i liiSiliilipiiiilili^

iii!iiij||iliil|il|iiii^ili!:lii

iiiliSiiiiiiS^^^^

iiiilftil|liii|l||i|iii9i|iii

iiiliiiiiliillii®^

li|j|i|||ililii|i||ij®|ililp{i

p?iS|Sliiliii§iii!HiiSiliilii8^

iililiillipiii*^

iliiliiiliiliii^

liliBiliiliillilii

|iiC® :i5i!

;iSl ;ISs S ;P ' i;

^■na!'Ss1 i^i ?'«<! Pi p?

Iii:f iB ' RiRi'iTi? i -iy w Pi ! R

i ! i f.l '1 1 i v/i v'v ii 1 !!!'■; 4

/TSBBp'BIftPSPilipifpi:

iiiiliiiiiliiiiliipi

:i|p'i;pplE| ;p|pPp5;ii;! : tip

f ! ip i j r; -ii?-;'^' i'' • ! ipi'ppv Jip'*

illililiiiiiliiiiiil

CC
<

o S2 o 52 Q 52 3

1 1 ES^SMITHS0NIAN“'|NSTITUTI0N N0liniliSNl“‘NVIN0SHillfllS^S3 I a va a n“'LI B RAR I ES^SMITI

2 ^ 2: ^ ^ r- ^ 2 r- 2

o o m 9 O

_ in ~ c/) ^' £ c/) z

ISNI NVINOSHilWS S3liJVdan LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINi
2 ^ ^ ^ 2 t c/> 2

<

2

O

ES^SMITHSONIAN lNSTITUTI0N“’N0liniliSNI_NVIN0SHilWs‘”s3 I a Va 8 H^LI BRAR I ES^SMITI

=; ^ 2 \ in

LSNI NVINOSHilWS S3 1 a Va 811 LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINI

z <- Z t-

00

50

>

73

m ^ ^ \<i2iiisjiX m __

I Es'^SMITHS0NIAN“1NSTITUTI0N NOliniliSNI NVINOSHilWS S3iaVaan“LIBRARIES*^SMITI

in 2 ,v,' in 2 in 2 . </>

O X . o

to in i in

X o’ X

ii

iSNI_NVIN0SHilWs‘”s3 I ava 3 n_ LI B RAR I ES^SMITHSONIAN_INSTITUTION^NOIiniliSNI_NVIN

c

^ ij

o " "^iiiS]^ o

! I ES^SMITHSONIAN'^INSTITUTION N0lini!iSNl'^NVIN0SHiII^S^S3 I B Vd 8 n"^LI B RAR I ES^SMIT
2 [2 2 r- 2 r- 2

N - CD /^^5\ - m <*. 2 = x^5aTSv o

o;v ^ H 33

CO

H

t/i \.''s '

m ~ XjvAsv^ rn ' iti

CO £ CO £ CO

iSNI NVINOSHill^S S3IBVaan LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVIN

2 9 ^ 5 '» ^ 2 ^ CO 2

'*^'' I \ss^ I '■’’ I %■ I Vs^y I ^ I" ■''" %

MES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHilWS S3iavaan LIBRARIES SMIT

CO — CO 2 \ ^ ^ in

< it 5 < it si B i ^ <

iSNi NviNosHilVMs S3iavaan libraries Smithsonian institution NoiiniiiSNi nvin

i i s d

</i

5 p

o:

CO

m

27!^

O x^N.oc>^ ™ O

^“INSTITUTION NOIinilJ.SNl"*NVINOSHillfllS^Sl I UMH 8 n“'L| B BAR I ES^SMITHSONIAN“INSTITUT
z r-, z r- z f~

CD
33

% I

3

IJ t

CD

33

>

33

rn x:<vosH^ 2 r»i ' z jn

I S3 I M VH a n“L* B R AR I ES SMITHSONIAN"’lNSTITUTION^NOIinillSNl“NVINOSHiIiMS^ S3 1 H Vd

to . 2 •- CfJ 2 « C« 2 CD ,

X
CD

o
2

>*' s

I^INSTITUTION^NOlin iSNI_NVIN0SHiIIAIS”S3 I UVMa n“LI BRAR I ES"SMITHS0NIAN“|NSTITUT

«— m X V CD ^ “ ee% —

CD

1 5 5 S'" ' 5

2 »J 2 J 2 ^

S3 1 a vd an libraries Smithsonian institution NoiiniiisNi NviNosHims saidvd

z r- z f" ^ z r- z

m ^ ^ —

riNSTITUTION ^ N01iniliSNrNVIN0SHill/\IS S3 I dVd a H^LI B R AR I ES^SMITHS0NIAN”!NST1TU7

z .>0. CD z 52 Z CD 2

5“’s3l8V8an^LIBRARIEs‘”sMITHSONIAN INSTITUTION NOliniliSNI NVIN0SHilWs‘”s3 I a Vd

CD 5 CD 2 ^ ^ ^ ^ == ^

q:

:^) <>>

q: _

^“^INSTITUTION N0lini!lSNl”^NVIN0SHillNS^S3 I dVd 8 11*^11 B RAR I ES^SMITHSONIAN”"lNSTITUT
r* 2 r*, 2 r- 2

CD
3J
>

33

5 S3 I dVd an “"libraries SMITHSONjAN^INSTITUTION^NOlinillSNI^NVINOSHillAIS S3ldVd

^ ^ Z > CD ^ ^ 2 ^ CD 2 CD

X RS O 3 3

> 2 • 2 >' 2

J INSTITUTION ‘^NOliniliSNI NVlN0SHill^s‘^S3 I d Vd 8 IT^LI B RAR 1 ES^^SMlTHSONlAN INSTITUT

CD ^ Z \ ^ ^ ^ 5

LJ CD fn y<^a V??S. CJ

„ 2 J Z Ij Z

5 S3ldVdan LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHillMS S3ldVd
2 r* 2 r- Z

^ S /Sipm £ e

f

I

4

The Journal of
ARACHNOLOGY

VOLUME 21

1993

NUMBER 1

THE JOURNAL OF ARACHNOLOGY

EDITOR: James W. Berry, Butler University

ASSOCIATE EDITOR: Gary L. Miller, The University of Mississippi

EDITORIAL BOARD: A. Cady, Miami (Ohio) Univ. at Middletown; J. E.
Carrel, Univ. Missouri; J. A. Coddington, National Mus. Natural Hist.; J. C.
Cokendolpher, Lubbock, Texas; F. A. Coyle, Western Carolina Univ.; C. D. Don-
dale, Agriculture Canada; W. G. Eberhard, Univ. Costa Rica; M. E. Galiano,
Mus. Argentine de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia,
Missouri; N. V. Horner, Midwestern State Univ.; D. T. Jennings, Garland, Maine;
V. F. Lee, California Acad. Sci.; H. W. Levi, Harvard Univ.; E. A. Maury, Mus.
Argentine de Ciencias Naturales; N. 1. Platnick, American Mus. Natural Hist.;
G. A. Polis, Vanderbilt Univ.; S. E. Riechert, Univ. Tennessee; A. L. Rypstra,
Miami Univ., Ohio; M. H. Robinson, U.S. National Zool. Park; W. A. Shear,
Hampden-Sydney Coll.; G. W. Uetz, Univ. Cincinnati; C. E. Valerio, Univ. Costa
Rica.

The Journal of Arachnology (ISSN 0160-8202), a publication devoted to the
study of Arachnida, is published three times each year by The American Arach-
nological Society. Memberships (yearly): Membership is open to all those inter-
ested in Arachnida. Subscriptions to The Journal of Arachnology and American
Arachnology (the newsletter), and annual meeting notices, are included with mem-
bership in the Society. Regular, $30; Students, $20; Institutional, $80 (USA) or
$90 (all other countries). Inquiries should be directed to the Membership Secretary
(see below). Back Issues: Susan E. Riechert, Department of Zoology, Univ. of
Tennessee, Knoxville, TN 37916 USA. Undelivered Issues: Allen Press, Inc.,
1041 New Hampshire Street, P.O. Box 368, Lawrence, Kansas 66044 USA.

THE AMERICAN ARACHNOLOGICAL SOCIETY

PRESIDENT: Allen R. Brady (1991-1993), Biology Department, Hope College,
Holland, Michigan 49423 USA.

PRESIDENT-ELECT: James E. Carico (1991-1993), Department of Biology,
Lynchburg College, Lynchburg, Virginia 24501 USA.

MEMBERSHIP SECRETARY: Norman 1. Platnick (appointed), American
Museum of Natural History, Central Park West at 79th St., New York, New
York 10024 USA.

TREASURER: Gail E. Stratton (1991-1993), Department of Biology, Albion
College, Albion, Michigan 49224 USA.

SECRETARY: Brent Opell (1991-1993), Department of Biology, Virginia Poly-
technic Institute and State University, Blacksburg, Virginia 24061 USA.
ARCHIVIST: Vincent D. Roth, Box 136, Portal, Arizona 85632 USA.
DIRECTORS: George W. Uetz (1991-1993), Charles E. Griswold (1991-1993),
Jackie Palmer (1992-1994).

HONORARY MEMBERS: C. D. Dondale, W. J. Gertsch, H. Homann, H. W.
Levi, A. F. Millidge, M. Vachon, T. Yaginuma.

Cover illustration: A male Tetragnatha extensa from Carlisle, Massachusetts. Original color photo
by Joe Warfel of Arlington, Mass. Photograph made with a handheld Olympus OM-1 35mm camera,
macro lens, telescoping extension tube and manual flash.

Publication date: 4 June 1993

THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER.

1993. The Journal of Arachnology 21:1-5

VISUAL BRIGHTNESS DISCRIMINATION OF THE JUMPING
SPIDER MENEMERUS BIVITTATUS (ARANEAE, SALTICIDAE)

Klaus Tiedemann: Departamento de Psicologia Experimental, Institute de Psicologia,
Universidade de Sao Paulo, 05508 Sao Paulo SP Brazil.

