrammopes animotus (arrow) monitoring her cylindrical egg sac.

surface of a female’s abdomen was made noticeably lighter by guanine deposits, I
selected the best match of the spider’s cephalothorax, legs and lateral abdominal
regions.

Each color in the NCG has a Munsell Notation that provides range, value, and
chroma indexes. Range has ten categories extending from red through red purple
and each category has ten subdivisions. All spiders and egg sacs collected were in
the yellow-red and yellow ranges. Value indicates the “darkness” of the color as it
would appear in the absence of color vision: black is 0, white is 10. Chroma is an
open-ended scale, specifying the intensity of a color’s hue. Figure 2 presents the
distribution of these three indexes for M. animotus females and egg sacs.

Female and egg sac colors are significantly correlated as measured by all three
indices: r = 0.35 (range), r = 0.58 (value), r = 0.31 (chroma); p < 0.003. Spiders
were more reddish yellow than their egg sacs (mean spider range 8.6, SD 1.8; egg
sac mean 9.4, SD 1.4), were darker than their egg sacs (mean spider value 4.1,
SD 0.7; egg sac mean 5.2, SD 1.1), and had less highly saturated colors than their
egg sacs (mean spider chroma 2.7, SD 0.8; egg sac mean 3.2, SD 1.2). Standard
errors for spider-egg sac differences were: range 0.2, value 0.1, and chroma 0.1

Just as background selection enhances the cryptic appearance of insects and
spiders (e.g., Malcom and Hanks 1973; Opell 1986; West and Hazel 1979) and
protects them from predation (e.g., Erichsen et al. 1980; Sims and Shapiro 1983;
West and Hazel 1982), the mechanism linking spider integument and egg sac silk
pigmentation probably optimizes the protective resemblance of the pair. The
similarity of spider and egg sac color range (“color”) should make the pair less
conspicuous to predators with color vision, whereas the similarity of color value
(“grayness”) should make them less conspicuous to predators that lack color
vision or to those vertebrates whose vision relies more heavily on rods than color-
sensitive cones under the low light conditions of the forest understory (Levine
1985).

Conspicuous guanine deposits were not present in most spiders and seemed to
be more common in lighter specimens, where there was little pigmentation to
mask them. Because of their mottled nature and dorsal concentration, these

EGG SAC CHROMA EGG SAC VALUE EGG SAC RANGE

1989. The Journal of Arachnology 17:110

6

5

4

3

6 3 3

2 2 1

14 6

2

8 1

6

1

1

3

light

4 5

SPIDER VALUE

6

dark

pale saturated

SPIDER CHROMA

Figure 2. The association between female and egg sac color indexes.

1989. The Journal of Arachnology 17:111

deposits probably do not create significant abdominal color dichotomies that are
visible ventrally or laterally.

The color similarity between M animotus females and their egg sacs may be
explained either genetically or dietarily, although in spiders there seems to be no
clear evidence of the latter (Holl 1986). If metabolized prey pigments are
incorporated into both a female’s integument and egg sac silk, then feeding
history could explain this similarity. If a spider’s integumental pigmentation is
genetically determined, then these genes may also regulate the color of egg sac
silk. During this study I often collected females having greatly different colors
from the same bamboo plant or dead palm frond. Although their diets should
have been very similar, this observation does not rule out the effects of single,
heavily pigmented prey on spider color. However, it does suggest that M.
animotus do not change their colors to match their backgrounds as do some
spiders (see Holl 1986 for a review).

I thank William Pfeiffer and the staff of the El Verde field station for their
help. Craig Hieber and David Wise provided helpful comments on this
manuscript. This study was supported by a small projects grant from the College
of Arts and Science, Virginia Polytechnic Institute and State University.

LITERATURE CITED

Erichsen, J. T, J. R. Krebs and A. I. Houston, 1980. Optimal foraging and cryptic prey. J. Anim.
EcoL, 49:27L276.

Foelix, R. F. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge.

Holl, A. 1986. Coloration and Chromes. Pp 16-25, In Ecophysiology of Spiders. (W. Nentwig, ed.),

Springer-Verlag, New York.

Levine, J. S. 1985. The vertebrate eye. Pp 317-337, In Functional Vertebrate Morphology. (M.

Hildebrand et al, eds.). Harvard Univ. Press, Cambridge.

Lubin, Y. D. 1986. Web building and prey capture in Uloboridae. Pp 132-171. In Spiders: Webs,
Behavior, and Evolution. (W. A. Shear, ed.), Stanford Univ. Press, Stanford.

Lubin, Y. D,, W. G. Eberhard and G. G. Montgomery. 1978. Webs of Miagrammopes (Araneae:
Uloboridae) in the Neotropics. Psyche, 85:1-23.

Malcom, W. M. and J. P. Hanks. 1973. Landing-site selection searching behavior in the micro-
lepidopteran Agonopteryx pulvipennella. Anim. Behav., 21:45-48.

Opell, B. D. 1984. Eggsac differences in the spider family Uloboridae (Arachnida: Araneae). Trans.
American Micros. Soc., 103:122-129.

Opell, B. D. 1986. The choice of web-monitoring sites by a green Miagrammopes (Araneae:
Uloboridae). Psyche, 93:167-176.

Opell, B. D. 1987. Changes in web-monitoring forces associated with web reduction in the spider
family Uloboridae. Canadian J. Zook, 65:1028-1034.

Sims, S. R. and A. M. Shapiro. 1983. Pupal color dimorphism in California Battus philenor
(Papilionidae): Morphology factors and selective advantage. J. Lepid. Soc., 37:236-243.

Smithe, F. B. 1975. Naturalists Color Guide. American Museum of Natural History, New York.

West, D. A. and W. N. Hazel. 1979. Natural pupation sites of swallowtail butterflies (Lepidoptera:
Papilioninae): Papilio polyxenes Fabr., P. glaucus L. and Battus philenor (L.). Ecol. Entomol.,
4:387-392.

West, D. A. and W. N. Hazel. 1982. An experimental test of natural selection for pupation site in
swallowtail butterflies. Evolution, 36:152-159.

Brent D. Opell, Department of Biology, Virginia Polytechnic Institute and
State University, Blacksburg, Virginia 24061 USA.

Manuscript received April 1988, revised June 1988.

1989. The Journal of Arachnology 17:112

MEASURING THE STICKINESS OF
SPIDER PREY CAPTURE THREADS

Comparative studies of spider orb-web architecture and function make the
simplifying assumption that the stickiness of these webs’ spiral, prey capture
elements is similar. The wide range of spider size, spiral spacing, and web
tensions suggests that this may not be true and that capture thread stickiness is
an important functional component of web design. Just as the length of capture
silk provides an index of material outlay and spiral spacing influences the size of
prey captured (Eberhard 1986), silk stickiness determines the web’s ability to
retain prey until they can be subdued by the spider.

Using a simple balance device, Eisner et al. (1964) compared the retention
ability of Nephila clavipes (Linn.) viscid thread for naked insect wings with those
beset with hairs and scales. However, the stickiness of different species’ capture
threads have not been compared. To study Uloboridae threads, I designed a
simple device for measuring thread stickiness that may be useful to other
investigators.

This instrument (Fig. 1) employs a glass needle strain gauge and is similar to
the devices used by Craig (1987) for measuring silk tensile strength and Opell
(1987) for comparing the web-monitoring forces expressed by spiders. The
instrument described here was fabricated from 6 mm thick plexiglass.

Silk samples are collected from a spider’s web using a microscope slide, along
whose length five 4 mm wide rectangular brass supports are glued at 4 mm
intervals. Double-sided tape applied to the top of each support holds the silk
under its original tension. This collecting device permits four replicate samples to
be taken from each web and the close spacing of its supports minimizes silk
extension during the process.

Two clips secure this collecting slide to a sliding platform that permits
positioning of each replicate cribellar silk sample under a narrow aluminum
contact plate. The latter is glued with epoxy to the tip of a fine glass needle
drawn from a hematocrit tube. Thumb screws secure the sliding plate after each
sample has been positioned.

The frame holding the calibrated needle can be raised and lowered relative to
the microscope slide with a small screw jack. After a small, standard downward
force is exerted on the contact plate, the strain gauge is slowly raised until the
contact plate attached to the needle’s tip pulls free of the capture silk. The
position of the needle on the strain gauge’s arbitrary scale immediately before the
plate pulls free of the thread is noted. When converted to its milligram equivalent
using a calibration graph and multiplied by the accelerating force of gravity, this
value yields the force in Newtons necessary to overcome the thread’s adhesion.

Although the frame is raised by a hand-operated jack, the need to continually
observe the glass needle’s position on the scale keeps the velocity of silk loading
low and fairly constant. In the example given below, scale unit spacing was 632
pm and, in order to record to the nearest half unit the point at which the contact
plate pulled free of the silk, the silk was loaded at a velocity of 287-316 pm per
second. This velocity is probably less than half that of a struggling prey’s
appendages. The rate of loading could be more precisely controlled by driving the
jack with a small motor, although a greater velocity than that listed above would

1989. The Journal of Arachnology 17:113

Figure 1. — The instrument used to measure capture silk adhesion. A = arbitrary scale, B =
mounting bracket for glass needle, C = clamps for securing microscope slide silk sampler, J = screw
jack for raising strain gauge, N = glass needle with contact plate near its free (left) end, P = pivot of
strain gauge, S = parallel stabilizing bars to reduce lateral movement of strain gauge, SP = sliding
platform for positioning silk samples.

make it difficult to accurately determine visually the point at which the contact
plate pulled free of the silk.

To test the reliability of this technique I measured the stickiness of cribellar silk
from 21 mature female Hyptiotes cavatus (Hentz) webs. All webs were spun in
the laboratory and the stickiness of their silk was measured 1-3 days after it was
produced. A 2.20 mm wide polished aluminum contact plate was first pressed
against each silk sample with a force of 3.03 x 10~^ Newtons. The mean stickiness
value of these 84 measurements was 3.251 x 10'^ N / mm of contact (SD = 6.7 x
lO’"^ N / mm, CV = 2.06). The mean of each web’s average stickiness value
(7V=21) had a SD of 4.4 x 10"^ N / mm and a CV = 1.34. Comparative values
for the four replicate measurements of each thread ranged from 0 to 5.76 and had
a mean of 1.30.

Although cribellar silk leaves no residue on the aluminum contact plate, viscid
threads do and the plate must be cleaned with acetone or other solvent before
each measurement. A length of dragline thread serves as both a control and a test
for a clean contact plate, as the plate should not stick to it.

Gabrielle Roth helped with measurements. John Anderson and Ruth Buskirk
provided helpful comments on this paper. This technique was developed during
studies supported by NSF grant BSR-8407979 and by a small projects grant from
Virginia Tech’s College of Arts and Sciences.

LITERATURE CITED

Craig, C. L. 1987. The ecological and evolutionary interdependence between web architecture and web
silk spun by orb-web weaving spiders. Biol. J. Linnean Soc., 30:135-162.

Eberhard, W. G. 1986. Effects of orb-web geometry on prey interactions. Pp 70-100, In Spiders: Webs,
Behavior, and Evolution. (W. Shear, ed.), Stanford Univ. Press, Stanford.

1989. The Journal of Arachnology 17:114

Eisner, T., R. Alsop and G. Ettershank. 1964. Adhesiveness of spider silk. Science, 146:1058-1061.
Opell, B. D. 1987. Changes in web-monitoring forces associated with web reduction in the spider
family Uloboridae. Canadian J. Zook, 65:1028-1034.

Brent D. Opell, Department of Biology, Virginia Polytechnic Institute and
State University, Blacksburg, Virginia 24061 USA.

Manuscript received April 1988, revised June 1988.

ON A NEW SPIDER OF THE GENUS DRASSYLLUS
(ARANEAE, GNAPHOSIDAE) FROM FLORIDA

A dozen species of the zelotine gnaphosid genus Drassyllus have been reported
from Florida (Platnick and Shadab, 1982, Bull. Amer. Mus. Nat. Hist., 173:1-97),
among which two, D. seminolus Chamberlin and Gertsch and D. alachua
Platnick and Shadab, appear to be endemic to the state. We report here on an
additional species collected from a pine flatwoods plant community in the St.
Johns River drainage. We thank the Exline-Frizzell Fund for Arachnological
Research of the California Academy of Sciences for supporting the field work
that lead to the discovery of this species, and Dr. M. U. Shadab of the American
Museum of Natural History (AMNH) for help with the illustrations. The format
of the description follows that used in the generic revision.

Drassyllus orlando, new species
Figs. 1-4

Types. — Male holotype and female allotype taken in a pitfall trap on the
University of Central Florida campus, 12 miles east of Orlando, SIO R31E T22S,
Orange Co., Florida (25 May 1983; D. T. Corey), deposited in AMNH.

Etymology. — The specific name is a noun in apposition taken from the type
locality.

Diagnosis. — Males can be distinguished from those of all other American
Drassyllus by the retrolateral depression on the palpal tibia (Fig. 2), females by
the widely separated median and anterior epigynal ducts (Fig. 4).

Male. — Total length 3.19-3.58. Carapace 1.43-1.50 long, 1.14-1.20 wide. Femur
II 0.94-0.98 long (four specimens). Eye sizes and interdistances: AME 0.06, ALE
0.09, PME 0.10, PEE 0.09; AME-AME 0.05, AME-ALE 0.01, PME-PME 0.01,
PME-PLE 0.03, ALE-PLE 0.03; MOQ length 0.22, front width 0.17, back width
0.21. TA small, lobular, EP prolonged retrolaterad of EMB (Fig. 1); RTA bent at
right angle above glabrous depression on retrolateral surface of tibia (Fig. 2). Leg
spination typical for genus.

Female. — Total length 2.87. Carapace 1.37 long, 1.11 wide. Femur H 0.87 long.
Eye sizes and interdistances: AME 0.05, ALE 0.07, PME 0.10, PLE 0.09; AME-
AME 0.06, AME-ALE 0.01, PME-PME 0.01, PME-PLE 0.03, ALE-PLE 0.04;
MOQ length 0.20, front width 0.16, back width 0.21. MP almost square (Fig. 3);

1989. The Journal of Arachnology 17:115

Figures 1-4. — Drassyllus orlando, new species: 1, palp, ventral view; 2 palp, retrolateral view; 3,
epigynum, ventral view; 4, epigynum, dorsal view.

MED and AED widely separated (Fig. 4). Leg spination: femora: II pO-0-0; IV
pO-0-0.

Other material examined. — Three males taken with the types (AMNH).

Distribution. — Known only from the type locality.

Norman I. Platnick, Department of Entomology, American Museum of
Natural History, Central Park West at 79th Street, New York, New York 10024
USA; and David T. Corey, Department of Biological Sciences, University of
Central Florida, Orlando, Florida 32816 USA.

Manuscript received May 1988, revised June 1988

SPERM DEPLETION IN THE ORB- WEAVING SPIDER
NEPHILA CLAVIPES (ARANEAE, ARANEIDAE)

In the spider literature, it is generally assumed that individual males can and do
inseminate more than one female. This would appear to be adaptive from the
male point of view, particularly since spider males invest nothing in care of egg
sac or young, and, in all three species examined so far, it seems that the first male
to mate with a given female fertilizes most of her eggs (Austad 1984; Christenson
and Cohn 1988). If a male defended a juvenile female for a period of time and
then mated with her just after her final molt, he might well be expected to search
for and mate with another female. However, the data showing that male spiders
can inseminate that second female are sparce. Breene and Sweet (1985) have

1989. The Journal of Arachnology 17:116

demonstrated that male black widow spiders can inseminate at least three
females. In this study, the limitation on insemination was determined by placing
male Nephila clavipes (L.) with up to three sexually receptive virgin females and
later examining both sexes for the presence of sperm. It would seem that any
assessment of male tactics, in any species, would benefit from taking into account
such information.

A number of ante-penultimate females (18-21 mm cephalothorax-abdomen
length) and newly-molted adult males (7-9 mm cephalothorax-abdomen length)
were gathered in July of 1986. To insure males were virgin, I gathered only those
on orbs with sperm webs and with the newly-molted animal’s coloration (see
Myers and Christenson 1988). All subjects were housed in 123 x 62 x 62 cm
boxes screened with Fiberglas®. These were situated in the hardwood forest at
the F. Edward Hebert Center of Tulane University, located about 20 km outside
of New Orleans, La. Females were fed two mealworm larvae per day.

Males were held in the field boxes, in male groups, for a few days to a week
prior to presentation to a female. About two days before the female’s molt,
determined by pattern of web deterioration, each virgin male {n = 11) was placed
in the upper barrier strands of a female’s orb. After the female’s molt, mating
behavior was time sampled at about ten-minute invervals for two hours. This is
sufficient observation to determine the nature of mating as copulations are
frequent after the female’s final molt, occurring an average of 28 minutes out of
the hour for more than a day (Christenson et al. 1985). Males were removed from
the first female’s web at different times, ranging from 5 to 35 days after the
female’s molt. Time of removal varied in order to increase the longevity of some
males, and to allow some males (n = 3) to mate with three females, and as a
consequence of availability of females who were ready to molt. After male
removal, first females {n - 11) were sacrificed, and the spermathacae removed,
squashed, and examined with a phase contrast microscope for sperm. Female
maturity was verified for each female by the sclerotized condition of the
spermathecae.

Immediately after removal from the first female’s cage, each male {n ~ 11) was
placed with a second female in the manner already described. Second females
molted anywhere from 10 to 34 days after the first female. On the day of second
female molt, mating behavior was observed during a one-hour serial record, to
give a more accurate account of mating frequency and time. Eight males were
removed and the palps checked for sperm. Second females {n = 11) were removed
and spermathecae checked in the manner already described.

The remaining three males were placed with a third female; those females
molted 47, 51, and 53 days after the first females. Mating was observed for one
hour after the female’s molt. The three males and females were removed a few
days after her molt and were checked for sperm. So, the eleven males had been
sacrificied and checked for sperm between 2 and 9 weeks after molting to
adulthood. This is about as long as adult male N. clavipes live under these
captive conditions (Cohn and Christenson 1987) and probably much longer than
they live under unrestrained conditions. To be included in the following analysis,
the male and all females he mated with had to have been examined for sperm.
Due to deaths and a few possible male escapes, data for an additional 8 males
and 15 females were excluded.

1989. The Journal of Arachnology 17:117

Nine of the eleven males were observed to mate with the first female. Mating
was of the usual vigor noted in N. clavipes just after the female’s final molt
(Christenson et al. 1985); mating was noted in 60% of the time samples and bulb
contraction frequencies averaged 41 per min. Insertions of the conductor were
numerous, and prolonged hematodochal bulb contractions were evident. All of
the eleven first females’ spermathacae contained many sperm.

Seven of the eleven males demonstrated some level of sexual behavior with the
second female. However, the vigor of mating was relatively low. Prolonged
conductor insertion with rhythmic and countable hematodochal bulb contractions
occurred in only two cases. In the first, the male mated eight times for a total of
16 minutes; the bulb contraction rate was 30 per min. In the second, the male
mated three times for almost 20 minutes; bulb contraction rate was 8 per min.
The matings of five males included an average of seven brief conductor insertions
with intromittent bulb contractions, the frequency of which were impossible to
count. None of the 1 1 second females’ spermathecae contained sperm.

None of the males with a third female showed any mating behavior. Third
females contained no sperm. Eventually, all 11 males tested were sacrificed; none
of their palps contained sperm. No sperm webs were noted in any female orb.

One can not take for granted that an active male spider can inseminate more
than one female. Robinson and Robinson (1980) noted that eunuchs, males with
terminal portions of the palps broken off, still actively fight with one another
while on a female’s orb. Although embolus damage is commonly thought to
reduce or eliminate the ability to transfer sperm (Foelix 1982), it should be
pointed out that Breene and Sweet (1985) found that even with damage to the tip
of the embolus, male black widows were capable of inseminating the female.
When a male N. clavipes mates, that does not necessarily mean that the male has
sperm available to the female. Two males in this study demonstrated mating with
a second female that could easily have been confused with mating typical of the
day after the female’s final molt (Christenson et al. 1985). The intermittent
insertions and bulb contractions noted in five other males could have been
confused with mating typically noted with older adult females (ibid.).

No sperm were found in the palps of any of the 11 males. This was somewhat
surprising; we had suggested at one time that because males near the hub (and
thus the female) had a feeding advantage over males remaining on the web
periphery, the hub male would have a relatively greater chance of moving to and
then mating with a second female (Christenson and Goist 1979). Anyway, it is
unclear why no sperm were available. Perhaps spermatogenesis is terminated at
the final molt, or there is some constraint on sperm transfer via sperm web
construction or via prolonged copulations (up to 48 hours; Christenson et al.
1985) leading to embolus damage which in this species might inhibit sperm
transfer. These possibilities are being examined histologically.

One might expect males of N. clavipes to be somewhat choosey concerning
females that could be assessed at the time of their final molt. I do not think that
males have an ample opportunity to do so. First, our boxed females do not
attract males, indicating that molting females do not appear to produce a
distance-acting pheromone, as can happen in other species (Olive 1982). Second,
males have a limited opportunity to visit adult females inhabiting aggregated
webs. Female N. clavipes aggregate later in the season, after having matured and

1989. The Journal of Arachnology 17:118

mated for the first time (Brown et al, 1985). Further study of male choice is,
however, warranted.

The discussion of alternative mating tactics in spiders must include limitations
on insemination. Consider what is perhaps the most fundamental decision a male
orb-weaving spider has to make, to remain on a female web or to search for
another female. Austad (1984) has argued that phyletic constraints, such as the
structure of the female sperm uptake and storage anatomy, may be closely related
to whether the male should guard the female. Results of the present study
indicate that constraints relating to insemination limits should also be considered.
Male N. clavipes deplete sperm after mating with a newly-molted female but do
not leave to search. Most remain with the female after mating (Cohn et at. 1988).
There is no compelling reason for males to leave their mate and search for
another because they could not father additional offspring by doing so. The
comparative study of spider species whose males routinely show limited guarding
and frequent movement between females would clarify the importance of
insemination limitation in the prediction of male behavioral tactics.

This study was supported by NSF grant BNS-83 17988. The author extends
thanks to M. Robinson for his review and S. Austad for his review and pointing
out a critical reference.

LITERATURE CITED

Austad, S. 1984. Evolution of sperm priority patterns in spiders. Pp. 223-249, In Sperm Competition
and the Evolution of Animal Mating Systems. (R. L. Smith, ed). Academic Press, New York.

Breene, R. G. and M. H. Sweet. 1985. Evidence of insemination of multiple females by the male black
widow spider, Latrodectus mactans (Araneae, Theridiidae). J. Arachnol., 13:331-335.

Brown, S. G., E. M. Hill, K. C. Goist, P. A. Wenzl and T. E. Christenson. 1985. Ecological and
seasonal variations in a free-moving population of the golden-web spider, Nephila clavipes. Bull.
British Arachnol. Soc., 6:313-319.

Christenson, T, S. G. Brown, P. A. Wenzl, E. M. Hill and K. C, Goist. 1985. Mating behavior of the
golden orb-weaving spider, Nephila clavipes: I. Female receptivity and male courtship. J. Comp.

Psych., 99:160-166.

Christenson, T and J. Cohn. 1988. Male advantage for egg fertilization in the golden orb-weaving
spider, Nephila clavipes. J. Comp. Psychol., 102:312-318.

Christenson, T. and K. C. Goist 1979. Costs and benefits of male-male competition in the orb-weaving
spider, Nephila clavipes. Behav. Ecol. SociobioL, 5:87-92.

Cohn, J., F. Balding, and T. Christenson. 1988. In defense of Nephila clavipes: Postmate guarding by
the male golden orbweaving spider. J. Comp. Psychol, 102:319-325.

Cohn, J. and T. Christenson. 1987. Utilization of resources by the male golden orb-weaving spider
Nephila clavipes (Araneae). J. Arachnol, 15:185-192.

Foelix, R. F. 1982. Biology of Spiders. Harvard University Press.

Myers, L. & T. Christenson. 1988. Transition from predatory juvenile male to mate-searching adult in
the orb-weaving spider Nephila clavipes (Araneae, Araneidae). J. Arachnol. 16:254-257.

Olive, C. 1982. Sex pheromones in two orb-weaving spiders, (Araneae, Araneidae): An experimental
field study. J. Arachnol, 10:241-245.

Robinson, M. H. and R. Robinson. 1980. Comparative study of the courtship and mating behavior of
tropical araneid spiders. Pacific Insects Monogr. No 36. Bishop Museum: Honolulu.

Terry Christenson, Department of Psychology, Tulane University, New
Orleans, Louisiana 70118 USA.

Manuscript received May 1988, revised July 1988.

1989. The Journal of Arachnology 17:119

FIELD OBSERVATIONS OF GASTERACANTHA CANCRIFORMIS
(ARANEAE, ARANEIDAE) IN A FLORIDA MANGROVE STAND

G aster acantha cancriformis (L.), sometimes known as the spiny-bellied
orbweaver, is a tropical-subtropical spider in the Americas (Levi and Levi 1968).
The biology of this spider is discussed in limited detail by Comstock (1940),
Gertsch (1979), and Kaston (1978). Muma (1971) presents detailed observations
on the seasonal history, web construction, mating, egg case construction, and
spiderling activity. However, most of Muma’s (1971) behavioral observations were
made in the laboratory, and he does not quantify web components. This note
quantifies the spider’s web components and describes some of its field behavior
and ecology.

Observations were conducted in 1975 from June to Sept, at the west end of the
Howard Franklin Bridge and Fla. Hwy 688, Pinellas Co., Florida, in a stand of
mixed mangroves, Avicennia germinans (L.) L., and Rhizophora mangle L.
Approximately 58 h were spent in observing 41 subadults and adult females. Web
component nomenclature is based on Savory (1977).

The 3c ± SD {n) of various web characteristics were as follows: web diameter,
29 cm + 6 cm (19); web angle, 71.3° ± 11.5° (20); spirals, 37.4 ± 7.5 (18); radii,
27.4 + 6.5 (20); frame threads, 2.8 ± 0.9 (13); mooring threads, 3.8 ± 1.0 (19);
bridge length, 1.1 m ± 0.5 m (23); radii with flocculent silk tufts, 2.9 ± 1.8 (26);
split radii with flocculent silk tufts, 1.0 ± 0.0 (3); frame thread with reinforced
silk layers, 2.2 ± 0.6 (11); and mooring thread with reinforced silk layers, 2.0 ±
0.0 (2). Muma (1971) reported 20 to 30 radii which agrees with my findings. He
reported also 10 to 30 spirals which is slightly less than what I observed. Levi and
Levi (1968) showed flocculent silk tufts on the spiral and radii. Neither Comstock
(1940), Muma (1971), nor I found these tufts on spirals. These tufts, or silken
adornments, are probably protective devices that warn birds or large insects of a
web’s presence (Ewer 1972; Eisner and Nowicki 1983).

Muma (1971) reported that webs are constructed at an angle perpendicular to
the ground, which I noted also. Females occupied the acute side of the incline
and were positioned at the hub so that their black ventral surface faced up and
their colored dorsal surface faced down. This may have been a counter-shading
device or a thermoregulatory response.

On three separate occasions all spiders dismantled their webs just before or
during rain, usually leaving from half of the web to only the bridge. This
response to rain is similar to that of Nephila maculata (Fab.) (Robinson and
Robinson 1973). When dismantling the web, C. cancriformis collected the silk
and pressed it to her mouth, where she rolled it into a lump. The silk color
changed from white to brown. Afterwards, the brownish lump was placed onto a
part of the old web, on vegetation, or dropped. This color change suggests that
rather than simply removing moisture from the silk, the spider may have
extracted nutrients before discarding the silk.

Prey items included three mosquitoes (Culicidae), a horsefly (Tabanidae), a
moth (Lepidoptera), two love bugs (Bibionidae), a small unidentified spider, an
antiion (Myrmeleontidae), three honey bees (Apidae), and a fly (Muscidae). One
female was observed drinking water after a rain storm by stroking her ventral
abdomen with her left fourth leg and pressing the leg to her mouth; she did this

1989. The Journal of Arachnology 17:120

twice. One spider ate the pollen from the corbiculla of a bee’s leg. Juvenile orb-
weaving spiders eat pollen (Smith and Mommsen 1984), but this is the first
account of an adult spider eating pollen.

The author thanks Ronnie Byford, Audry Cashio, Colwell Cook, and Amy
Korman for their reviews of this manuscript, and E. A. Heinrichs for covering
publication cost. Approved for publication by the Director of the Louisiana
Agricultural Experiment Station as manuscript number 88-17-2253.

LITERATURE CITED

Comstock, J. H. 1940. The Spider Book. Comstock Publishing Co., Ithaca, New York.

Eisner, T and S. Nowicki. 1983. Spider web protection through visual advertisement: Role of
stabilimentum. Science, 219:185-186.

Ewer, R. F. 1972. The devices in the web of the West African spider Argiope flavipalpis. J. Nat. Hist.,

6:159-167.

Gertsch, W. J. 1979. American Spiders, 2nd ed. Van Nostrand Reinhold, New York.

Kaston, B. J. 1978. How to Know the Spiders, 3rd ed. Wm. C. Brown, Dubuque, Iowa.

Levi, H, W. and L. R. Levi. 1968. A Guide to Spiders and Their Kin. Golden Press, New York.

Muma, M. H. 1971. Biological and behavioral notes on Gasteracantha cancriformis (Arachnida:
Araneidae). Florida Entomol., 54:345-351.

Robinson, M. H. and B. Robinson. 1973. Ecology and behavior of the giant wood spider Nephila
maculata (Fabricius) in New Guinea. Smithson Contr. Zook, 149:1-75.

Savory, T. 1977. Arachnida, 2nd ed. Academic Press, New York.

Smith, R. B. and T. P. Mommsen. 1984. Pollen feeding in an orb-weaving spider. Science, 226:1330-
1332.

Ben M. Gregory, Jr., Department of Entomology, Louisiana Agricultural
Experiment Station, Louisiana State University Agricultural Center, Baton

Rouge, Louisiana 70803 USA.

Manuscript received April 1988, revised September 1988.

SPIDER PREDATION ON VELVETBEAN CATEPILLAR MOTHS
(LEPIDOPTERA, NOCTUIDAE) IN A SOYBEAN FIELD

The velvetbean caterpillar (VBC), Anticarsia gemmatalis Htibner, is a major
pest of soybean {Glycine max [L.] Merrill) in the Gulf Coast area of the United
States (Herzog and Todd 1980). Little is known about predators of VBC adults:
Watson (1915, 1916) reported dragonflies in Florida but gave no names; Neal
(1974) listed the green jacket dragonfly, Erythemis simplicicollis (Say), and the
striped earwig, Labidura riparia (Pallas), in Florida; and Gregory (1987) reported
the spider Eriophora edax (Blackwall) in Honduras. This note provides further
documentation of VBC adult predation by spiders.

Observations were conducted by the senior author from 1981-1983 at the
University of Florida’s Green Acres Research Farm, Alachua County, FL. The
study site was an approximately 1 ha soybean field (variety Bragg). Most
observations were recorded to the nearest minute. Approximately 354 h were

1989. The Journal of Arachnology 17:121

Table I. — Records of spider predation on adult velvetbean caterpillar (VBC) at the Green Acres
Research Farm, Alachua County, FL (1981-1983). If exact time not given, then predation record
occurred during hyphenated times. E = field edge (+1 m), I = in field. Grass = unidentified grass,
bahiagrass = Paspalum notatum Flugge, hairy indigo = Indigofera hirsuia L., soybean = Glycine max
(L.) Merr., beggarweed = Desmodium tortuosum (Sw.) DC., sicklepod = Cassia obtusifolia L., Florida
pusley = Richardia scabra L., sandbur = Cenchrus sp. With regard to VBC sex, A = undetermined, M
= male, and F = female.

Species

Date

Time

Location

Substrate

VBC sex

Peucetia viridans (Hentz)

19 Sep. 81

2047

E

Bahiagrass

M

19 Sep. 81

2058

I

Soybean

M

24 Sep. 82

2112

I

Florida pusley

M

17 Sep. 81

2115

E

Bahiagrass

F

19 Sep. 83

2130

E

Sandbur

M

24 Sep. 82

2147

I

Soybean

F

24 Sep. 82

2223

I

Soybean

F

09 Sep. 81

2303

I

Soybean

M

03 Sep. 82

2323

E

Beggerweed

M

03 Sep. 82

2340

I

Sicklepod

M

25 Sep. 82

0023

I

Hairy indigo

M

04 Sep. 82

0052

I

Soybean

M

04 Sep. 82

0111

I

Sicklepod

M

04 Sep. 82

0136

I

Soybean

M

21 Aug. 81

0545-0701

I

Soybean

A

25 Aug. 81

0545-0703

E

Grass

M

15 Sep. 81

0545-0714

I

Soybean

M

Misumenops celer (Hentz)

03 Sep. 81

2230

E

Bahiagrass

M

16 Aug. 81

0000-0700

E

Grass

M

09 Oct. 82

0530

E

Hairy indigo

M

01 Sep. 81

0545-0607

E

Grass

F

Misumenoides formocipes (Walck.)

15 Sep. 81

0545-0714

E

Bahiagrass

M

Eriophora ravilia (C. L. Koch)

17 Sep. 81

2100

I

Soybean

M

Neoscona arabesca (Walck.)

24 Sep. 82

2332

E

Soybean

M

05 Oct. 82

0550

E

Soybean

F

Acanthepeira sp.

15 Sep. 83

0200

E

Bahiagrass

F

spent in the field during July through October, with 201 h during the day and 153
h during the night. Data were gathered by walking through the study site in a
systematic fashion. Small sections of the field, and the associated arthropods,
were disturbed during the day in the completion of other experiments; this may
have reduced the number of observed kills. A six-volt Everready® Freedom
Light™ was used for nocturnal observations; the lighting fixture was covered with
a section of Zip-a-Tone color sheet, Vermillion Hue #2545.

Six species of spiders are documented for the first time as predators of VBC
adults (Table 1). Ail 26 instances of predation were observed at night. The green
lynx spider, Peucetia viridans (Hentz), accounted for 65% of the records; crab
spiders (Misumenops, Misumenoides) for 19%; and orbweavers {Acanthepeira,
Eriophora, Neoscona) for 15%. Eighty-two percent of records for P. viridans
occurred between 2047 and 0136 hours, the time when VBC adults are most
active (Gregory 1986). Peucetia viridans was found throughout the field, usually
on dicotyledons and high above the ground (ca 1 m or higher). Crab spiders were
found only at the field edge, usually on monocotyledons and close to the ground
(ca 0.5 m or less). Most orbweavers were found at the field edge.

1989. The Journal of Arachnology 17:122

Twenty of the 26 captured moths were VBC males, a significant prey bias
(x^ = 7.84, df-l, P<0.01). Velvetbean caterpillar adult sex-ratio was ca 1:1 at the
study site (Gregory 1986). The nature of the prey bias may be due to aggressive
chemical mimicry of VBC mating pheromone by some of these spiders (see Foelix
1982). Based on the predation records, spiders were the principal mortality agents
of VBC adults; no other predators were observed at the study site.

This research was supported by EPA/SEA grant, Development of Comprehen-
sive, Unified, Economically, and Environmentally Sound Systems of Integrated
Pest Management. The authors thank Tad Hardy, Joon Ho Lee, and Fred
Whitford for their reviews of this manuscript. Spider vouchers are at the Florida
State Collection of Arthropods, Gainesville, FL. Florida Agricultural Experiment
Station Journal Series No. 8895.

LITERATURE CITED

Foelix, R. F. 1982. Biology of Spiders. Harvard University Press, Cambridge, Massachusetts.

Gregory, Jr., B. M. 1986. A model of adult and egg populations of Anticarsia gemmataiis Hiibner
(Lepidoptera: Noctuidae) in soybean. Ph.D, Dissertation, University of Florida, Gainesville,
Florida.

Gregory, Jr., B. M. 1987. Observations on velvetbean caterpillar (Lepidoptera: Noctuidae) in a
hyacinth bean field in Honduras. Florida Entomol., 70:296-297.

Herzog, D. C. and J. W. Todd. 1980. Sampling velvetbean caterpillar on soybean. Pp. 107-140, In
Sampling Methods in Soybean Entomology. (M. Kogan and D. C, Herzog, eds.) Springer-Verlag,
New York,

Neal, T. M. 1974. Predaceous arthropods in the Florida soybean agroecosystem. M.S. Thesis.
University of Florida, Gainesville, Florida.

Watson, J. R. 1915. The velvet bean catepillar {Anticarisa gemmataiis). Florida Agr. Exp, Stn. Ann.
Rep., 1914/1915:49-64.

Watson, J. R. 1916. Control of the velvet bean catepillar. Florida Agr. Exp. Stn. Bull, 130:46-58.

Ben M, Gregory, Jr,^ and Carl S* Barfield, Department of Entomology and
Nematology, University of Florida, Gainesville, Florida 32611 USA, and G. B.
Edwards, Florida State Collection of Arthropods, Division of Plant Industry,
FDACS, P.O. Box 1269, Gainesville, Florida 32602 USA.

'Current address: Department of Entomology, Louisiana State University, Baton Rouge, Louisiana
70803 USA.

Manuscript received April 1988, revised September 1988.

WATEM-MESISTANT SEX PHEROMONES
IN LYCOSID SPIDERS FROM A TROPICAL WET FOREST

The role of female sex pheromones for pre-copulatory communication in
lycosid spiders has been studied in Nearctic and Palearctic species that inhabit
deciduous forest and grassland biomes (review in Tietjen and Rovner 1982).
Substances involved are often bound to the silk dragline secreted by the female as

1989, The Journal of Arachnology 17:123

she wanders (Engelhardt 1964; Tietjen 1977; Tietjen and Rovner 1980), deposited
on the substratum (Bristowe and Locket 1926; Richter et al. 1971), or released
into the air (Tietjen 1979). The findings on another family of wandering spiders,
the Salticidae, recently have been reviewed (Pollard et al. 1987).

In the iycosids studied so far, the substances serving as contact sex pheromones
are inactivated (i.e., become ineffective for stimulating males) by water (Dondale
and Hegdekar 1973; Tietjen 1977). Consequently, rain or dew can limit the use of
such chemical signals in these spiders. On the other hand, Dolomedes triton
(Walckenaer), a member of the closely related family Pisauridae, produces a sex
pheromone that is effective on water, reflecting the special needs of this semi-
aquatic spider (Roland and Rovner 1983).

In Iycosids that inhabit a tropical wet forest life zone, where there is nightly
dew and frequent rainfall, any substrate-deposited or silk-bound pheromone that
could be effective for some hours would be expected to resist inactivation by
water. We tested this hypothesis in two species of Iycosids, Lycosa tristani Banks
(at one time placed in the genus Schizocosa) and Lycosa longitarsis F. Pickard-
Cambridge (a member of the Lycosa helluo species group). (Voucher specimens of
both species are deposited at the Biosystematics Research Centre in Ottawa).
Both Iycosids were abundant when collected at night as they rested waiting for
prey in a mown clearing at the Organization for Tropical Studies’ La Selva
research station near Puerto Viejo de Sarapiqui, Heredia Province, Costa Rica.
Research was conducted from mid-November through late December, 1987.

While male spiders placed in vacated female cages usually showed courtship
display, those placed in the vacated cages of females which had their spinnerets
sealed with paraffin rarely did so (Table 1). Presumably, the female pheromone
was bound to the silk, not deposited directly on the substrate by tarsal or body
contact. There was little or no courtship response by males to the vacated cages
of other males {L. tristani, 1/10; L. longitarsis, 0/10).

Next, males were tested with female draglines subjected to various treatments.
Tethered females were led along a cardboard track, traversing three glass rods (5
mm diameter) placed 7 cm apart and perpendicular to the path of the spider. The
dragline began at an attachment disk and was fastened at the other end with
adhesive tape. The taut and slightly elevated dragline was then either untreated,
allowed to age for 1 day, misted with water and allowed to dry, submerged in
water and allowed to dry after removal of the water from the tray, misted with
ethanol and allowed to dry, or misted with hexane and allowed to dry. (Drying
time was at least 0.5 h in all cases.) Male L. longitarsis also were tested with
untreated male draglines obtained in a similar manner to those of females. (Male
draglines of L. tristani, a much smaller lycosid, were not tested because the males
did not leave a sufficiently thick dragline as they were led along the track. Female

Table 1. — Occurrences of male courtship behavior in vacated cages of female conspecific lycosid
spiders with or without sealed spinnerets. Significance of differences: G-test with Yates’ correction. *P
<0.05; ***P< 0.001.

Spinnerets

Lycosa longitarsis

Lycosa tristani

Sealed

2/20

0/10

G = 5.820*

G = 19.782***

Not sealed

6/10

10/10

1989. The Journal of Arachnology 17:124

Table 2. — Number of males showing courtship display and dragline-following in response to
untreated or treated conspecific lycosid draglines. Values having asterisks are significantly different at
the 0.05 (*), 0.01 (**), or 0.001 (***) levels from the within-column value for untreated female
draglines (G-test with Yates’ correction). tV = 10 males of each species/ test.

Dragline

Treatment

Lycosa longitarsis

Lycosa tristani

Court

Follow

Court

Follow

Female

Untreated

7

7

9

10

Female

1-day old

2

3

4

5*

Female

Water spray

5

8

10

7

Female

Underwater

6

8

—

—

Female

Ethanol spray

2

3

7

8

Female

Hexane spray

1*

1*

1**

1***

Male

Untreated

0**

0**

—

draglines of L. tristani were not tested after being submerged, since they broke
too easily when the water was removed from the tray.)

The results of the above-listed tests are given in Table 2. The two lycosids not
only responded to the female pheromone by courting, but also showed dragline-
following, as in previously studied lycosids. (Neither response occurred when
draglines of male L. longitarsis were tested.) Loss of stimulatory efficacy with
time, perhaps due to volatility of the pheromone, was suggested by lower male
responsiveness toward 1 -day-old lines, significant in one category. Subjecting the
female draglines to water — even submergence — did not reduce the response levels
of the males significantly, nor did spraying the lines with ethanol. Male
reponsiveness in both species was reduced significantly after the lines were
sprayed with hexane, which, unlike water or ethanol, adhered to the silk.
Apparently, the dragline-bound pheromones of these lycosids are non-polar
compounds (being inactivated by hexane) and probably are not washed off or
inactivated by the nightly dew and frequent rain that typifies the tropical wet
forest life zone.

Our observations that penultimate male and female L. longitarsis constructed
dome-shaped nests in their cages before molting led to additional study of this
species. (Nest construction also was observed in some adult females with, as well
as without, egg sacs, although these nests appeared less dense than those used by
penultimate instars for molting.) We tested the nests of penultimate females for
possible stimulatory value by placing adult males in the vacated cages of such
females. Seven of 20 males tested showed courtship after contacting the nest. This
was not significantly different from the response level (6/10) shown by these
males to the vacated cages of adult females {G - 0.688). The construction of nests
with a sex pheromone bound to the silk is a behavior not reported before in
subadult lycosids, although it is well known in other families of wandering spiders
(Tietjen and Rovner 1982; Pollard et al. 1987). The presence of female sex
pheromone in penultimate female L. longitarsis also was suggested by the
attempts of four out of ten adult males to copulate with freshly killed penultimate
females.

We thank the Organization for Tropical Studies for making their facilities at
La Selva available to us. Charles D. Dondale and Allen R. Brady were kind
enough to provide taxonomic assistance, Robert B. Suter and William J. Tietjen

1989. The Journal of Arachnology 17:125

suggested improvements in the manuscript. This study was supported by a grant

from the Ohio University Research Committee to the junior author.

LITERATURE CITED

Bristowe, W. S. and G. H. Locket. 1926. The courtship of British lycosid spiders and its probable
significance. Proc. Zool. Soc. London, 1926:317-347.

Dondale, C. D. and B. M. Hegdekar. 1973. The contact sex pheromone of Pardosa lapidicina
Emerton (Araneida: Lycosidae). Canadian J. Zool., 51:400-401.

Engelhard!, W. 1964. Die mitteleuropaischen Arten der Gattung Trochosa C. L. Koch, 1848 (Araneae,
Lycosidae). Z. Morph. Okol. Tiere, 54:219-392.

Pollard, S. D., A. M. MacNab and R. R. Jackson. 1987. Communication with chemicals: pheromones
and spiders. Pp. 133-141, In Ecophysiology of Spiders (W. Nentwig, ed.). Springer-Verlag, New
York.

Richter, C. J. J., H. C. J. Stolting and L. Vlijm. 1971, Silk production in adult females of the wolf
spider Pardosa amentata (Lycosidae, Araneae). J. Zool, London, 165:285-290.

Roland, C. and J. S. Rovner. 1983. Chemical and vibratory communication in the aquatic pisaurid
spider Dolomedes triton. J. Arachnol., 11:77-85.

Tietjen, W. J. 1977. Dragline-following by male lycosid spiders. Psyche, 84:165-178.

Tietjen, W. J. 1979. Is the sex pheromone of Lycosa rabida (Araneae: Lycosidae) deposited on a
substratum? J. Arachnol., 6:207-212.

Tietjen, W. J. and J. S. Rovner. 1980. Trail-following behaviour in two species of wolf spiders:
sensory and etho-ecological concomitants. Anim. Behav., 28:735-741.

Tietjen, W. J. and J. S. Rovner. 1982. Chemical communication in lycosids and other spiders. Pp.
249-279, In Spider Communication: Mechanisms and Ecological Significance (P. N. Witt and J. S.
Rovner, eds.). Princeton University Press, Princeton.

Renee Lizotte and Jerome S. Rovner, Department of Zoological Sciences, Ohio
University, Athens, Ohio 45701 USA.

Manuscript received August 1988, revised September 1988.

DESCRIPTION OF THE MALE OF ARANEUS COCHISE
(ARANEAE, ARANEIDAE)

Levi (1973) described Araneus cochise from a single female collected at the
Southwestern Research Station in the Chiricahua Mountains of Arizona. The
species was placed in the Araneus sturmi group. The A. pegnia and A. sturmi
groups include most of the small species of Araneus. Recently A. cochise was
recorded from Texas (Agnew et al. 1985) based on some females swept from
Juniperus ashei Buchholz near Stephenville in 1982. More extensive collecting the
following spring yielded a larger series including males of Araneus cochise which
are described here for the first time.

The collection area consists of J. ashei and other shrubs in a landscape of
rocky hillsides. This habitat is located in the western section of the Cross Timbers
region of Texas (elevation = 399 m). The region grades from open grass savannah

1989. The Journal of Arachnology 17:126

Figures 1, 2. — Araneus cochise left palp: 1, ventral view; 2, mesal view. Scale = 0.5 mm.

to dense brush of post oak (Quercus stellata Wangenheim) and blackjack oak
{Quercus marilandica Muenchhausen).

Araneus cochise Levi
Figs. 1, 2

Araneus cochise Levi, 1973, p. 497, figs. 55-59, map 2 (female holotype from Cochise County,

Arizona, in American Museum of Natural History).

Diagnosis. — The male of A. cochise keys to A. mariposa Levi because the
conductor has a notch and tooth on the middle of the lateral side but differs from
the latter in that the median apophysis is more curved (Figs. 1, 2). A black stripe
extends from the middle anterior portion of the abdomen posteriorly to the
humps and branches outward to the folium which is more distinct than in other
Araneus. Markings in the female are similar to those of the male.

Description (Male). — The carapace and abdomen are similar to the female
except that the black longitudinal stripe on the anterior part of the abdomen is
more pronounced. Measurements of 15 specimens (mm): Total length (x = 2.87,
range = 2.5”3.3, SE = 0.05); carapace length (x = 1.58, range = 1.5-1. 7, SE = 0.01),
width (x = 1.41, range = L4-L5, SE = 0.01); first femur (x = 1.74, range = 1.6-2. 1,
SE = 0.03); patella and tibia (x = 1.97, range = 1. 9-2.1, SE = 0.02); metatarsus (x =
1.33, range = 1.2-1. 6, SE = 0.03); tarsus (x = 0.61, range = 0.6-0. 7, SE = 0.01).
Second patella and tibia (x = 1.76, range = 1.6-1. 9, SE = 0.02); third (x = 0.97,
range = 0.9-1.0, SE = 0.01); fourth (x = 1.31, range = L2-L4, SE = 0.02).

Description (Female). — Measurements of 33 specimens (mm): Total length (x =
3.59, range = 2. 8-4.3, SE = 0.06); carapace length (x = 1.56, range = 1.4-1. 7, SE =
0.01), width (x = 1.36, range = L3-L5, SE = 0.01); first femur (x - 1.54, range =
L4-L7, SE = 0.01); patella and tibia (x = 1.75, range = 1. 6-2.0, SE = 0.02);
metatarsus (x = 1.02, range = 1.0-1. 1, SE = 0.01); tarsus (x = 0.52, range = 0.5-0. 6,
SE = 0.01). Second patella and tibia (x = 1.51, range = 1.4-1. 7, SE = 0.01); third

1989. The Journal of Arachnology 17:127

(x = 0,92, range = 0.9“L0, SE = 0.01); fourth (x = 1.29, range = 1.2-L4, SE =
O.Oi).

Record. — TEXAS; Emih Co., 11 km NE of Stephenvilie (C. W. Agnew, American Museum of
Natural History, Museum of Comparative Zoology, Texas A&M University).

Note. — Males and females were first collected on 11 April Males were last
collected on 16 May and females occurred until 10 June. One egg sac was found
and 7 spiderlings emerged.

We thank H. W. Levi, J. A. Beatty, and H. R. Burke for reviewing the
manuscript.

LITERATURE CITED

Agnew, C. W., D. A. Dean and J. W. Smith, Jr. 1985. Spiders collected from peanuts and non-
agricultural habitats in the Texas west cross-timbers. Southv/est. Nat., 30:1-12.

Levi, H. W. 1973. Small orb-weavers of the genus Amneus north of Mexico (Araneae: Araneidae).
Bull. Mus. Comp. ZooL, 145:473-552.

D. A. Dean, C. W. Ageew and R. G. Breene, Department of Entomology,
Texas A&M University, College Station, Texas 77843 USA.

Manuscript received August 1988, revised October 1988.

THE AMERICAN ARACHNOLOGICAL SOCIETY

President:

William A. Shear (1987-1989)

Department of Biology
Hampden-Sydney College

Hampden-Sydney, Virginia 23943

Membership Secretary:

Norman I. Platnick (appointed)

American Museum of Natural History
Central Park West at 79th Street
New York, New York 10024

Secretary:

James W. Berry (1987-1988)

Department of Zoology
Butler University
Indianapolis, Indiana 46208

Directors:

James C. Cokendolpher (1987-1989), W. G. Eberhard (1986-1988), Jerome S.
Rovner (1987-1989).

Honorary Members:

P. Bonnet, W. J. Gertsch, H. Homann, H. W. Levi, G. H. Locket, A. F. Millidge,
M. Vachon, T. Yaginuma.

The American Arachnological Society was founded in August, 1972, to

promote the study of Arachnida, to achieve closer cooperation between amateur
and professional arachnologists, and to publish The Journal of Arachnology.

Membership in the Society is open to all persons interested in the Arachnida.
Annual dues are $30.00 for regular members, $20.00 for student members and
$50.00 for institutions. Correspondence concerning membership in the Society
must be addressed to the Membership Secretary. Members of the Society receive
a subscription to The Journal of Arachnology. In addition, members receive the
biannual newsletter of the Society, American Arachnology.

American Arachnology, edited by the Secretary, contains arachnological news
and comments, requests for specimens and hard-to-find literature, information
about arachnology courses and professional meetings, abstracts of papers
presented at the Society’s meetings, address changes and new listings of
subscribers, and many other items intended to keep arachnologists informed
about recent events and developments in arachnology. Contributions for
American Arachnology must be sent directly to the Secretary of the Society.

President-Elect:

George W. Uetz (1987-1989)
Department of Biological Sciences
University of Cincinnati

Cincinnati, Ohio 45221

Treasurer:

Gail E. Stratton (1987-1989)
Department of Biology
Albion College
Albion, Michigan 49224

Archivist:

Vincent D. Roth

Box 136

Portal, Arizona 85632

Research Notes

Scytodes poenitens Chamberlin, not pnoeitens (Araneae,

Scytodidae), Vincent D. Roth and H. D. Cameron 107

Do female Miagrammopes animotus (Araneae, Uloboridae) spin color-

coordinated egg sacs?, Brent D. Opell 108

Measuring the stickiness of spider prey capture threads, Brent D.

Opell 112

On a new spider of the genus Drassylus (Araneae, Gnaphosidae)

from Florida, Norman I. Platnick and David T. Corey, 114

Sperm depletion in the orb-weaving spider, Nephila clavipes

(Araneae, Araneidae), Terry Christenson 115

Field observations of Gasteracantha cancriformis (Araneae,

Araneidae) in a Florida mangrove stand,

Ben M. Gregory, Jr 119

Spider predation on velvetbean caterpillar moths (Lepidoptera,

Noctuidae) in a soybean field, Ben M. Gregory, Jr., Carl S.

Barfield and G. B. Edwards 120

Water-resistant sex pheromones in lycosid spiders from a tropical

wet forest, Renee S. Lizotte and Jerome S. Rovner 122

Description of the male of Araneus cochise (Araneae, Araneidae),

D. A. Dean, C. W. Agnew and R. G. Breene 125

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 17 Feature Articles NUMBER 1

Estudio de la evolucion de la tricobothriotaxia a traves del
desarrollo postembrionario de Theridion rufipes (Araneae,

Theridiidae), Alda Gonzalez 1

Species of spiders (Araneae) associated with the immature stages
of Mantispa pulchella (Neuroptera, Mantispidae), Kevin M.

Hoffman and Jeffrey R. Brushwein 7

Predatory behavior of three Japanese species of Metleucauge

(Araneae, Tetragnathidae), Makoto Yoshida 15

Spiders of Washington County, Mississippi, Orrey P Young,

Timothy C. Lockley and G. B. Edwards 27

Predation by Pisaurina mira (Araneae, Pisauridae) on Lygus
lineolaris (Heteroptera, Miridae) and other arthropods,

Orrey P Young 43

Nest acceptance by the crab spider Misumena vatia (Araneae,

Thomisidae), Douglass H. Morse 49

Two new species of kleptoparasitic Mysmenopsis (Araneae,

Mysmenidae) from Jamaica, Frederick A. Coyle and Thomas E.

Meigs 59

Spinneret silk spigot morphology: evidence for the monophyly of
orbweaving spiders, Cyrtophorinae (Araneidae), and the group

Theridiidae plus Nesticidae, Jonathan A. Coddington 71

Foliage-dwelling spiders in three central Florida plant
communities, David T Corey and Walter K. Taylor 97

(continued on back inside cover)

Cover photograph, female Cupiennius getazi feeding on Norops limifrons,
La Selva Biological Station of OTS, Costa Rica, by J. S. Rovner
Printed by the Texas Tech Press, Lubbock, Texas, USA
Posted at Lubbock, Texas, 15 May 1989

I ^

rxnl y IP

“ 1 he Journal or

ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 17

SUMMER 1989

NUMBER 2

THE JOURNAL OF ARACHNOLOGY

EDITOR: James E. Carico, Lynchburg College

ASSOCIATE EDITOR: J. E. Carrel, University of Missouri; Jerome S. Rovner,
Ohio University

EDITORIAL BOARD: J. A. Coddington, National Museum of Natural History,
Smithsonian Institution; J. C. Cokendolpher, Texas Tech University; F. A.
Coyle, Western Carolina University; C. D. Dondale, Agriculture Canada; W.
G. Eberhard, Universidad de Costa Rica; M. E. Galiano, Museo Argentine
de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia, Missouri; N.

V. Horner, Midwestern State University; D. T. Jennings, NEFES,
Morgantown, West Virginia; V. F. Lee, California Academy of Sciences; H.

W. Levi, Harvard University; E. A. Maury, Museo Argentine de Ciencias
Naturales; M. H. Muma, Western New Mexico University; N. 1. Platnick,
American Museum of Natural History; G. A. Polis, Vanderbilt University; S.
E. Riechert, University of Tennessee; A. L. Rypstra, Miami University, Ohio;
M. H. Robinson, U.S. National Zoological Park; W. A. Shear, Hampden-
Sydney College; G. W. Uetz, University of Cincinnati; C. E. Valerio,
Universidad de Costa Rica.

THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in
Spring, Summer, and Fall by The American Arachnological Society at Texas
Tech Press.

Individual subscriptions, which include membership in the Society, are $30.00
for regular members, $20.00^ for student members. Institutional subscriptions to
The Journal of Arachnology are $50.00. Correspondence concerning subscriptions
and memberships should be addressed to the Membership Secretary (see back
inside cover). Remittances should be made payable to The American
Arachnological Society. Inquiries about availability and current prices of back
issues should be sent to Dr. Susan E. Riechert, Department of Zoology,
University of Tennessee, Knoxville, TN 37916 USA. Correspondence concerning
undelivered issues should be addressed to the Texas Tech Press, Texas Tech
University, Lubbock, Texas 79409 USA.

Change of address notices must be sent to the Membership Secretary.

Detailed instructions for the preparation of manuscripts appear in the Fall
issue of each year, and can also be obtained from the Editor and the Associate
Editor. Manuscripts that do not follow those instructions will be returned to the
author(s) without benefit of review. Manuscripts and all related correspondence
must be sent to Dr. Jerome S. Rovner, Associate Editor, Department of
Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701 USA.

PROOFS, REPRINTS, and CHARGES: Authors will receive a reprint order
form along with their proofs. Reprints are billed at the printer’s current schedule
of costs. All authors will be billed by The American Arachnological Society for
page charges. The charge per journal page will be assessed as follows: $30.00-
nonmembers; $25. 00-members acknowledging grant, institutional, or other
support; $ 15.00-full members who do not have support; $ 10.00-student members
who do not coauthor with full members and who do not acknowledge support.

This publication is printed on acid-free paper.

GonzMez, A. 1989. Aoalisis del comportamiento sexual y produccion de ootecas de Theridion rufipes
(Araneae, Theridiidae), J. ArachnoL, 17:129-136.

ANALISIS DEL COMPORTAMIENTO SEXUAL Y PRODUCCION
DE OOTECAS DE THERIDION RUFIPES
(ARANEAE, THERIDIIDAE)*

Alda Gonzalez^

Facuitad de Cieecias Naturaies y Museo
Paseo dei Bosque s/n, 1900 La Plata, Argeeti

ABSTRACT

This paper deals with a laboratory study of the sexual behavior of Theridi
and 30 females were reared individually from the egg stage. Precopulatory, copulatory and
postcopulatory behaviors, were described. Behavior patterns of the male were examined, describing in
detail the 16 observed units. Females adopted a passive attitude during copulation. The positions
adopted by the male and female were drawn during the process. Thirty couples were observed; in 27,
copulation took place. The number of egg depositions by each fertilized female is given, as well as the
percentage of fertility and fecundity, the higher percentage occurring in the first egg sacs constructed.

EXTRACT©

En el preseete trabajo se realize el estudio del comportamiento sexual en el laboratorio de
Theridion rufipes. Se utiiizaron 30 machos y 30 hembras criadas en el laboratorio individualmente
desde el estado de huevo. Se describieron las unidades de comportamiento precopulatorio, copulatorio
y postcopulatorio. Se coefecciono el patron de comportamiento de macho, describieedose en detaile
las 16 unidades observadas. No se confecciono el patron correspondiente al comportamiento de ia
hembra porque esta asumio una actitud pasiva durante toda la experiencia. Se graficaron las cuatro
posiciones adoptadas por el macho y la hembra durante el proceso. De los 30 cases observados, en 27
se produjo copula. Se dio a conocer el eumero de desoves realizados por hembra fecundada, el que
oscilo entre 3 y 9 (x = 4) y el porceetaje de fertilidad y fecundidad, correspondiendole mayor
porcentaje a las primeras ootecas construidas.

INTRODUCCION

En el presente trabajo se realiza ei estudio del comportamiento sexual en el
laboratorio de Theridion rufipes Lucas. Se describee las unidades de
comportamiento observadas durante ios periodos precopulatorio, copulatorio y
postcopulatorio. Se da a conocer ei numero de desoves obtenido por hembra
fecundada, el ietervalo entre las puestas, la edad maxima de los ejempiares en su
primer desove, la edad maxima en que puede desovar, la vida reproductiva y la
vida fertil y ios porcentajes de fecundidad y fertilidad.

’Contribucion N° 178 del Centro de Estudios Parasitologicos y de Vectores (CEPAVE). Facuitad de
Ciencias Naturaies y Museo de ia Universidad Nacional de La Plata.

^Miembro de la Carrera del Investigador del Consejo Nacional de Investigaciones Cieetificas 'y
Tecnicas (CONICET).

130

THE JOURNAL OF ARACHNOLOGY

MATERIAL Y METODOS

Para realizar el estudio del comportamiento precopulatorio, copulatorio,
postcopulatorio y desoves, se utilizaron 30 machos y 30 hembras criadas en el
laboratorio individualmente desde el estado de huevo, segun metodos explicados
en una publicacion anterior (Gonzalez 1979).

Los ejemplares empleados en esta experiencia fueron seleccionados cinco dias
despues de su muda de maduracion. Cada hembra fue colocada en recipientes de
vidrio de 25 x 15 x 15 cm tres dias antes de la introduccion del macho. Se
impidio que los individuos se vieran entre si.

Las observaciones se efectuaron bajo las siguientes condiciones de laboratorio:
temperatura 23° C ± 3, fotoperiodo de extension normal entre 12 y 14 hs diarias
de luz natural y humedad relativa del 60% al 70%.

Los machos elegidos al azar fueron introducidos cuidadosamente en los
recipientes donde se hallaban las hembras y por un lugar opuesto al de ellas.

La denominacion de las unidades y de las series de movimientos estan en parte
inspirados en Costa (1975, 1979).

RESULTADOS

Comportamiento precopulatorio. — Al introducir el macho en el recipiente de la

hembra ambos permanecen inmoviles por el termino de 2 a 5 minutos. Luego, el
macho suspendido de la tela y sin desplazarse del lugar en que se encuentra,
efectiia una serie de movimientos (abajo-arriba) con los extremos de los tarsos en
forma alternada (derecho e izquierdo) y formando un angulo mayor de 90° y
menor de 120° con la articulacion femoro-patelar.

Conjuntamente con el movimiento de los palpos, el macho ejectua movimientos
lentos (sacudidas) arriba-abajo, adelante-atras con el primer par de patas sin
llegar a tocar la tela, formando un agulo abierto (mas de 100°) con la
articulacion femoro-patelar en el movimiento arriba-abajo y un angulo agudo
(menos de 90°) en el movimiento adelante-atras.

El tiempo empleado en ejecutar estos dos movimientos es breve, aproximada-
mente de 60 a 70 segundos, durante los cuales no se observo ningiin movimiento
abdominal.

Seguidamente, el macho detecta la presencia de la hembra y comienza un lento
desplazamiento hacia ella, que se mantiene suspendida de la tela con el dorso
hacia abajo. El macho avanza siguiendo una misma direccion en forma cautelosa.
Robinson (1982) explica esto mediante la presencia de feromonas aereas y de
lineas de arrastre dejadas por la hembra y seguidas mecanicamente por el macho.

El macho se acerca hasta 1 cm de distancia de la hembra y se detiene (quietud)
suspendido de la tela con el dorso hacia arriba y ejecuta movimientos con los
palpos a modo de pedaleo (pedaleo palpar). Se alternan 50 a 60 segundos de
pedaleo palpar con 30 a 40 segundos de quietud, observandose este proceso entre
9 y 15 veces.

La hembra hasta este momento ha permanecido inmovil. — Finalizado el

pedaleo palpar y al establecerse el contact© con el macho, la hembra,
bruscamente, cambia de postura y se coloca suspendida de la tela verticalmente
con el cefalotorax hacia abajo y la zona ventral hacia el macho. El macho

GONZALEZ— COMPORTAMIENTO SEXUAL DE THEM DION RU PIPES

131

?

posicidn 1

posicion 2

?

posicion 3

posicion 4

Figura 1. — Distintas posiciones adoptadas durante el apareamiento.

invierte su posicion y se mantieiie suspendido de la tela horizontalmente con la
zona dorsal hacia abajo y el cefalotorax hacia la hembra (Fig. 1 - posicion 1),
quedaedo perpendicularmente uno con respecto al otro^ de tal modo que los
palpos del macho se hallan enfrentados a la zona epiginal de la hembra, es decir
que adoptan la posicion de copula (toma de posicion).

Lo descripto hasta aqui ha sido observado en 27 de ios 30 cases estudiados; en
los ties restantes, al realizarse el acercamiento del macho, las hembras se
desplazaron rapidamente hacia el extremo opuesto del recipiente, no mostrandose
receptivas a la copula.

Comportamieiito copulatorio. — En la posicion 1, el macho extiende hacia
adeiante las patas 1 y 2 (derecha e izquierda) por debajo de la zona ocular de la
hembra como abrazandola (abrazo) formando uri angulo de 100° ± con la
articulacion femoro-patelar. Las patas 4 las mantieee extendidas hacia atras en
linea recta. Los palpos se hallan extendidos sin efectuar ninguii movimiento.
Acompafia al abrazo un movimiento ritmico de corto recorrido (arriba-abajo,
adelaete-atras) del abdomen. Emplea entre 3 y 4 minutos para llevarlos a cabo
con intervales de quietud de 30 a 40 sequedos.

Luego el macho frota sus palpos (frotamiento palpar) en forma aiternada
(derecho e izquierdo, indistintamente) sobre la zona del epigino de la hembra, sin
desplazarse, por espacio de 1 a 3 minutos. Kaston (1970) para especies del genero
Latrodectus y Gonzalez (1986) para Steatoda retorta describen un frotamiento
similar. La hembra continiia inmovil durante todo este tiempo.

Finalizado ei frotamiento palpar se produce la copula (insercion palpar). La
insercion es aiternada. Cuando un palpo es introducido en el orificio genital
femenino, el opuesto se halla extendido en linea recta hacia arriba. Acompafia a
la insercion un movimiento del abdomen. Al retirarse el palpo del epigino, el

132

THE JOURNAL OF ARACHNOLOGY

macho manifiesta suaves movimientos abdominales de adelante-atras y arriba-
abajo (2 6 3 segundos). La introduccidn es ntmica y dura de 2 a 4 minutos. Las
primeras inserciones son mas cortas que las ultimas. El numero de inserciones por
copula varia entre 4 y 12, son mas numerosas en las primeras y van
disminuyendo en las ultimas.

Concluida una copula los individuos de la pareja cambian de posicion. El
macho invierte la suya en un movimiento rapido, quedando con la zona dorsal
hacia arriba, gira 180° se aleja 2 6 3 cm (Fig, 1 - posicidn 2), gira nuevamente
180° y queda con el cefalotorax hacia la hembra nuevamente. La hembra se
coloca horizontalmente con la zona dorsal hacia abajo y el cefalotdrax dirigido
hacia el macho (Fig. 1 - posicion 3). Ambos ejemplares permanecen en posicion 3
(enfrentados pero invertidos) sin manifestar movimientos (descanso) durante 2 6 3
minutos, tras los cuales adoptan la posici6n 1 y se reinicia lo descripto.

Este proceso que comprende desde la toma de la posici6n 1 a la posici6n 3 con
c6pula incluida, se repite de 4 a 8 veces {x = 6), empleando para ello desde 38 a
49 minutos {x = 41).

Concluida la ultima c6pula, el macho que estaba en posici6n 1 se desliza
rapidamente hacia la zona superior del recipiente (retirada del macho) alejandose
de la hembra y ubicandose por encima de ella, a mas de 10 cm de distancia;
inmediatamente invierte su posici6n quedando suspendido de la tela horizontal-
mente, con el dorso hacia arriba. La hembra adopta su posici6n habitual
(horizontalmente con el dorso hacia abajo) (Fig. 1 - posici6n 4).

Comportamiento postcopulatorio. — El macho, luego de la retirada, se detiene
(posici6n 4) y efectua la limpieza de los palpos frotando uno contra otro
(limpieza palpar), empleando en ello de 5 a 7 minutos; conjuntamente realiza
movimientos bruscos con las patas 1 y 2 a modo de bicicleteo (bicicleteo pedial) y
movimientos con el abdomen (movimientos abdominales) de arriba-abajo de
corto recorrido,

En la Fig. 2 se halla representado el modelo total del comportamiento sexual
tipo del macho de Theridion rufipes.

En el 78% de las parejas que copularon no se observaron nuevos
acercamientos. En el 32% restante, entre los 33 y 39 minutos posteriores a la
limpieza palpar, el macho se desplaz6 nuevamente hacia la hembra y se repitieron
las etapas de quietud, pedaleo palpar y toma de posici6n del comportamiento
precopulatorio y las etapas de abrazo, movimientos abdominales y frotamiento
palpar del comportamiento copulatorio. Este ultimo frotamiento palpar es mas
prolongado (2 a 5 minutos) y con un ritmo mas acelerado que el de los primeros
acercamientos. En ningun caso se produjo c6pula, ya que finalizado el fotamiento
palpar, la hembra que parecia receptiva, comienza a efectuar movimientos
bruscos a modo de pedaleo con sus patas 1 y 2. Esto provoca el alejamiento
inmediato del macho, cambiando ambos ejemplares de posici6n 1 a posici6n 4.
Umaha (1987) tambien observa para la misma especie la existencia de una s61a
c6pula al inicio de la vida adulta.

A partir de aqui, y a pesar de haber dejado conviviendo por mas de quince
dias a las parejas en el mismo recipiente, no se registraron nuevos acercamientos
ni actitudes de ataque.

Desoves, fecundidad y fertilidad. — Las hembras fecundadas en el laboratorio
desovaron entre los 11 y 172 dias (x = 72 dias) despues del apareamiento. El
total de ootecas construidas por las 27 hembras fecundadas fue de 132, y la

GONZALEZ— COMPORTAMIENTO SEXUAL DE THERIDION RUFIPES

13

Limpieza paJpar

14

Bicideteo pedial

Figura 1. — Modelo tipo del comportamiento sexual del macho. Espesor de las flechas: 5 mm =
sucesiones. A = Comportamiento precopulatorio, B = Comportamiento copuiatorio, C
Comportamiento postcopulatorio.

134

THE JOURNAL OF ARACHNOLOGY

% 0
50 ^

ootecas

Figura 3. — Numero de desoves por hembra.

cantidad de desoves por hembra fue variable, desde 3 a 9 (x = 4) (Fig. 3). Segun
Umana (1987) el numero de ootecas construidas por hembra es de 1 a 10.

Teniendo en cuenta las hembras fecundadas y los desoves efectuados, se
consideraron los siguientes periodos: a) eclosion y primer desove (edad del
ejemplar en su primer desove), b) eclosion y ultimo desove (edad maxima en que
puede desovar), c) ultima muda y ultimo desove (vida reproductiva), d) primera
fecundacion y ultimo desove (vida fertil), e) ultimo desove y muerte (Tabla 1).

Umana (1987) proporciona datos sobre numero de mudas, duracion de estadios
y crecimiento. Debe tenerse en cuenta que esta autora llama “primera muda
verdadera” que da origen a un segundo estadio a la muda que en el presente
estudio, asi como en otros anteriores (GonzMez 1979, 1981, 1982, 1984, 1986) se
ha considerado como la tercer muda, que da lugar al cuarto estado (contando
como primera muda el desprendimiento de la cuticula embrional que arrastra
consigo al diente de eclosion).

Tabla 1. — Intervalos en dias entre: a = eclosion y primer desove, b = eclosion y ultimo desove, c =
ultima muda y ultimo desove, d = primera fecundacion y ultimo desove, e = ultimo desove y muerte.
N = numero de individuos, x = promedio, SD = desviacion tipica.

a

b

c

d

e

N

27

27

27

27

27

A

228-305

322-500

130-276

83-254

186-265

X

280.71

383.57

207.28

150.57

255.29

SD

32.50

55.87

45.41

56.54

26.07

GONZALEZ— COMPORTAMIENTO SEXUAL DE THERIDION RUFIPES

135

Tabla 2. — Porceetaje de fecundidad (Fe) y de fertilidad (Fr). N = numero de individuos, A =
amplitud, x = promedio, SD = desviacion tipica.

Primeras ootecas

Ultimas ootecas

Fe

Fr

Fe

Fr

N

10

10

10

10

A

107-71

107-70

73-78

70-32

SD

11.88

12.40

10.10

11.02

X

93.5

89.7

56.9

52.5

El promedio de huevos por ooteca fue de 89, correspondiendole a la primera el
mayor numero de huevos (107), el que fue descendiendo en las ootecas siguientes,
siendo en la ultima de 38. Esto concuerda con io citado por Umafia (1987).

Todos los huevos puestos no evolucionaron normalmente. En muy pocas
ootecas el desarrollo de los huevos aicanzo el 100%, mientras que en la gran
mayoria, el porcentaje de fecundidad (numero de huevos puestos) es mayor que el
porceetaje de fertilidad (oumero de huevos fertiles).

Kaston (1970) sostiene que para especies del genero Latrodectus el porcentaje
de fecundidad y de fertilidad no esta en relacion directa con el orden en que se
producen los desoves, ya que ootecas constrmdas tempranamente pueden no
mostrar desarrollo, mientras que otras constrmdas con posterioridad y por la
misma madre, poseen gran numero de huevos y aicanzan el mas alto porcentaje
de fertilidad.

Esto no sucede con los desoves de Theridion rufipes. Los estudios efectuados
demuestran que las primeras ootecas realizadas tienen mayor porcentaje de
fertilidad que las realizadas mas tardiamente (Tabla 2), coincidiendo esto con los
resultados de Umana (1987).

AGRADECIMIENTOS

A la profesora Maria Elena Galiano por la lectura critica del manuscrito, a la
Sra Maria C. Estivariz por el delineado de los dibujos y graficos y a la Dra
Monica Lopez Ruf por la claboracion brindada en las observaciones y registro de
las unidades de comportamiento.

LITERATURA CITADA

Costa, F. 1975. El comportamiento precopulatorio de Lycosa malMosa Tullgren (Araneae, Lycosidae).
Rev. Brasil Biol Rio de Janeiro, 35:359-368.

Costa, F. 1979. Analisis de la copula y de la actividad postcopulatoria de Lycosa malitiosa Tullgren
(Araneae, Lycosidae). Rev. Brasil. Biol. Rio de Janeiro, 39: 361-376.

Gonzalez, A. 1979. Observaciones bioecologicas sobre una especie del genero Latrodectus Walckenaer,
1805, del grupo mactans, de Sierra de la Ventana (prov. de Buenos Aires, Argentina) (Araneae,
Theridiidae). !IL Desarrollo postembrionario. Acta Zool. Lilloana, 35:97-1 !0.

Gonzalez, A. 1981. Desarrollo postembrionario de Latrodectus mirabilis, Latrodectus antheratus y
Latrodectus corallinus (Araneae, Theridiidae). Physis Secc. C, 39:83-91.

Gonzalez, A. 1982. El desarrollo postembrionario de Tidarren sisyphoides (Walckenaer) (Araneae,
Theridiidae). Physis Secc. C, 41:87-91.

Gonzalez, A. 1984. Desarrollo postembrionario y evolucion de los organos mecanoreceptores de
Latrodectus diaguita Carcavallo y estudio de la tricobotriotaxia de Latrodectus quartus Abalos
(Araneae, Theridiidae), Physis, Secc. C, 42:1-5.

136

THE JOURNAL OF ARACHNOLOGY

GonzMez, A. 1986. Ciclo biologico y desarrollo de Steatoda retorta GonzMez (Araneae, Theridiidae).
Rev. Soc. Entomol. Argentina, 44:191-203.

Kaston, B. J. 1970. Comparative biology of American black widow spiders. Trans. San Diego Soc.
Nat. Hist., 16:33-82.

Lucas, H. 1846. Historic naturelle des Animaux articules. In Exploration Scientifique de I’Algerie
pendant les annees 1840, 1841, 1842, publiee par ordre du Gouvernement et avec le concours d’une
commission academique. Sciences Physiques, Zoologie, 1:89-271.

Robinson, M. H. 1982. Courtship and mating behavior in spiders. Ann. Rev. Entomol. 27:1-20.
Umaha, G. M. 1987. Ciclo de vida y biologia de la reproduccion de Theridion rufipes Lucas
(Araneae, Theridiidae) en Costa Rica. Aracnologia, 7:1-9.

Manuscript received October 1987, revised June 1988.

Acosta, L. E. 1989. Pachyloides hades, nueva especie de opilion de la Argentina (Opiliones,
Gonyleptidae, Pachylinae). J. Arachnol., 17:137-142.

PACHYLOIDES HADES, NUEVA ESPECIE DE OPILION
DE LA ARGENTINA

(OPILIONES, GONYLEPTIDAE, PACHYLINAE)

Luis Eduardo Acosta

Catedra de Zoologia I

Facultad de Ciencias Exactas, Fisicas y Naturales
Universidad Nacional de Cordoba
Casilla de Correos 122, 5000 Cordoba, Argentina

ABSTRACT

Pachyloides hades, a new species of Gonyleptidae from the province of Tucuman, Argentina, is
described and illustrated. It differs from its nearest relative, Pachyloides tucumanus Canals, mainly in
the armature of the ocular tubercle and trochanter IV of the males. Differences from Pachyloides
thorellii Holmberg and Daguerreia maculata Canals are also noted. P. hades was collected at two
localities over 2,000 m above sea level.

RESUMEN

Se describe e ilustra Pachyloides hades, nueva especie de Gonyleptidae de la provincia de Tucuman,
Argentina. Se comentan las diferencias con su especie mas proxima, Pachyloides tucumanus Canals
(principalmente referidas a la armadura del oculario y el trocanter IV del macho), asi como con
Pachyloides thorellii Holmberg y Daguerreia maculata Canals. Pachyloides hades fue colectada en dos
localidades situadas por encima de 2,000 m s.n.m.

INTRODUCCION

Revisando material de Gonyleptidae del noroeste argentine, perteneciente al
Museo de La Plata, localice dos machos, mal conservados, de una posible nueva
especie, afin a Pachyloides tucumanus Canals. No obstante su parecido con esta,
los ejemplares mostraban algunas particularidades, pero en ese memento era
dificil determinar si ellas obedecian a variaciones individuales. Esto me motive a
realizar un viaje a la localidad de tales especimenes (El Infiernillo, provincia de
Tucuman), donde logre reunir un interesante lote. En proximidades de Tafi del
Valle capture otros ejemplares de la misma especie, en tanto obtuve lotes
adicionales en las colecciones del Institute Miguel Lillo de Tucuman. El estudio
de este material me permitio confirmar la validez de la especie, que describe a
continuacion con el nombre de Pachyloides hades.

Con esta suman seis las especies de Pachyloides Holmberg presentes en
Argentina, totalizando unas 14 para el genero, cuya area de distribucion abarca
tambien Uruguay, Paraguay y Brasil (Ringuelet 1959; Soares y Soares 1954).
Algunas de estas especies necesitan una cuidadosa revision de su status
taxonomico, lo que sera objeto de futuras contribuciones.

138

THE JOURNAL OF ARACHNOLOGY

Pachyloides hades, nueva especie
Figs. 1-8

Daguerreia maculata: Meyer y Weyrauch, 1966:60 (Fig. 43), 61 (error de identificacion).

Derivatio nominis. — El nombre especifico hades (en aposicion) refiere al dios
de los infiernos en la mitologia griega, en alusion a la localidad tipo.

Material tipico. — Holotipo macho (MACN 8620), El Infiernillo, Tucuman, 5
abril 1986 (L. Acosta); alotipo hembra, iguales datos, en el mismo tubo que el
holotipo. Un macho y una hembra paratipos (CZI), igual localidad, fecha y
colector; un macho y dos hembras paratipos (IML), igual localidad, 5 abril 1965
(W. Weyrauch); un macho y una hembra paratipos (MACN 8621), La
Quebradita, Tafi del Valle, Tucuman, 4 abril 1986 (L. Acosta). El ejemplar macho
de IML que aqui se designa como paratipo es presumiblemente el mismo
individuo ilustrado en Meyer y Weyrauch (1966), con el nombre de Daguerreia
maculata Canals.

Localidad tipo. — El Infiernillo (provincia de Tucuman, Argentina); 26°44'S/
65°47'W.

Distribucion geogrMica. — Argentina, provincia de Tucuman. Hasta el
momento, la especie ha sido colectada en dos localidades del sistema montanoso
del Aconquija: El Infiernillo (3,042 m s.n.m.) y La Quebradita (2,100 m, proxima
a Tafi del Valle). Desde el punto de vista biogeografico ambas corresponden a la
provincia altoandina (Cabrera y Willink 1973).

Habitat. — En sus dos localidades, Pachyloides hades fue hallada bajo piedras,
en ambiente humedo. El material de La Quebradita se colecto en las margenes de
un arroyo, bordeado por un bosquecillo de aliso {Alnus jorullensis H.B.K.); dicha
localidad representa el limite superior para esta betulacea, que entre los 1,700 y
1,800 m forma un piso altitudinal, con bosques casi puros. El ambiente en El
Infiernillo es, en cambio, un tipico pastizal de altura, dominado por especies de
Festuca. La precipitacion pluvial es aqui inferior a los 400 mm anuales, pero son
frecuentes las nieblas densas — especialmente al atardecer y la noche — , que al
condensarse sobre la vegetacion y las rocas, aseguran una importante cuota de
humedad al sustrato (Meyer y Weyrauch 1966).

Descripcion. — Coloracion general pardo amarillento a castano anaranjado,
coxa y femur IV mas oscuros; pedipalpos, patas I a III y metatarso y tarso IV
amarillo ciaro. Medidas de holotipo y alotipo en Tabla 1. Longitud del escudo
dorsal: machos de 5.30 a 7.49 mm (x = 6.49, n = 14), hembras de 6.06 a 6.98 mm
(x = 6.45, n = 14). Prosoma con escasos granulos, dispersos y sobre el borde
anterior; oculario bajo, con armadura representada por un par de granulos o
tuberculos muy pequenos. Escudo con surcos poco marcados, especialmente en el
macho, el surco longitudinal en el area I puede faltar o estar apenas insinuado
(en el holotipo este surco es muy tenue); en las hembras los surcos se destacan
con franjas de pigmento levemente mas oscuras. Pedipalpos: femur con uno y
trocanter con dos tuberculos setigeros ventrales, el primero con una espina
subterminal mesial. Numero de tarsitos: 6-8/ 9-7-7, con frecuencia el tarsito basal
en pata I algo dilatado (holotipo con 6-8/ 9-7-7 tarsitos, alotipo con 6-8-7-7;
variabilidad en Tabla 2). Macho: Granulacion poco destacada en areas I, II y
laterales, mas llamativa en las restantes areas y los tergitos libres, con granulos
anchos y bajos. Area I con dos hileras irregulares de pocos granulos; areas II y

ACOSTA— NUEVA ESPECIE DE PACHYLOIDES 139

Figuras 1-8. — Pachyioides hades, nueva especie: 1-6, holotipo macho (MACN 8620); 1, escudo
dorsal, tergitos libres, operculo anal, coxas IV, trocanter y femur derechos, vista dorsal; 2, coxa,
trocariter y femur IV derechos, vista ventral; 3, coxa, trocanter, femur y patela IV derechos, vista
lateral; 4, oculario, vista posterior; 5, pene, glande en vista dorsal; 6, id., vista lateral; 7-8, alotipo
hembra (MACN 8620); 7, escudo dorsal, tergitos libres, coxas IV, trocanter y femur derechos, vista
dorsal; 8, oculario, vista posterior. Figuras 9-10. — Pachyioides tucumanus Canals (macho de Horco
Molle, Tucuman; CZI): 9, oculario, vista posterior; 10, coxa, trocanter y base de femur IV derechos,
vista lateral La escala representa 1 mm en Figs. I, 2, 3, 7 y 10, y 0.5 mm en Figs. 4, 5, 6, 8 y 9.