ABSTRACT, It was observed that the jumping spider Menemerus bivittatus lives on light surfaces as well as
on dark surfaces, hunting prey which is lighter or darker than the surface the spider is on. From these observations
arises the question about the brightness or contrast discrimination abilities of this spider. The orientation response
was recorded for 14 spiders to a moving circular prey-stimulus varying from white, through grey, to black,
against a white, grey or black background. When the stimulus was darker than the background, there was a rapid
increase in response as the stimulus gets darker. This rapid change in response with stimulus brightness did not
occur when the stimulus was lighter than the background. These results reveal a high contrast discrimination
ability and also a dependence of the response on the overall stimulation conditions.

The visual system of jumping spiders (Salti-
cidae) is highly developed when compared to
other families of spiders. Like most spiders,
jumping spiders have four pairs of eyes. The most
specialized are the anterior median eyes (AM),
that are used in prey pursuit (Land 1971). It has
been suggested by Land that the AM are capable
of color vision. The other eyes function primarily
to detect prey movement and to elicit orientation
toward prey (bringing the prey into the visual
field of the AM eyes).

The jumping spider Menemerus bivittatus
(Dufour) is common in Southern Brazil and can
be found throughout the year in almost all homes
of Sao Paulo. Retreats are generally located in
the highest part of doors or window frames or
under an outside overhang.

Preliminary observations of the habitat pref-
erences of M. bivittatus revealed that (1) spiders
hunted on either a dark surface (e.g., black paint-
ed poles) or a light surface (e.g., wall of buildings)
and (2) on both of these surfaces both light col-
ored prey (e.g., small Diptera) and darkly colored
prey (e.g.. Mused) were taken. There appeared
to be no difference in prey catching efficiency
between the different backgrounds. Since M. bi-
vittatus has a light appearance (greyish brown
with black stripes) the spider is very conspicuous
on a black ground and almost invisible on light
colored walls.

These observations raised the question of
whether jumping spiders are capable of discrim-
inating differences in contrast between stimulus
and background. Since single visual receptor cells
are known to react with graded potentials to light

intensity, the physiological capability for con-
trast discrimination appears available. The most
likely mechanism is a neural circuit that en-
hances brightness differences and contrast in the
same way as is known for many vertebrates and
invertebrates. However, the question remains on
how contrast discrimination is integrated into
specific behaviors, such as prey catching and mate
recognition. The reflexive behavior of jumping
spiders to orient themselves and the AM eyes
toward the prey after its detection by the sec-
ondary eyes (Land 1972) is a very simple be-
havioral response that can be easily observed and
recorded in the laboratory.

The purpose of this experiment was to estab-
lish the psychophysical brightness discrimina-
tion function measured by this orientation re-
sponse for different levels of ground brightness.
Another question was to find if the function for
the discrimination of a light stimulus against a
dark background is symmetrical to a dark stim-
ulus against a light background. Lack of sym-
metry between the functions would indicate an
increased ability at prey detection in one of the
situations and might suggest that the spider would
have a preference for one of the hunting condi-
tions.

METHODS

Subjects.— Ten adult female and four adult
male jumping spiders, Menemerus bivittatus, were
collected on the campus of Sao Paulo University
and taken to the laboratory where they were held
individually in petri dishes in 12/12 h light/dark
illumination. Individuals ranged from 6-12 mm

2

THE JOURNAL OF ARACHNOLOGY

• L

Figure 1.— Apparatus used in the experiment. The spider (S) after being anesthetized was glued to the head
of the insect needle (I) that was inserted into the rod (R) that was adjusted so that the spider touched lightly the
styrofoam ball (B). Around the spider the painted cylinder (C) with the stimulus (St) that could be easily
interchanged, was positioned on the revolving turntable (T) driven by the motor (Mt). Observation was made
through the mirror (M). The lamp (L) provided constant and uniform light of 155 lux measured at 90° on the
surface of the styrofoam ball

in total length. Live Musca were offered to each
spider once a week. The spiders survived for
several months under these conditions making
it possible, if necessary, to divide the whole ex-
perimental procedure into several experimental
sessions on different days.

Experimental procedure. — Prior to an exper-
imental session the spider was lightly anesthe-
tized with carbon dioxide so that the head of an
insect needle could be glued with wax to the rear
part of the prosoma. Care was taken to avoid
covering the eyes with wax. After the spider com-
pletely recovered from anesthesia, the animal was
placed into the experimental apparatus (Fig. 1).
The experimental apparatus was basically a mo-
tor driven turntable on which three different cyl-
inders, constituting the background for the visual
stimuli presented to the spider, could be made
to turn in either direction. On the axis of the
cylinder a styrofoam ball (25 mm in diameter)
lay loosely in a Teflon cup. The spider, held by
the needle glued to its prosoma and an adjustable
rod, could walk on the styrofoam ball in any

direction, turning or rolling the styrofoam ball.
The slightest turning or walking movement of
the spider (and consequently of the styrofoam
ball) could be observed through a mirror at-
tached over the apparatus. Observation through
a mirror was preferred to direct observation in
order to reduce disturbances of the spider. Over-
all illumination was provided by a daylight ring
light (Toshiba, Japan ) producing an illumination
of 1 55 lux at the top of the styrofoam ball at 90°.

White, grey and black glass cylinders (250 mm
diameter, 300 mm height) were used for the
background against which the spider could see
the stimulus. The white cylinder had a reflection
density of d = 0. 1 1 , the grey 0.56, and the black
of 2.08. Density was measured with a reflection
densitometer (X-Rite Inc., Grand ville, Michi-
gan, USA, Model B318) calibrated for d = 0.00
with a standard calibration card (Ti02 coated).
The glass cylinders were painted from outside
(so that the inner glare was the same) except for
an 8 mm diameter “hole” left transparent, co-
planar to the spider, where the stimuli could be

TIEDEMANN-BRIGHTNESS DISCRIMINATION OF JUMPING SPIDERS

3

Table l.—Mean and SD of the response probability
for the white, grey and black backgrounds for each
presented stimulus, n = 1 4 for all cases.

Stim-
ulus
den- ,
sity

Background

White

Grey

Black

mean

SD

mean

SD

mean

SD

0.11

0.01

0.00

0.09

0.06

0.52

0.09

0.27

0.13

0.05

0.15

0.12

0.31

0.29

0.09

0.15

0.12

0.59

0.08

0.36

0.39

0.15

0.14

0.09

0.50

0.78

0.09

0.00

0.00

0.53

0.10

0.56

0.80

0.14

0.00

0.00

0.58

0.86

0.12

0.00

0.00

0.46

0.11

0.76

0.90

0.09

0.85

0.05

0.86

0.94

0.06

0.93

0.03

0.38

0.07

0.91

0.98

0.02

0.96

0.02

1.01

0.99

0.01

0.33

0.08

1.15

0.25

0.07

1.52

0.18

0.07

2.08

0.08

0.03

applied from outside without producing any
shadow or border. The stimulus thus subtended
an angle of about 4° at the spider’s eyes. With
the white background, stimuli of the following
reflection densities were used: 0.11, 0.27, 0.31,
0.36, 0.50, 0.56, 0.58, 0.76, 0.86 and 0.91. With
the grey background the same stimuli were used,
and also a darker one of 1.01 reflection density.
For the black background, stimuli of the follow-
ing reflection densities were used: 0. 1 1 , 0.3 1 , 0.50,
0.58,0.86, 1.01, 1.15, 1.52 and 2.08. The stimuli
were produced by painting white paper with the
same paint used for the glass cylinders, mixing
from the white and black paint in different quan-
tities.

Each animal was tested for the white, grey and
black backgrounds but in different orders. For
each background the stimuli were presented in
random order. Five trials were run on each stim-
ulus/background combination. Each trial con-
sisted of five complete turns of the cylinder (al-
ternating left and right turns). Each turn was
started with the stimulus exactly behind the spi-
der’s back, out of its visual field (this could be
verified by the fact that the spider never respond-
ed to a moving stimulus at that position) and
took around 20 seconds to be completed, re-
sulting in a mean angular velocity of the stimulus
of 1 8 deg/s. For every turn of the cylinder it was
manually recorded if the spider made a response
toward the stimulus or not. For each density val-

ue of the stimulus the overall response proba-
bility was calculated by dividing the number of
cylinder turns that elicited a response by the total
cylinder turns. With few exceptions, there was
no doubt that a movement of the spider was
directed to the moving dark or light spot. When,
for some reason the spider entered spontaneous
walking activity, the experiment was interrupted
until the spider assumed its typical alert posture.

After finishing the stimuli presentations for one
background density, the same procedure was re-
peated for a different cylinder. The spiders stayed
responsive for up to 3 or 4 hours.

RESULTS AND DISCUSSION

The mean and the standard deviation (SD) for
the response probabilities for each background
and stimulus are given in Table 1. The SDs cal-
culated among animals are fairly small, suggest-
ing that the recorded orientation response is rath-
er reflexive. The response probability curves for
the white, grey and black backgrounds are given
in Fig. 2.

The response curve for the white background
varies from almost zero for a white stimulus (zero
contrast with the background) to almost 100%
for a 0.9 1 stimulus density (maximum contrast).
This experimental condition would conform to
a situation were the spider hunts for dark prey
on a light background, which seems to be the
situation to which the spider is best adapted,
since the spider itself has a light color and most
of the prey like Musca are dark in color.

The response curve for the black background
has a quite different shape, not being just the
mirror image of the first curve as might have
been expected. Response probability declines very
slowly with increasing stimulus darkness. The
highest response probability is somewhat lower
than 60% for a light grey stimulus (reflection den-
sity of 0.27). For a lighter stimulus (0.11) that
produces a higher contrast with the black back-
ground, the response rate is even lower. The dif-
ference is statistically significant (Student, df =
13, r = 2.57, P < 0.05). Of the 14 spiders tested,
1 2 presented a lower response rate for the bright-
er stimulus. This suggests that the spider is not
responding just to the contrast value between
stimulus and background. The spider might be
responding in part to the appearance of the stim-
ulus disregarding the background, or some other
unknown factors are influencing its behavior.

The response curve for the grey background
shows, as expected, the lowest response proba-

4

THE JOURNAL OF ARACHNOLOGY

CO

<o

CO

a

o

a

U9

0.0 0.5 1.0 1.5 2.0

WHITE

GREY

BLACK

Relative Density

Figure 2.™ Response probability curves for the white, grey and black backgrounds (cylinders) to stimuli of
different greyness, from white to black.

bility for the stimulus density that equals the
background density (relative density of 0.50, 0.56
and 0.58) where the contrast is close to zero. The
part of the curve where the stimuli are darker
than the background has more or less the same
shape and steepness as the curve for the white
background. With reversed contrast, the highest
response probability is again not for the highest
contrast condition, but for stimuli in the relative
density range of 0.27-0. 36. Out of the 14 records,
for nine the response rate for the lightest stimulus
(0.1 1) was lower than for the next darker one,
two were equal and three reversed. The Student
test for the difference of the means for both stim-
uli revealed a near to significant probability (df
= 13, / = 1.98, P > 0.05). These results favor
the idea that the response is not only guided by
the contrast between prey and background, but
also by the absolute lightness of the prey.

The lower response maximum of 60% for the
stimuli against the black background could sug-
gest that the spiders are visually less alert or
aroused in this situation since the overall light
level within the black cylinder must be lower,
due to less reflection of the constant illumination
provided by the lamp on the top of the cylinder.

If this were the case, visual alertness would also
be lowered with the grey cylinder. Since in the
grey cylinder a response rate of 100% was
achieved for darker stimuli, visual alertness dif-
ferences, if any, caused by different illumination
levels, do not explain the results.

These overall results could be compared to the
description of Land (1972) who, in an experi-
ment where jumping spiders responded to black
or white stripes, found that the pursuit response
to black stripes was given to the leading edge,
while the response to white stripes to the trailing
edge. Land concluded that the stimulus must al-
ways move in such a way as to cause sequential
darkening of adjacent photoreceptors. The re-
sults of the present experiment could be ex-
plained in the same way, since the spider could
be responding to an edge of the stimulus, al-
though the stimulus was small. For the light stim-
ulus on the dark background, the spider would
be responding to the trailing edge of the stimulus
or, what would be exactly the same, to the leading
edge of the dark background. This would mean,
that the stimulus is now the “huge” background
which of course had lost its “prey characteristics”
to the spider, explaining the lower response rate

TIEDEMANN^BRIGHTNESS DISCRIMINATION OF JUMPING SPIDERS

5

for this condition. An alternative interpretation
would be that the spider in this experiment is
reacting to the small stimulus as a whole, as long
as it has enough contrast with the background,
but with a clear preference for dark stimuli on
lighter backgrounds. Drees (1952) showed that
the courtship behavior of Epiblemum scenicum
(Salticidae) reveals well-developed visual acuity.
Such acuity might also be important in prey cap-
ture. However, Blest (1985) found that the prey
capture sequence could be elicited by stimulus
shapes quite different from normal prey, as in
the study reported here. Therefore, it seems that
the orientation response which initiates both prey
capture and courtship is mainly guided by the
contrast of the stimulus against the background.

LITERATURE CITED

Blest, A. D. 1985. Retinal mosaics of the principal
eyes of jumping spiders (Salticidae) in some neo-

tropical habitats: optical trade-offs between sizes and
habitat illuminances. J. Comp. Physiol., 157:391-
404.

Drees, 0. 1952. Untersuchungen uber Bewegungsse-
hen und Optomotorik bei Springspinnen (Saltici-
dae). Z. TierpsychoL, 9:169-207.

Land, M. F. 1971. Orientation by jumping spiders
in the absence of visual feedback. J. Exp. Biol., 51:
471-493.

Land, M. F. 1972. Mechanisms of orientation and
pattern recognition by jumping spiders (Salticidae).
In Information processing in the visual systems of
arthropods, (R. Wehner, ed.). Springer- Verlag, Ber-
lin.

Manuscript received 15 October 1990, revised 3 No-
vember 1992.

1993. The Journal of Arachnology 21:6-22

CIRCADIAN RHYTHMICITY AND
OTHER PATTERNS OF SPONTANEOUS MOTOR ACTIVITY
IN FRONTINELLA PYRAMITELA (LINYPHIIDAE)

AND ARGYRODES TRIGONUM (THERIDIIDAE)

Robert B. Suter: Department of Biology, Vassar College, Poughkeepsie, New York
12601 USA

ABSTRACT. Endogenous biological rhythms are apparently found in all eukaryotic organisms. The most
ubiquitous of these, the circadian rhythm, functions to synchronize physiology and behavior with diel changes
in the environment. Data presented here demonstrate that, in the linyphiid spider Frontinella pyramitela (Wal-
ckenaer) and in the theridiid spidQV Argyrodes trigonum (Hentz), a circadian rhythm modulates locomotor activity
in some individuals but not in others. The data also show (a) that higher-frequency endogenous rhythms play
a part in determining the patterns of motor activity, and (b) that intervals between bouts of activity are influenced

by aperiodic processes that appear to be stochastic.

Much of the behavior of animals in the held
is closely correlated with environmental stimuli:
the detection of predators or prey stimulates flight
or pursuit, the presence of a rival results in agon-
ism, the arrival of a mate elicits courtship or
bonding rituals, and changes in the thermal en-
vironment lead to altered postures or positions.
The internal state of an animal (e.g., the time
since its last meal, the size of its gonads, or the
presents of parasites and pathogens) also influ-
ences behavior, sometimes profoundly.

Among the most ubiquitous of internal state
variables is the endogenous circadian clock, which
has been particularly well studied since the pi-
oneering work of Aschoflf (1954) and Biinning
(1963). Although the cellular and molecular bas-
es of biological clocks remain obscure, their util-
ity is well documented: they facilitate prediction
of periodic environmental events, make celestial
navigation possible, and enable the temporal or-
ganization of internal events. Because the overt
manifestations of these clocks are often subtle,
they are most easily and profitably studied in
animals isolated from confounding environmen-
tal stimuli, including conspeciflcs and such Zeit-
gebers as light cycles. Under those conditions of
isolation, an animal’s behavior is said to be fre-
erunning (i.e., showing the natural period of the
circadian clock).

In spiders, endogenous rhythmicity has been
suggested by data on diel rhythms under natural
conditions (e.g., in Amaurobius, Cloudsley-
Thompson 1957), and has been confirmed in a

very few species under constant conditions (e.g.,
in Cupiennius salei, Seyfarth 1980; for other ref-
erences, see Cloudsley-Thompson 1987). More-
over, in work with the linyphiid spider, Fronti-
nella pyramitela (Walckenaer), my laboratory has
found that internal programs quite distinct from
circadian clocks appear to govern the timing of
certain spontaneous (i.e., internally driven) be-
haviors related to courtship (Suter 1990) and co-
habitation (Suter & Walberer 1989). The present
study constitutes the beginning of an elucidation
of the programs, both circadian and other, that
underlie the onset and cessation of spontaneous
behaviors in F. pyramitela.

METHODS

I captured adult females of both F. pyramitela
andzl. trigonum in Dutchess County, New York,
during June, 1992, and immediately installed
each in an open-ended glass cylinder. I placed
the cylinder in an aluminum trough and inserted
that into a plexiglass holder which could house
eight such troughs. One end of each cylinder was
partially occluded by an infrared sensitive pho-
totransistor (Radio Shack SDP8403-301) and the
other end was sealed by a circular microscope
coverslip behind which was mounted an IR light-
emitting diode (Radio Shack SEP8703). The as-
sembled apparatus, with the simple electronic
circuits used to drive it, is shown in Fig. 1. I
enclosed the apparatus in a light-tight chamber
(20 cm X 24 cm X 18 cm) in which the relative
humidity was maintained at 1 00% and temper-

6

SUTER-PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

7

+12V

Figure 1. — Schematic diagram of the apparatus used to house spiders and detect their motion. IR emitters
(E) and detectors (D), coupled to a computer, formed the motion-detection system. The spiders themselves were
enclosed in glass cylinders 1 cm in diameter and 4.5 cm long.

ature was constant at 24 ±1 ®C. The electronic
driving circuits, which generate some heat, were
mounted outside the chamber. Output from each
IR phototransistor was digitized and recorded
by computer (hardware: Macintosh Ilci with Na-
tional Instruments NB-MIO-16 I/O board; soft-
ware: a customized data-logging program written
in Lab VIEW 2). At each of the eight channels,
amplitude was measured at 2 ms intervals (500
Hz) and the standard deviation of the amplitudes
collected during 1.6 s was the datum recorded as
an index of activity during that period. (Prelim-
inary tests had shown that, because of the mul-
tiple paths the IR light could take in a chamber,
grooming motions and other movements of ap-
pendages did not cause fluctuations in photo-
transistor output. Thus high-frequency fluctua-
tions were unlikely. Moreover, power spectra of
phototransistor outputs from chambers contain-
ing rapidly moving spiders recorded at 500 Hz
with no averaging revealed that there was no
significant energy at frequencies > 1.0 Hz).

The activity of 14 K pyramitela and 5 A. tri-
gonum was recorded during the study. The max-
imum continuous length of time in the chambers
was 1 1 days for a group of 7 F. pyramitela. While
in the chambers, the spiders were neither fed nor

watered. Because of the high relative humidity,
desiccation was not a problem for the spiders,
and fasting for periods longer than 1 1 days can
be tolerated by F. pyramitela (Suter 1985) and
by other species (Anderson 1974). Two separate
experiments involving F. pyramitela were run:
in one the spiders were maintained in constant
darkness (DD) throughout their 8 -day isolation,
and in the other the chamber enclosing the spi-
ders was lit for 8 hours each day for 1 1 days (LD,
onset of light daily at 1100 h; illumination by
orange, green, and red LEDs provided 7.7 lux at
each chamber). The single experiment that in-
volved A. trigonum was carried out in DD.