Ill con una hilera posterior compieta y un esbozo anterior irregular; areas IV, V y
tergitos libres, sendas filas de granules alargados, excepcionalmente en el area IV
se agrega un rudimento de segunda hilera anterior; areas iaterales, una hilera de
granulos bajos, poco llamativos; operculo anal, escasos granules en dos filas
transversales; esternitos con una hilera de granulos diminutos. Patas I a III

140

THE JOURNAL OF ARACHNOLOGY

Tabla 1. — Pachyloides hades nueva especie; medidas en mm de holotipo y alotipo.

Holotipo

Alotipo

Escudo

longitud

6.65

6.19

ancho maximo

5.57

5.11

Pata I

long, total

11.89

10.61

long, femur

2.88

2.55

Pata II

long, total

17.03

15.39

long, femur

4.13

3.86

Pata III

long, total

15.13

13.56

long, femur

3.83

3.54

Pata IV

long, total

20.82

17.88

trocanter

1.96

1.18

femur

4.58

4.26

patela

2.10

1.77

tibia

4.06

3.50

metatarso

5.57

4.91

tarso

2.55

2.26

Pedipalpos

long, total

7.83

7.56

long, femur

1.96

1.83

Queh'ceros

long, total

2.03

2.10

Oculario

ancho

1.05

0.98

alto

0.34

0.33

inermes, granulosas. Coxa IV con apofisis curva, dirigida hacia atras, de
extremos vueltos levemente hacia afuera; en norma dorsal, deja visible la
articulacion externa con el trocanter. Trocanter IV con apofisis lateral
lobuliforme; sobre la articulacion con el femur, apofisis dorsal subconica,
alargada, dirigida hacia arriba; pequena apofisis roma, ventromesial, en el margen
posterior del artejo. Femur IV suavemente curvado en vista lateral; superficie
dorsal con hileras longitudinales de granulos, destaca una hilera completa de
granulos perliformes en el borde dorsolateral; hilera mesial de apofisis
subtriangulares, mayores hacia el apice del artejo (en su extremo proximal son
tuberculos bajos); hilera ventrolateral similar, de apofisis mas pequenas; hilera
ventral de pocos tuberculos redondeados; tres apofisis apicales dorsales, la mesial
es la mayor del artejo. Pene: glande con parte ventral subrectangular, alargada,
con dos grupos de tres espinas en cada borde lateral, uno subterminal y uno
basal; un pequeno tuberculo entre ambos grupos de espinas, otro proximo al
grupo de espinas basal, y dos cercanos al grupo subterminal. Parte dorsal de base
ancha, cuyos hordes, en vista dorsal, ocultan parcialmente las espinas basales;
tubulo seminal con proceso ventral terminado en dos puntitas. Hembra: Escudo
dorsal con granulos mas pequenos y escasos que en el macho. Patas I a III
inermes. Coxa IV con una pequena apofisis conica lateral. Femur IV: hileras
longitudinales de granulos, algo mas destacados y aguzados en posicion
ventrolateral y ventromesial; apofisis apicomesial pequena. Ovipositor: extremo
apical tetralobado, con 10 espinas setiformes (tres espinas en los lobulos dorsales,
dos en los ventrales); en el alotipo se agrega una cuarta espina en el lobulo dorsal
derecho; vagina interna con cuatro pares de receptaculos seminales.

Comparacion y diagnosis. — La especie que mas se aproxima a Pachyloides
hades es Pachyloides tucumanus, presente tambien en la provincia de Tucuman
pero a menor altitud, en ambientes de selva. Las principals diferencias consisten
en la apofisis dorsal del trocanter IV del macho, que en P. hades termina en

ACOSTA— NUEVA ESPECIE DE PACHYLOIDES

141

Tabla 2. — Numero
material estudiado.

de

tarsitos en

Pachyloides hades

nueva

especie; frecuencias

halladas en el

Numero de tarsitos

5

6

7

8

9 10

n

Machos

T I

3

37

—

—

— —

40

T II

—

I

2

14

18 2

37

T III

2

1

37

—

— —

40

T IV

—

1

33

3

— —

37

Hembras

T I

I

36

—

—

— —

37

T II

—

—

2

22

15 —

39

T III

—

1

38

—

— —

39

T IV

—

—

39

—

— —

39

forma mas aguzada (Figs. 3 y 10), y en la armadura del oculario (Figs. 4 y 9),
mucho mas desarrollada en P. tucumanus; por otra parte, el pene aparece en la
nueva especie mas deprimido en vista lateral. Otras diferencias, algo sutiles,
afectan la armadura de coxa y femur IV del macho, la granulacion dorsal y el

tamano de los especimenes.

Tambien Pachyloides thorellii Holmberg, especie cercana a P tucumanus,
muestra similitudes con P. hades. Cabe senalar que Ringuelet (1959) considera a
P. tucumanus como subespecie de thorellii, criterio que no comparto; este autor
supone que la provincia de Tucuman es la “zona de intergraduacion” entre tales
“razas geograficas”. Los haliazgos seguros de P. thorellii, sin embargo, se limitan
a la provincia de Buenos Aires, Argentina, y al Uruguay. Los machos de esta
especie se reconocen facilmente pues la apofisis lateral del trocanter IV esta
reducida a un reborde esclerosado; la apofisis dorsal, en tanto, es identica a la de
P. tucumanus, y permite la distincion con P. hades. Las hembras de P. thorellii
poseen una pequeha apofisis dorsal en el trocanter IV (Canals 1943), ausente en
los dos Pachyloides de Tucuman.

Pachyloides hades fue confundida con Daguerreia maculata Canals-simpatrida
de P. tucumanus-, pero esta es una especie claramente distinta, por su coloracion
oscura, granules dorsales muy conspicuos, y su peculiar armadura en femur y,
especialmente, trocanter IV del macho.

Material estudiado. — ARGENTINA: Provincia de TUCUMAN; El Infiernillo, 5 abril 1986 (L.
Acosta), holotipo macho y alotipo hembra (MACN 8620); iguales datos, 1 macho y 1 hembra
paratipos (CZI); iguales datos, 7 machos, 5 hembras, 11 juv. (CZI); id. loc., 5 abril 1965 (W.
Weyrauch), 2 machos (MLP); iguales datos, 1 macho, 2 hembras paratipos (IML s/n), ""Daguerreia
maculata'’’ Ringuelet det. 1965; id. loc., 12 diciembre 1947 (R. Golbach), 3 machos (IML 00021); id.
loc. y coL, 28 noviembre 1947, 1 macho (IML 00019); id. loc. y fecha (O. Budin), 1 macho (IML
00020); La Quebradita, Tafi del Valle, 4 abril 1986 (L, Acosta), 1 macho y 1 hembra paratipos
(MACN 8621); iguales datos, 2 machos, 6 hembras, 4 juv. (CZI); “Departamento Tafi”, sin especificar
localidad, 21 octubre 1983 (Schocchi), 1 macho, 4 hembras, 1 juv. (IML s/n).

AGRADECIMIENTOS

Estoy muy reconocido al Lie. R. Arrozpide, Museo de La Plata (MLP), y al
Dr. Z. Tomsk y la Lie. F. Drahg, Instituto Miguel Lillo, Tucuman (IML), por el

prestamo de material de sus respectivas cotecciones. Agradezco tambien al Dr. A.

142

THE JOURNAL OF ARACHNOLOGY

Willink (IML), asi como a todo el equipo de entomologos de esa institucion, las
atenciones recibidas en mi viaje a la provincia de Tucuman. Los tipos de la nueva
especie, salvo un lote de paratipos perteneciente al IML, quedan depositados en
el Museo Argentino de Ciencias Naturales, Buenos Aires (MACN) y la Catedra
de Zoologia I, Universidad Nacional de Cordoba (CZI). El Dr. M. Doucet me
facilito gentilmente los equipos de optica para realizar las ilustraciones.
Finalmente, agradezco al Dr. E. A. Maury (MACN) y la Dra. N. V. de Argiiello
(CZI) sus criticas al manuscrito.

LITERATURA CITADA

Cabrera, A. L. y A. Willink. 1973. Biogeografia de America Latina. Colecc. Monogr. Cient. O.E.A.,
serie Biol., N° 13, pp. LVI, LI 17.

Canals, J. 1943. Descripcion de nuevos generos y especies de opiliones de la subfamilia “Pachylinae”.
Not. Mus. La Plata, Zook, 8(63): 1-22.

Meyer, T. y W. K. Weyrauch. 1966. Guia para dos excursiones biologicas en la provincia de Tucuman.

Miscel. Inst. Miguel Lillo, Tucuman, N° 23, pp. 1-127.

Ringuelet, R. A. 1959. Los aracnidos argentinos del orden Opiliones. Rev. Mus. Argentino Cienc.
Nat., C. Zook, 5(2): 127-439, lams. I-XX.

Soares, B. A. M. y H. E. M. Soares. 1954. Monografia dos generos de opilides neotropicos. Arq.
Zook, Sao Paulo, 8(9):225-302.

Manuscript received April 1988, revised September 1988.

Young, O. P. and T. C. Lockley. 1989. Spiders of Spanish moss in the delta of Mississippi. J.
Arachnol., 17:143-148.

SPIDERS OF SPANISH MOSS IN THE
DELTA OF MISSISSIPPI

Orrey P. Young

Southern Field Crop Insect Management Laboratory
USDA-ARS, RO. Box 346
Stoneville, Mississippi 38776 USA

and

Timothy C. Lockley

Imported Fire Ant Station
USDA-APHIS-PPQ, 3505 25th Avenue
Gulfport, Mississippi 39501 USA

ABSTRACT

Arthropods were collected monthly during a 13-month period from Spanish moss on water oak
trees in Leroy Percy State Park, Washington County, Mississippi. Of the ca 2000 arthropods
collected, spiders (600), bettles (600), chalcidoid wasps (500), and miscellaneous insects (300)
represented the major groups. Peak population levels of spiders occurred in spring and fall, as did the
occurrence of various crop pests. Spiders were represented by 13 families, 22 genera, and 29 species.
Seventy percent of the total spider sample was composed of two species that breed in Spanish moss,
Metaphidippus tillandsiae Kaston and Anyphaena maculata (Banks). Steadily declining non-spider
arthropod populations in Spanish moss throughout the fall-winter-spring period suggest that the
predatory activity of spiders may have an effect on over-wintering arthropod populations.

INTRODUCTION

Spanish moss {Tillandsia usneoides L.) (Bromeliaceae) is an epiphytic member
of the Pineapple Family that forms long hanging tufts on branches of trees in
damp or swampy areas and is distributed from Texas east to Florida and north to
Virginia on the Coastal Plain (Fernald 1950). During the winter, this plant can be
quite conspicuous on deciduous trees and locally abundant. During the course of
a search of the potential overwintering habitats of the tarnished plant bug, Lygus
lineolaris (Palisot) (Heteroptera: Miridae), samples from Spanish moss revealed a
diverse fauna of arthropods. Spiders were the most abundant taxonomic order in
the samples, and information is presented here on their seasonal composition,
abundance, and potential impact on other arthropod inhabitants.

METHODS AND MATERIALS

Samples were collected from May 1985 through May 1986 at Leroy Percy
(L.P.) State Park, Washington County, Mississippi. This park is a 1000 ha forest

144

THE JOURNAL OF ARACHNOLOGY

surrounded by agricultural lands planted in rice, soybeans, and cotton. Early-
successional habitats (e.g., oldfields and roadside margins) are abundant in the
park and contain plants supporting populations of crop pests, particularly
Erigeron spp. (Compositae) as a host of Lygus lineolaris. Approximately one-
third m^ of Spanish moss was collected monthly (except November, due to
flooding) from each of the same three water oak trees {Quercus nigra L.). These
trees were located at the south end of a 10 ha mowed field of mixed grasses that
was surrounded by low damp woods subject to periodic flooding. Each tree was
at least 10 m from the border of the woods and 15 m from each other. Sampling
was accomplished by standing on top of a truck cab under each tree, surrounding
a clump of moss with a 1/3 m^ fine-mesh organdy bag, cutting off the tree branch
at the bag opening, and continuing until the bag was full. The 3 bags were then
placed in a large plastic bag and transported to the laboratory, frozen at -20° C
for 12-36 hours, and then examined under magnification. Assistance with spider
identifications was provided by G. B. Edwards and voucher specimens were
deposited at Florida State Collection of Arthropods, Gainesville; Mississippi
Entomological Museum, Mississippi State University, Starkville; and USDA-
APHIS-PPQ-IFAS, Gulfport, Mississippi.

RESULTS AND DISCUSSION

Collections from Spanish moss over a 13-month period produced approxi-
mately 2000 arthropods. The majority of specimens were spiders (600),
Coleoptera (600), and chalcidoid Hymenoptera (500), with various other insect
groups totaling an additional 300. Crop pest species in the collections included
the tarnished plant bug (3); chinch bug, Blissus sp. (3); flea beetle, Altica sp. (3);
stink bug, Euschistus sp. (2); cereal leaf beetle, Oulema sp. (2); boll weevil,
Anthonomus sp. (1); and bean beetle, Cerotoma sp. (1). These data indicate that,
at least for the sites sampled, Spanish moss is not a significant refuge or breeding
habitat for insects known to be pests on field crops. Predators and parasites of
crop pests were present in low numbers and included Coleomegilla sp. and Lebia
sp. (Coleoptera), Orius sp. and Zelus sp. (Hemiptera), Chrysopa sp. (Neuroptera),
and Micropletis sp. (Hymenoptera).

Spiders were represented by 13 families, 22 genera, and 615 individuals (Table
1). Identification of immature spiders was particularly difficult and created some
tabulation problems. All adult, and most penultimate, specimens were identified
to species. Some immatures (e.g., Metaphidippus tillandsiae Kaston) could be
identified without confusion. Some immatures [e.g., Phidippus putnami (Peck-
hams)] were captured in a silken retreat with an adult female and thus were
identified by association. Some immatures [e.g., Anyphaena maculata (Banks)]
were identified to genus, and the presence of adult specimens of only one species
within the genus permitted a tentative species determination. When the adults of
two species within a genus were present (e.g., Philo dr omus), the immatures could
usually be associated with one of the species. Some immatures (e.g., Clubionoides
sp.) could be identified to genus, but the absence of adult specimens prevented a
species association. With these limitations. Table 1 lists 29 species, of which only
14 are represented by adult specimens. This list also contains 11 additions to the
state list as compiled by Dorris (1972).

YOUNG & LOCKLEY— SPIDERS OF SPANISH MOSS

145

The abundance of spiders in Spanish moss was greatest in September and May
with the appearance of large numbers of immatures and was at low levels in both
mid-summer and late winter when adults predominated. The number of species
was highest (12) in May and December and lowest (4) in July. Eighty-five percent
of the total spider sample was composed of only four species. These four
appeared to be the only species reproducing within Spanish moss and warrant
further comment.

Metaphidippus tillandsiae Kaston. — Two hundred forty-one specimens of this
species were obtained, representing all sampling periods. Overwintering occurs in
the adult stage, with an exclusively immature population in the summer and
adults first appearing in October. This species was first described by Kaston in

1973 and has only been recorded from Spanish moss. The distribution of M.
tillandsiae — North Carolina south to Florida, west to Louisiana — is entirely
within the range of Spanish moss (Fernald 1950; Kaston 1973). It is probable that
this species is restricted to Spanish moss and may be a major predator within that
microhabitat.

Anyphaena maculata (Banks). — One hundred eighty-six specimens were
obtained, representing all sampling periods except January. Overwintering occurs
in both adult and immature stages, with an exclusively immature population in
the summer and adults first appearing in October. This species was redescribed in

1974 and has been recorded from Spanish moss, sweeping in bottomland pine
and hardwood forests, sifting of leaf litter, and malaise trap (Platnick 1974), as
well as from cotton (Whitcomb and Bell 1964). The distribution of A. maculata
— New York south to Georgia, west to Illinois and Louisiana — is considerably
broader than Spanish moss (Platnick 1974), and thus is not restricted to this
microhabitat. This predator, combined with M. tillandsiae, may exert a dominant
influence on the abundance of arthropods in Spanish moss at the L.P. State Park
sites.

Philodromus keyserlingi Marx. — Fifty-three specimens were obtained, repres-
enting all sampling periods except June. Overwintering occurs in the immature
stages, with adults only present in May. This species was redescribed in 1961 and
has been recorded from wasp’s nests (Dondale 1961), as well as from corn
(Plagens 1985), peanuts ((Agnew et al. 1985), cotton (Skinner 1974), and soybeans
(LeSar and Unzicker 1978). The distribution of P. keyserlingi is very broad,
occurring from Massachusetts south to Florida, west to Iowa and New Mexico
(Dondale 1961). The temporal pattern of occurrence in the L.P. State Park
samples, combined with the published records of geographic and habitat
distribution, suggests that this species is an occasional vagrant and facultative
breeder in Spanish moss.

Nodocion floridanus (Banks). — Forty-four specimens were obtained, represent-
ing all sampling periods except July, January, and February. Overwintering may
occur in both the adult and immature stages, but the absence of specimens in
January and February precludes a definite determination. Adults were present
only in May and June, the same pattern of occurrence as reported from Arkansas
cotton and pine (Heiss and Allen 1986). This species was redescribed in 1980 and
has been reported from houses, wasp’s nests, under bark, on pine trees, in bowers
and nests (Platnick and Shadab 1980), as well as from cotton (Dean et al. 1982).
The distribution of N. floridanus is also very broad, occurring from
Massachusetts south to Florida, west to Minnesota and Arizona. (Platnick and

146

THE JOURNAL OF ARACHNOLOGY

Table 1. — Spiders of Spanish moss at Leroy Percy State Park, Mississippi, 1985-1986. Explanation
of symbols: Instar: M = male, F = female, PM = penultimate male, PF = penultimate female, IM =

immature (juvenile). * = no sample.

Taxon

ANYPHAENIDAE
Anyphaena maculata (Banks)

Teudis mordax (O.P.-Camb.)

ARANEIDAE

Eustala cepina (Walck.)
Neoscona hentzi (Keys.)

N. domiciliorum (Hentz)
CLUBINONIDAE
Clubionoides sp.
DICTYNIDAE
Dictyna sp.
GNAPHOSIDAE

Cesonia bilineata (Hentz)
Nodocion floridanus (Banks)

LINYPHHDAE

Ceraticelus sp.

LYCOSIDAE
Pardosa sp.
MIMETIDAE
Mimetus sp.
PHILODROMIDAE
Philodromus keyserlingi
Marx

P. vulgaris (Hentz)
PISAURIDAE
Pisaurina sp.
SALTICIDAE
Eris militaris (Hentz)
Hentzia sp.

H. mitrata (Hentz)
Metaphidippus sp,

M. tillandsiae Kaston

Phidippus sp.

P. puinami (Peckhams)

Platycryptus undatus (DeG.)

Instar MJ J ASONDJ FMAM Total

M

F

PM

PF

IM

M

IM

1

3

3 1

2

4 6 10 6 26 10

2 1
1

10

1 2 3

2 77

1

5

20

5

2

154

4

2

IM

10

10

IM

IM

M

F

PM

PF

IM

2 1 18 1

2 14 7

1

2

4

1

2

38

PM

PF

IM

IM

IM

2

1 1
2 1
1

M

F

PM

IM

F

1 113 4

20 5 8 1 1

2

1

2

48

1

IM

M

IM

M

IM

M

F

PM

IM

IM

M

F

IM

IM

14

1 15

8

4 12 21 10 30 11

1

25

1

1

1

14 14 11

22 9 13

1

1

2

2

1

1

66

78

8

89

1

1

1

25

1

YOUNG & LOCKLEY— SPIDERS OF SPANISH MOSS

147

Zygobalius sp.

IM

1

1

2

Z. nervosus (Peckhams)

M

1

1

THERIDIIDAE

Euryopis sp.

IM

1

I

E. funebris (Hentz)

M

2

2

Theridion sp.

IM

1111

4

THOMISIDAE

Misumenops sp.

IM

6

2 1

9

TOTALS No. Individuals

20 22 33 19 108 60 * 94

36 56

20 30 117

615

No. Species

5 5 4 5 7 5 * 12

7 8

6 8 12

29

Shadab 1980). This species is probably a facultative breeder in Spanish moss,
with individuals moving in and out of the microhabitat at various times of the
year.

SeventyTive years ago Rosenfeld (1911, 1912) published his Louisiana
investigations on the use of Spanish moss as an over-wintering habitat by the boll
weevil and its associated predators and competitors. From 12 samples (197 lbs)
collected in December, January, and June, he obtained 2614 insects and 287
spiders. The dominant insect was an aphid ant {Cremastogaster sp.), with 1481
individuals. Of the remaining insects, 343 (30.3%) were Anthonomus weevils and
25 (2.2%) were Nezara stink-bugs, with most of the rest beneficials such as
coccinellids and nabids. After noting the small number of insects obtained in
June, he concluded that Spanish moss “is much sought after by a large number of
insects as hibernating quarters”.

In 1930 and 1931, Rainwater (1941) collected Spanish moss from trees adjacent
to cotton fields in Louisiana and identified 139 insect species. Both the boll weevil
and the tarnished plant bug were present during late fall to early spring and
apparently “enter hibernation and survive the winter in this material”. Although
“numerous spiders were present”. Rainwater did not collect and identify them.

Based on the 1985-1986 collections, statements from Rosenfeld and Rainwater
may be broadened to indicate that arthropod populations in Spanish moss can be
both large and diverse at all times of the year. It may also be significant that a
relatively greater proportion of the arthropod population in the 1985-1986
samples was composed of spiders (predators), relative to the 1911-1912 samples.
This factor alone may be sufficient to account for the lower numbers of crop
pests in the 1985-1986 samples, though the relative proximity of cotton and the
number of trees sampled could also be important.

ACKNOWLEDGMENTS

The manuscript review provided by G. L. Bernon, G. B. Edwards, M. J.
Gaylor, R K. Lago, D. B. Richman, and W. L. Sterling is greatly appreciated.

LITERATURE CITED

Agnew, C. W., D. A. Dean and J. W. Smith. 1985. Spiders collected from peanuts and non-

agricultural habitats in the Texas west cross-timbers. Southw. Natur., 30:1-12.

Dean, D. A., W. L. Sterling and N. V. Horner. 1982. Spiders in eastern Texas cotton fields. J.

Arachnol, 10:251-260.

148

THE JOURNAL OF ARACHNOLOGY

Dondale, C.D. 1961. Revision of the aureolus group of the genus Philodromus (Araneae: Thomisidae)

in North America. Canadian Entomol., 93:199-222.

Dorris, R R. 1972. A list of spiders collected in Mississippi. Notes Arachnol. Southwest, 3:1-4.

Fernald, M. L. 1950. Gray’s Manual of Botany. American Book Co., New York. Ixiii + 1632.

Heiss, J. S. and R. T. Allen. 1986. The Gnaphosidae of Arkansas. Arkansas Agric. Exp. Sta. Bull.,
885:1-67.

Kaston, B. J. 1973. Four new species of Meiaphidippus, with notes on related jumping spiders
(Araneae: Salticidae) from the eastern and central United States. Trans. American Micros. Soc.,
92:106-122.

LeSar, C. D. and J. D. Unzicker. 1978. Soybean spiders: species composition, population densities,
and vertical distribution. Illinois Nat. Hist. Surv. Biol. Notes, 107:1-14.

Plagens, M. J. 1985. The corn field spider community: composition, structure, development and
function. Ph.D. thesis, University of Florida, Gainesville.

Platnick, N. 1974. The spider family Anyphaenidae in America north of Mexico. Bull. Mus. Comp.
ZooL, 146: 205-266.

Platnick, N. I. and M. U. Shadab. 1980. A revision of the North American spider genera Nodocion,
Litopyllus, and Synaphosus (Araneae, Gnaphosidae). American Mus. Novit., 2691:1-26.

Rainwater, C. F. 1941. Insects and spiders found in Spanish moss, gin trash, and woods trash, and on
wild cotton. U.S. Dept. Agriculture, Bur. Entomol. & Plant Quar., Bull. E-528:l-20.

Rosenfeld, A. H. 1911. Insects and spiders in Spanish moss. J. Econ. Entomol., 4:398-409.

Rosenfeld, A. H. 1912. Insects and spiders in Spanish moss (some additional data). J. Econ.
Entomol., 5:338-339.

Skinner, R. B. 1974. The relative and seasonal abundance of spiders from the herb-shrub stratum of
cotton fields and the influence of peripheral habitat on spider populations. M.S. thesis. Auburn
University, Albama.

Whitcomb, W. H. and K. Bell. 1964. Predaceous insects, spiders, and mites of Arkansas cotton fields.

Univ. Arkansas Agric. Exp. Stn. Bull., 690:1-84.

Manuscript received August 1988, revised September 1988.

Ware, A. D. and B. D. Opell. 1989. A test of the mechanical isolation hypothesis in two similar spider
species. J. Arachnol., 17:149-162.

A TEST OF THE MECHANICAL ISOLATION HYPOTHESIS
IN TWO SIMILAR SPIDER SPECIES

Amy D. Ware

Route 2, Box 1964
Bumpass, Virginia 23024 USA

and

Brent D. OpelP

Department of Biology

Virginia Polytechnic Institute and State University
Blacksburg, Virginia 24061 USA

ABSTRACT

External reproductive organs of spiders are often species-specific and are important taxonomic
characters in species identification. One explanation of this is the mechanical isolation or lock-and-key
hypothesis. It predicts that in closely related species with overlapping ranges, morphological character
displacement should occur in regions of sympatry to prevent costly interspecific copulation attempts.
To test this hypothesis, we measured homologous external genital sclerites of male Larinioides
(Nuctenea) sdopetaria and L. patagiata (Araneidae) and statistically compared their means and
variances for sympatric and allopatric regions of their distribution. Differences in both mean and
variance were observed, but the number of sclerites that differed between regions of sympatry and
allopatry was not greater than the number that differed between adjacent regions of sympatry. Thus,
these species failed to demonstrate the character displacement predicted by the mechanical isolation
hypothesis.

INTRODUCTION

In spiders, as in many invertebrate groups, genitalic differences have been
widely used to distinguish species. The “lock-and-key” mechanism of reproductive
isolation has traditionally been used to explain these differences in the sclerotized,
external genitalia of both male and female spiders. It postulates that the genitalia
are specifically shaped so that during mating they couple correctly only between
conspecifics (Mayr 1963). Although this hypothesis is commonly used to explain
the often small genitalic differences which separate species, it has never been put
to a critical test. Grasshoffs (1973a, b, 1975) descriptions of the mechanisms that
couple male and female genitalia make it plausible that small differences in one
or several genitalic parts (sclerites) could preclude mating between related species.
However, for most species the precise functions of genitalic sclerites are unknown
and the amount of difference sufficient to prevent successful mating is unclear.
Additionally, only Coyle (1985) has quantified the genitalic variation in a

'Address reprint requests to B. Opell.

150

THE JOURNAL OF ARACHNOLOGY

population of spiders and no study has analyzed the pattern of variation over a
species’ range,

Eberhard (1985) raises several objections to the traditional lock-and-key or

mechanical exclusion mechanism, the first being that it was originally proposed
for insects in which the female genitalia are relatively rigid, whereas the genitalia
in many other groups are not rigid and, therefore, probably incapable of
mechanical exclusion. Another objection is that species-specific sclerites of the
male genitalia of some species contact parts of the female genitalia that do not
differ among species, a situation inconsistent with the lock-and-key explanation.
A strong theoretical objection to the mechanical isolation hypothesis is that
selection should favor females that can identify a potential mate’s species early in
courtship (prior to physical contact), since courtship and copulation are costly for
a female (Daly 1978).

As alternatives to the lock-and-key mechanism, Eberhard (1985) suggests four
hypotheses to explain the often species-specific nature of animal genitalia: (1)
species isolation by genitalic stimuli, (2) pleiotropic effects of alleles selected in
other contexts, (3) mechanical “conflict of interest” between males and females,
and (4) sexual selection by female choice. Eberhard’s evidence more strongly
supports the latter hypothesis.

The objective of this study is to evaluate the mechanical isolation hypothesis
for species-specific genitalic differences by testing two of its predictions. If, as this
hypothesis predicts, genitalic differences prevent costly and potentially dangerous
interspecific copulation attempts between closely related species, then genitalic
features of partially sympatric species should show greater differences in regions
of sympatry than in regions of allopatry. Additionally, if two species with similar
genitalia are influenced by interspecific competition for mates, their genitalia
should be less variable within areas of sympatry than within areas of allopatry.
To test these predictions we measured and compared homologous genitalic
structures of two similar spider species.

MATERIALS AND METHODS

Criteria for selecting study species, — Before selecting study species, we
established four requirements. First, the species’ taxonomy must have been
recently revised to insure that they are valid morphological species and to provide
a current picture of their distribution. Second, the species must have similar
genitalic structures and occur in areas of sympatry and allopatry only with one
another, or both their ranges must be completely sympatric with one or more
closely related species. This is necessary to insure that any detected effects do not
result from interaction between a third species and only one of the two study
species. Third, species must have sufficient size and seasonal overlap to make
interbreeding a real possibility. Fourth, an ample number of museum specimens
must be available for study.

Surprisingly few spider species meet all of these requirements; however, two
species of Araneidae, Larinioides sclopetaria (Clerck) and Larinioides patagiata
(Clerck), qualify, the systematics of their genus having been recently revised by
Levi (1974). Subsequently, Grasshoff (1983) transferred these and two other
species from Nuctenea to Larinioides Caporiacco 1934. Although both species are

WARE & OPELL— MECHANICAL ISOLATION HYPOTHESIS

151

r.

j I 'i V,,sigisis

iii N. SCLOPETARIA • N. FWTAGIATA

Figure 1. — Ranges of Larinioides sdopetaria and Larinioides patagiaia, modified from Levi (1974).

found in Eurasia, this study focuses only on the specimens collected in North
America. Here, the ranges of both species are completely encompassed by that of
Larinioides cornuta (Clerck). Larinioides sdopetaria and L. patagiata have
regions of sympatry on both the east and west coasts and an area of allopatry for
L. patagiata between (Fig. 1).

Both the sizes and times of maturation of these two species overlap, Larinioides
sdopetaria males have a carapace length of 3.7“4.2 mm and females a carapace
length of 3. 9-4. 3 mm; L. patagiata has size ranges of 2. 9-3. 8 mm and 2. 5-4.0 mm,
respectively (Levi 1974). Size overlap can also be documented in specific
sympatric regions (blocks A-D, Fig. 4) of each species range. Within each region,
the smallest L. patagiata males had palpal femur and cymbium lengths that were
0. 8-1.0 times those of the smallest L. sdopetaria males and the largest L.
patagiata males had values that were 0.44-0.80 times those of the largest L.
patagiata males.

These two species show no evidence of temporal isolation. In Connecticut,
mature L. sdopetaria and L. patagiata can be found throughout the year (Kaston
1948). Larinioides sdopetaria is more commonly found around buildings than is
L. patagiata, although Kaston (1948) notes that the former species is also found

152

THE JOURNAL OF ARACHNOLOGY

Figure 2. — Scanning electron micrographs of Larinioides sclopetaria (A) and Larinioides patagiata

(B) left male pedipalps in ventral view. TA = terminal apophysis, C = conductor, CY = cymbium, E =

embolus, MA = median apophysis. Scale bars each represent 200 /im.

on bushes and rocks near streams. This observation could suggest either that L.
sclopetaria has a greater tolerance or is a more opportunistic species or that it
prefers a different habitat than L. patagiata.

Distributional history. — Although both L. sclopetaria and L. patagiata are
sympatric in Europe, Levi (1974) suggests that, due to its frequent association
with man-made structures, L. sclopetaria may have been introduced into North
America (L. patagiata’s native status is unquestioned). However, he also notes
that in Europe L. sclopetaria is also most commonly found on buildings. To
evaluate this species’ distributional history, we employed methods similar to those
used by Nyffeler et al. (1985). This involved examining and plotting the collecting
localities and dates of museum specimens of both species to detemine if there was
convincing evidence for the range expansion of L. sclopetaria.

Measurements. — Sclerite measurements were taken at 50X using a dissecting
stereo microscope fitted with an ocular micrometer scale. Values were recorded to
the nearest half unit, providing a resolution of 10 pm. Only measurements of
sclerites which could be positioned consistently and had well defined boundaries
were used. In order to determine which palpal indexes could be accurately taken,
three duplicate sets of preliminary measurements were taken on five males of the
same species over a period of ten days.

Four types of measurements were used in this preliminary study: overall spider
size (carapace length and first femur length), overall palp size (cymbium height
and width), sclerite dimensions, and distances separating one sclerite from
another. Coefficients of variation for all repeated measurements of the same
species except the latter were less than 0.13, indicating that only sclerite
separation could not be consistently measured.

In males of both species, one measurement (first femur length) of spider size
and 16 palpal measurements were taken. First femur length (FEE) was measured

WARE & OPELL— MECHANICAL ISOLATION HYPOTHESIS

153

CYMBiUM

EMBOLUS

C1

CONDUCTOR

retrolateral

M3

posterior

MEDIAN APOPHYSIS

Figure 3. — Male palpal sclerites (not to scale), as oriented for measurement, and the features
measured.

retrolaterally with the leg oriented perpendicular to the axis of measurement.
Palpal characters (Figs. 2 and 3) were measured on the left pedipalp after it was
removed from the body (unless the right pedipalp had been removed by a
previous examiner). For the following palpal measurements, the dimensions being
measured were oriented perpendicular to the axis of observation (directions refer
to palpal orientation): cymbium length (CYL), prolateral view; palpal femur
length (PFL), prolateral view; conductor width (Cl) and length (C2), retrolateral
view; terminal apophysis width (Al), apical (anterior) slightly retrolateral view (In
L. sclopetaria, the greatest width measurement was taken; in L. patagiata, width
was measured where the edges of the apophysis become parallel to one another.);
curvature of the terminal apophysis (A2 - the distance from the main body
beyond which the curved end projected), apical and slightly retrolateral view;
embolus length (El = from the tip to the point where the base was curved on
itself), anterior third width (E2), and basal width (E3), prolateral and ventral
views, perpendicular to axis of measurement in an anterior-posterior direction
and rotated laterally until the point where the base curved back on itself began to
be eclipsed; median apophysis projection separation (Ml) and depth (M2),
prolateral and perpendicular to the axis of measurement in a lateral direction,
rotated anterior-posteriorly until the main body just covered the upper edge of
the bottom projection; median apophysis length (M3), upper 14 width (M4),

154

THE JOURNAL OF ARACHNOLOGY

central width (M5), basal width (M6), and projection length (M7) and width
(M8), posterior and slightly ventral, then rotated until the gap between the two
projections and the main body was just barely eclipsed.

All statistical tests on these measurements were performed using SAS V (SAS
Institute, Inc., SAS Circle, PO Box 8000, Cary, NC, 2751P8000).

Establishing range blocks. — Using state lines as boundaries, species ranges were
divided into subunits (blocks) shown in Fig. 4. These blocks had similar areas
and, except for the western blocks of L. sclopetaria, were represented by
sufficient museum specimens to permit statistical tests. The scarcity of western L.
sclopetaria males (Fig. 4) does not preclude comparisons of L. patagiata in areas
of allopatry and sympatry with L. sclopetaria, as established by female specimens
of the latter species.

Preliminary tests. — Before comparative statistical tests were performed, three
preliminary steps were necessary. First, the normality of all measurements was
tested with the Kolmogorov goodness-of-fit test to assure that assumptions of the
statistical tests were met. Second, discriminate analysis was used to determine
how well specimens of each species could be assigned to the correct geographical
block. Only if most of the specimens from a block can be assigned to that block
on the basis of palpal features alone, are the proposed blocks appropriate for this
study. Third, F-tests comparing variability between the population and state
levels, and the state and block levels were performed to determine if, as

WARE & OPELL— MECHANICAL ISOLATION HYPOTHESIS

155

hypothesized, variability was greater in more inclusive units. If this is found to be
correct, it means that blocks can be analyzed without being biased by overly large
population or state variation.

Test of clinal variation. — We examined geographic variation because the
patterns of variation within a species must be understood in order to avoid the
problem of interpreting clinal variation as support for the mechanical isolation
hypothesis. A large difference in mean measurement values between regions of
sympatry and allopatry could actually be due to clinal variation, rather than the
effect of range overlap.

To reveal less conspicuous, but congruent patterns in sclerite size differences
indicative of clinal variation, we calculated sclerite measurement means for each
geographic block (Fig. 4). For L. patagiata, this created a matrix of nine rows
(blocks) and fifteen columns (one for each sclerite measurement except those
measuring overall size - FFL, CYL, PFL) and for L. sclopetaria, a matrix of
four blocks (the other two blocks included only one specimen each and were
excluded from this analysis) and fifteen sclerites was constructed. From this
matrix, we determined the number of sclerites for which each block had the
highest value, the next highest value, etc., until the ninth lowest position was
reached. These ranking variables were compiled into a summary table and from
these, consensus rankings (Table 3) for each species were derived. These rankings
were designed to detect any continuous trends in mean sclerite measurements
across the ranges of the two species. If sclerite measurements show east-west or
north-south clinal variation, consensus rankings will order the blocks of each
species into a logical, east-west or north-south geographical sequence. However,
this approach is not likely to detect other patterns, such as changes in sclerite
values from the center of a species’ range to its periphery or regional differences
due to altitude or precipitation differences.

Test of character displacement. — We evaluated evidence for character displace-
ment by comparing the means and variances of adjacent geographic blocks using
/"tests and F-tests, respectively. When variances differed significantly {P < 0.05),
/-tests for unequal variances were used.

RESULTS

Preliminary tests. — The Kolmogorov goodness-of-fit test shows the data to be
normally distributed (F < 0.05). Discriminate analysis of block divisions (Table 1)
shows a high percentage of specimens correctly identified, indicating that,
although arbitrary, these block divisions are appropriate ones for pooling the
data. Tests of variability between populations and states, and states and blocks
showed that variability was significantly greater (P < 0.05) in more inclusive
units.

Distributional history. — Museum specimens of both L. sclopetaria and L.
patagiata date to the 1860’s. Nothing about the pattern of acquisition of
additional specimens chronicles an expanding range for either species. Specimens
of both species collected in the eastern United States and Canada between 1864
and 1885 fully occupy the present-day eastern range of L. sclopetaria (Fig. 5).
Those added during the next two decades simply increase the density of points
within this range (Fig. 5), and do not, as Nyffeler, et al. (1986) found for the

156

THE JOURNAL OF ARACHNOLOGY

Table 1. — Discriminant analysis of palpal features among geographic blocks. Letters in parentheses
refer to those blocks into which the most frequently misidentified specimens were placed.

Species

Block

No.

% Identified
correctly

Highest % identified
incorrectly

L. sclopetaria

A

44

75

14(B)

B

41

90

5 (A & C)

C

40

73

12(A)

D

26

92

4 (A & B)

L. patagiata

A

30

77

13(E)

B

28

86

11 (D)

C

25

92

4 (D & E)

D

8

100

0

E

14

79

21 (D)

F

60

75

7 (B & G)

G

22

90

5 (E & F)

H

16

88

12(E)

I

14

100

0

introduced species, Steatoda bipunctata (L.); depict a range expansion from the
Northeastern coast of the United States inland and along the St. Lawrence River.
Likewise, the number of new specimens of each species added to museum
collections and the number of new state records resulting from this increase
(Table 2) does not portray the expansion of L. sclopetaria. In fact, these data
suggest that it is L. patagiata whose range expanded, a misconception explained
by an increase in the number of specimens collected from the western states.

Test of clinal variation. — For each species, the consensus ranking (Table 3) lists
in order the block that has the greatest number of sclerites placing it in the first,
second, third, etc. position. The percentage of sclerites which contributed to the
ranking of blocks was never much greater than 50% and averaged about 39%,
indicating that no strong consensus appeared in the ranking. Additionally, the
resulting consensus ranking shows no consistent trends of east-west or north-
south clinal variation in either species.

Test of character displacement. — Significant intraspecific differences in the
means of 0-8 palpal sclerite measurements were observed between adjacent blocks
of both L. sclopetaria and L. patagiata (Tables 4 and 5). However, the pattern of
sclerites that differed changed from block to block. In the eastern United States

Figure 5. — Early distributional records for Larinioides patagiata and Larinioides sclopetaria.

WARE & OPELL— MECHANICAL ISOLATION HYPOTHESIS

157

Table 2. — A historical summary of the number of new specimens and new state records of
Larinioides patagiata and Larinioides sciopetaria.

Time period

Number of specimens

Number of new state records

L. patagiata

L. sciopetaria

L. patagiata

L. sciopetaria

1864-1875

1

4 (36%)

(3)

(2) (40%)

1876-1885

12

9 (43%)

2

3 (60%)

1886-1895

0

5 (100%)

0

0

1896-1905

19

4(17%)

4

3 (43%)

1906-1915

21

24 (53%)

4

2 (33%)

1916-1925

16

5 (23%)

5

1 (17%)

1926-1935

147

18 (11%)

9

2(18%)

1936-1945

201

17 (8%)

6

0 (0%)

1946-1955

183

14 (7%)

3

0 (0%)

1956-1965

96

3 (3%)

3

1 (25%)

1966-1975

39

19 (33%)

0

3 (100%)

only M4, the median apophysis’s upper width,

differed significantly {P < 0.05)

between blocks

D and E of L. patagiata where the lock-and-key mechanism

predicts the greatest difference.

In contrast, in

blocks A through D where L.

patagiata and

L. sciopetaria are sympatric, 1-6 L. patagiata sclerites differ

significantly between adjacent

blocks. A significant overall L. patagiata size

(FFL) difference appears only

between blocks

A and B, indicating that the

interpretation of these results

is not confounded by interblock differences in

spider size.

The lock-and“key mechanism predicts that in the West the means of more L.
patagiata sclerites should differ between block I, which this species shares with L.
sciopetaria, and adjacent allopatric blocks F and H than between blocks E-H
where L. patagiata alone is found (Fig. 4). Table 5 shows that this is not the case.
In L. patagiata, only A2, the terminal apophysis' distal curvature, differs
significantly {P < 0.05) between blocks I and F and none of its sclerites differ

Table 3. — Consensus ranking of geographic blocks. This table lists in order the block that has the

greatest number of sclerites placing it in the first, second, third, etc. position.

% of Sclerites
contributing to

Species

Rank

Block

block rank

L. patagiata

1

G

53

2

F, I

20

3

F, H

33

4

A, F, H

27

5

A

33

6

C

47

7

B

40

8

B, D

27

9

E

40

L. sciopetaria

1

A

53

2

A, D

33

3

B

47

4

C

67

158

THE JOURNAL OF ARACHNOLOGY

Table 4. — Male Larinioides sclopetaria first femur and palpal sclerite means (in pm) and (standard
deviations) by geographical block. A appears between blocks whose means differ significantly {p <

0.05) and a between blocks whose variances differ significantly {p < 0.05)- Blocks F and G were
each represented by only a single specimen.

Feature

BLOCK

A

B

C

D

F

G

FFL

5852

+

5802

*

5418

♦

6024

7470

7802

(989)

(689)

(739)

(873)

CYL

1413

+

1325

+

1300

+

1348

1640

1640

(341)

(181)

(72)

(133)

Cl

253

248

+

238

*

252

300

280

(19)

(16)

(26)

(23)

C2

514

+

511

480

499

590

600

(59)

(41)

(48)

(66)

AI

185

+

180

+

186

187

240

200

(24)

(17)

(23)

(20)

A2

52

*

43

48

53

40

100

(21)

(16)

(18)

(22)

El

328

+ *

314

+

305

*

321

380

400

(35)

(17)

(31)

(27)

E2

108

+ *

99

+

102

*

111

120

120

(19)

(10)

(15)

(14)

E3

135

*

121

+ *

143

145

120

140

(15)

(14)

(25)

(21)

Ml

131

*

125

124

130

140

120

(13)

(12)

(14)

(14)

M2

242

*

232

230

*

243

260

260

(19)

(19)

(24)

(18)

M3

623

+

608

606

623

740

680

(49)

(35)

(44)

(56)

M4

241

236

229

*

243

240

240

(26)

(24)

(26)

(26)

M5

220

214

212

216

230

220

(19)

(14)

(18)

(22)

M6

306

313

306

305

320

300

(31)

(24)

(24)

To

oo

M7

229

226

*

204

*

223

280

280

(32)

(30)

(33)

(37)

M8

73

71

66

70

80

100

(12)

(13)

(15)

(16)

PFL

804

782

766

816

1000

1000

(86)

(73)

(92)

(113)

significantly between blocks I and H. In contrast, 3-8 sclerites differ significantly

between allopatric blocks E~H.

Likewise, intraspecific differences in variances do not support predictions of the
lock-and“key mechanism that greater variance should be permitted in allopatric
than sympatric regions. The variances of 1-4 (mean 2.3) L. patagiata sclerites
differs significantly {P < 0.05) between adjacent allopatric blocks A-D, as
compared with 1-4 (mean = 2.5) that differ between adjacent sympatric blocks E-
H (Table 5). The variance of only two sclerites differs significantly between
sympatric block D and block E, where L. patagiata alone is found. In the West,
the variances of 0-2 L. patagiata sclerites differs between sympatric block I and
adjacent allopatric blocks F and H. None of these observations lends support to
the lock-and-key mechanism.

WARE & OPELL— MECHANICAL ISOLATION HYPOTHESIS

159

Table 5. — Male Larinioides patagiata first femur and palpal sclerite means (in fxm) and (standard
deviations) by geographical block. A appears between the blocks whose means differ significantly
ip < 0.05) and a between the blocks whose variances differ significantly {p < 0.05).

BLOCK

Fea-

lure

A

B

C

D

E

F

G

H

I

F&H

F&I

FFL

4000

*

3666

3745

3920

3676

3689

3799

3577

3640

—

—

(333)

(336)

(480)

(263)

(321)

(377)

(364)

(358)

(389)

CYL

1335

*

1244

1251

1285

1214

+ *

1348

+ *

1422

*

1341

1341

—

+

(85)

(76)

(89)

(99)

(128)

(86)

(130)

(104)

(127)

Cl

196

*

183

186

180

184

*

213

216

208

216

—

—

(20)

(21)

(26)

(15)

(23)

(21)

(26)

(18)

(19)

C2

535

*

507

511

505

495

*

548

566

+ *

525

542

+ *

—

(42)

(41)

(42)

(54)

(52)

(55)

(59)

(35)

(40)

A1

159

+ *

140

+ 146

139

149

145

147

146

135

—

—

(31)

(19)

(31)

(24)

(19)

(26)

(33)

(21)

(18)

A2

28

21

23

+ 50

+ 19

23

24

+

21

19

+

+ *

(13)

(9)

(11)

(62)

(10)

(12)

(14)

(7)

(5)

El

310

299

304

287

304

*

338

354

347

339

—

—

(31)

(34)

(23)

(35)

(28)

(31)

(43)

(28)

(31)

E2

86

82

83

76

78

*

92

93

89

94

—

—

(12)

(12)

(11)

(7)

(12)

(12)

(16)

(13)

(14)

E3

100

+

93

* 105

97

87

*

102

+

108

+

106

101

—

—

(20)

(14)

(12)

(17)

(12)

(14)

(24)

(14)

(12)

Ml

212

206

203

210

197

*

218

*

246

*

216

208

—

—

(28)

(20)

(26)

(28)

(37)

(30)

(25)

(24)

(20)

M2

236

226

213

* 242

219

229

234

*

209

213

*

—

(34)

(37)

(32)

(25)

(30)

(29)

(28)

(29)

(31)

M3

481

+

469

480

492

447

475

+ *

506

485

504

—

—

(69)

(30)

(36)

(32)

(60)

(70)

(40)

(53)

(52)

M4

186

177

171

* 199

* 111

176

♦

193

199

186

*

—

(24)

(29)

(26)

(22)

(20)

(29)

(28)

(37)

(25)

M5

335

*

306

318

+ 315

+ 321

334

348

+

340

320

+

+

(35)

(31)

(30)

(60)

(32)

(50)

(56)

(29)

(28)

M6

272

265

278

275

253

*

283

287

290

287

—

—

(29)

(38)

(39)

(45)

(34)

(38)

(39)

(24)

(32)

M7

492

+

479

466

471

461

*

513

527

524

529

—

—

(90)

(49)

(57)

(70)

(53)

(51)

(51)

(59)

(38)

M8

72

*

94

80

74

77

+

97

93

97

101

—

—

(21)

(22)

(28)

(19)

(39)

(26)

(30)

(19)

(19)

PFL

697

*

626

653

676

638

673

+

711

*

644

673

—

—

(51)

(55)

(64)

(38)

(64)

(57)

(83)

(64)

(77)

DISCUSSION

In the absence of a clearer understanding of palpal sclerite function, the
mechanical isolating hypotheses predicts that, when adjacent blocks of a species
range are compared, a greater number of sclerites should differ between a
sympatric and an allopatric block than between two sympatric blocks. By failing
to show this difference, this study fails to demonstrate the character displacement
predicted by the lock-and-key mechanism. Even between sympatric blocks, there
is a shift in the sclerites whose means differ significantly, making it difficult to
argue that those L. patagiata sclerites that differ between regions of sympatry and
allopatry (blocks D and E and blocks I and either F or H) are of greater
importance in species isolation.

160

THE JOURNAL OF ARACHNOLOGY

There are several levels of sclerite function: general orientation, alignment, and
physical coupling. The last of these has the greatest potential to function in
reproductive isolation. Since the exact function of several of the sclerites is
unknown, we made an effort not to weight any sclerite beyond the limits of our
ability to measure it consistently. Additionally, several measurements were taken
on each sclerite to avoid neglecting any feature which might play a role in the
actual coupling of the palp with the female’s epigynum. Considering the large
number of measurements taken from each specimen, a number of parameters
involved in palp coupling were almost certainly analyzed. Two of the structures
measured (the embolus and the conductor) are known to be directly involved in
coupling (Grasshoff 1973a, b; Shear 1967). Therefore, the failure of this study to
find evidence for character displacement cannot be dismissed on the grounds that
it analyzes structures not involved in the coupling process.

This study emphasizes the danger inherent in establishing subspecies on the
basis of differences in a few characters. Although blocks were established as
arbitrary units for purposes of evaluating clinal variation, each meets the
traditional definition of a subspecies as: “an aggregate of local populations of a
species, inhabiting a geographic subdivision of the range of the species, and
differing taxonomicaily from other populations of the species” (Mayr 1963). As
spiders are not known to respect state boundaries, it is likely that these two
species’ ranges could be divided into other arbitrary units, each distinguished by
significant differences in one or more sclerites.

The use of museum specimens resulted in data from different localities and
dates being compiled into a single set. There are no assurances that specimens
found sympatrically at the species level are actually sympatric at the more
important population level. However, most of the collectors were not interested
specifically in one or the other of these large, conspicuous species and both
species probably had an equal chance of being collected if they occurred in the
same area.

Several problems are encountered in studies of character displacement (Grant
1972), the first being clinal variation. Some studies that have claimed to show
character displacement (e.g., Brown and Wilson 1956; Ficken et al. 1968) have
later been refuted because the morphological changes attributed to character
displacement could be explained by clinal variation (Grant 1972). Although this
study detected no north-south or east-west clinal variation in either species, the
block design it employs is not well suited to detect changes in sclerite values due
to local differences in such factors as altitude or precipitation. However, because
these local differences have a limited influence on a block’s values, they are less
likely to compromise the design of our study.

A second potential difficulty is that the historical events responsible for the
present distribution of most species are not well documented. If two species
evolved in isolation and then came into contact, there would be opportunity for
character displacement to occur. If they first originated sympatrically and then
one vacated part of the other’s range, there would be opportunity for character
release to occur in the vacated region. Two facts suggest that the present-day
distributions of the species studied have existed long enough to permit their
palpal sclerites to respond to such increased or relaxed selection pressure. First,
as early as the 1860’s both species’ eastern ranges appear fully occupied (Fig. 5).
Second, between adjacent blocks of each species (except F and G of L.

WARE & OPELL— MECHANICAL ISOLATION HYPOTHESIS

161

sclopetaria that were each represented by only a single specimen) there are
significant differences between sclerite means and variances (Tables 4 and 5).
However, there is the less likely possibility that these differences may result from
a series of founder effects.

Three isolating mechanisms can function to increase the efficiency of mating
where related species coexist: habitat isolation, ethological isolation, and
mechanical isolation (Bush 1975). The third problem with studies of character
displacement is that usually only one of these mechanisms is explored by the
study. Grant (1972) found that in some cases what passed for displacement
involving morphological characters was actually the result of habitat isolation
(Ripley 1959; Ficken et al. 1968). No comparative ecological data were available
for L. patagiata and L. sclopetaria. However, they may occupy slightly different
habitats, even though collected from the same locality.

Ethological isolation can also explain the separation of species. However,
Robinson and Robinson’s (1980) comprehensive study of courtship shows that
although most congeneric orb-weavers share a series of courtship behaviors, there
is much intraspecific variation. In such cases, differences in the external genitalia
of males and females would enhance reproductive isolation, even if they were not
the sole or even principal mechanism.

ACKNOWLEDGMENTS

We thank Bruce Turner, Bruce Wallace, and David West for guidance during
this study. They, Jonathan. Goddington, and William Eberhard provided helpful
comments and suggestions on earlier drafts of this manuscript. Specimens used in
this study were borrowed from the Museum of Comparative Zoology (Harvard
University), the American Museum of Natural History, the California Academy
of Sciences, and the Field Museum of Natural History. Grants from the
California Academy of Sciences Exline-Frizzell Fund for Arachnological
Research and from the Virginia Tech Department of Biology supported this
study.

LITERATURE CITED

Brown, W. L., Jr. and E. 0. Wilson. 1956. Character displacement. Syst. Zool., 5:49-64.

Bush, G. L. 1975. Modes of animal speciation. Ann. Rev. Ecol. Syst., 6:339-364.

Coyle, F. A. 1985. Two-year life cycle and low palpal character variance in a Great Smoky Mountain
population of the lamp shade spider (Araneae, Hypochilidae, Hypochilus). J. Arachnol., 13:211-
218.

Daly, M. 1978. The cost of mating. Am. Nat., 112:771-774.

Eberhard, W. G. 1985. Sexual Selection and the Evolution of Animal Genitalia. Harvard Univ. Press,
Cambridge, MA.

Ficken, R. W., M. S. Ficken and D. H. Morse, 1968. Competition and character displacement in two
sympatric pine-dwelling warblers {Dendroica, Parulidae). Evolution, 22:307-314.

Foelix, R. E. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge, MA.

Grant, P. R. 1972. Convergent and divergent character displacement. Biol. J. Linn. Soc., 4:39-68.

Grasshoff, M. 1973a. Konstruktions- und Funktionsanalyse an Kopulationsorganen einiger
Radnetzspinnen. Aufs. u. Red. Senckenberg Naturforsch. Ges., 24:129-151.

Grasshoff, M. 1973b, Bau und Mechanik der Kopulationsorgane der Radnetzspinne M angora
acalypha (Arachnida, Araneae). Z. Morphol. Tiere, 74:241-252.

162

THE JOURNAL OF ARACHNOLOGY

Grasshoff, M. 1975. Reconstruction of an evolutionary transformation: the copulatory organs of
Mangora (Arachnida, Araneae, Araneidae). Proc. 6th Int. Arachnol. Congr., Amsterdam,
IV. 1974: 12- 16.

Grasshoff, M. 1983. Larinioides Caporiacco 1934, der korrekte Name fur die sogennante Araneus
cornn/w^-Gruppe (Arachnida: Araneae). Senckenbergiana biol., 64:225-229.

Kaston. B. J. 1948. Spiders of Connecticut. Connecticut. St. Geol. Nat. Hist. Surv. Bull., 70:1-874.
Levi, H. W. 1974. The orb-weaver genera Araniella and Nuctenea (Araneae: Araneidae). Bull. Mus.
Comp. ZooL, 146:291-316.

Mayr, E. 1963. Animal Species and Evolution. Harvard Univ. Press, Cambridge, MA.

Nyffeler, M., C. D. Dondale and J. H. Redner. 1986. Evidence for displacement of a North American
spider, Steatoda borealis (Hentz), by the European species S. bipunctata (Linnaeus) (Araneae:
Theridiidae). Canadian J. ZooL, 64:867-874.

Ripley, S. D. 1959. Character displacement in Indian nuthatches (Sitta). Postilla, No. 42:1-11.
Robinson, M. H. and B. Robinson. 1980. Comparative studies of the courtship and mating behavior
of tropical araneid spiders. Pacific Insects Monograph, 36:1-217.

Shear, W. A. 1967. Expanding the palpi of male spiders. Breviora, Mus. Comp. ZooL, 259:1-27.

Manuscript received January 1988, revised November 1988.

Graze, M. J., A. A. Grigarick and K. A. Smith. 1989. Population ecology of Pardosa ramulosa
(Araneae, Lycosidae) in flooded rice fields of northern California. J. Arachnol, 17:163-170.

POPULATION ECOLOGY OF PARDOSA RAMULOSA
(ARANEAE, LYCOSIDAE) IN FLOODED RICE FIELDS
OF NORTHERN CALIFORNIA

Michael J. Oraze, Albert A. Grigarick and Kirk A. Smith

Department of Entomology
University of California
Davis, California 95616 USA

ABSTRACT

The phenology, abundance, and habitat preferences of Pardosa ramulosa (McCook) in California
rice fields were studied. Analysis of pitfall and floating sticky-trap samples showed that this spider
apparently had one generation per year and overwintered in the immature stage. The population on
the levees peaked in June and preceded the buildup in the paddies. Colonization of the paddies was
possible after emergent vegetation was present. Before this P. ramulosa was observed being blown
over the water surface. This may have contributed to its ability to disperse within the rice
agroecosystem. In the paddies, P. ramulosa was associated with patches of aquatic broadleaf weeds
where potential prey, including occasional pest species, was most abundant. P. ramulosa possesses
many attributes thought to be desirable for a natural enemy in an annual agroecosystem, and may be
an important predator of some pest species in rice.

INTRODUCTION

The importance of spiders as biological control agents has received
considerable attention over the past two decades (see Luczak 1979; Riechert and
Lockley 1984; Nyffeler and Benz 1987 for references). In spite of this, much
controversy still remains concerning their perceived benefit. Clearly, much more
information needs to be gathered (especially pertaining to the population and
feeding ecology of spiders) before any definitive statement can be made regarding
the role these generalist predators play in influencing prey levels. One habitat type
where spiders appear to be a particularly important predator group is in swamp
ecosystems, including flooded rice fields (Nyffeler and Benz 1987). We previously
reported that R ramulosa comprised ca 68% of all the spiders collected from
California rice paddies, and suggested that it was the spider most likely to
contribute to biological control in this agroecosystem (Oraze et al. 1988). Based
on this and the importance of spiders found in rice fields in other parts of the
world (Ito et al. 1962; Kiritani et al. 1972; Kiritani and Kakiya 1975; Kang and
Kiritani 1978; Chiu 1979; Kenmore 1980), we felt additional study of P. ramulosa
was warranted.

Various aspects of the population ecology of P. ramulosa have been studied in
a number of habitats (Leigh and Hunter 1969; Yeargan and Cothran 1974; Van
Dyke and Lowrie 1975; Hydorn 1977; Greenstone 1978), but little is known of its
ecology in rice. The present study was undertaken to determine the phenology.

164

THE JOURNAL OF ARACHNOLOGY

seasonal abundance and microhabitat preferences of P. ramulosa in California
rice fields so as to better understand the potential of this predator for reducing
insect pest levels.

MATERIALS AND METHODS

Phenology and seasonal abundance. — Ten pitfall and ten floating sticky traps
were used to monitor P. ramulosa population fluctuations on the levees and in
the paddies, respectively, at the Lattemore seedfield section of the Rice
Experiment Station near Biggs, (Butte County) California. The traps were
installed and serviced as described by Graze et al. (1988). Both trap types were
placed in designated locations for four to seven days once each month during the
growing season. Additional sampling with pitfall traps was conducted throughout
the fall and winter months between the 1984-85 growing seasons to determine
overwintering phenology and abundance.

Habitat preference (paddy). — The abundances of P. ramulosa and potential
prey items in vegetation types associated with the paddies were determined by
randomized complete-block design experiments conducted at the Rice Experiment
Station in 1984, and repeated with modifications in 1985. There were three
treatments (vegetation types) in 1984 and two in 1985. Four blocks were used in
both years. The experiments were located along the west margins of four adjacent
fields in 1984 and along the east margins of two adjacent fields in 1985. Plots
measured 6 by 6 m and were separated from one another by aluminum barriers
that were 38 cm high and 6 m long.

The three treatments — rice only, rice plus weeds and weeds only — were
achieved through various manipulations. In 1984 the rice was eliminated from the
weedy plots by hand removal. This was followed with the transplanting of
twenty-five ducksalad (Heteranthera limosa (Sw.) Willd.) plants. In 1985
glyphosate at 1.12 kg (AI)/ha with a 1% solution of Herbimax® oil adjuvant was
used to selectively remove the rice from the weedy plots. It was applied after the
rice but before the weeds had emerged through the water. Transplanting of weeds
was not necessary in 1985. In both years the species complex of the weedy plots
consisted primarily of ducksalad and monochoria {Monochoria vaginalis (Burm.
f.) PresL), with lesser amounts of California arrowhead {Sagittaria montevidensis
Cham, and Schlecht.), waterplantain {Alisma triviale Pursh.), and roughseed
bulrush {Scirpus mucronatus L.). The rice-only plots were treated with bentazon
or 4EC MCPA at 1.12 kg (AI)/ha as required to remove the above-mentioned
aquatic weeds. The rice plus weeds treatment of 1984 was left undisturbed.

The abundances of P ramulosa, aster leafhopper {Macrosteles fascifrons Stal)
and small flies (primarily Culicidae, Ephydridae and Chironomidae) were
obtained from floating sticky traps (one trap per plot) that were placed in the
plots for a single seven-day sampling period. Aphid densities were estimated by
taking five subsamples per plot with a UC-VAC suction device (Summers et al.
1984) and unit-area-sampler as described by Graze et al. (1988). The mean
number per sticky trap per day or the mean number per m^ within each plot were
transformed, (X + 1/2)*^^ and analyzed with a two-way analysis of variance
(1984) or paired Mest (1985).

GRAZE ET AL.— POPULATION ECOLOGY OF R RAMULOSA IN RICE

165

1984 1985

Figure 1. — Phenology and abundance of Pardosa ramulosa on rice field levees for 1984-85. Data

obtained from a 4-7 day sampling period each month. Adults were separated from immatures by the
presence of a fully developed palpal organ (males) or epigynum (females).

RESULTS AND DISCUSSION

Phenology and seasonal abundance. — It appears that Pardosa ramulosa had a
single generation per year and overwintered in the immature stage, based on
detailed analysis of pitfalLtrap samples for 1984-85 (Fig. 1). These data, along
with other data from trap catches, should be interpreted cautiously, as they reflect
spider activity as well as abundance. Even so, similar conclusions have been
reported for P. ramulosa in other habitats of northern California (Yeargan and
Cothran 1974; Hydorn 1977).

The population peak on the levees occurred in June in each of the three years
we sampled. In addition, the population decline on the levees coincided to some
degree with the rise of the population in the paddies (Fig. 2A and B). This
suggests that movement from the levees into the paddies contributed, at least in
part, to the decline of the levee population. Similar seasonal movements have
been described for P. ramulosa in other habitats (Hydorn 1977).

cd

"d

\

a.

cd

o

m

\

d

d

fcUD

1983 1984 1985

Figure 2. — Seasonal abundance of Pardosa ramulosa (adults and immatures) on levees (A), and in

paddies (B) of flooded rice fields. Vertical lines represent SEM. No measure of variation could be
calculated in “A” because of the combining of individual trap contents prior to counting.

The presumed dispersal from the levees into the paddies may be related to the
seasonal succession of the paddy environment. Early in the cropping cycle, before
emergent vegetation exists in the paddies, the spiders were apparently limited to
the terrestrial habitats of the agroecosystem. If a spider was to walk into the

GRAZE ET AL.— POPULATION ECOLOGY OF P. RAMULOSA IN RICE

167

paddy at this time it would be blown across the water surface. This “sailing” was
frequently observed as one of us (M.J.O.) walked along the paddy margins on a
breezy day. A similar phenomenon was observed in a California salt marsh where
P. ramulosa apparently “floated free” during intermittent tidal flooding. This
appeared to have a pronounced influence on the daily movements of these spiders
(Greenstone 1979a). “Sailing” may be the primary means by which P. ramulosa
colonizes rice fields. Ballooning may also be a means of colonization, although
the peak ballooning period of P. ramulosa does not coincide with the rice
growing season (Yeargan 1975a). If P. ramulosa can successfully cross a flooded
paddy by “sailing,” then it would be possible for this species to quickly infiltrate
large expanses of the agroecosystem. These spiders would probably initially
accumulate on the margins and levees. Later, as plants emerge through the water,
they could diffuse into or actively select the paddy habitat without being
displaced by the wind, allowing the paddy population to become established and
increase.

The abundance of P. ramulosa in the paddy typically declined late in the
summer before the draining of the fields in September (Fig. 2B). Several factors
may be responsible. Some that were observed, although not documented,
included: cannibalism, possible interference and resource competition with
another wolf spider, Pirata piraticus (Clerck) and predation from spider wasps
(Pompilidae). Determination of the relative importance of these and other factors
influencing the population dynamics of P. ramulosa in rice should be a
challenging yet worthwhile endeavor.

A review of similar studies shows that P. ramulosa is abundant in a wide
variety of habitats throughout California. These include annual agroecosystems
such as rice (Graze et al. 1988) and cotton (Leigh and Hunter 1969; Hickle 1981),
perennial agroecosystems such as alfalfa (Yeargan and Dondale 1974; Hickle
1981), relatively undisturbed areas with “wild” vegetation such as salt marshes
(Garcia and Schlinger 1972; Greenstone 1978) and sewage oxidation ponds
(Hydorn 1977), and finally backyard lawns (Van Dyke and Lowrie 1975). The
ability of P. ramulosa to inhabit such a diverse array of habitats (from coastal
salt marshes to irrigated desert cotton fields) indicates that it probably possesses
broad physiological tolerances and flexible behavior patterns. One might expect
to find this spider wherever its moisture requirements are satisfied within its
range. This may include more habitats than would be initially obvious. For
example, soil cracks appeared to provide P. ramulosa refuge between irrigations
in alfalfa and cotton fields (Yeargan and Cothran 1974; Hickle 1981) and in some
ephemeral aquatic habitats such as vernal ponds (Hydorn 1977; Greenstone 1980).

Habitat preference (paddy). — Plant type in the paddy had a significant
influence on P. ramulosa abundance. P ramulosa and nearly all of the potential
prey types sampled were significantly more abundant in the weedy plots in both
the 1984 and 1985 experiments (Tables 1 and 2). The weeds in the rice plus weeds
treatment of the 1984 experiment eventually perished. This resulted in plots that
were essentially the same as the rice-only plots by the time the samples were
taken. There were no significant differences in the arthropod species sampled
between these two treatments. Because of this, the rice plus weeds treatment was
not included in the 1985 experiment.

Pardosa ramulosa associated in microhabitats of the paddy where potential
prey, some of which are occasional pests of rice (e.g., aster leafhopper.

168

THE JOURNAL OF ARACHNOLOGY

Table 1. — Mean number ± SEM of Pardosa ramulosa and potential prey in vegetation types
associated with rice field paddies. Biggs, Calif. 1984. Samples were taken on 27 July 1984. Column
means followed by the same letter are not significantly different {P < 0.01, Tukey’s HSD method)
(Systat Inc. 1987). Transformed data, (X + 1/2)'^^, were analyzed with a two-way ANOVA, df - 2,3.

Vegetation type

Avg. no. per sticky trap per day

Avg. no. per m^

Par ado sa ramulosa

Aster leafhopper

Small flies

Aphids

Rice

0.6 ± 0.1 a

0.4 + 0.3 a

4.5 + 1.5 a

208.3 ± 125.8 a

Rice + weeds

0.3 ± 0.1 a

0.2 + 0.1 a

4.8 + 1.2 a

225.5 ± 110.4 a

Weeds

2.8 + 0.3 b

11.3 + 1.6 b

41.5 + 4.1 b

5795.3 + 3423.2 b

F statistic

73.221

55.148

82.933

8.825

P value

<0.000

<0.000

<0.000

0.018

mosquitoes, seed midges and leaf miners), were most abundant. This behavior
does not appear to be unusual for spiders (Riechert and Lockley 1984). However,
in a related study Greenstone (1978) found nb correlation between prey
availability and P. ramulosa density among the small pools from which he
sampled in an estuarine salt marsh. He noted that the spiders reached their
highest densities along the margins of the pools, irrespective of prey availability,
even though this varied enormously among pools. If P. ramulosa actively selected
weedy areas of the rice paddy, as the data of this study suggest, then it is not
clear exactly to what factor(s) (e.g., prey availability, temperature-humidity
regime, cover for protection, etc.) P ramulosa was responding. This would seem
to be a fruitful area for future research.

Pardosa ramulosa exhibited characteristics that are thought to be desirable for
a natural enemy in an annual agroecosystem (Ehler and Miller 1978). It was a
good colonizer, probably because of its “sailing” ability. It appears to have broad
physiological tolerances and the behavioral flexibility (based on its occurrence in
a number of very diverse ecosystems, as was discussed earlier) that probably help
it to survive the adverse conditions that typically exist during the early phases of
a cropping cycle. In addition, perceived limitations of generalist predators (like P.
ramulosa) such as cannibalism, territoriality (Hydorn 1977) and polyphagy
(Yeargan 1975b; Greenstone 1979b) might also enhance early season survivability
by mitigating the effects of temporal shortages of preferred prey. Finally, P.
ramulosa preceded potential pests into the paddy and later associated in
microhabitats where they were most abundant. Based on these findings, the role
of P. ramulosa in reducing levels of selected pests in California rice should be
investigated.

Table 2. — Mean number ± SEM of Pardosa ramulosa and potential prey in vegetation types
associated with rice field paddies. Biggs, Calif. 1985. Samples were taken on 30 July 1985. Column
means followed by the same letter are not significantly different {P < 0.05). Transformed data, (X +
1/2)'^^, were analyzed with a paired t-test, df - 1,3.

Vegetation type

Avg. no. per sticky trap per day

Avg. no. per m^

Paradosa ramulosa

Aster leafhopper

Small flies

Aphids

Rice

1.0 ± 0.1 a

14.0 ± 3.3 a

21.7 + 8.4 a

22.2 + 6.2 a

Weeds

2.9 + 0.6 b

38.9 ± 10.0 b

51.5 ± 13.2 a

474.3 + 169.8 b

t statistic

23.392

11.603

4.108

14.284

P value

0.017

0.042

0.136

0.032

GRAZE ET AL.— POPULATION ECOLOGY OF P. RAMULOSA IN RICE

169

ACKNOWLEDGMENTS

We thank O. G. Bacon, W. Johnson, and B. Brandon for technical assistance;
R. Stein and J. Lynch for their help with collecting and counting samples; the
staff at the Rice Experiment Station for their cooperation; and R. Karban, D.
Spiller and L. E. Ehler for reviewing the manuscript. This work was supported by
grants from the Entomology Department, Univ. of California, Davis and the
California Rice Research Board. This paper is part of a Dissertation submitted by
M.J.O. in partial fulfillment of a Ph.D. requirement.

LITERATURE CITED

Chiu, S. 1979. Biological control of the brown planthopper. Pp. 335-355, In Brown Planthopper:

Threat to Rice Production in Asia. International Rice Research Institute, Los Banos, Philippines.
Ehler, L. E. and J. C. Miller, 1978. Biological control in temporary agroecosystems. Entomophaga,
23:207-212.

Garcia, R. and E. I. Schlinger. 1972. Studies of spider predation on Aedes dorsalis (Meigen) in a salt
marsh. Pp. 117-118, In Proceedings and Papers 40th Annual Conference California Mosquito
Control Association.

Greenstone, M. H. 1978. A numerical response to prey availability of Pardosa ramulosa (McCook)
(Araneae: Lycosidae) and its relationship to the role of spiders in the balance of nature. Symp.
Zool. Soc. London, 42:183-193.

Greenstone, M, H. 1979a. A line transect density index for wolf spiders {Pardosa spp.), and a note on
the applicability of catch per unit effort methods to entomological studies. Ecol. EntomoL, 4:23-29.
Greenstone, M. H. 1979b. Spider feeding behavior optimizes dietary essential amino acid composition.

Nature, 282:501-503.

Greenstone, M. H. 1980. Contiguous allotopy of Pardosa ramulosa and Pardosa tuoba (Araneae:
Lycosidae) in the San Francisco Bay region, and its implications for patterns of resource
partitioning in the genus, American Midi. Nat., 104:305-311.

Hickle, L. A. 1981. The biology and ecology of Pardosa ramulosa (McCook), (Araneae, Lycosidae),
in Imperial Valley, California. Ph.D. Dissertation, Univ. of California, Riverside.

Hydorn, S. E. 1977. The biology of Pardosa ramulosa. Ph.D. Dissertation, Univ. of California,
Berkeley.

Ito, Y., K. Miyashita and K. Sekiguchi. 1962. Studies of the predators of the rice crop insect pests,
using the insecticidal check method. Japanese J. Ecol., 12:1-11.

Kang, J. and K. Kiritani. 1978, Winter mortality of the green rice leafhopper {Nephotettix cincticeps
Uhler) caused by predation. Japanese J. Appl. Entomol. ZooL, 22:243-249.

Kenmore, P. E. 1980. Ecology and outbreaks of a tropical insect pest of the green revolution. Ph.D.
Dissertation, Univ. of California, Berkeley.

Kiritani, K. and N. Kakiya. 1975. An analysis of the predator-prey system in the paddy field. Res.
Popul. Ecol., 17:29-38.

Kiritani, K., S. Kawahara, T. Sasaba and F. Nakasuji. 1972. Quantitative evaluation of predation by
spiders on the green rice leafhopper, Nephotettix cincticeps Uhler, by a sight-count method. Res.
Popul. Ecol, 13:187-200.

Leigh, T. F. and R. E. Hunter. 1969. Predacious spiders in California cotton. California Agric., 23:4-5.
Luczak, J. 1979. Spiders in Agrocoenoses. Polish Ecol. Studies, 5:151-200.

Nyffeler, M. and G. Benz. 1987. Spiders in natural pest control: A review. J. Appl. Entomol., 103:321-
339.

Graze, M. J., A. A. Grigarick, J. H. Lynch and K. A. Smith. 1988. Spider fauna of flooded rice fields
in northern California. J. Arachnol., 16:331-337.

Riechert, S. E. and T. Lockley. 1984. Spiders as biological control agents. Annu. Rev. Entomol,,
29:299-320.

Summers, C. G., R. E. Garrett and F G. Zalom. 1984. New suction device for sampling arthropod
populations. J. Econ. Entomol., 77:817-823.

Systat Inc. 1987. Systat User’s Guide. Systat Inc. Evanston, IL.

170

THE JOURNAL OF ARACHNOLOGY

Van Dyke, D. and D. C. Lowrie. 1975. Comparative life histories of the wolf spiders Pardosa
ramulosa and P. sierra (Araneae: Lycosidae). Southw. Nat., 20:29-44.

Yeargan, K. V. and W. R. Cothran. 1974. Population studies of Pardosa ramulosa (McCook) and
other common spiders in alfalfa. Environ. Entomol., 3:989-993.

Yeargan, K. V. and C. D. Dondale. 1974. The spider fauna of alfalfa fields in northern California.
Ann. Entomol. Soc. America, 67:681-682.

Yeargan, K. V. 1975a. Factors influencing the aerial dispersal of spiders (Arachnida, Araneida), J.
Kansas Entomol. Soc., 48:403-408.

Yeargan, K. V. 1975b. Prey and periodicity of Pardosa ramulosa (McCook) in alfalfa. Environ.
Entomol., 4:137-141.

Manuscript received August 1988, revised November 1988.

Gillespie, R. G. 1989. Diet-induced color change in the Hawaiian happy-face spider Theridion
gra//a?or (Araneae, Theridiidae). J, ArachnoL, 17:171-177.

DIET INDUCED COLOR CHANGE IN THE HAWAIIAN
HAPPY-FACE SPIDER THERIDION GRALLATOR,
(ARANEAE, THERIDIIDAE)

Rosemary G. Gillespie^

Department of Entomology, University of Hawaii,
Honolulu, Hawaii 96822 USA

ABSTRACT

The Hawaiian happy-face spider Theridion grallator Simon is a small spider, endemic to Hawaii,

where it is found under leaves in the wet and mesic forests. The abdomen is pale, translucent yellow,
but variable amounts of red, black or white pigment may be superimposed on this to generate a host
of patterned morphs. The translucence of the abdomen may enhance crypsis against predators
searching the underside of leaves; the variability in the superimposed pattern may serve to counteract
the development of a search image by the predator. The present study documents plasticity in base
coloration, which can change rapidly and markedly following ingestion of certain types of prey. This
may be merely a consequence of abdominal translucence. But it is interesting to note that it adds a
whole new dimension to the color polymorphism of the species.

INTRODUCTION

Color change in animals is a widespread phenomenon. It is generally associated
with a change in physiological state. This in turn may be induced by ontogenetic
or environmental changes, or stress. In spiders, color change is known to occur
under a variety of circumstances. Ontogenetic modifictions are widespread, with
the adult coloration being attained in the final molt (Bonnet 1933; Homann 1946;
Millot 1949). More rapid changes in color pattern have been noted in spiders of
the family Araneidae, which accumulate guanine beneath the cuticle during
periods of starvation, thereby developing a pattern of opaque white blotches over
the abdomen (Foelix 1979).

Reversible color change has been most extensively documented in the crab

spider Misumena vatia (Clerck) (Packard 1905; Gadeau de Kerville 1907; Rabaud
1923; Gabritschevsky 1927; Weigel 1941; Hinton 1976). This spider is usually
whitish and sits on white flowers. If it moves to a flower of another color it can,
by transferring a liquid pigmented material to the cuticle, change to the color of
its new substrate. However, the transformation takes about 10 days to occur, and
it is limited to pastel yellows and pinks. Similar reversible color changes have
been found to occur in other crab spiders (Heckel 1891; Bristowe 1958) as well as
the lynx spider Peucetia viridans (Hentz) (Oxyopidae) (Neck 1978).

The most rapid reversible changes to date are found in various Araneid spiders,
as a result of migration of chromatic inclusions or retraction of guanocytes

'Current address: Department of Biology, The University of the South, Sewanee, Tennessee 37375
USA.

172

THE JOURNAL OF ARACHNOLOGY

(Blanke 1975). The Australian Phonognatha wagneri Simon, for example, drops
from its web when disturbed, the abdomen simultaneously changing from cream
to brown (Roberts 1936). A similar phenomenon has been documented in
Leucauge subgemma Bos. et Str. and Chrysso venusta (Yaginuma) (Uyemura
1957), Floronia bucculenta (Clerck) (Bristowe 1958), Gea heptagon (Hentz)
(Sabath 1969) and Cyrtophora cicatrosa (Stoliczka) (Blanke 1975).

Rapid color change following dietary ingestion has not been documented in
spiders. In this paper I demonstrate the phenomenon in the Hawaiian happy-face
spider Theridion grallator (Simon). T. grallator is a small (up to 4.5 mm) spider
endemic to the Hawaiian Islands, where it inhabits the underside of leaves of a
variety of plants, especially the native Broussaisia arguta (Saxifragaceae) and
Clermontia arborescens (Campanulaceae). The spider has a base color of pale,
translucent yellow, but may exhibit a variety of abdominal color patterns (red,
black or white) superimposed on this base color. The predominant color morph
has no pattern (or a series of small black dots only); patterned morphs, although
generally less common, are diverse in form and extent of pigment.

The study was initiated following a chance observation: while monitoring
marked spiders (paint marks on certain legs), I observed an individual change
base coloration from translucent yellow to bright orange. Here I examine the
nature of diet-induced changes in base color, and possible functions of the
phenomenon. To what extent can spiders change the base color of the abdomen,
and do prey items that are normally included in their diet induce reversible color
change? The polymorphic nature of the pattern superimposed on the base color is
considered elsewhere (Gillespie and Tabashnik in press).

METHODS

The rate and extent to which T. grallator is capable of changing its color was
determined by keeping individuals in vials and allowing them to drink from
cotton swabs saturated with food coloring (red, green, blue and black). Spiders in
control vials had cotton swabs saturated with water only.

Individuals of T. grallator were then given a variety of insects that they might
capture in their natural environment. These were primrily Diptera: Drosophilidae
in the size range 2-4 mm (A = 11) and 5-6 mm (A = 5) and Dolichopodidae in
the size range 2-4 mm (A = 9). Lepidoptera (one adult moth and two larvae),
Homoptera (Cicadellidae, A = 3) and Araneae (Theridiidae, A = 1) were also
used to determine their effect on the base color of T. grallator.

Field observations were made on populations of T grallator in native forest
(mixed Ohi’a — Koa) in the Nature Conservancy of Hawaii’s Waikamoi Preserve
on Maui, near the common boundary of the Preserve and the Makawao State
Forest at 1360 m elevation. Spiders were located by thorough scrutiny of leaves
of Broussaisia arguta and Clermontia arborescens in an area of approximately 1
hectare, and were marked with small dots of paint on a leg. Individuals were
monitored once every two days between October 1987 and April 1988. Whenever
a spider was found to be feeding on prey, it was watched for any subsequent
color change. The duration of the color change (the period from the initial
observation of feeding until it reverted to the original translucent yellow) was
determined by checking color on sebsequent days. Color changes in marked

GILLESPIE— DIET-INDUCED COLOR CHANGE IN THERIDION

173

Table 1. — Change in color from pale yellow as a consequence of prey ingestion by the Hawaiian
happy-face spider Theridion grallator. Spiders were fed prey collected from the Waikamoi study site.
The color change induced in the opisthosoma of the spider by a particular prey item, and the duration
of the color change before the spider reverted to pale yellow, were all recorded. N = number of
spiders observed.

Prey

N

Color change

Duration (days)

DIPTERA

Drosophila, 2-4 mm

11

Orange

2

Drosophila, 5-6 mm

5

Orange

3

Dolichopodidae, 2-4 mm

9

Orange

2

LEPIDOPTERA

Adult moth

1

Dark orange

Caterpillar

2

Green

4-6

HOMOPTERA

Cicadellidae, 2-4 mm

3

Green

2-3

ARANEAE

Theridiidae, 2-4 mm

1

Orange

2

spiders were also monitored when feeding was not directly observed. This allowed
an estimate of the range of base colors the spiders could exhibit under natural
conditions, and their relative frequency.

In order to determine the generality of the phenomenon of color change
following dietary ingestion, species from two other genera of the Theridiidae and
one from the closely related Nesticidae, were tested for color changes induced by
intake of food coloring. The species were chosen on the basis of degree of
pigmentation of the abdomen, and were all from southeastern Tennessee. Nesticus
barri Gertsch is a pale, translucent yellow, esentially eyeless troglobite; it was
collected from Lost Cove Cave, near Sewanee, Tennessee. Achaearanea
tepidariorum (C. L. Koch), an extremely common spider found throughout the
United States, has a dirty white abdomen with gray markings. Latrodectus
mactans (Fabricius) has a shiny black abdomen, with a red “hourglass” mark on
the venter. Both A. tepidariorum and L. mactans were collected in Sewanee,
Tennessee. Individuals of each of these species (five N. barri, five A. tepidariorum
and two L. mactans) were placed in vials with swabs of red and green food
coloring (and water as control). Spiders were observed as they ingested the food
coloring, and monitored for changes in color over the next 7 days. Individuals
were then exposed to the second food color.

RESULTS

When placed in a vial, T. grallator readily ingested food coloring from the
cotton swabs. The color could be seen running into the abdominal section of the
intestine within 5-10 s, and infusing it with color within 1 min. The dye then
accumulated in the Malpighian tubules before being excreted after an average of
3.5 (SD = 1.6) days.

A similar effect was found when T. grallator was fed prey collected from the
Waikamoi study site. Table 1 summarizes the color changes that were observed.
The color generally changed from translucent yellow to either orange (following
ingestion of various Diptera or adult Lepidoptera) or green (after ingestion of
larval Lepidoptera and adult Homoptera). The abdomen retained the color for 2-

174

THE JOURNAL OF ARACHNOLOGY

Table 2. — Overall frequency of color changes in natural populations of T grallator: marked
individuals that were found to have changed color, whether or not the prey capture event itself was
observed {N = 82). Where prey capture was observed {N = 11), the approximate duration of each
color change (to the nearest day) was determined. The prey inducing the change of color to chocolate
was observed on neither occasion; but both these spiders retained their color for 7 days after the
change was initially observed.

Color

Proportion

Average duration (days)

Orange

81%

2-4

Green

17%

4-6

Chocolate

2%

>7

6 days, retention time probably being a function of the amount of pigment
ingested.

The frequency of color changes in natural populations of T. grallator (marked
individuals that were found to have changed color, whether or not the prey
capture event itself was observed) are shown in Table 2. As can be seen, the most
common color change was that from translucent yellow to orange, reflecting the
fact that small dipterans comprise approximately 70% of the dietary intake of T.
grallator (Gon 1985). On two occasions, the base color was found to have
changed to very dark chocolate brown. It is not known what prey item these
individuals had consumed.

The three other species tested for susceptibility to color change following
ingestion of food coloring varied widely in their response. N. barri changed color
in a manner similar to T. grallator. The red and green food coloring suffused the
abdomen within 1 minute; the dye was excreted after 2-3 days (average 2.7 days).
A. tepidariorum showed a definite, but very much weaker, response. The colors
became interspersed between the white dots of guanine, giving a mottled
appearance to the abdomen. In L. mactans, the food coloring could be seen as a
slight red or green tinge to the black abdomen. The color did not suffuse the
abdomen, but rather appeared as a folium, the pattern probably dictated by the
digestive diverticula.

DISCUSSION

The Hawaiian happy-face spider is capable of passively changing base
coloration according to dietary intake. Such rapid and reversible color changes
following ingestion have never before been documented in spiders. But, as I have
shown here, other species of spiders can also change color when they ingest dyes.
The degree to which spiders are capable of diet-induced color change appears to
be a function of their translucence. The heavily pigmented L. mactans showed a
barely discernible color change. In A. tepidariorum, the ingested dye was seen as
spots interspersed between the natural pigment and guanine crystals on the
abdomen. Under natural conditions, pigments from prey will tend to be ingested
in much smaller amounts, and their effects on abdominal coloration considerably
more subtle. In N. barri, however, the ingested dye suffused the entire abdomen
in a manner similar to that observed for T. grallator. N. barri may also
demonstrate observable color change under natural conditions if it ingests
pigmented prey. Indeed, although prey capture was not observed, two individuals

GILLESPIE— DIET-INDUCED COLOR CHANGE IN THERIDION

175

of this species have been found in the cave with black (as opposed to the usual
pale, translucent yellow) abdomens.

What are the selective forces responsible for degree of pigmentation in spiders?
Nearly all spiders — in common with insects — have some kind of integumental
pigment, which may serve to protect them from radiation (ultraviolet), to regulate
temperature, and may also function in protection from predation (Holl 1987).
Complete depigmentation is a characteristic of obligate cavernicoles (Gertsch
1984). Yet T grallator, an inhabitant of the Hawaiian forests, has a translucent
base color similar to the troglobitic N. barri. What evolutionary forces might be
implicated in the loss of pigment in this terrestrial species?

One of the most abundant groups of insectivores in Hawaiian forests are the
honeycreepers (Perkins 1913). In Maui’s Waikamoi Preserve, the most common
insect gleaner is the Maui Creeper, Parareomyza montana, whose behavior of
searching the underside of leaves resembles that of the creeper family Certhidae
(Scott et al. 1986). T. grallator occupies the underside of leaves. To the human
eye, the translucence of these spiders renders them almost invisible against the
light filtering through the leaf when viewed from below. Birds may encounter the
same difficulty when searching for prey. Avoidance of predation may well have
played a major role in the depigmentation of T. grallator.

Diet-induced color change in T. grallator may merely be an inevitable
consequence of its translucence. Yet it is interesting to note that this spider is
highly polymorphic. Although the base color of the abdomen is translucent
yellow and many individuals exhibit this base color alone, others exhibit a host of
superimposed patterns of red, black and white patches. These patterns are under
genetic control and are inherited as a simple Mendelian trait (Gillespie and
Tabashnik in press). Diet-induced color change adds an entirely new dimension to
the enormous array of color morphs that the species can exhibit. Could there be
any advantage to an individual in widening its spectrum of color morphs?

If there is any selective force operational here, it is most likely to be the
avoidance of predtion by birds. A common tendency in birds is the development
of a search image (Clarke 1962; Allen 1974, 1976; Murdoch and Oaten 1975;
Atkinson and Warwick 1983; Greenwood 1984; but see Guilford and Dawkins
1987). Studies have implicated the development of such search images in the
generation of interindividual variability in color patterns of the prey involved
(Endler 1978; Rettenmeyer 1970). Selection under these conditions will tend to
favor the less common morphs, which will consequently increase until numbers
are sufficient to allow the avian predators to develop a search image towards
them. It is possible that strong selection for abnormal color patterns is the
primary selective force generating variability in coloration of T. grallator, whether
this is genetically determined or environmentally plastic — a passive consequence
of dietary ingestion.

ACKNOWLEDGMENTS

The study was inspired and instigated through the work of Dr. Sam Gon. It
would have been quite impossible without an enormous amount of help from the
Nature Conservancy of Hawaii: Rob Rydell encouraged the initial efforts; Alan
Holt, Paul Higashino and Mark White allowed progress to continue throughout

176

THE JOURNAL OF ARACHNOLOGY

the year. Dr. Lloyd Loope and Art Medeiros (Haleakala National Park) gave
immense support and encouragement throughout the study. Thanks also to Bill
Mull for invaluable advise and discussion and to Dr. John Bordley and the
University of the South for use of computer equipment. Also David Wise and
Jan Weaver for comments on the first draft. Dr. Chris Parrish helped in
completion of the study. The work was funded by Whitehall Foundation grant
J86-47 to Dr. Bruce Tabashnik, with additional support from the Nature
Conservancy of Hawaii and the National Park Service.

LITERATURE CITED

Allen, J. A. 1974. Further evidence for apostatic selection by wild passerine birds: training
experiments. Heredity, 33:361-372.

Allen, J. A. 1976. Further evidence for apostatic selection by wild passerine birds: 9-1 experiments.
Heredity, 36:173-180.

Atkinson, W. D. and T. Warwick. 1983. The role of selection in the colour polymorphism of Littorina
rudis Maton and Littorina arcana Hannaford-Ellis (Prosobranchia: Littorinidae). Biol. J. Linn.
Soc., 20:137-151.

Blanke, R. 1975. Die Bedeutung der Guanocyten fiir den physiologischen Farbwechsel bei Cyrtophora
cicatrosa (Arachnida: Araneidae). Entomol. Germ., 2:1-6.

Bonnet, P. 1933. Cycle vital de Philaeus chrysops Poda. Arch. Zool. Exp. Gen., 71:129-144.

Bristowe, W. S. 1958. The World of Spiders. Collins, London.

Clarke, B. 1962. Balanced polymorphism and the diversity of sympatric species. Pp. 47-70, In
Taxonomy and Geography (D. Nichols, ed.). Oxford: Systematics Association.

Endler, J. A. 1978. A predator’s view of animal color patterns. Evol. Biol., 11:319-364.

Foelix, R. F. 1979. Biologie der Spinnen. Thieme, Stuttgart.

Gabritschevsky, E. 1927. Experiments on color change and regeneration in the crab spider Misumena
vatia. J. Exp. Zool. Phil, 47:251-266.

Gadeau de Kerville, H. 1907. Sur I’homochromie protectrice des femelles de Misumena vatia Clerck.
Bull. Soc. Entomol. France, 1907:145-146.

Gertsch, W. J. 1984. The spider family Nesticidae (Araneae) in North America, Central America and
the West Indies. Bulletin 31. Texas Memorial Museum, Univ. of Texas, Austin, pp. 1-91.

Gillespie, R. G. and B. E. Tabashnik. In press. What makes a happy face? Determinants of colour
pattern in the Hawaiian Happy-face spider Theridion grallator. Heredity.

Gon, S. M. III. 1985. Comparative behavioral ecology of the spider Theridion grallator Simon,
(Araneae: Theridiidae) in the Hawaiian archipelago. Ph. D. dissertation, University of California,
Davis.

Greenwood, J. J. D. 1984. The functional basis of frequency dependent food selection. Biol. J. Linn.
Soc., 23:177-199.

Guildford, T. and M. S. Dawkins. 1987. Search images not proven: a reappraisal of recent evidence.
Anim. Behav., 35:1838-1845.

Heckel, E. 1891. Sur le mimetisme de Thomisus onustus. Bull. Soc. Fr. Belgium, 23:347-354.

Hinton, H. E. 1976. Possible significance of the red patches of the female crab spider, Misumena
vatia. J. Zool. (London), 180:35-39.

Holl, A. 1987. Coloration and chromes. Pp. 16-25, In Ecophysiology of Spiders. (W. Nentwig, ed.)
Springer-Verlag, New York.

Homann, H. 1946. Uber die Jugendform von Micromata viridissima (Deg.) (Araneae). Bio. Zentralbk,
65:82-83.

Millot, J. 1949. Ordre des araneides (Araneae). Pp. 589-743, In Traite de Zoologie. (P. Grasse, ed.).
vol. VI. Masson, Paris.

Murdoch, W. W. and A. Oaten. 1975. Predation and population stability. Adv. Ecol. Res., 9:1-131.
Neck, R. W. 1978. Reddish coloration in a green spider: evolutionary origin and subsequent
adaptation. J. Zool. (London), 184:267-269.

Packard, A. S. 1905. Change of color and protective coloration in a flower spider {Misumena vatia
Thorell). J. New York Entomol. Soc., 13:85-86.

GILLESPIE— DIET4NDUCED COLOR CHANGE IN THERIDION

177

Perkins, R. C. L. 1913, Introduction (to Fauna Hawaiiensis). In Fauna Hawaiiensis Vol 1: xv-
ccxxviii. (D. Sharp, ed.). Cambridge Univ. Press, Cambridge.

Rabaud, E. 1923. Recherches sur la variation chromatique et Fhomochromie des arthropodes
terrestres. Bull. Biol. Fr. Belgium, 57:1-59.

Rettenmeyer, C. W. 1970. Insect mimicry. Ann. Rev, Entomol., 15:43-74.

Roberts, N, L. 1936. Colour change in the leaf curling spider {Araneus wagneri). Proc. R. Zool. Soc.
New South Wales, 28-29.

Sabath, L. E. 1969. Color change and life history observations in the spider Gea heptagon (Araneae:

Araneidae). Psyche, 76:367-374.

Scott, J. M,, S. Mountainspring, F. L. Ramsey and C. B, Kepler. 1986. Forest bird communities of

the Hawaiian islands: Their Dynamics, Ecology and Conservation. Studies in Avian Biology, 9.

Cooper Ornithological Society, Los Angeles.

Uyemura, T. 1957. Colour change of two species of Japanese spiders. Acta ArachnoL, 15:1-10.

Weigel, G. 1941. Farbung und Farbwechsel der Krabbenspinne Misumena vatia (L.) Z. Vergl.

Physiol, 29:195-248.

Manuscript received September 1988, revised November 1988.

■5':‘ ." ' -r I ■ r-*. I • ' ■■ ■ ■^' '

C*-' ^ *■ ‘ i . : . " ' ■ ■ ,

• ^%1 i ' ‘ -'I* • vV ■ • : ' 'p"p

V';. » :»tf ■■ ■^'^- ^ ^

4/*;_!

1U?J|

y«L 'x

P, ..:

4<>-

* -

j--

•.f - -• *«♦ *

• ■ *.1^

■ :»-

•■V:'

:> '''--I' ■'* •-'^’'9,-^-*

'■ 4 .-.'I .‘;v^ \ ^';,VVwSf- .i''

' ...r-'t ^-i»^ ^"'■

nK .r-r

" "

A»s-'

"if- . -.^v ., .^. ■■ '■ ( —

• n'v ■'^« . '• <'.<>v; '^'v m

^ ii. if -t^.^ i.-“.

f- ? J fv^ ‘■^A

i ■-■«*- -il*, # *- |v ^

^ s^v^ ' i- ■■ ■ ■• .

A A - t ' ‘’*-K

- y’^'^ % ^ ,-'

^4 - J "

Jennings, D, X and M. W. Houseweart. 1989. Sex-biased predation by web-spinning spiders (Araneae)
on spruce budworm moths. J. ArachnoL, 17:179-194.

SEX-BIASED PREDATION BY WEB-SPINNING SPIDERS

(ARANEAE) ON SPRUCE BUDWORM MOTHS

Daniel T. Jennings^

Northeastern Forest Experiment Station
USDA Building, University of Maine
Orono, Maine 04469 USA

and

Mark W. Houseweart^

Cooperative Forestry Research Unit

College of Forest Resources
University of Maine
Orono, Maine 04469 USA

ABSTRACT

Web-spinning spiders of 6 families, 12 genera, and at least 15 species preyed on spruce budworm,
Choristoneura fumiferana (Clem.), moths in Maine. Significantly more (G-tests, P ^ 0.05) male than
female moths were captured by 9 web-spinner species, and the overall capture by spiders was
significantly biased {P ^ 0.001) toward male moths. Most of the budworm-moth prey were found in
webs of Frontinella pyramiteia (Walck.) (30.6%) and Theridion picium (Walck.) (45.0%). Multiple
observations of the same web (temporal replication) indicated that T. pictum captured significantly
more (Z = —4.36, P = 0.000) budworm moths/ web than E pyramiteia. However, such differences in
prey-capture rates were not detected (Z = —1.49, P = 0.14) over several locations (spatial replication).
Web surveys during the spruce budworm’s moth-flight period indicated that percentages of webs with
budworm prey were about equal; F. pyramiteia {X = 19.5 ± 5.7), T. pictum {X= 18.9 ± 3.6), all
species {X = 16.0 ± 2.7).

Trees occupied by T. pictum were significantly taller {P ^ 0.001) and webs significantly higher {P ^
0.001) than trees and webs of F pyramiteia. For both spider species, mean relative web height was >
60% of tree height, possibly indicating nonrandom choices of foraging patch. However, tree height
and web height were not significantly {P ^ 0.05) related to prey-capture for F. pyramiteia webs; tree
height was significantly taller {P = 0.009) for T. pictum webs with budworm moths. Two species of
kleptoparasites, Argyrodes trigonum (Hentz) and A. fictilium (Hentz), were associated with host-
spider webs that captured spruce budworm moths.

Possible explanations for the observed sex-biased predation include: 1) sex-pheromone mimicry, 2)
uneven prey densities, 3) accidental capture, 4) moth behavior, and 5) moth-flight activity. Because of
the potential impacts on spruce budworm reproduction, predation by spiders gains increased
importance as a source of moth mortality.

'Present address: Northeastern Forest Experiment Station, 180 Canfield St., P. O. Box 4360,
Morgantown, West Virginia 26505 USA.

^Deceased, 8 April 1988.

180

THE JOURNAL OF ARACHNOLOGY

INTRODUCTION

The spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortrici-
dae), is the most destructive defoliator of coniferous forests in the northeastern
United States and Canada (Kucera and Orr 1981). Recent devastating outbreaks
of this forest pest have renewed interest in determining the natural enemies —
pathogens, parasites, and predators — that cause mortality to the spruce budworm.
Spiders have long been recognized as predators of the spruce budworm
(Johannsen 1913; Tothill 1923). Although all life stages of the spruce budworm —
eggs, larvae, pupae, and moths — are susceptible to predation by spiders (Jennings
and Crawford 1985), most observations concern spiders feeding on budworm
larvae (Jaynes and Speers 1949; Loughton et al. 1963; Renault and Miller 1972).
Few investigators have focused attention on spiders that prey on spruce budworm
moths; Jennings and Crawford (1985) listed 10 species of web-spinning spiders
that captured moths of C. fumiferana in their webs in Maine. Turnbull (1956)
observed predation by both web-spinning and vagrant spiders on budworm moths
in British Columbia; however, his observations pertain to the western spruce
budworm, C. occidentalis Freeman.

In this paper we describe predation by several species of web-spinning spiders
on moths of C fumiferana in Maine, compare differential mortality between
moth sexes, and discuss possible factors that contribute to sex-biased predation
on male moths.

METHODS

Study sites. — All observations were made in spruce-fir {Picea-Abies) forests
that were infested with the spruce budworm. Individual study sites were
distributed in a band from northwestern to southeastern Maine (Fig. 1). Study-
site abbreviations, locations (by organized town or township), and observation
years were:

(SOL) — Soldiertown Twp., T2 R3, Somerset County, 1979.

(WES) — West Middlesex, Canal Grant, Somerset County, 1979.

(TOM) — Tomhegan Twp., Somerset County, 1979.

(KOK)— Kokajo, T1 R13 WELS, Piscataquis County, 1977.

(GRE) — NE of Greenville, T2 R12 WELS, Piscataquis County, 1978.

(SHA) — Shawtown Twp., Piscataquis County, 1979.

(TEL) — Telos, T5 Rll WELS, Piscataquis County, 1978.

(COF) — Coffeelos, T6 Rll WELS, Piscataquis County, 1978.

(MIL) — Milo, Piscataquis County, 1979.

(MED) — Medford, Piscataquis County, 1978, 79, 80, 81.

(HOW) — Howland, Penobscot County, 1978, 79, 80.

(PAS) — Passadumkeag, Penobscot County, 1979.

(PEF) — Penobscot Expmt’l Forest, Bradley, Penobscot County, 1978, 79, 80,
81, 82.

(MAT) — Mattakeunk, Lee, Penobscot County, 1979.

(STE) — West of Steuben, T7 SD, Hancock County, 1985.

(T19)— T19 ED BPP, Washington County, 1979.

(T18)— T18 ED BPP, Washington County, 1979.

(P14) — No. 14 Plantation, Washington County, 1979, 81.

JENNINGS & HOUSEWEART— SEX-BIASED PREDATION

181

Most study sites were in cutover spruce-fir stands with abundant natural
regeneration of balsam fir, Abies balsamea (L.) Mill, and red spruce, Picea
rubens Sarg., or white spruce, P. glauca (Moench) Voss, in the understory.
Observations were confined to accessible webs in small trees (most ^ 3 m in
height) and in lower crowns of intermediate-height trees 5.5 m). All trees were
in the understory and grew in forest openings, e.g., along old logging roads and
trails. Forest-stand measurements were not taken; however, balsam fir was the
prominent understory species.

Study periods. — We observed spider predation on spruce budworm moths
during the peak years (1977-82) of a budworm epidemic in Maine, A few
observations were made during the decline phase of the same epidemic in 1985.
Each study year, webs of web-spinning spiders were examined for moth prey
during the spruce budworm’s moth-flight period. In Maine, budworm moth flight
begins in late June, peaks in early to mid-July, and ceases by late July or early
August (Houseweart et al. 1982; Kendall et al. 1982; Jennings et al. 1984). Male
moths of the spruce budworm generally emerge from pupae 1 or 2 days earlier
than female moths, and most moths live for about 10 days (Greenbank et al.
1980). Because of seasonal variation in development, moth activity at a site lasts
about 3 weeks, but can be extended longer by invasion of dispersing moths
(Greenbank et al. 1980). We made detailed observations every 2-10 days at three
sites, Greenville (GRE), Medford (MED), and Penobscot Experimental Forest
(PEF); observations at remaining sites were intermittent but within the
budworm’s moth-flight period.

Predator-prey collections. — At all study sites, we examined spider webs on
understory coniferous trees ad libitum (i.e., as encountered) and collected both

182

THE JOURNAL OF ARACHNOLOGY

spider(s) and any ensnared moth prey. Most of the spiders we studied in Maine
either left cadavers of spruce budworm moths in their webs {Frontinella) or
incorporated the cadavers into web retreats (Theridion). Occasionally, prey
cadavers were found wrapped in silk and dropped below the web {Theridion), or
hung in prey middens {Cyclosa), or hung in web supports {Neoscona). Some
unwrapped prey were found discarded immediately below the web {Prolinyphia).
To minimize disturbance, the resident spider was captured first and then prey
removed from the web or substrate with forceps. Field-collected spiders and their
prey were placed in 2-dram vials containing 70% ethanol, labeled, and
transported to the laboratory for identification.

Not all available webs were examined at each site, nor were all observed
spiders with budworm prey collected. However, when a spider-prey collection was
made, we removed all spruce budworm moths from the web or immediately
below the web. Our goal was to collect a representative sample of web-spinning
spiders that captured spruce budworm moths in the readily accessible understory.

Web surveys. — In addition to predator-prey collections, we inventoried spider
webs and their ensnared prey at various locations. Two survey methods were used
to provide temporal and spatial replication.

1) Temporal replication: At Greenville (GRE), Medford (MED), and
Penobscot Experimental Forest (PEF), webs of the two most commonly
encountered species — Frontinella pyramitela (Walck.) and Theridion pictum
(Walck.) — were identified and tagged on 20-22 small trees/ site. Numbers of webs
by species and site were: F pyramitela — MED {n = 20), PEF {n = 53); T
pictum — GRE {n = 55), MED {n - 10). Both species spin their webs in young
understory spruce or fir trees; each species builds a characteristic web. The
sheetline weaver F. pyramitela spins a “bowl and doily” web consisting of a
shallow bowl, a horizontal sheet beneath the bowl, and a meshwork of silk that
forms a barrier above the bowl (Comstock 1948). The theridiid T pictum
constructs a tangle web of many viscid threads, above which the spider ties
several spruce or fir needles together to form a dome-shaped tent in which to hide
(Emerton 1927). Cadavers of spruce budworm moths and other prey often are
incorporated into the silken walls of the tent; occasionally, prey is hung in the
tangle web or dropped to branches below the web.

Tagged trees and webs were revisited at intervals of 2-10 days {X - 5.5 days)
and data recorded on spider-web occupancy and prey captures. In most instances
we gently removed all ensnared moths from the tagged webs for prey
identification and prey-sex determination. There was minimal web disturbance.
Occasionally however, budworm moths were field-identified (but not sexed),
counted, and left intact in the webs. Tagged-web observations spanned the initial
period (late June, n = 2 visits/ site) and peak (mid-July, n = 3-4 visits/ site) in
spruce budworm moth-flight activity.

2) Spatial replication: At each of the aforementioned sites plus Howland
(HOW), Coffeelos (COF), and Telos (TEL), 30 webs (any species) were examined
ad libitum and numbers of budworm prey/ web recorded. Field identifications
were made of F. pyramitela and T pictum; specimens of unrecognized species
were collected for later identification. Cadavers of spruce budworm moths were
removed from webs for prey-sex determination. All 30-web inventories (1-3/site)
were made during the spruce budworm’s moth-flight period in late June {n = 2)
and July (n = 7).

JENNINGS & HOUSEWEART— SEX=BIASED PREDATION

183

Tree-web heights. — At some study sites, measurements were taken of tree and
web height (m) above ground for K pyramitela and T. pictum. Tree height was
measured from ground to uppermost terminal-shoot level; web height was
measured from ground to bowl level {Frontinella) and tent level {Theridion).

Spider identifications. — Collected spiders were identified by the senior author;
species determinations follow Kaston (1981) and other consulted sources
including revisions by Chamberlin and Gertsch (1958), Berman and Levi (1971),
and Levi (1957, 1974). Only sexually mature spiders were identified to species;
juveniles, including penultimate males, were identified to generic level.
Representative specimens of all identified species are deposited in the arachnid
collection, U.S. National Museum of Natural History, Washington, DC.

Prey-sex determinations. — All moth cadavers removed from spider webs were
examined microscopically and their genitalia compared with published descrip-
tions of the male (Outram 1970) and female (Outram 1971) reproductive systems
of the spruce budworm. Such examinations allowed confirmation of field-
identified prey species and determination of prey sex. A few collected moths (<
5%) were devoid of genitalia, i.e., the posterior abdominal segments were
missing — possibly lost during collection, discarded by the spider, or removed by
scavengers. Remains of these damaged moths were identified by their general
morphology (Freeman 1947) and compared with known, identified specimens;
they were sexed by tibial-spur length (long, male; short, female) and shape of the
scutellum.

Data analyses. — Because the spruce budworm has a sex ratio that does not
vary markedly from 1:1 (Miller 1963), we hypothesized an equal representation of
the sexes (male, female) among budworm moth-prey captured by web-spinning
spiders. We used the G-statistic for log-likelihood ratios (Sokal and Rohlf 1981)
to test observed vs. expected frequencies of moth sexes among prey captures for
individual and over all spider species. The null hypothesis for expected frequency
was 50% male, 50% female moths. Resultant G-values were compared with the
chi-square distribution at P = 0.05.

Means, standard errors, and coefficients of variation were calculated for tree-
and web-height measurements for F. pyramitela and T. pictum. We also calculated
relative web height (Toft 1987) for both species by the formula: rel. web ht. = web
ht./tree ht. x 100. The Wilcoxon’s two-sample test (Sokal and Rohlf 1981) was
used for comparisons of means (tree ht., web ht., rel. web ht.) between spider
species at P = 0.05. Spearman’s rank correlation coefficient (p) was used to
determine the degree of association between tree- and web-height variates for
each species.

RESULTS

Spider taxa. — Web-spinning spiders of 6 families, 12 genera, and at least 15
species were observed with spruce budworm moth prey in Maine (Table 1).
Almost all of the budworm prey captures were made by female spiders; a single
male of Agelenopsis utahana (Chamberlin & Ivie) was collected with budworm-
moth prey; a few webs of F. pyramitela (1), Prolinyphia marginata (C. L. Koch)
(2) and Theridion murarium Emerton (1) were cohabited by both spider sexes
and contained cadavers of spruce budworm moths. At Medford (MED), we

184

THE JOURNAL OF ARACHNOLOGY

Table 1. — Species of web-spinning spiders observed for spruce budworm moth prey, 18 localities,
spruce-fir forests of Maine; all study years combined. The G-statistic for log-likelihood ratios (Sokal
and Rohlf 1981) was used to compare observed vs. expected frequencies of moth sexes among prey
captures (loc = localities, obs = observed, SBW = spruce budworm, ns = not significant, P > 0.05, * =
includes multiple observations of some webs).

FAMILY n webs n webs with 2 SWB prey

Species n loc obs SBW prey Males Females G P

AGELENIDAE
Agelenopsis utahana
(Chamb. & Ivie)
Agelenopsis sp.
ARANEIDAE
Araneus nordmanni
(Thorell)

Araneus marmoreus Clerck

Araneus sp.

Araniella displicata (Hentz)
Cyclosa conica (Pallas)
Cyclosa sp.

Neoscona arabesca (Walck.)
Neoscona sp.

Nuctenea patagiata (Clerck)
Nuctenea sp.

DICTYNIDAE
Dictyna phylax
Gertsch & Ivie
Dictyna sp.

LINYPHIIDAE
Front inella pyramitela
(Walck.)

Pityohyphantes costatus

(Hentz)

Pityohyphantes sp.

Prolinyphia marginata
(C.L. Koch).

Prolinyphia sp.
TETRAGNATHIIDAE
Tetragnatha versicolor
Walck.

THERIDIIDAE

Theridion differens Emerton
Theridion frondeum Hentz
Theridion murarium
Emerton

Theridion pictum (Walck.)
Theridion spp.

All species

1

1

1

/

2

2

2

3

1

1

0

1

1

0

4

5

3

3

5

6

4

4

5

10

3

3

1

3

0

3

10

1

0

7

31

28

35

1

1

1

1

1

1

1

1

1

1

1

/

1

1

1

/

14

420*

98

129

7

11

8

9

5

7

0

7

31

18

18

1

1

0

4

6

2

2

4

26

4

5

4

5

3

3

12

17

10

8

11

564*

172

175

2

8

0

18

1170*

361

402

0 1.39 ns

0 4.16 ^0.05

0

4.16

^0.05

0

5.54

^0.05

0

4.16

^0.05

1

1.39

ns

2

35.73

^0.001

0

1.39

ns

0

1.39

ns

0

1.39

ns

0

1.39

ns

9

124.77

^0.001

0

12.48

^0.001

2

14.72

^0.00 1

0

2.77

ns

1

2.91

ns

1

1.05

ns

5

0.70

ns

28

118.53

^0.001

49

315.22

^0.001

observed a male and female of P. marginata sharing prey, i.e., both spiders were
feeding on the same male budworm moth. Juveniles of Agelenopsis, Araneus,
Dictyna, Neoscona, and Nuctenea were observed with budworm moth prey,
whereas juveniles of Cyclosa, Pityohyphantes, Prolinyphia, and Theridion were
not.

Moth-prey numbers. — Numbers of budworm moths/ web ranged from 1-5 for
E pyramitela, and from 1-14 for T. pictum', numbers for the remaining species

JENNINGS & HOUSEWEART— SEX-BIASED PREDATION

185

seldom exceeded 2 moths/ web. For multiple observations of the same web
(temporal replication), mean budworm moths/ web was significantly higher (Z =
—4.36, P - 0.000) for T pictum {X - 0.41 ± 0.07, n - 301) than for R pyramiteia
{X = 0.07 ± 0.02, n = 168). However, such differences in prey-capture rates
between spider species were not detected (Z = —1.49, P = 0.14) when several
localities (spatial replication) were considered; E pyramiteia {X = 0.52 ± 0.05, n -
252) and T pictum {X = 0.63 ± 0.06, n - 263).

Moth-prey sex. — A total of 402 male and 49 female moths of the spruce
budworm were collected from spider webs (all species) in Maine (Table 1). Most
of the budworm moths were retrieved from webs of F. pyramiteia (30.6%) and T
pictum (45.0%). Male spruce budworm moths were captured by at least 15 species
of web-spinning spiders, whereas female budworm moths were captured by only 7
species of web spinners. At least 9 (18.4%) of the ensnared female moths were
freshly emerged and gravid; most were captured by species of Theridion, i.e., T
pictum {n ~ 5), T murarium Emerton (n = 2), and T. frondeum Hentz {n ~ 1).
The T. frondeum capture was a male-female pair in copula; the gravid female
moth was dead whereas the male was still alive when observed and collected.

G-tests indicated that significantly more (P ^ 0.05) male than female moths of
the spruce budworm were captured by at least nine species of web-spinning
spiders (Table 1). The overall species total was significantly biased {P ^ 0.001)
toward capture of male spruce budworm moths. This differential mortality to
male moths of the spruce budworm previously has not been discovered and
reported.

Web surveys. — Because there were no significant differences in percentages of
webs with budworm prey by survey method, we pooled the data from web-
inventory sources. Over all spider species (?i = 1 1 observed during surveys), study-
site locations {n ~ 6), and observation dates {n ~ 15), the mean percentage of webs
with spruce budworm moths was 15.99 ± 2.69. Mean percentages of E
pyramiteia and T. pictum webs with budworm prey were 19.49 + 5.73 and 18.89
+ 3.56, respectively. The arcsine transformed means for these two species were
not significantly different (Wilcoxon 2-sample test, Z = 1.38, P ~ 0.17). Because
percentages of webs with budworm prey ranged from 0-100 for some dates, the
coefficients of variation were high over all species (166.77) and for both F.
pyramiteia (158.42) and T. pictum (92.47). Such variation was also evident when
web percentages were plotted by observation date (Fig. 2) for these two species.
Percentages of T. pictum webs with budworm moths peaked in early July and
remained at a relatively high level (^ 20%); F. pyramiteia webs declined sharply
after an initial peak in mid-July.

Tree-web heights. — Tree height for F pyramiteia ranged from 0.6-5. 5 m; web
heights for this species ranged from 0.3-2. 1 m. Correspondingly, tree heights for
T. pictum ranged from 1. 0-4.6 m; web heights ranged from 0.3-2. 3 m. Tree- and
web-height means were significantly different (Z = 7.62, tree; Z = 4.31, web)
between species (Table 2); trees occupied by T pictum were significantly taller {P
- 0.000) and webs significantly higher {P = 0.000) than those of F pyramiteia.
Mean relative web height was > 60% of tree height for both species; however,
mean percentages were not significantly different (Z = —1.05, P = 0.29) between
species (Table 2). Spearman’s rank correlation coefficients indicated a greater
degree of association between web and tree heights for E pyramiteia (p = 0.58, P
^ 0.001) than for T pictum (p = 0.24, P ^ 0.006). However, relative web height

186

THE JOURNAL OF ARACHNOLOGY

JUNE date ^ULY

Figure 2. — Percentages of Frontinella pyramitela and Theridion pictum webs with spruce budworm

moth prey by observation date; study sites and years combined.

was negatively correlated with tree height for T pictum (p = “-0.53, P ^ 0.001)
and less so for F. pyramitela (p = ~"0.13, P ^ 0.13).

Next we considered tree, web, and relative web-height means for both species
by prey-capture category (with, without budworm moth prey). For F. pyramitela,
none of the means were significantly different between prey-capture category
(Table 3); tree heights (Z - —1.82, P = 0.07), web heights (Z = —1.24, P = 0.22),
relative web height (Z = —0.74, P ~ 0.46). For T. pictum, mean tree height was
significantly taller (Z “ —2.58, P = 0.009) for webs with budworm moths;
however, web-height means did not differ significantly (Z = 0.31, P = 0.76)
between prey-capture category. Mean relative web height was significantly less (Z
= 2.46, P = 0.01) for T. pictum webs that captured spruce budworm moths (Table
3), possibly because tree height was significantly taller for these successful webs.

Kleptoparasites. — At least two species of kleptoparasites (Theridiidae) were
observed and collected; Argyrodes trigonum (Hentz) from webs of F pyramitela
{n = 4), T pictum {n ~ 2), and Neoscona sp. {n ~ 1); Argyrodes fictilium (Hentz)
and Argyrodes sp. each from webs of F. pyramitela {n ~ 2). Specimens of A.

Table 2, — Comparisons of mean (± SE) tree, web, and relative web heights of Frontinella
pyramitela and Theridion pictum, northeastern spruce-fir forests, Maine. Relative web height = web
ht./tree ht. x 100. Column means followed by different letters (a, b) are significantly different by
Wilcoxon 2-sample test, P - 0.05 (SAS Institute 1985).

Tree ht. (m)

Web ht. (m)

ReL web ht. (m)

Spider species

n

X

(+SE)

(±SE)

Z

(±SE)

Frontinella pyramitela

146

1.67a

0.05

1.06a

0.04

65.11a

1.76

Theridion pictum

134

2.14b

0.04

1.29b

0.03

62.61a

1.63

JENNINGS & HOUSEWEART-SEX-BIASED PREDATION

187

Table 3.— Comparisons of mean (+ SE) tree, web, and relative web heights of Frontinella
pyramitela and Theridion pictum by web-prey category, northeastern spruce-fir forests, Maine.
Relative web height = web ht./tree ht. x 100. Column means (within species) followed by different
letters (a, b) are significantly different by Wilcoxon 2-sample test, P = 0.05 (SAS Institute 1985).

Tree ht. (m)

Web ht. (m)

Rel. web ht. (m)

Spider species

n

(+SE)

(+SE)

X (+SE)

Frontinella pyramitela

with budworm moths

26

L58a

0.19

0.98a

0.10

63.36a 3.31

without budworm moths

120

1.69a

0.05

1.08a

0.04

65.49a 2.02

Theridion pictum

with budworm moths

83

2.22a

0.05

1.28a

0.04

59.18a 1.82

without budworm moths

51

2.02b

0.08

1.30a

0.05

68.18b 2.95

trigonum were collected at PEF,

MED, T18,

T19, and

P14;

A. fict ilium and

Argyrodes sp. were

found only

at

MED. Most (72.7%, n

= 11) of these

kleptoparasites were

associated

with

host-spider webs

that

had successfully

captured male moths of the spruce bud worm (1=3 moths/ web). We did not
observe Argyrodes spiders feeding directly on budworm prey.

DISCUSSION

Spider habitat associations. — The species of web-spinning spiders that we
collected from understory balsam fir and spruce differ markedly from the
terricolous spider fauna of Maine’s spruce-fir forests (Jennings et al. 1988;
Hilburn and Jennings 1988). However, many of the understory species also are
known to inhabit crowns of mature trees. For example, Araniella displicata
(Hentz), Pityohyphantes costatus (Hentz), Dictyna phylax Gertsch & Ivie, and
Theridion murarium Emerton have been collected from crowns of dominant/
codominant red spruce and balsam fir in Maine (Jennings and Collins 1987;
Jennings and Dimond 1988). The orb weaver Cyclosa conica (Pallas) was found
in foliage samples clipped from crowns of mature red spruce in Maine (Jennings
and Dimond 1988); webs of this species also are found suspended between small
trees in the understory. Species common to arboreal habitats of mature tree
crowns and to small, understory tree strata of Maine’s spruce-fir forests include
Araniella displicata, Dictyna phylax, Tetragnatha versicolor Walckenaer, and
Theridion differens Emerton.

Interestingly, the two species that we most commonly observed with budworm
moth prey, E pyramitela and T. pictum, apparently prefer the herb-shrub-small
tree-zone in spruce-fir forests of Maine. Neither species has been taken in pitfall
traps (Jennings et al. 1988; Hilburn and Jennings 1988), or among extensive
foliage samples clipped from crowns of dominant/ codominant red spruce
(Jennings and Collins 1987), balsam fir, red and white spruce (Jennings and
Dimond 1988), and hemlock (Jennings, unpublished data). In Maine’s spruce-fir
forests, webs of R pyramitela most frequently are found on small trees and
shrubs, and occasionally on low-growing forbs (^ 0.3 m). Webs of T pictum
most commonly are found on small trees, less frequently on shrubs, and seldom
on low-growing forbs in these forests.

Our data on web heights and relative web heights for F. pyramitela and T
pictum (Table 2) indicate a possible nonrandom selection of foraging patch by

188

THE JOURNAL OF ARACHNOLOGY

both species, i.e., mean relative web height was > 60% of tree height for both
species. And both species built their webs near branch apices where flying insects
are apt to alight; webs of neither species were seldom observed within interiors of
tree crowns. Selection of foraging patches near branch tips may favor predation
on male moths of the spruce budworm because males ‘"buzz” (i.e., slow, hovering
flight) near the periphery of branches and near tree crowns (Sanders and Lucuik
1972; Greenbank et al. 1980). Other species of web-building spiders are known to
make web-site selections based on prey abundance; see Riechert and Gillespie
(1986) for a review on nonrandom selection of foraging patch by spiders.

Sex-biased predation. — Our observations over several years indicate that male
moths of the spruce budworm are much more susceptible to predation by web-
spinning spiders than female moths of this coniferous-tree defoliator, at least in
the understory. We have no data on possible budworm-moth predation by web-
spinning spiders that inhabit mature tree crowns; however, such predation is
likely to be extensive, especially during budworm epidemics, because of the high
densities of both spiders (Jennings and Collins 1987) and budworms (Morris
1955) found in crowns of mature conifers. For small coniferous trees in the
understory, the current study confirms that significantly more (P ^ 0.05) male
than female moths of the spruce budworm are captured by at least nine species of
web-spinning spiders. The factors that cause and influence this differential
mortality to male moths of the spruce budworm are unknown. We consider and
discuss the following possibilities: 1) attraction of male moths of the spruce
budworm to prey-mimicking pheromones emitted by spiders; 2) uneven densities
and distributions of male-female moths in the understory; 3) accidental capture of
male moths responding to “calling” female moths located near spider webs; 4)
differences in male-female moth behavior; and 5) sexual differences in flight
activities of spruce budworm moths. These factors are not necessarily
independent, but may be interactive.

1) Sex-pheromone mimicry: There is mounting evidence that some spiders
produce compounds that attract male moths (Hacker 1935; Eberhard 1977;
Horton 1979; Yeargan 1988); the elicited responses are similar to the effects
caused by sex pheromones of target prey species (review in Stowe 1986). Most
observations of sex-pheromone mimicry concern species of araneid spiders that
attract and capture only male moths; attracted Lepidoptera include species of
Geometridae, Noctuidae, Pyralidae, and Olethreutidae (Stowe 1986), and
Saturniidae (Horton 1979). Apparently, the ability to mimic sex pheromones of
moths has evolved more than once (Stowe 1986), and the attractants emanate
from the spiders (Eberhard 1977; Stowe 1986; Stowe et al. 1987) or from their
webs (Horton 1979; Eberhard 1981). The preponderance of male budworm moths
captured in spider webs in Maine suggests that sex-pheromone mimicry might be
involved; however, field bioassays of male-moth attraction and olfactory-choice
studies are needed to confirm such mimicry. Conversely, the capture of both male
and female moths in some webs, and the apparent lack of prey specialization (i.e.,
captures included Homoptera, Diptera, Coleoptera, and Lepidoptera), lends
minimal support to a sex-pheromone mimicry hypothesis. Additionally, several
species of spiders of diverse families (Table 1) captured male moths of the spruce
budworm; thus far, sex-pheromone mimicry by spiders has been confirmed for
only a select few species, mostly Araneidae (Stowe 1986).

JENNINGS & HOUSEWEART— SEX-BIASED PREDATION

189

2) Uneven prey densities: Virtually nothing is known about the density and
distribution of spruce budworm moths in the understory of spruce-fir forests.
During outbreaks, small understory trees are defoliated by budworm larvae, and
female moths deposit some eggs on needles of understory trees (Morris 1955;
Jennings and Jones 1986). Most observations of spruce budworm moths concern
dispersal flights and estimates of aerial densities of moths above the forest canopy
(Greenbank et al. 1980). Measurements of moth densities within forest stands are
extremely difficult because of intervening foliage layers, moth-flight activities, and
differential responses of moth sexes to light traps (Greenbank et al. 1980).
Despite these limitations, investigators have noted a markedly consistent 50%
male: 50% female ratio for the budworm population in stands over the season
(Miller 1963; Greenbank et al. 1980). However, exodus flights (emigration) from
forest stands and mass invasions (immigration) of budworm moths into stands
contain a high proportion of egg-carrying female moths (Greenbank et al. 1980).
Because invading moths often drop to the ground, such invasions may affect
moth densities in the understory. But in Maine, we never witnessed a mass
invasion of moths at any study site. Although we cannot exclude possible unequal
distributions and densities of moth sexes in the understory, we suspect that sexual
differences in behavior and moth-flight activities were more influential in affecting
sex-biased predation by web-spinning spiders.

3) Accidental capture: Sex-pheromone communication by the spruce bud-
worm has received a great deal of attention from investigators (Sanders 1971,
1975; Sanders and Lucuik 1972, 1975), but the possible interrelationships between
“calling” female moths and predators has not been studied. We suspect that the
increased activities of male moths responding to sex pheromones greatly
magnifies their chances of capture by web-spinning spiders. Spider webs in situ
near “calling” female moths are apt to capture the most male moths; however,
this hypothesis needs to be tested under field conditions. Female moths of the
spruce budworm emerge from pupae in midafternoon (Sanders and Lucuik 1972),
and remain sedentary near their place of emergence (Sanders 1975). The female
moths start “calling” near dusk and continue calling throughout the night
(Sanders 1971, 1975). Hence, female moths that emerge and emit sex pheromones
near spider webs are apt to attract potential prey to these webs. The “buzzing”
flight activity of male moths is continuous throughout the day but increases in
intensity by mid-afternoon (Sanders and Lucuik 1972), and corresponds with the
peak calling activity of female moths (Sanders 1971). Although we did not
measure hourly capture rates for spider webs in Maine, we suspect that most
moths were caught during the time of increased male-moth activity. We often
observed spruce budworm moths flying into spider webs and being captured
successfully by resident spiders during daylight hours; no observations were made
at night.

Newly emerged, mated female moths of the spruce budworm also remain
sedentary near their place of emergence (Sanders and Lucuik 1975) and call
intermittently on successive days (Sanders and Lucuik 1972). Hence, spider webs
in situ near such resident female moths may make multiple moth captures on
successive days. This might explain why some webs captured male moths
successfully during more than one observation period in Maine.

4) Moth behavior: Recent studies have shown that some insects are attracted
to ultraviolet (UV) light reflected from spider webs (Craig and Bernard 1987);

190

THE JOURNAL OF ARACHNOLOGY

however, such attraction has not been demonstrated for spider webs and spruce
budworm moths in northeastern spruce-fir forests. The '‘buzzing” flights of male
moths, from dawn until shortly after midnight (Greenbank et ah 1980) and near
tree-crown peripheries within centimeters of branch tips (Greenbank 1973),
suggests that: (a) male moths are active during the diurnal period of greatest light
reflectance, and (b) male moths frequent zones where spider webs are common
(i.e., near branch apices). Further, flying male moths of the spruce budworm are
consistently photopositive to a discrete light source, whereas female moths
generally are photonegative (Wellington 1948). These behavioral differences,
coupled with heightened male moth-flight activity, might explain why male moths
are more susceptible to predation by web-spinning spiders. Phototatic responses
of male and female moths to UV-reflected light emanating from spider webs need
to be tested in northeastern spruce-fir forests.

5) Moth-flight activity: Finally, greater male than female moth-flight activity
probably is the factor that contributes most to spider predation on spruce
budworm moths. Increased levels of potential prey activity should greatly
influence the foraging success of “sit and wait” predators like some web-spinning
spiders. Male moths of the spruce budworm are more active, both spatially and
temporally, than female moths (Greenbank et al. 1980); such activity increases the
likelihood of male-moth encounters with spider webs, and particularly with webs
built near branch apices where male moth-flight activity is high. In laboratory
experiments with spiders preying on aphids, Provencher and Coderre (1987)
concluded that the most active prey, not necessarily the most abundant, will be
the most captured. Because female moths of the spruce budworm generally
remain sedentary, with few short, interbranch flights until 50% of their eggs are
laid (Greenbank et al. 1980), female-moth encounters with spider webs initially
are reduced. Males and spent females of the spruce budworm are more active
than gravid females (Greenbank et al. 1980).

We conclude that the minimal flight activities of female moths after emergence
reduces their susceptibility to spider predation — at least by web-spinning species.
Conversely, the intensified flight activities of male moths — whether induced by
sex-mimicking pheromones, moth sex pheromones, or UV-reflected light-
increases their susceptibility to spider predation. Quantification of potential prey
activities and their influences on spider predation need further investigation.

Web surveys. — The peaks in percentages of F. pyramitela and T. pictum webs
with budworm prey (Fig. 2) generally coincide with peaks in pheromone-trap
catches of spruce budworm moths in Maine’s spruce-fir forests (Houseweart et al.
1981; Jennings et al. 1984). These results support our hypothesis that male-moth
activity plays a significant role in predation by web-spinning spiders. Because
pheromone traps measure male moth activity (Sanders 1971), daily trap catches
may be useful for predicting captures of spruce budworm moths by web-spinning
spiders. This assumes that predation by web-spinning spiders on spruce budworm
moths is both density and activity dependent.

Future web surveys should take into account differences in web-residence times
of host spiders; unoccupied webs are less likely to capture spruce budworm moths
than occupied webs. During our sequential visits to tagged webs in Maine, we
noted that as time progressed, E pyramitela webs were more likely to be
unoccupied than T pictum webs. In New Jersey, Janetos (1984) observed a mean
residence time of only 5.4 days for F. pyramitela. Scavenging of moth prey from

JENNINGS & HOUSEWEART— SEX-^BIASED PREDATION

191

spider webs by ants and kleptoparasites also may confound survey results, i.e.,
underestimate predation by web-spinning spiders. Because some theridiid spiders
feed on ants, ants may be more successful at removing budworm moths from
sheetwebs of E pyramitela than from tangle webs of T picium. Ants were
included among the prey items removed from T pictum webs in Maine.

Kleptoparasitism. — In addition to stealing prey from host spiders. Trail (1980)
observed araneophagy by species of Argyrodes\ the kleptoparasite attacks and
kills the host spider (Wise 1982). Because of systematic destruction of several host
spiders, Shear (1986) concluded that A. fictilium may be a specialist on
Frontinella communis (Hentz) (= F. pyramitela). However, in Maine we found A.
trigonum more commonly associated with E pyramitela webs; only one specimen
of A. fictilium was collected. Rypstra (1981) indicated that Argyrodes spiders may
cause the host spider to move its web. We suspect that some of the web
“abandonments'’ in Maine may be attributed to invasion and occupancy by
Argyrodes, and possibly host-spider mortality induced by these kleptoparasites.
For example, Argyrodes spiders were observed in 3 of 4 webs “abandoned” by T.
pictum at MED. Both species of Argyrodes found during this study, A. trigonum
and A. fictilium, have been recorded from Maine (Exline and Levi 1962).

Spider-budworm impacts. — What are the overall effects of spider predation on
spruce budworm moths? Does sex-biased predation on male moths adversely
affect the reproductive potential of the spruce budworm, and, consequently,
population trend for the next generation? These and other questions need to be
answered. This study shows that web-spinning spiders cause differential mortality
to male moths of the spruce budworm; however, the consequences of such
mortality on population dynamics of the budworm are unknown.

Predation by web-spinning spiders on male moths could hamper mating success
and reproductive potential of the spruce budworm. Male moths of the spruce
budworm mate only once, and female moths rarely mate more than once per 24-h
period (Sanders 1975). Although males are capable of multiple matings (i.e., with
more than one female), an increasing proportion of second and subsequent
matings are infertile (Outram 1971). It follows then that significant mortality to
male moths increases the likelihood that virgin female moths will remain unmated
or be inseminated by “experienced” males with reduced fertility. Both unmated
females and females from infertile matings of the spruce budworm lay few eggs
(Outram 1971); hence, female oviposition success and reproductive potential are
indirectly affected by male-moth survivorship. Likewise, predation on gravid
female moths and moths in copula, as observed in this study, directly influences
reproductive potential of the spruce budworm. Because of these potential impacts
on spruce budworm reproduction, predation by spiders gains increased
importance as a source of moth mortality. Additional studies are needed to assess
age-interval survivorships of both predators and prey.

ACKNOWLEDGMENTS

We thank our former technicians for field assistance: Janice E. Littlefield,
David M. Kendall, Susan L. May, Susan M. Sheffer, Denise E. Stephens,
Michael Wissenbach, and Wesley A. Wright. Members of our personal families
also assisted with field-data collection: Nancy B., Cynthia A., and Diane L.

192

THE JOURNAL OF ARACHNOLOGY

Jennings; Tania R, Megan K., and Chad E. Houseweart. Thomas A. Skratt
provided technical assistance with data analyses; Janet J. Melvin provided word
processing service. Statistical review was provided by Donald W. Seegrist,
Northeastern Forest Experiment Station, Broomall, PA. Constructive reviews of
an early draft were given by: A. Randall Alford, Barton M. Blum, John B.
Dimond, Charles D. Dondale, Richard R. Mason, and Mark K. Stowe. This
research was supported by the USDA Forest Service, Northeastern Forest
Experiment Station, RWU-4151, Orono, ME, and by the Cooperative Forestry
Research Unit, University of Maine, Orono, ME.

This paper was in preparation when the junior co-author Mark W. Houseweart
died unexpectedly. Although inadequate, this paper is dedicated as a memorial to
his acute observational abilities, high standards of academic excellence, and
superior research performance. Any errors of data analyses and interpretation are
my own (D.T.J.).

LITERATURE CITED

Berman, J. D. and H. W. Levi. 1971. The orb weaver genus Neoscona in North America, Mus. Comp.
Zool. Bull., 141:465-500.

Chamberlin, R. V. and W. J. Gertsch. 1958. The spider family Dictynidae in America north of
Mexico. Am. Mus. Nat. Hist. Bull., 116. 152 pp.

Comstock, J. H. 1948. The Spider Book. Comstock Publishing Co., Ithaca, New York.

Craig, C. L. and D. G. Bernard. 1987. Evolution of alternative foraging modes: spiders spin silks and
weave webs that deceive prey. Pp. 13, In American Arachnology No. 37, (Abstract; Spider Silk
Symposium, American Arachnological Society Meetings, Harvard Univ., Cambridge, Massachu-
setts, June 15-20, 1987).

Eberhard, W. G. 1977. Aggressive chemical mimicry by a bolas spider. Science, 198:1173-1175.
Eberhard, W. G. 1981. The single-line web of Phoroncidia studo Levi (Araneae, Theridiidae): A prey
attractant? J. Arachnol., 9:229-232.

Emerton, J. H. 1927. A Maine spider. The Maine Naturalist, 7(2):35-37.

Exline, H. and H. W. Levi. 1962. American spiders of the genus Argyrodes (Araneae, Theridiidae).
Mus. Comp. Zool. Bull, 127:75-204.

Freeman, T. N. 1947. The external anatomy of the spruce budworm, Choristoneura fumiferana
(Clem.) (Lepidoptera, Tortricidae). Canadian Entomol., 79:21-31.

Greenbank, D. O. 1973. The dispersal process of spruce budworm moths. Canadian For. Serv., Marit.

For. Res. Cent. Inf. Rep. M-X-39. 25 pp.

Greenbank, D. O., G. W. Schaefer and R. C. Rainey. 1980. Spruce budworm (Lepidoptera:

Tortricidae) moth flight and dispersal: new understanding from canopy observations, radar, and
aircraft. Mem. Entomol. Soc. Canada 110, 49 pp.

Hacker, H. 1935. A novel moth lure. Queensland Agric. J., 43:549-550.

Hilburn, D. J. and D. T Jennings. 1988. Terricolous spiders (Araneae) of insecticide-treated spruce-fir
forests in west-central Maine. The Great Lakes Entomol., 21:105-114,

Horton, C. C. 1979. Apparent attraction of moths by the webs of araneid spiders. J. Arachnol, 7:88.
Houseweart, M. W., D. T. Jennings and C. J. Sanders. 1981. Variables associated with pheromone
traps for monitoring spruce budworm populations (Lepidoptera: Tortricidae). Canadian Entomol,
1 13:527-537.

Houseweart, M, W., S. G. Southard and D. T. Jennings. 1982. Availability and acceptability of spruce
budworm eggs to parasitism by the egg parasitoid, Trichogramma minutum (Hymenoptera:
Trichogrammatidae). Canadian Entomol, 114:657-666.

Janetos, A. C. 1984. Comparative ecology of two linyphiid spiders (Araneae, Linyphiidae). J.
Arachnol, 11:315-322.

Jaynes, H. A. and C. F. Speers. 1949. Biological and ecological studies of the spruce budworm. J.
Econ. Entomol, 42:221-225.

Jennings, D. T. and J. A. Collins. 1987. Spiders on red spruce foliage in northern Maine. J.

Arachnol, 14:303-314.

JENNINGS & HOUSEWEART— SEX=BIASED PREDATION

193

Jennings, D. T. and H. S. Crawford, Jr. 1985. Predators of the spruce budworm. U.S. Dep. Agric.,
Agric. Handb. 644, 77 pp.

Jennings, D. T. and J. B. Dimond. 1988. Arboreal spiders (Araneae) on balsam fir and spruces in
east-central Maine. J. Arachnol., 16:223-235,

Jennings, D. T,, R. M. Frank and M. W. Houseweart. 1984. Attraction of male spruce budworm
moths, Choristoneura fumiferana (Clemens), to pheromone-baited traps in small-tree thinnings. J.
Chem. EcoL, 10:125-133.

Jennings, D. T., M. W. Houseweart, C. D. Dondale and J. H. Redner. 1988. Spiders (Araneae)
associated with strip-clearcut and dense spruce-fir forests of Maine. J. Arachnol., 16:55-70.

Jennings, D. T. and R. L. Jones. 1986. Field tests of kairomones to increase parasitism of spruce
budworm (Lepidoptera: Tortricidae) eggs by Trichogramma spp. (Hymenoptera: Trichogrammati-
dae). The Great Lakes Entomol., 19:185-189.

Johannsen, O. A. 1913. Spruce bud-worm {Tortrix fumiferana, Clemens). Maine Life Sci. Agric. Exp.
Stn. Bull. 210, 31 pp.

Kaston, B. J. 1981. Spiders of Connecticut, revised edition. Connecticut Geol, Nat. Hist. Surv. Bull.
70, 1020 pp.

Kendall, D. M., D. T. Jennings and M. W. Houseweart. 1982. A large capacity pheromone trap for
spruce budworm moths (Lepidoptera: Tortricidae). Canadian Entomol., 114:461-463.

Kucera, D. R. and R W. Orr. 1981. Spruce budworm in the eastern United States. U.S. Dep. Agric.,
For. Serv., For. Insect Dis. Leafl. 160. 7 pp.

Levi, H. W. 1957. The spider genera Enoplognatha, Theridion, and Paidisca in America north of
Mexico. Am. Mus. Nat. Hist. Bull., 112:5-123.

Levi, H. W. 1974. The orbweaver genera Araniella and Nuctenea. Mus. Comp. Zool, Bull, 146:291-
316.

Loughton, B. G., C. Derry and A. S. West. 1963. Spiders and the spruce budworm. Pp. 249-268, In
The Dynamics of Epidemic Spruce Budworm Populations. (R. F. Morris, ed.). Mem. Entomol.
Soc. Canada 31. 332 pp.

Miller, C. A. 1963. The spruce budworm. Pp. 12-19, In The Dynamics of Epidemic Spruce Budworm
Populations. (R. F. Morris, ed.). Mem. Entomol. Soc. Canada 31. 332 pp.

Morris, R. F. 1955. The development of sampling techniques for forest insect defoliators, with
particular reference to the spruce budworm. Canadian J. Zool., 33:225-294.

Outram, 1. 1970. Morphology and histology of the reproductive system of the male spruce budworm,
Choristoneura fumiferana. Canadian Entomol., 102:404-414.

Outram, 1. 1971. Morphology and histology of the reproductive system of the female spruce
budworm, Choristoneura fumiferana (Lepidoptera: Tortricidae). Canadian Entomol., 103:32-43.

Provencher, L. and D. Coderre. 1987. Functional responses and switching of Tetragnatha laboriosa
Hentz (Araneae: Tetragnathidae) and Clubiona pikei Gertsch (Araneae: Clubionidae) for the
aphids Rhopaiosiphum maidis (Fitch) and Rhopalosiphum padi (L.) (Homoptera: Aphididae).
Environ. Entomol., 16:1305-1309.

Renault, T. R. and C. A. Miller. 1972. Spiders in a fir-spruce biotype: abundance, diversity, and
influence on spruce budworm densities. Canadian J. Zool., 50:1039-1046.

Riechert, S. E. and R. G. Gillespie. 1986. Habitat choice and utilization in web-building spiders. Pp.
23-48, In Spiders — Webs, Behavior, and Evolution. (W. A. Shear, ed.). Stanford Univ. Press,
Stanford, California. 492 pp.

Rypstra, A. L. 1981. The effect of kleptoparasitism on prey consumption of the spider Nephila
clavipes. Oikos, 37:179-182.

Sanders, C. J. 1971. Daily activity patterns and sex pheromone specificity as sexual isolating
mechanisms in two species of Choristoneura (Lepidoptera: Tortricidae). Canadian Entomol.,
103:498-502.

Sanders, C. J. 1975. Factors affecting adult emergence and mating behavior of the eastern spruce
budworm, Choristoneura fumiferana (Lepidoptera: Tortricidae). Canadian Entomol., 107:967-977.

Sanders, C. J. and G. S. Lucuik. 1972. Factors affecting calling by female eastern spruce budworm,
Choristoneura fumiferana (Lepidoptera: Tortricidae). Canadian Entomol., 104:1751-1762.

Sanders, C. J. and G. S. Lucuik. 1975. Effects of photoperiod and size on flight activity and
oviposition in the eastern spruce budworm (Lepidoptera: Tortricidae). Canadian Entomol.,
107:1289-1299.

SAS Institute. 1985. SAS User’s Guide: Statistics, SAS Institute, Cary, North Carolina. 1290 pp.

194

THE JOURNAL OF ARACHNOLOGY

Shear, W. A. 1986. The evolution of web-building behavior in spiders: a third generation of
hypotheses. Pp. 364-400, In Spiders — Webs, Behavior, and Evolution. Stanford Univ. Press,
Stanford, California. 492. pp.

Sokal, R. R. and F. J. Rohlf. 1981. Biometry. The Principles and Practice of Statistics in Biological
Research, second edition, W. H. Freeman and Co., New York.

Stowe, M. K. 1986. Prey specialization in the Araneidae. Pp. 101-131, In Spiders — Webs, Behavior,
and Evolution. Stanford Univ. Press, Stanford, California. 492 pp.

Stowe, M. K., J. H. Tumlinson and R. R. Heath. 1987. Chemical mimicry; bolas spiders emit
components of moth prey species sex pheromones. Science, 236:964-967.

Toft, S. 1987. Microhabitat identity of two species of sheet-web spiders: Field experimental
demonstration. Oecologia, 72:216-220.

Tothill, J. D. 1923. Notes on the outbreaks of spruce budworm, forest tent caterpillar, and larch
sawfly in New Brunswick. Proc. Acadian Entomol. Soc., 8:172-182.

Trail, D. S. 1980. Predation by Argyrodes (Theridiidae) on solitary and communal spiders. Psyche,
87(3/4):349-355.

Turnbull, A. L. 1956. Spider predators of the spruce budworm, Choristoneura fumiferana (Clem.) at
Lillooet, B. C., Canada. Proc. 8th Pac. Sci. Congr., HIA:1579-1593.

Wellington, W. G. 1948. The light reactions of the spruce budworm, Choristoneura fumiferana
Clemens. (Lepidoptera: Tortricidae). Canadian Entomol., 80:56-82A.

Wise, D. H. 1982. Predation by a commensal spider, Argyrodes trigonum, upon its host: an
experimental study. J. ArachnoL, 10:110-116.

Yeargan, K. V. 1988. Ecology of a bolas spider, Mastophora hutchinsoni: phenology, hunting tactics,
and evidence for aggressive chemical mimicry. Oecologia, 74:524-530.

Manuscript received September 1988, revised December 1988.

Dahl, R. D. and A. M= Granda. 1989. Spectral sensitivities of photoreceptors in the ocelli of the

tarantula, Aphonopeima chakodes (Araeeae, Theraphosidae). J. ArachnoL, 17:195-205.

SPECTRAL SENSITIVITIES OF PHOTORECEPTORS
IN THE OCELLI OF THE TARANTULA
APHONOPELMA CHALCODES (ARANEAE, THERAPHOSIDAE)

R. Dennis Dahl^ and A* M. Granda

Program in Neuroscience
School of Life Sciences
University of Delaware
Newark, Delaware 19716 USA

ABSTRACT

Spectral sensitivities of primary and secondary eyes in the Theraphosid spider, Aphonopeima
chakodes Chamberlin, were investigated by recording intracellular receptor potentials from single
photoreceptors. The responses of all cells were graded depolarizations, monophasic in waveform. All
ceils showed dual spectral sensitivities, with a primary peak near 500 nm and a secondary peak in the
near ultraviolet at 370 nm. The 500 nm peaks were fit well by a Dartnall nomogram.

Spectral sensitivity curves were similar under both dark and light adaptation suggesting the presence
of a single photopigment. Intensity-response functions with white light showed sensitivity differences
between primary and secondary eyes. Secondary eyes had greater sensitivity ranges and smaller slope
coefficients showing them to be more sensitive than primary eyes.

INTRODUCTION

Differing spectral sensitivities of visual cells are of considerable interest because
they are an essential condition for wavelength discriminations in visual behavior.
In several species of spider, multiple and differentiable sensitivities have been
reported to both visible and to near-ultraviolet spectral regions: in the wolf
spider, Lycosa (DeVoe, Small and Zvargulis 1969; DeVoe 1972); in the jumping
spider, Phidippus (DeVoe 1975), in Menemerus (Yamashita and Tateda 1976,
1981), in Plexippus (Blest, Hardie, McIntyre and Williams 1981); and in Argiope
(Yamashita and Tateda 1976, 1978, 1981; Tiedemann, Ventura and Ades 1986).

In this paper, we report spectral sensitivities from the visual cells of ocelli in
another spider, the New World Tarantula Aphonopeima chakodes Chamberlin.
The arrangement of ocelli consisting of primary and secondary eyes is shown in
Figure 1. Spectral sensitivities were measured from single photoreceptor cells in
each of the ocelli by means of intracellular recording. Both intensity-response
functions and action spectra were derived from the basic recordings. Comparative
evaluations were made among the several ocelli; comparisons were also drawn
between Aphonopeima and other arthropods.

'Present address: Department of Pharmacology, Uniformed Services University of the Health
Sciences, Bethesda, Maryland 20814 USA.

196

THE JOURNAL OF ARACHNOLOGY

Figure 1. — Cormorus of Aphonopelma chalcodes. The eight eyes are arranged on the cormorus of
the anterior portion of the cephalothorax as four sets of paired eyes: the primary Anterior Medial
Eyes (AME); and the secondary Anterior Lateral Eyes (ALE), Posterior Medial Eyes (PME), and
Posterior Lateral Eyes (PLE).

METHODS

Experimental animals. — Thirty-one female spiders were purchased from

commercial sources. Their weights and pubescence were suggestive of pre-adult
instars.

Preparation. — After an experimental animal was rendered tractable with
carbon dioxide, the legs and the abdomen were cut off near the cephalothorax.
The isolated cephalothorax was pinned dorsal surface down to a cork block in an
especially constructed lucite chamber, and then immersed in spider saline solution
(Rathmayer 1965). The paturons of the chelicerae and endites were removed,
revealing the ocelli attached to the hemocoelic surface of the cormorus. The optic
nerves were left intact. Access to the ocelli then required only minor dissection of
the musculature. A 1.0 - 1.5% w/w solution of Fungal Type VI protease in saline
was applied to the ocelli for one to three minutes to soften the ocellar capsule
and to allow easy penetration by the electrodes.

Optical system. — A two-channel optical system was employed. One channel
served for monochromatic stimulation, while the other provided light for
chromatic adaptation. In Channel 1, the light from a 150W xenon arc lamp was
passed through a monochromator, collimated, and focused on the tip of an
ultraviolet, light-conducting, fiber-optic guide. Light intensity was controlled by
neutral density wedges continuously variable over 6.65 optical density units. Light
duration was controlled by an electrically operated solenoid with attached flag.
An auxiliary shutter was placed in the light path of Channel 1 and used during
chromatic adaptation. In Channel 2, light at either 546 nm or 365 nm from a
mercury lamp was isolated by using appropriate cut-off filters; the beam was then
directed onto a beam splitter in the path of Channel 1, and the combined lights
were focused on the tip of the fiber optic.

The stimulus and chromatic adaptation lights were calibrated using a black-
body thermopile with attached microvoltmeter. All light fluxes were specified in
log photons . cm~^ . s"^ emergent from the light-guide tip.

DAHL & GRANDA— SPECTRAL RESPONSES OF APHONOPELMA

197

Recording system. — Intracellular potentials were recorded from single photore-
ceptor cells by means of glass micropipette electrodes filled with 3M KCl. The
electrodes had measured tip resistances in the range 60-80 megohms. All
recordings were conventional. Permanent records were captured on film from
which the measurements were made.

Experimental protocol. — Once a cell was penetrated and stable at a membrane
potential between ”30 mV and —40 mV, it was allowed to dark adapt for 30 min.
Rapid spectral sensitivity scans, consisting of 100 ms flashes of monochromatic
light over a 3 log unit range, were then made. For all spectral scans, low criterion
amplitudes of 4 mV were elicited to avoid light adaptation of the preparation.
Scans were done in balanced order using 370 nm and 520 nm flashes as control
wavelengths. The first four flashes of each scan were control flashes followed by
test flashes that in turn were interrupted by additional control flashes every fifth
flash. Spectral scans spanned the spectrum from 360 nm to 640 nm, in 20 nm
increments, for a total of 23 flashes per complete scan. Each scan was followed
by an 11-point intensity-response series at the two control wavelengths. With
interflash interval at 40 s, a complete scan took from 40 to 50 min from the time
of initial impalement. Cells were also scanned under chromatic adaptation in
order to determine visible and ultraviolet contributions. Such experiments were
begun with a dark-adapted scan that was followed by scans under chromatic
adaptation at 365 nm or 546 nm. These chromatic lights were of an intensity that
elicited dark-adapted potentials of approximately 50% saturation amplitude.

RESULTS

Receptor potential. — Intracellular receptor potentials consisted of graded
depolarizations to light; no regenerative responses were observed. Figure 2 shows
the receptor potential of an anterior lateral eye (ALE) to monochromatic light
flashes at 520 nm for three durations. These responses are typical of responses
recorded from all cells. The response waveform was characterized by an initial
transient depolarization that occurred to light onset, followed by a slow return of
the membrane potential to its resting level. The form of the response is seen most
clearly to the 100 ms flashes of Fig. 2. The transient there forms a peak which
then decays slowly to the baseline from its maximum value. To longer stimulus
flashes (500 and 900 ms), peak responses show longer time courses that merge with
overall response decays. Even to flashes longer than 900 ms, response decays were
gradual and never showed abrupt returns to baseline upon flash cessation (cf.
DeVoe 1972, 1975). No differences in waveform were observed in the responses to
different wavelengths of light. Occasional, irregular, transient variations in
potential (bumps), associated with less intense lights of long duration were seen.
They were similar to those reported in Limulus by Yeandle (1958) and probably
result from membrane responses to single photon absorptions.

Peak response amplitudes varied directly as the intensity of light, although the
period of depolarization was longer for longer flashes. The latency of response
was inversely related to light intensity.

Intensity-response functions. — Figure 3 shows the functional relationship
between response peak amplitude and light intensity. The responses were recorded
to white light which was attenuated over a seven log unit range. The curve for the

198

THE JOURNAL OF ARACHNOLOGY

Flash Duration

100 msec soo msec 900 msec

2.0x10'* /\

*o

0)

w

'E

o

3.98x10*® /V.
rL__

c

o

o

JZ

£L

1.41x10*®

3.98x10® ^

PL.

5,62x10®

rL____ r L-.

Figure 2. — Responses of ALE to monochromatic light of 520 nm. Flash durations of 100, 500, and
900 ms are listed above each column. Light intensities over a three log unit range are listed as ordinal
values. An indicator for light onset and offset is shown beneath each receptor potential; its vertical
displacement measures a calibrated 10.0 mV.

primary anterior median eye (AME) consists of 16 points, while the curve for a
secondary anterior lateral eye (ALE) consists of 20 points. These points were fit
by the Michaelis-Menten equation of format: V/ Fmax = /"/ (/" + A:), where V/ Vmax
is the normalized intracellular response potential, k is the intensity of light
required for one-half maximum response potential, and I is an increment of light
intensity (Naka and Rushton 1966; Stryer 1988:189). When comparing the two
curves, the AME curve shows a greater slope coefficient than the ALE curve: the
value of n is 0.71 for AME, and 0.55 for ALE. The slope differences between the
two curves indicate a one log unit stimulus intensity difference at the one-half
maximum response potential point in Fig. 3. This horizontal separation reflects
differences in sensitivity for the two classes of eyes (Glantz 1971). The ALE cells
possess an overall sensitivity advantage for dimmer lights, whereas this difference
is nearly gone at about —1.5 log units where the curves begin to overlap. The
linear range of sensitivity for AME is 1.5 log units, while the range for the
corresponding ALE curve is 3.0 log units.

Figure 4 displays sample intensity-response functions for several sample
wavelengths that plot peak amplitudes of response against light intensities for

DAHL & GRANDA— SPECTRAL RESPONSES OF APHONOPELMA

199

Figure 3. — Intensity-response functions to white light for AME and ALE cells. Amplitudes were
normalized and plotted against relative log light intensity. Each point is a mean for three cells with

standard deviations not larger than +/— 0.05.

AME and ALE cells. Within each kind of eye, intensity-response functions are
remarkably linear, and for each eye, alike in slope regardless of stimulus
wavelength. The two sets of curves differ however in their degree of slope (see
Discussion).

Spectral sensitivity. — Reciprocal values of photon flux necessary to elicit
criterion responses of 4.0 mV at each stimulus wavelength formed the spectral
sensitivity curves. Responses of this amplitude permitted use of dim lights that
did not result in appreciable light-adaptation of the cells. Action spectra for dark-
adapted cells of AMEs, ALEs, PMEs, and PLEs are shown in Fig. 5. The cells of
both primary and secondary eyes possess spectral sensitivity curves that are
virtually identical in shape. All curves possess major sensitivity peaks near 500
nm as well as peaks of lower sensitivity in the near ultraviolet at 370 nm. Points
for the longer wavelengths agree well with the Dartnall (1953) nomogram for a
vitamin Ai-based photopigment peaking at 500 nm.

The measured range of wavelength sensitivity for all cells lies between 350 and
640 nm. The logarithmic difference between the sensitivity peak in the 440 nm to
640 nm range, and the sensitivity peak in the near ultraviolet, 350 nm to 400 nm,
varied from 0.60 to 0.70 log unit. The overall range of sensitivity within the
experimental scan, from the most sensitive point near 500 nm and the least
sensitive point at 640 nm, measured 2.5 log units for all eyes.

Chromatic adaptation. — To isolate the visible or near-ultraviolet peaks, stable
cells in both AME and ALE were adapted with monochromatic lights of 365 nm
or 546 nm (see Methods). A representative cell is shown for each group in Figs. 6
and 7. For all light-adapted cells, overall sensitivities were reduced relative to
their dark-adaptive states, but the shapes of the curves remained virtually the
same. Accompanying the cells’ progressive reduction in sensitivity, when adapted
to 365 nm or 546 nm light, was a decrease in sensitivity difference between longer
wavelength and near ultraviolet peaks. In dark-adaptation, this sensitivity

200

THE JOURNAL OF ARACHNOLOGY

A

Figure 4. — A. Selected intensity-response functions for an AME cell. Response amplitude is plotted
against optical density attenuation indicated as unit values on the abscissa; the response curves are
separated for clarity. B. Intensity-response functions for an ALE cell is displayed as in A.

difference was 0.70 log unit, which then decreased to 0.56 log unit under
chromatic adaptation to 365 nm light and to 0.35 log unit under 546 nm light.

DISCUSSION

Although primary and secondary eyes of Aphonopelma chalcodes display
anatomical dimorphism, no functional differences were found between the two
sets of eyes for two of three parameters studied. Both intracellular responses, and
the spectral sensitivity functions derived from them, showed remarkable
similarities. Time courses and general shapes of the waveforms did not differ
between the two types of ocelli; neither did spectral sensitivity functions differ in
their regions of peak sensitivity. Both primary (AME) and secondary ocelli (ALE,
PME, and PLE) showed sensitivity peaks near 500 nm, with prominent peaks in

DAHL & GRANDA— SPECTRAL RESPONSES OF APHONOPELMA

201

Figure 5. — Spectral sensitivity curves for AME, ALE, PME, and PLE cells. The curves are
displaced vertically for clarity. Data points for each curve are mean spectral sensitivities for 12 cells
normalized at 520 nm to a log relative photon sensitivity of 0.0. Vertical lines are ranges of the data
points. The solid lines are derived from the nomogram of Dartnall (1953) for an Ai photopigment
absorbing maximally at 500 nm. Ordinal values related to each curve are sensitivity values derived
from -log relative photon fluxes that produce 4 mV responses by 100 ms light flashes at the stimulus
wavelengths in nanometers (nm) indicated on the abscissa. Mean light intensity at 500 nm was 1.2 x
10® photons • cm~^ « s”V

202

THE JOURNAL OF ARACHNOLOGY

Figure 6. — Spectral sensitivity curves for a dark- and chromatically light-adapted AME cell. Solid
line is derived from the Dartnall nomogram as in Fig. 5. The actual placement of individual curves
derives from changes in adaptation.

the near ultraviolet at 370 nm. There was, however, a difference with respect to
overall sensitivity to light. As shown in Fig. 3, the intensity-response function for
the secondary ocellus ALE, showed a shallower slope coefficient (0.55) than did
the function for the primary ocellus AME (0.71). The differing slopes with
wavelength as a parameter can be seen in Fig. 4. Secondary ocelli are apparently
able to efficiently integrate photon absorptions over a more extensive range of
light intensities than do primary ocelli.

Receptor potentials and waveforms. — Receptor potentials of tarantula photore-
ceptors to flashes of light consist of smoothly graded depolarizations. The graded
changes in membrane potential are consistent with intracellular receptor
potentials recorded from other arthropod eyes and are of the same polarity
(DeVoe 1975; Bruno, Mote and Goldsmith 1973). There were transient decreases
in membrane potentials to light onset followed by slow restorations to initial
resting levels, even with flashes longer than 100 ms. These waveforms are in
contrast to those reported for wolf spider by DeVoe (1972), where prolonged
flashes of light produced marked OFF responses.

Spectral sensitivity. — All photoreceptor potentials recorded from cells in AME,
ALE, PLE, and PME showed a common sensitivity peak in the visible
wavelengths at 500 nm, together with a lesser peak at 370 nm in the near
ultraviolet, the two peaks differing in sensitivity by about 0.7 log unit. These
results are in close company with the dual sensitivities that are maximum at 360-

DAHL & GRANDA— SPECTRAL RESPONSES OF APHONOPELMA

203

Figure 7. — Spectral sensitivity curves for a dark- and chromatically light-adapted ALE cell. See
legend to Figure 6.

370 nm and at 510 nm in the principal eyes and in the anterior lateral eyes in
wolf spider (DeVoe 1972). Principal eyes of the orb-weavers, Agiope bruennichii
and Agiope amoena, on the other hand, have three types of receptor cells with
maximum sensitivities at 360 nm, 480-500 nm, and 540 nm. Posterior lateral eyes
for these same spiders show similar peak sensitivities (Yamashita 1985). In
general, principal and secondary eyes of Salticidae and Lycosidae, when assessed
intracellularly, have cells that correspond to Aphonopelma (cf. Yamashita 1985,
tables 1 and 2). The excellent fit of the data points in Fig. 5 by the Dartnall
nomogram supports the idea that measured spectral sensitivities most likely derive
from a single vitamin Ai-based photopigment (Dartnall 1953). In further support
of this idea is the fact that chromatic adaptations with 365 nm and 546 nm lights
produced no differential effects. There is also evidence that parametric changes in
stimulus intensity can result in matched waveforms regardless of wavelength, an
argument for simple intensity effects but not one that reflects spectral changes.
From all these lines of evidence, it is unlikely that Aphonopelma possess any
ability to discriminate wavelengths.

Information over optic nerves in most animals is coded as regenerative action
potentials. That is normally not the case in wolf spiders (DeVoe 1972) nor is it
for Aphonopelma (personal observations). Photoreceptor excitation is conducted
to the supraesophageal ganglion in the form of decrementing graded potentials.
Although rare, nonregenerative neural activity is known to occur in other
arthropod visual systems (lonnides and Walcott 1971; Shaw 1972). The quality of

204

THE JOURNAL OF ARACHNOLOGY

information transmitted to optic centers in this fashion must of necessity be
primitive. Aphonopelma may therefore respond simply to the fundamental

dimension of ambient light intensity. Complexities of wavelength discrimination
and contour perception are apparently not involved, for the visual system here is
functionally homogeneous and shows none of the response complexities seen in
color discriminating eyes.

Ocellar structure and function. — Similar to many spiders, secondary ocelli in
Aphonopelma possess tapeta. Tapeta provide a mechanism for lightpath
doubling; they lie very close to the rhabdomeres so that reflected light

immediately retraverses these photopigment-bearing cells (Land 1972). There are
no tapeta in primary ocelli, and light traversing these rhabdomeres is absorbed by
the heavily pigmented capsule of the ocellus after a single passage. When
intensity-response functions for primary and secondary ocelli are superimposed,
these curves reveal that at equivalent lights, and also for less intense lights,
responses of secondary ocelli are larger than those of primary ocelli. This finding
is consistent with the idea that secondary ocelli and associated structures have
evolved as detectors of dim lights. The functions for secondary ocelli are less

steep than those for primary ocelli, and this property suggests a more efficient

response range for a given intensity range of light input. There appears to be in
these structures a simple duplex arrangement that accounts for a greater range of
light sensitivity than would otherwise be possible, an arrangement that performs
for Aphonopelma what rods and cones do in the intensity domain for the
vertebrate retina.

ACKNOWLEDGMENTS

We are grateful to Margie L. Barrett, A. A., for figure illustration and
photography, and to Mary A. Boyer, A.S., for skillful manuscript editing.

LITERATURE CITED

Blest, A. D., R. C. Hardie, P. McIntyre and D. S. Williams. 1981. The spectral sensitivities of
identified receptors and the function of retinal tiering in the principal eyes of a jumping spider. J.
Comp. Physiol., 145:227-239.

Bruno, M., M. Mote and T. H. Goldsmith. 1973. Spectral absorption and sensitivity measurements in
single ommatidia of the green crab. J. Comp. Physiol., 82:151-163.

Dartnall, H. J. A. 1953. The interpretation of spectral sensitivity curves. British Med. Bull., 9:24-30.
DeVoe, R. D., R. J. W. Small and J. Zvargulis. 1969. Spectral sensitivities of wolf spider eyes. J. Gen.
Physiol., 54:1-32.

DeVoe, R. D. 1972. Dual sensitivities of cells in wolf spider eyes at ultraviolet and visible wavelengths
of light. J. Gen. Physiol., 59:247-269.

DeVoe, R. D. 1975. Ultraviolet and green receptors in principal eyes of jumping spiders. J. Gen.
Physiol., 66:193-207.

Glantz, R. M. 1971. Toward a general theory of visual adaptation. Doc. Ophthal., 30:245-252.
lonnides, A. C. and B. Walcott. 1971. Graded illumination potentials from retinula axons in the bug
Lethocerus. Z. vergl. Physiol., 71:315-327.

Land, M. F. 1972. The physics and biology of animal reflectors. Pp. 75-106, In Progress in Biophysics
and Molecular Biology. (J. A. V. Butler and D. Noble, eds.). Pergamon, New York, vol. 24.

Naka, K. I. and W. A. H. Rushton. 1966. S-potentials from luminosity units in the retina of fish
(Cyprinidae). J. Physiol., 85:587-902.

Rathmayer, W. 1965. Neuromuscular transmission in a spider and the effect of calcium. Comp.
Biochem. Physiol., 14:673-687.

DAHL & GRANDA— SPECTRAL RESPONSES OF APHONOPELMA

205

Shaw, S. R. 1972. Decremental conduction of the visual signal in barnacle lateral eye. J. Physiol.,
220:145-160.

Stryer, L. 1988. Biochemistry. W. H. Freeman & Co., New York.

Tiedemann, K. B., D. F. Ventura and C. Ades. 1986. Spectral sensitivities of the eyes of the orb web
spider Argiope argentaia (Fabricus). J. Arachnol., 14:71-78.

Yamashita, S. 1985. Photoreceptor cells in the spider eye: spectral sensitivity and efferent control. Pp.

103-117, In Neurobiology of Arachnids. (F. G. Barth, ed.). Springer- Verlag, New York.

Yamashita, S. and H. Tateda. 1976. Spectral sensitivities of jumping spider eyes. J. Comp. Physiol.,
105:29-41.

Yamashita, S. and H. Tateda. 1978. Spectral sensitivities of the anterior median eyes of the orb web
spiders, Argiope bruennichii and A. amoena. J. Exp. Biol., lAAl-Sl .

Yamashita, S. and H. Tateda. 1981. Efferent neural control in the eyes of orb weaving spiders. J.
Comp. Physiol, 143:477-483.