The digitized activity records of individual
spiders were analyzed in two ways. (1) Spectral
Analysis I: A Hamming window (to minimize
the artifacts caused by truncation of a signal) was
applied to the recorded string of activity /inac-
tivity periods which was then analyzed by fast
Fourier transform (FFT) and displayed as a pow-
er spectrum showing the proportion of explained
variance in activity as a function of frequency.
Significant peaks were identified by a chi-square
method described elsewhere (Suter & Forrest, in
press) which could reveal time-based periodici-
ties such as a circadian rhythm. For the detection

8

THE JOURNAL OF ARACHNOLOGY

Frequency (cycles/clay)

0 6 12 18 24

/"Aa/A aaA ^_^A_

aa/V.a-j>\.aW\ /VA\

.aA^AA\ AjVWyv A A^__^^>A_wW\A^^UlA/V_AJwlrwlwWv^^

^Ay^MAA^A/'^MAA^y' ^ajVIAvAAa-vWv/V.^^

_V\A/WS.lVyiA_^.A_A._A

12

18

24

Frequency (cycles/day)

Figures 2-8.— Activity of F. pyramitela under conditions of constant darkness (DD). Graphs on the left show
activity as a function of time, with days shown sequentially from top to bottom. FFT analysis of these data
resulted in the power spectra shown on the right in each figure. The power spectra are shown at two scales to
elucidate both the overall pattern (lower panel) and the peaks between 0 and 1 0 cycles/d (upper panel). In each
of these power spectra (as in all others in this study), the first component (furthest to the left) should be ignored
because it reflects the entire data set and not oscillations within the data set. Much of the explained variance is
concentrated near 1.0 cycle/d, but significant peaks are also found at higher frequencies. Peaks at or above the
horizontal lines are significant at a = 0.0 1 .

SUTER-PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

9

jI/lUIX

u

— I ill 1_1^

J jl

JlU.

lJI

YJ

U

0

6 12

18

24

La/\aAj^iwAjAJ'

a M a. aIk .Aj^AjAJ'

AJ\y\Ao^JJU_JLA__^/^V_A^M_AJ^^

AjAA^hL.^jJ\rK.

.A/Vl.i4_A.iyL

12

18

24

Frequency (cycles/day)

, 008 - — —

1

).004 -

!.002 - 1

,000

^AhaW

ill

0

,23456789 1i

8 8 8 g

o ro 0

k

0 20 40 60 80 1 00

Frequency (cycles/day)

Figures 2--8 . — Continued.

0 6 12 18 24

THE JOURNAL OF ARACHNOLOGY

Frequency (cycles/day)

0 6 12 18 24

Figures 2-8.— Continued.

SUTER- PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

11

/'A

.jUla^aaIja

L

hAi\

40

60

80

Frequency (cycles/day)

12

24

Figures 2-8. — Continued.

Aaa-M™<\.-Ca — Aa

AvWvwv

AMMiJiLhll

MIiMJa

iIamAIwwii^JxlJ A_jj__7vk^

vuvAaAal IU^_J — Jt_JU_vvAJAAI

A /

jvL-kjJi fWJL-_l\

^ iLlikl t T

12

18

24

Figures 9-15.— Activity of F. pyramitela under conditions of a 24-h photoperiod (LD) in which lights were
on between 1 1 00 and 1 900 h (arrows). See the legend for Figs. 2-8 for an explanation of the layout.

12

THE JOURNAL OF ARACHNOLOGY

A>VxA_A
JwnAla/^-xA^

jAh^jAjn

/UMyA/V^vjWV^^

wv_n lmva-aj^'^A^xA /lj\a aA^IATV^

/VJ\^ui/L-JU hJJ^^J\AAAJ\K

^A_Av^^_/■''^/^AA__ ± L

6 12 18 24

10

Frequency (cycles/day)

^1\J\}\kJ^

Aa_^^/v^ AjiAA fHj

uvA_n /''''\ J[aw^

..jvwJl AiyfN

w\UVaJ'A^vv-jv_jv_,^^

wVVWjJVa

aAaMMaa^L I M aa

aA JW lAWwLjiji

JXXAM^

K^^v~a,v^,-^AaA/IA l\j\^ AVlL_-AmA

awJA^|1w,_A. t T

12

18

24

0.008

0.006

0.004

0.002

t.

<u

® 0.000

012345678910

Oi

.> 0.008

13

Pi 0.006

0.004

0.002

0.000

0 20 40 60 80 1 00

Frequency (cycles/day)

Figures 9-15. “Continued.

SUTER=== PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

13

0.007

1 00

Frequency (cycles/day)

0 6 12 18 24

0.012-1

0.010-

0.008 -

0.006 -

0.004 "

4>

^ 0.002 -
©

ft.

J[.

0.000 4

>

1 1

2 3 4 56 7 8911

cq 0.012-

^ 0.010-

0.008 -

0.006 •

0.004 -

0.002 -1

0.000 -1

0 20 40 60 80 100

Frequency (cycles/day)

Figures 9- 1 5 . — Continued.

14

THE JOURNAL OF ARACHNOLOGY

vJ Aj'aJ^

aa/V/u

jv^ IHJ\fh\}v^ AjVi A

V_Uv.^Aaa,->UaAlJu_^ /’'^'VkaTWH

jVL_UWJ\i ^ .

T

0 6 12 18 24

0.006

14

Frequency (cycles/day)

0 6 12 18 24

Frequency (cycles/day)

Figures 9-=- 1 5 . Continued.

SUTER^PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

15

JLILJUTWI

"irM /itfi. nn prYsy-vv^Mv^

... „ ^■ . AftAlto/U,

n Pin A n n» ,

12

18

24

Frequency (cycles/day)

iVjuAvWiliA#

-J\ /\_J\JlAJU\i\jVAA A_

_jx/uM.nA Ai A A A fl

0 6 12 18 24

Frequency (cycles/day)

Figures 16-20.— Activity of trigonum under conditions of constant darkness (DD). See the legend for Figs.

2-8 for an explanation of the layout.

16

THE JOURNAL OF ARACHNOLOGY

J . ii JjJkl

llii

.iikbi'Ui AkiiAriuiii

.JiiUik

JUlyAhA>^ A

0.010'

0.008

0.006

0.004

:

hJ

• 1 ' 1

18

0.006

liliidA

Frequency (cycles/day)

12

18

24

ll -'ll

*1

_Jjjl

liilll

1

IjUI

1 aJl

a1 Aj JiLi

illLjllllliL

0 6 12 18 24

Frequency (cycles/day)

Figures 1 6-20. — Continued.

SUTER^PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

17

►

Frequency (cycles/day)

0 6 12 18 24

Figures 16-20. —Continued.

of slow rhythms (fewer than 24 cycles per day)
activity in each 3 -min period was summed and
the string of 3-min sums was analyzed by FFT.
For the detection of more rapidly cycling rhythms,
subsets of the full data string were analyzed with-
out summing (i.e., at 0.625 Hz, one sample every
1.6 s). The two procedures were required because
the FFT algorithm could accommodate no more
than 4096 data points in a single analysis. (2)
Spectral Analysis II: the inactivity periods alone
but in order, abstracted from the time-based string
above, were analyzed in the same way by FFT
and displayed as a power spectrum as described
above; this procedure could reveal periodicities
in a program which controlled the durations of
inactivity periods. For each of the spectral anal-
ysis protocols, a was set at 0.01 to decrease the
probability of type II errors.

RESULTS

The activity of both species of spiders in the
isolation chambers consisted of bouts of nearly
continuous activity alternating with periods dur-
ing which no activity was detectable. Complete
activity records for all tested spiders are shown
in Figs. 2-20. Presented with each of the activity
records in Figs. 2-20 are graphs of explained vari-
ance as a function of frequency [power spectra,
from Spectral Analysis (I)]. In all but one of the

graphs that represent spider activity in constant
darkness {F. pyramitela. Figs. 2-8; A. trigonum,
Figs. 16-20), significant {P < 0.01) peaks occur
near 1/day (1.15 x 10-5 Hz), an indication that
both species should be added to the long list of
taxa in which the presence of endogenous cir-
cadian rhythms is confirmed. Some of the F.
pyramiteia under LD conditions (Figs. 9-1 5) were
relatively inactive during the light phase of the
cycle (Figs. 9-11), but others appeared to be un-
influenced by the dim lighting. The suppression
of activity levels in some spiders by relatively
dim light suggests that the animals are function-
ally nocturnal in the field.

Figures 2-20 also show that higher frequency
periodicities participate in generating the ob-
served activity patterns. In all of the power spec-
tra, power is concentrated in significant peaks at
frequencies between 1 cycle/day and 1 00 cycles/
day: in Fig. 2 1 , which shows cumulative power
as a function of frequency for two F. pyramiteia
activity records, 50% of total power is at fre-
quencies below 40 cycles/day and at least 75%
is at frequencies below 1 00 cycles/day. Some of
the significant high frequency (i.e., higher fre-
quency than 1/day) periodicities are prominent
enough to be easily seen in the activity records
themselves. Perhaps the most prominent occurs
in Fig. 18 in which a periodicity at about 40

18

THE JOURNAL OF ARACHNOLOGY

Frequency (cycles/day)

Figure 21. —Cumulative explained variance in the
spectral analyses of the activity records of two F. pyr-
amitela. The heavy line corresponds to the analysis of
the data shown in Fig. 2 and the light line corresponds
to the analysis of the data shown in Fig. 7. The most
rapid rise in explained variance occurs very near the
origin, indicating that high frequencies (> 75 cycles/
day) are of minor importance (see text).

cycles/day is evident during the 24 h beginning
at 2100 h on day 1, and during day 6. That 24-h
period and all of day 6 for the same trigonurn
are shown with the corresponding power spectra
in Fig. 22.

The intervals between activity bouts for all
spiders in DD are shown and analyzed in Figs.
23-34. Spectral Analysis (II) of these intervals
alone, kept in order but abstracted from a time-
based activity series indicates that in F. pyr-
amitela there is no indication of significant pe-
riodicity in the patterning of inter-activity inter-
vals: the variance explained by periodicities is
distributed relatively evenly across each spec-
trum of harmonics and is below the 0.01 level
of significance for any particular peak above the
first harmonic.