Yeandle, S. 1958. Electrophysiology of the visual system. Amer. J. Ophthalmol, 46:82-87.

Manuscript received June 1988, revised January 1989.

>'■ it- '''

? -

'jiT^ji^'l^ S’iint li^; ; i ,.¥^^|%«t-^.'^'>-A«x'^

iSiK^k' - • - ' <> ■'^v ; feVi' a « - n ; J*

■? <5-¥'^t^;m
CF. . ^ ■

1-? ’ .sT ■''*^' '!*''■*

^ ^ ■*'*^ , V %’/, v-f*^

,f-M.^: 7.^ - :. '• ,. .1

•’#>'.n.v '- :. ,* ♦'i^fi’v- , : 5i;’:: ’ i

„- . -^-V. -■ ^ •' :jf ,g • ■ '.'... ■'^P?yii?lfii %

»j4i4i^“'- ' *'i- ' \ ‘ ■■ '^'- . .^‘-.vyrriv

', ,. sifl. ‘ '■■ V, >^'*'^1

«, ^ .el , ^ ’ ,>?:?.

i •■ > V i^jCl

■ -A''"' tVM 'lii ^ ? 4 , It '.

■• t*. .,t;^^l* ivf

' - * ^V-’ .■7*'

.1

-.'1.

-

Briggs, T. S. and D. Ubick. 1989. The harvestman family Phalangodidae. 2, The new genus,
Microcina (Opiliones, Laniatores). J, ArachnoL, 17:207=220.

THE HARVESTMAN FAMILY PHALANGODIDAE. 2.
THE NEW GENUS, MICROCINA
(OPILIONES, LANIATORES)

Thomas S. Briggs and Darrell Ubick

Department of Entomology
California Academy of Sciences
Golden Gate Park

San Francisco, California 94118-4599 USA

ABSTRACT

A new phalangodid genus, Microcina, is described to accommodate Sitalcina tiburona Briggs and
Horn and five new species {edgwoodensis, homi, jungi, ieei, and lumi). The species are diagnosed and

illustrated and their relationships hypothesized.

The six species of Microcina, known only from the San Francisco Bay region, are all paedomorphs
restricted to xeric habitats. Two species groups are present: the first {tiburona) is characterized by a
unique form of sexual dimorphism (males have enlarged eye tubercles); the second {homi) by unique
male genitalia, structurally intermediate between Calicina and the remaining Nearctic phalangodid
genera.

INTRODUCTION

The phalangodid genus Sitalcina, established by Banks (1911) for Shakes
calif ornicus Banks, was revised and greatly enlarged by Briggs (1968). In our
recent study (Ubick and Briggs 1989), Sitalcina was shown to be polyphyletic and
most of the species were transferred to a newly established genus, Calicina. At
that time three species remained unplaced. One of these species, Sitalcina
tiburona Briggs and Horn, represents yet another new genus, Microcina, the focus
of this paper.

This study is based on the examination of 140 specimens, almost all from the
collection of TSB and now deposited in the California Academy of Sciences;
additional material was borrowed from and is deposited at the American
Museum of Natural History (New York). The specimens were collected by the
authors and the following persons: Lee Cheng, C. Fox, A. Gray, Kevin Horn,
Albert Jung, Jason Lee, Vincent F. Lee, Bill Lum, Toshiro Ohsumi, and Ken
Wang. The specimens were prepared and examined as in Ubick and Briggs (1989).
The species descriptions are brief, partly because character states shared at higher
levels (genus and species group) are not repeated. More important is the apparent
absence of significant intraspecific variation in Microcina for the characters
examined. Some variation in size has been recorded and is given in Table 1.
There is also little interspecific variation between the members of the tiburona
group and no characters have been found to reliably distinguish the females. All
specimen measurements are in mm.

208

THE JOURNAL OF ARACHNOLOGY

Table 1. — Measurements (in mm) of Microcina species giving ranges, means, and standard
deviations. Abbreviations are TBL = total body length, SL = scute length, SW = scute width, LIIL =
leg II length, ETL = eye tubercle length, ETW = eye tubercle width.

Species

Sex

N

TBL

SL

sw

LIIL

ETL

ETW

homi

m

13

0.82-1.00

0.55-0.72

0.58-0.69

1.40-1.90

0.13-0.18

0.12-0.18

0.888+0.053

0.625+0.044

0.626+0.035

1.594+0.132

0.145+0.013

0.139+0.016

f

19

0.79-1.05

0.55-0.74

0.56-0.78

1.51-1.92

0.12-0.18

0.13-0.18

0.904+0.075

0.640+0.049

0.661+0.067

1.651+0.133

0.145+0.015

0.150+0.015

tiburona

m

17

0.97-1.21

0.76-0.87

0.69-0.85

1.90-2.44

0.22-0.28

0.24-0.33

1.095+0.060

0.810+0.035

0.759+0.041

2.078+0.144

0.251+0.017

0.274+0.021

f

13

0.97-1.22

0.67-0.80

0.71-0.85

1.82-2.18

0.18-0.21

0.18-0.23

1.074+0.062

0.765+0.035

0.779+0.041

1.984+0.133

0.187+0.012

0.211+0.013

leei

m

3

0.87-1.10

0.64-0.70

0.59-0.67

1.74-1.85

0.18-0.19

0.21-0.23

0.973+0.117

0.677+0.032

0.637+0.042

1.797+0.055

0.183+0.006

0.217+0.012

f

3

0.80-0.97

0.63-0.69

0.57-0.69

1.60-1.70

0.12-0.15

0.15-0.18

0.863+0.093

0.650+0.035

0.620+0.062

1.663+0.055

0.140+0.017

0.167+0.015

lumi

m

5

0.97-1.10

0.72-0.79

0.68-0.72

1.69-1.87

0.18-0.26

0.23-0.28

1.040+0.054

0.760+0.029

0.712+0.018

1.804+0.074

0.224+0.029

0.250+0.021

f

4

0.95-1.18

0.64-0.72

0.68-0.77

1.77-1.87

0.13-0.15

0.15-0.18

1.070+0.112

0.675+0.037

0.720+0.042

1.815+0.042

0.145+0.012

0.170+0.014

jungi

m

4

1.05-1.20

0.77-0.88

0.73-0.83

1.97-2.13

0.21-0.26

0.26-0.31

1.107+0.065

0.847+0.052

0.785+0.042

2.082+0.076

0.232+0.021

0.280+0.022

f

2

0.97-1.10

0.75-0.77

0.78-0.80

1.93-2.00

0.17-0.18

0.18-0.19

1.035+0.092

0.760+0.014

0.790+0.014

1.965+0.050

0.175+0.007

0.185+0.007

edge-

woodensis

m

3

0.90-0.97

0.67-0.70

0.62-0.70

1.50-1.87

0.18-0.21

0.22-0.26

0.940+0.036

0.683+0.015

0.650+0.043

1.733+0.203

0.193+0.015

0.247+0.023

Microcina, new genus

Sitalcina: Briggs and Horn 1966. Briggs 1968 (in part).

Diagnosis. — Species of Microcina appear to be unique among the Nearctic
Phalangodidae in having an areolate body cuticle (Fig. 1). They are further
distinguished from other phalangodids with reduced tarsal counts (3-4-4-4) in
having a penis with a folding glans and an ovipositor cuticle completely covered
with microspines.

Type species. — Sitalcina tiburona Briggs and Horn, 1966.

Etymology. — The generic name is a contraction of micro and Sitalcina,
referring to the small size of the species, and is feminine in gender.

Description. — Color of body pale orange; appendages yellowish white.
Abdominal integument (of preserved specimens) somewhat transparent; irregular
white masses visible beneath cuticle. Body length 0.8 to 1.2 mm. Carapace cuticle
with honeycomb network of ridges (areolate); flattened tubercles present in
cephalic region. Tergites with posterior margins tuberculate {homi group) or
smooth {tiburona group). Eye tubercle low and rounded (enlarged in males of the
tiburona group); cornea and retina absent. Anterior margin of carapace with 1
pair of anterior tubercles; ozopores lateral. Venter similarly sculptured as dorsum.
Palp with typical number and arrangement of megaspines: tarsus and tibia with
two pairs each, patella with one (mesal), and femur with one mesoapical and
three ectobasal. Tarsal count 3-4-4"4. Ovipositor surface with prominant ridges;
completely covered with microspines; with 7 pairs of apical setae; setae apically
hooked. Penis with folding glans, which also telescopes in the homi group. Stylus
curved, spinelike; basally surrounded by a pair of lobes. Ventral plate rounded

BRIGGS & UBICK— NEW PHALANGODID GENUS MICROCINA

209

apically, with 1 pair of stout, lateral setae and 3-4 pairs of smaller, ventral setae;
homi group with additional 2 pairs of lateral setae.

Natural history. — All species are winter active paedomorphs known only from
xeric habitats.

Distribution. — The San Francisco Bay region of California.

Species. — The six species, representing two species groups, can be separated by
the following key. No characters were found to distinguish females of the
tiburona group.

KEY TO SPECIES OF MICROCINA

1. Male eye tubercle subequal to that of female. Penis stylus telescopes during
expansion (Figs. 16-18); oviposter setae with simple tips (Fig. 10)

{homi group) . homi

Male eye tubercle distinctly larger than that of female (Figs. 2-5). Stylus does
not telescope during expansion (Figs. 21-33); ovipositor setae with trifurcate
tips (Figs. 12, 13) {tiburona group). .2

2. Penis apical lobes pointed, stylus slightly sinuous (Figs. 24, 32). .......... 3

Apical lobes rounded, stylus strongly sinuous (Figs. 22, 26, 30). 4

3. Apical lobes narrow, surface lightly fringed (Fig. 32) edgewoodensis

Apical lobes wide, surface strongly fringed (Fig. 24). ................. leei

4. Penis substylar knob rounded (Fig. 26) lumi

Substylar knob angular (Figs. 22, 30), 5

5. Apical lobes with serrate ventral margin (Fig. 22)
Apical lobes with smooth ventral margin (Fig. 30),

tiburona
. . ,jungi

210

THE JOURNAL OF ARACHNOLOGY

The homi species group

Diagnosis. — The single species representing this group, unlike other Microcina,
has tubercles on the posterior tergite margins (Fig. 1). Males are unique among
the Nearctic Phalangodidae in having a glans which both unfolds and telescopes
during expansion (Figs. 14, 15, 18). Males are also distinguished from other
Microcina in lacking enlarged eye tubercles (Fig. 1), in having the ventral plate
with both ventral and lateral setal series (Fig. 17), in having a glans with apical
lobes bearing stout fringe flaps (Fig. 7), and in lacking a substylar knob and
dorsal flap on the glans (Figs. 6). Females have ovipositor setae with simple tips
(Fig. 10).

Distribution. — Known only from Santa Clara County (Fig. 34).

Microcina homi, new species
Figs. 1, 6, 7, 10, 11, 14-20, 34.

Sitalcina minor Briggs and Horn, 1966:263 (in part, all individuals from Santa Clara County).
Diagnosis. — Same as for species group.

Etymology. — Named after Kevin Horn, collector of these and numerous other
rare and unusual phalangodids.

Description. — Male (Holotype): Total body length, 0.85. Scute length, 0.55;
width, 0.61. Eye tubercle length, 0.14; width, 0.12. Leg II length, 1.40. Penis as
illustrated (Figs. 6, 7, 14-20).

Female: Total body length, 0.88. Scute length, 0.60; width, 0.62. Eye tubercle
length, 0.12; width, 0.13. Leg II length, 1.60. Ovipositor as illustrated (Figs. 10,

11).

Natural history. — Known only from grassland habitats. Specimens from Santa
Teresa Park were collected beneath Franciscan sandstone; all others are from
serpentine. This species is fully sympatric with Calicina serpentinea (Briggs and
Horn) and at one locality (0.9 mi. S Junction of Silver Creek and San Felipe
Roads) with Microcina jungi, new species.

Material examined. — Holotype: U.S.A.: CALIFORNIA: Santa Clara Co., 1.8 miles N of Highway
101 on Metcalf Road, 2 January 1983 (T. S. Briggs, V. F. Lee, and D. Ubick), male (CAS).

Paratypes: U.S.A.: CALIFORNIA; Santa Clara Co., same data as holotype, 25 males, 20 females
(CAS); 1 mile NW of Morgan Hill, 26 February 1966 (T. S. Briggs and K. Horn), male (CAS); 0.5
miles NW of Santa Teresa County Park, 27 February 1966 (T. S. Briggs), 2 females (CAS); 0.9 miles
S of Junction of Silver Creek and San Felipe Roads, 27 November 1966 (T. S. Briggs and A. Jung),
male, 3 females (CAS); San Jose, Silver Creek Road, 5 miles SE of Tully Road, 27 February 1966 (T.
S. Briggs and K. Horn), 2 males, 12 females (CAS); San Jose, W side Silver Creek Road, 5 mi SW
Tully Road, 27 November 1966 (T. S. Briggs and C. Fox), 7 males, 3 females (AMNH).

The tiburona species group

Diagnosis. — Members of this group have the posterior tergite margins smooth,
lacking tubercles. Males differ from homi in having enlarged eye tubercles (Figs.
2-5) and non-telescoping styli (Figs. 8, 9, 2L33). Males may be further
differentiated in having the ventral plate with only one pair of lateral setae (Figs.
21, 23, 25, 29, 31), in having a glans with apical lobes bearing a fine fringe (Figs.

BRIGGS & UBICK— NEW PHALANGODID GENUS MICROCINA

211

Figures 2-5. — Microcina tiburona, topotypes, cephalic region: 2, male, lateral view; 3, female, lateral
view; 4, male, dorsal view; 5, female, lateral view. Scale bar = 0.30 mm.

22, 24, 26, 30, 32), and in having a substylar knob and a dorsal flap on the glans
(Fig. 8, 9). Females have ovipositor setae with trifurcate tips (Figs. 12, 13).
Distribution. — Known only from the San Francisco Bay region (Fig. 34).
Species. — The five species representing this group can be distinguished by the
male genital characters. We have not been able to discover characters for
differentiating females.

212

THE JOURNAL OF ARACHNOLOGY

Figures 6-9. — Male genitalia of Microcina species, topotypes: 6, 7, M. homi; 6, penis, lateral view;
7, glans, ventral view; 8, 9, M. tiburona\ 8, penis, lateral view; 9, glans, lateral view.

Microcina tiburona (Briggs and Horn), new combination
Figs. 2-5, 8, 9, 12, 13, 21, 22, 34

Sitalcina tiburona Briggs and Horn, 1966:265, PI 1 (figs. 2, 6). Briggs, 1968:27, (figs. 28, 58, 89).

Diagnosis.^ — This species is distinguished from others in the group by the
following combination of male genitalic characters: stylus strongly sinuous; apical

BRIGGS & UBICK— NEW PHALANGODID GENUS MICROCINA

213

080122 15KV 60um I 080124 15KV 4.3um

Figures 10-13. — Female genitalia of Microcina species, topotypes: 10, 11, M. homi; 10, ovipositor,
lateral view; 11, ovipositor surface; 12, 13, M. tiburona; 12, ovipositor, lateral view; 13, ovipositor,
apical seta.

lobes rounded with serrate ventral margin; and substylar knob angular (Figs. 8, 9,

21, 22).

Description. — Male (Allotype): Total body length, 1.10. Scute length, 0.80;
width, 0.75. Eye tubercle length, 0.24; width, 0.28. Leg II length, 1.90. Penis as

illustrated (Figs. 8, 9, 21, 22).

214

THE JOURNAL OF ARACHNOLOGY

Figures 14-20. — Microcina homi, male holotype (Figs. 16-18, 20), male paratopotype (Figs. 15, 19),
male paratype, Silver Creek Road (Fig. 14): 14, unexpanded penis, lateral view; 15, partially expanded
penis, lateral view; 16, fully expanded penis, dorsal view; 17, fully expanded penis, ventral view; 18,
fully expanded penis, lateral view; 19, partially expanded glans, lateral view; 20, fully expanded stylus,
lateral view. Scale = 0.25 mm (Figs. 14-18), 0.10 mm (Figs. 19, 20).

Female (Holotype): Total body length, 1.10. Scute length, 0.80; width, 0.75.
Eye tubercle length, 0.20; width, 0.22. Leg II length, 2.00 Ovipositor as illustrated
(Figs. 12, 13).

Natural history. — All collections are from serpentine grassland. Sitalcina''
cockerelli Goodnight and Goodnight occurs in adjacent chaparral-grassland
ecotone.

Material examined. — Holotype: U.S.A.: CALIFORNIA; Marin Co., Tiburon, Ring Mountain,
spring about 0.5 miles NE of Bel Aire School, 15 January 1966 (T. S. Briggs and K. Horn), female
(CAS).

BRIGGS & UBICK— NEW PHALANGODID GENUS MICROCINA

215

Paratypes: U.S.A.: CALIFORNIA; Marin Co., same data as holotype, male (allotype), 4 males
(CAS); 7 males, 3 females (AMNH), 22 January 1966 (T. S. Briggs and K. Horn), 5 males, 4 females
(CAS); Tiburon, 0.5 miles S of El Campo, 22 January 1966 (K. Horn), 3 females (CAS).

Other material: U.S.A.i CALIFORNIA; Marin Co., Tiburon, Ring Mountain, 19 December 1968
(T. S. Briggs), 2 males, 4 females (CAS); Tiburon, Ring Mountin, near stream between Reed Ranch
Road and Ring Mountain Reserve boundary, 27 January 1985 (T. S. Briggs and K. Wang), male
(CAS); 14 November 1987 (T S. Briggs and J. Lee), 4 males, 2 females (CAS); Tiburon, end of
Miraflores Lane off Avenida Mirafiores, 21 December 1988 (T. S. Briggs and L. Cheng), 2 males, 2
females (CAS).

Microcina ieei, new species
Figs. 23, 24, 34

Diagnosis. — This species is distinguished from others in the group by the
following combination of male genital characters: stylus slightly sinuous, apical
lobes pointed, and substylar knob indistinct.

Etymology. — Named after Vincent F. Lee, collector of this and many other
phalangodids.

Description. — Male (Holotype): Total body length, 1.10. Scute length, 0.70;

width, 0.65. Eye tubercle length, 0.19; width, 0.21. Leg II length, 1.80. Penis as
illustrated (Figs. 23, 24).

Female: Total body length, 0.82. Scute length, 0.63; width, 0.60. Eye tubercle
length, 0.12; width, 0.17. Leg II length, 1.60. Ovipositor as in M. tiburona.

Natural history. — Found beneath sandstone rocks in open oak grassland where
it is sympatric with Calicina polina (Briggs). Another phalangodid, Sitaldna
californica (Banks), has been collected in adjacent, thickly forested areas.

Material examined. — Holotype: U.S.A.: CALIFORNIA: Alameda Co., Berkeley, Woolsey Canyon
(E of LeCoete Street), N side of canyon adjacent to Lawrence Berkeley Laboratory Parking Lot, 21
December 1983 (T. S. Briggs, V. F. Lee, and D. Ubick), male (CAS).

Paratypes: U.S.A.: CALIFORNIA; Alameda Co., Berkeley, N side of Woolsey Canyon, 17
February 1960 (A. Gray), male, female (AMNH); Oakland, 0.7 miles NE of Ashby on Claremont
Avenue, 29 January 1983 (T. S. Briggs), male, female (CAS).

Microcina lumi, new species
Figs. 25-28, 34

Diagnosis.— This species is distinguished from others in the group by the
following combination of male genital characters: stylus strongly sinuous, apical
lobes rounded, and substylar knob rounded.

Etymology. — Named after Bill Lum, one of the collectors of this species.

Description. — Male (Holotype): Total body length, 1.10. Scute length, 0.78;
width, 0.68. Eye tubercle length, 0.22; width, 0.25. Leg II length, 1.80. Penis as
illustrated (Figs. 25-28).

Female: Total body length, 1.00. Scute length, 0.65; width, 0.68. Eye tubercle
length, 0.13; width, 0.15. Leg II length, 1.80. Ovipositor as in M. tiburona.

Natural history. — Found beneath serpentine rocks in grassland biomes;
sympatric with Calicina polina (Briggs).

Material examined. — Holotype: U.S.A.: CALIFORNIA; Alameda Co., 500 feet S of intersection of
Lake Chabot Road and Fairmont Drive on W facing slope (400 feet eiev.), 6 April 1982 (T. S. Briggs
and D. Ubick), male (CAS).

216 THE JOURNAL OF ARACHNOLOGY

Figures 21-28. — Male genitalia of Mkrodna species: 21, 22, M. tiburona, topotype; 21, expanded
penis, lateral view; 22, glans, lateral view; 23, 24, M. led, holotype; 23, expanded penis, lateral view;
24, glans, lateral view; 25-28, M. lumi, holotype; 25, expanded penis, lateral view; 26, glans, lateral
view; 27, unexpanded penis, lateral view; 28, expanded penis, ventral view. Scale bar = 0.25 mm (Figs.
21, 23, 25, 27, 28), 0.10 mm (Figs. 22, 24, 26).

Paratypes: U.S.A.: CALIFORNIA; Alameda Co., same data as holotype, 3 females (CAS); NE San
Leandro, near abandoned military road N of Fairmont Drive, 26 January 1969 (T. S. Briggs and B.
Lum), 3 males, female (CAS).

Microcma jumgi, new species
Figs. 29, 30, 34.

Diagnosis. — This species is distinguished from others in the group by the
following combination of male genital characters: stylus strongly sinuous, apical
lobes rounded with smooth ventral margin, and substylar knob angular.

BRIGGS & UBICK— NEW PHALANGODID GENUS MICROCINA

111

Figures 29-33. — Male genitalia of Microcina species, holotypes: 29, 30, M. jungi\ 29, expanded
penis, lateral view; 30, glans, lateral view; 31-33, M. edgewoodensis; 31, expanded penis, lateral view;
32, glans, lateral view; 33, expanded penis, ventral view. Scale bar = 0.25 mm (Figs. 29, 31, 33), 0.10
mm (Figs. 30, 32).

Etymology. — Named after Albert K. S. Jung, collector of this and many other
phalangodids.

Description. — Male (Holotype): Total body length, 1.20. Scute length, 0.88;
width, 0.78. Eye tubercle length, 0.26; width, 0.31. Leg II length, 2.10. Penis as
illustrated (Figs. 29, 30).

Female: Total body length, 1.10. Scute length, 0.75; width, 0.78. Eye tubercle
length, 0.18; width, 0.19. Leg II length, 2.00. Ovipositor as in M. tiburona.

Natural history. — Found beneath serpentine rocks in grassland biomes;
sympatric with M. homi, new species, and probably Calicina serpentinea (Briggs
and Horn).

Material examined. — Holotype: U.S.A.: CALIFORNIA; Santa Clara Co., 0.9 miles S of junction of
Silver Creek and San Felipe Roads, 27 November 1966 (T. S. Briggs and A. K. S. Jung), male (CAS).

Paratypes: U.S.A.: CALIFORNIA; Santa Clara Co., same locality as holotype, male, 4 females
(CAS).

218

THE JOURNAL OF ARACHNOLOGY

Figure 34. — Map of the San Francisco Bay Region, California, showing the distribution of
Microcina species. M. homi (circles), M. tiburona (hexagon), M. lumi (open triangle), M. jungi (filled

triangle), M. edgewoodensis (open squares), M. leei (filled squares).

Microcina edgewoodensis, new species

Figs. 31-34

Diagnosis. — This species is distinguished from others in the group by the
following combination of male genital characters: stylus slightly sinuous; apical
lobes strongly pointed, apically serrate; and substylar knob rounded.

Etymology. — Named after the type locality, Edgewood Park.

Description. — Male (Holotype): Total body length, 0.95. Scute length, 0.68;
width, 0.62. Eye tubercle length, 0.19; width, 0.22. Leg II length, 1.50. Penis as
illustrated (Figs. 31-33).

Female: Unknown.

BRIGGS & UBICK— NEW PHALANGODID GENUS MICROCINA

219

Natural history. — Found beneath serpentine rocks in grassland adjacent to
scrub oaks; sympatric with Calicina minor (Briggs and Horn).

Material examined. — Holotype: U.S.A.: CALIFORNIA; San Mateo Co., Edgewood Park, Canada
and Edgewood Roads, 13 April 1985 (T. S. Briggs and T. Ohsumi), male (CAS).

Paratypes: U.S.A.: CALIFORNIA; San Mateo Co., Edgewood Park, canyon between Sylvan and
Serpentine Trails, 2 January 1987 (T. S. Briggs, V. F. Lee, and D. Ubick), 2 males (CAS).

ECOLOGY

All species of Microcina live in open grassland biomes and are conspicuously
modified for life in xeric environments. The species are small and lightly
pigmented, lack eyes, and have (relative to other phalangodids) a reduced tarsal
count (3“4“4"4) and a reduced number of anterior tubercles (1 pair). Such
modifications, found in only one other Nearctic phalangodid genus, Calicina,
appear to result from two phenomena: paedomorphosis (probably progenesis) and
troglobism (Ubick and Briggs 1989).

With the exception of M. tiburona, all species of Microcina have been collected
with (and apparently share the habitats of) the sympatric species of Calicina: M.
leei and M. lumi with C. polina (Briggs); M. jungi and M. homi with C.
serpentinea (Briggs and Horn); and M. edgewoodensis with C minor (Briggs and

Horn).

PHYTOGENY

Monophyly. — Microcina appears to be monophyletic on the basis of three
possible syeapomorphies. (1) Body cuticle sculpturing. The body cuticle of
Microcina is areolate (Figs. 1-5), composed of an intricate honeycomb of ridges
(similar to that found on the appendages of all phalangodids examined), whereas
that of Calicina and Sitalcina is tuberculate. Superficial examination of the
remaining phalangodids suggests that the areolate cuticle is unique to Microcina.
Both on the basis of uniqueness and an apparently more orderly nature, the
areolate cuticle appears to be derived. (2) Paedomorphic-troglobitic adaptations.
The species of Microcina are all small and show reduction in structures
(pigmentation, eyes, anterior tubercles, and tarsal count). These adaptations are
absent in most phalangodids and may be synapomorphic for Microcina.
However, the strength of this character is weakened since parallel paedomorphic-
troglobitic modifications are found in species of Calicina. (3) Apical lobe
ornamentation. The glans of Microcina has the apical lobes ornamented with a
scale-like fringe; large in M. homi (Figs. 6, 7), small in the tiburona group (Figs.
8, 9). These structures have not been observed in species of Calicina or Sitalcina;
it remains to be seen whether they are unique to Microcina.

Species groups. — Microcina contains two well defined species groups. The
tiburona group, with five species, is characterized by a unique (at least among the
North American and European phalangodids) sexual dimorphism (males have
enlarged eye tubercles) and several male genitalic characters (see Diagnosis). The
second group contains a single species, M. homi, which has an unusual glans,
possibly unique, which both unfolds and telescopes during expansion.

Sister group. — The species of Microcina share with all other Nearctic
Phalangodidae, except Calicina, a folding glans. Although the widespread

220

THE JOURNAL OF ARACHNOLOGY

distribution of this state suggests plesimorphy, we have argued previously (Ubick
and Briggs 1989) that the folding glans is probably derived for two reasons: (1)
the folding glans is structurally more complex than the telescoping glans and (2)
the folding glans is usually associated with additional and clearly derived genitalic
characters, such as the bifurcate ventral plate found in Banksula, Texelia, and all
Appalachian genera. The folding glans is thus presumed to be a synapomorphy.

Two possible symplesiomorphies join Microdna and Calidna: (1) The
telescoping glans, shared by all species of Calidna, also occurs in M. homi. In
this species the glans unfolds, as in all other Microdna, but subsequently
telescopes for complete expansion (Figs. 16-18, 20). This type of glans appears to
be unique; it is not known to occur in any other species of Nearctic
Phalangodidae nor in the European species examined {Ptychosoma vitelUnum
Soerensen and Scotolemon lespesi Lucas). (2) Both Microdna and Calidna have
ovipositors bearing microspines (Figs. 10-12), which appear to be absent in
Sitaldna, The significance of this character will be elucidated as other genera are
examined. However, the character state found in Microdna (microspines
completely cover ovipositor) appears to be plesiomorphic in Calidna (because of
correlation with presumably plesiomorphic male genitalic characters) and may
likewise be so in Microdna.

BIOGEOGRAPHY

Microdna is known only from the San Franciso Bay region (Fig. 34). The six
species are ailopatric; some are restricted to unusually small areas of suitable
habitat. The single known instance of sympatry is between M. homi and M.
jungi, which belong to different species groups. This suggests that the initial
barrier which isolated the groups occured in the southern part of the Microdna
range. If dispersal did not play a significant role in the formation of the present
distribution, then this barrier would have been located in what is now Santa
Clara County.

ACKNOWLEDGMENTS

The authors wish to thank the various collectors for their diligent field work.
Thanks go to Mary Ann Teoorio for producing the SEMs. We thank James C.
Cokendolpher and William A. Shear for kindly reading, and commenting upon
this paper. This study was partially funded by a grant from the In-House
Research Fund of the California Academy of Sciences.

LITERATURE CITED

Banks, N. 1911. The Phalangida of California. Pomona College J. EntomoL, 3:412-421.

Briggs, T S. 1968. Phalangids of the laniatorid genus Sitakina (Phalangodidae: Opiliones). Proc.
California Acad. Sci., 36:1-32.

Briggs, T. S. and K. Horn. 1966, Five new species of Phalangodidae from California (Opiliones). Pan-
Pacific Entomol, 42:262-269.

Ubick, D. and T. S. Briggs. 1989. The harvestmen family Phalangodidae, 1. The new genus Calidna,
with notes on Sitakina (Opiliones: Laniatores). Proc. California Acad. Sci., 46:95-136.

Carico, J. E. 1989. Descriptions of two new species of the genus Architis (Araneae, Pisauridae) and
the female of A. vilhena. J. ArachnoL, 17:221-224.

DESCRIPTIONS OF TWO NEW SPECIES OF THE GENUS
ARCHITIS (ARANEAE, PISAURIDAE) AND THE
FEMALE OF A, VILHENA

James E. Carico

Biology Department
Lynchburg College
Lynchburg, Virginia 24501 USA

ABSTRACT

Two new species of the genus Architis are described: A. sinops from a male from Brazil and A.
suarez from a male from Colombia. The female of A. vilhena Carico is described for the first time.

Since the revision of Architis Simon (Carico 1981), a collection of specimens
containing previously undescribed material has been received from the American
Museum of Natural History. Herein are descriptions of two new species and of
the previously unknown 'female of A. vilhena Carico extracted from this

collection.

Architis sinops, new species
Figs. 1-3

Type . — Male holotype from Sinop, Mata Grosso, Brazil (October, 1976; M.
Alvarenga), deposited in the American Museum of Natural History, New York.

Etymology. — The specific name is a noun in apposition derived from the name
of the type locality.

Diagnosis. — Architis sinops is distinguished from all known species, except A.
capricorna Carico, by sharing the following combination of characters: anterior
eye row slightly procurved, the ALE distinctly smaller than the AME, and the
clypeus height about twice the diameter of an ALE. This species is the smallest
Architis yet discovered (total body length = 3.58 mm) which is only slightly more
than half the size of its closest relative, A. capricorna (total body length = 6.2).

Description. — Male holotype: Carapace high, unmarked except for indistinct
median band, each eye surrounded by distinct dark ring (Fig. 1); length 1.7, width
1.45. Sternum dusky, unmarked. Eye measurements in Table 1, anterior row
slightly procurved. Leg measurements in Table 2, dusky with faint annulations.
Abdomen dorsally (Fig. 1) dusky, mottled with lighter areas, cardiac area light,
lighter ventrally, length 1.88. Palpus (Figs. 2, 3) with complex tibial apophysis
with elements located distally on ventral, retrolateral and dorsal sides. Female:
Unknown.

222

THE JOURNAL OF ARACHNOLOGY

Figures 1-7. — Morphology of species of Architis: 1-3, A. sinops, male holotype; 1, dorsal pattern;
2,3, right palpus; 2, ventral view; 3, retrolateral view; 4, 5, A. suarez, right palpus of male holotype; 4,
ventral view; 5, retrolateral view; 6, 7, A. viihena, female, epigynum; 6, ventral view; 7, dorsal view.
Scales: Fig. 1 = 0.5 mm; Figs. 2-7 = 0.1 mm.

CARICO— NEW ARCHITIS SPECIES

223

Table 1. — Eye measurements of species of Architis. a = holotype male, b = holotype male, c =
female.

Diameters Row length Ocular quadrangle

Species

ALE

AME

PLE

PME

Ant.

Post.

Ant.

Post.

Height

Clypeus

A. sinops^

0.08

0.11

0.15

0.13

0.42

0.73

0.27

0.40

0.33

0.17

A. suarez"

0J3

0.13

0.15

0.15

0.54

0.72

0.27

0.42

0.35

0.05

A. vilhena'"

0.10

1.11

0.22

0.22

0.59

0.85

0.30

0.49

0.41

0.18

Natural history. — Unknown.

Distribution and material examined. — Known only from seven adult males

taken in a single collection at the type locality.

Architis suarez, new species
Figs. 4, 5

Type. — Male holotype from Rio Suarez, Colombia, 1000 m, (October 1946),
deposited in the American Museum of Natural History, New York.

Etymology. — The name is a noun in apposition taken from the name of the
type locality.

Diagnosis. — The eyes of the procurved anterior eye row are equal in size and
are only slightly smaller than eyes of the posterior row. The male of this species is
further distinguished from A. cymatilis Carico, its sister species, by details of the
palpal bulb and tibial apophysis. An additional distinction is that A. suarez has
relatively longer legs: carapace length/ leg I 0.12 compared with 0.14 in A.
cymatilis.

Description. — Male (Holotype): Specimen with pattern not well preserved as a
result of apparent drying out sometime after it was preserved. Carapace
moderately high, each eye surrounded by a dark ring; length 1.69, width 1.50. Eye
measurements in Table 1, anterior row procurved, ALE on prominent tubercles.
Leg measurements in Table 2, very slender, indistinct annulations ventrally,
several short stout macrosetae on ventral side of coxa and trochanter 1. Abdomen
somewhat shriveled, length 1.83. Palp (Figs. 4, 5) with prominent tibial apophysis
projecting distally, with two distal subdivisions, ventral one acute, thickly
sclerotized, dorsal one rounded, not so heavily sclerotized. Female: Unknown.

Natural history. — Unknown.

Distribution and material examined. — Known only from the type specimen.

Table 2. — Leg measurements in species of Architis. a = holotype male, b = holotype male, c =
female.

Species

I

11

III

IV

A. sinops^

5.63

5.78

5.25

6.14

A. suarez^

14.56

15.12

9.66

12.44

A. vilhend"

14.0

13.9

10.6

12.2

224

THE JOURNAL OF ARACHNOLOGY

Architis vilhena Carico

Figs. 6, 7

A. vilhena Carico, 1981:150, figs. 2, 7, 20, 21; map 2.

Female, — Color light with indistinct pattern as in male. Shape and proportions

of body agrees closely with those of the male (Carico 1981). Carapace length 2.2,
width 1.8. Eye measurements in Table 1. Leg measurements in Table 2.
Epigynum, see Figs. 6, 7.

Note. — This is the first description of the female of this species. The collection
contained 2 females and a single male from Caninde, Rio Gurupi, Para, Brazil, 7-
15 April 1963, collected by B. Malkin (AMNH).

Additional new localities. — BRAZIL: MATO GROSSO; Barra do Taparape, 1-
12 January 1963, (B. Malkin), 1 male (AMNH), Sinop, Oct. 1975, (M.
Alvarenga), 5 males, (AMNH).

ACKNOWLEDGMENTS

Thanks are extended to N. Platnick and L. Sorkin of the American Museum of
Natural History for the loan of the collection, and to M. Muma and J. Beatty
who reviewed the text.

REFERENCE CITED

Carico, J. E. 1981. The Neotropical spider genera Architis and Staberius (Pisauridae). Bull. American
Mus. Natur. Hist., 170:140-153.

Manuscript received February 1989, revised March 1989.

Bennett, R. G. 1989. “‘Emerit’s glands” in Cybaeota (Araneae, Agelenidae). J. ArachnoL, 17:225-235.

‘DEMERIT’S GLANDS” IN CYBAEOTA
(ARANEAE, AGELENIDAE)

Robert G. Bennett

Department of Environmental Biology
University of Guelph
Guelph, Ontario NIG 2W1 Canada

ABSTRACT

A new type of spider integumentary gland with distinctive cuticular morphology was recently
described in the Telemidae. Strikingly similar glands have subsequently been found in the
Leptonetidae, some Salticidae, and now in the genus Cybaeota (Agelenidae). The spatial distribution
of the glands in Cybaeota is described. Their placement on the bodies of these spiders supports the
repugnatorial secretion hypothesis proposed for them in the Telemidae. The phylogenetic implications
of the scattered distribution of this character in spiders are discussed.

INTRODUCTION

In 1981 Emerit described a new type of integumentary structure from the legs
of the cave-dwelling telemid spider Telema tenella Simon. In this spider he found
up to twenty tiny “cupules gaufrees”, or cuticular plates, randomly distributed
along the middorsal length of each tibia (except on the pedipalps). Noting the
presence of a minute pore in each, he suggested that the plates are either
chemosensors or glands. Later Emerit and Juberthie (1983) demonstrated that the
plates are glandular and produce a non-proteinaceous secretion. In two further
papers (1984, 1985) Emerit described similar glands from other telemids of the
genera Apneumonella Page, Usofila Keyserling, Seychellia Saaristo, Cangoderces
Harington, and Jocquella Baert. He concluded that the glands are most likely
repugnatory in nature in spite of their lack of accumulation reservoirs normally
associated with chemical defense secretions (Emerit 1984). In summary Emerit
(1985) suggested that these glands are apomorphic for the family Telemidae and
divide the family into two groups: Usofila plus Telema with oval plates (Figs.
2, 3), and the four others with transverse furrows.

Thinking that perhaps the glands are an adaptation to troglobiosis, Emerit
(1984) studied one cave-dwelling leptonetid but found no tibial glands. Platnick,
however, observed glands similar to those on Telema and Usofila on the tibiae of,
and Forster found them on the patellae of, some other leptonetid species
(Platnick 1986). Platnick (1986) subsequently reviewed the distribution of tibial
and patellar glands in the Leptonetidae and concluded that (1) a few leptonetids
have oval-type tibial glands, (2) patellar glands, although lacking in the
Telemidae, are found in most Leptonetidae, (3) the glands are a synapomorphy of
the two families (a previously accepted but cladistically untested sister grouping),
and (4) the oval morphology is probably plesiomorphic for the two families.

226

THE JOURNAL OF ARACHNOLOGY

Figures 1-4. — Emerit’s gland distribution in Cybaeota spp. and Telema sp.: 1, C. nana, on tibia,
patella, and femur I (arrows indicate two glands on femur); 2, 3, Telema sp. (Victoria, British
Columbia), dorsally on tibiae; 2, tibia III; 3, tibia I; 4, C. shastae, distodorsally on male palpal
patella.

Oval glands strikingly similar to those in the telemids and leptonetids are now
known to occur in at least two other families. Wanless (1984, 1987) and Wanless
and Lubin (1986) have found them in the Salticidae (in a dorsal abdominal
cluster in Portia Karsch, Cyrba Simon, Cocalus C. L. Koch, Mintonia Wanless,
and Gelotia syringopalpis Wanless; dorsally on the tibiae in Spartaeus Thorell;
and on all legs, especially femur I and patella I of males in Diolenius minotaurus
(Wanless and Lubin). This paper describes them from the cybaeine agelenid genus
Cybaeota Chamberlin and Ivie.

At least until their true function is determined, it is proposed here that this
type of gland be termed “Emerit’s glands.”

BENNETT— “EMERIT’S” GLANDS IN CYBAEOTA

111

Figures 5-8. — Emerit’s gland distribution in Cybaeota spp.: 5, 6, C. calcarata, proximodorsally on
tibia I; 5, at least 20 glands of variable morphology; 6, two glands indicated by arrows in Fig. 5; 7, 8,

C. shastae, mid-dorsally on tibia I; 7, five glands; 8, gland indicated by arrow in Fig. 7.

EMERIT’S GLANDS IN CYBAEOTA
PLACEMENT, FUNCTION, AND PHYLOGENETIC IMPLICATIONS

During the preparation of a revision of Cybaeota (Bennett 1988) oval tibial
glands were found on C shastae Chamberlin and Ivie (Figs. 7, 8) which strongly
resemble, in morphology and distribution, those of Usofila, Telema, and the
Leptonetidae. Subsequent work with Cybaeota has shown that Emerit’s glands
occur on many parts of the body surface of C shastae and at least two of the
three other known species: C. calcarata (Emerton) and C. nana Chamberlin and
Ivie.

In Cybaeota the morphology of Emerit’s glands is quite variable. There are
three basic variations: distinctively keeled medially (Fig. 1), flattened with little

228

THE JOURNAL OF ARACHNOLOGY

Figures 9-12. — Emerit’s gland distribution in Cybaeota spp.: C. shastae, on anterior prolateral
margin of chelicera; 10-12, distodorsally on patellae; 10, C. nana, patella II, female; 11, same, male;

12, C. calcarata, patella I.

ornamentation (Fig. 6), and laterally elongated (Fig. 25). However, intermediates
and other morphotypes (e.g., Fig. 28) exist. A wide range of variation may be
observed on a single specimen. This seems especially true for C. calcarata (Fig, 5).

All glands have a single pore which is directed distally on leg segments,
anteriorly on the carapace, and posteriorly on the abdomen. The glands are most
numerous and heavily concentrated on the legs - dorsally on the tibiae (Figs. 5-8)
and patellae (Figs. 1, 10-12) and distolaterally on the femora (Figs. 1, 13) - and
adjacent to the eye group (Figs. 17-20, 22). Glands also are encountered
sporadically on the dorsal surfaces of the patellae and tibiae of palps (Figs. 4, 14-
16), anteriorly on the chelicerae (Figs. 9, 21), ventrally on the coxae of legs (Figs.
25, 26), and dorsally on the abdomen (Figs. 27-29, 31, 32) or rarely ventrally (Fig.

BENNETT— “EMERIT’S” GLANDS IN CYBAEOTA

229

Figures 13-16. — Emerit’s gland distribution in Cybaeota spp.: 13, C. nana, prolaterally on femur,
patella, and tibia I (arrows indicate some of at least 20 glands present); 14-16, on female palps; 14, C.
nana, patella (two glands indicated by arrows); 15, C. shastae, proximodorsal margin of tibia; 16,
same, distodorsal margin.

30). Glands are also scattered about dorsally on the metatarsi of the legs (Figs.
33-36) and on the carapace (Figs. 23, 24).

In Cybaeota, Emerit’s glands are unknown on the sternum, trochanters, tarsi,
ventrally on leg segments (except the coxae), on the palpi (except as noted
above), or dorsally on the femora. As in most other taxa where they are known
to occur, the glands are distributed equally among males and females (see
Wanless and Lubin 1986).

No evidence of Emerit’s glands has been found in the following agelenids which
exploit habitats similar to those occupied by Cybaeota: Dirksia cinctipes (Banks),
Ethobuella tuonops Chamberlin and Ivie, Cryphoeca montana Emerton, Cicurina
brevis (Emerton), various species of the genera Cybaeus L. Koch and Cybaeina

230

THE JOURNAL OF ARACHNOLOGY

Figures 17-20. — Cybaeota spp., Emerit’s glands around eyes: 17, C. calcarata, posterior to PME; 18,
gland indicated by arrow in 17; 19, C shastae, anterior and retrolateral to left eyes; 20, same species,
different specimen, posterior to PME. Arrows indicate some of glands present. P=PME.

Chamberlin and Ivie, and an unidentified species of each of Blabomma
Chamberlin and Ivie, Calymmaria Chamberlin and Ivie, and one unidentified
genus. Similarly, the clubionid Phrurotimpus borealis (Emerton), which shares
conspicuous paired ventral tibial macrosetae and habitat type with Cybaeota,
lacks this type of gland.

The placement of these glands on Cybaeota gives strong circumstantial support
to Emerit’s (1984) repugnatory secretion hypothesis. Like many other spiders,
Cybaeota will often “feign death” when disturbed, folding its legs up above the
carapace and bending them at the femur/ patella joints. In this posture the most
exposed regions of an individual are the dorsal surfaces of the tibiae and patellae
and to a lesser extent the cephalic region and the ventral lateral parts of the
femora. These are the same areas in which the heaviest concentrations of Emerit’s

BENNETT— “EMERIT’S” GLANDS IN CYBAEOTA

231

Figures 21-24. — Emerit’s gland distribution in Cybaeota spp.: 21, 22, C. nana; 21, two on proximal
retrolateral margin of chelicera; 22, two (indicated by arrows) below left AME; 23, 24, C. shastae\
mid-dorsally on carapace between eye group and dorsal groove; 23, eight glands (indicated by arrows);
24, gland in Fig. 23 indicated by horizontal arrow. A=AME.

glands occur. Relatively unexposed areas (sternum, dorsal surfaces of coxae and
femora, and ventral surfaces of the tibiae) or less “important” segments (tarsi)
have no glands.

From the above overview it can be concluded that Emerit’s glands appear in
several distantly related groups of spiders either as a retained plesiomorphy or a
frequent convergence. Because of their recently demonstrated wide distribution, it
is important to reexamine the status of Emerit’s glands as the sole synapomorphy
of Leptonetidae plus Telemidae. Marshall (1987) has argued that in the absence
of a known outgroup phylogeny the frequent occurrence of a character in the
outgroup makes an equivocal polarity decision more probable and increases the
likelihood that what is being interpreted as homology is actually homoplasy.

232

THE JOURNAL OF ARACHNOLOGY

Figures 25-28. — Emerit’s gland distribution in Cybaeota spp.: 25, 26, C. nana, ventrally on coxae;

25, coxa IV; 26, coxa III; 27, 28, C. caicarata, dorsally on abdomen; 27, at least 16 glands present; 28,
gland indicated by arrow in Fig. 27.

Outgroup relationships at the family level in spiders are uncertain. It is likely that
the distribution of Emerit’s glands is wider than currently documented (Platnick
pers. comm., Wanless pers. comm.). The probabilities of polarity error and
misinterpreted homoplasy with respect to Emerit’s glands are increased because of
this and render this character unreliable as an indicator of monophyly. It is quite
possible this character is a true synapomorphy of Leptonetidae plus Telemidae.
No synapomorphies are known which contradict this grouping (Platnick 1986).
For these reasons Platnick’s hypothesis (1986) stands.

The discovery of Emerit’s glands in Cybaeota does nothing to resolve its
relationship with other spiders. Cybaeota remains in the Agelenidae, Cybaeinae
but incertae sedis.

BENNETT— “EMERIT’S” GLANDS IN CYBAEOTA

233

Figures 29-32. — Cybaeota nana, Emerit’s glands on abdomen; 29, dorsal, five glands (two indicated

by arrows); 30, ventral, one gland, indicated by arrow; 31, dorsal, one gland; 32, same.

ACKNOWLEDGMENTS

The following people are warmly thanked for their interest in my findings: Drs.
M. Emerit, S. A. Marshall, F. R. Wanless, and especially N. I. Platnick whose
paper provided the stimulus for this one and who provided initial copies of the
pertinent references, valuable information, and valued discussion. Sandy Smith
provided technical advice in the use of a Hitachi S-570 SEM. Assistance in the
preparation of plates was given by D. J. Hamilton. John Heraty, F. A. Coyle,
and especially S. A. Marshall and N. I. Platnick reviewed and criticized drafts of
the manuscript.

234

THE JOURNAL OF ARACHNOLOGY

Figures 33-36. — Cybaeota spp., Emerit's glands proximodorsally on metatarsi: 33, C. shastae, two

glands (indicated by arrows), leg IV; 34, C nana, one gland, leg IV; 35, C. cakarata, one gland; 36, C.

nana, one gland proximal to gland in Fig. 34.

LITERATURE CITED

Bennett, R. G. 1988. The spider genus Cybaeota (Araneae, Agelenidae). J. Arachnol. 16:103-119.

Emerit, M. 1981. Sur quelques formations tegumentaires de la patte de Telema tenella (Araignee,
Telemidae), observees au microscope electronique a'balayage. Atti Soc. Toscana Sci. Nat., Mem.,
ser. B, 88 (suppl.):45-52.

Emerit, M. 1984. Les glandes tibiales des Telemidae: uee structure inedite localisee sur I’appendice
locomoteur des araignees. Compt. Rendu Acad. Sci. Paris, ser. 3, 299:1-4.

Emerit, M. 1985. L’appareil glandulaire tegumentaire de la patte des Telemidae (Araneae). Un critere
phylogenique? Mem. Biospeol., 12:91-96.

Emerit, M. and C. Juberthie. 1983. Mise en evidence d’un equipement glandulaire tegumentaire sur la
patte d’une araignee troglobie: Telema tenella (Telemidae). Mem. Biospeol, 10:407-411.

Marshall, S. A. 1987. Systematics of Bitheca, a new genus of New World Sphaeroceridae (Diptera).
Syst. Ent., 12:355-380.

BENNETT— “EMERIT’S’^ GLANDS IN CYBAEOTA

235

Platnick, N. L 1986. On the tibial and patellar glands, relationships, and American genera of the
spider family Leptonetidae (Arachnida, Araneae). American Mus. Novitates, 2855, pp. 1-16.

Wanless, F. R, 1984. A revision of the spider genus Cyrba (Araneae; Salticidae) with the description
of a new presumptive pheromone dispersing organ. Bull. British Mus. nat. Hist. (ZooL), 47:445-
481.

Wanless, F. R. 1987. Notes on spiders of the family Salticidae. 1. The genera Spartaeus, Mintonia and
Taraxelia. Bull. British Mus. nat. Hist. (ZooL), 52:107-137.

Wanless, F. R. and Y. D. Lubin. 1986. Diolenius minotaurus sp. nov., a remarkable horned jumping
spider from Papua New Guinea (Araneae: Salticidae). J. Nat. Hist., 20:1211-1220.

Manuscript received January 1989, revised March 1989.

1989. The Journal of Arachnology 17:237

RESEARCH NOTES

SPIDER VS. SPIDER:

FRONTINELLA PYRAMITELA DETECTS ARGYRODES
TRIGONUM VIA CUTICULAR CHEMICALS

Interspecific chemical communication is known among spiders in the contexts
of their use of chemotactile information in identifying prey species (e.g.,
Robinson and Olazarri 1971) and their attraction of prey species with chemical
lures (Eberhard 1977, 1981; Horton 1979). We know of no reports, however, of
spiders detecting and discriminating among heterospecifics that might, themselves,
prey on the spiders doing the chemical sensing.

Several researchers have reported that members of the genus Argyrodes
(Theridiidae) sometimes prey upon spiders with which they cohabit in an
otherwise kleptoparasitic or commensal manner (e.g., Exline and Levi 1962;
Lamore 1958). Wise (1982), in an experimental field study, showed that the
predatory (i.e., host-killing) behavior of Argyrodes trigonum (Hentz) is more
common than was previously realized, at least when the host is relatively small.
Predation on larger spiders such as the agelenid Agelena Umbata Thorell is also
known (Tanaka 1984), and Larcher and Wise (1985) have shown that some of the
relationships between A. trigonum and its host species can be quite complex.

One of the common hosts of A. trigonum in the northeastern United States is
the bowl and doily spider, Frontinella pyramitela (Walckenaer) (Linyphiidae).
The relationship between these two species is frequently kleptoparasitic but is also
predatory. Archer (1946) reported A. trigonum preying on F. pyramitela, and
more recently Suter (1985) reported the same phenomenon. Whatever the
character of the interaction between the two species, it is certainly intense: during
part of the summer, the kleptoparasite/ predator inhabits about 20% of all bowl
and doily webs and causes the departure or death of many of the hosts (Suter
1985). Moreover, indirect evidence suggests that the two species have interacted
frequently and over a long period of time: female bowl and doily spiders, though
larger than their mates, permit males to capture prey on the females’ webs, and
the function of that permissiveness may be to deflect (onto the males) the risk of
being captured by Argyrodes and other prey-mimicking spiders (Suter 1985).

In the work reported here, we sought to determine whether physical contact
between a bowl and doily spider and its kleptoparasite/ predator was chemically
informative from the host’s perspective, and whether the information gained was
predator-specific.

Female bowl and doily spiders and the kleptoparasitic A. trigonum were
captured on F pyramitela webs in Poughkeepsie, NY, during June, 1985. Juvenile
instars of a crab spider, Misumenoides sp. (Thomisidae) that were approximately
the same size as adult A. trigonum were also captured in June. These eventually
served as a check on whether the responses of F. pyramitela to contact with the

1989. The Journal of Arachnology 17:238

kleptoparasites’ carcasses were species specific. The bowl and doily spiders were
maintained in the laboratory using methods described elsewhere (Suter 1985).
Soon after their capture, both A. trigonum and Misumenoides sp., as well as
several E pyramiteia, were killed by freezing and stored frozen until they were
prepared for testing.

A. trigonum carcasses were either left intact (not chemically modifed) or
washed in hexane for 30 minutes. Because of our experience with the cuticular
pheromones of F. pyramiteia (Suter et al. 1987), we expected that treatment with
hexane would remove behaviorally active chemicals found on the surfaces of the
carcasses.

The assay arena consisted of the web of an adult female bowl and doily spider
with the carcass of another spider cemented to it. The carcass was cemented with
droplets of Tester's paint (behaviorally neutral after drying) near the center of the
underside of the bowl in the location normally occupied by the owner of the web.
Several hours later, the completed arena was used in behavioral tests. At the start
of a test, we released a female bowl and doily spider onto the bowl of a web that
contained a heterospecific or a conspecific female carcass. Oriented search
behavior (Suter 1984) usually brought the assay female into contact with the
carcass after less than 30 seconds. The first contact with the carcass began the
behavioral assay, which ended 5 minutes later. The initial contact and subsequent
behaviors were videotaped from above while a voice record of the behaviors
visible from the side of the web was made on one of the tape’s audio channels.
We analysed both immediate post-contact behavior and later behaviors, though
we did not consider any behaviors that took place >5 minutes after initial
contact.

Apart from normal locomotion, three distinct behaviors followed the assay
spider’s contact with a carcass: flinch — the rapid withdrawal of the first two pairs
of legs from the immediate vicinity of the carcass without displacement of the
assay spider’s body; touch-retreat — a rapid leap away from the carcass resulting
in displacement of the assay spider’s body and a change in the orientation of the
body relative to the carcass; and feeding — the insertion of the assay spider’s fangs
into the body of the carcass and the beginning of ingestion.

Contact with the four classes of test carcasses (intact E pyramiteia, intact and
washed A. trigonum, and intact Misumenoides sp.) resulted in different arrays of
behaviors from the assay spiders (Table 1). Contact with an intact carcass of the
predatory kleptoparasite, A. trigonum, nearly always elicited touch-retreats from
the assay spider whereas contact with F. pyramiteia carcasses seldom did and
contact with hexane-washed A, trigonum and with Misumenoides sp. never did.
The assay spiders rarely fed on the carcass of an intact A. trigonum but often fed
upon the carcasses of the other three classes of spiders.

During the five minutes of any particular test, the number of touch-retreats
reflected the identity of the carcass: the host species elicited significantly fewer
touch-retreats {N = 35, median = 0) than did the kleptoparasitic predator (A = 71,
median = 3; 1 -tailed Mann- Whitney test, Z = -6.09, P < 0.001). The orientation of
the assay spider immediately after a touch-retreat also reflected, though less
strongly, the identity of the carcass: following a touch-retreat, the assay spider
was significantly more likely to be facing away from the carcass if the carcass was
A. trigonum {N = 70, median = 120°) than if it was E pyramiteia {N = 16, median
= 97°; 1 -tailed Mane- Whitney test, Z = -1.96, P < 0.025). (Note that in the

1989. The Journal of Arachnology 17:239

Table 1. — Incidences of behaviors performed by female F. pyramitela during the 5 minutes
immediately following contact with the carcass of a conspecific or heterospecific spider. Decimals
indicate the fraction of trials during which a particular behavior was performed.

Incidence of

Touch/

Retreats

Flinches

Feeding

N

A. trigonum

0.96

0.18

0.04

28

Washed A. trigonum

0

0.20

0.30

20

F. pyramitela

0.27

0.10

0.32

30

Misumenoides sp.

0

0.10

0.50

20

2

X

68.11

1.57

9.45

p

<0.001

>0.50

<0.01

preceding analysis of orientation, circular statistical tests could not legitimately be
applied because the data were semicircular.) Finally, although the posFcontact
orientation of the assay spider and the retreat distance were positively correlated
(r = 0.373, P < 0.01) over all touch-retreats, the retreat distance did not vary
significantly with carcass identity.

Our results indicate that female bowl and doily spiders respond differentially to

contact with the surfaces of conspecific females and two different taxa of
heterospecifics. Because the major differences are eliminated after Argyrodes
trigonum carcasses are washed in hexane, we conclude that the relevant
differences are chemical rather than structural.

Because 1) the carcasses of the theridiid and the thomisid are treated very
differently by E pyramitela and 2) responses to carcasses of the kleptoparasitic
predator are least like responses to the two other types of intact spiders, we
further conclude that the assay spiders respond to the A. trigonum carcasses in a
way that is specific to the genus or species of the carcass. The validity of this
conclusion would be suspect if it were not for the apparent appropriateness of the
host spider’s response to contact with the kleptoparasitic predator. We know that
contact with a living A. trigonum is sometimes fatal to the host (references above)
and thus is to be avoided. We should expect, then, to find that chemotactile
recognition of A. trigonum would be followed immediately by flight or other
rapid withdrawal, and that is exactly what the data show. Because feeding
requires intimate contact between the host and the carcass, we should also expect
feeding on A. trigonum to be suppressed, and it apparently is.

The ability of bowl and doily spiders to discriminate among potential predators
based on chemical information is not surprising given their similar abilities in the
realm of intraspecific communication (Suter et al. 1987 and references therein).
Because the detection of predators’ chemicals by prey is well documented in other
arthropod taxa (e.g., ants; Carlin and Johnston 1984), the near absence (Tretzel
1959) from the literature of other examples of spider-to-spider interspecific
chemical communication suggests that the phenomenon demonstrated here is
either rare or has been rarely sought.

The specificity and appropriateness of F. pyramitela'^ responses to contact with
A. trigonum carcasses suggests that the two species have interacted intensely over
relatively long periods of time. The same logic suggests that one might expect to
find interspecific communication of a specific and appropriate nature whenever

1989. The Journal of Arachnology 17:240

two spider species are known to have interacted frequently over the course of
many generations. Because such relationships are probably rare, the accompany-
ing communication may also be found only infrequently.

We thank Mark K. Stowe for his constructive comments on this manuscript.

LITERATURE CITED

Archer, A, F. 1946. The Theridiidae or comb-footed spiders of Alabama. Alabama Mus. Nat. Hist.
Papers, 22:1-67.

Carlin, N. F. and A. B. Johnston. 1984. Learned enemy specification in the defense recruitment system

of an ant. Naturwissenschaften, 71:156-157.

Eberhard, W. G. 1977. Aggressive chemical mimicry by a bolas spider. Science, 198:1173-1175.
Eberhard, W. G. 1981. The single line web of Phoroncida studo Levi (Araneae, Theridiidae): A prey
attractant? J. Arachnol., 9:229-232.

Exline, H. and H. W. Levi. 1962. American spiders of the genus Argyrodes (Araneae, Theridiidae).

Bull. Harvard Mus. Comp. Zool., 127:75-204.

Horton, C. C. 1979. Apparent attraction of moths by the webs of araneid spiders. J. Arachnol, 7:88.
Lamore, D. H. 1958. The jumping spider, Phidippus audax Hentz, and the spider Conopistha trigona
Hentz as predators of the basilica spider, Allepeira lemniscata Walckenaer in Maryland. Proc. Ent.
Soc. Washington, 60:286.

Larcher, S. F. and D. H. Wise. 1985. Experimental studies of the interactions between a web-invading

spider and two host species. J. Arachnol, 13:43-59.

Robinson, M. H. and J. Olazarri. 1971. Units of behavior and complex sequences in the predatory
behavior of Argiope argentata (Fabricius):(Araneae: Araneidae). Smithson. Contrib. Zool, 65:1-
36.

Suter, R. B. 1984. Web tension and gravity as cues in spider orientation. Behav. Ecol Sociobiol,

16:31-36.

Suter, R. B. 1985. Intersexual competition for food in the bowl and doily spider, Frontinella
pyramitela (Araneae, Linyphiidae). J. Arachnol, 13:61-70.

Suter, R. B., C. M. Shane and A. J. Hirscheimer. 1987. Communication by cuticular pheromones in a
linyphiid spider. J. Arachnol, 15:157-162.

Tanaka, H. 1984. Rate of predation by a kleptoparasitic spider, Argyrodes fissifrons, upon a large
host spider, Agelena Umbata (Araneae). J. Arachnol, 12:363-368.

Tretzel, E. 1959. Zum Begegnungsverhalten von Spinnen. Zool. Anz., 163:194-205.

Wise, D. H. 1982. Predation by a commensal spider, Argyrodes trigonum, upon its host: an
experimental study. J. Arachnol, 10:111-116.

Robert B. Suter, Cari M. Shane and Andrea J. Hirscheimer, Department of

Biology, Vassar College, Poughkeepsie, New York 12601 USA.

Manuscript received April 1988, revised July 1988.

AGELENA CONSOCIATA (ARANEAE, AGELENIDAE) AND
ITS NEST ASSOCIATES: INSECT CLEANERS

The density of cooperative spiders within their communal nests can be high:
hundreds and sometimes thousands of individuals. Refuse, such as insect
carcasses and metabolic waste, present a problem to these colonies. While the
cooperative spiders themselves are known to perform cleaning activities (pers.

1989. The Journal of Arachnology 17:241

obs.), Other arthropods associated with the nests, in particular lepidopteran
larvae, have been reported to play a role in refuse removal (Pocock 1909;
Robinson 1977). Pocock (1909) observed small larvae of the moth, Batrachedra
stegodyphobius Walsingham, eating insect carcasses in the nest of an unidentified
species of social eresid, Stegodyphyus sp. in South Africa. Robinson (1977)
studied this cleaning behavior under laboratory conditions in another moth larva,
Neopalthis madates Druce, that is found in the nests of the new world social
theridiid, Anelosimus eximius (Keyserling).

We report here on our field observations of arthropod nest associates of the
West African social spider, Agelena consociata Denis. Five nests of this species
located on the M’Passa Reserve of The Institut de Recherche en Ecologie
Tropicale, near Makokou, Gabon, were intensively studied during the period
from October 16 to December 5, 1987. Forty-eight hours during this period,
made up of 30-minute segments, was devoted to direct observation on the nests
for prey encounters. The presence and activity of insects in the nest area as well
as the spiders’ reactions to this activity, was recorded for each encounter. Each
nest was rotated through a schedule such that all nests were observed for equal
amounts of time at the same set times each day. Field identifications only were
made of most of these taxa because we wished to avoid disruption of colony
activities. We were able to collect ants without disruption because of their
numbers. These were identified to genus.

A total of 791 individual insects were seen in the nests of A. consociata during
the course of these observations. Fifty-two of these insects were attacked and
eaten. Others escaped or were repelled. Three types of insect were tolerated,
lepidopteran larvae, tenebrionid beetles and some small ant species.

Lepidopteran larvae were only observed in the largest, and presumably oldest,
nest under observation. Brach (1977) noted that the older nests of Anelosimus
studiosus (Hentz) were more likely to contain inhabitant arthropods. Brach
observed these larvae to move in rapid and seemingly random spaced jerks with
long intervals of motionlessness. Catepillars in the Agelena nest were never seen
to move across the surface of the webbing but remained in the leaf and web
tangle that made up the central retreat, where there tends to be a build-up of
detritus. There were no observations of spiders approaching these larvae or even
orienting toward them.

In addition to the lepidopterans that may be common to social spider nests,
beetles were frequently seen in all nests regardless of nest size and presumed age.
At least three species of tenebrionids were observed to have free access to most
areas of the colony, walking across the surface of the sheet or inside the nest
retreats. These beetles were occasionally touched by spiders but for the most part
were otherwise ignored. Twice an attack occurred when a tenebrionid beetle came
very close to a cluster of egg sacs. The beetles were driven away by groups of
mature female spiders, that tended to be found hanging from the underside of
these eggs. In these cases the beetles were not killed but merely expelled from the
immediate proximity of the eggs. Agelena consociata discriminated this beetle
type from others such that upon contact, varying forms were eaten or carried to
the edge of the sheet and dropped over the side.

All of the nests under regular observation housed ants which were seen to be
active in all but one nest in at least 20% of the observation periods. The ants, of
the genera Pheidole and Technomyrmex, were approached by the spiders as they

1989. The Journal of Arachnology 17:242

walked across the surface of the webbing. They were also sometimes touched by
Agelena but were never attacked. Other ants discovered by the spiders on the
sheet, even much larger species, were attacked.

A simple experiment, completed at six nests other than those under daily
observation, demonstrated the nest cleaning activity of these ants. Plastic petri
dishes enclosing freshly killed (by freezing) moths were placed on the webbing of
each nest. The lids were closed such that Agelena could not enter them. After 6
hours the nests were revisited. All dishes were found to contain ants and in two
of the cases the carcasses had already been completely consumed.

In summary, lepidopteran larvae inhabiting the webs of social spiders in the
families Theridiidae and Eresidae have been reported. To our knowledge this is
the first account of a similar system in the social agelenids. The lepidopteran
larvae reported for other social spiders were less prominent than ants and beetles
in this system. Although we have documented cleaning activity only for the ants,
observations by others suggest that the larvae at least function in this capacity as
well.

We thank Marcel Bilombi for help in locating Agelena nests and Carole Capel
whose assistance in data collection was invaluable. We also wish to thank Dr. M.
Lacey at the systematic entomology laboratory at the USD A who identified the
ants for us. This report was supported by a National Science Foundation grant
(BSR-83 14900) to S. E. Riechert.

LITERATURE CITED

Brach, V. 1977. Anelosimus studiosus (Araneae: Theridiidae) and the evolution of quasi-sociality in
the theridiid spiders. Evolution, 31:154-161.

Pocock, R. I. 1903. Notes on the commensalism subsisting between a gregarious spider Stegodyphyus
sp. and the moth Batrachedra stegodyphobius. Wlsm. Ent. Mon. Mag., 39:167-170.

Robinson, M. H. 1977. Symbioses between insects and spiders: an association between lepidopteran
larvae and the social spider Anelosimus eximius (Araneae: Theridiidae). Psyche, 83:225-232.

Robert E. Furey and Susan E. Riechert, Graduate Program in Ethology,
Department of Zoology, M3 13 Walters Life Sciences Building, University of
Tennessee, Knoxville, Tennessee 37996-0810 USA.

Manuscript received August 1988, revised November 1988.

SUBMERSION SURVIVAL IN AERIAL WEB- WEAVING
SPIDERS FROM A TROPICAL WET FOREST

INTRODUCTION AND METHODS

Previously I found that several species of temperate zone, web-weaving spiders
that dwell beneath stones possess morphological and behavioral adaptations that
enhance underwater survival (Rovner 1986). These spiders retain large air stores
on the body surface in their dense setal coats; furthermore, they usually show
minimal or no locomotor activity when submerged. Both factors adapt them for

1989. The Journal of Arachnology 17:243

surviving rain-caused flooding. On the other hand, the aerial web-weavers I
tested — Achaearanea tepidariorum (C. L. Koch), Araneus marmoreus Clerck,
Micrathena gracilis (Walckenaer), and Prolinyphia marginata (C. L. Koch) — were
less able to resist drowning. This may have resulted from the fact that their setal
coats retained either a small ventral air store or none at all Also, the araneids
showed intense locomotor activity at the beginning of the test, which was
followed by immotility very early in the test period. After removal from the
water, these four species either became active only after a relatively long time or
not at all, i.e., they had drowned. The high metabolic rate of araneids (Anderson
1970) may have played a role in their drowning so quickly. However, the relative
size of the air store on the body surface and the level of activity were the only
factors I examined.

I subsequently hypothesized that these two factors that conferred drowning
resistance are sufficiently plastic within spider families as to be significantly
modified by selection imposed by local conditions. I anticipated that aerial web-
weaving spiders from a life zone typified by frequent, heavy rainfall (with possible
local flooding) would possess greater drowning resistance than temperate
deciduous forest species.

Here I tested seven species from a lowland tropical wet forest site; six were
araneids and one a theridiid. Work was conducted during a 6-week stay at the
Organization for Tropical Studies’ La Selva station near Puerto Viejo de
Sarapiqui, Heredia Province, Costa Rica (mid-November to late December,
1987). Female spiders were collected near trails in the forest — Micrathena
brevipes (O. P. Cambridge), Micrathena molesta Chickering, Micrathena
schreibersi (Perty), Nephila clavipes (Linn.), and Achaearanea toeniata (Keyse-
rling); in the clearing surrounding the station buildings — Argiope savignyi Levi
and N. clavipes; and on a footbridge over the river — Metazygia sp. Voucher
specimens are deposited in Harvard University’s Museum of Comparative
Zoology.

Within a few days of capture, 10 individuals of each species were submersed
one at a time for 1 h in 25° C distilled water, aerated prior to use. This occurred
in a completely filled, covered, all-glass miniature aquarium of an appropriate
size (27 or 70 ml) for the species being tested. (The inside bottom and back wall
had a white paper liner that provided a surface to which the spider could cling).
Adult N. clavipes were tested in a 260-ml paper cup with a plastic lid.

Each spider was anesthetized with CO2 to facilitate placement in the water;
timing began when the spider resumed movement. As pointed out by J. F.
Anderson (pers. comm.) the use of CO2 for anesthesia complicates the analysis
since the gas affects respiratory processes. However, CO2 anesthesia was the only
means available to me to achieve a uniform initiation of each test. Grasping an
unanesthetized spider to hold it beneath the water during placement of the lid on
the aquarium could have caused overexcitement or even injury in some spiders.
(An aquarium with a bottom inlet and top outlet to allow filling after placement
of the spider and lid closure would prevent the need for anesthesia in future
studies.)

I also wished to conduct some preliminary observations on species differences
in reaction to contact with water and in buoyancy. For this purpose, several
individuals of each species were placed on the surface of water contained in a
plastic basin. After observing the nature of possible behaviors that would improve

1989. The Journal of Arachnology 17:244

Table 1. — Activity, typical behaviors, and mortality of spiders subjected to a 1-h submersion (25° C),
listed in rough order of increasing likelihood of drowning after submersion. Percent of h spent in
locomotion is given as x + SD (range); N = 10.

Species

Percent locomotion

Typical behaviors

Deaths

Micrathena molesta

3.6 ± 2.7 (0.8

-7.8)

rapid crawl/ inactive;

mobile < 15 min

2/10

Metazygia sp.

6.1 + 4.1 (0.1

- 13.2)

rapid crawl/ inactive;
mobile < 25 min

6/10

Achaearanea taeniata

3.6 ± 2.2 (1.4

- 8.8)

rapid crawl/ inactive;
mobile < 15

7/10

Micrathena brevipes

3.1 ± 0.8 (2.2

-4.5)

thrash on bottom;

mobile < 10

7/10

Micrathena schreibersi

2.7 ± 0.8 (1.5

- 4.0)

thrash on bottom;
mobile — 10

7/10

Argiope savignyi

2.2 + 1.1 (0.6

- 4.0)

slow or rapid crawl;
mobile < 10 min

7/10

Nephila clavipes juv.

2.3 ± 1.9 (0.1

-5.4)

slow crawl or thrash;
mobile < 10 min

9/10

Nephila clavipes adult

0.9 ± 0.9 (0.1

-2.4)

slow crawl or thrash;
mobile < 5 min

9/10

survival, I gently pushed each spider beneath the surface to see if it showed
buoyancy after losing contact with the surface film.

RESULTS AND DISCUSSION

In six of the seven species, more than half of the individuals drowned as a
result of the 1-h submersion (Table 1). Such high mortality levels had occurred in
only one of the four temperate zone aerial web-weavers {Micrathena gracilis) that
I tested previously. On this basis, it is evident that the aerial web weavers that I
tested from the tropical wet forest are not any better adapted for drowning
resistance than are those tested from the temperate deciduous forest biome.

During submersion, most of the wet forest species were motile for less than 15
min (Table 1), a level of endurance similar to that shown by the two temperate
zone araneids tested previously. Metazygia sp., which had built webs on a
footbridge above the river, was able to remain motile the longest and, of all the
species tested here, spent the greatest percentage of the hour in locomotion.
Micrathena molesta was the most resistant of the spiders to drowning during
submersion; however, there was no obvious factor to account for its superior
endurance. Least resistant to drowning were adult Nephila clavipes, which
became immotile after a very brief initial period of slow crawling or very slow
thrashing. If submerged during rain-caused flooding in the field, adults of this
species would survive only briefly while attempting to climb above the water’s
surface. Juvenile N. clavipes showed longer periods of motility but nonetheless
drowned as often as the adults in the 1-h test (Table 1). The remaining araneids —
M. brevipes, M. schreibersi, and A. savignyi — fared little better than N. clavipes.
The single theridiid tested, A. taeniata, was similar to the best-adapted araneid,
M. molesta, as to percent locomotion and duration of motility but drowned as
frequently as the last-mentioned group of araneid species.

I also studied the effects of placing individuals of the seven species on the
surface of an open pool of water in a basin. Pushing each floating spider beneath

1989. The Journal of Arachnology 17:245

the surface revealed a difference in buoyancy between two groups of species. The
three Micrathena species, Metazygia sp., and N. davipes sank to the bottom,
whereas Argiope savignyi and Achaearaneataeniata usually were buoyed up,
apparently by having a temporary air store on the body surface of relatively large
size for the mass of the spider. The gas store in the booklungs and the
hydrophobic nature of the cuticle also may have influenced buoyancy (J. F.
Anderson pers. comm.).

Of additional interest were differences among the species’ responses to contact
with the water’s surface film. Three species — A. taeniata, Metazygia sp., and N.
davipes — floated on their sides with their legs drawn in, a behavior that kept one
or both lung slits exposed to the air. The three Micrathena species and A.
savignyi usually floated upright, with their lung slits below the surface. Although
A. savignyi had an opisthosomal air film that probably maintained an air supply
channel from its slits to the air above the water, that did not seem to be the case
for the three Micrathena species, which have relatively few opisthosomal setae.
However, one of these, M. brevipes, showed a behavior that solved the dilemma.
By lifting its massive opisthosoma to a position above the carapace, the spider
then further rolled into a partial “flip” and ended up in an inverted position.
Thus, now floating on its dorsal surface, the spider avoided drowning.

The data presented here and in my previous study (Rovner 1986) suggest that
araneids — even those of tropical wet forests — can withstand only relatively brief
periods of submersion, compared to web-weavers that dwell beneath stones.
Apparently, the features that provide drowning resistance in spiders are general to
the families possessing them and cannot be readily modified to adapt species of
other families to local conditions. Indeed, the marsh-dwelling araneid Acanthe-
peira venusta (Banks) in Florida, which climbs down into the water when
disturbed, has relatively brief submersion periods of 3 min or less (Folkerts and
Mullen 1987). On the other hand, the absence of adaptations for aquatic
conditions in aerial web weavers of the tropical wet forest may suggest that these
spiders are usually able to avoid falling into floodwaters during heavy rain and
that, if they do fall onto the water, they may have adaptations reducing the
possibility of submersion or drowning after being trapped in the surface film.

I thank the Organization for Tropical Studies for permitting me to use the
facilities at their La Selva research station, Brent D. Opell for the loan of an
equipment item, Herbert W. Levi for taxonomic assistance, and Renee S. Lizotte
for assistance in collecting spiders. John F. Anderson and Brent D. Opell
suggested a number of improvements in the original manuscript. This study was
supported by National Geographic Society grant no. 3532-87 to the author.

LITERATURE CITED

Anderson, J. F. 1970. Metabolic rates of spiders. Comp. Biochem. Physiol., 33:51-72.

Folkerts, D. R. and G. R. Mullen. 1987. Aquatic submergence of Acanthepeira venusta (Banks)
(Araneae, Araneidae)., J. ArachnoL, 15:137-138.

Rovner, J. S. 1986. Spider hairiness: air stores and low activity enhance flooding survival in inland
terrestrial species. Actas X Congr. Int. ArachnoL Jaca/Espana, 1:123-129.

Jerome S. Rovner, Department of Zoological Sciences, Ohio University,
Athens, Ohio 45701 USA.

Manuscript received July 1988, revised November 1988.

1989. The Journal of Arachnology 17:246

MITE PARASITISM OF THE POLYMORPHIC SPIDER,
ENOPLOGNATHA OVATA (ARANEAE, THERIDIIDAE),
FROM COASTAL MAINE

INTRODUCTION AND METHODS

The theridiid spider Enoplognatha ovata (Clerck) is a common inhabitant of
weedy vegetation along the northeastern seaboard of North America. Recent
research into the ecological genetics of color polymorphism in E. ovata in Maine
populations (Reillo and Wise 1988a, b, c) has uncovered considerable parasitism
of this species by larval Parasitengona mites. Here I present parasitism
frequencies for color morphs of mature female spiders from 15 natural
populations.

Thirty-five coastal Maine populations of E. ovata were censused during mid-
August from 1986-1987. Populations were distributed between Boothbay, ME
(43° 50' N. Lat./69°37'W. Long.) and Acadia National Park at Mt. Desert Island
(44°25' N. Lat./68°15' W. Long.) (see map in Reillo and Wise [1988b]).
Descriptions of the color phenotypes and life history of E. ovata can be found
elsewhere (Seligy 1971; Oxford 1976; Wise and Reillo 1985), and censusing
techniques for estimating morph frequencies are discussed in Oxford (1976,
1985a, b) and Reillo and Wise (1988b). Spiders were examined for mites in the

Figures 1-4. — Scanning electron micrographs of E. ovata host and Parasitengona mites: 1, adult
female spider with trombidiid mites clustered on opisthosoma (32X magnification); 2, trombidiid mite
with gnathosoma attached to host at left (260X); 3, gnathosoma of erythraeid mite, dorsal view
(940X); 4, gnathosoma, ventral view, showing subcapitalum and palps (1500X).

1989. The Journal of Arachnology 17:247

field with the naked eye. Samples of hosts and mites were preserved in alcohol or
by freezing. Specimens were identified by W. Calvin Welbourn, curator at the
acarology laboratory of Ohio State University.

RESULTS AND DISCUSSION

Two families of mites were found among the samples: Trombidiidae, likely
Trombidium auroraense or a close relative (Figs. 1, 2); and Erythraeidae,
probably of the genus Leptus (Figs. 3, 4). Mites were encountered only as larvae
(six legs), bright orange in color, usually attached to the host on the dorsal side
along the margins of the carapace or on top the opisthosoma, with clusters often
nestled on or adjacent to the pedicel (Fig. 1). Total body length ranged from 0.20
mm to >1.0 mm and varied with the extent of engorgement. The number of mites
per host was not scored in the field, but examination of preserved specimens
usually revealed one or several larvae/ host, with occasional heavy loadings in
excess of 15 larvae/ host.

Since not all mites scored in the field were collected and identified, I will
present gross parasitism frequencies for both mite families collectively. Mites were
found among 15 of the 35 censused populations (Table 1). The frequency of
parasitism among mature females was highly variable for affected populations,
ranging from 0.1% to 20.7% (mean ± SE for all populations and years = 0.057 ±
0.012). Mites were also observed among immature spiders of both sexes and
mature males, but small sample sizes for these categories prohibited estimating
parasitism frequencies. No clear association between mite incidence and
environment could be detected; however, populations inhabiting edge vegetation
along open fields or areas beneath sparse canopy appeared to consistently contain
mites whereas shaded populations were generally found to be mite-free.

Parasitism frequencies did not change significantly between 1986 and 1987 for
three of five populations having frequencies > 3% for both years (populations
BPP, NLl, MP; Table 1; Chi-squares < 1.46, df - \, continuity correction, P >

0. 226). For the other two populations, parasitism in one (population DM) more
than doubled (Chi-square = 6.74, df - continuity correction, P - 0.009), while
in the other (population NL2) it decreased by nearly half (Chi-square = 7.73, df -

1, continuity correction, P = 0.005).

I found no evidence of differential parasitism of color morphs of the host E.
ovata. For populations in which phenotype and parasitism frequencies were
sufficient to conduct contingency chi-square tests (populations MP, NLl, NL2,
DM; Table 1), parasitism was random with respect to phenotype in 1986 (Chi-
squares < 4.07, df= 2, P> 0.131) and 1987 (Chi-squares < 4.66, # = 2, P >
0.097). I also found no evidence that parasitized females reproduced less
successfully than non-parasitized females. For populations with total number of
parasitized females > 10 (Table 1), there was no difference between the relative
proportions of unparasitized females with egg sacs and parasitized females with
egg sacs in 1986 (Chi-squares < 0.106, df - I, continuity correction, P > 0.744) or
1987 (Chi-squares < 0.544, df - I, continuity correction, P > 0.461). It is of
course impossible to determine from my data whether parasitism may have
adversely affected fecundity via decreased egg production or may have rendered
females more susceptible to mortality prior to ovipositing. However, these data

1989. The Journal of Arachnology 17:248

Table 1. — Morph frequencies and incidence of mite parasitism among coastal Maine populations of
mature female E. ovata. LIN = lineata phenotype; RED = redimita phenotype; OVA = ovata
phenotype; ES = total number of females with egg sacs; “M” refers to number of individuals with
mites in each category; %M = frequency of parasitism for sample size N. Egg sac data given only for
populations with total #M > 10.

POPN/YR

LIN

RED

OVA

Combined

ES

N

(M)

N

(M)

N

(M)

N

(M)

(%M)

N

(M)

GCP/86

181

(12)

5

(0)

0

(0)

186

(12)

(0.065)

184

(12)

SS/86

221

(1)

3

(0)

0

(0)

224

(1)

(0.004)

MH/86

197

(2)

6

(0)

0

(0)

203

(2)

(0.010)

MH/87

170

(1)

4

(0)

0

(0)

174

(1)

(0.006)

BPP/86

202

(23)

4

(0)

0

(0)

206

(23)

(0.112)

206

(23)

BPP/87

196

(21)

3

(0)

0

(0)

199

(21)

(0.106)

131

(10)

DC/87

2485

(4)

203

(0)

87

(0)

2775

(4)

(0.001)

DM/86

155

(9)

17

(1)

5

(0)

177

(10)

(0.056)

175

(10)

DM/87

194

(24)

29

(7)

4

(1)

227

(32)

(0.141)

152

(24)

NLl/86

840

(35)

189

(10)

75

(6)

1104

(51)

(0.046)

1088

(48)

NLI/87

1535

(53)

244

(9)

138

(8)

1917

(70)

(0.037)

1038

(44)

NL2/86

435

(82)

105

(29)

35

(8)

575

(119)

(0.207)

543

(106)

NL2/87

481

(54)

93

(8)

55

(H)

629

(73)

(0.116)

305

(40)

NL3/86

204

(2)

63

(0)

27

(0)

294

(2)

(0.007)

NL3/87

253

(3)

47

(2)

26

(2)

326

(7)

(0.021)

MP/86

238

(25)

25

(1)

4

(0)

267

(26)

(0.097)

263

(26)

MP/87

576

(67)

59

(10)

17

(0)

652

(77)

(0.118)

369

(46)

OCR 1/86

127

(3)

32

(2)

4

(0)

163

(5)

(0.031)

OCR2/86

59

(2)

15

(0)

1

(0)

75

(2)

(0.027)

OCR4/86

38

(2)

7

(0)

2

(0)

47

(2)

(0.043)

J/86

209

(I)

13

(0)

2

(0)

224

(1)

(0.004)

C/87

194

(0)

38

(2)

9

(0)

241

(2)

(0.008)

suggest that parasitism by Parasitengona larvae to not provide a selective
mechanism for maintenance of color polymorphism in coastal Maine E. ovata
populations.