DISCUSSION

The behavior of an organism at a particular
instant is a function of the interaction between
its internal state and information the organism
possesses about its surroundings. These two el-
ements are not entirely distinct from each other,
however: first, the internal state is in part an
evolutionary construct, a consequence of histor-
ical responses to the organism’s surroundings;
and second, the organism’s current information
about its surroundings is available to it only after
filtration through systems of sensation and per-

Frequency (cycles/day)

1

-iwu

0 1 0 20 30 40 50 60 70 80 90 100

Jjji

0 100 200 300 400

Frequency (cycles/day)

Figure 22.— Power spectra of subsets of the activity
data shown in Fig. 18. In that data set, a periodicity
at about 40 cycles/day is evident during the 24 h be-
ginning at 2100 h on day 1, and during day 6. The
corresponding power spectra are shown here in panels
a and b, respectively. As in other power spectra in this
study, peaks above the horizontal lines are significant
at a = 0.01. In panel b, the peak at 36 cycles/day is
significant at a = 0.05.

ception that are sensitive to (and part of) the
internal state. In this context, the endogenous
rhythms and other behavioral programs of an
organism should be seen as parts of the internal

SUTER-PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

19

H

1

C

.2

u

Q

u

0)

o

“S

05

0 100 200 300 400 50

Sequence

0 600

700 800

0 20 40 60 80 100 120 140 160 180 200

Harmonic

Harmonic

20 160 200 240 280 320 360 400

Sequence

60 80 1 00 120 140 1 60 1 80 200

Harmonic

00 200 300 400 500 600

Sequence

20 40 60 80 100 120 140 160 180 200

Harmonic

Figures 2 3“=2 9.— Patterns in the durations of intervals between bouts of activity by F. pyramitela in constant
darkness (DD). The durations of the intervals themselves, in each figure, are shown (in min) in the top panel,
and the power spectrum for that sequence of intervals is shown in the bottom panel. Peaks above the horizontal
lines are significant at a = 0.01. In none of the analyses was there significant power at harmonics above DC,
an indication that the sequences of durations were not different from random.

state that are evolutionary responses to the pe-
riodic or probabilistic structure of the environ-
ment.

Endogenous rhythms.^ The best known en-
dogenous rhythms in nature are those that ap-
proximate geophysical rhythms with respect to
period length. Their original function was prob-
ably to provide a means by which organisms
remain synchronized with their environments,
and they now function, in addition, in photo-
period measurement and in navigation. Circa-
dian rhythms and other low frequency endoge-
nous rhythms (e.g., circumlunar, circannual) are
nearly ubiquitous organizers of activity in or-
ganisms. Among spiders, circadian rhythms have
been identified (reviewed in Cloudsley-Thomp-
son 1987) but neither higher nor lower frequency

rhythms have been implicated in spider behav-
ior. Because most spider species are short-lived,
one would not expect to discover among them
endogenous rhythms with very long periods (e.g.,
the lunar month or the solar year). Nor would
one necessarily expect to discover endogenous
rhythms with very short periods because the en-
vironment does not contain biologically impor-
tant geophysical cycles shorter than the 12.25 h
tidal period.

Because of the relatively brief durations of the
studies described herein, I could not have dem-
onstrated rhythms with periods longer than about
two days. The data do make clear, however, that
circadian and higher frequency rhythms in F.
pyramitela and A. trigonum participate in or-
ganizing spontaneous motor activity (Figs. 2-20).

20

THE JOURNAL OF ARACHNOLOGY

Harmonic

0 20 40 60 80 100 120 140 160 180 200

Harmonic

Harmonic

Figures 23-29.— Continued.

The evidence for an endogenous circadian clock
is neither surprising nor particularly interesting
given the ubiquity of these clocks in biological
systems and their obvious efficacy in maintaining
synchrony between the organism and its envi-
ronment. The presence, and in some instances
the predominance, of higher frequency rhythms,
in contrast, is quite interesting.

Because of the absence of biologically impor-
tant geophysical cycles with frequencies from 3-
100 cycles/day, the significant periodicities in
spider motor activity uncovered in that range
cannot be of use to the spiders as environmental
synchronizers. Moreover, because the spiders in
these studies were given neither food nor water,
the most likely physiological rhythms (those as-
sociated with the filling and emptying of the nu-
trient and water pools) must be discounted. I am
left without a strong hypothesis with which to
explain the function(s) of the high frequency pe-

riodicities demonstrated in the spontaneous mo-
tor activity of these isolated spiders. Of particular
interest are the relatively stable oscillations that
persist for many cycles and appear to be stable
after several days (e.g., Figs. 18, 22): these are
likely to be both important in the lives of the
spiders and amenable to experimental investi-
gation.

Other behavioral programs.— The periodici-
ties described above are time-based; that is, they
are detected as peaks of activity which are re-
peated at regular intervals in the time domain.
Quite a distinct type of behavioral program would
be one that obeyed the following algorithm: when
behavior x begins, continue x until t seconds
have elapsed, then begin behavior y ; assign t a
new value based on some specified trigonometric
function; repeat. The resulting periodicity in the
durations of x would be undetectable if there
were uncorrelated variability in the durations of

Relative Power Duration (min) Relative Power Duration (min) Relative Power Duration (min)

SUTER^PATTERNS OF SPONTANEOUS MOTOR ACTIVITY

21

20 40 60 80 100 120 140 160 180 200

Harmonic

200 300 400 500 600

Sequence

O)

p 0.02

20 40 60 80 100 120 140 160 180 200

Harmonic

0 20 40 60 80 100 120 140 160 180 200

Harmonic

nAiyiiiiiiiiiilLiiiiJ

32 .s

^ 20 -

i :

T ' 1 « 1 r 1 ' 1 ' 1 ' i ’

0 100 200 300 400 500 600 7(

Sequence

30 ® 0

! 1 i 1 . j ^=^-1 1 1 1— 5 — 1 M

100 200 300 400 500 600

Sequence

-

Relative Power

§ 1 2 1 ^

1 ~

i ^

33

0 20 40 60 80 100 120 140 160 180 200

Harmonic

Figures 30““34.-- Patterns in the durations of intervals between bouts of activity by A. trigonum in constant
darkness (DD). See the legend to Figs, 23-29 for details of the layout. In none of the analyses was there significant
power at harmonics above DC, an indication that the sequences of durations were not different from random.

22

THE JOURNAL OF ARACHNOLOGY

V. For example: suppose y is motor activity and
X is inactivity; if activity occurs in bouts that
vary in duration according to some non-periodic
function, then peaks in the durations of inactivity
bouts would appear nonperiodic; if, in contrast,
inactivity bouts were extracted in order from the
time series, the periodicity of peaks in the du-
rations of inactivity bouts would become appar-
ent. This is the reasoning behind Spectral Anal-
ysis IT In the linyphiid spider, F. pymmitela,
and in the theridiid spider, A. trigonum, the du-
rations of inactivity bouts show no evidence of
periodicity (Figs. 23-34) which leads to the con-
clusion that these bouts are not generated by the
sort of behavioral program outlined above.

Thus there is no evidence from the data pre-
sented here that the inactivity bouts of F. pyr-
amitela and A. trigonum are regulated by any
sort of non-random program. What, then, de-
termines the duration of a particular period of
inactivity in these spiders? The possibility that
they are coupled to, or driven by, some process
the output of which is itself random is intriguing
but well outside the scope of this study.

Conclusion,— My analysis of the motor activ-
ity patterns of F. pyramitela and A, trigonum
reveals that endogenous oscillators participate in
determining the timing of activity but not the
durations of inter-activity intervals. Some non-
periodic processes, which may be the equivalent
of random-number generators, are also impor-
tant in determining the durations of inter-activ-
ity intervals. Both the nonperiodic processes and
the endogenous oscillators with periods much
shorter than 24 h are particularly interesting be-
cause they constitute an unexplored set of be-
havioral programs that may influence much of
the behavior of these spiders.

ACKNOWLEDGMENTS

I am indebted to Steve Clark (at Vassar Col-
lege) and Jeff Cynx (at The Rockefeller Univer-

sity) for several of the conversations that led to
this study, to Thomas Getty (at Michigan State
University) for his helpful comments on the
analysis of apparent noise, and to Tim Forrest
(at the National Center for Physical Acoustics)
for his guidance in the use of the discrete Fourier
transform.

LITERATURE CITED

Aschoff, J. 1954. Zeitgeber der tierischen Tagesper-
iodik. Naturwiss., 41:49-56.

Anderson, J. F. 1974. Responses to starvation in the
spiders Lycosa lenta (Hentz) and Filistata hibernalis
(Hentz). Ecology, 55:576-585.

Biinning, E. 1963. Die physiologische Uhr. Springer,
Berlin.

Cloudsley-Thompson, J. L. 1957. Studies in diurnal
rhythms. V. Nocturnal ecology and water-relations
of the British cribellate spiders of the genus Ciniflo
(BL). J. Linn. Soc. (ZooL), 43:134-152.
Cloudsley-Thompson, J, L. 1987. The biorhythms of
spiders, Pp.371-379 In Ecophysiology of Spiders
(W. Nentwig, ed.). Springer- Verlag, Berlin.
Seyfarth, E-A. 1980. Daily patterns of locomotor ac-
tivity in a wandering spider. Physiol. EntomoL,
5:199-206.

Suter, R. B. 1985. Intersexual competition for food
in the bowl and doily spider, Front inel la pyramitela
(Linyphiidae). J. ArachnoL, 13: 61-70.

Suter, R. B. 1990. Courtship and the assessment of
virginity by male bowl and doily spiders. Anim.
Behav., 39:307-313.

Suter, R. B. & T. G. Forrest, in press Vigilance in
the Interpretation of spectral analyses. Anim. Be-
hav.,00:000-000.

Suter, R, B. & L. Walberer. 1989. Enigmatic cohab-
itation in bowl and doily spiders, Frontinella pyr-
amitela (Araneae, Linyphiidae). Anim. Behav., 37:
402-409.

Manuscript received 5 March 1991, revised 13 Novem-
ber 1992.

1993. The Journal of Arachnology 21:23--28

PREDATION BY SPIDERS ON GROUND-RELEASED
SCREWWORM FLIES, COCHLIOMYIA HOMINIVORAX
(DIPTERA: CALLIPHORIDAE)

IN A MOUNTAINOUS AREA OF SOUTHERN MEXICO'

John B. Welch^: Screwworm Research, Agricultural Research Service, United States
Department of Agriculture, Tuxtla Gutierrez, Chiapas, Mexico

ABSTRACT. Predation by spiders on ground-released adult screwworms, Cochliomyia hominivomx (Co-
querel), was studied near Tuxtla Gutierrez, Chiapas, Mexico, during 13 August 1984-23 January 1985. Obser-
vations of predatory behavior and manual collections of spiders during September provided the majority of the
data. Species in 1 2 genera of spiders were confirmed as predators of screwworm flies. Nephila clavipes, Eriophora
ravilla, Neoscona oaxacensis and Leucauge spp. were the most important predators. Spiders caused an estimated
4.5% mortality to flies in a 2250 m^ area during September. Capture of screwworm flies in webs up to 10 m
above ground suggests the need to investigate the importance of forest canopies in screwworm ecology.

The screwworm fly, Cochliomyia hominivorax
(Coquerel), native to the western hemisphere, is
an obligate parasite of warmblooded animals.
Studies of the ecology of screwworms in the trop-
ics have been conducted since the late 1970’s
(e.g., Krafsur et al. 1979; Spencer et al. 1981;
Brenner 1985; Mangan & Thomas 1989). How-
ever, information on predation of screwworms
is generally lacking.

My interest in predation of screwworms began
during an unpublished study of dispersal of
ground-released adult screwworm flies in a
mountainous area of southern Mexico. On the
occasions of the first two releases of flies (22 May
and 6 July 1984), spider populations were ap-
parently low because few spiders and webs were
seen during the collection of the fly samples. Nu-
merous spiders and webs were encountered dur-
ing trips to the study area the week before the
planned third release (14 August). Spiders con-
tinued to be numerous prior to the fourth release
(25 September). Spider populations appeared to
decline after this release, as few spiders and webs
were observed while collecting fly samples during
the fifth and sixth releases (13 November 1984
and 22 January 1985).

The results of an investigation into predation

‘ Mention of a proprietary product does not imply an
endorsement or a recommendation for its use by USDA.

^ Current address: USDA-ARS Screwworm Research,
Center for Space Research, WRW 402, The University
of Texas at Austin, Austin, Texas 78712-1085 USA

on adult screwworm flies during the third and
fourth releases are presented herein. Additional
data concerning spider predation during the fifth
and sixth releases and other screwworm fly dis-
persal studies are included. Information on the
predatory behavior of spiders is also presented.

METHODS

The study was conducted in the Sumidero
Canyon National Park located ca. 15 km north
of Tuxtla Gutierrez, Chiapas, Mexico. Spiders
were also collected from Finca San Rafael, a study
site ca. 32 km south of Tuxtla Gutierrez during
another trapping study of C. hominivorax (Welch
1988).

The study site in the Sumidero Canyon Na-
tional Park was located on the southern face of
the mountain at an elevation of ca. 1 040 m above
mean sea level. The habitat was a low deciduous
forest (Miranda 1 975) comprised predominantly
of trees ca. 3-8 m tall, with scattered, emergent
trees above the canopy. The study site at Finca
San Rafael was also situated in a low deciduous
forest, although the trees were taller (ca. 4-10 m)
and the vegetation was more dense.

Sterile screwworm pupae of the A- 8 2 strain
obtained from the sterile-fly production plant of
the Joint Mexico-U. S. Commission for the Erad-
ication of Screwworms were placed in the field
on 25 September (and the other release dates)
and allowed to emerge. Pupae were marked with
20 g/liter of fluorescent powder (Dayglo Color,
Cleveland, Ohio) and distributed in open card-

23

24

THE JOURNAL OF ARACHNOLOGY

board cartons (1.45 x 18 x 4 cm; 0.5 liter/car-
ton) stacked inside a wooden crate (26.5 x 36
X 49.5 cm). The crate was covered with a cor-
rugated tar-paper roof for protection and sus-
pended 1.5 m above ground by wire from a tree
limb. This is termed “ground-released” as op-
posed to being released from an airplane. Per-
centage of emergence was determined from con-
trols (0.5 liter carton of marked pupae enclosed
within a screen bag inside the release crate) (W elch
1988). All flies were examined under longwave
ultraviolet light for fluorescent markings on the
frontal suture (Brenner 1984).

Predation data were obtained by direct obser-
vation and by collection of spider and fly sam-
ples. Spider and fly activities were monitored
visually at the Sumidero release site 5 from 0630
to 1600 h (time period due to park hours) on 26
and 27 September. Three pitfall traps filled with
70% ethyl alcohol were operated from 25-29
September, 13-17 November, and 22-26 Janu-
ary. Fly emergence ended and manual collections
of spiders and sweep samples of the vegetation
in the vicinity of release site 5 were conducted
on 28 September. Spiders with screwworm flies
in their webs were also collected from other re-
lease and trap sites within the Sumidero Canyon
National Park and Finca San Rafael throughout
the remainder of the study period.

A search for screwworm fly cadavers and spi-
ders was made along a transect (1 50 m) heading
south of release site 5 on 28 September. The
transect was located along an existing trail de-
scending the mountain because cutting of new
trails or paths within the park was prohibited.

Specimens were sent to William B. Peck for
identification. Voucher specimens are main-
tained in a collection at the United States De-
partment of Agriculture, Agricultural Research
Service, Screwworm Research Laboratory at El
Alto de Ochomogo near San Jose, Costa Rica (as
is typical in Costa Rica, the actual laboratory site
has no street or mailing address; requests for
information concerning the voucher specimens
should be directed to the author at his listed
mailing address).

RESULTS

A total of 1 26 spiders representing 1 2 families
and ca. 26 species were collected during this study.
Most (78.6%) of the spiders belonged to the fam-
ilies Tetragnathidae and Araneidae, with Nephila
clavipes L., Leucauge spp., Neoscona oaxacensis
(Keyserling) and Eriophora ravilla (C. L. Koch)

accounting for 19.0, 19.0, 14.3, and 13.5%ofthe
total spiders collected, respectively. Other ara-
neid genera included Micrathena, Verrucosa, and
Mangora. Spiders belonging to the families Fil-
istatidae, Tengellidae {Zorocrates), Plectreuridae
(Plectreurys), Theridiidae (Argyrodes), Lycosidae
(Lycosa), Oxyopidae (Peucetia), Clubionidae,
Sparassidae, Selenopidae (Selenops), Thomisi-
dae {Misumenoides, Misumenops) and Salticidae
{Phidippus) were also collected. The majority (103
specimens, 7 families) of the spiders collected
were captured manually, while 5 specimens rep-
resenting 4 families were collected by pitfall trap,
and 18 specimens representing 3 families were
collected by sweep sample.

Predation by spiders on screwworms was ev-
idenced by dead flies in the webs of spiders (Table
1) and by direct observation of flies being cap-
tured. Species of Misumenops sp. and Misume-
noides sp. were collected, each with one screw-
worm fly. Additionally, a Peucetia viridans
(Hentz) was discovered with two dead flies on
one occasion. Species of 11 genera of spiders
captured screwworm flies in webs during the
study. The four most commonly collected groups
of spiders listed above also accounted for the
most (91.5%) webs containing screwworm flies
(Table 1). Webs of N. clavipes accounted for
43.9% of those with ensnared flies. Eriophora
ravilla, N. oaxacensis, and Leucauge spp. webs
accounted for 19.0, 17.8 and 9.5%, respectively,
of the webs with flies. All webs of N, clavipes and
88.9 and 93.7% of webs of E. ravilla and N.
oaxacensis, respectively, contained screwworm
fly cadavers. Only 40.0% of the webs of Leucauge
spp. contained dead flies.

A rough estimate of percent predation of
screwworms may be obtained from the collec-
tions of data along the transect heading south of
the release crate at release site 5. A total of 389
screwworm flies were observed in webs along the
transect within 75 m of the release crate. Webs
were located within ca. 3 m of the transect and
up to 10 m above the ground. Based on emer-
gence of controls, an estimated 8710 flies were
released from the site, thus resulting in a calcu-
lation of 4,5% mortality of flies by spiders within
the 2250 m^ area of the transect. Inspection of
the surrounding area resulted in few additional
webs being located and no screwworms were seen
in the webs. Therefore, the estimate would only
be valid for the area of the transect; and it is still
imprecise because flies killed and removed from
the webs, and flies that were bitten and escaped

WELCH-- ADULT SCREWWORM PREDATION BY SPIDERS

25

from the web, but then died, etc. were not count-
ed.

Although not along the transect, a fly released
from site 1 got caught in a web of N. oaxacensis
located 1 50 m to the south, the record distance
between release site and point of entanglement
in a spider web during this study.

Due to the small sample size, fly capture data
were pooled for all spider species. Generally, more
flies per web were ensnared from 1-10 m from
the release crate than from 1 1-20+ m for most
species of spiders (Fig. 1). Most webs of all spe-
cies (64.6%) along the transect were 1-3 m above
ground and accounted for the most screwworm
flies ensnared. More webs (27.7%) were 7-10 m
than 3. 5-6. 5 m above ground (7.7%), and more
flies were caught in the higher webs than those
located within the 3. 5-6. 5 m range (Fig. 2). This
pattern was exhibited by N, clavipes and E. rav-
illa webs. Entanglement in webs of Neoscona
oaxacensis occurred 1-3 m above ground, and
the webs of the unidentified species were mostly
1-3 m above ground.

Observations of fly activity at the release site
indicated that flies began leaving the release crate
at 0800 h on 26 September and continued until
observations were stopped at 1 600 h. Fly activity
began at0637hon27 September and continued
until 1437 h, at which time dispersal from the
release crate was complete. On both days, fly
activity increased when direct sunlight reached
the release crate, and slowly decreased when di-
rect sunlight was blocked by clouds. No flies left
the release crate while it rained, and flies outside
of the release crate moved to the undersides of
leaves and rocks during the rain.