I am indebted to W. C. Welbourn for identifications and for much-needed
attention to an earlier version of this note. I also thank E Stroup and B. Burd
for field work, and P. Rutledge for the S.E.M.’s. This work was supported in part
by an NSF Graduate Fellowship.

LITERATURE CITED

Oxford, G. S. 1976. The colour polymorphism in Enoplognatha ovatum (Clerck) (Araneae:
Theridiidae) — Temporal stability and spatial variability. Heredity, 36:369-381.

Oxford, G. S. 1985a. A countrywide survey of colour morph frequencies in the spider Enoplognatha
ovata (Clerck) (Araneae: Theridiidae); evidence for natural selection. Biolog. J. Linnean Soc.,
24:103-142.

Oxford, G. S. 1985b. Geographical distribution of phenotypes regulating pigmentation in the spider
Enoplognatha ovata (Clerck) (Araneae: Theridiidae). Heredity, 55:37-45.

Reillo, P. R. and D. H. Wise. 1988a. Genetics of color expression in the spider Enoplognatha ovata
(Araneae: Theridiidae) from coastal Maine. American Midland Nat., 119:318-326.

Reillo, P. R. and D. H. Wise. 1988b. Temporal and spatial patterns of morph-frequency variation
among coastal Maine populations of the polymorphic spider Enoplognatha ovata (Araneae:
Theridiidae). American Midland Nat., 119:337-354.

Reillo, P. R. and D. H. Wise. 1988c. An experimental evaluation of selection on color morphs of the
polymorphic spider Enoplognatha ovata (Araneae: Theridiidae). Evolution, 42:1172-1189.

1989. The Journal of Arachnology 17:249

Seligy, V. L. 1971. Postembryonic development of the spider Enoplognatha ovata (Clerck) (Araneae:
Theridiidae). Zool. J. Linnean Soc., 50:21-31.

Wise, D. H. and P. R. Reillo. 1985. Frequencies of color morphs in four populations of Enoplognatha
ovata (Araneae: Theridiidae) in eastern North America. Psyche, 92:135-144.

Paul R. Reillo^ Department of Biological Sciences, University of Maryland,
Baltimore County Campus, Catonsville, Maryland 21228 USA.

Manuscript received May 1988, revised December 1988.

'Current address: Department of Zoology, University of Maryland, College Park, Maryland 20742
USA.

NOCTURNAL PREDATION BY MISUMENA VATIA
(ARANEAE, THOMISIDAE)

Spiders of the family Thomisidae are typically small ground or vegetation
inhabitants which capture their prey by ambush and do not spin webs (Gertsch
1979). Because of their dependence on vision, thomisids are primarily active
during the day (Foelix 1982). The crab spider Misumena vatia (Clerck) is one of
the most abundant and widely distributed of the “flower spiders” in North
America. It is commonly observed during the day on flowers such as goldenrod
in the late summer and fall (Kaston 1978). Misumena vatia has been reported to
feed in the daytime on dragonflies and butterflies (Lovell 1915), as well as on
bumblebees, honeybees, hoverflies, and various unidentified insects (Morse 1979).
To our knowledge, Morse (1981) presents the only record of nocturnal predation.
He does not indicate when in the nocturnal period it occurred or the species of
the “moth” prey. While conducting nocturnal sampling of cotton insects, we
occasionally observed evidence of predation by M. vatia. Although these
observations are preliminary, we believe they provide further evidence for the
nocturnal activities of this spider.

Observations were conducted in a Uha fallow crimson clover field in
Washington Co., Mississippi, from 4-9 September 1987. Weather conditions
during this period were moderate; temperatures ranged between lows of 16°C and
highs of 33° C; there was no precipitation; incidence of solar radiation averaged
over 400 langleys/day; and the wind blew occasionally (0-10 km/h) from the
southwest. The region is largely agricultural, with nearly all the surface area
covered by catfish ponds, roads and drainage ditches, and row crops including;
cotton, corn, soybeans, grain sorghum and rice. This particular field had been
undisturbed since spring and was overgrown with Johnsongrass. Scattered
throughout the field and protruding above the canopy were isolated individuals of
the Common Sunflower, Helianthus annuus L. (Compositae). These plants were
1. 7-2.0 m tall and contained 8-12 inflorescences each, from 0.3 to 2.0 m above the
ground. One particular plant, the only one in full bloom, was examined daily at

1989. The Journal of Arachnology 17:250'

Table I. — Prey of Misumena vatia and time observed.

Date

Time

Taxon of prey

Prey stage

4 Sept.

2100

Trkhoplusia ni (Hubner) (Lepidoptera: Noctuidae)

adult

2130

T. ni

adult

2200

Diabrotica undecimpunctata howardi Barber

(Coleoptera: Chrysomelidae)

adult

2230

Colomychus talis (Grt.) (Lepidoptera: Pyralidae)

adult

2300

T ni

adult

9 Sept.

2100

T ni

adult

2130

T ni

adult

2230

D. undec. howardi

adult

0700-0800 for the presence of M. vatia and at no time were more than two
individuals detected, none with prey. This same plant, when observed after dark,
contained as many as nine individuals, some with prey.

Local sunset during early September is approximately 1915 with moonrise
occurring at 1500 on 4 Sept, and 2130 on 9 Sept. On both nights from 2000-2400,
the Helianthus plant was visually searched for spiders, with the aid of a red-
filtered headlamp. Numbered tape markers were attached to a stem within 10 cm
of each individual. At 30~min intervals each spider was examined and removed
from the plant if prey had been captured.

On 4 Sept., nine M. vatia were observed on one Helianthus plant, with five
capturing prey between 2000 and 2400 (Table 1). On 9 Sept., six M. vatia were
observed on the same plant, with three individuals obtaining prey during the
observation period. All were observed as solitary occupants of an inflorescence,
positioned with the center of the vertically oriented flower and facing downward,
with the first two pair of legs spread wide in an “alert” posture.

A cursory comparison of diurnal and nocturnal observations at Helianthus
during early September indicated considerably more night-time insect activity.
Not only were the inflorescences attractive to various insects, but moths were also
observed feeding at the extra-floral nectaries located on the petal bracts.
Misumena vatia appears to have responded to this nocturnal peak in prey activity
by expanding its typical diurnal predatory activity into at least the early portion
of the night. This same general phenomenon was reported by Morse (1981), who
demonstrated that nocturnal predation by M. vatia on milkweed, but not on
goldenrod, was associated with high nocturnal insect activity on milkweed
compared to low nocturnal activity on goldenrod. We therefore suggest that M.
vatia does not depend on visual cues and has the ability to be a successful
predator outside the typical thomisid activity period.

LITERATURE CITED

Foelix, R. F. 1982. Biology of Spiders. Harvard Univ. Press, Cambridge.

Gertsch, W. J. 1979. American Spiders, 2nd ed. Van Nostrand Reinhold, New York.

Kaston, B. J. 1978. How to Know the Spiders, 3rd ed. Wm. C. Brown Co., Dubuque, Iowa.

Lovell, J. H. 1915. Insects captured by Thomisidae. Canadian EntomoL, 47:115-116.

Morse, D. H. 1979. Prey capture by the crab spider Misumena calydna (Araneae: Thomisidae).

Oecologia, 39:309-319.

Morse, D. H. 1981. Prey capture by the crab spider Misumena vatia (Araneae: Thomisidae).

American Midi. Natur., 105:358-367.

1989. The Journal of Arachnology 17:251

Timothy C. Lockley, USDA-APHIS=^PPQ, 3505-25th Ave., Gulfport, MS
39501 USA; Orrey R Young and Jane Leslie Hayes, UDSA-ARS-SFCIML, P.O.
Box 346, Stoneville, MS 38776 USA.

Manuscript received December 1988, revised March 1989.

A SATHROCHTHONIUS NORTH OF THE EQUATOR
(PSEUDOSCORPIONIDA, CHTHONIIDAE)

The genus Sathrochthonius was established by J. C. Chamberlin (1962) on the
basis of S. tuena, n. sp., from near Sydney, N.S.W., Australia. Subsequently, 2
other species were described from Australia {S. crassidens Beier, 1966a, from New
South Wales, and S. webbi Muchmore, 1982, from Queensland), 1 species from
New Caledonia (5'. kaltenbachi Beier, 1966b). and 2 species from New Zealand
{S. maoricus and S. insulanus Beier, 1976). Unexpectedly, a species was
discovered in central Chile {S. pefauri Vitali-di Castri, 1974); and now a new
Sathrochthonius has been found in southeastern Venezuela at latitude 5°N.

Sathrochthonius venezuelanus, new species
Figs. 1-3

Type data. — Holotype female (WM7067. 01001), from VENEZUELA: Bolivar,
La Gran Sabana, 9 km from Chivaton Hotel toward Kavanayen (about 5°30'N,
6P30'W), (1370 m), 29 June 1987, (M. A. Ivie), under bark of dead tree;
deposited in Florida State Collection of Arthropods, Gainesville.

Diagnosis. — A 2-eyed species much like N. pefauri but smaller (palpal femur <
0.45 mm), with a distinct spinneret on the movable finger of the chelicera, and
with no trace of coxal spines or granulations on pedal coxae.

Description of female holotype (male unknown). — Generally typical of the
genus (Chamberlin 1962:303). Body and appendages pale tan. Carapace mostly
smooth, finely reticulated laterally; entire anterior margin finely denticulated,
slightly depressed at middle, with no obvious epistome; 2 corneate eyes;
chaetotaxy 6-4-4-2-2. Coxal area typical except that there are no coxal spines and
no granulations on any of the coxae; chaetotaxy 2-2-3:2-6:2-6(7):2-8:2-9;
intercoxal tubercle bisetose. Abdomen typical; tergites and sternites smooth;
tergal chaetotaxy 6:6:8:9:9:8:8:8:8:8:T2T:0; sternal chaetotaxy
10:(3) 10(3):(2) 10(2): 11:11:11: 12:9:10:0:2.

Chelicera 0.65 as long as carapace; hand with 6 setae; flagellum of about 10
pinnate setae; spinneret a distinct projection from movable finger (Fig. 1); serrula
exterior of about 18 blades.

Palp robust (Fig. 2); femur 3.15, tibia 1.9, and chela 3.3 times as long as broad;
hand 1.6 times as long as deep; movable finger 1.16 times as long as hand.
Surfaces smooth except for small granules on bases of chelal fingers and some
elevated setal areoles on femur and trochanter. Trichobothria as shown in Fig. 3.

1989. The Journal of Arachnology 17:252

Figures 1-3. — Sathrochthonius venezuelanus, n. sp., holotype: 1, tip of movable finger of chelicera;
2, right palp, dorsal view; 3, left chela, lateral view.

Fixed chelal finger with 37 teeth, all cusped; movable finger with 39 teeth, cusped
distally and proximally but rounded in middle of row; fixed finger with an
accessory denticle on internal surface near distal end.

Legs typical. Leg IV with entire femur 2.6 and tibia 3.65 times as long as deep;
tactile seta near proximal end of both basitarsus and telotarsus; telotarsus lacking
a tooth at distal end of upper margin.

Measurements (mm). — Body length 1.30. Carapace 0.45/0.43. Palpal trochanter
0.215/0.125; femur 0.41/0.13; tibia 0.245/0.13; chela 0.63/0.19; hand 0.32/0.20;
movable finger 0.37 long. Leg IV: entire femur 0.40/0.155; tibia 0.31/0.085;
basitarsus 0.14/0.06; telotarsus 0.245/0.05.

Remarks. — It was a great surprise to find this specimen in a collection from
Venezuela, as Sathrochthonius has been known in the Western Hemisphere only
from central Chile, well south of the equator. Evidently the genus is distributed
over a great area of South America, not confined to the southwestern part, to the
Nothofagus forest, as suggested by VitalLdi Castri (1973, 1974). Additional
collecting may well show that Sathrochthonius, like Austrochthonius, is
represented in South Africa as well (cf. Vitali-di Castri, 1973).

Sathrochthonius venezuelanus does not possess the small tooth at the end of
the upper margin of telotarsus IV as described by VitalLdi Castri for S. pefauri
(1974:199 and fig. 10). However, reexamination of some paratypes of S. webbi
Muchmore (1982) reveals the presence of such a projection in that Australian
species. Here, the '‘tooth” is actually a spinelike projection of the rim of the
areole of the large terminal seta of the upper tarsal margin, not an extension of
the tarsal surface itself. Such an elaboration of the terminal areole is more or less
developed in some other chthoniids as well (personal observation). And it is

1989. The Journal of Arachnology 17:253

similar to the projection of the areoles of the coxal spines in the North American
genera Apochthonius and Kleptochthonius (cf. Benedict and Malcolm 1973),

I am much indebted to Michael A. Ivie for providing the material on which
this description is based.

LITERATURE CITED

Beier, M. 1966a. On the Pseudoscorpionidea of Australia. Australian J. Zool, 14:275-303.

Beier, M. 1966b. Ergebnisse der osterreichischen Neukaledonian-Expedition 1965. Pseudoscorpioni-
dea. Ann. Naturhistor. Mus. Wien, 69:363-371.

Beier, M. 1976. The pseudoscorpions of New Zealand, Norfolk, and Lord Howe. New Zealand J.
Zook, 3:199-246.

Benedict, E. M. and D. R. Malcolm. 1973. A new cavernicolous species of Apochthonius
(Chelonethida: Chthoniidae) from the western United States with special reference to troglobitic
tendencies in the genus. Trans. American Microsc. Soc., 92:620-628.

Chamberlin, J. C. 1962. New and little-known false scorpions, principally from caves, belonging to the
families Chthoniidae and Neobisiidae (Arachnida, Chelonethida). Bull. American Mus. Nat. Hist.,
123:299-352.

Muchmore, W. B. 1982. A new cavernicolous Sathrochthonius from Australia (Pseudoscorpionida;
Chthoniidae) Pacific Insects, 24:156-158.

Vitali-di Castri, V. 1973. Biogeography of pseudoscorpions in the mediterranean regions of the world.
Pp. 295-305, In Mediterranean Type Ecosystems, Origin and Structure. (F. di Castri and H. A.
Mooney, eds.) Ecological Studies. Analysis and Synthesis, vol. 7. Springer-Verlag, New York.
Vitali-di Castri, V. 1974. Presencia en America del Sur del genero Sathrochthonius (Pseudoscorpio-
nida) con descripcidn de una nueva especie. Physis (Buenos Aires), 33:193-201.

William B. Muchmore, Department of Biology, University of Rochester,
Rochester, NY 14627 USA.

Manuscript received January 1989, revised March 1989.

APHID PREDATION BY HARVESTMEN

IN POTATO FIELDS IN SCOTLAND

Several species of aphids damage potatoes in Scotland. Dixon (1986) has
studied the natural enemies of potato aphids, paying particular attention to
polyphagous predators like harvestmen, spiders and ground beetles. The
harvestmen, Leiobunum rotundum (Latreille) (Todd 1950) and Phalangium opilio
Linne (Bristowe 1949) are known to prey on aphids, but little information exists
on harvestmen as predators of aphids on potatoes. This paper reports on the
predation of aphids by harvestmen in potato fields in Scotland.

Harvestmen were caught in pitfall traps during the summers of 1983-85
inclusive in fields of potatoes (cv. Maris Piper) near Edinburgh, Scotland. Pitfall
traps were clear polystyrene containers (perimeter: 25 cm) placed in the potato
furrows with their rims at soil level. Traps were emptied weekly. Forty-five ml of
10% formalin and a drop of liquid detergent were added to each trap at each

1989. The Journal of Arachnology 17:254

Table 1. — Total numbers of harvestmen caught in pitfall traps in Scottish potato fields during 1983
and 1984.

Species

1983

1984

Phalangium opilio Linne

65

61

Opilio saxatilis C. L. Koch

47

3

Mitopus morio (Fabricius)

16

7

Paroligolophus agrestis (Meade)

11

0

Immature harvestmen

1

19

Leiobunum rotundum (Latreille)

1

1

Oligolophus tridens (C. L. Koch)

1

0

Opilio parietinus (DeGeer)

1

0

sampling occasion. One hundred and twenty traps were spaced over 1.3 ha in
1983, but this trapping intensity was too time consuming and in each of 1984 and
1985, 56 traps were used in 2.0 ha.

Harvestmen were dissected under a binocular microscope and stomachs with
attached gastric caecae were removed, placed on slides in 40% glycerine (after
Loughridge and Luff 1983), and examined at lOOX bright-field. Aphid remains
were counted and identified to species {Macrosiphum euphorbiae (Thomas),
Myzus persicae (Sulzer) (Homoptera: Aphididae)) where possible. Ail harvestmen
were dissected except during late summer when aphids were not present in the
field.

Seven species of harvestmen were trapped during 1983 and 1984, but none in
1985 (Table 1). Of a total of 233 individuals, 54% were Phalangium opilio, a
longdegged, highly active species which usually inhabits woodland and bushes
(Sankey and Savory 1974). Dempster (1967) suggested that R opilio was an
important predator of Pieris rapae (L.) (Lepidoptera: Pieridae) in Brussel sprout
crops. This harvestmen has been recorded from potato fields in England (Foster
1972). The data for other species trapped are combined over time because the
numbers trapped were low. Phalangium opilio was present over most of the
summer, but Opilio saxatilis, the next most frequently trapped species, occurred
almost entirely in September. Although populations were probably underesti-
mated by pitfall trapping, the method will have indicated which species were
present. The usual methods of sampling harvestmen include beating and sweeping
foliage or using funnels (Berlese, Tullgren) to assess soil and litter samples
(Sankey and Savory 1974).

Of the species trapped in potato fields, all contained aphid remains except the
single Oligolophus tridens. Fifty-four percent of 113 P. opilio examined, had
eaten aphids. The maximum number of aphids found in a harvestman was in a P.
opilio which contained at least five (30 legs, 4 siphunculi, and 3 rostra).

Opilionids have been neglected and probably undervalued as predators of crop
pests. The present study has shown that several species prey on aphids in potato
crops. Harvestmen may have great potential in natural pest control and warrant
further research.

LITERATURE CITED

Bristowe, W. S. 1949. The distribution of harvestmen (Phalangida) in Great Britain and Ireland, with
notes on their names, enemies and food. J. Anim. EcoL, 18:100-114.

1989. The Journal of Arachnology 17:255

Dempster, J. P. 1967. A study on the effects of DDT applications against Pieris rapae on the crop
fauna. Proc. 4th British Insect, and Fung. Conf., 1:19-25.

Dixon, P. L. 1986. Pesticides and natural enemies (particularly ground beetles) of aphids on potato.
Ph. D. Thesis, Univ. of Edinburgh.

Foster, G. N. 1972. The population dynamics of aphids infesting potato. Ph. D. Thesis, Univ. of
Newcastle-upon-Tyne.

Loughridge, A. H. and M. L. Luff. 1983. Aphid predation by Harpalus rufipes (DeGeer) (Coleoptera:

Carabidae) in the laboratory and field. J. appl. EcoL, 20:451-462.

Sankey, J. H. P. and T. H. Savory. 1974. British Harvestmen. Linn. Soc. London, Acad. Press, 76 pp.
Todd, V. 1950. Prey of harvestmen (Arachnida, Opiliones). Ent. Mon. Mag., 86:252-254.

P. L. Dixon^ and R, G. McKinlay, Edinburgh School of Agriculture, West
Mains Road, Edinburgh, Scotland, U.K. EH9 3JG.

Manuscript received January 1989, revised March 1989.

Present Address: Forestry Canada, P.O. Box 6028, St. John’s, Newfoundland, Canada AlC 5X8.

THE AMERICAN ARACHNOLOGICAL SOCIETY

President:

William A. Shear (1987-1989)

Department of Biology
Hampden-Sydney College
Hampden-Sydney, Virginia 23943

Membership Secretary:

Norman I. Platnick (appointed)

American Museum of Natural History
Central Park West at 79th Street
New York, New York 10024

Secretary:

James W. Berry (1987-1988)

Department of Zoology
Butler University
Indianapolis, Indiana 46208

Directors:

J. A. Coddington (1988-1990), James C. Cokendolpher (1987-1989), Jerome S.
Rovner (1987-1989).

Honorary Members:

P. Bonnet, W. J. Gertsch, H. Homann, H. W. Levi, G. H. Locket, A. F. Millidge,
M. Vachon, T. Yaginuma.

The American Arachnological Society was founded in August, 1972, to
promote the study of Arachnida, to achieve closer cooperation between amateur
and professional arachnologists, and to publish The Journal of Arachnology.

Membership in the Society is open to all persons interested in the Arachnida.
Annual dues are $30.00 for regular members, $20.00 for student members and
$50.00 for institutions. Correspondence concerning membership in the Society
must be addressed to the Membership Secretary. Members of the Society receive
a subscription to The Journal of Arachnology. In addition, members receive the
biannual newsletter of the Society, American Arachnology.

American Arachnology, edited by the Secretary, contains arachnological news
and comments, requests for specimens and hard-to-find literature, information
about arachnology courses and professional meetings, abstracts of papers
presented at the Society’s meetings, address changes and new listings of
subscribers, and many other items intended to keep arachnologists informed
about recent events and developments in arachnology. Contributions for
American Arachnology must be sent directly to the Secretary of the Society.

President-Elect:

George W. Uetz (1987-1989)
Department of Biological Sciences
University of Cincinnati
Cincinnati, Ohio 45221

Treasurer:

Gail E. Stratton 0987-1989)
Department of Biology
Albion College
Albion, Michigan 49224

Archivist:

Vincent D. Roth

Box 136

Portal, Arizona 85632

Research Notes

Spider vs. spider: Frontinella pyramitela detects Argyrodes trigonum via
cuticular chemicals, Robert B. Suter, Cari M. Shane and

Andrea J. Hirscheimer 237

Agelena consociata (Araneae, Agelenidae) and its nest associates: insect

cleaners, Robert E. Furey and Susan E. Riechert 240

Submersion survival in aerial web-weaving spiders from a tropical wet

forest, Jerome S. Rovner 242

Mite parasitism of the polymorphic spider, Enoplognatha ovata

(Araneae, Theridiidae), from coastal Maine, Paul R. Reillo 246

Nocturnal predation by Misumena vatia (Araneae, Thomisidae),

Timothy C. Lockley, Orrey R Young and Jane L. Hayes 249

A Sathrochthonius north of the equator (Pseudoscorpionida,

Chthoniidae), William B. Muchmore 251

Aphid predation by harvestmen in potato fields in Scotland, R L. Dixon
and R. G. McKinlay 253

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 17 Feature Articles NUMBER 2

AnMisis del comportamiento sexual y produccion de ootecas de

Theridion rufipes (Araneae, Theridiidae), Alda Gonzalez 129

Pachyloides hades, nueva especie de opilion de la Argentina

(Opiliones, Gonyleptidae, Pachylinae), Luis E. Acosta 137

Spiders of Spanish moss in the delta of Mississippi, Orrey R Young and

Timothy C. Lockley 143

A test of the mechanical isolation hypothesis in two similar spider

species, Amy D. Ware and Brent D. Opell 149

Population ecology of Pardosa ramulosa (Araneae, Lycosidae) in flooded
rice fields of northern California, Michael J. Graze, Albert A. Grigarick

and Kirk A. Smith 163

Diet-induced color change in the Hawaiian happy-face spider

Theridion grallator, (Araneae, Theridiidae), Rosemary G. Gillespie 171

Sex-biased predation by web-spinning spiders (Araneae) on spruce

bud worm moths, Daniel T Jennings and Mark W. Houseweart 179

Spectral sensitivities of photoreceptors in the ocelli of the tarantula,

Aphonopelma chalcodes (Araneae, Theraphosidae), R. Dennis Dahl

and A. M. Granda 195

The harvestman family Phalangodidae. 2. The new genus, Microcina

(Opiliones, Laniatores), Thomas S. Briggs dind Darrell Ubick 207

Descriptions of two new species of the genus Architis (Araneae, Pisauridae)

and the female of A. vilhenq, James E. Carico 221

“Emerit’s glands” in Cybaeota (Araneae, Agelenidae),

Robert G. Bennett 225

(continued on back inside cover)

Cover photograph, female Cupiennius getazi feeding on Norops limifrons,
La Selva Biological Station of OTS, Costa Rica, by J. S. Rovner
Printed by the Texas Tech Press, Lubbock, Texas, USA
Posted at Lubbock, Texas, 29 September 1989

The Journal of

ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 17

FALL 1989

NUMBER 3

THE JOURNAL OF ARACHNOLOGY

EDITOR: James E. Carico, Lynchburg College
ASSOCIATE EDITOR: Jerome S. Rovner, Ohio University
EDITORIAL BOARD: J. E. Carrel, University of Missouri; J. A. Coddington,
National Museum of Natural History, Smithsonian Institution; J. C.
Cokendolpher, Texas Tech University; F. A. Coyle, Western Carolina
University; C. D. Dondale, Agriculture Canada; W. G. Eberhard, Universidad
de Costa Rica; M. E. Galiano, Museo Argentine de Ciencias Naturales; M. H.
Greenstone, BCIRL, Columbia, Missouri; N. V. Horner, Midwestern State
University; D. T. Jennings, NEFES, Morgantown, West Virginia; V. F. Lee,
California Academy of Sciences; H. W. Levi, Harvard University; E. A.

Maury, Museo Argentine de Ciencias Naturales; M. H. Muma, Western New
Mexico University; N. 1. Platnick, American Museum of Natural History; G.
A. Polis, Vanderbilt University; S. E. Riechert, University of Tennessee; A. L.
Rypstra, Miami University, Ohio; M. H. Robinson, U.S. National Zoological
Park; W. A. Shear, Hampden-Sydney College; G. W. Uetz, University of
Cincinnati; C. E. Valerio, Universidad de Costa Rica.

THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in
Spring, Summer, and Fall by The American Arachnological Society at Texas
Tech Press.

Individual subscriptions, which include membership in the Society, are $30.00
for regular members, $20.00 for student members. Institutional subscriptions to
The Journal of Arachnology are $50.00. Correspondence concerning subscriptions
and memberships should be addressed to the Membership Secretary (see back
inside cover). Remittances should be made payable to The American
Arachnological Society. Inquiries about availability and current prices of back
issues should be sent to Dr. Susan E. Riechert, Department of Zoology,
University of Tennessee, Knoxville, TN 37916 USA. Correspondence concerning
undelivered issues should be addressed to the Texas Tech Press, Texas Tech
University, Lubbock, Texas 79409 USA.

Change of address notices must be sent to the Membership Secretary.

Detailed instructions for the preparation of manuscripts appear in the Fall
issue of each year, and can also be obtained from the Editor and the Associate
Editor. Manuscripts that do not follow those instructions will be returned to the
author(s) without benefit of review. Manuscripts and all related correspondence
must be sent to Dr. Jerome S. Rovner, Associate Editor, Department of
Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio 45701 USA.

PROOFS, REPRINTS, and CHARGES: Authors will receive a reprint order
form along with their proofs. Reprints are billed at the printer’s current schedule
of costs. All authors will be billed by The American Arachnological Society for
page charges. The charge per journal page will be assessed as follows: $50.00-
nonmembers; $35. 00-members acknowledging grant, institutional, or other
support; $25. 00-full members who do not have support; $20. 00-student members
who do not coauthor with full members and who do not acknowledge support.

This publication is printed on acid-free paper.

Jimenez, M.-L. 1989. Nuevas especies del genero Philodromus (Araneae, Philodromidae) de la region
del Cabo, B.C.S., Mexico. J. Arachnol, 17:257-262.

NUEVAS ESPECIES DEL GENERO PHILODROMUS
(ARANEAE, PHILODROMIDAE) DE LA
REGION DEL CABO, B.C.S., MEXICO

Maria-Luisa Jimenez

Centro de Investigaciones Biologicas
de Baja California Sur, A.C.

Division de Biologia Terrestre
Ap. Postal 128, La Paz, B.C.S. 23060 Mexico

ABSTRACT

Two new species of the genus Philodromus infuscatus group. P. jimredneri, P. pericu and the male
of P. coachellae from the Cape Region, Baja California Sur, Mexico, are described and illustrated.

RESUMEN

Se describen e ilustran dos especies nuevas del genero Philodromus, grupo infuscatus: P jimredneri
y P. pericu, ademas el macho P. coachellae de la Region del Cabo, Baja California Sur, Mexico.

INTRODUCCION

El grupo infuscatus del genero Philodromus comprende 20 especies de aranas,
la mayoria ampliamente distribuidas en las zonas aridas de Estados Unidos y
Mexico (Dondale y Redner 1969). De ellas, diez ban sido citadas para la
Republica Mexicana: Philodromus pratariae (Scheffer 1904), P. pratarioides, P
separatus Dondale y Redner 1969, P. infectus Dondale y Redner 1969, P.
pseudanomalus Dondale y Redner 1969, P. cavatus Dondale y Redner 1969, P.
mexicanus Dondale y Redner 1969 y P. albicans O. Pickard-Cambridge, 1897.
(Bonnet 1958; Dondale y Redner op. cit.).

Banks (1898) en su trabajo sobre los aracnidos de Baja California y otras
partes de Mexico, senala “no haber encontrado representantes del genero
Philodromus en la fauna de la region del Cabo”. Jimenez (1988) durante el
estudio sobre las aranas de la “Reserva de la Biosfera Sierra de la Laguna”, cita
por primera vez a P. coachellae Schick, 1964 y P. infuscatus utus Chamberlin,
1921 para esta region.

En este trabajo se describen dos especies nuevas del grupo infuscatus y se dan a
conocer los caracteres del macho de P coachellae Schick. Estas aranas habitan la
selva baja caducifolia y bosque de pino-encino de la Sierra de la Laguna y del
matorral xerofilo del Comitan, Baja California Sur.

258

THE JOURNAL OF ARACHNOLOGY

Philodromus jimredneri, especie nueva
Figs, 1-4

Dates del tipo. — Holotipo macho recolectado en Comitan, Baja California Sur,
matorral xerofilo, 17 de febrero 1987 (M. Jimenez) y los siguientes paratipos: una
hembra y un macho, 24 de febrero 1987, 17 de febrero 1987 (M. Jimenez) todos
de la misma localidad del tipo. El tipo sera depositado en la coleccion del
Laboratorio de Acarologia (Facultad de Ciencias, Universidad Nacional
Autonoma de Mexico); los paratipos sera depositados en la coleccion
aracnologica del Centro de Investigaciones Biologicas de Baja California Sur,
A.C

Macho. — Longitud total 2,72-3.25 mm, longitud del prosoma 1.12-1.25 mm y
anchura 1.07-1.32 mm, femur II 1.87-2.02 mm (dos ejemplares). Prosoma
dorsoventralmente aplanado, un poco mas elevado al nivel de la coxa III, de
color amarillo claro, moteado de cafe rojizo y bianco, con escasas sedas; area
ocular con dos bandas medias oscuras que pasan entre los ojos medios
posteriores, despues del area ocular hay una mancha en forma de “V” bordeada
de cafe en el extremo anterior; ojos posteriores en tuberculos grises. Patas
pilosas, amarillo claro con anulaciones, lateralmente mas oscuras y escasa
escopula; femur I 1.62-1.70 mm con tres macrosedas dorsales, dos prolaterales y
tres retrolaterales; tibia I 1.32-1.35 mm con dos macrosedas prolaterales, dos
retrolaterales y dos pares ventrales; basitarso I 1.12-1.15 mm con dos macrosedas
prolaterales, dos retrolaterales y dos pares ventrales; tibia III 1.12-1.15 mm con
una macroseda prolateral, una retrolateral y dos pares ventrales. Opistosoma
ovalado, truncado anteriormente, con una fisura media, de color bianco, moteado
de cafe grisaceo, con tres bandas transversas cafes en forma de “V” invertida,
costados mas oscuros y el vientre bianco. Tibia del pedipalpo con la apofisis
ventral corta y delgada, apofisis retrolateral con dos dientes, uno corto y
descansando sobre la apofisis ventral y el otro largo y puntiagudo; embolo corto
y recto; conductor curvo y delgado; cimbio sin proyeccion dorsal conica y en la
base con una prominencia angular en el lado retrolateral (Figs. 1 y 2).

Hembra. — Longitud total 3.67 mm, longitud del prosoma 1.17 mm y anchura
1.22 mm (un ejemplar). Estructura general similar a la del macho pero con
coloracidn mas clara; femur II 2.00 mm; femur I 1.65 mm con dos macrosedas
dorsales, dos prolaterales, ninguna ventral; tibia I 1.65 mm con tres macrosedas
prolaterales, tres retrolaterales y dos pares ventrales; basitarso I 1.17 mm con tres
macrosedas prolaterales, tres retrolaterales y tres pares ventrales; tibia III 1.17
mm con una macroseda prolateral, una retrolateral, dos pares ventrales. Epiginio
tan largo como ancho, con una ligera depresion y un septo medio ancho que va
disminuyendo posteriormente (Fig, 3); espermatecas contiguas alargadas y ovales
con un pequeno abultamiento en el margen anterior; conducto del organo
espermatico delgado y corto, que surge en la parte media lateral del cuerpo
principal; tubos de fertilizacion presentes (Fig. 4).

Discusion. — Philodromus jimredneri presenta caracteristicas afines a P.
pseudanomalus Dondale y Redner, pero difiere en lo siguiente: el embolo es mas
corto, el conductor es mas largo y curvo y el cuerpo principal de las espermatecas
presenta abultamientos en las margenes anteriores. Es menos parecida a P.
anomalus Gertsch, porque el lobulo de la base del embolo, es menos agudo y el

JIMENEZ— NUEVAS ESPECIES DEL GENERO PHILODROMUS

259

Figures 1-8. — Especies del genero Phiiodromus: 1-4, Philodromus jimredneri sp. nov.; I, pedipalpo,
vista ventral; 2, pedipalpo, vista lateral; 3, epiginio, vista ventral; 4, epiginio vista dorsal; 5, 6,
Philodromus pericu sp. nov.; 5, epiginio, vista ventral; 6, epiginio, vista dorsal; 7, 8, Philodromus
coachellae Schick, 1965; 7, pedipalpo, vista ventral; 8, pedipalpo, vista lateral.

260

THE JOURNAL OF ARACHNOLOGY

conducto de los organos espermaticos surge en las mitad media lateral del cuerpo
principal de las espermatecas.

Distribudon. — Conocida solo para la localidad del tipo.

Etimologia, — Esta especie esta dedicada al Sr. James Redner del Biosystematics
Research Centre, Ottawa, Canada, como reconocimiento a su gran labor dentro
del estudio de las Philodromidae y en agradecimiento por la ayuda brindada.

Philodromus pericu, nueva especie
Figs. 5, 6

Datos del tipo. — Holotipo hembra recolectado en la Sierra de la Laguna,
Canon de la Zorra, selva baja caducifolia, 840 m, 6 de marzo 1986 (F. Cota) y los
siguientes paratipos: dos hembras, 1685 m, 15 de enero 1988 (F. Cota), 840 m 17
de enero 1988 (V. Roth) todos de la misma localidad del tipo. El tipo sera
depositado en la Coleccion del Laboratorio de Acarologia, (Faculta de Ciencias,
Universidad Nacional Autonoma de Mexico); un paratipo sera depositado en el
Centro de Investigaciones Biologicas de Baja California Sur, A.C. y otro en el
Biosystematic Research Centre, Ottawa, Canada.

Hembra. — Longitud total 3.42-3.95 mm; longitud y ancho del prosoma 1.30-
1.45 mm, femur II 2.00-2.32 mm (tres ejemplares). Prosoma dorsoventralmente
aplanado, un poco mas elevado a nivel de coxa III, amarillo cafe, finamente
moteado con una banda media longitudinal blanca, margenes negros y frente con
una linea blanca que continua hacia los costados a nivel de coxa II; ojos laterales
en tuberculos grises, los lateroposteriores sobre una mancha blanca; esternon,
labio y enditos blancos con sedas hialinas. Patas claras con bandas oscuras
laterales, patelas mas oscuras y femures con macrosedas sobre manchas oscuras;
femur I 1.70-1.82 mm con dos macrosedas dorsales, dos prolaterales, dos
retrolaterales y dos distodorsales; tibia I 1.37-1.52 mm, sin macrosedas dorsales,
tres prolaterales, tres retrolaterales y dos pares ventrales; basitarso I 1.17-1.45 mm
con tres macrosedas prolaterales, tres retrolaterales y tres pares ventrales; tibia III
1.12-1.17 mm con dos macrosedas prolaterales, dos retrolaterales y dos pares
ventrales. Opistosoma de forma ovalada y larga, region cardiaca con una banda
media dorsal que se ensancha y se interrumpe formando tres llneas transversas en
forma de “V” invertida, costados cafe oscuro salpicados de bianco con una banda
blanca que se interrumpe a la mitad del opistosoma; vientre bianco. Epiginio con
un septo medio amplio truncado posteriormente; atrio poco profundo con
margenes laterales redondos (Fig. 5), cuerpo principal de la espermateca alargado
y sin lobulaciones y con abultamientos en su margen anterior; conducto del
organo espermatico muy largo y delgado que se dobla hacia la parte anterior
(Fig. 6).

Discusion. — Philodromus pericu es similar a P pratairoides Dondale y Redner,
pero difiere en que presenta las patas con bandas laterales y la coloracion del
cuerpo es mas oscura; el septo medio del epiginio es truncado posteriormente, el
conducto del organo espermatico es mas delgado y el cuerpo principal de las
espermatecas no es lobulado y presenta abultamientos anteriores.

Distribucion. — Conocida solo para la localidad del tipo.

Etimologia. — El nombre es derivado de la tribu “pericu”, indios nativos de
Baja California Sur.

JIMENEZ— NUEVAS ESPECIES DEL GENERO PHILODROMUS

261

Philodromus coachellae Schick, 1965
Figs. 7, 8

Philodromus coachellae Schick 1965. Dondaie y Redner 1969 Canadian Entomol. 101:933, figs. 58 y

59 hembra.

Datos del macho. — Nueve machos colectados en varias altitudes: 840 m, 4
octubre de 1986, 1 noviembre de 1986, 786 m, 29 octubre de 1987; 420 m, 1
noviembre de 1987 (M. Jimenez y F. Cota cols.) Los especimenes seran
depositados en la Coleccion aracnologica del Laboratorio de Acarologia
(Facultad de Ciencias, Universidad Nacional aut6noma de Mexico) asi como
representantes en el Centro de Investigacines Biologicas de Baja California Sur,
A.C., en el Biosystematics Research Centre, Ottawa, Canada y en la coleccion
particular de la autora.

Macho. — Longitud total 2.50-3.80 mm, longitud del prosoma 1.20-1.62 mm y
anchura L22-1.57 mm (9 ejemplares), femur II 2.32-3.02 mm. Prosoma amarillo
con una mancha clara en forma de “V” en frente del surco dorsal; areas laterales
moteadas y con dos manchas blancas a nivel de coxas I y II; con escasas sedas
cortas y fuertes; margenes oscuros; ojos encerrados en circulos blancos sobre
tuberculos grises; labio y enditos amarilios y esternon bianco salpicado con
pequenas manchas oscuras en sus margenes. Patas largas, delgadas y moteadas,
femures mas claros y tarsos oscuros, con escasa escopula; femur I 1.80-2.57 mm,
con tres macrosedas dorsales, tres prolatrales, tres retrolaterales; tibia I 1.62-2.10
mm, con una macroseda dorsal, tres prolaterales, tres retrolaterales, dos pares
ventrales; basitarso I 1.37-1.82 mm, con tres macrosedas prolaterales, tres
retrolaterales y tres pares ventrales; tibia III 1.17-1.50 mm con una macroseda
dorsal, dos prolaterales, tres retrolaterales y dos pares ventrales. Opistosoma gris,
dos veces mas largo que ancho sin sedas, truncado en el frente, con una muesca
media anterior; dorso con una mancha clara en la region cardiaca, finamente
moteado de cafe rojizo, con una banda ancha transversal posterior blanca;
costados blancos con lineas oscuras inclinadas, vientre claro. Tibia del pedipalpo
mas iarga que ancha; apofisis ventral delgada y poco esclerosada y mas larga que
la apofisis retrolateral, esta ultima presenta dos dientes, el mas largo descansa
sobre la apofisis ventral y el corto presenta una cuspide bifida; embolo corto y
abultado en su base con una punta delgada y en ocasiones engrosada, que se
aloja lejos del conductor que es delgado pequeno y con el extremo curvo; cimbio
sin saliente angular en el lado retrolateral y sin proyeccion conica dorsal en la
base (Figs. 7-8).

Discuslon. — Los machos de Philodromus coachellae Schick se parecen a los de
P. mexicanus Dondaie y Redner, en que ambos tiene el cuerpo alargado y
delgado, el conductor del pedipalpo es delgado y largo, pero difieren en que la
primera especie presenta el embolo mas delgado y con la punta recta, la apofisis
retrolateral bicuspide y la formula de las macrosedas de las patas es diferente. El
macho descrito como P. coachellae ha sido colectado con las hembras de esta
especie en varias ocasiones.

AGRADECIMIENTOS

La autora agradece al Dr. C. D. Dondaie y al Sr. James H. Redner
(Biosystematics Research Centre, Ottawa, Canada) por su ayuda en la

262

THE JOURNAL OF ARACHNOLOGY

reconfirmacion de las especies descritas, por sus valiosos comentarios y
sugerencias al manuscrito, al Dr. J. Palacios-Vargas (Laboratorio de Acarologia,
Facultad de Ciencias, UNAM), asi como por la revision del mismo. A los
tecnicos Sres: F. Cota y A. Cota por la ayuda en la colecta de los ejemplares
estudiados.

Este trabajo fue realizado con el apoyo economico del Centro de
Investigaciones Biologicas de Baja California Sur, A.C., el Consejo Nacional de
Ciencia y Tecnologia (CONACyT) y la Secretaria de Programacion y
Presupuesto (SPP).

LITERATURA CITADA

Banks, N. 1898. Arachnida from Baja California, and other parts of Mexico. Proc. California Acad.
Sci., (ser. 3) 1:205-308.

Bonnet, P. 1958. Bibliographia Araneorum, Toulouse 2(4):3027~4230.

Dondale, C. D. and J. H. Redner. 1969. The infuscatus and dispar groups of the spider genus
Philodromus in North and Central America and the West Indies (Araneida: Thomisidae) Canadian
Entomol., 101:921-954.

Jimenez, M. 1988. Nuevos registros de arahas en Baja California Sur, Mexico, Fol. Entomol. Mexico,
74:197-204.

Manuscript received July 1988, revised November 1988.

Perez-Miles, F. Variacion relativa de caracteres somaticos y genitales en Grammostola moUicoma
(Araneae, Theraphosidae). J. Arachnoi., 17:263-274.

VARIACION RELATIVA DE CARACTERES SOMATICOS Y
GENITALES EN GRAMMOSTOLA MOLLICOMA
(ARANEAE, THERAPHOSIDAE)

Fernando Perez- Miles

Institute de Investigaciones Biologicas Clemente Estable
Division Zoologia Experimental
Av. Italia 3318, Montevideo Uruguay

ABSTRACT

A quantitative analysis of somatic and genital characters on a sample of Grammostola moUicoma
(Ausserer, 1875) showed higher intraspecific variation in genitalic than somatic characters, especially
in females. Experimental copulation within a subsample confirmed that all were the same species.
These results seems to be uncommon in spiders and may result from: i) female moulting after maturity
eedysis; ii) allometric growth of spermathecae; iii) multiple sexual selection by female choice
(according to female size) and its effects on male genitalic evolution.

RESUMEN

Un anMisis cuantitativo de caracteres somaticos y genitales realizado en una muestra de Grammostola
moUicoma (Ausserer, 1875) mostro una variacion intraespecifica mayor en caracteres genitales que en
los somaticos, especialmente en hembras. Copulas experimentales realizadas en una submuestra
confirmaron la coespecificidad. Estos resultados parecen ser poco comunes en aranas y podrian
deberse a: i) que las hembras continuan mudando despues de adultas; ii) crecimiento alometrico de las
espermatecas; iii) seleccion sexual multiple por eleccion de la hembra (condicionada por el tamano de
la hembra) y sus efectos sobre la evolucion de la genitalia de los machos.

INTRODUCCION

El geoero Grammostola Simon 1892, incluye especies de gran tamano y de

distribucion exclusivamente sudamericana. Este genero ha sido utilizado repetidas
veces para estudiar el valor de caracteres sistematicos de Mygalomorphae
(Bucherl 1951, 1957; Schiapelli y G. de Pikelin 1960, 1962a, b; G. de Pikelin y
Schiapelli 1969).

En los ultimos 40 anos han predominado dos tendencias controvertidas en la
utilizacion de caracteres considerados especificos en Grammostola: (i) el uso de
caracteres morfometricos somaticos (Bucherl 1951) y (ii) el uso de caracteres
morfologicos genitales (Schiapelli y G. de Pikelin 1960).

El argumento mas utilizado en la literatura para dilucidar el valor espedfico de
los caracteres en Grammostola ha sido su variabilidad. Sin embargo, los estudios
de Bucherl (1951) y Schiapelli y G. de Pikelin (1960, 1962), se han realizado en
individuos cuya coespecificidad es dudosa o en taxones de estabilidad discutida.
Otros inconvenientes detectados en trabajos previos es que no presentan

264

THE JOURNAL OF ARACHNOLOGY

resultados cuantitativos de variacion ni vinculan sus resultados de caracteres
genitales con las hipotesis clasicas de aislamiento reproductor.

La variacion relativa de caracteres genitales y somaticos ha sido objeto de
estudios recientes en Araneae (e.g., Vollrath 1980; Coyle 1985; Blasco-Feliu 1986)
surgiendo interpretaciones explicativas novedosas (Eberhard 1983, 1985). Tales
aportes dan elementos adicionales para la interpretacion de la variacion de
caracteres en Grammostola que no han podido tenerse en cuenta en el pasado.

En este trabajo se estudia la variabilidad de dos grupos de caracteres (genitales
y somaticos) en una muestra de G. molUcoma de Uruguay y Sur de Brasil. Diez
copulas experimentales en una submuestra confirmaron la coespecificidad y
permitieron el estudio detallado de la variacion de genitales de los individuos que
copularon. Los objetivos de este estudio son: i) confirmar el status especifico de
la muestra; ii) analizar la variabilidad intraespecifica de caracteres genitales y
somaticos en G. mollicoma; iii) analizar el grado de asociacion entre caracteres
mediante tecnicas de agrupamiento y iv) agregar elementos para evaluar las
hipotesis de seleccion sexual.

Los resultados de este estudio permitiran establecer una base mas solida y
objetiva de discusion, con utilidad potencial en la sistematica de este genero y en
Theraphosidae.

MATERIAL Y METODOS

Cuarenta y dos individuos (24 machos y 18 hembras) de G. mollicoma
(Ausserer, 1875), procedentes de Uruguay y Sur de Brasil (al sur del paralelo 28
S) fueron examinados y depositados en las colecciones del Museo Nacional de
Historia Natural, Montevideo, de Instituto Butantan (IB) Sao Paulo. Los
ejemplares fueron determinados por comparacion con el material tipico (holotipo
macho y paratipo hembra, depositados en el British Museum of Natural History
[BMNH], Londres).

Una submuestra de 4 machos y 5 hembras fueron utilizados en experiencias de
comportamiento sexual. Las observaciones se realizaron en recipientes de vidrio
de 39x16x20 cm, con una capa de arena en el fondo, registrando el relato por
grabacion magnetofonica. Durante las experiencias la temperatura del laboratorio

vario entre 23 y 25° C.

Caracteres: Fueron registrados nueve caracteres somaticos: (1) longitud de pata
I, (2) longitud de pata II, (3) longitud de pata III, (4) longitud de pata IV, (5)
longitud de cefalotorax, (6) ancho maximo de cefalotorax; indices entre
longitudes de : (7) pata I/pata IV, (8) cefalotorax/ pata I, (9) cefalotorax/ pata IV.

Los caracteres genitales registrados en machos (Fig. 1) fueron: (10) longitud del
bulbo, (11) ancho del bulbo a !4 de su longitud (comenzando desde su extremo
basal), (12) ancho del bulbo a V2 de su longitud, (13) ancho del bulbo a % de su
longitud (comenzando del extremo basal). Todos los anchos fueron tornados
perpendicularmente a la longitud. Indices entre: (14) caracter 11/caracter 10, (15)
caracter 12/ caracter 10, (16) caracter 13/ caracter 10.

Los caracteres genitales registrados en hembras (Fig. 2) fueron: (10) longitud de
espermatecas, (11) ancho de la base de espermatecas, (12) ancho maximo del
fundus, (13) cuello minimo de espermatecas; indices entre: (14) caracter 11/
caracter 10, (15) caracter 12/ caracter 10, (16) caracter 13/ caracter 10.

PEREZ-MILES— VARIACION EN GRAMMOSTOLA

265

o

1

2

Figuras 1-2. — Bulbos y espermatecas de G. mollicoma mostrando las medidas tomadas, como se
definen en el texto; 1, bulbo vista prolateral, con trama el esclerito I, en bianco esleritos I + II.; 2,
espermatecas (vista ventral), la trama indica el area glandular.

Las mediciones fueron realizadas con micrometro ocular (apreciacion 0.1 mm).
Las patas fueron medidas artejo por artejo, por su parte media dorsal; tomandose
en cuenta unicamente las partes esclerificadas. Otros caracteres fueron tornados
de dibujos de todos los ejemplares, realizados con camara clara y aumento fijo.
Las medidas corresponden a bulbos y espermatecas izquierdos. En las
espermatecas se tomo en cuenta particularmente el area glandular (De Carlo
1973), que fue dibujada (Fig. 2). Los terminos de morfologia de bulbo siguen a
Kraus (1978, 1984).

Metodos estadisticos: Los estadisticos de los caracteres somaticos fueron
estudiados por separado y luego comparados. Las medidas se compararon
mediante test de t de Student para diferencia de medias, con restricciones para las
varianzas (test de F de Snedecor). Los indices fueron comparados a partir del
analisis de regresion entre los datos originales, mediante comparacion de
pendientes e interceptos de acuerdo a Phillips (1983). La variabilidad de
caracteres fue comparada por el metodo de Lewontin (1966), y graficada en
perfiles de variabilidad segun Sokal y Braumann (1980).

Los caracteres no homologables (somaticos con genitales y genitales entre
sexos) fueron comparados todos contra todos por el test de Lewontin (1966).

Se realizaron analisis de agrupamiento para caracteres, por separado para
machos y hembras, Los fenogramas fueron elaborados a partir de una matriz de
correlacion multiple entre caracteres (estandarizados por transformacion
logaritmica neperiana) y utilizando ligamiento promedio (UPGMA). Se tomaron
los caracteres como OTUs (tecnica R), lo que permite agrupar caracteres
jerarquicamente sobre la base de su correlacion, e inferir la informacion que
brindan. Estos procedimientos siguen en lineas generales a Sneath y Sokal (1973)
y Crisci y Armengol (1983). El programa de computos multivariantes utilizado
fue el paquete PRESTA, desarrollado en el Centro Ramon y Cajal, Espana.

CARACTERES SOMATICOS

Las medidas (caracteres 1 a 6) presentaron diferencias significativas (p < 0.05)
entre machos y hembras (Tabla 1).

266

THE JOURNAL OF ARACHNOLOGY

Tabla 1. — Estadisticos de los caracteres somaticos y genitales en Grammostola mollicoma: Car. =
numero de caracter; X = promedio; DT = desvio tipico; CV = coeficiente de variabilidad; / = valor de /
de Student caiculado (a = para interceptos; b = para pendientes).

Car.

Machos {N =

24)

Hembras {N -

18)

/

P

A

DT

CV

X

DT

CV

1

62.59

4.76

7.61

57.16

5.76

10.06

3.25

<0.01

2

60.78

3.75

6.18

53.04

5.29

9.98

4.03

<0.01

3

57.10

4.64

8.13

50.68

4.86

9.60

4.32

<0.01

4

68.96

5.34

1J5

62.02

5.89

9.50

3.93

<0.01

5

19.67

1.71

8.71

22.37

2.66

11.88

3.76

<0.01

6

18.14

1.76

9.72

20.17

2.47

12.24

2.97

<0.01

7

0.91

0.05

5.02

0.92

0.02

2.65

14.89‘‘

<0.001

14.51"

<0.001

8

0.31

0.01

4.25

0.39

0.01

3.62

4.65"

<0.001

0.19"

0.90-0.80

9

0.29

0.02

7.82

0.36

0.02

5.12

5.14"

<0.001

1.12"

0.30-0.20

10

5.20

0.11

10.42

1.38

0.28

20.11

11

1.52

0.14

9.38

0.80

0.25

3L02

12

0.85

0.11

13.37

0.46

0.11

24.58

13

0.23

0.04

17.08

0.27

0.39

14.52

14

0.29

0.02

7.82

0.60

0.13

21.71

IS

0.17

0.01

13.79

0.34

0.07

20.16

16

0.05

0.01

15.65

0.20

0.04

18.18

Los anMisis de regresion entre los datos originaies utilizados para los indices
rechazaron la hipotesis nula b = 0 (p < 0.05). Los indices (caracteres 7 a 9)
mostraron diferencias estadisticamente significativas entre machos y hembras en
la comparacion de interceptos (p <0.001). La comparacion de pendientes (Tabla
1) evidencio diferencias significativas entre sexos en el caracter 1 (p < 0.001) pero
no en los caracteres 8 v 9 (i? > 0.20).

Los caracteres 1 a 6 presentaron mayores CVs en las hembras que en los
machos (Fig. 3, Tabla 1). Los caracteres 7 a 9 (indices) presentaron CVs menores
en las hembras que en los machos (Fig. 3, Tabla 1). A1 comparar la variabilidad
de caracteres somMicos entre sexos, usaedo el test de Lewontin (1966), las
diferencias no resultaron ser estadisticamente significativas (p > 0.05).

CARACTERES GENITALES

L Espermatecas. — Las espermatecas de todas las hembras de (j. mollicoma
estudiadas comparten un mismo patron morfologico: presentan un apice (fundus)
globuloso a subglobuloso y un conducto mas o menos curvado, que se eesancha
gradualmente en sentido basal. Dentro de este patron morfologico se observe un
amplio rango de variacion continua intraespecifica (Fig. 6), Dicha variacion fue
evidenciada por los altos CVs de los caracteres 10 a 16 (Tabla 1, Fig. 5).

2. Bulbos. — Los bulbos de los iodividuos de G. mollicoma presentaron un
mismo patron morfologico pero coo ciertas variacioees (Fig. 7). La region
proximal (esclerito I) es fusiforme y espiralada; la region distal (escleritos II + III)
presenta forma subconica, terminada en un estilo largo y fino (Fig. 1). El surco
de la ‘‘zona fr%il” (“weak zone” de Kraus 1978) es aproximadamente

PEREZ^MILES— VARIACION EN GRAMMOSTOLA 267

CV

Caracteres

Figuras 3-5, — Perfiles de variabilidad {sensu: Sokal y Braumann 1980) de los caracteres estudiados:
3, caracteres somaticos (• machos, o hembras); 4, caracteres genitaies de machos; 5, caracteres
genitales de hembras.

perpendicular al eje mayor del bulbo. La region basal del estilo presenta una aleta
coiispicua.

3. Comparaddn de bulbos y espermatecas.— La variacion ' de caracteres
genitales de machos y hembras fue comparada (todos contra todos) utilizando el
metbdo de Lewoetie (1966). De 49 comparaciones realizadas , 31 demostraron

6

Figura 6. — Espermatecas de hembras de la moestra de G. moilicoma estudiada, vista ventral

268

THE JOURNAL OF ARACHNOLOGY

7

Figura 7. — Bulbos izquierdos de machos de la muestra de G. mollicoma estudiada, vista prolateral.

una variabilidad significativamente mayor {p < 0.05) ee ios caracteres genitales de

hembras que de machos y 18 no demostraron diferencias significativas {p > 0.05).
Las comparaciones que no demostraron diferencias significativas de variabilidad
fueron; caracter 10 de hembras con 12, 13, 15 y 16 de machos; 13 de hembras con
10, 12, 13, 15 y 16 de machos; 14 de hembras con 13 de machos; 15 de hembras
con 12, 13, 15 y 16 de machos; 16 de hembras con 12, 13, 15 y 16 de machos.
Coincidentemente estas comparaciones implican a los caracteres genitales mas
variables en los machos (12, 13, 15, 16) y menos variables en las hembras (10, 13,
14, 15 y 16)(Figs. 4y 5).

4, Estudio experimental. — Se obtuvieron 10 copulas experimentales entre 4
machos y 5 hembras de G. mollicoma (Fig. 8). Las uniones de individuos
mediante copulas sustento con mayor grado de certeza su coespecificidad (aunque
no pudo determinarse si ocurrio inseminacidn y fecundacion). El estudio
detallado de las espermatecas de las hembras que copularon permitio confirmar la
extrema variabilidad observada en el material muerto. Estas experiencias sugieren
que tales variaciones no corresponden a diferencias interespecificas, ya que son
observables aun entre hembras que copularon con el mismo macho. Los bulbos
de los machos que copularon presentaron una variabilidad similar a lo observado
en el material muerto.

PEREZ-MILES— VARIACION EN GRAMMOSTOLA

269

Figura 8. — Esquema de morfologia de bulbos y espermatecas de una submuestra de G. mollicoma.
Las flechas unen los individuos que copularon entre si. La trama indica el area glandular.

COMPARACION DE CARACTERES SOMATICOS Y GENITALES

1. Comparacion de variaciones. — La variacion de todos los caracteres genitales
fue comparada con la variacion de todos los caracteres somaticos en hembras y
machos.

En hembras; de 63 comparaciones realizadas, 57 demostraron una
variabilidad significativamente mayor {p < 0.05) en caracteres genitales que en los
caracteres somaticos y 6 comparaciones no demostraron diferencias significativas
{p > 0.05). No demostraron diferencias significativas las comparaciones del
caracter genital 13 (cuello minimo de espermateca) con los caracteres somaticos 1
a 6. Coincidentemente, el caracter 13 presento el menor valor de CV dentro de
los caracteres genitales de las hembras (Fig. 5, Tabla 1).

En los machos; de 63 comparaciones realizadas, 41 demostraron una
variabilidad significativamente mayor {p < 0.05) de los caracteres genitales que
los somaticos y 22 comparaciones no demostraron diferencias significativas de
variabilidad. No demostraron diferencias significativas las comparaciones de los

270

THE JOURNAL OF ARACHNOLOGY

0

u

1

9

Figura 9. — Fenograma de caracteres en hembras de G. mollicoma, tecnica R, 1 6 caracteres, « = 17,
coeficiente de correlacion, UPGMA.

caracteres: 10 con 1 a 6 y 9; 11 con 1 a 6 y 9; 14 con 1 a 6 y 9; 15 con 6. Estas
comparaciones incluyen a los caracteres genitales con menores CVs (10,11,15) y a
los caracteres sornaticos con mayores CVs (1 a 6 y 9) (Figs. 3 y 4, Tabla 1).

2. Analisis de Agrupamiento (AAg). — En las hembras (Fig. 9) todas las
medidas somaticas se encueetran altamente correlacionadas (r > 0.924) y a un
nivel menor (r = 0.672) con el caracter genital 10 (longitud de espermatecas). A
este grupo se incorpora el formado por el nucleo 8-9 (r = 0.794) mas el caracter 7
(r = 0.506), indices sornaticos que se unen con las medidas a nivel r = 0.394.

Se reconocen dos grupos mas, formados por caracteres genitales y sus indices;
el de los caracteres 12, 15, 13 que se une al cluster anterior a nivel r = 0.156 y el
grupo de los caracteres 11, 14, 16 que se incorpora a nivel r = 0.011. Las
correlaciones del caracter 10 con las medidas del cuerpo resultaron significativas
al test Spearman {p < 0.05).

En los machos (Fig. 10) las medidas de patas (1-4) se unen a un nivel de

correlacidn alto (r > 0.729). Este grupo se une a nivel r = 0.490 con el nucleo
previo entre los caracteres genitales 10 y 11 (r = 0.829) que corresponden al largo
y ancho basal del bulbo. Las correlaciones del caracter 10 con los caracteres 1, 5
y 6 (medidas) resultaron significativas al test Spearman {p < 0.05). Los caracteres
5 y 6 (largo y ancho de cefalotorax) se unen a un nivel alto de correlacidn (r =
0.837). Dicho nucleo se une a niveles mas bajos con el caracter 8 (r = 0.860) y a
nivel r - 0.445 con el nucleo 7-9 previamente unido (r = 0.860). Este grupo
(caracteres 5-9) se une con el anterior a nivel r = 0.104.

Se reconoce un tercer grupo de caracteres (13, 16, 12, 15, 14) unido al cluster
anterior a nivel r - 0.019.

DISCUSION

La uniformidad relativa de los caracteres genitales en algunas especies de

aranas ha sido senalada en las revisiones sistematicas de Blasco-Feliu (1986),

PEREZ-MILES— VARIACION EN GRAMMOSTOLA

271

10

Figura lO.-Fenograma de caracteres en machos de G. mollicoma, tecnica R, 1 6 caracteres, n - 24,

coeficiente de correlacion, UPGMA.

Coyle (1968, 1971, 1974, 1985), G. de Pikelin y Schiapelli (1969), Schiapelli y G.
de Pikelin (1962) y Vollrath (1980). Tales hechos han tenido consecuencias sobre
la utilidad sistematica de los caracteres genitales como especificos de especie.
Datos semejantes han sido utilizados para la reinterpretacion de las hipotesis de
seleccion sexual (Eberhard 1985). Particularmente en Grammostola, la
variabilidad relativa de los caracteres genitales ha sido causa de controversia
(Schiapelli y G. de Pikelin 1962a, b; G. de Pikelin y Schiapelli 1969; Bucherl
1951; Lucas y Bucherl 1965).

A diferencia de lo senalado por Schiapelli y G. de Pikelin (1962a, b) en (j.
mollicoma los caracteres genitales presentaron CVs mayores o iguales a los de los
caracteres somaticos, particularmente en las hembras. Las hembras de G.
mollicoma continuan mudando y creciendo despues de la muda de maduracion y
es posible explicar la mayor variacion de las dimensiones genitales de las hembras
por mecanismos de crecimiento alometrico de las espermatecas. Datos tornados
de seis mudas consecutivas de un individuo adulto de Grammostola sp.
(Schiapelli y Gerschman 1962) iedican un mayor crecimiento relative de las
espermatecas en relacion al cefalotorax y patas I y IV. El crecimiento total de los
caracteres somaticos en esas seis mudas fue: cefalotorax 27.08%, pata I 19.01% ,
pata IV 31.75%; mientras que los caracteres genitales (tornados de los dibujos de
dichas autoras) se incrementaron asi: 10: 40.69%, 11:66.57%, 12:24.89%,
13:19.88%. Al igual que en los resultados de la muestra estudiada, durante el
crecimiento el caracter 13 (cuello minimo de la espermateca) es el menos variable
dentro de los genitales. La variacion de los caracteres genitales en los machos es
tambien mayor o igual a la de los somaticos. La variacion de caracteres genitales
fue mayor en las hembras que en los machos, lo que esta de acuerdo con el
estudio experimental. La variacion de caracteres en machos probablemente este
determinada evolutivamente por la variacidn de genitalia en las hembras. No
obstante no hay evidencia a favor de un mecanismo de aislamiento reproductor
tipo llave-cerradura, dada la relativa simplicidad de estructuras. Los resultados
podrian explicarse por las hipotesis de selecci6n sexual por eleccion de la hembra,

272

THE JOURNAL OF ARACHNOLOGY

como ha sido postulado por Eberhard (1983, 1985). Las copulas observadas
sugieren que un macho es capaz de copular con hembras que presentan
espermatecas de morfologia variable, lo que podria contraponerse a la hipotesis
de seleccion sexual por eleccion de la hembra. No obstante, no hay pruebas
definitivas sobre el exito de tales copulas (no se determine si ocurrio
inseminacion y fertilizacion) por lo que la hipotesis de eleccion de la hembra
mantiene su validez para explicar los resultados. Por otra parte, no puede
descartarse una disminucion en la selectividad de las hembras debido a las
condiciones experimentales.

La seleccion sexual por eleccion de la hembra estaria condicionada en G.

mollicoma por el tamano de la hembra y de las espermatecas (correlacionados en
el AAg). Para cada tamano de hembra existiria un tamano de bulbo a ser
seleccionado, correspondiente a la optima estimulacion. Machos con bulbos
pequenos pueden copular exitosamente con hembras pequenas; estas hembras
tienen una abundancia esperada mayor que las mas grandes, de mayor edad.
Machos con bulbos mayores pueden copular exitosamente con hembras mas
grandes, de abundancia esperada menor pero con mayor produccion de huevos
(Petersen 1950; Kessler 1973; Wise 1975, 1979: in Joeque 1983). Machos y
hembras de tamano intermedio tendrian posibilidades de encuentro y
descendencia intermedias. De este modo las posibilidades de descendencia entre
machos con bulbos de diferentes tamanos serian semejantes. La menor variacion
de caracteres genitales en los machos no contradice la hipotesis de eleccion de la
hembra (Eberhard 1985:143). Esta diferencia de variacion podria deberse tambien
a la ausencia de mudas posteriores a la maduracion en los machos.

Las proporciones corporales indudablemente tienen un papel importante en el
exito de la copula dado el sistema de sujecion de la hembra por las apofisis
tibiales del macho (e.g., Bucherl 1951). Machos pequenos no pueden sujetar y
elevar hembras grandes sin adoptar posiciones riesgosas y machos grandes tienen
dificultad para sujetar y elevar hembras pequenas sin hacerles perder el equilibrio
(Costa y Perez-Miles en prep.). Estos hechos sugieren una primera seleccion por
tamano que actuaria en fases tempranas del cortejo y precopula. La correlacion
encontrada entre algunos caracteres del bulbo y somaticos indicaria una posible
seleccion indirecta del tamano del bulbo a traves del tamano del macho, relativa
al tamano de la hembra. En las hembras se encontro que los caracteres somaticos
covarian mas entre si que con los genitales. Probablemente esto obedezea al
crecimiento alometrico de las espermatecas. Otras correlaciones encontradas entre
indices y algunas medidas podrian ser efectos laterales introducidos por los
indices, como lo senala Atchley et al. (1976).

AGRADECIMIENTOS

A P. D. Hillyard (BMNH) y S. Lucas (IB) por el prestamo de material
utilizado en este estudio. Al Cr. Emilio Conforte (PLUNA) por su gestion para la
obtencion del material de estudio. A A. Bentos-Pereira por su colaboracion en los
dibujos. A R. M. Capocasale, F. G. Costa y E. Gudynas por sus valiosos
comentarios y sugerencias. A W. E. Eberhard y J. Coddington por la revision
critica del manuscrito.

PEREZ-MILES— VARIACION EN GRAMMOSTOLA

273

LITERATURA CITADA

Atchley, W. R., Gaskins, C. T y D. Anderson. 1976. Statistical properties of ratios. I. Empirical
results. Syst. Zool, 25:137-148.

Ausserer, A. 1875. Zwiter Beitrag zur Kenntnis Arachniden-Familie der territelariae Thorell
(Mygalidae autor). Verhandl. K.K. Zool. Gesell. Wien, 21:177-224.

Blasco Feliu, A. 1986. Les spermatheques des Nemesia (Araneae, Ctenizidae); valeur taxonomique de
ce caractere. Bull. Mus. natn. Hist. nat. Paris, (4e ser.,), 8:343-350.

Bucherl, W. 1951. Estodos sobre a biologia e a sistematica do genero Grammostola Simon, 1892.
Monog. Inst. Butantan, 1:1-126.

Bucherl, W. 1957. Sobre a importancia dos bulbos copuladores e das apofisis tibiais dos machos na
sistematica das aranhas carangejeiras (Orthognatha). An. Acad. Brasil. Ci., 29:337-416.

Coyle, F. A. 1968. The mygalomorph spider genus Atypoides (Araneae: Antrodiaetidae). Psyche,
75:157-194.

Coyle, F. A. 1971. Systematics and natural history of the mygalomorph spider genus Antrodiaetus and
related genera (Araneae: Antrodiaetidae). Bull. Mus. Comp. Zool., 141:269-402.

Coyle, F. A. 1974. Systematics of the trapdoor spider genus Aliatypus (Araneae: Antrodiaetidae).
Psyche, 81:431-500.

Coyle, F. A. 1985. Two year life cycle and low palpal character variance in a great smoky mountain
population of the lamp-shade spider (Araneae, Hypochilidae, Hypochilus). J. Arachnol., 13:211-
218.

Crisci, J. V. y M. F. Lopez Armengol. 1983. Introduccion a la Teoria y Practica de la Taxonomia
Numerica. OEA, Monografias, Ser. Biol. 26, 128 pp., Washington.

De Carlo, J. M. 1973. Anatomia microscopica de las espermatecas de los generos Grammostola y
Acanthoscurria (Araneae, Theraphosidae). Physis, Secc. C, 32:343-350.

Eberhard, W. G. 1983. Why are genitalia usually useful species characters? Pp. 53-59, Proc. IX Int.
Arachnol. Congress Abstracts, Panama. Smithsonian Inst. Press.

Eberhard, W. G. 1985. Sexual Selection and Animal Genitalia. Harvard Univ. Press, Cambridge,
Massachusetts.

Gerschman de Pikelin, B. S. y R. D. Schiapelli. 1969. Discusion de los caracteres validos en la
sistemMica de las aranas Therasomorphae. Bull. Mus. natn. Hist, nat., Paris, ser. 2, 41 (supp.
1):150-154.

Jocque, R. 1983. A mechanism explaining sexual size dimorphism in spiders. Biol, Jb. Dodonaea,
51:104-115.

Kessler, A. 1973. A comparative study of the production of eggs in eight Pardosa species in the field
(Araneida, Lycosidae). Tijdschrift voor Entomologie, 1 16:23-41.

Kraus, O. 1978. Liphisiius and the evolution of spider genitalia. Symp. zool. Soc. London, 42:235-254.

Kraus, O. 1984. Male spider genitalia: evolutionary changes in structure and function. Verh.
naturwiss. Ver. Hamburg, 27:373-382.

Lewontin, R. C. 1966. On the measurement of relative variability. Syst. Zool., 15:141-142.

Lucas, S. y W. Bucherl. 1965. Importancia dos orgaos sexuais na sistematica de aranhas. Mem. Inst.
Butantan, 32:89-94.

Petersen, B. 1950. The relation between size of mother and number of eggs and young in some spiders
and its significance for the evolution of size. Experientia, 6:96-98.

Phillips, R. B. 1983. Shape characters in numerical taxonomy and problems with ratios. Taxon,
32:535-544.

Schiapelli, R. D. y B. S. Gerschman de Pikelin. 1960. Las especies del genero Grammostola Simon
1892, en la Republica Argentina (Araneae, Theraphosidae). Actas Trab. I. Congr. Sudam., Zool.,
La Plata, 3:199-208.

Schiapelli R. D. y B. S. Gerschman de Pikelin. 1962a. Estudio de seis mudas de un ejempiar hembra
del genero Grammostola Simon, 1892 (Araneae, Theraphosidae). Physis, 23:17-19.

Schiapelli, R. D. y B. S. Gerschman de Pikelin. 1962b. Importancia de las espermatecas en la
sistematica de las aranas del suborden Mygalomorphae (Araneae). Physis, 23:69-75.

Simon, E. 1891. Etudes arachnologiques 23e. memoire XXXVIII. Descriptions d’especes et de genres
nouveaux de la famille des Aviculariidae. Ann. Soc. Ent. France, 60:300-312.

Sneath, P. H. A. y R. R. SokaL 1973. Numerica! Taxonomy. The Principles and Practice of
Numerical Classification. Freeman, San Francisco.

274

THE JOURNAL OF ARACHNOLOGY

Sokal, R. R. y C. A. Braumann. 1980. Significance test for coefficients of variation and variability
profiles. Syst. ZooL, 29:50-66.

Vollrath, F. 1980. Male body size and fitness in the web-building spider Nephila clavipes. Z.
Tierpsychol., 53:61-78.

Manuscript received February 1988, revised December 1988.

Willey, M. B. and P. H. Adler. 1989. Biology of Peucetia viridam (Araneae, Oxyopidae) in South
Carolina, with special reference to predation and maternal care. J. ArachnoL, 17:275-284.

BIOLOGY OF PEUCETIA VIRIDANS
(ARANEAE, OXYOPIDAE) IN SOUTH CAROLINA,
WITH SPECIAL REFERENCE
TO PREDATION AND MATERNAL CARE

Marianne B. Willey'

Department of Biological Sciences
Clemson University
Clemson, South Carolina 29634 USA

and

Peter H. Adler

Department of Entomology
Clemson University

Clemson, South Carolina 29634 USA

ABSTRACT

We studied the biology of the green lynx spider, Peucetia viridans, in South Carolina, with
emphasis on the selection factors that maintain extended maternal care. Nocturnal predation on the
clutch was greater than diurnal predation. Chiracanthium inclusum (Araneae, Clubionidae) was the
primary predator of eggs and emerged spiderlings. After producing a fertile egg sac in the field, female
P. viridans produced at least three fertile sacs in the laboratory. In the field, females produced a
second egg sac if their first sac was lost early in the season. In contrast with other studies, spiderlings
were unable to exit from about 74% of the egg sacs without the aid of the mother.

INTRODUCTION

Maternal care in spiders ranges from short-term guarding of egg sacs in solitary
species to long-term communal care of young in social species. The most common
form of maternal care is a brief guarding period after the spiderlings have
emerged from the egg sac (Brach 1976). In most families, this gregariousness is
short-lived and usually disappears after the first post-eclosion molt. At this time,
maternal behavior often disappears, and conflicts between mother and young can
result in spiderlings being eaten (Foelix 1982).

The green lynx spider, Peucetia viridans (Hentz), mates only once and guards
its egg sac and young until spiderling dispersal. After mating, the female feeds for
several weeks, and oviposits 25-600 eggs into an egg sac which she anchors to the
vegetation with silk (Whitcomb et ai. 1966). Spiderlings emerge from the egg sac
as second instars and remain on or around the sacs until dispersal by ballooning.

’Present address: Department of Entomology, Clemson University, Clemson, South Carolina 29634-
0365 USA.

276

THE JOURNAL OF ARACHNOLOGY

The purpose of this study was to examine the biology of P. viridans, with
emphasis on the role of predation as a selection factor favoring extended
maternal care.

MATERIALS AND METHODS

Study site. — The study site is located approximately 2.3 km SE of Clemson
University in Clemson, Pickens County, South Carolina. The site consists
primarily of grasses and forbs and is bordered on the east by Lake Hartwell and
on the west by woody vegetation.

We made observations every day from 25 September through 20 December
1987 and once per week from 21 December through 18 January 1988 for
approximately 2-4 h, starting at 1230 EST. In addition, nocturnal observations
were made from 28 September through 13 October 1987 for 2-4 h, beginning at
2230 EST. By the latter date, predation ceased because of low nighttime
temperatures. The study area was divided into four sections, and a different
section was selected sequentially as the starting location each day. Additional
field observations were made once per day from 4 September through 6
December 1988.

From 7 October 1987 through 20 May 1988 and from 9 September through 9
December 1988, we conducted laboratory studies in a rearing chamber
maintained at 25 ± 2°C and 65 ± 4% relative humidity, with a 14L:10D
photoperiod. In experiments lasting longer than 48 h, we fed each green lynx
female one muscid fly three times per week.

Natural history observations. — Laboratory studies: To determine whether P.
viridans females could produce more than one egg sac, we removed 46 females
from their field sacs in 1987 and housed them individually in plastic containers
(6.1 cm deep x 4.0 cm diameter) in the laboratory. We also removed females from
sacs produced in the laboratory within one day after spiderlings emerged. For
each egg sac, we recorded the date of construction, date of spiderling emergence,
number of unhatched eggs, and number of spiderlings. When females died, we
counted mature (chorionated) eggs in the ovaries. We recorded how many days
spiderlings required to emerge from successive sacs and how many days females
required for subsequent oviposition (empty sacs disregarded). To determine
whether spiderlings could emerge from egg sacs without the aid of the mother, we
removed 43 egg sacs from guarding females in September 1988 and housed the
sacs in the laboratory. Sacs were not removed until 24 h after construction to
allow time for the silk to harden. When females constructed second egg sacs in
the field, we also removed these sacs 24 h after construction. If no spiderlings
emerged from egg sacs within 65 days after construction, we opened the sacs to
determine the fate of the clutch.

Field studies: In the field, we determined the percentage of guarding females
occupying each species of plant {N - 82) and the height of the egg sacs above the
ground (A = 81). Some females pulled leaves around the sacs, and tied the leaves
together with silk, which formed shelters that hid the sacs from view. Using two
treatments, sheltered sacs {N ~ 19) and unsheltered sacs {N ~ 11); we tested the
null hypothesis that predation on egg sacs was independent of the presence of a
shelter in Rubus argutus Link (blackberry), the primary plant used for egg-sac
anchorage.

WILLEY & ADLER— BIOLOGY OF PEUCETIA VIRIDANS

111

Prey items of females and spiderlings were recorded. Feeding rates of females
and spiderlings were calculated by dividing the number of feeding observations on
each group (gravid females, females after egg-sac loss, females guarding
spiderlings, females guarding egg sacs, and unguarded spiderlings) by the total
number of observation periods for each group.

Egg sacs consist of a flat disc attached to a round bowl Using the bowl as a
reference point, we determined the orientation of the egg sacs once every 24 h
until the first spiderlings emerged. An orientation change was recorded when the
bowl direction differed by one major compass subdivision (22.5°) or more from
one observation period to the next. For each female, we calculated the percentage
of observation periods during which orientation changed. These percentages were
ranked and analyzed with analysis of variance to compare changes in bowl
orientation in the absence of the female (A = 25) versus changes in orientation in
the presence of the female (A = 53).

To determine whether females could produce a second egg sac in the field, we
located 43 gravid females in September 1988 and tracked them until construction
of the first sac, removed the first sac, and noted whether additional sacs were
produced. To aid in identification of females, each female was given a unique
mark on the abdomen or legs with orange, water-proof enamel.

Predation experiments, — Field studies: In the field, we randomly assigned each
spider to one of three treatments. Each plant occupied by a female was marked
with a numbered ribbon, and the coloration and markings of each female were
recorded. In treatment 1, we removed females from their egg sacs on 25
September (N = 26). In treatment 2 (A = 28), we removed females from
spiderlings after approximately 90% of the young had emerged from the egg sac.
Treatment 3 served as a control (N = 29), with females left to guard both egg sacs
and spiderlings. Mortality was only attributed to predation if we observed a
predator consuming eggs or spiderlings or if dissection of the predator revealed
eggs or young.

Over the course of the study, many of the samples fit the criteria of more than
one treatment (e.g., disappearance of a guarding female would move the sample
from treatment 3 to treatment 1). Because of this lack of independence between
treatments, data were compared on a percentage basis. The total percentage of
predation was calculated for unguarded egg sacs, unguarded spiderlings, guarded
egg sacs, and guarded spiderlings. These percentages were calculated by dividing
the total number of predators for each of the above four groups by the total
number of observation periods for each group. When the same species of
predator appeared on the same clutch in consecutive observation periods
(observation periods began daily at 1230 EST and 2230 EST), it was counted
only once. No predators were observed after 18 November, so observation
periods beyond this point were not included in analyses.

Percentages of diurnal versus nocturnal predation were calculated for the four
groups. Percentages of diurnal predation were calculated by dividing the total
number of predators for each group by the total number of diurnal observation
periods on that group. Percentages of nocturnal predators were calculated in a
similar manner. Diurnal plus nocturnal predators do not always equal the total
number of predators because if the same species of predator appeared on the
same clutch in consecutive observation periods, it was recorded as a diurnal and a
nocturnal predator. A 24-h absence was required between diurnal observations

278

THE JOURNAL OF ARACHNOLOGY

and between nocturnal observations before counting the same species of predator
on a particular clutch twice. Egg-sac and female disappearances were also

recorded.

We deposited voucher specimens of P. viridans, predators, and prey items in
the Clemson University Arthropod Collection.

RESULTS

Natural history observations. — Laboratory studies: The fertility (number of
spiderlings plus number of unhatched eggs) of P. viridans decreased with
consecutive egg sacs (Table 1). In the laboratory, a second egg sac was
constructed by 93.5% of the females (16.3% of the sacs were empty), a third sac
by 45.6% of the females (19.0% were empty), a fourth sac by 8.7% of the females
(25.0% were empty), and a fifth sac (empty) by 2.2% of the females. Females did
not open sacs without spiderlings (N = 22), whereas, females opened sacs with
spiderlings {N = 39) except when the females died prematurely {N ~ 4).
Spiderlings in these unguarded sacs were unable to emerge. At death, females (N
- 26) contained an average of 11.7 ± 1.9 {X ± SE) mature eggs. Spiderling-
emergence times from successive egg sacs were relatively constant (21-28 days).
The amount of time required for oviposition decreased with successive sacs (Table
2).

Of the 43 egg sacs (second sacs excluded) taken into the laboratory in 1988,
27.9% of the sacs had spiderlings emerge without maternal aid. An average of
81.1 ± 22.4 spiderlings (53.9 ± 12.6% of the total number of spiderlings in these
sacs) emerged. Spiderlings made no exit holes in 27.9% of the sacs, whereas
62.8% of the sacs had one exit hole, and 9.3% of the sacs had two exit holes. Of
the sacs with single exit holes, 29.6% had spiderlings lodged in the holes; 77.8%
of spiderlings trapped in the holes failed to complete ecdysis. No spiderlings were
lodged in the exit holes of sacs with two openings. Of the 6136 spiderlings
trapped in the 43 egg sacs, 763 (12.4%) were trapped in their exuviae while trying
to molt (Table 3). After 65 days, an average of 3.0 ± 1.3 live spiderlings was
found in six of the egg sacs (range = 1-7 spiderlings/ sac), and three of these sacs
had exit holes.

Table 1. — Clutch sizes and spiderling emergence times from consecutive egg sacs constructed by
Peucetia viridans in 1987 and 1988. a = after spiderlings hatched, shed chorions were counted; sacs
were not included in the analysis when spiderling number was greater than the number of shed
chorions or when eggs were oviposited without the protection of an egg sac, b = sacs were constructed
in the field; all others were constructed in the laboratory, c = sacs were constructed September-
October 1988. (SE = standard error). NA = not applicable because spiderlings did not emerge from
sacs.

No. spiderlings
{X± SE)

PT

No. unhatched eggs
(X + SE)

N

Emergence times (days)
(X + SE)

N

Sac

144.7 ± 10.04

34

5.7+ 3.19

46

—

—

Sac #2

18.2+ 2.19

42

5.6+ 1.40

43

27.0 + 0.28

31

Sac #3

11.6+ 3.79

19

12.2+ 3.24

20

27.8 + 0.70

8

Sac #4

2.0 + 2.00

4

7.2+ 6.59

4

21.0

I

Sac #5

0.0

1

0.0

1

NA

0

Sac

165.3 + 8.68

43

7.7 + 10.98

43

24.8 + 1.50

12

Sac #2*’’'^

57.2 + 33.28

4

52.2 + 25.32

4

28.0

1

WILLEY & ADLER— BIOLOGY OF PEUCETIA VI RI DANS

279

Table 2. — Interval between sequential ovipositions of Peucetia viridam. a = egg sacs were
constructed in the field; all others were constructed in the laboratory, b = sacs were constructed in
1987, and females were allowed to guard until spiderlings emerged, c = sacs were constructed in 1988,
d = value is a minimum, as determined from the first day each field sac was located, (SE = standard
error).

No. days between oviposition events {X + SE)

N

Sac #P’'’-Sac #2"

73.8 ± 7.84'

9

Sac #2'’=Sac #3^

45.8 ± 2.05

16

Sac #3‘’-Sac #4"

40.0 ± 3.46

3

Sac #r’LSac #2"’"

27.2 + 2.50

4

Ooiy two of the second egg sacs {N ~ 4) from 1988 contained spiderlings (two
contained only eggs) (Table 1); 120 spiderlings (96.8% of the total) emerged from
one sac, whereas all 105 spiderlings failed to emerge from the other sac despite
the fact that this sac had three exit holes.

Field studies: Females used seventeen species of plants for oviposition (Willey
1988), with one female per plant. Rubus argutus (N = 30) and Eupaiorium
hyssopifolium L. {N - 13) predominated. The average egg-sac height was 73.8 ±
3.0 cm above ground (range = 21.5-161.5 cm), which represented the top 1/4 of
all plants. When females constructed shelters, they placed their sacs inside the
shelters and remained in direct physical contact with the sacs until spiderlings
emerged; they then moved to the outside of the shelters to guard. Predation on
the sacs was independent of the presence of R. argutus shelters (x^ = 0.151, df =
1, 0.6979).

Adults fed on prey items in five insect orders, and spiderlings fed on insects in
three orders and cannibalized conspecifics (Willey 1988). Hymenoptera (N = 32)
were the primary prey of adult females. We observed 15 females feeding in the
presence of spiderlings, but the females moved at least 3 cm away from the clutch
and did not share the prey. We observed six spiderlings feeding on small insects;
two fed simultaneously on the same ant. Gravid females had the highest feeding
rate (10.6%), followed by females after egg-sac loss (6.2%), females guarding
spiderlings (3.8%), unguarded spiderlings (1.2%), females guarding egg sacs (1.1%)
and guarded spiderlings (0.0%).

Guarded egg sacs were reoriented significantly more often than unguarded sacs
(F = 56.70, df = 1, P = 0.0001). Of the 53 guarding females, 51 reoriented their
egg sacs an average of 67.7 ± 2.8% of the time; the remaining two sacs had
spiderlings emerge on 28 September and 5 October, so reorientation may have
occurred prior to initiation of the study. Of the 25 unguarded egg sacs, eight
showed no reorientation, and the remaining 17 had orientation changes of <

Table 3. — Fate of Peucetia viridam spiderlings in unguarded egg sacs in 1988. a = calculated using
total spiderlieg number, b = calculated using number of spiderlings trapped in egg sacs, c = egg sacs
containing no spiderlings were omitted from analyses. (SE = standard error)

N

None trapped
in egg sacs
{X ± SE)

% trapped
in egg sacs
(X ± SE)*

None trapped
in exuviae
(X ± SE)

% trapped
in exuviae
(X ± SEf

Sac#l

43

142.7 ±11.24

84.9 ± 5.06

17.7 ± 4.49

12.4 ± 2.80

Sac #2

2"

54.5 ± 50.50

51.6 ±48.38

0.0 ± 0.00

0.0 ± 0.00

280

THE JOURNAL OF ARACHNOLOGY

22.5°; these changes were probably due to a gradual weakening of the silk
attachment lines from wind and rain.

The earliest egg sac was constructed on 8 September 1988; in 1987, the first egg
sac was located on 10 September. After losing their first egg sac, all 43 females
relocated to different plants. We were able to find and follow only four of these
females until they constructed second sacs. Oviposition into second egg sacs
required less time in the field than in the laboratory (Table 2).

Predation experiments. — Field studies: Predation was greater when the female
was absent (Table 4). The total percentage of predation was greatest for
unguarded egg sacs (6.5%), followed by unguarded spiderlings (1.2%), guarded
egg sacs (0.1%), and guarded spiderlings (0.0%).

Chiracanthium indusum (Hentz) was the most frequently observed predator of
unguarded egg sacs and spiderlings, and secondarily used the sac as a retreat.
This predator commonly constructed a silk retreat against the egg-sac disc,
chewed a hole through the disc, and fed on the clutch from the protection of its
retreat. Chiracanthium indusum was observed inside egg sacs where it
presumably fed on the sac contents. This predator was seen only once in the
presence of a guarding female, but was not observed feeding.

In contrast, we observed Cesonia bilineata (Hentz), a gnaphosid spider, remain
undetected within a clutch of guarded spiderlings for 15 h. Although the predator
was never observed eating spiderlings, it was surrounded by dead spiderlings and
moved within the clutch until the guarding female approached, at which time it
ceased movement until the female moved away.

The female also was unable to protect the clutch from ants when large numbers
were present. We observed a colony of Crematogaster sp. swarm over a guarded
egg sac. The female moved 6 cm from the egg sac while ants entered the sac and
carried spiderlings away. Approximately 12 h later, the female relocated her egg
sac. At this time, the egg sac contained two holes, and Crematogaster sp. were
still detectable within the sac.

Unguarded clutches attacked by orthopterans and hymenopterans (with the
exception of Vespula sp.) were completely destroyed. Orthopterans consumed
entire clutches plus the sac in less than 30 min; ants carried away whole clutches
overnight. In contrast, predators such as C indusum, Cesonia bilineata^ and
larvae of Chauliognathus pennsylvanicus (De Geer) (Coleoptera, Cantharidae)
spent several days feeding on a clutch without consuming all of the eggs or
young. Gagrellids fed at sacs for less than 1 h and consumed from one to two
eggs.

The percentage of nocturnal predation was greater than diurnal predation in all
treatments but one (Table 5). Both diurnal and nocturnal predation were greater
in the absence of the female. We observed 17 incidents of diurnal predation and
21 incidents of nocturnal predation (Table 4). Chiracanthium indusum was the
most common diurnal and nocturnal predator.

During the study, 15 unguarded egg sacs disappeared, presumably from rain,
wind, or predation, and 10 sacs disappeared simultaneously with their guarding
females, presumably because of relocation or predation. Nine egg sacs
disappeared from guarding females, although the females remained. In six
instances, guarding females disappeared from their egg sacs, presumably because
of predation; the remaining clutches eventually disappeared or suffered mortality
from predation.

WILLEY & ADLER— BIOLOGY OF PEUCETIA VI RI DANS

281

Table 4. — Predators of Peucetia viridans clutches, a = total predation does not always equal diurnal
plus nocturnal predation (see text).

Number of predators

Diurnal

Nocturnal

Totaf

UNGUARDED SACS

Opiliones, Gagrellidae, undet.

—

3

3

Araneae

Chiracanthium inclusum (Hentz)

3

6

8

Orthoptera

Campylacantha olivacea olivacea Scudder

1

—

1

Melanopius sp.

1

—

1

Oecanthus sp.

1

2

3

Coleoptera

Chauliognaihus pennsylvanicus

2

5

6

(De Geer), larval stage

Hymenoptera

Crematogaster clara Mayr

1

—

1

Crematogaster minutissima Mayr

—

1

1

Crematogaster sp.

1

1

1

Paratrechina parvula (Mayr) —

UNGUARDED SPIDERLINGS

1

1

Araneae

Chiracanthium inclusum

2

1

3

Cesonia bilineata (Hentz)

1

—

1

Orthoptera

Campylacantha olivacea olivacea

1

—

1

Hymenoptera

Paratrechina parvula

1

—

1

Vespula sp.

GUARDED SACS

1

—

1

Hymenoptera

Crematogaster sp.

1

1

1

DISCUSSION

In South Carolina, the most frequent predator of R viridans eggs was C.
inclusum, followed by Ch. pennsylvanicus larvae, Orthoptera, and gagrellids. In
northern Florida, Fink (1986, 1987) found that the two major sources of
mortality of P. viridans egg sacs were ant predation, and parasitism by Mantispa
viridis Walker (Neuroptera, Mantispidae); however, ants were the primary factor
favoring maternal care of egg sacs because the female’s presence did not
significantly reduce mantispid parasitism. We did not observe mantispid
parasitism in South Carolina and, although mantispids are common in the area,

Table 5. — Diurnal and nocturnal predation (%) on Peucetia viridans clutches (% = No. predators/
No. observation periods), a = no. of observation periods.

Diurnal predation (N)^

Nocturnal predation (N)

Unguarded sacs

3.8 (262)

13.8 (138)

Unguarded spiderlings

1.3 (470)

1.0 (98)

Guarded sacs

0.1 (865)

0.2 (577)

Guarded spiderlings

0.0 (325)

0.0 (65)

282

THE JOURNAL OF ARACHNOLOGY

Brushwein (1986), in 4 years, located only two sacs that contained mantispids. In
eastern Texas, mantispid parasitism did not constitute a significant source of
mortality of P. viridans eggs (Killebrew 1982).

In South Carolina, the most frequent predator of emerged spiderlings was C
indusum. In northern Florida, C. indusum was observed in or on five unguarded
egg sacs (10.9%), but Fink (1987) stated that C. indusum was not observed
feeding on eggs or spiderlings and might be using empty sacs only as a retreat.
Fink (1986) concluded that in northern Florida, salticids were the major
predators of emerged spiderlings; however, Fink (1986, 1987) made only diurnal
observations, so nocturnal predators were not detected.

Although predation on egg sacs was independent of the presence of R. argutus
shelters, this is the first report of R viridans females constructing foliage shelters
for their young. It is possible that the shelters protect the egg sacs from predation
by making the sacs difficult to locate. Conversely, if a predator finds the sac, it
may be protected while it forages. Pisaurina spp. and Dolomedes spp. (Araneae,
Pisauridae) also construct shelters by pulling leaves over the sacs and tying them
together with silk (Gertsch 1979).

Although females fed in the presence of spiderlings, we found no evidence that
they directly provided food for the young, as reported by Whitcomb et al. (1966).
However, spiderlings did feed on small insects that became trapped in the
female’s silk. We observed two cases of cannibalism among spiderlings in the
field, whereas Fink (1984) did not observe any aggression among the young.

In the field, P viridans females consistently reoriented their egg sacs
throughout the guarding period. This behavior may regulate temperature of the
sac contents and ensure that the eggs develop at the same rate. Randall (1977)
found that the adaptive significance of maternal care in P. viridans was directly
related to the female’s role in reorienting the egg sac; when sacs were prevented
from being reoriented, spiderlings emerged approximately four days late, which
resulted in reduced clutch sizes because of cannibalism within the sacs.

Unguarded spiderlings failed to emerge from 74% of the egg sacs, and
subsequently died within the sacs, suggesting that opening the sac might be an
important component of maternal care in P. viridans; however, low laboratory
humidity (65%) might have adversely affected the ability of spiderlings to emerge.
In contrast, Randall (1977) and Fink (1986) found that spiderlings were able to
emerge from unguarded sacs. Our 1987 results indicate that females need a cue
from inside the egg sacs in order to open them, because guarded sacs with no
spiderlings were never opened, and guarded sacs with spiderlings were always
opened. Randall (1977) also found that females received cues from within the sacs
when it was time for the spiderlings to emerge.

Laboratory females in South Carolina produced up to three fertile egg sacs
after producing a fertile sac in the field, but clutch size decreased in successive
sacs. This decline could be due to sperm depletion, decreased egg production, or
an inadequate food supply. The latter is likely because second egg sacs produced
in the field in 1988 had a clutch size approximately 2.5 times greater than second
sacs produced in the laboratory. Over half of the South Carolina females were
gravid at death, so, physiologically, it was possible to produce more eggs,
although other factors might have been limiting. Whitcomb et al. (1966) reported
that females in the laboratory in Arkansas constructed up to six egg sacs, and

WILLEY & ADLER— BIOLOGY OF PEUCETIA VIRIDANS

283

that successive sacs contained fewer eggs. Fink (1984) also reported a decrease in
egg number in second egg sacs in northern Florida.

In South Carolina, oviposition into second egg sacs in the laboratory occurred
after a minimum of 74 days and oviposition into second egg sacs in the field
occurred after an average of 27 days; the difference may reflect food availability.
Spiderlings in the laboratory required approximately 27 days within the egg sacs
prior to emergence. Therefore, unless the first egg sac was lost early in the season,
low temperatures, decreased foliage for sac anchorage, scarcity of prey items, and
female mortality would lower the probability of a second sac being successful.
Whitcomb et al. (1966) reported that in the laboratory in Arkansas (20-24° C),
females constructed second egg sacs one to two months after construction of the
first sac. These authors, therefore, hypothesized that in Arkansas, P. viridans did
not have time to produce a second sac before the first frost. Fink (1984) noted
that in north Florida, females constructing second sacs would not live long
enough to guard them successfully, and she (1986) presented a detailed discussion
of the adaptive significance of guarding the first egg sac in northern Florida.
Second egg sacs have been reported in the field in southern California (Turner
1979; Polis in Fink 1986) and southern Florida (Fink 1986). We suggest that
production of successive egg sacs at higher latitudes, such as in northern South
Carolina, is adaptive only when the original sac is lost early in the season.

In P. viridans, maternal care reduces clutch mortality from predation; however,
our observations indicate that predation pressures in South Carolina differ from
those in Florida, and that protection from nocturnal predation and possibly
aiding the young in emergence from the egg sac are important factors influencing
the retention of maternal care.

ACKNOWLEDGMENTS

We thank the following specialists for identifying specimens: D. A. Nickle
(Acrididae) and D. R. Smith (Formicidae), Systematic Entomology Laboratory,
USD A. We also thank D. A. Aborn and E. E. Robertson for technical assistance.
We are grateful to J. R. Brushwein and L. S. Fink (University of Florida), S. A.
Gauthreaux (Clemson University), and G. W. Uetz (University of Cincinnati) for
reviewing the manuscript. This study was supported, in part, by a Grant-in-Aid of
Research from Sigma Xi, The Scientific Research Society, to MBW. This is
Technical Contribution No. 2862 of the South Carolina Agricultural Experiment
Station, Clemson University.

LITERATURE CITED

Brach, V. 1976. Subsocial behavior in the funnel-web wolf spider Sosippus floridanus (Araneae:
Lycosidae). Florida EntomoL, 59:225-229.

Brushwein, J. R. 1986. Bionomics of Mantispa viridis Walker (Neuroptera: Mantispidae). Ph.D.
Dissertation, Clemson Univ., South Carolina.

Fink, L. S. 1984. Maternal guarding behavior in the green lynx spider, Peucetia viridans (Hentz).
M.S. thesis, Univ. Florida, Gainesville.

Fink, L. S. 1986. Costs and benefits of maternal behaviour in the green lynx spider (Oxyopidae,

Peucetia viridans). Anim. Behav., 34:1051-1060.

Fink, L. S. 1987. Green lynx spider egg sacs: sources of mortality and the function of female guarding
(Araneae, Oxyopidae). J. Arachnol., 15:231-239.

284

THE JOURNAL OF ARACHNOLOGY

Foelix, R. F. 1982. Biology of Spiders. Harvard Ueiv. Press, Cambridge, Massachusetts.

Gertsch, W. J. 1979. American Spiders. Van Nostraod Reinhold Co., New York.

Killebrew, D. W. 1982. Mantispa in a Peucetia egg case. J. ArachnoL, 10:281-282.

Randall, J. B. 1977. New observations of maternal care exhibited by the green lynx spider, Peucetia
viridans Hentz (Araneida: Oxyopidae). Psyche, 84:286-291.

Turner, M. 1979. Diet and feeding phenology of the green lynx spider, Peucetia viridans (Araneae:
Oxyopidae). J. Arachnol., 7:149-154.

Whitcomb, W. H., M, Hite and R. Eason. 1966. Life history of the green lynx spider, Peucetia
viridans (Araneida; Oxyopidae). J. Kansas Entomol. Soc., 39:259-267.

Willey, M. B, 1988. Factors favoring maternal care in the green lynx spider, Peucetia viridans (Heetz)
(Araneae: Oxyopidae), with special reference to predation. M.S. thesis, Clemson Univ., South
Carolina.

Manuscript received January 1989, revised March 1989.

Richman, D. A. 1989. A revision of Hentzia (Araneae, Salticidae). J. Arachnol., 17:285-344.

A REVISION OF THE GENUS HENTZIA
(ARANEAE, SALTICIDAE)

David B. Richman

Department of Entomology, Plant Pathology and Weed Science
New Mexico State University
Las Cruces, New Mexico 88003 USA

ABSTRACT

The genus Hentzia belongs in the subfamily Dendryphantinae of the family Salticidae. It appears to
be closely related to '‘‘Beata” wickhami and the genus Anicius. Twenty species are recognized in this
revision, occurring from Nova Scotia and Quebec in the north to northern South America in the
south. The genus ranges along the coastal areas of Mexico on both sides of the continent north to
Arizona in the west and through central Texas to Minnesota. Six new species are described. These are
H. calypso from Jamaica, H. chekika from Florida, the Bahamas and Cuba, H. cubana from Cuba,
H. pima from Arizona, H. whitcombi from Guadeloupe, Puerto Rico and a few other Caribbean
islands, and H. zombia from Hispaniola. The genera Parahentzia Bryant 1943, and Maeviobeaia
Caporiacco 1947, are here made junior synonyms of Hentzia. Parahentzia insignita Chickering 1946, is
made a junior synonym of Anoka paraUela Peckham and Peckham 1894. Anoka peckhami Cockerell
1893, and Wala albovittata Keyserling 1885, become junior synonyms of Icius vittatus Keyserling
1885. All of these are placed in the genus Hentzia. Balmaceda peckhami Bryant 1940 is found to be
the female of H. tibialis Bryant 1940. The female paratype of Hentzia tibialis Bryant 1940, is found to
be the female of Hentzia chekika n. sp. Finally Wala noda Chamberlin 1916, is transferred from
Hentzia to Corythalia.

INTRODUCTION

The genus Hentzia is composed of somewhat elongate jumping spiders with a
primarily circum-Caribbean distribution. The genus was erected by Marx (1883),
with Epiblemum palmarum Hentz 1832 as the type species. H. mitrata was
described by Hentz as an Attus in 1846, the Peckhams erected the genus Anoka
for H. vernalis in 1893 and Cockerell described Anoka peckhami from Jamaica in
the same year. Roewer (1954) listed 14 species in the genus, including antiliana
Bryant, audax Bryant, fimbriata (F. O. Pickard-Cambridge), footei (Petrunke-
vitch), grenada (Peckham and Peckham), noda (Chamberlin), parallela (Peckham
and Peckham), peckhami (Cockerell), poenitens (Chamberlin), squamata (Petrun-
kevitch), tibialis Bryant, vernalis (Peckham and Peckham), mitrata (Hentz), and
palmarum (Hentz). Of these, "'Hentzia'' noda was found during the current study
not to belong to the genus at ail, belonging instead to the genus Corythalia, and
H. peckhami was discovered to be a junior synonym of Icius vittatus Keyserling
(now Hentzia vittata). Examination of material in the collections of the Museum
of Comparative Zoology, the American Museum of Natural History, the Florida
State Collection of Arthropods, and the United States National Museum of
Natural History, has produced a total of six undescribed species, four represented
by both sexes and two only by the female. In addition, the two species of

286

THE JOURNAL OF ARACHNOLOGY

Parahentzia were added to Hentzia, with R insignita Chickering made a junior
synonym of Hentzia parallela. This brings the total number of species to 20.

The affinities of the genus Hentzia are somewhat difficult to determine. The
genus appears to be a member of the subfamily Dendryphantinae and resembles
''Beata'' wickhami (Peckham and Peckham) and Anicius dolius Chamberlin in
genitalic structure. The former species probably does not belong to the genus
Beat a, where it was placed by Edwards (1980) or to Icius, in which it was
originally placed (Peckham and Peckham 1894). Its proper placement will have to
wait for future studies. The latter species is the type species for the genus Anicius.
Assigning a sister genus for Hentzia is problematic, but ''Beata"^ wickhami and
Anicius dolius are reasonable approximations (see section on relationships of
Hentzia for a more complete discussion). The most extreme member of the genus,
H. vernalis, was at first thought to belong to a sister genus - Anoka (it is the
type for this genus), but further analysis seems to lead to the conclusion that it
was probably derived from Hentzia footei and is thus a valid member of Hentzia.
The so-called Parahentzia, H. mandibularis, H. parallela and H. vittata, are at
another extreme for the genus.

The distribution of species in the Caribbean is probably the result of island
speciation, some active dispersal and recent accidental transport by man. The
genus is probably relatively recent in origin and the ease with which the spiders
are transported probably explains the present wide distributions of such species as
H. antillana and H. vittata more than any other factor. Other species, such as H.
audax and H. squamata are more localized and are probably endemic island
productions.

The species in this genus are commonly found on shrubs or trees, such as
willows, oaks, palmettoes, palms, and mangroves, and on commercial crops such
as citrus, soybeans and cotton (personal observation and Whitcomb et al. 1963).
Populations can be relatively dense, as they were on black mangroves near Cedar
Key (Way Key), Levy Co., Florida, in 1975 (Richman 1977, 1982). Most species
have adults present throughout the year, although this is not true in the northern
part of the ranges of H. palmarum and H. mitrata.

Egg sacs are produced during much of the year in the tropical and subtropical
parts of the distribution of the genus. In Florida egg sac records from captive
specimens ranged from March to October (Richman 1977). The eggs are laid in a
silken sac, within a resting sac of silk constructed by the female. Records for H.
palmarum indicate that the average egg clutch size for this species is around 15
(specimens from near Cedar Key, Levy County, Florida).

METHODS

In the following species accounts, most are represented by 10 measurements for
each sex. The following measurements were found to be of use: total length,
carapace length, width and height, width of ocular region anteriorly and
posteriorly, and length of ocular region from bases of anterior median eyes
(AME) to posterior edges of posterior lateral eyes (PEE), and length and width of
chelicerae. All measurements are in mm. The leg formula refers to the
comparative length of each of the four pairs of walking legs. Thus a formula of
1234 indicates that the legs are ranked exactly in the order of anterior to

RICHMAN— REVISION OF HENTZIA

287

posterior placement, whereas 2134 would indicate that the second pair was the
longest, followed by the first, third and fourth. The placement of the posterior
median eyes (PME) in relation to the anterior lateral eyes (ALE) and the PLE is
also noted in the descriptions.

Sources for specimens examined are indicated by the following: Academy of
Natural Sciences of Philadelphia (ANSP), American Museum of Natural History
(AMNH), British Museum (Natural History) (BMNH), Bruce Cutler Collection
(BC), Canadian National Collection (CNC), David B. Richman Collection
(DBR), Exline-Peck Collection (now at the California Academy of Sciences)
(EPC), Florida State Collection of Arthropods (FSCA), Museum of Comparative
Zoology (MCZ), Texas A. & M. University (TAM), United States National
Museum of Natural History (Smithsonian Institution) (USNMNH), University of
California, Berkeley (UCB), and Universidad de Costa Rica (UCR).

The maps show the ranges of the twenty species, based on locality records that
could be found on current maps. Some records were found that I could not place
on the maps. This is especially true of records from Cuba and the Bahamas. For
example, in a recent atlas of Cuba (Atlas Nacional de Cuba 1970) at least three
listings are found for “Soledad”, none of which match the locality for the
specimens labeled “Soledad, Cuba.” This locality is really part of the modern city
of Cienfuegos. Other localities are more obscure. Many of the smaller Bahama
Islands are also difficult to find on a map. I have, however, listed all locality
records in the descriptive section, whether I could locate them on a map or not.

RELATIONSHIPS OF HENTZIA

The genus Hentzia is here defined by the presence of both pencils of hair below
the posterior median eyes and spatulate hairs on the ventral retromargin of the
first patella and distal femur (Figs. 3, 4). These characteristics are most pro-
nounced in the female, especially in regard to the hair pencils. Males often have
elongated chelicerae and somewhat elongate bodies. The male palpi of most
species exhibit a two-lobed bulb and a thin, curved embolus. This characteristic is
shared with ''Beata'' wickhami. Males of Hentzia usually have simple or truncated
retromarginal cheliceral teeth, whereas females have bicuspid retromarginal teeth.

As noted earlier, the genus Hentzia appears to be most closely related to
''Beata^" wickhami and Anicius dolius. Both species lack the hair pencils and
spatulate hairs found in the females of Hentzia (Figs. 1-4) and also differ in the
males having bicuspid (or bifurcate) retromarginal teeth (Figs. 5, 10 ). The
retromarginal tooth of the male of A. dolius is barely bicuspid, however (Fig. 10).
It is doubtful that B. wickhami, which was described in the catch-all genus Icius,
really belongs in Beata. Its similarity to Hentzia in the structure of the palpus
(Figs. 6, 7) is very striking. The female epigynum (Figs. 8, 9) is also quite similar.
The male of Anicius dolius has a palpus that is different from that of Hentzia,
but still the embolus is long and thin and generally similar to at least some of the
Hentzia species (Figs. 11, 12). The female genitalia (Figs. 13, 14) also resemble
that of some Hentzia species. Both sexes of A. dolius are close to Hentzia in
general appearance. ^"Beata"^ wickhami is known from southern Florida, Cuba
and the Bahamas. Anicius dolius and an apparent undescribed species of the
same genus are from southern Mexico.

288

THE JOURNAL OF ARACHNOLOGY

Figures 1-4. — Anatomical structures of Anicius dolius and Hentzia: 1, carapace of female A. dolius,
showing lack of hair pencils; 2, ventral view of female A. dolius, showing lack of spatulate hairs on
posterior ventral edge of femur and patella I; 3, carapace of Hentzia antillana, showing hair pencils
below posterior median eyes; 4, ventral view of first leg of Hentzia antillana, showing spatulate hairs
on femur and patella.

A cladogram (Fig. 15) shows the apparent relationships of ” wickhami and

A. dolius, taken as a group, and the species groups of the genus Hentzia.
Resolutions of species relationships within a group are based on characters not
shown on the cladogram but are discussed in the text. For example, the two
species H. chekika and H. poenitans have very similar epigynal structure, with the
j“Shaped tubes much shorter than in the next closest species H. grenada. They
also lack the sperm tube loop found on the bulb of H. grenada male palpi. They
are thus placed closer to each other than to H. grenada on the cladogram.
Similarly, H. audax and H. cubana appear to be more closely related to each
other than to the other members of the palmarum species group because the male
chelicerae are nearly identical in location of teeth and both have male palpi with
a hooked tibial apophysis. The males of this species are so similar, except in size,
that for a long time I considered them to be just size varients of the same taxon.
The females are, however, quite different in their epigynal structure. On the other
hand, the relationships of the vittata group are not resolved, and many of the
relationships within the palmarum group remains uncertain. I thus thought it best
not to use more than the seven characters to separate the species groups for the
cladistic analysis.

Figures 5-9. — Beata wickhami (Peckham and Peckham): 5~7, male from Habana, Cuba; 5, left
chelicera, ventral view, showing bicuspid retromarginal tooth; 6, 7, palp; 6, ventral view; 7, retrolateral
view; 8, 9, epigynum of female from Bahamas, 8, ventral view; 9, dorsal view.

The characters used in the cladogram are: 1) presence or absence of female hair
pencils, 2) presence or absence of female spatulate hairs, 3) male retromarginal
tooth type (bifurcate or bicuspid, simple, truncate [it should be noted that H.
footei occasionally exhibits a bifurcate retromarginal tooth, but as the tooth is
usually just truncate, the latter character was used in the analysis]), 4) male bulb
shape (lobes equal, prolateral reduced), 5) female epigynum (with central cone-
like structure, without central conedike structure), 6) male promarginal teeth
(equal, not equal), 7) female epigyeal tubes (j~shaped, other). The weights and
consistency index (c.i.) are represented on the legend of the cladogram. There are
apparently four species groups in Hentzia, based on this analysis. These are: the
paimarum group, the grenada group, the vernalis group and the vittata group.
The grenada group and the paimarum group are perhaps the most closely related,
based on the structure of the male palpi and cheiicerae, but differ strongly in the
structure of the female epigynum.

It should be noted that as this is a first attempt to construct a cladogram for
Hentzia, the relationships should not be considered as proven. Because of the
uncertainty involved in this cladogram and the lack of separation between species,
there is no doubt in my mind that a better one can be, and probably will be,
produced in the future. This can be considered as only a preliminary attempt.

290

THE JOURNAL OF ARACHNOLOGY

Figures 10-14. — Anicius dolius Chamberlin: 10, left chelicera of male from Michoacan, Mexico,
ventral view; 11, 12, palp of holotype male from Guadalahara, Mexico; 11, ventral view; 12, retrolateral
view; 13, 14, epigynum of female from Michoacan, Mexico; 13, ventral view; 14, dorsal view.

KEY TO THE SPECIES OF HENTZIA

Note: Any specimen found north of southern Georgia and east of a line from Minnesota south to west
Texas is almost certainly either H. palmarum or H. mitrata. From southern Georgia through Florida
H. grenada is also possible and H. chekika may be encountered in southern Florida.

L Males ,2

Females 19

2. Tibial apophysis sinuate (Fig. 92); with 2-3 apparent retromarginal teeth
(actually only one truncate retromarginal tooth on distal 1/4) (Figs. 89,
90); basal part of embolus reduced and more firmly attached to the

RICHMAN— REVISION OF HENTZIA

291

p a 1 m a r u m

grenada vernalis

-i I-

^ - o =2

o5oi.*3Xco

• CO-===33~

fiQaC(a.<».^ooe

I * i

Q. « U

I I 1 i I ?

V 1 1 1 a t a

5

2 o

=3 f I
s.

^ 1 1

1

2

3

4

5

6
1

O-O-

0-0

1.

F hair p«^i Is

i,ypB

r

•eiqpt.

2

states

2

mtmps

C. 1 .

2.

F spatulatw hairs

r

3

2

i

i.ie

3.

h rutrolooth

u

2

3

2

I.W

4.

ri ^Ib

r

f

2

2

0.5i

5.

F

r

1

2

2

0.^

6.

h prole«th

r

!

2

2

0.50

?.

F «pigtub«s

r

1

2

1

i.m

1. F hair perils: ab^nt; pr^ent;

2. F spalufate hairs: ob^t; preseil;

3. M retrotooth: bifurCTte; simple; truncate;

4. M ^ib: lobes equal; prolate^ai reduced;

5. F ^igyrtua: with cone; without cone;

5. H proleeth: equal; not e^ial;

?. F i^iglubes: j“-»haped; otfw;

Figure 15. — Cladogram of the genus Hentzia, based on seven morphological characters. Only
species groups are separated by the characters used on the cladogram. Species separations are based
on characters defined in the species group descriptions in the text. The four species groups defined are
1) palmarum, 2) grenada, 3) vernalis, and 4) viitata. See text section on the relationships of Hentzia
for discussion of characters used.

292

THE JOURNAL OF ARACHNOLOGY

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

tegulum (Fig. 91) (northern South America, St. Vincent, Trinidad and

Barbados) vernalis

Tibial apophysis curved, truncate or straight, never sinuate; always with one
retromarginal tooth; embolus not reduced 3

Retromarginal tooth straight or curved and truncate or rarely bifurcate

(Figs. 96, 97, 104, 109) .4

Retromarginal tooth simple .6

Fang with at least one tooth (Figs. 96, 97) (Dominica, St. Lucia) .... ,footei

Fang without teeth ..... 5

Distal promarginal tooth equidistant from proximal promarginal and
retromarginal tooth as seen from below (Fig. 109) (northern Lesser

Antilles and Puerto Rico) whitcombi

Distal promarginal tooth closer to retromarginal tooth (Fig. 104) (Cuba to
St. Lucia) antiliana

Tibial apophysis of palpus truncated and slanted (Fig. 47) (Cuba) tibialis

Tibial apophysis usually spike-like or curved 7

Tibial apophysis distinctly curved (Cuba) 8

Tibial apophysis straight or nearly so 9

Larger than 5.25 mm found only in the mountains of southeastern Cuba

audax

Never larger than 5.25 mm; usually 3-4 mm; widespread in Cuba cubana

Chelicerae covered with white scales (Fig. 146) (Mona Island, Puerto

Rico) squamata

Chelicerae not covered with white scales ........ 10

First legs without pigment (Fig. 29); chelicerae rarely elongate; bulb
noticeably expanded distally (Fig. 33) (Quebec south to Florida and west

to Kansas, Oklahoma and Texas) mitrata

First legs pigmented 1 1

Chelicerae robust in larger males, being nearly as wide as long 12

Chelicerae elongate in larger males, much longer than wide .............. 15

Chelicerae with dorsal tubercle (rarely absent) (Fig. 140) (Hispaniola)

mandibularis

Chelicerae without dorsal tubercle 13

Tiny species (ca. 3 mm) with distinctive dorsal abdominal markings (Fig.

130) (Jamaica) calypso

Larger species (>3 mm), without distinctive dorsal markings (males difficult
to distinguish) ............. 14

RICHMAN— REVISION OF HENTZIA

293

14, Range from Panama along South American coast to Trinidad and

Guyana .parallela

Range Jamaica, Hispaniola, Cuba and the Bahamas vittata

15. Bulb with distinctive sperm duct loop (Figs. 70, 71); elongated species (Fig.

67) associated with small palms and cycads (Florida, Georgia) . . . .grenada
Bulb without sperm duct loop; sperm duct only slightly curved ........... 16

16. Dorsal abdominal pattern with 8-10 dark spots (Figs. 80, 81); associated

with tail palm trees such as Cocos (Florida, Bahamas, Cuba) ...... chekika

Pattern without 8-10 dark spots; not associated with tall palms .......... 17

17. Retromarginal cheliceral tooth opposite or nearly opposite proximal

promarginal tooth (Figs. 18, 19) (Nova Scotia, east to Bermuda, south to
Cuba, west to Oklahoma and Texas, northeastern Mexico) . . . , .palmarum
Cheliceral teeth approximately equidistant (Figs. 37, 75) 18

18. Anterior bulb lobe slanted and smaller than posterior (Fig. 76); relatively

tiny species (ca. 3 mm) known only from Sonora ............. .poenitens

Anterior bulb lobe not slanted; lobes approximately equal in size; larger
species (usually > 4 mm) (Tamaulipas south to Costa Rica, north along
the Pacific coast to Nayarit) .fimbriata

19. Covered with pink iridescent scales; epigynum with internal ‘‘cloverleaf”

structure (Fig. 155) (Hispaniola) .zombia

Without pink iridescent scales; epigynum not as in Fig. 155 ............ .20

20. Epigynum as in Figs. 65, 66, with c-shaped openings; dorsal abdominal spot

pattern block-like (Fig. 64) (Arizona) ............................ pima

Epigynum unlike Figs. 65, 66, with openings not c-shaped .............. .21

21. Tiny species (ca. 3-4 mm); dorsal abdominal pattern distinctive (Fig. 131);

epigynum with large round openings less than one diameter apart (Figs.
135, 136) (Jamaica) ......................................... calypso

Larger species (usually > 4 mm); epigynum with relatively smaller openings
or openings not round ........................................... 22

22. Legs unpigmented; epigynum with long central tubular structure (Fig. 35)

(Southern Canada south to Florida and west to Texas) .......... .mitrata

Legs pigmented; epigynum with short, belHike, or no central structure ... .23

23. With short belHike structure on epigynum; epigyeal openings rounded or

forming atrium (Figs. 24, 26) (Nova Scotia to Bermuda and Cuba, west to
Texas and northeastern Mexico) ........................... palmarum

With vague bell4ike structure on epigynum, or with none; openings
variable, but not as in palmarum .................................. 24

24. Epigynum as in Figs. 93, 94, with apparent double lateral tubes in dorsal

view; dorsal abdominal pattern lacking (Northern South America,
Trinidad, St. Vincent, Barbados) ............................. vernalis

294

THE JO'URNAL OF ARACHNOLOGY

Epigynum unlike Figs. 93, 94, with only one pair of tubes; with or without

dorsal abdominal pattern ........................................ 25

25. Epigynum with trumpet-shaped openings (Figs. 107, 152) ... ........... .26

Epigynum otherwise ............................................... 27

26. Spermathecae larger in area than trumpet-shaped part of epigynum (Figs.

152, 153) (found only on Mona Island) ..................... ,squamata

Spermathecae equal in area to trumpet-shaped part (Figs. 107, 108) (Cuba,

Hispaniola and Puerto Rico to St. Lucia) antillana

27. Epigynai openings circular or oval ................................... 28

Epigynal openings not circular or oval ............................... 30

28. With vague central bell-shaped structure on epigynum (Fig. 112) (Puerto

Rico to Guadeloupe) ..................................... whitcombi

Without vague bell-shaped structure ................................. 29

29. Epigynal openings circular (Fig. 144) (Hispaniola) ........... .mandibularis

Epigynal openings oval (Fig. 128) (Central America, northern South

America, east to Trinidad) .................................. parallela

30. Epigynum with characteristic “hump' anteriorly (Figs. 40, 42) (Tamaulipas

south to Costa Rica and north to Nayarit) ................... .fimbriata

Epigynum not so constructed ........................................ 31

3 1 . Epigynum wider than 1 ong ......................................... 32

Epigynum longer than wide ......................................... 34

32. Epigynum with complicated series of tubes (Fig. 51) ................ tibialis

Epigynum relatively simple internally ................................ 33

33. Abdomen lacking dorsal pattern (Fig. 74); small species - ca. 3 mm; found

in Sonora, Mexico ....................................... .poenitens

Abdomen with distinctive pattern of five pairs of brown spots (Fig, 81);
larger species - 5-6 mm; found in Cuba, the Bahamas and Florida. . .chekika

34, Epigynum with funnel-like or angulate openings (Cuba) 35

Epigynum without funneHike openings ............................... 36

35. Epigynum with straight, trumpet-like tubes, leading to slit-like openings

(Figs. 57, 58); large species (5.25 mm or larger) found in southeastern
mountains of Cuba .......................................... audax

Epigynum with tubes twisted laterally; openings oval; small species (usually
3"4 mm); widespread in Cuba; epigynum diagnostic (Figs. 62, 63) cubana

36. Epigynum with elongate tube leading to slit-like openings (Fig. 72) (Florida

and southern Georgia) ..................................... .grenada

Epigynum with bent tube leading to openings ......................... 37

RICHMAN— REVISION OF HENTZIA

295

37. Epigynal openings slanting posteriorly (Fig. 100) (Dominica to St. Lucia)

footei

Epigynal openings not slanting posteriorly, but laterally (Jamaica through
Hispaniola and Cuba to the Bahamas) vittata

SPECIES ACCOUNTS

Hentzia Marx 1883

Attus Walckenaer 1805 (applied to nearly all salticids - junior synonym of Salticus Latreille 1804).
Epiblemum Hentz 1832 (applied to Salticusas well as Hentzia).

Hentzia Marx 1883, type species palmarum (Hentz) 1832.

Wala Keyserling 1885, type species palmarum (Hentz) 1832. Synonymy Bryant 1940.

Anoka Peckham and Peckham 1893, type species vernalis Peckham and Peckham 1893. Synonymy
Bryant 1940.

Parahentzia Bryant 1943, type species mandibularis Bryant 1943. NEW SYNONYMY.

Maeviobeata Caporiacco 1947, type species charitonovi Caporiacco 1947. NEW SYNONYMY.

As noted in an earlier section, this genus is characterized by the primarily
female characters of hair pencils below the PME and spatulate hairs on the
ventral retromargin of the first patella and distal femur. Members of the genus
are also characterized by males having long, thin emboli arising on the prolateral
side of the tegulum and an anterior separation on the bulb, forming two lobes.
Hentzia vernalis is an exception to the latter character, having the bulb partially
reduced. Most males have elongate chelicerae and the first legs considerably
darker than posterior three pairs. Females usually have spermathecae with a
series of spiral twists and always have chelicerae with two simple promarginal
teeth and one bicuspid retromarginal tooth. The retromarginal teeth of the male
chelicerae are usually simple or truncate, the latter form often being bent distally.

SPECIES MISPLACED IN THE GENUS HENTZIA

Wala noda Chamberlin 1916 = Corythalia noda (Chamberlin), NEW
COMBINATION. Holotype and paratype females from Torontoy, PERU, and
female from Conservidayo River, PERU (MCZ) examined.

/MLAfyli? C/M SPECIES GROUP

The palmarum species group consists of seven species of slightly elongate
spiders with short to moderately elongate chelicerae and a bell-shaped or tubular
central structure on the epigynum of the females. With the exception of H.
mitrata, all of the known males have darkly pigmented or marked legs. The
group is the most widespread of the species groups of Hentzia, being found from
Nova Scotia to Bermuda, south to Cuba and South America, west to Oklahoma
and Texas and north along the west Mexican coast as far as Nayarit and Sinaloa.
One species, H. pima n. sp., is known from only one female, but appears to be a
member of this group and occurs in southern Arizona.

At present, the species within this group and within the vittata group are
perhaps the most difficult to relate by synapomorphic characters. In the case of

296

THE JOURNAL OF ARACHNOLOGY

the palmarum group this is, at least in part, because several of the species are
from Cuba and relatively little material is available. It appears that H. palmarum,
H. fimbriata and H. tibialis are closely related, based on similarities in the
structure of the palpi, such as the long, relatively thin and slightly curved tibial
apophysis. It could be equally argued that H. mitrata would fit here as well, and
I have left it next to H. palmarum in the species descriptions for convenience, but
it differs from all the other species in the group in having long, white-fringed
front legs and thus appears on the cladogram as a separate line (Fig. 15). Both H.
audax and H. cubana have extremely similar males, except for size and thickness
of the palpal tibial apophysis, but the females differ significantly (see species
descriptions). Finally the male of H. pima is unknown and its placement on the
cladogram is problematic. Until more material becomes available, especially from
Cuba, the relationships of the species within the palmarum species group can only
be assigned provisionally.

Hentzia palmarum (Hentz)

Figs. 16-28, Map 1

Epiblemum palmarum Hentz 1832:108 (type lost); 1846:366; Peckham and Peckham 1883:28.

Attus marginatus Walckenaer 1837:466 (type lost) (see Maddison 1986:142, 143).

Attus ambiguus Walckenaer 1837:467 (type lost); Chamberlin and Ivie 1944:201.

Hentzia palmarum Marx 1883:26; Kaston 1948:491; Chickering 1944:161.

Icius palmarum Peckham and Peckham 1888:46; Emerton 1891:232; Emerton 1902:56.

Anoka palmarum Peckham and Peckham 1894:126; Simon 1901:614.

Wala palmarum Peckham and Peckham 1909:508; Comstock 1912.

NOTE: Icius vittatus Keyserling 1885 and Wala albovittata Keyserling 1885 were incorrectly assigned

as synonyms of H. palmarum and are instead senior synonyms of Hentzia peckhami (Cockerell).

Diagnosis. — Chelicerae of male with retromarginal tooth usually directly in line
with proximal promarginal tooth (Figs. 18, 19) (separates this species from all but
//. mitrata), bulb of male palpus with edges of distal half parallel or expanded
(Figs. 20, 22), first legs pigmented (Fig. 16) (separates this species from H.
mitrata), female epigynum with bell-like median structure (Figs. 24, 26).
Chelicerae of male usually elongate; male chelicerae and first legs pigmented;
median dorsal marking on posterior 1/3 of carapace if present thin (willow leaf
shape); female often with herring-bone pattern; bodies of both male and female
distinctly less elongate than those of H. grenada or H. chekika n. sp.

Male. — Total length 4.00-5.30. Carapace 1.60-2.00 long, 1.30-1.70 wide, 0.80-
1.00 high at PLE. Ocular area 0.70-0.90 long, 1.00-1.30 wide anteriorly, 1.10-1.40
posteriorly. Chelicerae 0.75-1.80 long 0.30-0.45 wide (10 males from Cedar Key
Levy Co., Florida). Carapace length and cheliceral length measurements (124
specimens from The Bahamas, Florida, Pennsylvania, Texas, Bermuda, Georgia,
New Jersey, Ontario, Nova Scotia, North Carolina, South Carolina, Arkansas,
Missouri, Massachusetts and Michigan, not including above specimens) 1.36-2.35
and 0.40-2.20 (Fig. 28). PME equidistant to ALE and PLE. Leg formula 1423.
Carapace red to orange-brown with white scales forming willow-leaf marking on
dorsal posterior 2/3 and lateral bands. Margin of carapace black, with dark area
becoming wider toward posterior. Clypeus covered with white hairs. Eyes ringed
with black except AME which are ringed with dark brown. Chelicerae dark red
brown with pale promarginal 1/3. Labium dark brown with pale tip. Sternum

RICHMAN— REVISION OF HENTZIA

297

Figures 16-19. — Hentzia palmarum (Hentz): 16, male from Plymouth Co., Mass., dorsal view; 17,
female from Highlands Co., Fla., dorsal view of abdomen; 18, left chelicera of male from Tulsa Co.,
Okla., ventral view; 19, left chelicera of male from Levy Co., Fla., ventral view.

298

THE JOURNAL OF ARACHNOLOGY

orange. Abdomen either with central dark band, lateral white scale bands on
either side and venter brown or with 3-4 dark spots on the central band.
Abdomen covered with iridescent scales. First leg red-brown, tarsus yellow,
proximal metatarsus yellow, dorsal femur lighter brown. Other legs yellow. Palpi
red“brown, cymbium yellow.

Female. — Total length 4.70-6.10. Carapace 1.90-2.40 long, 1.40-2.05 wide, 0.70-
1.00 high at PLE. Ocular area 0.80-1.00 long, 1.20-1.45 mm wide anteriorly, 1.30-
1.70 mm wide posteriorly. Chelicerae 0.50-0.80 long, 0.35-0.55 wide (10 females
from Cedar Key, Levy Co., Florida). PME slightly closer to ALE than to PLE.
Leg formula 1423. Carapace red-brown with lighter center which ends before
posterior margin. Eyes ringed with black except AME ringed with brown.
Carapace covered with white scales. Clypeus covered with white hairs. Chelicerae
red-brown. Endites dark brown, prolateral 1/3 lighter. Labium dark brown with
pale tip. Sternum orange-brown. Abdomen yellowish with dark central pattern,
sometimes with herring-bone pattern or with three to four brown spots, similar to
pattern of female H. mitrata. Front legs orange-brown with femur darker
anteroventrally and patella and tibia darker distally. Other legs yellow. Palpi
yellow with dorsal dark spots on proximal patella, tibia and tarsus.

Distribution. — Nova Scotia and Ontario south to Cuba and the Bahamas, east
to Bermuda and west to Minnesota, Nebraska, Texas and Tamaulipas, Mexico.
(Map 1).

Variation. — Males differ in size and in length of the chelicerae (Fig. 28). There
does not appear to be an allometric relationship between these two parameters
and the plotted dots fit a straight line with the formula y = 0.498697x + 1.194149.
Males of other species appear to have similar relationships between carapace and
chelicera measurements (differing somewhat in placement along the curve and in
the exact formula for the curve), and were represented by fewer data points.
Because of these factors, the graph of H. palmarum measurements is presented
here as a representative for the genus. Generally the pattern of the carapace is
relatively uniform, but both males and females have some variation in abdominal
patterns; the males sometimes having central spots, but may lack them and the
females sometimes have a herring-bone pattern or brown spots like female H.
mitrata. There seems to be no geographic association with these patterns.

Natural history. — Males and females have been collected throughout the year,
especially in the southern part of their range. Immatures seem to be also present
throughout the year. This species is primarily associated with shrubs and small
trees and has been collected on black mangroves, red mangroves, white
mangroves, willows, scrub oaks and various understory shrubs in Florida.
Females from Florida (most from black mangroves near Cedar Key, Levy Co.)
produced an average of 15.1 eggs in the laboratory (range 6 to 27, SD = 5.5, n =
24) from March through October. Populations on black mangrove were relatively
dense throughout the year in 1975. This species is often found sympatrically with
H. grenada (Peckham and Peckham), but the latter is almost always found on
saw palmettoes or dwarf Sabal palms. Hentzia mitrata (Hentz) is also found
sympatrically with H. palmarum and long series of both species were collected in
Bucks Co., Pennsylvania by Wilton Ivie, unfortunately without any ecological
notes. Males of H. palmarum exhibit a characteristic courtship and agonistic
display (described by Richman 1982).

RICHMAN— REVISION OF HENTZIA

299

Map I. — North and Central America, showing distribution of Hentzia palmarum (closed circles,
open circles = state record only), H. poenitens (closed square), H. pima (open square), and H.
fimbriata (closed triangles).

Specimens examined. — (Material from Bruce Cutler Collection not examined marked with an
asterisk) BAHAMAS: Great Exuma (UCB), Harbor Island (MCZ). BERMUDA: Grasmere (MCZ).
CANADA: NOVA SCOTIA; Digby (CNC), Kentville (CNC). ONTARIO; London (CNC). CUBA:
Cabanas (AMNH), Cristo (AMNH), Habana (AMNH, MCZ), Holguin (MCZ), Pinar Rio (AMNH),
Santa Clara (AMNH), Santiago de Cuba (MCZ), San Vicente (AMNH), Soledad (MCZ). MEXICO:
TAMAULIPAS; Reynosa (AMNH), Tampico (AMNH). U.S.A. (county records only, except for the
District of Columbia): ALABAMA; Baldwin (AMNH), Mobile (AMNH). ARKANSAS; Bradley
(EPC), Conway (EPC), Crawford (EPC), Hempstead (ANSP, AMNH, EPC), Jefferson (EPC),
Mississippi (EPC), Perry (EPC), Washington (EPC, MCZ). CONNECTICUT; Fairfield (AMNH),
Hartford (AMNH, USNMNH), Litchfield (AMNH, USNMNH), New Haven (AMNH, USNMNH),

300

THE JOURNAL OF ARACHNOLOGY

Figures 20-27. — Hentzia palmarum (Hentz): 20, 21, palp of male from Levy Co., Fla., 20, ventral
view; 21, retrolateral view; 22, 23, palp of male from Tulsa Co., Okla.; 22, ventral view; 23,
retrolateral view; 24, 25, epigynum of female from Levy Co,, Fla., 24, ventral view; 25, dorsal view;
26, 27, epigynum of female from Highlands Co., Fla., 26, ventral view; 27, dorsal view.

New London (AMNH). DELAWARE; Sussex (FSCA). DISTRICT OF COLUMBIA; “East Branch’
(USNMNH). FLORIDA; Alachua (AMNH, FSCA), Baker (FSCA), Bay (FSCA), Broward (FSCA),
Citrus (AMNH, FSCA), Collier (FSCA), Dade (AMNH, FSCA, MCZ), Dixie (FSCA), Duval
(FSCA), Escambia (FSCA), Gadsden (FSCA, MCZ), Glades (AMNH), Hendry (FSCA), Highlands
(AMNH, DBR, FSCA, MCZ, UCB), Hillsborough (FSCA), Indian River (FSCA), Jackson (AMNH,
FSCA), Jefferson (MCZ), Uike (AMNH, FSCA), Lee (AMNH), Leon (AMNH, MCZ), Levy (EPC,
FSCA), Maddison (MCZ), Manatee (FSCA), Marion (FSCA), Martin (AMNH, FSCA), Monroe

RICHMAN— REVISION OF //^7VrZ/.4 301

m m

Chelicera Length

Figure 28. — Regression of carapace length vs. chelicera length for males of Hentzia palmarum,
showing linear relationship and corresponding lack of allometry. Other species in the genus show
similar regressions.

(AMNH, FSCA, MCZ, USNMNH), Nassau (AMNH), Okaloosa (AMNH, FSCA), Orange (AMNH,
FSCA, USNMNH), Palm Beach (FSCA, MCZ), Pasco (FSCA), Pinellas (AMNH, EPC, FSCA,
MCZ), Polk (FSCA), Putnam (FSCA), St. Lucie (FSCA), Sarasota (FSCA), Seminole (AMNH),
Volusia (AMNH, FSCA), Wakulla (FSCA). GEORGIA; Baker (FSCA), Bibb (AMNH), Brooks
(MCZ), Charlton (AMNH, FSCA), Chatham (AMNH), Clarke (MCZ), Glynn (AMNH, USNMNH),
Liberty (AMNH), Lowndes (AMNH), Morgan (USNMNH), Raybun (AMNH), Screven (AMNH),
Tift (AMNH), Turner (AMNH), Ware (AMNH). ILLINOIS; Fayette (FSCA), Jackson (BC*) Kane
(AMNH) Hardin (AMNH), Madison (FSCA). INDIANA; (state record only) (AMNH). IOWA; Story
(AMNH), Woodbury (FSCA). KANSAS; Bourbon (AMNH), Meade (AMNH), Sedgewick (EPC).
KENTUCKY; Trigg (FSCA) and state record (AMNH). LOUISIANA; East Baton Rouge (AMNH),
Iberville (AMNH). MARYLAND; Montgomery (AMNH). MASSACHUSETTS; Barnstable
(AMNH), Hampshire (AMNH), Plymouth (USNMNH), Suffolk (EPC), Worchester (MCZ).
MICHIGAN; Clinton (FSCA), Lenawee (AMNH), Livingston (FSCA), Macomb (AMNH), Midland
(AMNH), Oakland (AMNH), St. Clair (AMNH), St. Joseph (USNMNH), Shiawassee (BC*).
MINNESOTA; LeSeur (BC*), Wabasha (BC*). MISSISSIPPI; Forest (AMNH), George (AMNH),
Hancock (AMNH), Harrison (AMNH). MISSOURI; Johnson (EPC), Newton (EPC), Phelps (EPC),
Vernon (EPC). NEBRASKA; Dawson (AMNH), Jefferson (DBR), Hamilton (DBR), Harlan (DBR).
NEW JERSEY; Bergen (AMNH), Burlington (AMNH, MCZ), Middlesex (AMNH, CNC), Ocean
(AMNH). NEW YORK; Suffolk (AMNH). NORTH CAROLINA; Avery (MCZ), Dare (CNC),
Durham (AMNH), Macon (USNMNH), New Hanover (MCZ), Wake (AMNH). OKLAHOMA;
Carter (AMNH), Choctaw (AMNH), Comanche (AMNH), Grady (AMNH), Le Flore (AMNH),
Marshall (AMNH), Payne (DBR), Pittsburgh (AMNH), Tulsa (DBR). PENNSYLVANIA; Bucks
(AMNH), Carbon (ANSP), Centre (MCZ) Luzern (ANSP), Schuylkill (USNMNH). SOUTH

302

THE JOURNAL OF ARACHNOLOGY

CAROLINA; Aiken (AMNH), Anderson (CNC), Berkeley (USNMNH), Charleston (AMNH),
Dorchester (AMNH), Kershaw (AMNH). TEXAS; Aransas (AMNH), Austin (AMNH), Bell
(AMNH), Bexar (AMNH, MCZ), Brazos (TAM), Cameron (AMNH, MCZ), Comal (AMNH), Dallas
(AMNH, USNMNH), Erath (TAM), Grayson (AMNH), Gillespie (AMNH), Hidalgo (AMNH, TAM,
USNMNH), Karnes (MCZ), Kenedy (TAM), Kerr (AMNH, FSCA), Limestone (AMNH), Llano
(AMNH), Lubbock (AMNH), Medina (AMNH), McLennan (AMNH), San Patricio (AMNH, FSCA,
MCZ), San Saba (AMNH), Shelby (AMNH), Travis (ANSP, AMNH), Walker (TAM), Wichita
(FSCA), Wilbarger (AMNH), Williamson (AMNH), Uvalde (AMNH). VIRGINIA; Accomack
(USNMNH), Arlington (AMNH, USNMNH), Augusta (USNMNH), Montgomery (FSCA), Norfolk
(MCZ).

Hentzia mitrata (Hentz)
Figs. 29-36, Map 2

Attus mitratus Hentz 1846:363 (type destroyed).

Attus morigerus Hentz 1846:365 (type destroyed).

Maevia sulphurea C. L. Koch 1848:71 (type not examined).

Maevia pallida C. L. Koch 1848:79 (type not examined).

Icius mitratus Peckham and Peckham 1888:48; Emerton 1891:232; 1902:57.

Anoka mitrata Peckham and Peckham 1894:125.

Wala mitrata Peckham and Peckham 1909:507.

H. mitrata Chickering 1944:159; Kaston 1948:492.

Diagnosis. — Male chelicerae generally short and lacking pigmentation, male
first legs with white hair fringe and unpigmented (Fig. 29). Male palpus with bulb
expanded toward distal end (Fig. 33). Female with elongate central bell-shaped
structure on epigynum (Fig. 35), generally with legs unpigmented. These
characteristics separate this species from all other members of the genus.

Male. — Total length 3.50-4.10. Carapace 1.50-1.70 long, 1.20-1.40 wide, 0.65-
0.85 high at PLE. Ocular area 0.65-0.75 long, 0.90-1.02 wide anteriorly and 1.00-

I. 10 wide posteriorly. Chelicerae 0.50-0.60 long, 0.30-0.35 wide (10 males from
NE of Jamison, Florseshoe Bend, Neshaminy Creek, Bucks Co., Pennsylvania).
Carapace length and cheliceral length measurements (49 specimens from Quebec,
Ontario, Georgia, New Jersey, Missouri, Mississippi, Florida and North Carolina)
1.45-2.05 and 0.50-1.15. The males with the longest chelicerae from Florida. PME
equidistant between ALE and PLE. Leg formula 1423. Carapace orange with red-
brown bands on either side of center, extending around eyes. Eyes ringed in black
except AME which are ringed with brown. Clypeus covered with white hairs.
Chelicerae whitish. Endites, labium and sternum yellow-cream. Abdomen yellow
colored with brown markings (Fig. 29). Venter gray to cream. All legs yellow.
Palpi yellow. In life males of this species from Florida appear to be white with
white legs and orange dorsum.

Female. — Total length 2.90-4.50. Carapace 1.50-1.80 long, 1.20-1.40 wide, 0.60-
0.75 high at PLE. Ocular area 0.65-0.80 long, 0.95-1.12 wide anteriorly, 1.00-1.20
posteriorly. Chelicerae 0.40-0.50 long, 0.26-0.35 wide (10 females from NE of
Jamison, Horseshoe Bend, Neshaminy Creek, Bucks Co., Pennsylvania). PME
closer to ALE than to PLE. Leg formula 1423. Carapace orange-brown, covered
with white hairs, scales. Clypeus covered with white hairs. Endites and labium
lighter distally. Abdomen yellowish with brown pattern (Fig. 30). First legs
orange. Other legs and palpi yellowish. In life females from Florida are white
with brown markings on the body and white legs.

RICHMAN— REVISION OF HENTZIA 303

31

Figures 29-32. — Hentzia mitrata (Hentz): 29, male from Washington Co., Minn., dorsal view; 30,
female from Washington Co., Minn., dorsal view of abdomen; 31, 32, ventral views of male left
chelicera; 31, from Bucks Co., Penn., 32, from Dade Co., Fla.

Distribution. — Ontario and Quebec south to Florida and the Bahamas and
west to eastern Texas and Minnesota (Map 2).

Varation. — Males generally have short chelicerae in most of the species range
except in parts of Florida, where they may have chelicerae half as long as the
carapace. Males from Florida tend to have sparser fringes on the first legs and

Figures 33-36. — Hentzia mitrata (Hentz): 33, 34, palp of male from Dade Co., Fla.; 33, ventral
view; 34, retrolateral view; 35, 36, epigynum of female from Bucks Co., Penn.; 35, ventral view; 36,
dorsal view.

narrower palpi, but otherwise match with more northern specimens. Females vary
in the dorsal markings, with some similar to H. palmarum females. The epigyna
are relatively uniform. Some specimens from Cuba (MCZ) may have to be
assigned to this species, but I have not seen a male that looked exactly like
mainland specimens. Until more material is available from Cuba, I feel that it is
best to not include the records of H. mitrata from this island.

Natural history. — Adults have been collected in every month of the year,
especially in the southern part of the range. They usually have been collected on
shrubs, such as wax myrtle, or trees, such as live oak. Natural enemies include
the wasp Trypoxylon (Trypargilum) clavatum Say (Hymenoptera: Sphecidae).
Adults and immatures were found as prey in the nests of this wasp under a
wooden bridge in Leon Co., Florida in July along with some H. palmarum (Coll.
G. B. Edwards). The courtship behavior was described by Peckham and Peckham
(1889, 1890) and Richman (1982).

RICHMAN— REVISION OF HENTZIA

305

Specimens examined. — (Material from Bruce Cutler Collection not examined marked with an
asterisk) BAHAMAS: Great Exuma (MCZ), Rum Cay (MCZ). CANADA: ONTARIO; Belleville
(FSCA), Essex Co. (CNC), London (CNC), Pelee Island (AMNH), Turkey Point (FSCA). QUEBEC;
Frelighsburg (CNC). U.S.A.: (County records only, except for the District of Columbia) ALABAMA;
Baldwin (AMNH), Coosa (AMNH), Madison (AMNH), Morgan (AMNH). ARKANSAS; Arkansas
(EPC), Conway (EPC), Hempstead (ANSP, EPC), Jefferson (EPC), Mississippi (EPC), Polk (EPC),
Washington (EPC, MCZ). CONNECTICUT; London (AMNH), Middlesex (AMNH), New Haven
(AMNH, USNMNH). DISTRICT OF COLUMBIA; “Chain Bridge” (USNMNH), “Battery”
(USNMNH). FLORIDA; Alachua (FSCA, UCB), Dade (AMNH, FSCA, MCZ), Gadsden (MCZ),
Glades (AMNH), Hernando (FSCA), Highlands (AMNH), Hillsborough (FSCA), Indian River
(MCZ), Jackson (FSCA), Jefferson (MCZ), Lake (AMNH), Levy (FSCA), Leon (FSCA, MCZ),
Martin (AMNH), Monroe (AMNH, FSCA), Orange (AMNH), Palm Beach (FSCA, MCZ), Pasco
(FSCA), Polk (FSCA), St. Lucie (FSCA), Sarasota (FSCA). GEORGIA; Charleton (USNMNH),
Dekalb (AMNH), Hall (AMNH, USNMNH), Thomas (CNC, DBR). ILLINOIS; St. Clairl (AMNH)
also state record (AMNH). INDIANA; State record (AMNH). IOWA; Clayton (MCZ). KANSAS;
Riley (FSCA). LOUISIANA; Madison (AMNH). MARYLAND; Montgomery (AMNH, USNMNH).
MASSACHUSETTS; Plymouth (USNMNH), Suffolk (MCZ). MICHIGAN; Livingston (FSCA), St.
Claire (AMNH), Shiawassee (BC*). MINNESOTA; Anoka (FSCA), Blue Earth (MCZ), Chisago
(BC*), Hennepin (BC*), Ramsey (BC*), Wright (BC*), Washington (DBR). MISSISSIPPI; Forest

306

THE JOURNAL OF ARACHNOLOGY

(AMNH), Humphreys (AMNH), Panola (EPC), Sharkey (EPC). MISSOURI; Johnson (EPC), Phelps
(EPC). NEW JERSEY; Bergen (AMNH), Burlington (AMNH), Camden (CNC), Hunterdon
(AMNH), Middlesex (CNC, FSCA), Monmouth (AMNH), Ocean (AMNH), Sussex (AMNH),
Warren (FSCA). NEW YORK; Suffolk (AMNH). NORTH CAROLINA; Cumberland (FSCA), New
Hanover (MCZ), Orange (AMNH), Transylvania (MCZ), Wake (AMNH), Washington (MCZ).
OHIO; Lake (AMNH). PENNSYLVANIA; Bucks (AMNH), Philadelphia (ANSP). SOUTH
CAROLINA; Oconee (AMNH). TEXAS; Jasper (AMNH), Sabine (MCZ), Walker (TAM).
VIRGINIA; Amherst (CNC, FSCA), Arlington (USNMNH), Campbell (USNMNH), Fairfax
(USNMNH) Montgomery (USNMNH) Page (AMNH). WEST VIRGINIA; Hamshire (AMNH).
WISCONSIN; LaCrosse (FSCA), St. Croix (BC*), and state record (MCZ).

Hentzia fimbriata (R O. P. - Cambridge)
Figs. 37=43, Map 1

Anoka fimbriata F. O. P. - Cambridge 1901:256 (types from Guatemala in BMNH examined).

Anoka grenada F. O. P. - Cambridge 1901:256 (not Anoka grenada Peckham and Peckham 1894)

(specimens from Guatemala in BMNH examined).

Wala fimbriata Petrunkevitch 1911:716.

H. fimbriata Roewer 1954:1217.

Diagnosis. — Male similar to H. palmarum but with retromarginal cheliceral
tooth separated from the first promarginal by at least the distance between
promarginal teeth as seen ventrally (Fig. 37). Female with characteristic bell-
shaped central epigynal structure with a “hump” between lateral openings (Figs.
40=43).

Male. — Total length 3.10=4.90. Carapace 1.50-1.90 long, 1.15-1,60 wide, 0.60-
0.80 high at PLE. Ocular area 0.60-0.80 long, 0.95-1.20 wide anteriorally and

I. 00-1.30 wide posteriorly. Chelicerae 0.50-1.50 long, 0.30-0.50 wide (14 males
from Guatemala, Mexico and Nicaragua). Carapace and cheliceral length
measurements (23 males from Mexico and Central America) 1.50-2.38 and 0.50-
1.89. Males of Cambridge’s ^'grenada'' with long chelicerae as opposed to his type
specimens of A. fimbriata. As far as is known there are no valid records for H.
grenada from Central or South America (see description of H. grenada). PME
slightly closer to ALE than to PLE. Leg formula 1423. Carapace yellow, with
lateral bands of white hairs. Black around all eyes but AME; brown around
AME. Clypeus with white hairs. Chelicerae red to yellow-brown. Endites and
labium yellow brown. Sternum orange-brown. Abdomen yellow with or without
darker markings. First legs yellow-brown, darker on anterior ventral femora,
patellae and tibiae. Other legs yellow. Pedipalpi yellow with dark spot on distal
patellae.

Female. — Total length 3.60-5.15. Carapace 1.40-2.15 long, 1.10-1.40 wide, 0.50-
0.70 high at PLE. Ocular area 0.60-0.75 long, 0.88-1.50 wide anteriorly and 0.95-
1.15 wide posteriorly. Chelicerae 0.30-0.40 long, 0.20-0.30 wide (14 females from
Guatemala, Mexico and Nicaragua), PME closer to ALE than to PLE. Leg
formula 1423. Carapace orange with white hairs laterally. Black around eyes
except AME. Brown around AME. Clypeus covered with white hairs. Chelicerae
orange. Endites and labium orange. Sternum orange. Abdomen light brown with
gray-brown markings. Venter yellow. First legs yellow to orange. Pedipalpi
yellow.

Distribution. — East coast of Mexico from southern Tamaulipas south to

Colombia and north along the West coast to Nayarit (Map 1).

RICHMAN— REVISION OF HENTZIA

307

Figures 37-43, — Henizia fimbriata (F. 0. P.-Cambridge): 37--39, “Syntype” male from Guatemala; 37,
left chelicera, ventral view; 38, palp, ventral view; 39, palp, retrolateral view; 40, 41, epigynum of
female from Panzos, Guatemala; 40, ventral view; 41, dorsal view; 42, 43 epigynum of female from
Campeche, Mexico; 42, ventral view; 43, dorsal view.

Natural history. — Males have been collected in January, March and June-
November. Females have been collected in February-March, May- August and
October. Adults probably can be found through the year, as with H. palmarum.
Specimens from Veracruz were collected by W. Maddison on bushes along
roadside in marsh.

Specimens examined.— COLOMBIA: Isla San Andres (UCR). COSTA RICA: GUANACASTE; 4
km NW Canyas (UCB), Comunidad (MCZ). GUATEMALA: Amatitian (AMNH), Panzos (AMNH),
Petapa (BMNH), Rabinal (AMNH). HONDURAS: Roatan (FSCA). MEXICO: CAMPECHE; 6 km
W Francisco Escarcega (MCZ). COLIMA; Santiago (AMNH). NAYARIT; Tepic (AMNH).
OAXACA; Soledad (AMNH), 13 km S Tuxtepec (MCZ). SAN LUIS POTOSI; 20 mi. E Cuidad del
Maiz (AMNH), Tamazunchale (AMNH). TAMAULIPAS; Naciemento del Rio Frio nr. Gomez Farias

308

THE JOURNAL OF ARACHNOLOGY

Figures 44-51. — Henizia tibialis Bryant: 44, male from San Vincente, Cuba, dorsal view; 45, left
chelicera of male from Pinar, Cuba, ventral view; 46, 47, palp of male from Soledad, Cuba; 46,
ventral view; 47, retrolateral view; 48, 49 epigyeum of female from San Vincente, Cuba; 48, ventral
view; 49, dorsal view; 50, 51, epigynum of female from Habana, Cuba; 50, ventral view; 51, dorsal
view.

(MCZ), 34 km E Tula (MCZ). VERACRUZ; Alamo (AMNH), 40 km E Coatzacoaicos (MCZ), Rio
Bianco (MCZ), San Rafael (MCZ), Tlacotalpan (AMNH). YUCATAN; Grutas de Lolton 7 km S
Oxkutzca (MCZ), 12 km N Piste (MCZ), Uxmal (AMNH). NICARAGUA: Granada (MCZ),
Masachapa (AMNH).

RICHMAN— REVISION OF HENTZIA

309

Map 3. — West Indies, showing distribution of Hentzia antilhna (dosed circles) and H. tibialis (open
circles).

Hentzia tibialis Bryant
Figs. 44-51, Map 3

Balmaceda peckhami Bryant 1940:464 (holotype female from Soledad, Cuba, in MCZ examined -
Hentzia peckhami preoccupied by Anoka p. Cockerell 1893). NEW SYNONYMY.

H. tibialis Bryant 1940:498 (holotype male from Soledad, Cuba, in MCZ examined, allotype female =
H. chekika n. sp.).

Diagnosis. — Males differ from all other Hentzia spp. in having a truncated
tibial apophysis on the palpus (Fig. 47). Otherwise they resemble H. palmarum in
general structure. The apparent female is Balmaceda peckhami Bryant. The
female paired by Bryant (1940) with H. tibialis is H. chekika n. sp. The female of

H, tibialis differs from all other Hentzia in having wide spaced openings in the
epigynai plate and in the structure of the spermathecal ducts (Figs. 48-51).

Male. — Total length 3.90-4.92. Carapace 1.35-1.89 long, 1.35-1.80 wide, 0.70-
0.86 high at PLE. Ocular area 0.70-0.82 long, 1.05-1.23 wide anteriorly and 1.10-

I. 31 wide posteriorly. Cheiicerae 0.90-1.89 long, 0.35-0.53 wide (five males from
Cuba). PME closer to ALE than to PLE. Leg formula 1432 (leg 3 and 2 very
close in length). Carapace red-brown with white hairs laterally. Black around eyes
except dark brown around AME. Clypeus red-brown. Cheiicerae red-brown with
anterolateral dark brown line. Fang black with distal 1/4 yellow. Endites and
labium dark brown. Sternum orange. Abdomen orange-brown covered with

310

THE JOURNAL OF ARACHNOLOGY

iridescent scales, white laterally. Venter brown,; epigastric furrow with lateral
orange streaks. First legs red-brown, darker anteroventrally. Tarsi yellow. Other
legs orange. Pedipalpi yellow; bulb very dark, almost black.

Female. — Total length 3.40-4.65. Carapace 1.40-1.70 long, 1.10-1.38 wide, 0.60-
0.80 high at PLE. Ocular area 0.65-0.85 long, 0.95-1.18 wide anteriorly and 1.00-
1.25 wide posteriorly. Chelicerae 0.40-0.50 long, 0.25-0.35 wide (seven females
from Cuba). PME closer to ALE than to PLE. Leg formula 1423 (leg 1 and leg 4
close in length). Carapace orange-brown with white scales and hairs laterally.
Black around eyes except brown around AME. Clypeus covered with white hairs.
Chelicerae orange to red-brown. Endites and labium orange-brown with lighter
tips. Sternum orange to brown. Abdomen yellow with darker brown central band
of block-like markings. Lateral areas with brown blotches. Venter yellow. First
legs yellow-brown, darker around distal patellae, tibiae and metatarsi; tarsi
yellow. Other legs yellow. Pedipalpi yellow with proximal half annulae except on
femora.

Distribution. — Known only from Cuba (Map 3).

Natural history. — Males collected from May and July-September. Females
collected from February and July.

Specimens examined. — CUBA: Habana (MCZ), Luis Lazo (MCZ), Pinar del Rio (AMNH), San
Vincente (AMNH), Soledad (MCZ).

Hentzia audax Bryant
Figs. 52-58, Map 4

Hentzia audax Bryant 1940:496 (holotype male and allotype female from Pico Turquino, Cuba in

MCZ examined).

Diagnosis. — Males differ from other members of the genus, except H. cubana,
by the curved and often massive tibial apophysis on the palpus (Figs. 55, 56) and
the pattern of cheliceral teeth (Fig. 54). It differs from H. cubana in being much
larger (over 5.25 mm as opposed to 3-4 mm) and in the structure of the palp
(Figs. 55, 56). The female epigynum differs from that of H. cubana in having
trumpet-like straight lateral tubes, with slit-like openings (Figs. 57, 58).

Male. — Total length 5.30-6.00. Carapace 2.30-2.54 long, 1.90-2.10 wide, 0.90-
1.10 high at PLE. Ocular area 0.90-1.10 long, ,L35- 1,45 wide anteriorly, 1.45-1.60
wide posteriorly. Chelicerae 1.60-2.40 long, 0.60-0.66 wide (six males from Pico
Turquino, Cuba). PME closer to ALE than to PLE. Leg formula 1423. Carapace
red-brown, with yellow area between eyes and dark around eyes. White scales
laterally. Clypeus and chelicerae red-brown. Endites and labium light red-brown.
Sternum light red-brown to yellow-brown.' Abdomen yellow-brown with an
iridescent sheen caused by scales. Faint dorsal markings, lateral white lines.
Venter yellow. First legs red-brown, slightly lighter on dorsal proximal femora
and on tarsi. Other legs yellow. Pedipalpi red-brown, cymbium darker.

Female, — Total length 6.07-6.80. Carapace 2.20-2.21 long, 1.80-2.00 wide, 1.10-
1.15 high at PLE. Ocular area 0.90-1.05 long, 1.39-1.45 wide anteriorly and 1.64-
1.65 wide posteriorly. Chelicerae 0.82-0.90 long, 0.49-0.60 wide (two females Pico
Turquino, Cuba). PME slightly closer to ALE than to PLE. Leg formula 1423.

RICHMAN— REVISION OF HENTZIA

311

Figures 52-58. — Hentzia audax Bryant, types from Pico Turquino, Cuba: 52, 53, paratypes; 52,
male, dorsal view; 53, female, dorsal view of abdomen; 54, left chelicera of male holotype, ventral
view; 55, 56, palp of male paratype; 55, ventral view; 56, retrolateral view; 57, 58, epigynum of female
paratype; 57, ventral view; 58, dorsal view.

Carapace red-brown; yellow areas between eyes; eyes dark, Chelicerae red-brown.
Endites and labium red-brown. Sternum yellow-brown. Abdomen yellow with
central pattern of brownish anterior paired streaks, followed by three central
streaks or triangles. Venter yellow. First legs red-brown. Other legs yellow.
Pedipalpi red-brown with yellow tips.

Distribution. — Known only from mountains and coast of southeastern Cuba.

(Map 4)

Natural history,— Males and females collected in June.

Specimens examined. — CUBA: Pico Turquino (MCZ), coast below Pico Turquino (MCZ).

312

THE JOURNAL OF ARACHNOLOGY

Map 4. — West Indies and Florida, showing distribution of Hentzia grenada (closed circles), H.
mandibularis (open circles), H. audax (closed squares), and H. vernalis (closed triangles).

Hentzia cubana, new species
Figs. 59-63, Map 5

Types. — Holotype male and allotype female from Soledad, Cuba (August 1931,

N. Banks) deposited in the Museum of Comparative Zoology.

Etymology. — The name is derived from the island on which the species occurs.
Diagnosis. — Males differ from all other species of Hentzia except audax in
having a hooked tibial apophysis combined with nearly equidistant cheliceral
teeth. They differ from H. audax in being much smaller (less than 4 mm), in the
more slender tibial apophysis of the male pedipalp (Figs. 60, 61), and in the
associated females. Females differ from all other species in the details of the
epigynum (Figs. 62, 63). This species may prefer the lowlands more than H.
audax.

Male. — Total length 3.25-3.75. Carapace 1.35-1.55 long, 1.10-1.30 wide, 0.60-

O. 75 high at PLE. Ocular area 0.60-0.75 long, 0.90-1.10 wide anteriorly and 1.00-
1.12 wide posteriorly. Chelicerae 0.40-1.10 long, 0.30-0.42 wide (seven males from
Soledad, Cuba). PME closer to ALE than to PLE. Leg formula 1423. Carapace
red-brown with lateral white scales. Black around eyes except red-brown around
AME. Clypeus covered with white hairs. Chelicerae red-brown. Endites very
dark, with anterolateral edge white. Labium dark brown, tip white. Sternum

RICHMAN— REVISION OF HENTZIA

313

palp of male from Soledad, 60, ventral view; 61, retrolateral view; 62, 63, epigynum of female; 62,
ventral view; 63, dorsal view.

orange. Abdomen brown^ with shiny scales; laterally lighter. Venter light brown.
First legs red-brown; tarsus and metatarsus lighter and metatarsus with slightly
darker distal end. Other legs yellow-orange. Pedipalps orange-brown, cymbium
lighter.

Female. — Total length 4.20-5.00. Carapace 1.55-1.90 long, 1.28-1.60 wide, 0.65-
0.75 high. Ocular area 0.70-0.80 long, 1.08-1.20 wide anteriorly and 1.10-1.30
wide posteriorly. Chelicerae 0.40-0.55 long, 0.30-0.40 wide (five females from
Soledad, Cuba). PME equidistant between ALE and PEE. Leg formula 1423.
Carapace orange-brown; white scales laterally. Black around eyes except brown
around AME. Clypeus covered with white hairs. Chelicerae orange. Endites and
labium dark brown, with pale tips. Sternum dark brown. Abdomen yellowish
with brown central band; central band with two light streaks forming inverted
“V” anteriorly, followed by lateral notches near middle and narrowing into a
triangular and then block-like mark toward posterior. Venter yellow. First legs

314

THE JOURNAL OF ARACHNOLOGY

Map 5. — West Indies, showing distribution of Hentzia whitcombi (closed circles), H. cubana (open
circles), H. squamata (closed square), and H. calypso (closed triangles).

orange-brown with tarsus-metatarsus lighter; metatarsus with darker distal ring.
Other legs yellow-orange. Pedipalpi yellow with dark dorsal markings on
proximal tarsus and tibia.

Distribution. — Known only from Cuba (Map 5).

Natural history. — Males collected in February-March, May, July-September
and November. Females in February-March and July-August. As in other
tropical species, adults are probably found throughout the year.

Specimens examined. — CUBA: Buenos Aires (MCZ), Cienaga de Zapata (MCZ), Habana (MCZ),
Pinar del Rio (MCZ), Punta San Juan (MCZ), Soledad (MCZ), 5 mi. E of Soledad (MCZ), Trinidad
Mountains (MCZ).

Hentzia pima, new species
Figs. 64-66, Map 1

Types. — Holotype female from Brown Canyon, Baboquivari Mountains, Pima

Co., Arizona (8 June 1958, W. J. Gertsch) deposited in the American Museum of
Natural History. The male is not known.

Etymology, — The name is derived from the American Indian tribe which also
gave the name to the county.

Diagnosis. — Different from all other members of the genus in the c-shaped

RICHMAN— REVISION OF HENTZIA

315

Figures 64-66. — Hentzia pima n. sp., female holotype from Brown Canyon, Baboquivari Mts., Pima
Co., Ariz.; 64, dorsal view of abdomen; 65, 66, epigynum; 65, ventral view; 66, dorsal view.

epigynal openings (Figs. 65, 66). Dorsal abdominal pattern (Fig. 64) similar to
other members of the palmarum group.

Female. — Total length 5.61. Carapace 2.12 long, 1.77 wide, 0.89 high at PLE.
Ocular area 0.89 long, 1.30 wide anteriorly and 1.48 wide posteriorly. Chelicerae
0.77 long, 0.47 wide (only holotype). PME closer to ALE than to PLE. Leg
formula 1423. Carapace orange with scattered black hairs and white scales
laterally. Dark around eyes. Clypeus covered with white hairs. Chelicerae orange,
red-brown ventrally. Endites red-brown with white prolateral edges. Labium red-
brown. Sternum orange. Abdomen yellow with distinctive brown markings
consisting of two curved v-shaped marks in center of dorsum (Fig. 64). First legs
orange. Other legs and pedipalpi yellow.

Distribution. — Known only from southern Arizona (Map 1).

Natural history. — Holotype collected in June.

Specimens examined. — Only from the type locality (AMNH).

GRENADA SPECIES GROUP

The grenada species group consists of only three species, H. grenada, H.
chekika and H. poenitens. They have distinctive female epigyna (Figs. 72, 73, 78,
79, 85, 86), with elongate to short J-shaped tubes leading to the spermathecae.
Males of H. grenada differ from other members of the species group and from all
other known Hentzia males by having a sperm tube loop in the bulb (Figs. 70,
71). It also differs from the other two species in the species group by the length of
the female J-shaped tubes. The possible relationship of these three species is
indicated on the cladogram (Fig. 15). Both H. grenada and H. chekika are very
elongate, with moderately long chelicerae in large males. Hentzia poenitens is one
of the smallest of the species of Hentzia and is widely separated geographically
from the other species, being found in Sonora on the west coast of Mexico.

Hentzia grenada (Peckham and Peckham)

Figs. 67-73, Map 4

Anoka grenada Peckham and Peckham 1894:126 (holotype male from “New Grenada” in MCZ

examined).

316

THE JOURNAL OF ARACHNOLOGY

Figures 67-73. Hentzia grenada (Peckham and Peckham): 67, 68, 70-73, from Highlands Co., Fla.;
67, male, dorsal view; 68, female, dorsal view of abdomen; 69, left chelicera of holotype male from
“New Grenada”, ventral view; 70, 71, palp of male; 70, ventral view; 71, retrolateral view; 72, 73,
epigynum of female; 72, ventral view; 73, dorsal view.

Wala grenada Banks 1904:138; Peckham and Peckham 1909:507.

H. grenada Roewer 1954:1217.

Diagnosis. — Males differ from all other members of the genus by the presence
of a loop in the sperm duct as viewed from the retromargin of the palpal bulb
(Figs. 70, 71). Females have a characteristic epigynal structure, including two
long, slightly J-shaped ducts (Figs. 72, 73). Both males and females have very
elongate bodies and males possess a white band on the central dorsal carapace
which widens toward the eyes (Fig. 67). Only H. chekika has a similar
appearance, but the latter species is larger, lives on tall palm trees and differs in

RICHMAN—REVISION OF HENTZiA

317

lacking the loop on the bulb of the male palpus and in the structure of the female
epigynum,

Male. — Total length 4. 08-5. 70. Carapace 1.58-1.80 long, 1.20-1.70 wide, 0.66-
0.80 high at PLE. Ocular area 0.65-0.90 long, 1.00-1.23 wide anteriorly and 1.00-
1.23 wide posteriorly. Chelicerae 0.90-1.80 long, 0.30-0.48 wide (10 males from
Highlands and Martin counties, Florida). PME slightly closer to ALE than to
PLE. Leg formula 1423. Carapace red-brown with central band of white scales
and hairs widening toward AME. Lateral white band followed by marginal dark
band. Black around eyes except brown around AME. Clypeus covered with white
hairs. Chelicerae dark red-brown, fangs lighter on distal 1/3. Endites and labium
dark brown. Sternum orange. Abdomen yellow dorsally with red-brown markings
(Fig. 67). Venter light brown with lateral areas lighter. First legs yellow with red-
brown prolateral and retrolateral femora, distal patellae, distal and proximal
tibiae. Other legs yellow. Pedipalpi red-brown with cymbium yellow.