Observations on the predatory activity of spi-
ders in webs within three m of the release crate
were made in relation to screwworm flies emerg-
ing from the release crate at release site 5. In-
dividuals of N. clavipes (five females) began feed-
ing on the flies immediately as the flies became
trapped in the webs on the morning of 26 Sep-
tember. This activity continued from 0800 h un-
til ca. 1020 h, when the spiders began repairing
the webs. Predation and web repair continued
throughout the afternoon while the flies were ac-
tive. However, spiders began ignoring some of
the new flies caught in the webs ca. midday (e.g.,
one spider with 29 flies in its web began ignoring
the flies at 1148 h and another spider with 14
flies in its web began ignoring other flies at 1151
h). Flies that were ignored and not killed im-
mediately had time to make a possible escape.

Table L— Species of spiders and number of webs
containing at least one cadaver of Cochliomyia hom-
inivorax after field release. Cadavers also were found
in the webs of an unidentified theridiid and other un-
identified species.

Number of webs

With Without

Species prey prey

Plectreurys sp.

1

0

Argyrodes sp.

1

3

Araneus sp.

1

0

Eriophora ravilla

16

2

Nephila clavipes

36

0

Micrathena spp.

2

2

Verrucosa arenata

2

0

Neoscona oaxacensis

15

1

Leucauge spp.

8

12

Total

82

20

Approximately 27.5% (19 of the 69) of the flies
caught in the webs of N. clavipes escaped during
the study period on 26 September.

At 1420 h, a female E. ravilla climbed onto a
web of a female N. clavipes, and the N. clavipes
rapidly retreated to the vegetation to which the
web was anchored. The E. ravilla began exam-
ining the dead flies (1 1) in the web and initiated
repairs to the web. After repairs were completed,
the spider began preying on newly captured flies
and maintained the web for the remainder of the
afternoon. Upon my arrival at the study site on
the morning of 27 September, the female E. rav-
illa had vacated the web, and what appeared to
be the original N. clavipes owner (based on size
and appearance) had returned and was repairing
the web. This specimen continued predation in
the web and the E. ravilla did not return.

The web of the largest female N. clavipes had
not been repaired by the morning of 27 Septem-
ber, had several large holes and was cluttered
with the corpses of flies. This individual made
no attempt to repair the web and did not react
to any new flies hitting her web during the day.
All of the flies that were caught in her web on
27 September escaped.

The percentage of flies escaping the webs of N.
clavipes on 21 September could not be estimated
because the numbers of flies that were hitting the
webs and escaping were too numerous to count.
At one point during the morning (0745 h), during
peak dispersal of the flies, it was estimated that
ca. 40 flies per minute were hitting the webs.

26

THE JOURNAL OF ARACHNOLOGY

25

8

Q

CL

Z)

H-

CL

<

O

if)

y

o

50

O Nephila clavipes
# Eriophora ravilla
A Neoscona oaxacensis

o

8

o

0 5 10 15 20 25 30 35 40 45 50 55 60 65

DISTANCE (m)

Figure L— Number of screwworm fly cadavers found in webs of three species of spiders located from 1-75
m from the release box.

Nineteen screwworm flies were observed being
preyed upon during the hours of 0655-1 130 by
N. clavipes.

Observations on the predation activity of eight
female E. ravilla were made in relation to the
release crate at release site 5. Predation activity
of E. ravilla also began on the morning of 26
September when the dispersing screwworm flies
began hitting the webs. At first, webs were re-
paired immediately after the flies were killed and
removed. However, beginning around 1020 h,
the spiders ceased repairing the webs. Then at
1108 h some individuals of this species began
ignoring live flies entangled in their webs. Flies
were ensnared and killed throughout the day un-
til observations were stopped at 1600 h.

A total of 88 screwworm flies was ensnared in
webs of E. ravilla on 26 September. Seventeen
flies escaped from the webs, resulting in an es-
timated 19.3% escape.

Although more flies were dispersing on 27 Sep-
tember, E. ravilla began ignoring the flies hitting
the webs after 0848 h. Three of seven screwworm
flies ensnared before 0848 h escaped, resulting
in an estimated 42.9% escape for that time pe-
riod.

Predation of screwworm flies dispersing from

the release crate at release site 5 by three female
N. oaxacensis showed the same pattern as that
by N. clavipes and E. ravilla. Predation on 26
September began immediately when the flies be-
gan hitting the webs and continued until 1 530 h.
Hies began escaping from the webs at 1035 h
with an overall 12.7% escape (7 of 55). Spiders
also began repairing their webs around 1020 h.

Most of the flies caught in the webs of N. oax-
acensis on 26 September had been removed by
0630 h on 27 September. Ensnarement of flies
on 27 September began at 0654 h and continued
until 1000 h. Two flies escaped during that time
period, resulting in 1 1.8% escape.

DISCUSSION

Predation of screwworm flies by species of 1 2
genera of spiders was observed in this study.
Specimens of E. ravilla, N. clavipes, N. oaxa-
censis, V. arenata and P. viridans were the only
spiders within these genera identified to species,
with the first three being the most important
predators.

Nephila clavipes apparently build their webs
in areas within the forests which are probable
flight paths of insects (Robinson & Mirick 1971),
Collections of flies in webs up to 10m above the

WELCH -ADULT SCREW WORM PREDATION BY SPIDERS

O Nephila clavipes
# Erlophora ravilla
ANeoscono ooxocensis

27

25-

o*

20-

Q

LlJ

cr

? 15-

Cl

<

o

Li_

5-

0-^

0

O#

2 4 6

HEIGHT (m)

O

O

O

•

8

— I —

8

O

o

8

o

o

10

Figure 2.— Number of screwworm fly cadavers found in webs of three species of spiders from 0.5-10 m above
the ground.

ground, in the area of the forest canopy, suggest
that the forest canopy may be important to the
ecology of screwworms. More screwworm flies
have been collected in forest habitats than in
pastures (Mangan & Thomas 1989). Studies of
screwworms have been confined to ground level,
so further investigations of the vertical distri-
bution of screwworms in relation to habitat are
needed.

The predatory behavior exhibited by the N.
clavipes under observation agrees with the de-
scriptions in the literature with one major dif-
ference. It was reported that N. clavipes always
transported its prey to the hub of the web after
immobilization (Robinson et al. 1 969; Robinson
& Mirick 1971). None of the five female N. cla-
vipes that were monitored at release site 5 in my
study exhibited this behavior: flies were never
moved from the site of capture. Initially, when
prey numbers were low, an immobilization bite
was given and wrapping occurred at the capture
site followed by apparent feeding on the prey in
situ. When large numbers of flies were hitting the
webs or when the spiders apparently became sa-
tiated, post-immobilization wrapping was omit-
ted. The latter is in agreement with Robinson et
al. (1969) and Robinson & Mirick (1971). This
apparent feeding may have been what was re-

ferred to as a ‘Tong bite” by Robinson & Mirick
(1971); however, because considerable time was
spent by the spiders with the prey and later prey
were ignored (suggesting the spiders’ hunger was
satiated), it appeared that the spiders were feed-
ing. Upon my return to release site 5 on 27 Sep-
tember, corpses of flies were still present at the
site of capture in the webs and none were present
at the hub where the spiders were resting. Also,
the bite and back-oflfbehavior described by Rob-
inson & Mirick (1971) was not exhibited by N.
clavipes during my study, but this was probably
due to the smaller size of the prey (i. e., flies vs.
crickets) between the two studies.

Differences in percentages of prey escaping
from N. clavipes webs between earlier studies and
mine were probably due primarily to the unusu-
ally high density of screwworm flies in the area.
An estimated 27.5% of the screwworm flies es-
caped from the webs on 26 September. Robinson
et al. (1969) reported that 46% of the stingless
bees {Trigona sp.) ensnared, escaped from the
webs, primarily while N. clavipes was occupied
at the hub of the web. However, no estimation
was calculated for 27 September because too
many flies to be counted were hitting the webs
and were being ignored because the spiders were
apparently satiated.

28

THE JOURNAL OF ARACHNOLOGY

ACKNOWLEDGMENTS

The technical assistance of Eliseo Broca E.,
Alfredo Matias E., Carlos Moises E., Carlos Oje-
da E. and Ramiro Penagos R. is greatly appre-
ciated. Sterile flies used in this study were pro-
vided by the Joint Mexico-U. S. Commission
for the Eradication of Screwworms. Permission
to work in the Sumidero Canyon National Park
was granted by Felipe Barbosa Rivera of the Se-
cretaria de Desarrollo Urbano y Ecologia. I am
especially grateful to William B. Peck for iden-
tification of spiders collected during this inves-
tigation. I also thank Frank D. Parker and David
A. Dean for their reviews of an earlier version
of this manuscript, and Matthew H. Greenstone,
David Wise and an anonymous reviewer for sug-
gesting changes incorporated in the revision.

LITERATURE CITED

Brenner, R. J. 1984. Dispersal, mating, and ovipo-
sition of the screwworm (Diptera; Calliphoridae) in
southern Mexico. Ann. Entomol. Soc. America, 77:

779=-788.

Brenner, R. J. 1985. Distribution of screwworms
(Diptera: Calliphoridae) relative to land use and to-
pography in the humid tropics of southern Mexico.
Ann. Entomol Soc. America, 78:433-439.

Krafsur, E. S., B. G. Hightower, & L, Leira. 1979. A
longitudinal study of screwworm populations,
Cochiiomyia hominivorax (Diptera: Calliphoridae)
in northern Veracruz, Mexico. J. Med. Entomol,
16:470-481.

Mangan, R. L. & D. B. Thomas. 1989. Habitat pref-
erence and dispersal patterns in native female screw-
worms (Diptera: Calliphoridae). Ann. Entomol Soc.
America, 82:332-339.

Miranda, F. 1975. La vegetacion de Chiapas, primera
parte. Ediciones del Gobiemo del Estado, Tuxtla
Gutierrez, Chiapas, Mexico.

Robinson, M. H., & H. Mirick. 1971. The predatory
behavior of the golden-web spider Nephila clavipes
(Araneae: Araneidae). Psyche, 78:123-139.

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

Spencer, J. P., J. W. Snow, J. R. Coppedge & C. J.
Whitten. 1981. Seasonal occurrence of the primary
and secondary screwworm (Diptera: Calliphoridae)
in the Pacific coastal area of Chiapas, Mexico during
1978-1979. J. Med. Entomol, 18:240-243.