Female. — Total length 4.10-5.50. Carapace 1.60-2.00 long, 1.20-1.52 wide, 0.55-
0.80 high at PLE. Ocular area 0.70-0.90 long, 1.00-1.22 wide anteriorly and 1.05-
1.28 wide posteriorly. Chelicerae 0.40-0.60 long, 0.28-0.42 wide (10 females from
Highlands and Martin counties, Florida). PME slightly closer to ALE than to
PLE. Leg formula 1423. Carapace orange, darker laterally back of eyes and
lighter down middle and in band along either side. Margin line black. Black
around eyes except brown around AME. Clypeus covered with white hairs.
Chelicerae orange-brown. Endites and labium orange-brown with light distal tips.
Sternum yellow. Abdomen yellow with brown markings (Fig. 68). Venter yellow.
First legs yellow with brown retrolateral streak on femora and dark spots on the
promargin of femora, distal and prolateral patellae, and distal and proximal
tibiae. Other legs yellow. Pedipalpi yellow with dorsal spots on proximal patellae,
tibiae and tarsi.

Distribution. — Florida and south Georgia (Map 4).

Discussion. — It appears that the type specimen of this species was mislabeled at
some point, as it matches perfectly with males from Florida but does not
compare with any specimens known from Central America, northern South
America or the West Indies. New Grenada is now the modern state of Colombia.
A similar problem was discovered with two of Keyserling’s types (see H. vittata).

Natural history. — Males have been collected in every month but October and
females from every month but January, August and October. There are thus
probably adults throughout the year. The courtship is quite similar to that of H.
palmarum and males will readily mate with females of the latter species under
laboratory conditions (Richman 1982). Hybrids are unknown in the wild and
immatures from crosses were not successfully raised in the laboratory.

Specimens examined. — U.S.A.: (County records only) FLORIDA; Alachua (FSCA), Baker (FSCA),
Collier (AMNH, FSCA, MCZ), Dade (FSCA, MCZ), Duval (FSCA), Gadsden (FSCA), Glades
(FSCA), Hamilton (AMNH), Highlands (AMNH, DBR, FSCA), Hillsborough (FSCA), Indian River
(AMNH, FSCA), Lake (AMNH), Lee (AMNH), Manatee (DBR), Marion (FSCA), Martin (FSCA),
Monroe (AMNH, FSCA, MCZ), Pinellas (FSCA), Nassau (FSCA), Putnam (FSCA). GEORGIA;
Chatham (USNMNH), Lowndes (AMNH).

Hentzia poenitens (Chamberlin)

Figs. 14-79, Map 1

Wala poenitens Chamberlin 1924:680 (holotype male from Guaymas, Sonora, consisting of only the
right palpus, in MCZ examined).

318

THE JOURNAL OF ARACHNOLOGY

Figures 74-79. — Hentzia poenitens (Chamberlin) from Sonora, Mexico: 74, female from Los
Algodones, dorsal view of abdomen; 75, left chelicera of male from Guaymas, ventral view; 76, 77,
palp of male from Guaymas, 76, ventral view; 77, retrolateral view; 78, 79, epigyeum of female from
Los Algodones; 78, ventral view; 79, dorsal view.

H. poenitens Roewer 1954:1218.

Diagnosis. — Male palpus with slanted anterior bulb (Fig, 76) and internal
structure of female epigynum with short J-shaped ducts diagnostic (Fig. 79). The
smallest member of the genus other than H. calypso. Lacking markings.

RICHMAN—REVISION OF HENTZIA

319

Male. — Total length 3.60. Carapace 1.59 long, 1.30 wide, 0.65 high at PLE.
Ocular area 0.71 long, 0.94 wide anteriorally and 1.06 wide posteriorly. Chelicerae
1.18 long, 0.41 wide (one male from Guaymas, Sonora). PME closer to ALE than
to PLE. Leg formula 1423. Carapace orange with scattered black hairs; white
scales laterally. Eyes as in other species of the group. Clypeus covered with white
hairs. Chelicerae brown, darker and mottled ventrally; with corrugated
iridescence. Endites brown, lighter toward prolateral edge. Labium brown, lighter
anteriorly. Sternum yellow. Abdomen yellow with no markings. First legs yellow;
femora dark brown laterally and ventrally with scattered white scales and dorsal
yellow stripe. Other segments darker ventrally, lighter dorsally. Other legs yellow.
Pedipalps brown mottled with yellow, tips yellow.

Female. — Total length 3.30-4.00. Carapace 1.36-1.50 long, 1.13-1.20 wide, 0.57-
0.62 high at PLE. Ocular area 0.60-0.66 long, 0.82-0.90 wide anteriorly and 0.90-
0.98 wide posteriorly. Chelicerae 0.35-0.45 long 0.24-0.35 wide (four females from
Sonora, Mexico). PME closer to ALE than to PLE. Leg formula 1432. Carapace
orange, black around eyes, lighter laterally. Clypeus with white hairs. Chelicerae
red-brown. Endites and labium red brown. Sternum yellow-brown. Abdomen
yellow, occasionally with faint central stripe. All legs yellow-brown. Pedipalpi
yellow with proximal dorsal dark spots on all segments.

Distribution. — Known only from Sonora, Mexico (Map 1).

Natural history. — Males collected in April and July, females in March and
August.

Specimens examined.— MEXICO: SONORA; Guaymas (AMNH, MCZ), 4-8 mi. S Peon (FSCA), 6
mi. N San Carlos Bay (UCB).

Hentzia chekika, new species
Figs. 80-86, Map 6

Types.— Male holotype and female allotype from Matheson Hammock Park,
Miami, Dade Co., Florida (7 May 1985; G. B. Edwards), deposited in the FSCA.
Male and female paratypes in the FSCA, AMNH and MCZ.

Etymology. — The specific name refers to one of the last chiefs of the Calusa
Indians of south Florida.

Diagnosis. — Hentzia chekika is one of the largest of the Hentzia species;
members of both sexes usually being over 5 mm in length. Males resemble those
of H. grenada, but are larger, lack the loop on the retromargin of the palpal bulb
(Figs. 83, 84), have a distinctive pattern of elongated spots connected with a
central tree-like marking on the dorsal abdomen (Fig. 80), and the epigynum has
longer J-shaped ducts than H. poenitens, but much shorter than those of H.
grenada (Fig. 86).

Male. — Total length 4.35-6.30. Carapace 1.70-2.30 long, 1.25-1.80 wide, 0.65-
1.00 high at PLE. Ocular area 0.70-1.05 long, 1.10-1.45 wide anteriorly and 1.05-
1.45 wide posteriorly. Chelicerae 0.70-2.00 long, 0.33-0.50 wide (10 males from
Dade Co., Florida, and 1 male from Cuba). PME closer to ALE than to PLE.
Leg formula 1423. Carapace yellow-brown with red-brown bands on either side.
Dorsum with narrowing central band of white hairs nearly reaching AME. White
spots of hairs between anterior eyes; otherwise black around eyes except brown
around AME. Lateral white bands followed by black margin around carapace.

Figures 80-86. — Hentzia chekika n. sp.; 80-84, from Dade Co., Fla., 80, male, dorsal view; 81,
female, dorsal view of abdomen; 82, left chelicera of male, ventral view; 83, 84, palp of male; 83,
ventral view; 84, retrolateral view; 85, 86, epigynum of female from Soledad, Cuba; 85, ventral view;
86, dorsal view.

Clypeus covered with white hairs. Chelicerae red to yellow brown, iridescent,
often darker prolaterally. Fang of holotype with distal 1/4 lighter. Endites brown,
holotype with light prolateral 1/2. Labium brown. Sternum orange. Abdomen
yellow-brown with distinctive darker pattern (Fig. 80). Venter yellow. First legs
yellow-brown, dark brown anteroventrally except tarsi and metatarsi. Metatarsi
brown, darker distally. Tarsi yellow. Other legs yellow. Pedipalpi yellow; bulb
brown and proximal brown bands on patellae and tibiae.

Female. — Total length 4.60-7.00. Carapace 1.90-2.40 long, 1.35-1.80 wide, 0.65-
0.90 mm high. Ocular area 0.85-1.00 long, 1.20-1.40 wide anteriorly, 1.20-1.40

RICHMAN— REVISION OF HENTZIA

321

Map 6. — West Indies and Florida, showing distribution of Hentzia chekika (closed circles), H.
zombia (open circles), and H. footei (closed squares).

wide posteriorly. Chelicerae 0.50-0.80 long, 0.40-0.50 wide (11 females from Dade
Co., Florida, and 6 from Cuba). PME closer to ALE than to PLE. Leg formula
1423. Carapace orange with brown lateral area, followed by orange and then a
black margin. White hairs scattered throughout. Black around eyes except brown
around AME. Clypeus covered with white hairs. Chelicerae red-brown. Endites
red-brown, lighter medially. Labium red-brown, lighter anteriorly. Sternum
orange to yellow with dark lateral bands in allotype. Abdomen yellow with
brown pattern (Fig. 71). Venter yellowish. First legs orange with brown prolateral
femora and patellae, dark brown distal band on tibiae and metatarsi, white scales
on proventral femora. Other legs yellow to orange. Pedipalpi yellow-orange with
dark dorsal bands on proximal tibiae, patellae and tarsi.

Distribution. — Florida, Bahamas and Cuba (Map 6).

Natural history. — Males have been collected from March to May, July to
September, November to December. Females from January, March, May to
August and November. Probably both sexes can be collected in any month of the
year. Records which include habitat data indicate that this species is found
primarily on tall feather palms such as Cocos nucifera and Roystonia regia.

Specimens examined. — BAHAMAS: Abaco Cays (AMNH), Andros (AMNH), Freeport (AMNH),
Grand Bahama (AMNH), North Bimini (AMNH), Stirrup Cay (FSCA). CUBA: Soledad (MCZ),
Cienaga de Zapata (MCZ). U.S.A.: FLORIDA; Broward Co., Ft. Lauderdale (FSCA); Dade Co.,
Miami-Doral Country Club (FSCA), Matheson Hammock (FSCA), Kendall (AMNH), Crandon Park
(AMNH), Perrine (AMNH); Lee Co., Sanibel Island (MCZ). Monroe Co., Flamingo (FSCA).

322

THE JOURNAL OF ARACHNOLOGY

VERNALIS SPECIES GROUP

This group takes the name of its most aberrant member by virtue of priority.
Hentzia vernalis appears to have been derived from footei-Ekt ancestors, based
on the structure of the male palpi and chelicerae (Figs. 89-92, 96-99). All
members of this species group have truncated retromarginal teeth in the male
which distinguishes them from all other species in the genus (Figs. 89, 90, 96, 97,
104, 109). Most large males in this species group, with the possible exception of
H. footei, have very elongate chelicerae. The four species in this group, H.
vernalis, H. footei, H. antiliana and H. whitcombi (n. sp.) range from Cuba to
Colombia. The two species pairs within the species group (Fig. 15) can be
separated by the structure of the retromarginal teeth of the males. Both H.
vernalis and H. footei have relatively short, straight, truncated or (rarely)
bifurcate retromarginal teeth (Figs. 89, 90 and 96, 97), whereas H. antillana and
H. whitcombi have angled and truncate retromarginal teeth (Figs. 104, 109). The
species can be further divided by the structure of the epigyna and arrangement of
cheliceral teeth.

Hentzia vernalis (Peckham and Peckham)

Figs. 87-94, Map 4

Anoka vernalis Peckham and Peckham 1893:701 (holotype male from St. Vincent in the MCZ

examined).

H. vernalis Roewer 1954:1218.

Diagnosis. — Males are unlike any other Hentzia in the structure of their
chelicerae (Figs. 89, 90) and their palpi (Figs. 91, 92). They appear to form the
most extreme end of the spectrum for their species group. The females have hair
pencils like other Hentzia, but the epigynum is unlike any other member of the
genus in that there appears to be two tubes on either side (Figs. 93, 94).

Male. — Total length 3.75-4.85. Carapace 1.55-2.00 long, 1.20-1.50 wide, 0.65-
0.80 high at PEE. Ocular area 0.70-0.80 long, 1.05-1.25 wide anteriorly, 1.08-1.30
wide posteriorly. Chelicerae 0.85-1.50 long, 0.35-0.46 wide (7 males from St.
Vincent, Trinidad, Grenada and Barbados). PME closer to ALE than to PEE.
Leg formula 1423. Carapace orange, ocular area yellow with lateral white areas
from posterior to eyes. Eyes surrounded by black, except AME red-brown.
Clypeus covered with white hairs. Chelicerae orange. Endites, labium and
sternum flesh-colored. Abdomen yellow with four vague darker spots. Venter
yellow with two parallel darker lines. First legs orange. Other legs yellow.
Pedipalpi yellow.

Female. — Total length 3.50-5.50. Carapace 1.40-1.90 long, 1.10-1.50 wide, 0.55-
0.70 high at PEE. Ocular area 0.60-0.85 long, 0.95-1.20 wide anteriorly and 1.00-

I. 30 wide posteriorly. Chelicerae 0.30-0.60 long, 0.25-0.50 wide (six females from
St. Vincent and Grenada). AME equidistant between ALE and PEE. Leg formula
1432. Carapace orange, black around eyes except brown around AME. Darker
around lateral areas of carapace and dorsally to midline. Clypeus covered with
white hairs. Chelicerae, endites and labium orange. Sternum yellow. Abdomen
yellow with pattern of speckles followed by band 3/4 way to spinnerets, similar to
that found on many H. antillana females. First legs yellow, slightly darker
dorsally on distal patellae and tibiae. Other legs and pedipalpi yellow.

Figures 87-94. — Hentzia vernatis (Peckham and Peckham): 87, 89, 9F94, from St. Vincent; 87, holotype
male, dorsal outline view; 88, female, dorsal view of abdomen; 89, left chelicera of holotype male,
ventral view; 90, left chelicera of male from Barbados, ventral view; 91, 92, palp of holotype male; 91,
ventral view; 92, retrolateral view; 93, 94, epigynum of female; 93, ventral view; 94, dorsal view.

Distribution.— Northern South America, Trinidad, Barbados, St. Vincent and
Grenada (Map 4).

Natural history. — Males collected in April and August-September (Other
specimens with no dates). Females in February and August-October. Probably
found as adults all year.

324

THE JOURNAL OF ARACHNOLOGY

Specimens examined.— BARBADOS: Barbados (USNMNH), Bridgetown (AMNH). COLOMBIA:
Cienaga (FSCA), Mag. Tasajera (FSCA). GRENADA: Broadway (MCZ), St. Georges (AMNH). ST.
VINCENT: Island record (MCZ), Kingston (AMNH). TRINIDAD: Port-oFSpain (MCZ).

Hentzia footei (Petrunkevitch)

Figs. 95-101, Map 6

Wala footei Petrunkevitch 1914:330 (holotype male from Dominica in AMNH examined).

H. footei Roewer 1954:1217.

Diagnosis. — Males differ from all other Hentzia in having at least one tooth on
the fang, and occasionally having a bifurcate retromarginal tooth (Figs. 96, 97).
The female epigynal structure (Figs. 100, 101) differs from other Hentzia in
having slender downturned tubes and deeply indented openings. In this character
the epigynum is similar to that of Anicius dolius.

Male. — Total length 2.90-4.43. Carapace 1.30-1.89 long, L05-L48 wide, 0.55-
0.82 high at PLE. Ocular area 0.65-0.90 long, 0.93-1.23 wide anteriorly and 0.93-

I. 31 wide posteriorly. Chelicerae 0.42-1.07 long, 0.25-0.49 wide (10 males from
Dominica, Martinique and St. Lucia). PME closer to ALE than to PLE. Leg
formula 1423. Carapace red-brown; lighter scales and white band of hairs
laterally. Black around eyes and margin. Clypeus dark brown. Chelicerae yellow
to red-brown. Endites gray to dark brown with prolateral distal 1/3 cream.
Labium dark brown with distal 1/6 cream. Sternum yellowish. Abdomen yellow-
brown, slightly metallic, with indistinct cross bands (especially just anterior from
spinnerets). Lateral dorsal white bands, followed by gray toward venter. Venter
brown with two narrow posterior gray stripes. First legs yellow to red-brown with
1/2 prolateral and ventral femora, patellae and tibiae dark brown and proximal
and distal patellae and tibiae dark brown. Tarsi sometimes with yellow tip.
Holotype with only indistinct markings on first legs. Other legs yellow. Pedipalpi
yellow with orange bulb with dark band or streak on distal femora.

Female. — Total length 4.70-5.05. Carapace 1.90-2.15 long, 1.48-1.68 wide, 0.75-
0.80 high at PLE. Ocular area 0.85-0.90 long, 1.30-1.40 wide anteriorly and 1.30-
1.45 side posteriorly. Chelicerae 0.50-0.60 long, 0.40-0.45 wide (three females from
St. Lucia). PME closer to ALE than to PLE. Leg formula 1432. Carapace
orange-brown. Black around eyes, except brown around AME. Clypeus covered
with white hairs. Chelicerae orange brown. Endites red-brown with lighter
prolateral anterior edge. Labium red-brown with lighter anterior 1/5. Sternum
yellow. Abdomen yellow with brown central pattern. Anteriorly there is a yellow
streak ending in a claw-like mark on the midline followed by a dark brown
inverted V and three connected triangle decreasing in size ending at the
spinnerets. Lateral to these markings the abdomen is spotted and blotched with
brown. Venter yellow with dark triangular marking anterior to the spinnerets and
black ring around spinnerets. First legs yellow-brown; prolateral brown bands on
distal femora, patellae, tibiae and metatarsi, and proximal tibiae. Other legs
yellow. Pedipalpi yellow.

Distribution. — Dominica, Martinique and St. Lucia in the southern Lesser

Antilles (Map 6).

Natural history — Males from January, June-July and October. Females from
October. Probably found as adults all year.

RICHMAN—REVISION OF HENTZiA

325

I-

0.1

Figures 95-101. — Hentzia footei (Petrunkevitch): 95, male from St. Lucia, dorsal view; 96, 97, left
chelicerae of males; 96, holotype from Dominica, ventral view; 97, Martinique, ventral view; 98, 99,
palp of holotype male from Dominica; 98, ventral view; 99, retrolateral view; 100, 101, epigynum of
female from St. Lucia; 100, ventral view; 101, dorsal view.

Specimens examined.— BRITISH WEST INDIES: St. Lucia (MCZ). DOMINICA: Roseau
(AMNH), Springfield (USNMNH). MARTINIQUE: Fond la Haye (AMNH), Pointe Ferret (AMNH).

326

THE JOURNAL OF ARACHNOLOGY

Hentzia antiliana Bryant
Figs. 102-108, Map 3

Wala vernalis Petrunkevitch 1930:139 (misideetification).

Hentzia antiliana Bryant 1940:494 (holotype male and allotype female from Antigua in MCZ
examined).

Diagnosis. — Males can be distinguished from all other members of the genus
except H. whitcombi n. sp. and some specimens of H. footei by the presence of a
curved, truncated retromarginal tooth on the chelicerae. It can be distinguished
from H. whitcombi by the pattern of promargioal teeth in relationship to the
retromarginal tooth and from H. footei by the general structure of the chelicerae
(Fig. 104). The female epigynal structure has characteristic trumpet-shaped tubes
leading to the spermathecae (Fig. 108).

Male. — Total length 3.60-5.00. Carapace 1.50-2.00 long, L30-L70 wide, 0.70-
LOO high at PLE. Ocular area 0.70-0.90 long, 1.00-1.25 wide anteriorly, 1.15-1.40
wide posteriorly. Chelicerae 1.10-2.00 long, 0.30-0.43 wide (10 males from Virgin
Gorda, British Virgin Islands). PME slightly closer to ALE than to PLE. Leg
formula 1423. Carapace red-brown with white hairs laterally. Dark around eyes.
Ciypeus with white hairs. Chelicerae yellow brown, promargin red-brown; fang
darker. Endites red-brown with lighter inner edge. Labium dark brown. Sternum
yellow-brown. Abdomen golden brown, slightly metallic. Venter and lateral areas
yellow-gray. First legs yellow-brown; darker under prolateral femora, patellae and
tibiae. Indistinct bands on distal patellae, tibiae and metatarsi. Other legs yellow.
Pedipalpi red-brown; cymbium and bulb darker.

Female. — Total length 4.05-5.50. Carapace 1.50-2.10 long, 1.25-1.70 wide, 0.60-
0.90 high at PLE. Ocular area 0.70-0.80 long, 1.00-1.30 wide anteriorly and 1.10-
1.45 wide posteriorly. Chelicerae 0.45-0.80 long, 0.30-0.50 wide (10 females from
Virgin Gorda, British Virgin Islands). PME closer to ALE than to PLE. Leg
formula 1423. Carapace red-brown with lateral white hairs. Eyes dark. Ciypeus
with white hairs. Chelicerae red-brown. Endites red-brown with yellow prolaterai
tip. Labium red-brown. Sternum yellow-brown. Abdomen yellow with red-brown
markings. A posterior dark band (also seen in H. vernalis and H. squamata
females) is almost always present. First legs red-brown. Other legs yellow.
Pedipalpi yellow; annulate at proximal joints. -
Distribution. — West Indies from Dominica to Cuba (Map 3).

Natural history. — Males and females have been collected in every month but
December. It is probably found in all months as adults. In Puerto Rico the
species was common in citrus groves in May and was also occasionally swept
from tall grass. In Guadeloupe they were abundant in citrus in February. It is
sympatric with H. whitcombi over much of its range.

Specimens examined. — ANTIGUA: Antigua (MCZ, USNMNH), Crosbies (MCZ), St. Johns
(AMNH), Shirley Heights (MCZ). BARBUDA: Codrington (AMNH). BRITISH VIRGIN
ISLANDS: Amegada (AMNH, USNMNH), Beef Island (AMNH), George Dog Island (AMNH),
Ginger Island (AMNH), Greater Thatch Island (AMNH), Green Cay (AMNH), Guana Island
(AMNH, USNMNH), Little Comonoe (AMNH), Little Thatch Island (AMNH), Little Tobago
(AMNH), Sandy Key (AMNH), Tortola (AMNH, MCZ, USNMNH), Virgin Gorda (AMNH, MCZ).
BRITISH WEST INDIES: St. Kitts (MCZ), St. Lucia (MCZ), St. Nevis (MCZ). CUBA: Ceiba
(AMNH). DOMINICA: country record (MCZ), Fond Sophie (AMNH), Roseau (AMNH);
GUADELOUPE: Maire Galant (FSCA), Petit Bourg (FSCA), Pointe-a-Pitre (AMNH). HAITI: Port-
au-Prince (MCZ). LEEWARD ISLANDS: Saint Maarten (AMNH), Saba (MCZ). MARTINIQUE:

RICHMAN— REVISION OF HENTZIA

327

Figures 102-108. — Hentzia antiilana Bryant; 102, 103, from Isabela, Pureto Rico; 102, male, dorsal
view; 103, female, dorsal view of abdomen; 104-106, holotype male from Antigua; 104, left chelicera,
ventral view; 105, 106, palp; 105, ventral view; 106, retrolateral view; 107, 108, epigynum of female
from Virgin Gorda, British Virgin Islands; 107, ventral view; 108, dorsal view.

Fond la Haye (AMNH), Pointe Ferret (AMNH) Trois Ilets (AMNH). MONTSERRAT: Plymouth
(AMNH). PUERTO RICO: Aibonito (AMNH), Arecibo (AMNH), Barros (AMNH), Blanquilla
(AMNH), Cabeza de Parro Island (AMNH), Cambalache Forest (AMNH), Camuy (AMNH), Cana
Gorda (DBR), Cayo Ahogado (AMNH), Cayo Caracoles (AMNH), Cayo San Cristobal (AMNH),
Chicken Island (AMNH), Coamo Springs (AMNH), Corozal (AMNH), Desecheo Island (AMNH),
Isabela (DBR), Isla Palominires (MCZ), Isla Ramos (AMNH), Isleta Marina (AMNH), Laguna

328

THE JOURNAL OF ARACHNOLOGY

Cartegena (MCZ), La Parquera (MCZ), Loma Tinaja (AMNH), Luquillo Mountains (MCZ), Manati
(AMNH), Maricao Bosque National (MCZ), Mayaguez (AMNH, DBR, MCZ), Muertes Island
(AMNH), Nicacos Island (MCZ), Quebradilia (AMNH), Pico Atalaya (AMNH), Pineros Island
(AMNH), Ratones Island (AMNH), Rio Piedras (MCZ), San Juan (AMNH, MCZ), Vega Baja
(AMNH), Vivevero de Catalina (AMNH). UNITED STATES VIRGIN ISLANDS: Anegada
(AMNH), Hassel Is. (AMNH), St. Croix (MCZ, USNMNH), St. John (FSCA, MCZ, USNMNH),
St. Thomas (MCZ).

Hentzia whitcombi, new species
Figs. 109413, Map 5

Types. — Holotype male and allotype female from Petit Bourg, Basse Terre,
Guadeloupe (25 February 1977. W. H. Whitcomb) deposited in the FSCA.

Etymology. — The species is named for Dr. W, H. Whitcomb, Professor
Emeritus, University of Florida, who collected the type series during his study on
the natural enemies of citrus pests.

Diagnosis. — Males differ from all other Hentzia except H. antillana in having
long chelicerae with a curved truncated retromarginal tooth. They differ from H.
antillana males in the arrangement of the cheliceral teeth (Fig. 109). The female
epigynum differs from that of H. antillana by having round openings (Figs. 112,
113).

Males. — Total length 3.20-4.90. Carapace L50-1.90 long, 1.20-1.50 wide, 6.50-
8.50 high at PLE, Ocular area 0.70-0.92 long, LOO- 1.22 wide anteriorly and 1.00-
1.28 wide posteriorly. Chelicerae 0.65-1.70 long, 2.40-4.00 wide (10 males from
Guadeloupe). PME closer to ALE than to PLE. Leg formula 1423. Carapace red-
brown with central dorsal spearhead-like light marking with point toward
posterior. Lateral areas with white scales. Black around eyes except dark brown
around AME. Carapace with scattered iridescent scales. Clypeus brown with
white hairs. Chelicerae red brown, darker proiaterally, with fringe of white hairs.
Distal 1/5 of fang yellowish. Endites dark brown, tip pale. Labium dark brown,
tip pale. Sternum orange. Abdomen dark brown with darker markings, similar to
those of H. antillana. Dorsum with iridescent scales. Lateral area with pattern of
white and brown stripes. Venter light brown. First legs yellow; dark brown
proiaterally and ventrally. Other legs yellow. All legs with iridescent scales.
Pedipalpi yellow; bulb and cymbium brown.

Female. — Total length 3.80-5.40. Carapace 1.70-2.00 long, 1.30-L60 wide, 0.70-
0.80 high at PLE. Ocular area 0.70-0.85 long, 1. 15-1.30 wide anteriorly and 1.20-
1.40 wide posteriorly. Chelicerae 0.40-0.60 long, 0.30-0.45 wide (10 females from
Guadeloupe). PME closer to ALE than to PLE. Leg formula 1423. Carapace
orange; white scales laterally. Black around eyes except brown around AME.
Clypeus with white hairs. Chelicerae orange-brown. Endites dark brown,
prolateral tip pale. Labium dark brown with pale tip. Sternum yellow. Abdomen
yellow with brown markings like H. antillana. Venter yellow. First legs yellow
with light brown markings on prolateral distal femora, patellae and tibiae, and
retrolateral femora. Other legs yellow. Pedipalpi yellow with brown dorsal
proximal markings on patellae, tibiae and tarsi.

Natural history. — Males have been collected in Jaeuary-April and Juee-July.
Females have been collected from January-March and May-August. They
probably occur throughout the year.

RICHMAN— REVISION OF HENTZIA

329

Figures 109-113. — Hentzia whitcombi n. sp. from Basse Terre, Guadelopue: 109, left chelicera of
male, ventral view; 110, 111, palp of male; 110, ventral view; 111, retrolateral view; 112, 113,
epigynum of female; 112, ventral view; 113, dorsal view.

Specimens examined.— BRITISH VIRGIN ISLANDS: Peter Island (AMNH). BRITISH WEST
INDIES: St. Kitts (MCZ). DOMINICA: Portsmouth (AMNH), Roseau (USNMNH). GUADE-
LOUPE: Pointe-a-Pitre (AMNH), Petit Bourg (FSCA), Lamentin (FSCA). MONTSERRAT: Galways
Estate (AMNH). PUERTO RICO: Aguadilla (USNMNH), Culebrita Island (AMNH), La Pauquera
(MCZ), Levin’s Rock (AMNH), Maricao (MCZ), Maracao National Forest (AMNH), Mayaguez
(AMNH, MCZ), Rio Piedras (AMNH), Toro Negro (MCZ). UNITED STATES VIRGIN ISLANDS:
St. John (FSCA), St. Thomas (MCZ).

330

THE JOURNAL OF ARACHNOLOGY

VITTATA SPECIES GROUP

This group contains two species formerly placed in the genera Parahentzia
Bryant and Maeviobeata Caporiacco. It also contains the Hentzia or Wala species
squamata and vittata, as well as two new species, calypso and zombia, for a total
of six species. All the known males have characteristic cheliceral teeth, with a
spike-like retromarginal tooth and usually two smaller promarginal teeth (Figs.
115, 116, 125, 132, 139, 148, 149). Most males in this species group have robust
chelicerae and are less elongate than other Hentzia species. Hentzia calypso is one
of the smallest species of the genus, but the rest are larger and comparable to
most other Hentzia species in size. The species range from the Bahamas and
Cuba to northern South America and Central America.

The vittata group part of the cladogram (Fig. 15) is more or less arbitrary in
that H. zombia is known only from females, H. squamata is somewhat aberrant
in that it has males with long chelicerae, and H. calypso is more elongate and
much smaller than other species in the group. In addition, H. mandibularis males
often have a strange tubercle on the dorsal chelicerae. The only two species that
can be easily related, other than by the characters found in the group as a whole,
are H. vittata and H. parallela. The males of these are difficult to separate and it
is possible that they represent variations of one widespread species. At present I
am treating them as separate species, but I would not be totally surprised if they
turned out to be synonyms, although the females seem to exhibit some differences
that may separate these similar species (see species descriptions).

Hentzia vittata (Keyserling)

Figs. 114-122, Map 7

Icius vittatus Keyserling 1885:504 (holotype female from “United States” in MCZ examined).

Wala albovittata Keyserling 1885:517 (holotype male from “United States’ in MCZ examined). JSEW
SYNONYMY.

Anoka peckhami Cockerell 1893:221 (holotype and paratypes from Jamaica in MCZ examined). NEW
SYNONYMY.

Anoka moneagua Peckham and Peckham 1894:127 (holotype and paratypes in MCZ examined).
NEW SYNONYMY.

Hentzia peckhami Bryant 1943:488. NEW SYNONYMY.

Diagnosis. — Males differ from all other Hentzia males except for H. parallela
and H. calypso in the structure of the chelicerae (Figs. 115, 116). They differ from

H. calypso in size and pattern. H. vittata and H. parallela are very close and can
be distinguished primarily by the female epigyna of H. vittata (Figs. 119-122),
which have wider spermathecal areas than either H. parallela or H. calypso.

Male. — Total length 2.95-5.17. Carapace 1.39-2.21 long, 1.23-1.80 wide, 0.66-

I. 15 high at PLE. Ocular area 0,66-0.90 long, 1.07-1.31 wide anteriorly, 1,15-1.48
wide posteriorly. Chelicerae 0.41-1.23 long, 0.37-0.51 wide (11 males from
Jamaica plus the holotype of Wala albovittata from “United States”). PME closer
to ALE than to PLE. Leg formula 1423. Carapace red-brown; white hairs
laterally. Eyes surrounded by black except brown around AME. Clypeus covered
with white hairs. Chelicerae red-brown, metallic. Endites and labium dark red-
brown, sometimes with lighter tips. Sternum light red-brown to orange. Abdomen
orange to golden brown, often with slight indication of markings dorsally and

RICHMAN— REVISION OF HENTZIA

331

115

Figures 114-122 — Hentzia vittata (Kerseriieg): 114, female from Torrington, Jamaica, dorsal view of
abdomen; 115, left chelicera of male (holotype of Anoka peckhami Cockerell) from Jamaica, ventral
view; 116, left chelicera of male from Freeport, Grand Bahama, ventral view; 117, 118, palp of male
(holotype of Anoka rnoneagua Peckham and Peckham) from Moneague, Jamaica; 117 ventral view;
118, retrolaterai view; 119-122, epigyna of females; 119, 120, from Jamaica; 119, ventral view; 120,
dorsal view; 121, 122, from Cuba; 121, ventral view; 122, dorsal view.

332

THE JOURNAL OF ARACHNOLOGY

I:

Map 7. — West Indies and Central America, showing distribution of Hentzia vittata (closed circles)
and H. parallela (closed squares).

sometimes with lateral lighter bands. Lateral area darker; venter gray-brown to
brown. First legs red-brown, darker ventrally and laterally. Tarsus orange. Other
legs yellow-brown to yellow. Coxae yellow. Pedipalpi red-brown, femora and
patellae sometimes yellow.

Female. — Total length 3.44-5.00. Carapace 1.48-1.89 long, 1.23-1.72 wide, 0.65-
0.82 high at PLE. Ocular area 0.74-0.98 long, 0.98-1.30 wide anteriorly and 1.15-
1.39 wide posteriorly. Chelicerae 0.40-0.57 long, 0.30-0.41 wide (10 females from
Jamaica, Haiti and the holotype of Icius vittatus from ‘‘United States”). PME
closer to ALE than to PLE. Leg formula 1432. Carapace yellow to orange-brown
with lateral white hairs. Eyes surrounded by black, except brown around AME.
Glypeus white to yellow. Chelicerae light red-brown to orange. Endites and
labium red-brown to orange, lighter distally. Sternum red-brown to yellow-
brown. Abdomen yellow with either four dark brown spots dorsally followed by a
dark cross-band or with a pattern of two brown spots anteriorly, followed
posteriorly by two to three triangular to square markings, ending at spinnerets.
Venter yellow to yellow brown. First legs yellow-brown to orange. Other legs
yellow to yellow-brown. Pedipalpi yellow with darker proximal dorsal spots on
all but femora.

Distribution. — Bahamas, Cuba, Hispaniola and Jamaica (Map 7).

RICHMAN— REVISION OF HENTZIA

333

Natural history. — Males have been collected in January, March-April, June-
July and October-December. Females in January-June and August-December.
Adults are present throughout the year.

Specimens examined. — BAHAMAS: Abaco Cays (AMNH), Andros (AMNH), Cat Island
(AMNH), Elbow Cay (AMNH), Grand Bahama (AMNH), Great Exuma (MCZ), Harbor Island
(MCZ), Little Harbor Cay (AMNH), Lucaya (AMNH), Nassau (AMNH), New Providence Island
(AMNH), North Bimini (AMNH), South Bimini (AMNH), Stirrup Cay (FSCA), West Caicos
(AMNH). CUBA: Buenos Aires (MCZ), San Bias (MCZ), Soledad (MCZ), 7 km N Vinales (AMNH).
DOMINICAN REPUBLIC: Bara Hora (MCZ), Boca Caica (AMNH), La Matica (AMNH), Largo
Enriquillo (FSCA), La Vega (MCZ), Las Waitas (MCZ), Nisibon (FSCA), Puerto Plata (MCZ),
Santo Domingo (AMNH), S of Santiago (MCZ). HAITI: Carrefour (AMNH), Diguini (MCZ),
Formond (FSCA), Grand Anne (MCZ), Grand Riviere (MCZ), Kenscoff (AMNH), La Boule
(AMNH), La Hoatte (MCZ), Les Anglais (FSCA), Port-au-Prince (AMNH, MCZ), Post Terre Rouge
(MCZ). JAMAICA: Christina (AMNH, MCZ), Discovery Bay (AMNH), The Great Morass
(AMNH), Hanover Askenish (MCZ), Holland Bay (AMNH), James (MCZ), Kingston (AMNH,
MCZ), Manderville (MCZ), Moneague (MCZ), 2.5 mi. E Ocho Rios (MCZ), Port Antonis (MCZ),
Port Henderson (MCZ), Portland (MCZ), Robin’s Bay (MCZ), Roundhill (AMNH), St. Andrew
(MCZ), St. Ann (MCZ), St. Catherine (MCZ), St. Elizabeth (MCZ), St. Mary (MCZ), St. Thomas
(MCZ), Unity Valley (AMNH), Westmoreland (MCZ), Whitehouse (AMNH).

Hentzia parallela (Peckham and Peckham)
Figs. 123429, Map 7

Anoka parallela Peckham and Peckham 1894:129 (holotype male from Trinidad in MCZ examined).
Wala parallela Petrunkevitch 1911:717.

Parahentzia insignita Chickering 1946:316 (holotype male and allotype female from Panama Canal
Zone in MCZ examined). NEW SYNONYMY.

Maeviobeata charitonovi Caporiacco 1947:34; 1948:734 (holotype male and paratype male and female
from Guyana in the Museo Zoologico “La Specola,” Florence, Italy, examined by M. E. Galiano -
synonymy based on her drawings). NEW SYNONYMY.

H. parallela Roewer 1954:1218.

Diagnosis. — Males differ from all other Hentzia except H. vittata in having
massive chelicerae with a slanted, spike4ike retromarginal tooth (Fig. 125). They
differ from H. calypso in size dorsal pattern and distribution. Female epigynum
(Figs. 128, 129) diagnostic, with trumpet-like openings and rectangular

spermathecal area.

Male. — Total length 3.95-5.15. Carapace 1.60-2.07 long, 1.35-1.85 wide, 0.70-

I. 06 high at PLE. Ocular area 0.70-0.90 long, 1.05-1.42 wide anteriorly, LI 5-1. 53
wide posteriorly. Chelicerae 0.50-1.00 long, 0.35-0.60 wide (10 males from
Boquete, Panama and 8 from Canal Zone, Aruba, Curasao and Trinidad. Total
18). PME closer to ALE than to PLE. Leg formula 1423. Carapace red-brown;
band of white hairs laterally, often followed by a wide black band to margin.
Black around eyes, except brown around AME. Clypeus red-brown covered with
white hairs. Chelicerae red-brown. Endites and labium, red-brown, often with
lighter distal tips. Sternum red-brown. Abdomen orange to red-brown, often
without markings or with four dark spots dorsally followed by two bands just
anterior to spinnerets. There is occasionally a lighter longitudinal band on each
side of the dorsum. Darker lateral, areas. Venter gray to brown, in at least one
case with three rows of small light dots. First legs red-brown, darker on
prolateral surface of femora, patellae and tibiae. Other legs yellow. Pedipalpi
yellow to red-brown, cymbium slightly darker.

334

THE JOURNAL OF ARACHNOLOGY

129

Figures 1 23- 1 29. — Hentzia parallela (Peckham and Peckham): 123, 124, from Fort Randolph, Canal
Zone, Panama; 123, male, dorsal view; 124, female, dorsal view of abdomen; 125, left chelicera of
male (holotype of Parahentzia insignita Chickering) from Canal Zone, Panama, ventral view; 126,
127, palp of male from Oranjstad, Aruba; 126, ventral view; 127, retrolateral view; 128, 129, epigynum
of female from Panama; 128, ventral view; 129, dorsal view.

Female. — Total length 3.75-5.70. Carapace 1.70-1.95 long, 1.30-1.60 wide, 0.65-
0.77 high at PLE. Ocular area 0.65-0.85 long, 1.05-1.30 wide anteriorly and 1.15-
1.39 wide posteriorly. Chelicerae 0.35-0.50 long, 0.30-0.41 wide (12 females from
Panama and Canal Zone). PME closer to ALE than to PLE. Leg formula 1423.

RICHMAN— REVISION OF HENTZIA

335

Carapace orange with darker paired dorsal posterior and lateral longitudinal
bands. Black around eyes, except brown around AME. Clypeus orange.
Chelicerae orange. Endites and labium orange, lighter on distal tips. Sternum
orange. Abdomen yellow with two longitudinal rows of four brown spots each,
followed by a thin band of thick dark brown spots just anterior to the spinnerets.
All legs yellow-orange. Pedipalpi yellow, unniarked.

Distribution. — Trinidad, Guyana, Aruba, Curasao, Costa Rica, Honduras and
Panama (Map 7).

Natural history. — Males have been collected in January, June-August and
November-December. Females from June-August and November. They are
probably found as adults all year.

Specimens examined. — ARUBA: Oranjested (AMNH). COSTA RICA: Monteverde (AMNH).
CURASAO; Piscadera Bay (AMNH). HONDURAS: Lancetilla (MCZ). PANAMA: Boquete (MCZ),
Canal Zone (Barro Colorado) (MCZ), El Valle (AMNH). TRINIDAD: East Coast (MCZ), Piarco
(AMNH).

Hentzia calypso, new species
Figs. 130-136, Map 5

Types. — Male holotype from 2 mi. SW Unity Valley, Jamaica (18 March 1955;
A. M. Nadler), deposited in the AMNH. Male and female paratypes from
Jamaica deposited in the AMNH and the MCZ.

Etymology. — The specific name refers to the sea nymph in Homer’s Odyssey
and also a style of music found in the southern Caribbean.

Diagnosis. — Hentzia calypso is one of the smallest species of the genus, males
being less than 4.0 mm and females from 3.5 to 4.5 mm in length. The structure
of the male chelicerae is similar to that of H. vittata, but the body is much more
elongate and the dorsal pattern is distinctive (Figs. 130, 131). Female epigynum
diagnostic, with large openings and oval spermathecae internally (Figs. 135, 136);
externally similar to those of H. chekika and H. poenitens (Fig. 135). It differs
from the former species in size, internal structure of epigynal tubes, pattern and
distribution and from the latter in internal structure of epigynum, pattern and
distribution.

Male. — Total length 3.15-3.90. Carapace 1.25-1.50 long, 0.92-1.12 wide, 0.50-
0.60 high at PLE. Ocular area 0.55-0.70 long, 0.80-0.95 wide anteriorly and 0.82-
0.95 wide posteriorly. Chelicerae 0.25-0.50 long, 0.23-0.32 wide (four males from
Jamaica). PME slightly closer to PLE than to ALE. Leg formula 1423. Carapace
red-brown with white lateral bands. Black around eyes except AME, which is
dark brown. Clypeus red-brown. Chelicerae dark brown. Endites dark brown
with prolateral 1/3 lighter. Labium dark brown with lighter anterior 1/3.
Sternum red-brown. Abdomen dark red-brown with darker squarish spots and
white lateral stripes dorsally (Fig. 130). Venter gray-brown. First legs yellow-
brown with darker brown on prolateral and ventral femora, patellae and tibiae.
Darker bands on distal patellae, distal and proximal tibiae and distal metatarsus.
Tarsus yellow. Other legs yellow. Pedipalpi with yellow cymbium, dark brown
patellae and tibiae, and femur with yellow dorsally, dark brown prolaterally.

Female. — Total length 3.50-4.45. Carapace 1.50-1.65 long, 1.10-1.30 wide, 0.55-
0.70 high at PLE. Ocular area 0.60-0.70 long, 0.95-1.00 wide anteriorly and 0.95-
1.10 posteriorly. Chelicerae 0.25-0.40 long, 0.25-0.30 wide. PME closer to PLE

336

THE JOURNAL OF ARACHNOLOGY

Figures 130-136. — Hentzia calypso n. sp.; 130-134, from Unity Valley, Jamaica; 130, male, dorsal
view; 131, female, dorsal view of abdomen; 132, left chelicera of male, ventral view; 133, 134, palp of
male; 133, ventral view; 134, retrolateral view; 135, 136, epigynum of female from Roundhill, Jamaica;
135, ventral view; 136, dorsal view.

than to ALE. Leg formula 1423. Carapace orange, black around eyes except
AME; brown around AME; darker lateral to midline, lighter along margin.
Clypeus covered with white hairs. Chelicerae, eedites and labium orange.
Sternum yellow. Abdomen yellow with dark brown markings as in male. Streaks
laterally forming rough stripes. Venter yellow. First legs yellow with dark stripe
on prolateral femora and prolateral distal tibiae. Spots on retrolateral distal and
proximal tibiae, distal femora and patellae. Other legs and pedipalpi yellow.

Distribution. — Known only from Jamaica (Map 5).

Natural history. — Males from March and November. Females from July and
November. Probably adults can be collected all year.

Specimens examined. — JAMAICA: Buff Bay (MCZ), Manderville (FSCA), New Castle (MCZ),
Roundhill (AMNH), St. Andrew (MCZ), St. Catherine (AMNH).

RICHMAN— REVISION OF HENTZIA

337

Figures 137-141. — Hentzia mandibularis (Bryant): 137, 138, from Dept. L’Oeste, Haiti; 137, male
from Morne d’Enfer, dorsal view: 138, female from Parc National de la Vicite, dorsal view of
abdomen; 139, left chelicera of male from Port-au-Prince, Haiti, ventral view; 140, 141, holotype male
from the Dominican Republic; 140, left chelicera, dorsal view, showing tubercle; 141, anterior
sternum, ventral view showing depressions.

Hentzia mandibularis (Bryant)

Figs. 137445, Map 4

Parahentzia mandibularis Bryant 1943:500 (holotype male from foothills of Cordillera Central S of
Santiago, Dominican Republic, in MCZ examined).

Diagnosis. — Although H. mandibularis is atypical in that males may have a
pronounced tubercle on the dorsal chelicerae, it is certainly a Hentzia and is
related to H. vittata. They may also have the depressions in the sternum
mentioned by Bryant, or they may lack them. The males differ from other

Figures 142-145. — Hentzia mandibularis (Bryant): 142, 143, palp of holotype male; 142, ventral
view; 143, retrolateral view; 144, 145, epigynum of female from La Decouverte, Haiti; 144, ventral
view; 145, dorsal view.

Hentzia males with thick and short chelicerae by the position and size of the
retromarginal tooth (Fig. 139) the presence in some males of a dorsolateral
tubercle on the chelicerae (Fig. 140), the narrow labium and the presence in some
males of two depressions in the anterior sternum (Fig. 141). The structure of the
epigynum, with circular openings, is diagnostic for females (Figs. 144, 145).

Male. — Total length 3.70-5.17. Carapace 1.55-2.13 long, 1.20-2.05 wide, 0.70-
1.15 high at PLE. Ocular area 0.75-0.90 long, 1.00-1.31 wide anteriorly and 1.18-
1.48 wide posteriorly. Chelicerae 0.50-1.15 long, 0.30-0.57 wide (seven males from
Haiti and the Dominican Republic). PME closer to ALE than to PLE. Leg
formula 1423. Carapace red-brown with white bands laterally. Eyes surrounded
by black, except AME, surrounded by dark brown. Clypeus covered with white

RICHMAN— REVISION OF HENTZIA

339

hairs. Chelicerae red to orange-brown. Endites, labium and sternum red to
orange-brown with two indentations on anterior sternum (Fig. 141). Abdomen
red to orange-brown with white dorsolateral bands. Lateral area gray-brown;
venter red to gray-brown. Scattered iridescent scales and often with darker
markings in central dorsal portion. First legs red-brown, darker anteriorly and
ventrally. Tarsus lighter. Other legs whitish to orange. Pedipalpi red-brown, bulb
and cymbium darker.

Female. — Total length 4.15-4.84. Carapace 1.71-1.80 long, 1.35-1.40 wide, 0.70-
0.77 high at PLE. Ocular area 0.77-0.80 long, 1.06-1.10 wide anteriorly and 1.20-
1.25 wide posteriorly. Chelicerae 0.50-0.90 long 0.30-0.35 wide (three females
from Haiti). PME closer to ALE than to PLE. Leg formula 1423. Carapace red-
brown with scattered black hairs, white scales laterally and iridescent scales
dorsally, especially between eyes. Clypeus covered with white hairs. Chelicerae
orange-brown, without tubercles. Endites dark brown with pale tips. Labium dark
brown. Sternum red-brown. Abdomen red-brown with white streaks and spots
and lateral bands. Lateral area red-brown, venter whitish. First legs yellow-
brown, darker ventrolaterally on femur and on distal tibia, patella and
metatarsus. Metatarsus-tarsus generally yellow. Other legs white. Pedipalpi
yellow; proximal brown spots on dorsal tibia, patella and tarsus.

Distribution. — Known only from the island of Hispaniola (Map 4).

Variation. — Specimens may have or lack distinct depressions in the sternum.
Males may have dorsal cheliceral tubercles or may lack them, or have only traces.

Natural history, — Males have been collected in March and May-July, Females
are known only from March and May.

Specimens examined.— DOMINICAN REPUBLIC: Foothills Cordillera Central S of Santiago
(MCZ). HAITI: Dept. L’Oeste,; Parc National de la Vicite (FSCA), Morne d’Enfer (FSCA), Kenscoff
(AMNH), Port-au-Prince (MCZ).

Hentzia squamata (Petrunkevitch)
Figs. 146-153, Map 5

Wala squamata Petrunkevitch 1930:146 (holotype and paratypes from Mona Island in AMNH

examined).

Hentzia squamata Bryant 1941:93.

Diagnosis. — Males differ from all other Hentzia in having white scales nearly
covering chelicerae (Fig. 146). Females resemble females of H. antillana closely
and can be easily separated only by association with males. Males also differ
from males of H. antillana in the arrangement and structure of the cheliceral
teeth (Figs. 148, 149), which place it in the vittata species group.

Male. — Total length 3.69-5.44. Carapace 1.77-2.48 long, 1.48-2.12 wide, 0.89-
1.18 high at PLE. Ocular area 0.83-1.06 long, 1.21-1.53 wide anteriorly and 1.30-
1.59 wide posteriorly. Chelicerae 0.59-2.12 long, 0.44-0.65 wide (eight males from
Mona Island). PME closer to ALE than to PLE. Leg formula 1423. Carapace
red-brown, yellowish in ocular area and with lateral white scales. Eyes ringed
with black, Clypeus covered with white hairs. Chelicerae yellow-brown, covered
with pearly scales. Endites and labium yellow-brown with light distal tips.
Sternum yellow-brown. Abdomen yellow-brown with darker markings (Fig. 146)
and with light dorsolateral stripes, followed by dark lateral stripes. Venter yellow-

340

THE JOURNAL OF ARACHNOLOGY

155

Figures I46-I53 — Hentzia squamata (Petrunkevitch) from Mona Island: 146, male, dorsal view; 147,
female, dorsal view of abdomen; 148, left chelicera of male, ventral view; 149, left dielicera of another
male, ventral view; 150, 151, palp of holotype male; 150, ventral view; 151, retrolateral view; 152, 153,
epigynum of female; 152, ventral view; 153, dorsal view.

Figures 154-155. — Hentzia zombia n. sp., epigynum of female from Roche Platte, Haiti; 154, ventral
view; 155, dorsal view.

RICHMAN— REVISION OF HENTZIA

341

brown. First legs yellow-brown. Other legs lighter yellow-brown. Pedipalpi yellow
brown with dark dorsal spot on tibiae.

Female. — Total length 4.13-6.67. Carapace 1.89-2.48 long, 1.59-2.18 wide, 0.83-
1.06 high at PLE. Ocular area 0.83-1.06 long, 1.24-1.53 wide anteriorly and 1.36-
1.77 wide posteriorly. Chelicerae 0.53-0.89 long, 0.35-0.59 wide (11 females from
Mona Island), PME closer to ALE than to PLE. Leg formula 1423. Carapace
red-brown, yellow in ocular area. Eyes ringed in black except AME ringed in
brown. Clypeus covered with white hairs. Chelicerae red or yellow-brown. Endites
and labium red-brown with lighter distal tips. Sternum yellow-brown. Abdomen
yellow with brown markings much like H. antillana (Fig. 147). Venter yellow.
First legs yellow-brown. Other legs lighter yellow-brown. Pedipalpi yellow with
dorsal proximal spots except on femora.

Distribution. — Known only from Mona Island (Map 5).

Natural history. — Males collected in February, April and August. Females in
August. Probably found all year, but the island is little collected because of
difficulties of access.

Specimens examined. — PUERTO RICO: Mona Island (AMNH, MCZ).

Hentzia zombia, new species
Fig. 154, 155, Map 6

Types. — Female holotype from Morne d’Enfer, Dept. L’Oeste, Haiti (5 May
1984, M. C. Thomas) deposited in the FSCA.

Etymology. — The specific name refers to the “living dead” of West Indian
voodoo.

Diagnosis. — While females have the major characteristics of the genus — hair
pencils, spatulate hairs on femora and external genitalia similar to several
described species, such as H. fimbriata (Fig. 154), it is easily distinguished from
all other species by its large size (about 7 mm), stocky build and covering of pink
iridescent scales. The internal epigynal structure (Fig. 155) is unlike any other
Hentzia. It is at present difficult to place this species in a species group, but it is
assumed to be related to H. vittata because of its general body shape and the
structure of the external epigynum.

Female. — Total length 6.60-7.10. Carapace 2.60-2.90 long, 2.20-2.45 wide, 1.20
high at PLE. Ocular region 1.15-1.20 long, 1.45-1.60 wide anteriorly and 1.60-1.80
wide posteriorly. Chelicerae 0.90-1.00 long, 0.70-0.80 wide (two females from
Haiti). PME closer to ALE than to PLE. Leg formula 1423. Carapace dark
brown. Eye region dark with black around all eyes except dark brown around
AME. Clypeus dark brown. Chelicerae dark red-brown. Endites and labium dark
brown with white tips. Sternum orange. Coxae orange. Abdomen dark brown to
yellow, with darker markings consisting of four dark patches on the posterior
dorsal two-thirds and two dark streaks just anterior to spinnerets which nearly
join at midline. Venter brown. First legs brown; lateral femora darker; tarsus
yellow. Other legs yellow. Pedipalpi dark brown; tarsi lighter on distal 2/3. Body
covered with iridescent pink scales.

Distribution. — Known only from Haiti (Map 6).

Natural history. — Collected in February and May.

Specimens examined. — HAITI: Morne d’Enfer (FSCA), Roche Platte (FSCA).

342

THE JOURNAL OF ARACHNOLOGY

ACKNOWLEDGMENTS

I am especially thankful for the help of Wayne Maddison, who made the
habitus drawings and offered a number of helpful suggestions as well as providing
the McClade® computer program for cladistic analysis. Suggestions were also
offered by Bruce Cutler, G. B. Edwards, Carol Sutherland and J. G. Watts. The
present study is based largely on the collections of the American Museum of
Natural History (AMNH), The Museum of Comparative Zoology (MCZ), the
Florida State Collection of Arthropods (FSCA), and the United States National
Museum of Natural History (USNMNH). I am indebted to Norman 1. Platnick
(AMNH), H. W. Levi (MCZ), G. B. Edwards (FSCA) and Jonathan Coddington
(USNMNH) for access to this material. Additional material was lent by Charles
Griswold (Essig Museum, University of California at Berkeley), D. A. Dean
(Texas A&M University), Charles Dondale (Canadian National Collection), W. B.
Peck (H. Exline-Peck Collection), Paul Hillyard and Fred Wanless (British
Museum-Natural History), Carlos Valerio (Universidad de Costa Rica), and
Donald Azuma (The Academy of Natural Sciences, Philadelphia). At the MCZ I
would also like to thank John M. Hunter and Laura Leibensperger. I am
indebted to Maria Elena Galiano, Museo Argentine Ciencias Naturales
“Bernardino Rivadavia” in Buenos Aires, for providing drawings of the type of
Maeviobeata charitonovi. Other people who helped in various ways include Jon
Reiskind (University of Florida), W. H. Whitcomb (University of Florida), Ellis
W. Huddleston, John Owens, Grant Kinzer, David Thompson, Brad Lewis, and
James Zimmerman (New Mexico State University). I would like to especially
thank my wife, Lynda Goin, for her encouragement in this project.

This is Journal Article 1424 of the Agricultural Experiment Station, New
Mexico State University, Las Cruces, NM.

LITERATURE CITED

Banks, N. 1904. The Arachnida of Florida. Proc. Acad. Sci. Philadelphia, 56:120-147.

Bryant, E. B. 1940. Cuban spiders in the Museum of Comparative Zoology. Bull. Mus. Comp. Zool.
Harvard, 86:247-554.

Bryant, E. B. 1943. The salticid spiders of Hispaniola. Bull. Mus. Comp. Zool. Harvard, 92:445-522.

Bryant, E. B. 1947. A list of spiders from Mona Island, with descriptions of new or little-known
species. Psyche, 54:86-99.

Caporiacco, L. di. 1947. Diagnosi preliminari delle species di Arachnidi della Guiana Britannica
raccolte da Beccari e Romini. Monitore Zool. ItaL, 56:20-34.

Caporiacco, L. di. 1948. Arachnida of British Guiana collected by Prof. Beccari. Prop. Zool. Soc,
London, 118:607-747.

Chamberlin, R. V. 1916. Results of the Yale Peruvian Expedition of 1911. The Arachnida. Bull. Mus.
Comp. Zool. Harvard, 60:177-299,

Chamberlin, R. V. 1924. The spider fauna of the shores and islands of the Gulf of California. Proc.
California Acad. Sci., 12:561-694.

Chamberlin, R. V. and W. Ivie. 1944. Spiders of the Georgia Region of North America. Bull. Univ.

Utah, 35:1-267.

Chickering, A. M. 1944. The Salticidae of Michigan. Pap. Michigan Acad. Sci., 29:139-222.
Chickering, A. M. 1946. The Salticidae of Panama. Bull. Mus. Comp. Zool. Harvard, 97:1-474.
Cockerell, T D, A. 1893. Description of a new Jamaican spider. J. Inst. Jamaica, 1:221-222.

Comstock, J. H. 1912. The Spider Book. Garden City, New York.

Edwards, G. B. 1980. Jumping spiders of the United States and Canada: Changes in the key and list
(4). Peckhamia 2:1 1-14.

RICHMAN— REVISION OF HENTZIA

343

Emerton, J. H. 1891. New England spiders of the family Attidae. Trans. Connecticut Acad. Sci.,
8:220-252.

Emerton, J. H. 1902. The Common Spiders of the United States. Ginn and Co., Boston.

Hentz, N. M. 1832. On North American spiders. American J. Sci., 21:99-122.

Hentz, N. M. 1846. Descriptions and figures of the Araneides of the United States. Boston J. Natur.
Hist., 5:352-370.

Instituto de Geografia de la Academia de Ciencias de la Cuba y URSS. 1970. Atlas Nacional de
Cuba. Habana.

Kaston, B. J. 1948. Spiders of Connecticut. Bull. Connecticut St. Geol. Natur. Hist. Surv., 70:1-874.
Keyserling, E. 1885. Neue spinnen aus Amerika. VI. Verb. Zool.-Bot. Ges. Wien, 34:489-534.

Koch, C. L. 1848. Die Arachniden. 14. Nurnberg.

Maddison, W. 1986. Distinguishing the jumping spiders Eris militaris and Eris flava in North America
(Araneae: Salticidae). Psyche, 93:141-149.

Marx, G. 1883. Araneina. Pp 21-26. In A list of the Invertebrate Fauna of South Carolina. (L. O.
Howard, ed.). Charleston.

Peckham, G. W. and E. G. Peckham. 1883. Descriptions of new or little known spiders of the family
Attidae from various parts of the United States of North America. Milwaukee.

Peckham, G. W. and E. G. Peckham. 1888. Attidae of North America. Trans. Wisconsin Acad. Sci.,
7:1-104.

Peckham, G. W. and E. G. Peckham. 1889. Observations on sexual selection in spiders of the family
Attidae. Occ. Pap. Natur. Hist. Soc. Wisconsin, 1:1-60.

Peckham G. W. and E. G. Peckham. 1890. Additional observations on sexual selection in spiders of
the family Attidae, with some remarks on Mr. Wallace’s theory of sexual ornamentation. Occ. Pap.
Natur. Hist. Soc. Wisconsin, 1:117-151.

Peckham, G. W. and E. G. Peckham. 1893. On the spiders of the family Attidae of the island of Saint

Vincent. Proc. Zool. Soc. London, 1893:692-704.

Peckham, G. W. and E. G. Peckham. 1894. Spiders of the Marptusa group. Occ. Pap. Natur. Hist.
Soc. Wisconsin, 2:85-156,

Peckham, G. W. and E. G. Peckham. 1909. Revision of the Attidae of North America. Trans.
Wisconsin Acad. Sci., 16:355-646.

Petrunkevitch, A. 1911. A synonymic index-catalogue of spiders of North, Central and South America
with all adjacent islands, Greenland, Bermuda, West Indies, Terra del Fuego, Galapagos, etc. Bull.
American Mus., 29:1-791.

Petrunkevitch, A. 1914. Attidae of the Yale Dominica Expedition. J. New York Entomol. Soc.,
22:329-331. .. ;

Petrunkevitch, A. 1930. The spiders of Porto Rico. Part three. Trans. Connecticut Acad. Sci., 31:1-

191.

Pickard-Cambridge, F. O. 1901. Biologia Centrali-Americana, Araneida. London.

Richman, D. B. 1977. The relationship of epigamic display to the systematics of jumping spiders
(Araneae: Salticidae). Ph.D, Dissertation, University of Florida.

Richman, D. B. 1982. Epigamic display in jumping spiders (Araneae, Salticidae) and its use in
systematics. J. ArachnoL, 10:47-67.

Roewer, C. F. 1954. Katalog der Araneae. Vol. 2. Brussels.

Simon, E. 1901. Histoire naturelle des Araignees. Part 2, Fas. 3. Paris.

Walckenaer, C. A. 1805. Tableau des Araneides ou Caracteres essentiels des tribus, genres, families et
races que renferme le genre Aranea de Linne, avec la designation des especies comprises dans
chacune de ces divisions. Paris.

Walckenaer, C. A. 1837. Histoire naturelle des Insectes Apteres. Vol. 1. Paris.

Whitcomb, W. H., H. Exline and R. T. Hunter. 1963. Spiders of the Arkansas cotton field. Ann.
Entomol. Soc. America, 56:653-660.

Manuscript received November 1988, revised April 1989.

344

THE JOURNAL OF ARACHNOLOGY

INDEX

Anicius 287

Beata 287

Hentzia 295

antiliana 326

audax .310

calypso 335

chekika .319

cubana 312

fimbriata 306

footei. 324

grenada 315

mandibularis 337

mitrata 302

palmarum 296

parallela 333

pima .314

poenitens 3 1 7

squamata 339

tibialis 309

vernalis 322

vittata 330

whitcombi 328

Zambia 341

Hentschel, E. and W. B. Muchmore. 1989. Cocinachernes foUosus, a new genus and species of
pseudoscorpion (Chernetidae) from Mexico. J. Arachno!., 17:345-349.

COCINACHERNES FOLIOSUS, A NEW GENUS AND SPECIES
OF PSEUDOSCORPION (CHERNETIDAE)

FROM MEXICO

Edna Hentschel

Departmeet of Biology
Facuitad de Cieecias

Universidad Nacional Autoeoma de Mexico
Mexico, D.E Mexico

and

William B. Muchmore

Department of Biology
University of Rochester
Rochester, New York 14627 USA

ABSTRACT

The new genus and species Cocinachernes foliosus is described, based on specimens from Isla
Cocinas, Estado de Jalisco, Mexico, It is unique in the form of the spermathecae of the female.

INTRODUCTION

In. recent years increasing attention has been paid to the spermathecae of
females in the taxonomy of pseudoscorpions in the family Chernetidae (see
Muchmore 1975; Muchmore and Hentschel 1982; Mahnert 1979, 1985). A
number of species with similar external features have proved to possess quite
different female genitalia (spermathecae), for example, Epactiochernes compared
to Dinocheirus (Muchmore 1974), and Americhernes compared to Lamprochernes
(Muchmore 1976). Here we present the description of a genus which is quite
similar externally to IlUnichernes Hoff or Ceriochernes Beier but which has a
unique form of spermatheca.

Cocinachernes, new genus
Type species. — Cocinachernes foliosus, new species.

Diagnosis. — A genus of the family Chernetidae Chamberlin. Carapace and
pedipalps moderately sclerotized, deep reddish-brown in color. Surface of
carapace, palps and abdominal sclerites reticulate, with stout, wide, pinnately
feathered and leaflike vestitural setae. Carapace with 2 transverse furrows,
without eyes or ocular spots, and with about 60 pinnately feathered setae more or

346

THE JOURNAL OF ARACHNOLOGY

less arranged in 8 transverse rows. Tergites 1-10 divided; sternites 4-10 divided;
pleural membranes longitudinally rugose and slightly papillose. Tergites with 6 to
12 pinnately feathered, palmate setae arranged in a single row; sternites with 8 to
12 smaller marginal setae, slightly denticulate on anterior ones, changing
gradually to pinnately feathered shape on posterior ones. Setae on stigmal and
anal plates short and acuminate. Cheliceral hand with 7 setae, b and sb terminally
denticulate, es short and acuminate, both accessory setae acuminate (proximal
one possibly denticulate); flagellum of 4 setae, the 2 basal ones short and lying
close together, the distal one unilaterally serrate along distal margin; galea of
female moderate in size, with 6 small branches, that of male smaller and without
branches. Palps robust, slightly larger in female; surface of palps reticulate except
for fingers, with pinnately feathered setae on all segments except movable finger.
Two especially long, slightly feathered setae on inner surface of chelal hand near
base of fixed finger, and 3 similar long setae arranged in a row on dorsal side of
fixed finger. Trichobothrium st on movable finger closer to t than to sb and near
middle of finger; ist on fixed finger distinctly distad of est, which is near middle
of finger. Venom apparatus well developed in movable chelal finger, vestigial or
absent in fixed finger; each finger with 29-32 marginal teeth and with several
external and 1 or no internal accessory teeth. Legs rather slender, with pinnately
feathered, denticulate, and acuminate setae. Tarsus of leg IV without any tactile
seta. Anterior genital operculum of male with a group about 35 setae, including 4
larger ones medially; posterior operculum with 20 setae on face and along
posterior margin. Female anterior genital operculum with 23-24 short setae on
face; posterior operculum with a row of 15-16 setae on face and along posterior
margin; spermatheca in form of four short, broad tubes attached to a spherical
base.

Etymology. — The genus is named after the island Cocinas, where it was

collected.

Remarks. — This new genus shares a number of important characters with
Illinichernes Hoff (1949) from the United States: four setae in cheliceral
flagellum, absence of a tactile seta from tarsus IV, presence of accessory setae on
the cheliceral hand, the peculiar leaflike setae on body and appendages, and the
long clavate setae on the fixed chelal finger. But it differs radically from
Illinichernes in the form of the spermathecae: four short tubules extending from a
central globose chamber in Cocinachernes, two long thin tubules with expanded
ends in Illinichernes (see Benedict and Malcolm 1982: fig. 5). Also Cocinachernes
is quite similar in respect to its integument and leaflike setae to Ceriochernes
foliaceosetosus Beier (1974) from Brasil, but differs in having four rather than
three setae in the flagellum. It differs from Ceriochernes^}) amazonicus Mahnert
(1985) in the nature of both the vestitural setae and the spermathecae.

Cocinachernes foliosus, new species
Figs. 1-7

Types. — Mexico, Estado de Jalisco, Islas Cocinas, in dry litter 200 m from
shore, 9 September 1981 (J. G. Palacios): male holotype and 2 female paratypes,
deposited in the Arachnid Collection of the Laboratorio de Acarologia, Facultad
de Ciencias, Universidad Nacional Autonoma de Mexico, Mexico DF.

HENTSCHEL & MUCHMORE— NEW PSEUDOSCORPION GENUS

347

Figures. 1-7. — Cocinachernes foliosus, new species: 1-5, male; 1, part of tergite, showing reticulated
surface and leaflike setae; 2, genital opercula; 3, right palp, dorsal view; 4, left chela, lateral view; 5,

leg IV; 6, 7, female; 6, genital opercula; 7, spermathecae.

Description of male (based on holotype). — Surface of carapace, tergites,
posterior sternites palps (except fingers), and legs lightly to moderately sclerotized
and moderately to heavily reticulate, each space with more or less hexagonal
shape, except for the small area surrounding the insertion of each seta, which is
smooth (Fig. 1). Vestitural setae on these parts of the body broad, pinnately
feathered, leaflike, with a middle axis and 9 to 12 lateral ribs, which are often
arranged opposite one another, the pair of basal ribs always much shorter and
thinner than the rest; each seta standing on a small tubercle (Fig. 1). Carapace
subtriangular, as long as posterior breadth, with two distinct transverse furrows,

348

THE JOURNAL OF ARACHNOLOGY

without eyes or ocular spots; surface heavily reticulate, with 62 vestitural setae
more or less arranged in 8 transverse rows, 7 setae near anterior margin and 6
near posterior margin, setae along lateral margins smaller than those on face of
carapace. Tergites 1^10 and sternites 4-10 divided; surfaces pf tergites heavily
reticulate; anterior sternites almost smooth, changing gradually to a heavily
reticulate surface on the posterior ones; pleural membranes longitudinally rugose
and slightly papillose; all dorsal setae well developed in a pinnately feathered
shape, except for the acuminate ones on the anal plate; setae of sternites
acuminate on the genital, stigmal and anal plates, denticulate on anterior
sternites, and changing gradually to a pinnately feathered shape on posterior
sternites, but always smaller than setae on the dorsum of body. Tergal chaetotaxy
of holotype 6:8:6:8:8:9:9:11:10:11:10:2; and sternal chaetotaxy 37:(3)20(3):
(1)8(1): 17: 13: 12: 10: 10:8:6:2. 37 setae on face of anterior genital operculum, of
which the 4 in the middle are longer; posterior operculum with 20 setae, arranged
more or less in 2 rows on face and posterior margin (Fig. 2). Internal genitalia of
usual chernetid type, well sclerotized and distinct.

Chelicera 1/3 as long as carapace; hand with 7 setae, b and sb terminally
denticulate, es short and acuminate, distal accessory seta acuminate proximal one
slightly denticulate; flagellum of 4 setae, the 2 basal ones short and lying close
together, and the distal one heavily dentate along distal margin; galea short and
simple, without denticles.

Palp stout (Fig. 3), with chelal hand a little broader than deep; femur 2.75,
tibia 1.8, and chela (without pedicel) 2.5 times as long as broad; hand (without
pedicel) 1.4 times as long as deep; movable finger 0.97 as long as hand. Surface of
palp moderately reticulate except for fingers; setae pinnately feathered except for
those on ventral margin of hand, which are less developed or denticulate and
those on fingers, which are acuminate; 5 especially long, pinnately feathered setae
on fixed finger, a row of 3 on dorsal side near middle and 2 on internal face near
base of finger. Trichobothria as indicated in Fig. 4. Fixed finger with 30
contiguous, cusped marginal teeth, and 5 external plus 1 internal accessory teeth;
movable finger with 3 1 similar marginal teeth and 5 external (no internal)
accessory teeth; venom apparatus well developed only in movable finger.

Legs slender, with moderately reticulate surfaces and with pinnately feathered
setae except on internal margins, where setae are smaller and denticulate or
acuminate; leg IV (Fig. 5) with entire femur 3.95 times as long as deep; tarsus
lacking a tactile seta.

Description of female (based on the 2 paratypes). — Much like male, but slightly
more robust; genital opercula as shown in Fig. 6; anterior operculum with a
compact O-shaped group of 23-24 short setae on face, posterior operculum with
15-16 setae on face and along posterior margin; spermathecae in form of four
short, broad tubes extending from a globose chamber with a rugose surface (Fig.
7). Cheliceral galea better developed than in male, with 6 small rami. Palp much
as in male, but slightly larger; femur 2.8, tibia 2.05-2.1, and chela (without
pedicel) 2.45-2.55 times as long as broad; hand (without pedicel) 1. 3-1. 4 times as
long as deep; movable finger 0.94-1.03 times as long as hand. Fixed finger with
29-31 marginal teeth, and 5-7 external and 1 or no internal accessory teeth;
movable finger with 32 marginal teeth, 3-4 external and no internal accessory
teeth.

HENTSCHEL& MUCHMORE— NEW PSEUDOSCORPION GENUS

349

Measurements (mm). — Figures given first for holotype male, followed in
parentheses by those for the paratype females. Body length 2.13 (2.41-2.55).
Carapace length 0.68 (0.74-0.78). Chelicera length 0.225. Palpal femur 0.59 (0.67)
by 0.215 (0.24); tibia 0.525 (0.58-0.59) by 0.29 (0.28); chela (without pedicel) 0.83
(0.93-0.94) by 0.335 (0.37-0.38); hand (without pedicel) 0.44 (0.48-0.495) by 0.315
(0.36-0.37); pedicel length about 0.10; movable finger length 0.425 (0.47-0.495).
Leg IV: entire femur 0.495 (0.54-0.57) by 0.125 (0.12-0.14); tibia 0.41 (0.44-0.45)
by 0.08 (0.095-0.10); tarsus 0.335 (0.355-0.37) by 0.07 (0.08).

Etymology. — The species is named foliosus for the leaflike appearance of the
vestitural setae.

Remarks. — In this species the female is larger than the male, but the
proportions of the palps are about the same in the two sexes. It is interesting to
note the considerable reduction in the number of setae, a condition apparently

related to the increase in breadth of the setae.

ACKNOWLEDGMENT

We thank J. G. Palacios for providing the specimens upon which this work is
based.

LITERATURE CITED

Beier, M. 1974. Brasilianische Pseudoscorpione aus dem Museum in Genf. Rev. suisse Zoo!., 81:899-
909.

Benedict, E. M. and D. R. Malcolm. 1982. Pseudoscorpions of the family Chernetidae newly
identified from Oregon (Pseudoscorpionida, Cheliferoidea). J. Arachnol., 10:97-109.

Hoff, C. C. 1949. The pseudoscorpions of Illinois. Bull. Illinois Nat. Hist. Surv., 24:407-498.

Mahnert, V. 1979. Pseudoskorpione (Arachnida) aus dem Amazonas-Gebiet (Brasilien). Rev. suisse
ZooL, 86:719-810.

Mahnert, V. 1985. Weitere Pseudoskorpione (Arachnida) aus dem zentralen Amazonasgebiet
(Brasilien). Amazoniana, 9:215-241.

Muchmore, W. B. 1974. Pseudoscorpions from Florida. 3. Epactiochernes, a new genus based upon
Chelanops tumidus Banks (Chernetidae). Florida Entomol., 57:397-407.

Muchmore, W. B. 1975. Use of the spermathecae in the taxonomy of chernetid pseudoscorpions. Proc.
6th Int. Arachnol Congr.: 17-20.

Muchmore, W. B. 1976. Pseudoscorpions from Florida and the Caribbean area. 5. Americhernes, a
new genus based upon Chelifer oblongus Say (Chernetidae). Florida Entomol, 59:151-163.

Muchmore, W. B. and E. Hentschel. 1982. Epichernes aztecus, a new genus and species of
pseudoscorpion from Mexico (Pseudoscorpionida, Chernetidae). J. Arachnol, 10:41-45.

Manuscript received January 1989, revised April 1989.

f ’ ^;*.*

■:pi **-,*

w. -
'-r:?

■ ■?. • - ' . ^ '

7 ...

vy*

,.• fH.^a

, u >i1

■ '

$

•'*t

^ i'

mi

iS

.«*» ^=“ .

■*' ’ -^ » '■ # ' ’,i* , ’ .' • ’ ^ ji t.v JkbJf .'^,' ' i'a •*-'>.. * -■ '. t' '^.= i

i.-;- ! i- ■

^.•^•«%.v- !<.*{. ..:-}i,.t ,.,1

•'■■; v' -ni'viw ^ r,i.’ ,J. .

'V.i ':■>>'' ■ 'Jt- •♦:>i'''v>!..^’'^w*jjtij ■■.

■•• >.< . ■'■•'.h-su' •»;/ . r''‘ l«?"Sa?9*2 2.^A*-V.‘''*.‘i1IHI'*. :*’-'l ‘Vt;

J i »

,v'.

' iiViliSwi

>I ■' ■' j .. ■■ ,,

Vj-' • r,.' > T ‘•^. ,.^t ■ ' * ■■■ !

.. -.-e*

•'> • i^..j';

-_ _ /-■v ‘ '•»:- ' • ■

i.-. '■•'

r* .i ■i.-'ftf* %» . . ■■

> v-l-r.'. V'V. r: •■

r ■ ■' ' - 1. wori^fi

V. .

> } J. u fil '•• ^ '* ♦ T,

r- '■ ^ "v":,-; » r-.i- i xnjf /, u

^ .. i ^ j^

'■V^'' .’^ ■••

,ff •*

. ,♦.■■•■ ». . '''if" ■■•'

.... i^-.

V.,;;." S'*l*t."l ^ '‘’ * -. '■ ^li '.,

tV-t- > *• . ; \ ••liW ''•

.. ■ */ . ■ v-''. ^‘.‘. . >. -u..

. ^ V r^,^\

t; . '“‘■■J;

^ /wCi::.! <5, -

k.v:-r. xi||^

. ' XSmm:

A

'i ffl

Till. I ^ ‘Vj-nv/MC ■

4

i"

;i H^jort 'J'

C .l»j '.^! • jh^ B,^

■. ,. pt tw-Vv j

Curcic, B. P. M. 1989. Further revision of some North American false scorpions originally assigned to
Microcreagris Balzan, (Pseudoscorpiones, Neobisiidae). J. Arachnol., 17:351-362.

FURTHER REVISION OF SOME NORTH AMERICAN
FALSE SCORPIONS ORIGINALLY ASSIGNED TO
MICROCREAGRIS BALZAN
(PSEUDOSCORPIONES, NEOBISIIDAE)

B. P. M. Curcic

Institute of Zoology, Faculty of Science, Studentski Trg 16,
YU~11000 Beograd, Yugoslavia

ABSTRACT

The type specimens of six North American species, originally described in the genus Microcreagris
Balzan have been re-examined. Of these, five species belong to four Microcreagris-xtXdLitd genera.
These are: T lata (Hoff) in Tuberocreagris Curcic, C laudabilis (Hoff) in Cryptocreagris Curcic, A.
mortis (Muchmore) in Alabamocreagris Curcic, and A. reddelli (Muchmore) and A. texana
(Muchmore) in Australinocreagris Curcic. A new genus, Minicreagris, has been proposed, with
Microcreagris pumila (Muchmore) as its type species. Genital areas and trichobothrial patterns are
figured, and brief diagnoses and some taxonomic and distributional data for the genera and species
examined are given. A key to the Microcreagris-itl^itA genera of North America north of Mexico is
presented.

Considerable confusion has surrounded the identity of members of the genus
Microcreagris Balzan, 1892. This was in part due to the difficulty in obtaining
type specimens, but to a greater extent to inadequate original description of the
type species, M. gigas Balzan, 1892, from Asia. In addition, the long used
character of galeal form has shown to be of limited value. It was recently that the
holotype of M. gigas was adequately redescribed by Mahnert (1979), and Curcic
(1983) did the same for the other known species, M. herculea Beier, 1959. As
currently defined, the genus Microcreagris is restricted to China and Afghanistan
(Curcic 1983, 1985, 1986).

Some of the available pseudoscorpions, previously assigned to the genus
Microcreagris and present in North America north of Mexico (Beier 1931, 1932;
Hoff 1945, 1958; Chamberlin 1952, 1962; Muchmore 1966, 1969), have been
revised by Curcic (1978, 1981, 1982a, 1984). The type material of some United
States species, along with some undetermined U.S. species, has been re-examined.
From this study, ten new genera have been proposed and described (Curcic 1984).
Thus the outstanding heterogeneity of Microcreagris'^ in that area has been
demonstrated.

In the present study, I have re-examined the type specimens (holotypes and
allotypes) of six additional pseudoscorpion species of the ""Microcreagris"-
complex, which are deposited in the collection of the American Museum of
Natural History in New York (AMNH). Of these, five species belong in four
Microcreagris-YQlaiQd genera already reported from this region and one belongs to
a new genus, Minicreagris.

352

THE JOURNAL OF ARACHNOLOGY

The purpose of this paper is to further demonstrate the heterogeneity of
Microcreagris'" in North America and to present objective criteria for the
identification of some North American genera of this complex. This study should
also stimulate an analysis of the taxonomic rank of all other North American
pseudoscorpions currently assigned to "'Microcreagris.""

Family NEOBISIIDAE Chamberlin, 1930
Genus Tuberocreagris Curcic, 1978

Diagnosis. — Galea with subtermieal spinule(s). Abdominal sternites with one
row of setae. Male genital area: sternite II with a group of median and posterior
setae, sternite III with a transverse row of anterior setae, some intermediary setae,
and a row of posterior marginal setae. Female genital area: sternite II with a
group of setae on either side of the midline, sternite III with a transverse row of
posterior setae.

Manducatory process with 3 setae. Femur and chelal palm of pedipalp with
distinct granulations. Interiorly, an accessory tubercle on femur and two such
tubercles on tibia. Trichobothriotaxy esb distal to eb; ib-ist-isb in proximal part
of finger; est closer to it than to ist; sb and st equidistant from b and t,
respectively.

Leg IV: tibia, basitarsus and telotarsus with one tactile seta each.

Type species. — Meobisium rufulum Banks.

Subordinate taxa. — Tuberocreagris rufula (Banks), T. lata (Hoff).

Distribution. — District of Columbia and North Carolina, USA.

Remarks. — A thorough study of some congeneric specimens from Washington,
D.C., has showed that the interior tubercles on pedipalpal femora and tibiae may
be of different sizes, and sometimes inconspicuous and even difficult to establish
(particularly when these podomeres are studied on fixed slide preparations)
(Curcic, MS).

Tuberocreagris lata (Hoff), new combination
Figs. 1, 2, 9, 10

Microcreagris lata Hoff, 1945:323, 1958:12.

Diagnosis. — Epistome small and apically rounded. Carapace with 4 + 4 + 6 + 6
+ 6 = 26 (male) and 4 + 6 + 6 + 6 + 7 = 29 setae (female). Two pairs of eyes
resembling eye-spots. Galea with several terminal branchlets.

Tergites FX with 6-9-1 FI FI F12-1 M3ri2ri 1 (male), or 6-94 FI F124 F1242-
13-12 setae (female). Male genital area: sternite II with 12 setae (of these, six
setae are found along the posterior sternal margin), sternite III with 4 anterior, 3
intermediary and 10 posterior setae, and 4 or 5 microsetae along each stigma.
Female genital area: sternite II with 3 setae on each side of the midline, sternite
III with 10 posterior setae and 5 or 6 suprastigmal microsetae on each side. Other
sternites with one row of setae.

Pedipalps: fixed chelal finger with 51 or 52 (male) and 56 close-set, , small and
rounded and eventually square-topped teeth (female); some distal teeth are

CURCIC -REVISION OF MICROCREAGRIS

353

/

0.25

5

</

/-

4

Figures 1-5, — Genital area. 1, Tuberocreagris lata (Hoff), holotype male; 2, T. lata (Hoff), allotype
female; 3, Cryptocreagris iaudabilis (Hoff), holotype male; 4, C. laudabilis (Hoff), allotype female; 5,
Minicreagris pumila (Muchmore), holotype male. Scale in mm.

asymmetrically pointed. Movable chelal finger with 51 (male) or 60 small, close-
set and rounded teeth (female); few distal teeth are asymmetrical
Distribution. — North Carolina, USA.

354

THE JOURNAL OF ARACHNOLOGY

Specimens examined. — Holotype male (C. Clayton Hoff 5610-S-419.1) and allotype female (C.
Clayton Hoff 561 1-S-4I9.2), from oak litter, sandy soil, Duke Forest, Durham, North Carolina, 16
September 1944 (A. S. Pearse).

Genus Cryptocreagris Curcic, 1984

Diagnosis. — Galea with apical branchlets. Abdominal sternites VI and VII each
with 2 anterior discal setae. Male genital area: sternite II with a group of median
and posterior setae, sternite III with a row of anterior, some intermediate and a
series of posterior setae. Female genital area: sternite II with a group of small
setae on each side of the midline, sternite III with a row of posterior setae.

Manducatory process with 4 setae. Femur and chelal palm of pedipalp with
inconspicuous granulations. Trichobothriotaxy: esb distal to eb\ ist-isb-ib
clustered on finger base; it and et located distally on finger tip; est nearer to it
than to ist\ st slightly closer to t than to sb\ sb slightly closer to b than to st.

Leg IV: tibia, basitarsus and telotarsus with one tactile seta each.

Type sp^cits.—Microcreagris laudabilis Hoff.

Subordinate taxa. — Cryptocreagris laudabilis (Hoff), C. magna (Banks).

Distribution. — California and New Mexico, USA.

Remarks. — In this study, the holotype and allotype of C laudabilis have been

restudied, in addition to the earlier redescription, based on the analysis of seven
paratype specimens (Curcic 1984).

Cryptocreagris laudabilis (Hoff)

Figs. 3, 4, 12, 13

Microcreagris laudabilis Hoff, 1956:4, 1958:12.

Diagnosis. — Epistome low and rounded apically, carapace with 4 + 6 + 4 + 4 +
6 = 24 setae. Two pairs of eyes with flattened lenses. Galea with terminal
branchlets. Flagellum with 8 anteriorly pinnate blades.

Tergites I-X with 6^94 1-12-1242G347H7G5 (male) and 6-8™l MOG 1-1342-13^
11-12 setae (female). Male genital area: sternite II with 27 median and posterior
setae, sternite III with 5 anterior, 3 intermediate and 16 posterior setae, and 6
microsetae along each stigma. Female genital area: sternite II with a group of 4
small setae on either side of midline, sternite III with 12 posterior setae and 6
small setae along each stigma. Sternites VI and VII each with 2 anterior discal
setae.

Pedipalps: fixed chelal finger with 71 (female) and 78 small and close-set teeth
(male), distal teeth are asymmetrical, gradually changing from square-topped to
eventually slightly asymmetrical teeth. Movable chelal finger with 72 (female) and
80 teeth (male), similar in form and size to those on the fixed chelal finger, but
only few distal teeth asymmetrical.

Distribution. — New Mexico, USA.

Remarks. — In the holotype female of this species, two teratologies were noted.
First, on the movable finger of the left chelicera, two galeal setae are developed
(instead of one). Second, the right manducatory process bears 6 setae, while the
left has the normal complement of 4 setae.

CURCIC— REVISION OF MICROCREAGRIS

355

The material of C. magna has been redescribed elsewhere (Curcic 1984),
Specimens examined. — Holotype male (C. Clayton Hoff 9152-S-2083.5) and allotype female (C.
Clayton Hoff 9152-S-2083.9), under rocks, fir litter, near top of Mt. Taylor, Valencia Co., New
Mexico, 11,150 ft. 21 July 1953.

Genus Minicreagris Curcic, new genus

Name derivation. — Named for the small body size of its type species

Diagnosis. — Galea stylet-like. Flagellum of 7 or 8 blades, of which the
proximal 3 are smooth and reduced in size from distal to proximal. Male genital
area: sternite II with a group of median and posterior setae (of these, some setae
are found along the posterior sternal margin), sternite III with a transverse row of
some anterior and a series of posterior setae. Female genital area: one group of
small setae on each side of the midline. Other abdominal sternites with one row
of setae.

Manducatory process with 3 setae. Chelal palm of pedipalp with inconspicuous
granulations (not “smooth” sensu Muchmore 1969). Trichobothriotaxy: esb distal
to eb\ ib-ist-isb in proximal part of finger; est closer to it than to isv, sb,
equidistant from st and b; st somewhat closer to t than to sb, and closer to the

interior finger margin (or teeth) than other trichobothria of the movable chelal
finger.

Leg IV: tibia, basitarsus and telotarsus with one long tactile seta each.

Type species. — Microcreagris pumila Muchmore.

Subordinate taxon. — Minicreagris pumila (Muchmore).

Distribution. — Alabama and Tennessee, USA.

Remarks. — The study of the paratype male and paratype female of M. pumila
(Curcic 1984) has made me think that the taxonomic status of this species was
debatable. The reason for this opinion was the inadequate number of males
studied. The subsequent analysis of the holotype male has enabled me to
reconsider M. pumila as the type species of a new genus. Its diagnostic characters
include the number of eyes, the trichobothrial pattern, the setation of the
manducatory process, the form and dermal structure of pedipalpal articles, the
setation of the sternites and the form of flagellar blades. At present, Minicreagris
is monotypic.

Minicreagris pumila (Muchmore), new combination
Figs. 5, 11

Microcreagris pumila Muchmore, 1969:2.

Diagnosis. — Epistome small and rounded. Carapace: 4 + 6 + 4 + 4 + 6"24
setae. With two small eyes. Galea simple.

Tergites I-X with 6-6-8-11-10-12-11-10-10-9 setae. Male genital area: sternite II
with 10 setae. Of these, 6 are along the posterior sternal margin. Sternite III with
5 anterior and 10 posterior setae, and 3 suprastigmal setae on each side. Other
sternites with one row of setae.

Pedipalps: fixed chelal finger with 43 small and close-set teeth, distal ones
asymmetrical; proximal teeth rounded or slightly asymmetrical. Movable chelal

356

THE JOURNAL OF ARACHNOLOGY

6

8

f

/

Figures 6-8. — Genital area: 6, Alabamocreagris mortis (Muchmore), holotype female; 7,
Australinocreagris reddelli (Muchmore), holotype female; 8, Australinocreagris texana (Muchmore),
holotype female. Scale in mm.

finger with 42 small and close-set teeth; distal teeth asymmetrically pointed, and
proximal ones square-topped or rounded.

Distribution. — Alabama, Georgia, and (?) Tennessee, USA.

Remarks. — The descriptions of this species by Muchmore (1969) and by Curcic
(1984) define the species adequately, except for some details of the setation and
dermal structure.

Muchmore (1969) suggested that this species is epigean and is only fortuitously
collected in caves. That it may be widespread and more variable than indicated

CURCIC— REVISION OF MICROCREAGRIS

357

by the same author is suggested by two other specimens from caves in Tennessee
and Alabama.

Specimen examined. — Holotype male (WM 1009.01001), from Bryant Cave, Blount Co., Alabama,
19 March 1966 (S. Peck).

Genus Alabamocreagris Curcic, 1984

Diagnosis. — Galea simple. Abdominal sternites uniseriate, Male genital area:
sternite II with some median and posterior setae, sternite III with an anterior row
of few setae and a posterior setal row. Female genital area: sternite II with a
group of small setae on either side, sternite III with a transverse series of
posterior setae.

Manducatory process with 4 long setae. Pedipalpal articles smooth, except for
chelal palm which is inconspicuously granulate internally. Trichobothriotaxy: esb
not distal to eb', ist-isb-ib at finger base; it and et located distally; est closer to it
than to ist\ st equidistant from t and sF, sb closer to b than to st.

Leg IV: tibia, basitarsus and telotarsus each with one or two tactile setae.

Type species. — Microcreagris pecki Muchmore.

Subordinate taxa. — Alabamocreagris pecki (Muchmore), A. mortis (Much-

more).

Remarks. — The type species has been described elsewhere (Curcic 1984).

Alabamocreagris mortis (Muchmore), new combination

Figs. 6, 14

Microcreagris mortis Muchmore, 1969:8.

Diagnosis. — Epistome knob-like. Carapace with :4 + 6 + 4 + 4 + 6=:24 setae.
Neither eyes nor eye-spots present. Galea unbranched.

Tergites I-X with 6-6-7-9-9-13-13-12-12-10 setae. Female genital area: sternite II
with 3 small setae on either side of the middle, sternite III with 10 posterior setae
and 3 suprastigmal microsetae on each side. Other sternites with single row of
setae.

Pedipalps: fixed chela! finger with 71 close-set and small teeth, distal ones
asymmetrical; proximal teeth square-topped. Movable chelal finger with 81 teeth
which are small, close-set, distally asymmetrical, and rounded proximally; basal
teeth square-topped.

Remarks. — Alabamocreagris mortis has a single tactile seta on telotarsus IV,
whereas A. pecki bears two such setae.

Specimen examined. — Holotype female (WM 1555.01001), from “Morgue” Cave nr. Fern Cave,
Jackson Co., Alabama, 22 June 1968 (W. Torode).

Genus Australinocreagris Curcic, 1984

Diagnosis. — Galea simple, stylet-like, or branched apically. Male genital area:
sternite II with a cluster of median and posterior setae, sternite III with a group
of anterior and median setae, followed by few intermediary setae, and a
transverse row of posterior setae. Female genital area: sternite II with a group of

358

THE JOURNAL OF ARACHNOLOGY

Figures 9-13. — Trichobothrial pattern: 9, Tuberocreagris lata (Hoff), holotype male; 10, T. lata
(Hoff), allotype female; 11, Minicreagris pumila (Muchmore), holotype male; 12, Cryptocreagris
laudabilis (Hoff), holotype male; 13, C laudabilis (Hoff), allotype female. Scale in mm.

small setae on either side of the middle, sternite III with a row of posterior setae.
Abdominal sternites VI-VIII each with a pair of anterior discal setae.

Manducatory process with 3 (occasionally 4) setae. Pedipalpal articles
inconspicuously granulate. Trichobothriotaxy: esb distal to eb, both setae on bulb
of chela; ist and isb close to each other; ib somewhat closer to ist than to esb; est~

CURCIC— REVISION OF MICROCREAGRIS

359

Figures 14-16. — Trichobothrial pattern: 14, Alabamocreagris mortis (Muchmore), holotype female;
15, Australinocreagris texana (Muchmore), holotype female; 16, Australinocreagris reddeUi
(Muchmore), holotype female. Scale in mm.

it~et in distal finger half; st closer to t than to sb', sb slightly closer to st than to b,
or equidistant from these setae.

Tibia IV and basitarsus IV each with one long tactile seta, and telotarsus IV
with one or two such setae.

Type species. — Microcreagris grahami Muchmore.

Subordinate taxa. —Australinocreagris grahami (Muchmore), A. ozarkensis
(Hoff), A. reddelli (Muchmore), and A. texana (Muchmore).

Distribution. — Arkansas, California and Texas, USA.

Remarks. — In the description of Australinocreagris, Curcic (1984) erroneously
stated that sternite III of the male is beset with ‘Tew intermediary setae”. Actually
this sternite has only a small group of anterior and median setae and a transverse
row of posterior setae, as illustrated in fig. 7 of the same paper.

360

THE JOURNAL OF ARACHNOLOGY

Australinocreagris reddelli (Muchmore), new combination
Figs. 7, 16

Microcreagris reddelli Muchmore, 1969:17.

Diagnosis. — Epistome absent. Carapace with 4 + 6 + 2 + 5 + 4 = 21 setae.
Neither eyes nor eyespots developed. Both galeae broken and lost. Flagellum
eight-bladed, all blades pinnate anteriorly.

Tergites I-X with 8-10-12-1 FI Fll-l 1-1 FI 1-9 setae. Female genital area:
sternite II with a small group of 3 setae on either side of the midline, stemite III
with 13 posterior setae and 6 microsetae along each stigma. Male genital area:
unknown. Sternites VFVIII each with a pair of anterior discal setae.

Pedipalps: fixed chelal finger with 85 small and close-set, asymmetrically
pointed teeth. Movable chelal finger with 98 small, asymmetrically pointed and
close-set teeth.

Distribution. — Texas, USA.

Remarks. — In this species, basitarsus IV has two tactile setae. In view of the
fact that A. texana has one such seta on this podomere, it is probable that the
number of tactile setae on basitarsus IV reflects intrageneric variability, just as is
seen in the members of the genus Neobisium Chamberlin, 1930 (Curcic 1982b).

Specimen examined. — Holotype female (WM 171.01001), under small rock on silt in darkness,
Shultz Cave, 2 miles E Valente, Travis Co. Texas, 21 August 1963 (Bill Russell).

Australinocreagris texana (Muchmore), new combination
Figs. 8, 15

Microcreagris texana Muchmore, 1969:18.

Diagnosis. — Epistome minute and knob-like. Carapace with 4 + 6 + 7 + 6 = 23
setae. Neither eyes nor eye-spots present. Galea biramous, each ramus with 2 or 3
terminal branchlets. Flagellum nine-bladed, all blades pinnate anteriorly.

Tergites FX with 8-8-12-14-13-14-13-12-13-10 setae. Female genital area:
sternite II with 4 small setae on each side of the midline, sternite III with a series
of 13 posterior setae and 6-8 small setae along each stigma. Male genital area:
unknown. Sternites VI-VIII each with a pair of anterior discal setae. Even on
sternite IX two median setae are found slightly anterior to posterior setal row.

Pedipalps: fixed chelal finger with 99 low, close-set and small teeth; distal ones
asymmetrically pointed and proximal ones square-topped but still slightly
asymmetrical. Movable chela! finger with 108 teeth of the same form and size as
those on the fixed finger.

Distribution. — Texas, USA.

Remarks. — This species has one tactile seta on telotarsus IV, and A. reddelli
has two such setae. Until more examples are available, the occurrence of the
different number of tactile setae on telotarsus IV in the members of
Australinocreagris may be regarded as the case of intrageneric variability.

Specimen examined. — Holotype female (WM 849.01001), from Tooth Cave, Travis Co., Texas, 16
May 1965 (T. Barr, R. Mitchell, W, Andrews),

CURCIC— REVISION OF MICROCREAGRIS

361

Key to Microcreagris-icldLlQd Genera of North
American Pseudoscorpions

The newly established genus and its related genera (Curcic 1984) may be
distinguished by means of the following key:

L Some abdominal sternites with two rows of setae 2

Abdominal sternites with one row of setae 5

2. Sternites VI and VII with two rows of setae Cryptocreagris Curcic

Sternites VKVIII with two rows of setae .............................. 3

3. Stemite III of the female with a transverse row of posterior setae 4

Sternite III of female with a transverse row of posterior setae and a few

setae on face near middle ....................... Tartarocreagris Curac

4. Sternite II of female with a group of setae on either side of midline.

Manducatory process with 3 setae ............ > Australinocreagris Curcic

Sternite II of female with a unique group of setae in the form of an inverted
'TJ’f Manducatory process with 4 setae ......... Saetigerocreagris Curcic

5. Stemite III of male with an anteromedian groove in the form of a “V”,

bordered by two pairs of small setae ............... Fissiiicreagris Curac

Sternite III of male with an ungrooved anterior margin: anterior setae
clustered or arranged in a row ..................................... 6

6. Pedipaipal articles smooth .......................... Lissocreagris Curac

Some pedipaipal articles granulate .................................... 7

1. Chelal trichobothria sb and st equidistant from b and / respectively ........ 8

Chelal trichobothrium sb closer to b than to st, or somewhat closer to st
than to h ........................................................ 9

8. Manducatory process with 3 long setae. Pedipaipal femur granulate, femur

and tibia with one and two accessory tubercles respectively ..............

.............................................. Tuberocreagris Curac

Manducatory process with 4 long setae. Pedipaipal femur and tibia
smooth ..................................... Alabamocreagris Curcic

9. Chelal trichobothrium st somewhat closer to the interior finger margin than

are jh or h .................................. . Minicreagris, new genus

Chelal trichobothrium st as close to the interior finger margin as or b . . . . .

...............................................................10

10. Sternite III of male with a cluster of anteromedian setae and a row of
posterior setae. .............................. .Americocreagris Curcic

Sternite III of male with an anteromedian group of setae, a series of
intermediary setae and a transverse row of posterior setae ...............

............................................... Globocreagris Curac

362

THE JOURNAL OF ARACHNOLOGY

ACKNOWLEDGMENTS

I am grateful to the reviewers, W. B. Muchmore and Vincent F. Lee for the
constructive criticism and valuable comments on the problem studied. I am also
indebted to Norman Platnick of the American Museum of Natural History in
New York for the loan of the type specimens considered herein.

LITERATURE CITED

Balzan, L. 1891-1892. Voyage de M. E. Simon au Venezuela (Decembre 1887 - Avril 1888). 16® mem.

Chernetes (Pseudoscorpiones). Ann. Soc. Entomol. France (1891), 60:497-552.

Beier, M. 1931. Neue Pseudoscorpione der U.O. Neobisiinea. Mitt. zool. Mus. Berlin, 17:299-318.
Beier, M. 1932. Pseudoscorpionidea 1. Subord. Chthoniinea et Neobisiinea. Tierreich, 57:1-258.

Beier, M. 1959. Zur Kenntnis der Pseudoscorpioniden-Fauna Afghanistans. Zool. Jahrb., Abt. Syst.
Oekol. Geogr. Tiere, 87:257-282.

Chamberlin, J. C. 1952. .New and little-known false scorpions (Arachnida, Chelonethida) from
Monterey County, California, Bull. American Mus. Nat. Hist., 99:259-312.

Chamberlin, J. C. 1962. New and little-known false scorpions, principally from caves, belonging to the
families Chthoniidae and Neobisiidae (Arachnida, Chelonethida). Bull. American Mus. Nat. Hist.,
123:299-352.

Curcic, B. P. M. 1978. Tuber ocreagris, a new genus of pseudoscorpions from the United States
(Arachnida, Pseudoscorpiones, Neobisiidae). Fragm. Balcanica, 10:111-121.

Curcic, B. P. M. 1981. A revision of some North American pseudoscorpions (Neobisiidae,
Pseudoscorpiones). Bull. Mus. Hist. Natur. Belgrade, ser. B, 36:101-107.

Curcic, B. P. M. 1982a. Americocreagris, a new genus of pseudoscorpions from the United States.

Bull. Acad. Serbe Sci. Arts, CL Sci. Math. Natur. (Sci. Natur.), 80, 22:47-50.

Curcic, B. P. M. 1982b. Postembryonic development in the Neobisiidae (Pseudoscorpiones,
Arachnida). Serbian Acad. Serbe Sci. Arts, Monogr., 545, Dep. Sci., 56:1-90.

Curcic, B. P. M. 1983. A revision of some Asian species of Microcreagris Balzan, 1892 (Neobisiidae,
Pseudoscorpiones). Brit. Arachnol. Soc. Bull., 6:23-36.

Curcic, B. P. M. 1984. A revision of some North American species of Microcreagris Balzan, 1892
(Arachnida : Pseudoscorpiones: Neobisiidae). Brit. Arachnol. Soc. Bull., 6:149-166.

Curcic, B. P. M. 1985. A revision of some species of Microcreagris Balzan, 1892 (Neobisiidae,
Pseudoscorpiones) from the USSR and adjacent regions. Brit. Arachnol. Soc. Bull., 6:331-352.
Curcic, B. P. M. 1986. On the taxonomy and biogeography of Microcreagris-rtXdLiQd genera in Eurasia
(Neobisiidae, Pseudoscorpiones). Mem. Soc. R, Beige Entomol., 33:75-79.

Hoff, C. C. 1945. New species and records of pseudoscorpions from Arkansas. Trans. American
Microsc. Soc., 64:34-57.

Hoff, C. C. 1945. Pseudoscorpions from North Carolina. Trans. American Microsc. Soc., 64:311-327.

Hoff, C. C. 1956. Diplosphyronid pseudoscorpions from New Mexico. American Mus. Novit., 1875:1-
50.

Mahnert, V. 1979. The identity of Microcreagris gigas Balzan (Pseudoscorpiones, Neobisiidae). Brit.
Arachnol. Soc. Bull., 4:339-341.

Muchmore, W. B. 1966. A cavernicolous pseudoscorpion of the genus Microcreagris from southern
Tennessee. Entomol. News, 77:97-100.

Muchmore, W. B. 1969. New species and records of cavernicolous pseudoscorpions of the genus
Microcreagris (Arachnida, Chelonethida, Neobisiidae, Ideobisiinae). American Mus. Novit., 2392:1-
21.

Manuscript received November 1988, revised May 1989.

1989. The Journal of Arachnology 17:363

RESEARCH NOTES

A NEW SPIDER, PARATHEUMA MAKAI
(ARANEAE, DESIDAE), FROM HAWAII

In two previous papers (Beatty and Berry 1988a, 1988b) we dealt with the seven
known species of the desid genus Paratheuma Bryant, and we suggested that the
discovery of additional species might be expected. One such species was
subsequently collected on two of the Hawaiian Islands and is described here.

The specimens occurred on rocky ocean beaches, as is characteristic for the
genus; but the Hawaiian beaches, in contrast to many other localities where
Paratheuma occurs, were almost devoid of broken coral. Instead, a dark, smooth
heavy volcanic rock (probably basalt) (Fig. 1) or abrasive lava was common. On
Kauai the spiders were found on the sand under basalt rocks varying from about
10-50 cm in diameter. On Hawaii they were on sand in the spaces between
smaller pieces of rough porous lava.

Four adults of each sex were measured. Total length was measured from
anterior carapace margin (or anterior eye margin when eyes projected beyond
carapace) to the base of the spinnerets. Some specimens from Kauai were reared

Figure 1. — Type locality of Paratheuma makai, beach at Kapaa, Kauai, Hawaii.

1989. The Journal of Arachnology 17:364

in the laboratory but are not listed below. Juveniles listed here are not considered

paratypes.

Paratheuma makai, new species

Holotype. — Male from Hawaii; Kauai Island, at mile marker 9.4 N of Kapaa,
under beach rocks, 14 Jan. 1988 (J. W. and E. R. Berry) in Bishop Museum,
Honolulu, Hawaii. The name, regarded as a noun in apposition, is a Hawaiian
word meaning “toward the ocean”.

Diagnosis. — Male with relatively short (length less than twice basal width),
erect, almost triangular tibial apophysis, and long arcuate retrolateral branch of
conductor tip. Female with epigynal depressions curved slightly forward
anteriorly, the hoods flanked medially by distinct notchdike depressions. Similar
to P. australis Beatty and Berry, R insulana (Banks), and P. rangiroa Beatty and
Berry, but clearly distinguished from them by the characteristics cited.

Figures 2-5. — Genitalia of Paratheuma makai: 2, left pedipalp, ventral; 3, tibia of left pedipalp,
lateral; 4, epigynum, ventral; 5, epigynum, dorsal, cleared.

1989. The Journal of Arachnology 17:365

Descriptive notes. — Male: Total length 3. 3-4.0 mm, carapace length 1.7-1. 9 mm,
maximum carapace width 1.26-1.32 mm. Embolus arising on medial side of bulb
near middle of bulb’s length, its structure as in other species of the genus. End of
conductor with short medially directed branch and long arcuate retrolateral
branch curving down over distal retrolateral surface of tibia (Fig. 2). Tibial
apophysis erect, shorter than in any other species except P. interaesta (Roth and
Brown), triangular, its length less than twice its basal width (Fig. 3.). Bristle
pattern as in other Pacific species.

Female: Total length 3. 5-4. 3 mm, carapace length 1.6- 1.7 mm, maximum
carapace width 1.08-1.26 mm. Epigynal depressions pale, extending obliquely
forward, arching slightly mediad at anterior ends. Epigynal hoods distinct,
flanked medially by depressions. Anterior border of epigynum broadly
rectangular (Fig. 4). Ducts of epigynum narrow, twisted tightly around themselves
so that their course is very difficult to determine (Fig. 5). Seminal receptacle and
duct complex smaller than in other Pacific species.

Discussion. — Many localities on Kauai were searched, but the spiders were
found only at the beach at Kapaa, where they were abundant. No specimens were
found on Oahu, and few on Hawaii. However, only a small portion of Hawaii’s
beach was searched.

Specimens from Hawaii were smaller than the mean for specimens from Kauai;
but only one adult of each sex was taken on Hawaii, and the male matured in the
laboratory. In all other characters they agreed with the Kauai specimens.
Genitalic structure in both sexes is most similar to those of P. australis and P.
rangiroa, and somewhat less to P insulana. Other structures are almost uniform
thoughout the genus.

This species conforms to the distribution pattern shown by other species of the
genus — only one species of Paratheuma on any given island, although a given
species may occur on more than one island or island group. Further investigation
is needed to confirm this pattern and to determine the total area occupied by the
genus. Islands in need of investigation are those in Polynesia east and south of
Rangiroa and Manihi, and in Melanesia, Indian Ocean and the Atlantic Ocean.

Distribution. — HAWAII; Kauai and Hawaii Islands.

Specimens examined. — HAWAII: Kauai Island, under beach rocks at mile marker 9.4 N of Kapaa, 14
Jan. 1988 (J. W. and E. R. Berry), 7 males, 5 females, 4 immatures. Hawaii Island, in lava on beach,
25 road miles NW of Kailua at Anaehoomalu, 18 Jan. 1988 (J. W. and E. R. Berry), 1 male, I female,
2 immatures.

LITERATURE CITED

Beatty, J. A. and J. W. Berry. 1988a. The spider genus Paratheuma Bryant (Araneae, Desidae). J.
Arachnol., 16:47-54.

Beatty, J. A. and J. W. Berry. 1988b. Four new species of Paratheuma (Araneae, Desidae) from the
Pacific. J. Arachnol, 16:339-347.

James W. Berry, Department of Biological Sciences, Butler University,
Indianapolis, Indiana 46208 USA; and Joseph A. Beatty, Department of Zoology,
Southern Illinois University, Carbondale, Illinois 62901 USA.

Manuscript received December 1988, revised March 1989.

1989. The Journal of Arachnology 17:366

PREY OF THE CRIBELLATE SPIDER, DICTYNA ANNULIPES
(ARANEAE, DICTYNIDAE), ON APPLE TREE FOLIAGE

Cribellate spiders belonging to the genus Dictyna are often abundant on the
foliage of fruit trees and other plants where they spin their webs on the upper
surface of partially rolled leaves, in the axils of twigs and on the bark (Chant
1956; Putman 1967; McCaffrey and Horsburgh 1980; Temerak 1981; Nuessly and
Golden 1983; Bostanian et al. 1984). Putman (1967) reported that D. annulipes
(Blackwall) was a common spider on peach bark, and also recovered D. foliacea
(Hentz) and D. sublata (Hentz) by limb jarring.

In 1988, D. annulipes was the predominant spider {162%, N - 235 spiders) on
the foliage of dwarf apple trees, cultivars McIntosh and Empire, at Jordan
Station, Ontario, To assess the importance of this spider as a general predator in
apple orchards, determination of its prey was undertaken by analysis of carcasses
in its web and by assays of gut contents by immunoelectroosmophoresis as
described by Allen and Hagley (1982). Webs (x no. per tree - 32, N = 1 trees)
were examined at two- to three-day intervals and prey carcasses identified in situ,
or the webs removed and examined microscopically in the laboratory. Spiders
removed from the webs were placed individually in No. 000 gelatin capsules and
frozen (— 15®C) immediately. Serological tests were subsequently performed on
individual spiders to determine if they had fed on insects in the orders Diptera,
Hemiptera/ Homoptera, Hymenoptera and Lepidoptera using antisera prepared to
the apple maggot, Rhagoletis pomonella (Walsh), the green apple aphid. Aphis
pomi DeGeer, the braconid, Pholetesor ornigis (Weed) and the leafminer,
Phyllonorycter blancardella (Fabr.), respectively.

Chironomids (Diptera), (species undetermined), were the major prey (70.7%, N
= 222 prey cadavers) of D. annulipes as determined by web analysis. Putman
(1967) also suggested that chironomids were the major prey of web-spinning
spiders, including Dictyna spp., on peach trees. The second largest group of prey
were Hemiptera/ Homoptera (15.3%) of which the leafhopper, Typhlocyba
pomaria McAtee (5.88%) and the mirid, Campylomma verbasci (Meyer) (5.0%)
were the most frequently recovered species. Chant (1956) reported that D.
arundinacea L. fed on the mirid, Plagiognathus arbustorum Fab,, but the
anthocorid, Anthocoris nemorum L., was rejected. Chant (1956) also stated that
aphids and coccids were not particularly favored by D. arundinacea. Dictyna
annulipes apparently did not favor the green apple aphid, A. pomi, which
comprised only 2.7% of the prey in the webs. Although McCaffrey and
Horsburgh (1980) did not analyse the contents of the webs of D. foliacea and D.
sublata, they suggested that the webs would be most effective for the capture of
small, weak-flying insects, such as leafhoppers and aphids.

Other groups of prey in the webs of D. annulipes included Araneae (4.5%),
Hymenoptera (2.7%), Lepidoptera (1.8%), Neuroptera (0.9%) and other Diptera
(4.5%).

The proportion of D. annulipes giving positive serological reactions for various
prey groups is shown in Table 1. The greatest proportion of spiders tested
positive for the Hemiptera/ Homoptera group. The number of spiders that were
serologically positive to the antisera for Diptera decreased from May through

1989. The Journal of Arachnology 17:367

Table 1. — The proportion (N = 177) of D. annulipes that were serologically-positive for four prey
groups.

Month

No. Z). annulipes
tested

Diptera

Hemiptera/

Homoptera

Hymenoptera

Lepidoptera

May

24

12.5

41.7

8.3

8.3

June

89

9.0

36.0

16.9

19.1

July

33

6.1

39.4

30.3

15.2

August

31

1.2

14.3

13.0

31.2

August, probably due to the fact that chironomids were most abundant in late
April and May (unpublished data). Chironomid carcasses accumulated in the
webs over time, as they were not removed by D. annulipes. Feeding on
Hymenoptera and Lepidoptera increased as the season progressed and probably
reflects an increase in the density and activity of both predator and prey
(unpublished data). Of 177 D. annulipes that had fed, 75.7% consumed
individuals of one prey group, 20.9% of two prey groups and 3.4% of three prey
groups.

We thank D. Barber, R. Luft and R Richards for technical assistance and C.
D. Dondale, Biosystematics Research Centre, Agriculture Canada, Ottawa, for
identifying the spiders and for his comments on an earlier draft of this paper.

LITERATURE CITED

Allen, W. R. and E. A. C. Hagley. 1982. Evaluation of immunoelectroosmophoresis on cellulose
polyacetate for assessing predation of Lepidoptera (Tortricidae) by Coleoptera (Carabidae) species.
Can. Ent., 114:1047-1054.

Bostanian, N. J., C. D. Dondale, M. R. Binns and D. Pitre. 1984. Effects of pesticidae use on spiders
(Araneae) in Quebec apple orchards. Can. Ent., 116:663-675.

Chant, D. A. 1956. Predacious spiders in orchards in south-eastern England. J. Horticultural Sci.,
31:35-46.

McCaffrey, J. P. and R. L. Horsburgh. 1980. The spider fauna of apple trees in Central Virginia.
Environ, EntomoL, 9:247-252.

Nuessly, G. S. and R. D. Golden. 1983. Spider predation on Coleophora parthenica (Lepidoptera:
Coleophoridae), a moth imported for the biological control of Russian Thistle. Environ. EntomoL,

12:1433-1438.

Putman, W. L. 1967. Prevalence of spiders and their importance as predators in Ontario peach

orchards. Can. Ent., 99:160-170.

Temerak, S. A. 1981. Beneficial and harmful insects as prey of the web-building spider Dictyna sp.
(Arachnida, Dictynidae) on pomegranate trees in Egypt. Anz. Schadlingske, Pflanzenschutz,
Ummeltschutz, 54:90-93.

Elmer A. C. Hagley and Wayne R. Allen, Agriculture Canada, Research
Station, Vineland Station, Ontario, Canada LOR 2E0.

Manuscript received January 1989, revised March 1989.

1989. The Journal of Arachnology 17:368

OBSERVATIONS ON THE SOCIAL SPIDER,
ANELOSIMUS DOMINGO (ARANEAE, THERIDIIDAE),
IN SOUTHWESTERN PERU

Although Anelosimus domingo Levi (Araneae, Theridiidae) was described in
1963 (Levi 1963), little is known about its natural history. It is now evident that
Anelosimus saramacca Levi and Smith (1982) is a synonym of A. domingo Levi
(1963). (NEW SYNONYMY: Levi, per. comm.). Levi and Smith (1982) mention
that the species is cooperatively social and include some habitat features in their
description. Here we intend to definitively document the cooperative foraging in
this species and report some preliminary observations of its web structure and
behavior patterns in relation to other Anelosimus species.

In 1987 and 1988 we discovered a total of five communal webs of A. domingo
in the forest undergrowth vegetation of the Tambopata Reserved Zone (12°50'S,
069°17'W), Madre de Dios, Peru. This 5,500 hectare reserve consists of pristine
Amazonian forest on the border between the Tropical and Subtropical Zones as
classified by Holdridge et al. (1971). All of the A. domingo webs that we found
were located in the upper flood plain forest type at Tambopata. This forest type
is created by flooding that occurs periodically in certain areas of the Amazon
basin creating a secondary flood plain with soils richer than upland forest (Erwin
1984). This habitat, which has a very diverse plant community, is dominated by
palms (Iriartea) of medium height (25-35 m tall). Each web was positioned in the
undergrowth such that it was exposed to direct sunlight from zero to a maximum
of three hours each day. A. eximius appears to have a much broader habitat
range as we found its webs in five of the seven forest types described at
Tambopata as well as along lagoons and paths where they could be exposed to
sunlight 6 or 7 hours in a day. Another social species in the same genus found at
Tambopata, A. rupununi, was only found in clearing areas or along bodies of
water where they were exposed to direct sunlight nearly all day (pers. observ.).
We found the first two webs of A. domingo in June 1987 and observed them
periodically between that time and August 1988. By July 1988, one of those webs
had deteriorated and disappeared and we had found three additional webs. Webs
were isolated and uncommon even within the appropriate forest type (smallest
nearest neighbor distance was 2.5 km).

The appearance of A. domingo webs is very similar to that of A. eximius (see
Brach 1975; Vollrath 1982; Christenson 1984 for descriptions of A. eximius webs).
All of the webs we found consisted of a single basket-shaped sheet of webbing
with dead leaves and adjacent vegetation incorporated as retreats. Above this
sheet was a non-sticky tangle of barrier webbing that can extend into neighboring
vegetation several meters (Table 1). We estimate that there were several hundred
individuals housed in each web, however, at any moment in time, only about 20-
40 spiders were distributed in obvious areas (Table 1). The rest of the spiders
were in retreats under leaves where many of them were situated in such a way
that they could efficiently respond to prey or other disturbances in the web. Also
in these retreats some adult females would be guarding egg sacs and caring for
very young juveniles. Because it was difficult to see into the retreats in the A.
domingo webs, we were unable to determine if females fed spiderlings by

1989. The Journal of Arachnology 17:369

Table 1. — Measurements in cm of A. domingo webs. Sheet measurements represent the length,
width and depth of that basket-shaped structure. The height of the web is given as the lowest point of
the sheet. The barrier extends up from the sheet into vegetation.

Web

Sheet

Height

Barrier

# Spiders visible
(mean ± SD)

1

80 X 81 X 22

40

200

32.5 + 11.3

2

65 X 70 X 40

70

140

27.5 ± 5.5

3

41 X 38 X 15

32

79

21.4 ± 12.3

4

110 X 140 X 35

12

290

39.3 ± 18.3

5

22 X 18 X 10

84

45

11.3 ± 6.1

regurgitation in a manner similar to A. eximius (Christenson 1986 and pers.
observ.).

Just as for A. eximius (Nentwig 1985), the cooperative behavior and large webs
constructed by A. domingo enables them to capture prey many times their own
body size and mass (Table 2). The prey capture sequence is virtually identical to
that we have observed in a sympatric population of A. eximius. When an insect
enters the web several spiders emerge and begin to subdue the insect by circling
and wrapping it with silk. Periodically individuals will move in and bite until the
insect is killed. Although not significant with our small sample size, it appears
that a larger number of spiders react in a shorter period of time to larger prey
(Table 2). It took more individuals of A. domingo significantly longer to subdue
large prey items (35-40 mm in length) in comparison to smaller prey (5-15 mm in
length) (Mann Whitney [/-test, p < 0.05) (Table 2). If the insect is relatively small
it is moved from an exposed area of the web to one of the more protected
retreats where spiders in all age classes can feed. Larger insects are left in the
lower portion of the barrier web and the spiders emerge from retreats to feed on
them there.

Interspecific tolerance and even interspecific cooperation have been reported
for other social spider species (Fowler and Levi 1979; Krafft 1970, 1975). We
found, however, that A. domingo and A. eximius are not tolerant of one another.
Adult females of A. eximius did not survive when introduced into the field webs
of A. domingo, although we were very successful at introducing spiders into
different colonies of the same species. Six laboratory experiments were conducted
in which one female of each species (matched by size) was introduced into a six-
dram vial measuring 2.5 X 4.5 cm (3-6 spiders of a single species lived well for
extended periods of time in vials of this size). In two experiments we introduced
the spider at the same time. In the first vial, the A. domingo had attacked the A.
eximius within the first five minutes and they had five aggressive interactions

Table 2. — Number of spiders and duration of the reaction period and handling time of prey items
introduced into A. domingo webs (mean + SD). a = Values for largest prey size are significantly

different from values for smaller categories by Mann-Whitney t/-test {P < 0.05).

Prey size

N

Reaction
time (sec)

Number of
spiders reacting

Handling
time (sec)^

Number of
spiders handling^

5-10 mm

4

0.6 ± 0.5

3.1 ± 1.2

201 ± 91

2.3 ± 3.3

10-15 mm

5

0.3 ± 0.5

3.3 ± 1.3

375 + 158

4.0 ± 8.2

35-40 mm

4

0.2 ± 0.7

7.1 ± 5.2

982 + 315

10.1 ± 3.2

1989. The Journal of Arachnology 17:370

within an hour. That female A. domingo killed the A. eximius in the third day.
In the other vial no activity was observed and, by the third day, the two spiders
had divided the space approximately equally with one spider at each end. In four
other vials we introduced one female first and allowed her 24 hours to establish a
small web inside the vial before we introduced the female of the other species.
Two vials received A. domingo first and two vials received A. eximius first. The
results of these experiments were symmetrical. Of the two vials in which A.
domingo established webbing prior to A. eximius, one ended up with a single A.
domingo and the other ended up with a single A. eximius after three days. The
outcome was identical for the vials in which A. eximius was introduced first. The
results of these experiments indicate that A. domingo and A. eximius will not
coexist peacefully or cooperate with one another. It is apparent, however, that
these species are capable of using one another’s web. Also there appears to be no
home bias among these spiders. That is, the spider spinning the initial web had
no advantage in the contests we set up.

The observations we have made firmly places A. domingo among the
cooperatively social spiders of the genus. Because sociality is well-developed
among Anelosimus spiders, comparative studies of all aspects of their biology
become important to our understanding of the evolution of social behavior both
within this genus and in other spiders as well. Certainly more detailed
investigations of A. domingo will help us understand some of the ecological
factors that influence the evolution of social behavior.

We would like to thank K. R. Cangialosi, G. J. Binford, J. L. Whitis, D. Silva
and K. Renton for assistance in the field. J. Coddington offered advice and
suggestions as well as identifying the spiders for us. H. W. Levi, T. E. Christenson
and W. Nentwig offered valuable suggestions on an earlier draft of this note.
Specimens were deposited at the National Museum of Natural History,
Smithsonian Institution. This research was supported by NSF grant BSR-
8604782, by the Hamilton Campus of Miami University and by the Department
of Zoology, Miami University, Oxford Campus.

LITERATURE CITED

Brach, V. 1975. The biology of the social spider Anelosimus eximius (Araneae: Theridiidae). Bull. So.
Calif. Acad. Sci., 74:37-41.

Christenson, T. E. 1984. Behaviour of colonial and solitary spiders of the theridiid species Anelosimus
eximius. Anim. Behav. 32:725-734.

Erwin, T. L. 1985. Tambopata Reserved Zone, Madre de Dios, Peru: history and description of the
reserve. Rev. per. Ent., 27:1-8.

Fowler, H. G. and H. W. Levi. 1979. A new quasisocial Anelosimus spider (Araneae, Theridiidae)
from Paraguay. Psyche, 86:11-18.

Holdridge, L. R., W. C. Grenke, W. H. Hatheway, T. Liang and J. A. Tosi, Jr. 1971. Forest
Environments in Tropical Life Zones: A Pilot Study. Pergamon press, New York, NY.

Krafft, B. 1970. Contribution a la biologie et a Pethologie d'Agelena consociata Denis (araignee
sociale du Gabon). 1. Biologica Gabonica, 4:197-301.

Krafft, B. 1975. La tolerance reciproque chez I’Araignee, sociale Agelena consociata Denis. Proc. 6th
Int. Arachn. Congr., VI: 107-1 12.

Levi, H. W. 1963. The American spiders of the genus Anelosimus (Araneae, Theridiidae). Tran. Am.
Micros. Soc., 91:30-48.

Levi, H. W. and D. R. R. Smith. 1982. A new colonial Anelosimus spider from Suriname. Psyche,
89:275-278.

1989. The Journal of Arachnology 17:371

Nentwig, W. 1985. Social spiders catch larger prey: a study of Anelosimus eximius (Araneae;

Theridiidae). Behav. Ecol. SociobioL, 17:79-85.

Volirath, F. 1982. Colony foundation in a social spider. Z. TierpsychoL, 60:313-324.

Ann L. Rypstra and R. Scott Tirey, Miami University, 1601 Peck Blvd.,

Hamilton, OH 45011 USA.

Manuscript received January 1989, revised April 1989.

EGG PREDATION BY CATOLACCUS PROB. N. SR
(HYMENOPTERA, PTEROMALIDAE) ON THE SPIDER,
DICTYNA COLORADENSIS (DICTYNIDAE)

This note documents predation by the wasp, Catolaccus prob. n. sp. on the
eggs of Dictyna coloradensis Chamberlin in northern Idaho and eastern
Washington. Egg sacs of D. coloradensis were collected from the rangeland
weeds, spotted knapweed {Centaurea maculosa Lamarck) and yellow starthistle
(C. solstitialis L.) (Asteraceae). One or the other of these plant species dominated
each collection site (see below). Web height, when collected from spotted
knapweed and yellow starthistle, were an average of (± SE) 64.7 (+ 7.2, « = 18)
and 39.2 (± 2.8, n = 36) cm, respectively. In spotted knapweed, the webs were
found in the stems, whereas in yellow starthistle they were located among the
flower spines.

Collections were made in three areas: (1) north Idaho panhandle (Farragut
State Park [8.3 km E of Athol] and Athol [1.6 km E of Athol] sites; Kootenai
Co.) where spotted knapweed comprised the primary web substrate; (2) eastern
Washington (Chief Timothy State Park [8.3 km W of Clarkston]; Asotin Co.)
and (3) north central Idaho (the Pond [5.1 km NW of Culdesac] and Central
Grade [6.4 km NE Hatwai] sites; Nez Perce Co.) where egg sacs were collected
from yellow starthistle substrates. Twenty webs were collected ca. weekly from
May through September and egg sacs present were examined for the presence of
the wasp.

Dictyna coloradensis egg sacs collected at various sites were found to be
infested with a predacious pteromalid, Catolaccus prob. n. sp. (based on the
taxonomic determination by Eric E. Grissell). Larvae of this pteromalid were
found feeding upon eggs within the egg sac. When an egg sac was infested, all
eggs were consumed by a single pteromalid larva. Although the pteromalid eggs
were never found, they were either layed singly or multiple ovipositions resulted
in the mortality of all but one individual as only a single pteromalid adult was
reared from all parasitized egg sacs.

The spider egg sacs were flocculent white, constructed of dense silk and always
attached adjacent to the web retreat in the center of the hackled-band, mesh web.
No external evidence of parasitism could be identified on the web or egg sacs.

1989. The Journal of Arachnology 17:372

thus we suspect wasp eggs were layed within the egg sac by the thin wasp
ovipositor (ca. 1 mm long).

Most species of Catolaccus are secondary parasites of ichneumonid or braconid
parasites of Lepidoptera or Coleoptera (Burks 1954). We found no other records
of spider egg predation by members of this genus. Another pteromalid,
Arachnopteromalus dasys Gordh, has been reported as an egg sac parasite egg
predator) of several uloborid spider species (Gordh 1976, 1983; Peaslee and Peck
1983); thus, predation of cribellate spider eggs by pteromalids may have evolved
more than once along independent lineages (G. Gordh, pers. comm.).

The percent parasitism of egg sacs collected at the different sites during 1982
were: Athol, 19% {n - 183); Farragut State Park, 23% {n = 13); Chief Timothy
State Park, 24% {n = 17); and 0% {n = 49) at the Pond site. During the 1983
season, for unknown reasons, the percent parasitism dropped dramatically:
Farragut State Park, 7% {n = 84); Central Grade, 0% (n = 26); and the Pond site,
0% {n = 22). Percent parasitism over the sites was 16% (n = 262) during 1982 and
5% {n - 132) during 1983. Wasp parasitism of 2 and 3 egg sacs located in single
webs occurred three times each. Dictyna coloradensis females produce an average
of 2-3 egg sacs (Wheeler 1985); thus, when attack occurs, egg predation may
significantly reduce reproductive success of this spider. However, the cause(s) of
the decline in parasitism rates (e.g., weather, defensive response, alternate hosts)
between the two years is yet unknown.

Adult wasps were occasionally observed alighting and walking freely on D.
coloradensis webs in the laboratory and in the field, apparently without
provoking the spider. The pteromalid was never found ensnared in the spider
webs. Field observations of D. coloradensis indicate that the adult females remain
in the web with the eggs and the first two instars until spiderling dispersal,
perhaps to assist in prey capture or to guard the offspring against natural enemies
(Wheeler 1985). On two occasions ants entering the webs of mature females were
confronted with apparently defensive charging and web tugging by the spider.
After 5 min of observation in both cases, a standoff resulted where neither the
ants nor the spiders appeared to gain an advantage. However, based upon our
wasp observations, protection from this egg predator may be ineffective due to
wasp adaptations for walking freely on webs and the lack of spider defensive
responses to wasp infestation.

Gratitude is extended to W. J. Gertsch and E. E. Grissell for determination of
the spider and pteromalid species, respectively. Gordon Gordh provided
additional taxonomic assistance with the pteromalid. Wasp and spider specimens
are deposited in the W. F. Barr museum, University of Idaho. Financial support
was provided, in part, by a grant (USDA-ARS, Fed. Grant #58-91H2-4-0007)
awarded to R. H. Callihan and J. B. Johnson and Idaho Agricultural Experiment
Station Project 061-R838 (J. P. McCaffrey), a contributing project to Western
Regional Project, W-84. This is University of Idaho Agricultural Experiment
Station Publication number 8976. We appreciate the critical reviews of earlier
drafts of this manuscript by James B. Johnson and Dennis J. Schotzko.

LITERATURE CITED

Burks, B. D. 1954. Parasitic wasps of the Catolaccus group in the Americas. USDA Tech. Bull. No.

1093.

1989. The Journal of Arachnology 17:373

Gordh, G. 1976. A new genus of Pteromalidae from Missouri, the type-species of which parasitizes
Uloborus octonarius Muma (Hymenoptera: Chalcidoidea; Araneida: Uloboridae). J. Kansas
Entomol. Soc., 49:100-104.

Gordh, G. 1983. New distributional and host records for Arachnopteromalus dasys Gordh
(Hymenoptera: Pteromalidae), an egg-sac parasite of uloborid spiders (Araneida). Proc. Entomol.
Soc. Washington, 85:181.

Peaslee, J. E. and W. B. Peck. 1983. The biology of Octonoba octonarius (Muma) (Araneae:
Uloboridae). J. Arachnol., 11:51-67.

Wheeler, G. S. 1985. The bionomics of two spiders, Dictyna coiorademis Chamberlin and D. major
Gertsch (Araneae: Dictynidae): potential antagonists to the biocontrol of weeds. M.S. Thesis,
University of Idaho, Moscow, 103 pp.

G. S, Wheeler and J. P. McCaffrey, Department of Plant, Soil, &
Entomological Sciences, University of Idaho, Moscow, Idaho 83843 USA.
(present address of senior author: Department of Entomology and Nematology,
Bldg. 339, IFAS (0711), University of Florida, Gainesville, Florida 3261U0711
USA).

Manuscript received February 1989, revised May 1989.

1989. The Journal of Arachnology 17:374

BOOK REVIEWS

Ono, Hirotsugu. 1988. A Revisional Study of the Spider Family Thomisidae
(Arachnida, Araneae) of Japan. National Science Museum, Ueno Park 7-20,
Tokyo 110, Japan (Unpriced).

This is a coherent and critical treatment of the Thomisidae found in Japan.
Based on a solid background of painstaking genus-by-genus revisionary work
extending over the past nine years, the book provides workers with the
taxonomic knowledge needed to identify 22 genera and 53 species of Japanese
crab spiders. It is noteworthy that Ono has studied material and literature from
regions beyond the confines of his own country, thus presenting a broad
treatment that will interest arachnologists the world over.

On opening the book the reader is greeted by three strikingly beautiful color
photographs. One shows Xysticus croceus with its fangs buried in the thorax of a
lycaenid butterfly, another a pair of Oxytate striatipes suspended in copula on a
thread. The body of the book is illustrated with high-quality line and stipple
drawings showing the habitus, the male palpus in two views, and the epigynum in
external and internal views. These are supplemented by a table of anatomical
terms which Ono uses in a sense different from that of other workers. The range
maps show only collection localities within the country itself, though range
statements give the world distribution. There is no attempt to illustrate
intraspecific variation. For illustrations of eight of the species the reader is
referred to earlier papers by Ono.

The descriptions and synonymies are quite full. The inclusion of nomina nuda
is a dubious practice, however, and there are no references to the world
catalogues of Bonnet, Roewer, Brignoli, or Platnick. There are many new
synonyms based on examination of types or of presumably reliable publications.
Ono gives leg setal placement in great detail, though based on single specimens,
and no use is made of these in keys or diagnoses. Measurements are given as
simple ranges of values with no indication of sample size; a mean and standard
deviation are preferable, as these enable users of the book to make decisions
about size differences. The biological notes on each species are useful.

The approach to the problem of intergrading genera is interesting. For
example, the assemblages known under the names Xysticus (with 350 described
world species) and Coriarachne (with only six) if distinguished in the usual way
leave some species that do not fit very well. Ono’s solution is to interpose a third
genus, Bassaniana, for some of the intermediates, based on relative degree of
body flattening and relative lengths of the third and fourth pairs of legs. Thus the
East Asiatic Coriarachne decorata sits uneasily with the North American C.
versicolor, C. utahensis, and C. floridana in Bassaniana, and the European C.
depressa with some Xysticus spp. in Coriarachne. As expected, it becomes more
difficult to distinguish the three genera than the two. The characters are subject
to exceptions, and for some species the character has never been made known.
Such a classification needs more work to make it endure.

1989. The Journal of Arachnology 17:375

Ono proposes the Clubionoidea, or part thereof, as sister group of the
Thomisidae. He recognizes the heterogeneity of the clubionoids. The most
primitive subfamily of thomisids is thought to be the Stephanopinae, and the
most specialized the ant mimics of the subfamily Aphantochilinae. The
synapomorphy for Thomisidae is the ambush type of prey capture, which arose
with the exaggerated development of the first two pairs of legs as a powerful
grappling device. Members of the family also share an unusual development of
the anterior lateral eyes and diurnal hunting behavior. The most speciose
subfamily in Japan, as in the world, is the Thomisinae, with 17 genera and 44
species. Ono provides an arrangement of the world thomisine genera into 13
tribes, but without characters to distinguish them.

The island chain forming Japan extends approximately from latitude 25° to
45° north, and experiences climates from the humid subtropics to the cold
temperate. In an interesting section at the end of the book, the author speculates
on the origins of the Japanese thomisids. The majority of species are thought to
represent the Old World tropics, several of which have spawned Japanese
endemics. Xysticus daisetsuzanus is quite unique, being found only on Mt.
Daisetsu-zan in central Hokkaido and having its closest relatives in the
circumpolar region; it is evidently a postglacial relict. Interesting also are species
like Xysticus saganus, whose closest relatives are found in North America, much
as in Antrodiaetus and certain Opiliones. Only two (not three as stated) are
Holarctic, namely, Misumena vatia and Ozyptila sincera.

All in all, the work is valuable, and the author is to be congratulated.

Charles D. Dondale, Biosystematics Research Center, Research Branch,
Agriculture Canada, Ottawa, Ontario, KIA 0C6, Canada.

Hammen, L. van der 1989. An Introduction to Comparative Arachnology. SPB
Academic Publishing, The Hague. 576 pp. 302 figs. (Price $150.00).

Progress in comparative arachnology suffers from the absence of an
authoritative and accessible introductory text that outlines our current
understanding of arachnid evolution, explains important controversies and
suggests profitable avenues for research. Despite the promise of its title, van der
Hammen’s book is not an attempt to fill this vacuum. In fact, the author informs
us at the outset that it is “not a handbook but a general survey of personal
insights.” It is primarily a review of van der Hammen’s descriptive studies, and
the author makes no real effort to summarize discoveries or opinions of other
workers. Arachnologists familiar with van der Hammen’s research will find little
new information, but the book may serve as a reference for those requiring access
to a summary of van der Hammen’s contributions to arachnology.

The book is divided into two sections, a 70-page “general part” and a 500-page
“systematic part”. There is a list of references that includes most of the important
work in comparative arachnology.

The general part summarizes basic aspects of arachnid biology, concentrating
on areas of particular interest to systematists (external morphology, reproduction,
postembryonic development and phytogeny), but there are no discussions of
paleontology or biogeography. Van der Hammen gives little attention to internal

1989. The Journal of Arachnology 17:376

morphology or to arachnids as living organisms (i.e., behavior, functional
morphology, ecology, physiology).

In his overview of arachnid morphology, van der Hammen summarizes the
Bauplan of each order, concentrating on tagmata, mouthparts, appendages, coxal
glands and genital structures. This section is not appropriate as an introduction
to arachnid morphology, as van der Hammen assumes that the reader is already
familiar with the higher arachnid taxa and presents his own controversial
interpretations with an air of certainty generally reserved for time-tested
hypotheses (e.g., “The problem of the homologization of leg segments is now
definitely solved”). It is unlikely that a nonspecialist would be able to discriminate
arachnological “facts” from van der Hammee’s proposals. Throughout this
summary the author uses an original system of terminology that he hopes will
facilitate comparisons between arachnid orders, but he does a great disservice to
himself and the reader by not including a glossary or index. Terms are sometimes
defined parenthetically when introduced, but this is clearly inappropriate for what
is basically a reference book that will not be read from cover-to-cover.

The section on postembryonic development outlines a general framework for
the comparative study of arachnid life cycles, Acarologists are ahead of most
other arachnologists in extracting information from development for use in
systematics and for addressing general evolutionary topics, such as heterochrony.
The issues van der Hammen addresses may inspire araneologists,
scorpionologists, etc. to pay closer attention to preadult instars.

As van der Hammen is best known for championing the view that Acari is
diphyletic and for his novel reconstructions of arachnid ordinal relationships, I
had expected a more thorough treatment of these subjects within the section
devoted to phylogeny. Here the author describes his latest phylogenetic
hypothesis and outlines the characteristics of new superordinal taxa, but he does
not provide the kind of explicit analysis that most systematists have come to
expect. Although he occasionally uses the language of phylogenetic systematics,
van der Hammen rejects cladistics as an “atomistic” methodology that leads to
“highly artificial classifications”. He favors a “structuralist” approach in which
higher taxa are recognized by unexpressed potentialities (“deep structure”) rather
than their observable manifestations. In the absence of a more complete
discussion of his methodology and its application to arachnids, van der
Hammen’s phylogenetic hypotheses are not convincing.

In the systematic part, van der Hammen describes the external morphology of
one or two representatives from each of the generally recognized arachnid orders
and major subordinal taxa, giving special attention to 'primitive’ species. Each
order is treated in a separate chapter, and chapter contents are organized to
facilitate comparisons. In keeping with the author’s specialty, over 300 pages are
devoted to mites. His descriptions of several phylogenetically significant mite taxa
(opilioacarines, holothyrids and a primitive actinedid) are especially thorough.
Chapters dealing with the remaining arachnids and Limulus are either exact
reproductions or slightly expanded versions of the author’s four-part series
“Comparative Studies in Chelicerata”. These chapters contain some novel
observations and insights, but it seems that their main purpose is to encompass
nonacarines within the author’s system of terminology rather than to make
original contributions. The descriptions are accompanied by numerous well-
executed line drawings.

1989. The Journal of Arachnology 17:377

Van der Hammen’s comparative method has benefits and drawbacks. By
providing intensive descriptions of primitive or ‘‘typical” representatives from
each major group, he includes the basic morphology of each taxon as well as
potentially significant minutiae that a more synthetic approach would omit. One
has the satisfaction of being introduced to a real organism rather than a
generalized cartoon. On the other hand, van der Hammen tends to get bogged
down in details and leaves the task of critical synthesis largely to the reader.
There is a tendency to gloss over variation within diverse groups, and little
mention is made of phylogenetic relationships within orders.

In summary, the book is inappropriate as an introductory text. The author
tends to ignore or trivialize research and opinions other than his own, and he
frequently presents his untested proposals as accepted facts. In contrast, van der
Hammen’s talent for description and the phylogenetically significant taxa he
includes make this a useful reference work for comparative arachnologists. The
book’s price will probably limit its distribution to libraries, but, as it is basically a
review of van der Hammen’s earlier work, most of the valuable information is
available free of charge from the primary literature.

Jeffrey W. Shultz, Department of Zoology, The Ohio State University, 1735

Neil Ave., Columbus, Ohio 43210 USA.

Coddington, J. A. (Ed.). 1989. Spider literature: a computer bibliography,
version 1.0. Available from the editor at the Department of Entomology,
National Museum of Natural History, Smithsonian Institution, Washington, DC
20560 USA.

Arachnology, like many other endeavors, is being tremendously advanced by
the application of affordable and approachable microcomputer technology. The
bibliographic tool provided by this compilation is a superb case in point. Spiders,
and collections of them, are myriad; the problems arachnologists face in dealing
with tens of thousands of specimens and taxa are enormous. The difficulties
added in dealing with an explosively enlarging literature are equally acute.
Various aids exist, of course, including catalogs of taxa and indices of
publications, such as the annual compilations provided by the Zoological Record
and the Centre International de Documentation Arachnologique. Those aids
suffer, however, from limitations imposed by the printed form in which they are
distributed. Most notably, they individually cover limited time spans, and permit
efficient searching through relatively few (and predefined) routes.

In this computerized bibliography, most of those limitations are overcome. This
first public release includes over 11,000 citations to published papers, mostly from
1965 to date (although the complete bibliographies of Brignoli’s catalog,
extending back to 1940, as well as more recent supplements to it, are included).
All post-1965 publications cited in the CIDA lists and all post-1977 publications
cited in the Zoological Record (and many omitted by both those sources) are
included. The bibliography is distributed in the form of ASCII files on floppy
disks. It is most readily obtainable for MS-DOS systems, as the convenient IBM
extended character set, including many frequently encountered accented

1989. The Journal of Arachnology 17:378

characters, is used. The files can easily be moved to other hardware systems,
however, with special editing required only for those accented characters.

One could use these files as normal word-processing documents, and simply
search repetitively for particular combinations of characters (names of authors,
taxa, or places, for example). But they gain maximal utility when imported into
modern database software, including text-based systems (such as AskSam, by
Seaside Software), full relational databases (such as R:BASE, by Microrim), or
specifically bibliographic software (such as Paperbase, by Wight Scientific).
Importing the files usually requires minimal editing; for example, I was able to
import the full 1.8 million characters into R:BASE in less than 30 minutes, using
a powerful ASCII text editor (XyWrite III Plus, by XyQuest) to globally change
the characters marking the ends of each field and record. As distributed, each
reference includes three fields: one for the author(s) and year of publication, one
for the title, and one for the journal reference (or publisher) and keywords (if
available). Text-based software can absorb these fields directly; relational
database software may require users to specify a maximum number of characters
per field (which can waste much disk space), and deal individually with overly
long entries. Some relational databases, such as R:BASE, have a field type that is
flexible in size and can therefore handle all these references automatically, using
minimal disk space. The files are in native Paperbase format, and that software
(obtainable at a discounted price with the files) allows flexible output (in the
formats specific to particular journals) as well as multiple string searches and easy
file editing, indexing, and maintenance.

Once imported, one can easily obtain lists of all papers whose titles or keyword
references include any string of characters. Thus, one has at one’s fingertips
citations to all papers that meet a specified criterion: all papers authored or
coauthored by a given individual, for example (I’ve noted problems only in
alternative transliterations of author’s names from non-Roman — especially
Russian — alphabets), or all papers whose titles or keywords include a given taxon
name and/or geographic area (with allowances for linguistic differences, e.g.,
Brasil or Brazil), or all papers published in a given journal during specified years
(here again, standardization of journal abbreviations is occasionally a problem).

Coddington has recruited an international coterie of collaborators who are
hard at work extending the bibliography back in time (primarily through
incorporating the bibliographies in Bonnet’s catalog and similar sources), and
annual updates are planned. The compilation and updating are far from simple
processes; because of the variety of sources involved, discrepancies in format and
duplications of titles require careful attention — byte by byte, character by
character — to each entry. Persons wishing copies of the files can obtain them
merely by agreeing to help with these efforts.

The usefulness of this database cannot be overestimated. To give just one
example. I’ve recently been concerned with some kleptoparasitic spiders of the
family Mysmenidae; by merely searching the title and keyword fields for all
references that include the character string ‘"klepto” I was able to find several
important papers that might otherwise have been overlooked.

In short, Coddington has provided arachnologists with an enormously powerful
tool that should do much to overcome the hazards of the modern “information
explosion.” That he has done so within an academic framework not yet prepared
to reward such “electronic” publications is even more laudable. The advent of

1989. The Journal of Arachnology 17:379

such powerful handles on the literature makes it imperative that authors devote
some effort to incorporating, in their titles, taxon names and keywords that will
allow citations to be retrieved efficiently (hopefully, the days of “One Hundred

New Species of Spiders” are long gone!).

Norman 1. Platnkk, American Museum of Natural History, New York, New
York 10024 USA.

*. - tfc .t>i*S'

‘.tf <4)^

s ■'

V ••

r. ^

s : • ■ r.. I r ■ \i, , .:

'•■ ■-"’ : r V , V . i! :♦

■I ■ "JSVV-. :

■% «il

• r.-»-.-

*»• Vi-A;

% ‘.li

I i .. ' i: -’».f

:■ n-

t-v

T

; ”4

381

THE JOURNAL OF ARACHNOLOGY

Instructions to Authors

GENERAL COMMENTS

Manuscripts are acceptable in English, French, or Spanish, and must be TYPED DOUBLE- OR
TRIPLE-SPACED THROUGHOUT. Good quality paper should be used and erasable bond is
specifically excluded. Leave margins at least 1.5 in. (4 cm) on the left and I in. (2.5 cm) on the top,
bottom, and right. Do not hyphenate any words at the right margin. Avoid using right justification in
manuscripts prepared on word processors. Do not edit your own manuscript. Italics are permitted
only to indicate scientific names; only the TITLE, PRIMARY HEADINGS (e.g., INTRODUCTION,
etc.) and RUNNING HEADS should be typed in capital letters. Follow the recommendations of
Council of Biological Editors (CBE) Style Manual, Third Edition, unless indicated otherwise below.
To facilitate prompt review by two or more referees, send to the Associate Editor three identical
copies of the typed material, one of which should be designated as the “original,” together with copies
of the illustrations.

Manuscripts longer than about 1500 words (five double-spaced typewritten pages in Elite, six
double-spaced typewritten pages in Pica) should be prepared as Feature Articles, shorter papers as
Research Notes. Manuscripts and all related correspondence must be sent to the Associate Editor: Dr.
Jerome S. Rovner, Department of Zoological Sciences, Ohio University, Irvine Hall, Athens, Ohio
45701.

FEATURE ARTICLES

Arrange the various part of feature articles in the following sequence: (1) mailing address, (2) title,
(3) by-line, (4) abstract, (5) body of text, (6) acknowledgments, (7) literature cited, (8) figure legends,
(9) footnotes, (10) running head, (11) tables with legends, (12) figures. Put only parts 1, 2 and 3 on
page I, and start page 2 with part 4. All other pages are numbered consecutively.

Mailing address. — Include the complete address and telephone number of that author with whom
all correspondence with the Editors should be handled.

Title. — When using common or scientific names in the title, be sure to include in parentheses the
appropriate higher taxa (order, family, . . .). Include footnote indication if appropriate (e.g., to
acknowledge grant support), but type footnote separately (see part 9 above). The title should be typed
in capital letters and may not exceed 48 characters and spaces per line.

By-line. — Include the name(s) of author(s) as customarily written (less titles) and complete
address(es), including zip code or other postal designation. Include footnote indication(s) if
appropriate (e.g., to indicate change of address), but type footnote separately (see part 9 above).
Leave three spaces between title and by-line, and four spaces between name(s) and address(es).

Abstract. — The abstract should be a summary of the basic findings, and should not exceed 2 to 3
percent of the text in length. Papers in a language other than English must be accompanied by an
English abstract as well as an abstract in the language of the text. Papers written in English but
focusing on a subject pertaining to a country where another of the accepted languages is used, may
include an abstract in that language.

Body of text. — Use whatever form seems best to accommodate the necessary description of
research. Be concise. Conform to standard references for abbreviations, i.e. the CBE Style Manual for
scientific abbreviations, and a good language reference such as The Little, Brown Handbook for
abbreviations of English words. Use the metric system for all measurements (the English system is
acceptable only when transcribing locality data accompanying museum specimens), and note that
abbreviations of metric units of measurements are not punctuated (e.g., mm and km, but ft. and mi.).
Citations should be in the following form: Bellrose (1950); Bellrose (1950:33); or (Bellrose 1950). The
complete scientific name of a species or genus, including author(s), must be given the first time they
are mentioned in the text. Use single-line notation for fractions [e.g., 1/4 and not 'A; (4-12)/ 3 and not

4U2 j

The following special directions apply to authors of taxonomic papers:

(a). Do not use abbreviations in a primary heading to indicate that a new name or a new combination
is being proposed (e.g., A-us x-us, new species, rather than A-us x-us, n. sp. or comparable
abbreviations).

382

(b) . Keys must be typed as follows:

1. Use Arabic numerals to designate the leading entry of the couplet. 2

Do not designate the second entry of a couplet, either by means of numbers, letters or other
marks x-us

2. Type numbers flush to left margin, and start entry on fifth space. A-us y-us

Subsequent lines of any entry must also be idented five spaces. A-us z-us

(c) . Synonymies must follow the abbreviated style shown below:

A-us x-us Jones 1930:3, 1935:9; Russell 1945:453; Smith 1954a: 16, 1954b:678; Cooper and Lim
1955:18 (in part).

A-us y-us Bates 1932:18, fig. 4. NEW SYNONYMY.

A-us z-us\ Miranda 1948:98 (misidentification); Harris 1951:3 (in part?), {nec A-us z-us
Zimmer).

(d) . Lists of specimens examined of a given taxon must be the last items typed in the treatment of
that taxon as they will be set in smaller type. Adhere to the following style for listing specimens
examined: Country: State or comparable political subdivision; County or District, detailed locality
(elevation), 14 July 1945 (collector): 2 males, 5 females (acronym of institution where specimens are
deposited), next detailed locality within that County, etc.; next County in the same State; etc.: next
State in the same Country: etc. Next Country: etc. Punctuation rules are very simple. Use a period to
separate countries, a colon to separate states, a semi-colon to separate counties, and a comma to
separate specific localities.

Acknowledgments. — Avoid overlooking persons who have in some substantial way assisted with the
work. Authors of taxonomic papers should spell out the name and indicate parenthetically the
acronym of institutions where specimens studied are deposited, if not mentioned elsewhere in the text.

Literature Cited. — Include only those publications to which reference is made in the text. Adhere to
the CBE Style Manual and refer to a previous issue of The Journal of Arachnology for style. Do not
abbreviate place names in journal citations. Repeat name(s) of author(s) in case of multiple entries.

Figure Legends. — Provide one legend for each illustration which will be reproduced separately, or
for each “plate” consisting of several illustrations. Adhere to the following styles:

Figs. 1-4. — A-us x-us, male from Timbuktu: 1, left leg; 2, right chelicera; 3, dorsale aspect of
genitalia; 4, ventral aspect of abdomen.

Figs. 27-34. — Right chelicerae of species of A-us from Timbuktu: Figs. 27, 29, 31, 33, — Dorsal
views; Figs. 28, 30, 32, 34. — Prolateral views of movable finger; Figs. 27-28: A-us x-us, holotype male;
Figs. 29-30: A-us w-us, male; Figs. 31-32: A-us z~us, holotype male; Figs. 33-34; A-us t-us, male. Scale
= I.O mm.

Type all figure legends consecutively on the same page(s), double spacing lines within each legend
and leaving 4 spaces between legends. Keep in mind that 99 characters and spaces represent one
printed line, and that 4 mm for each printed line must be subtracted from the maximum height
permissible on full-page illustrations.

Footnotes. — Footnotes will be permitted on page 1 when it is appropriate to acknowledge grant
support and to indicate change of address, etc. No footnotes will be permitted on any other page.
Type footnotes together on a separate page.

Running head. — At the top of each right hand page will be printed the author(s)’ last name(s) and
an abbreviated title. This running head may not exceed 60 characters and spaces, must be typed in
capital letters, and a hyphen must separate the name(s) from the title.

Tables with legends. — Prepare all tables precisely as they are to be typeset. Construct tables as
simply as possible. Include the legend at the top of each. Make marginal notations in the text which
clearly indicate the appropriate point of insertion of each table. Note that in these instructions
“graphs” are regarded as “figures,” not “tables.”

The size of the printed page imposes a limit on the size of tables that can be accepted. In a normal,
upright table these limits are 99 characters and spaces per line and 56 lines in height, including legend.
Sideways tables are difficult to typeset and present problems during paste-up; therefore, they will be
accepted in exceptional cases only, as most tables that are “too wide” can usually be rearranged to fit
into the more desirable, upright position. Any table that exceeds the dimensions given above will not
be accepted for publication. Tables must be typed double-spaced throughout, and the legend must be
in the following style:

Table 1. — Incorporate into the legend all information necessary to make the data presented clearly
understandable. Do not use footnotes.

383

Figures, — Figures are the only parts of the paper for which the author is responsible for preparation
of camera-ready material, and therefore they must be executed with great care. Drawings should be
made with black drawing ink on high quality, smooth, white, heavy paper. Drawings made in some
manner not employing black ink (e.g., shaded pencil drawings) are difficult to print, more so than
photographs. When such a technique is used, the drawings should be fairly dark, have good contrast,
and be well executed. The largest dimensions on the printed page on which illustrative material (with
legends) will be printed is 127 by 200 mm. Decisions as to size of originals, placement of several
smaller drawings on a single, page, etc,, should be based upon this format. Where several smaller
drawings are to appear together on a single page, it is the responsibility of the author(s) to furnish an
assembled “plate” precisely as it should appear in print. An effort should be made to efficiently utilize
the available space within the limits of the original “plate.” The “plate” must be prepared to allow for
inclusion of the legend beneath the drawings on the same printed page. Number all drawings
consecutively using Arabic numerals; do not designate illustrations with letters, and never designate
plates. Labeling should be neat, secure, and done with a method which produces fiat characters, e.g.,
pressure-sensitive transfer characters or a lettering apparatus employing black ink. Photographs
should have good contrast, be glossy or semi-giossy, and printed slightly larger (up to 2X) than
anticipated published size. Photographs and inked drawings should not be grouped together in a
single plate, because they will be difficult to print. Illustrative material of any kind larger than two
times the anikipated printed size will not be accepted. Larger illustrations are difficult to handle and
are easily damaged or lost in the mail. Authors preparing larger illustrations should consult local
photo shops and print firms about the availability of PMT’s (PhotoMechanicai Transfer) for reducing
their illustrations to an acceptable size, PMT’s are high contrast, direct positive-to-positive
reproductions on semi-glossy paper, which will also give the author(s) a good idea of how the
illustrations will look when printed. PMT’s can be obtained at a reasonable cost in a number of sizes,
usually 8 by 10 in. or 9 by 12 in., with any amount of figure reduction desired within that space.
Seeding PMT’s rather than originals avoids the loss of art work in the mail, and if the PMT’s get lost
or damaged they can be replaced. Illustrations, PMT’s, and photographs must be securely mounted on
stiff white paper or mounting board (Scotch Magic Transparent Tape can be used on the white
portions of illustration paper). On the back of each plate include author’s name(s), abbreviated title of
manuscript (or running head), figure numbers included, and orientation (= topside). Illustrative
material that is poorly prepared, too large, or not mounted properly will be returned to the author.

RESEARCH NOTES

Arrange the various parts of your research notes in the following sequence: (1) mailing address, (2)
title, (3) body of text, (4) by-line, (5) figure legends, tables with legends, and illustrations. Follow
instructions given above for feature articles unless otherwise indicated below. Do not include
footnotes anym^here. References to grant support and all other acknowledgments should be made in a
statement in the body of the text. If an indication of change of address is desired, it should be
included parenthetically after the credited institution as “present address.” For citation of literature,
follow the same instructions as for Feature Articles. The by-line must be typed in paragraph form
after the body of the text or the literature cited when one is present. When possible, Research Notes
will take priority over Feature Articles in the publication sequence.

0

i\

'■ r.vr-.^

.... V 4

r 2 ‘ ' ^'^•* ' .i(^' ^

f. {- . 'v.4 1.. *j4:v i* .'i*'^ i-ibo^'i

t- u ii

,)0*:,'. . L':..'.mi ,'•!*- .‘^-A»u

«•■ ' *'*-j^* ‘ ■' " I''*'

•' I' ^!j .#!?

.j', .‘u-. '45^

. tV..; . !»■>.- C;..i%t‘^ *:•

C^'t* <

r-GX" ' '** '■ ■*

'< ' ‘ ... - .I-..:. ,jrT-.i-..- '"•■■■■!^ '■• tj.' '

''•’■• '- : '' %^s\i.'>'l I f»-, ".J.I Cii'v'*,’'n'\; r», v*'i

. '.-■ ■* '...Y : 'If Ik; .’ j .. ., ‘ ■

‘ ' V ' - ^

;.,*• ' > -i^- ■. ■'■ -: ■••-f4^»1rti8lAl . ■>\:*.;rW^,. _. Ill

;»f w I '?’4.'*!ti3'5^p^^ 'f ^

•^' v . ^ >.

^3-.

^.' ». ' ^ - V -y^ ,;V7

' '1'

^. '. j

•I’l

■ Jff" , 'Vf^V «

>. 4m "

4m "r^ .,;.

■•«■>.’■>' ■fi/'ioi-i^sk:. f f-’' .. .-1; ;■'(>' ^ V.

^^'>^^•■'■5’

1 '■ .nr..: ;<* .Y-.; fj ^f4l illi' 6 " -'.f 'i j

• ' ' ' ■' 1 *' •
f.'v' ■' '^*i^f*-^''' -- ^

"■.."{4-- ' •• •..- j,.u.^-,-''';i'NjT-.*; i'l-4.^ tw ;.;;< ‘ 1

■ ' •■ ' .r..' .:* ■> r YS.>; .... ..rY'vn- ,S ^

"•■iv-.uV i,|H;,j.5 »'• V ■ i ^‘f***,^ y{r<-‘^^';J”.-,.. ;*s

y vi;?. ;.

' .V ■' * ' •'• qi|

’ ' ■' >’■■ ' ' 4 ' *■ Wv»ri ^ .oi'

^ N'-'..v 4^.,.- r. . .-

A,.!- ‘ ^ . . V '■'•* C"

', x - A"'

' ' “”:4 ■■ * . - ' '► ■'. |■♦*^^< ■ .r, ,*j.. jf/. .,

• ■' V f ^4 A'*’^ ^ . •' , #

■ .- ; 4:f ,- ,

f»..

1

f.- <»’‘.>‘iiy'« T^V-*- w^*y'

'.’, k ' ■ ■111 1 ■

M- .^%4^f|riispWp

THE AMERICAN ARACHNOLOGICAL SOCIETY

President:

Willkm A. Shear (1987-1989)

Department of Biology
Hampden-Sydney College
Hampden-Sydney, Virginia 23943

Membership Secretary:

Norman I. Platnick (appointed)

American Museum of Natural History
Central Park West at 79th Street
New York, New York 10024

Secretary:

James W. Berry (1987-1989)

Department of Zoology
Butler University
Indianapolis, Indiana 46208

Directors:

J. A. Coddington (1988-1990), James C. Cokendolpher (1987-1989), Jerome S.
Rovner (1987-1989).

Honorary Members:

P. Bonnet, W. J. Gertsch, H. Homann, H. W. Levi, G. H. Locket, A. F. Millidge,
M. Vachon, T. Yaginuma.

The American Arachnological Society was founded in August, 1972, to
promote the study of Arachnida, to achieve closer cooperation between amateur
and professional arachnologists, and to publish The Journal of Arachnology.

Membership in the Society is open to all persons interested in the Arachnida.
Annual dues are $30.00 for regular members, $20.00 for student members and
$50.00 for institutions. Correspondence concerning membership in the Society
must be addressed to the Membership Secretary. Members of the Society receive
a subscription to The Journal of Arachnology. In addition, members receive the
biannual newsletter of the Society, American Arachnology.

American Arachnology, edited by the Secretary, contains arachnological news
and comments, requests for specimens and hard-to-find literature, information
about arachnology courses and professional meetings, abstracts of papers
presented at the Society’s meetings, address changes and new listings of
subscribers, and many other items intended to keep arachnologists informed
about recent events and developments in arachnology. Contributions for
American Arachnology must be sent directly to the Secretary of the Society.

President-Elect:

George W. Uetz (1987-1989)
Department of Biological Sciences
University of Cincinnati
Cincinnati, Ohio 45221

Treasurer:

Gail E. Stratton (1987-1989)
Department of Biology
Albion College
Albion, Michigan 49224

Archivist:

Vincent D. Roth

Box 136

Portal, Arizona 85632

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 17 Feature Articles NUMBER 3

Nuevas especies del genero Philodromus (Araneae, Philodromidae)

de la region del Cabo, B.C.S., Mexico, Maria- Luisa Jimenez 257

Variacion relativa de caracteres somaticos y genitales en Grammostola

mollicoma (Araneae, Theraphosidae), Fernando Per ez-M ties 263

Biology of Peucetia viridans (Araneae, Oxyopidae) in South Carolina,
with special reference to predation and maternal care, Marianne B.

Willey and Peter H. Adler 275

A revision of the genus Hentzia (Araneae, Salticidae), David B. Richman 285

Cocinachernes foliosus, a new genus and species of pseudoscorpion
(Chernetidae) from Mexico, Edna Hentschel and William B. Muchmore .... 345
Further revision of some North American false scorpions originally
assigned to Microcreagris Balzan, (Pseudoscorpiones, Neobisiidae),

B. P M. Curcic .351

Research Notes

A new spider, Paratheuma makai (Araneae, Desidae), from Hawaii,

James W Berry and Joseph A. Beatty 363

Prey of the cribellate spider, Dictyna annulipes (Araneae, Dictynidae),

on apple tree foliage, Elmer A. C. Hagley and Wayne R. Allen 366

Observations on the social spider, Anelosimus domingo (Araneae,

Theridiidae), in southwestern Peru, Ann L. Rypstra and R. Scott Tirey 368

Egg predation by Catolaccus prob. n. sp. (Hymenoptera, Pteromalidae)
on the spider, Dictyna coloradensis (Dictynidae), G. S. Wheeler and
J. P McCaffrey 371

Book Reviews

A Revisional Study of the Spider Family Thomisidae (Arachnida, Araneae)

of Japan, by Hirotsugu Ono, Charles D. Dondale 374

An Introduction to Comparative Arachnology, by L. van der Hammen,

Jeffrey W. Schultz 375

Spider Literature: A Computer Bibliography, version 1.0,
by J. A. Coddington, Norman /. Platnick 377

Other

Instructions for Authors 380

Cover photograph, female Cupiennius getazi feeding on Norops limifrons,

Le Selva Biological Station of OTS, Costa Rica, by J. S. Rovner
Printed by PrinTech, Lubbock, Texas, USA
Posted at Lubbock, Texas, 8 December, 1989

J

00 _ 00 ' £ (j) _

VINOSHIIIAIS SBI^Vdail LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHimS

^ . . ^ ^ 2

< \^ ^ y<^yiru>^ <r ^ e- <r

2 \T' ^

(f> ^ </) ■ " z c/)

MITHSONIAN INSTITUTION NOlinillSNI _ NVlNOSHillAIS S3ldVdan LIBRARIES SMITHSONIAN

cn

C/)

{/iNosHims saiavaan libraries smithsonian institution NoiiniiiSNi nvinoshiiims

m pq - NvOiii^ m

_(/)_(/)£ — (/)

\/1ITHS0NIAN INSTITUTION NOlinillSNI NVIN0SH1I^MS S3iavaan LIBRARIES SMITHSONIAN
z in z; ^ c/5 z .V . in

<C ^

>

(/> •“ Z CO Z CO “ ^

tfiNosHims S3 lava an libraries smithsonian institution NouniiiSNi nvinoshtiws

CO =; CO ~ CO

2 _J Z -J _

MITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIINS S3IBVaan LIBRARIES SMITHSONIA^

r- Z r- Z

Vk CD

m ^ m ' pq

00 £ 00 ' £ CO

viNosHims S3 lava an libraries smithsonian institution NoiiniiiSNi NviNOSHims

CO 2 , ^ Z V CO z

O

2 2 > ViwsS' 5 • )^'’ > ' 5

MITHSONIAN INSTITUTION NOlinillSNI NVINOSHil S 3 I a V H a IT L I B R AR I E s'^SMlTHSONIAN

CO A: CO (/)

- H

<1 <

viNOSHiims S3 1 a va an libraries smithsonian institution NoiiniiiSNi nvinoshiiws

Z r- z r- Z r"

m ^ XlOia5^ m

tl CO ~ CO £ CO

MITHSONIAN INSTITUTION NOlinillSNI NVlNOSHillAIS S3iavaan LIBRARIES SMITHSONIA^

CO z CO z ^

< i: .< s ,< , s:

^ C/)

ryan libraries Smithsonian institution NoiiniiiSNi nvinoshiiins SBiBVdan

z. ^ (/i z tn

LI I

CO ^ 2 CO •**“ Z w 2

UTION NOIinillSNI_NVINOSHimS SBIdVdail LIBRARIES SMITHSONIAN INSTITUTION

CO \ CO !:;</) —

aan libraries smithsonian institution NoiiniiiSNi NviNosHims saidvaan

r- 21 ^ 2 r- 2:

CD
70
>

TO

m ^'v. uA^ " "W*' 'CTu^ ^ rn

CO ± CO ± — CO

UTION NOlinillSNI NVINOSHlIl^S SBldVdail LIBRARIES SMITHSONIAN INSTITUTION

(o 5 , ^ IT r s

, _ X o

—

5>

CO 2 CO _

2 t i

r 2 \55L£S-^ > 2 >

ryan^LIBRARlEs'^SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHlIWs'^S 3 I d V d 3 11

~ (O X CO ~ . CO

X

CO .V
r-\

CO

UTION NOIini!lSNl"'NVlNOSHimS S3 I d V d 3 11 "" L I B R A R I ES^SMITHSONlAN^'lNSTlTUTION

dan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHlIl^S SlldVdail

X CO 2 V CO 2 CO

^ Q ^ O X O X

LIE

V-- X o X o o

z v?C4AJ^^ t 2! , t vMUW±7 z .

-g 2 N^OSV^^i?^ > ’

(O •^' X Eo 2 CO 2

JTION NOIinillSNI_NVlNOSHlllMS S3ldVdai1 LIBRARIES SMITHSONIAN INSTITUTION

^ CO X CO X CO X

^ ;;: /WW\ ^ M

^lA oc /i-A X ■— cc -w>

^ xm. <

'yan libraries smithsonian institution NoiiniusNi nvinoshliws saiyvaan

rn tn

— CO — CO . _

UTION NOlinillSNI NVINOSHilliMS SBIdVBan LIBRARIES SMITHSONIAN INSTITUTION

CO 2 CO 2 CO 2

^ . V - &<s 5 -

r\

CO