Welch, J. B. 1988. Effect of trap placement for de-
tection of Cochiiomyia hominivorax (Diptera: Cal-
liphoridae). J. Econ. Entomol, 81:241-245.

Manuscript received 7 December 1990, revised 10 Feb-
ruary 1993.

1993. The Journal of Arachnology 21:29-39

THE NATURAL HISTORY OF THE CALIFORNIA TURRET SPIDER
ATYPOIDES RIVERSI (ARANEAE, ANTRODIAETIDAE):
DEMOGRAPHICS, GROWTH RATES,
SURVIVORSHIP, AND LONGEVITY

Leonard S. Vincent: Division of Biological Sciences, Fullerton College, 321 E.
Chapman Avenue; Fullerton, California 92632 USA

ABSTRACT. A large and dense population of over 500 burrows of Atypoides riversi in a 2.0 x 3,2 m area
was monitored for two years to indirectly determine demographics, growth rates, survivorship and longevity of
the spiders. Twelve size classes of spiders were designated by correlating spider size to burrow size. All size
classes were present simultaneously throughout the year. Variable gro\vth rates were recorded for spiders in each
size class, and survivorship was lowest for spiders in the smallest size classes. It is estimated, based in large part
on growth rates, that A. riversi can live at least 16 years in the field.

The only long term comprehensive study of
the genealogy and demography of a large popu-
lation of mygalomorph spiders concerns the Aus-
tralian citnizidAnidiops villosus (Rainbow) (Main
1978). In another study, Marples & Marples
(1972) observed a population of several species
of New Zealand ctenizids for six years. The dem-
ographics of burrowing wolf spiders have been
examined in detail by McQueen (1978, 1983),
Humphreys (1976) and Miller & Miller (1991).

Herein I describe the natural history of the
fossorial mygalomorph spider Atypoides riversi
O. P- Cambridge, the California turret building
spider (Rivers 1892). Unlike the long term field
studies of Main and the Marples, my objectives
were to determine, in a two-year period, the dem-
ographics, growth rate, survivorship, and lon-
gevity of^, riversi. Unlike Main (1978) and Mar-
ples & Marples (1972), who measured burrow
and door diameters but did not correlate these
measurements to spider size, I measured and cor-
related burrow entrance size to spider size.
McQueen (1978), Humphreys (1976) and Miller
& Miller (1984) found positive correlations for
certain burrowing wolf spiders. Decae et al.
(1982), studying the burrow structure of a cten-
izid, also found a positive correlation between
carapace length and burrow diameter but did not
associate this with longevity. Using the correla-
tion and following all burrow size changes through
the two year period, I derived life history infor-
mation comparable to following a single cohort
of riversi through its long life. Miller & Miller

(1991) used a similar approach to study Geoly-
cosa turricola.

Additional natural history information on var-
ious antrodiaetids can be found in Atkinson
(1886a, 1886b), Coyle (1971, 1986), Rivers
(1891), Smith (1908), Vincent (1980, 1985, 1986),
and Vincent & Rack (1982).

METHODS

Study sites.— Two study sites at the University
of California’s Blodgett Forest Research Station,
located in the American River watershed on the
western slope of the Sierra Nevada in El Dorado
County, approximately 10 miles west of George-
town and at an elevation 1275 m were chosen
for their high density of burrows and uniform
ground cover. The population dynamics of spi-
ders in both areas were similar; therefore, this
paper reports on only one. Population data for
the other study area and vegetation descriptions
for both are in Vincent (1980).

The study site measured 2.0 x 3.2 m and con-
sisted of 1 60 20 cm square quadrats formed by
a grid system composed of nylon string and
wooden stakes. The stakes were placed at 20 cm
increments around the perimeter of the plot, and
string was placed on or slightly above the ground
connecting facing stakes. Ground cover was
mostly pine and cedar needles with occasional
pine seedlings present during the spring and sum-
mer months. A barbed-wire fence enclosed the
study area to exclude deer and other large ani-
mals.

29

30

THE JOURNAL OF ARACHNOLOGY

Correlation of spider size and burrow entrance
size,— A series of 15 non-metric ball bearings
ranging in diameter from 2/32 inch (1.59 mm)
to 16/32 inch (12.7 mm) in increments of 1/32
inch (0.79 mm) were hard-soldered to thin single
fiber wire “handles”. These ball bearings were
then used to measure the internal diameter of
spider burrow entrances. The internal diameter
was considered equivalent to the diameter of the
ball bearing that fit (or came the closest to fitting)
the narrowest section of the tapered burrow en-
trance. For convenience, burrow size classes were
designated by the numerators that fit the entranc-
es (sizes 2-“16). Attempts to measure burrow en-
trances accurately with a caliper or ruler proved
to be difficult and damaging to the flexible and
fragile entrance.

Six to 14 burrows {n = 128) representing each
size class were arbitrarily chosen near the study
site for measurement (Vincent 1980). After each
burrow was measured, the resident spider was
dug from its burrow, anesthetized by cooling with
crushed ice (large spiders) or CO2 (small spiders),
and measured. Spiders were measured with a
stereomicroscope fitted with an ocular microm-
eter accurate to 0.039 mm. Measurements of
maximum width of both the carapace and ster-
num were correlated to the internal diameter of
the burrow entrance.

Observation platform,— A portable observa-
tion platform consisting of a 2 x 1.33 m sheet
of plywood was supported approximately 1 2 cm
over the plot by planks and blocks. The leading
edge of the platform coincided with the trailing
edge of the row being examined to allow a de-
tailed view of one 20 cm square quadrat. After
examining all quadrats in a row, I advanced the
platform to the trailing edge of the next row, etc.
Since A, riversi, like some other fossorial my-
galomorphs, is sensitive to vibrations, successful
observations necessitated moving slowly on the
platform. Adjusting the platform caused some
spiders to retreat temporarily down their bur-
rows.

Burrow observations.— The position of each
burrow was noted and its entrance diameter was
measured. Burrows with flexible and freshly silked
turrets were measured for size class designation.
The following burrow conditions were recorded:
(1) occupied [spider was seen in its burrow]; (2)
abandoned [burrow appeared in use, but no spi-
der was detected during the immediate obser-
vation period, approximately five minutes]; (3)
closed [entrance was folded closed and sealed

with silk]; (4) missing [burrow could not be
found]; (5) old [burrow was in a state of disrepair,
the turret was stiff and/or tom or non-existent].
These and other relatively rare burrow condi-
tions are discussed in detail in Vincent (1980).

Data collection dates.— Burrows were initially
censused 5-23 September 1976. On 22-24 April
1977 new burrows and burrow conditions for a
random sample (n= 1 53) of previously censused
burrows of sizes 3 through 1 1 were recorded.
Also, burrow conditions for all burrows of sizes
12, 13, and 14 were recorded {n = 13). The ran-
dom-sample size for each size class was deter-
mined so that the standard deviation of the es-
timated proportion would be no greater than 0.30.
Confidence intervals for the tme proportions were
calculated by a formula given in Bickel & Daksun
(1977, formula 5.1.13), and modified to account
for sampling without replacement (Cochran 1977,
sec. 2. 1 5). During 28-3 1 July 1 977, new burrows
were mapped and recorded, and previously re-
corded burrows were measured again. On 20 Au-
gust 1977 a random sample of burrows present
on 28“3 1 July, 1977 was censused to confirm the
presence of spiders in the burrows previously
sized. Sample size and confidence intervals were
determined as above for the April 1977 random
sample. All burrows present on 28-31 July 1977
were recensused 19-20 May 1978 to see if they
contained spiders. Burrows that contained spi-
ders on 19-20 May 1978 were recensused and
remeasured 3-6 August 1978, and all burrows in
odd-numbered rows were observed at night with
a dim unfiltered flashlight as a further check on
spider presence. If a spider was not immediately
visible in its burrow, I waited several minutes
for it to appear; if it still did not appear, I con-
sidered the burrow abandoned. Again, all new
burrows were recorded and mapped.

Survivorship calculations.— Survivorship of A.
riversi was indirectly determined by subtracting
from the initial number of burrows censused the
number of burrows missing or considered “old”
during each consecutive census. Survivorship of
eggs and emerging spiderlings is unknown.

RESULTS AND DISCUSSION

Correlation of spider size and burrow entrance

size.— Burrow entrances ranged from 3/32 inch
(2.38 mm) through 14/32 inch (11.06 mm). A
regression of spider size (carapace widths) against
burrow entrance size was highly significant {P <
0.001) (Fig. 1).

Assumptions.— In estimating the following

VINCENT=NATURAL HISTORY 0¥ ATYPOIDES RIVERSI

31

Figure 1.— Regression of ball bearing size on carapace width (Y = 8.3X -3.1, R^ = 0.934, P < 0.001).

demographics, survivorship, growth rates and
longevity for A. riversi, I assume each burrow
has had only one occupant who had enlarged and
maintained it over time, that burrows present
and in good shape contain a living spider, and
that missing burrows are a measure of spider
mortality.

Various field observations support the as-
sumption of “burrow fidelity”. Throughout most
of the year I have measured increases in burrow
width for all size classes and have often observed
excavated soil adjacent to burrows. Further, dur-
ing hundreds of hours of observing A. river si in
the field, only on one occasion was a spider seen
totally outside its burrow. (This spider was later
found to be parasitized by a nematode). Indeed,
the reluctance of river si to leave their burrows
was evident by the difficulty encountered in
coaxing them completely out even with tethered
prey; once outside, they rapidly find their way
back to their burrow. Additionally, I excavated
hundreds of burrows but never found more than
one spider in a burrow. Lastly, pitfall traps placed
adjacent to or within 25 m of the study sites,

during most of 1977, recovered only adult male
spiders and only during the fall mating season.

The above observations suggest that A. riversi
does not leave its burrow to enter another burrow
to evict its resident in a competitive interaction
(as in Riechert 1978), to search out larger vacated
burrows, or to establish new burrows in better
areas. Other antrodiaetids (F. A. Coyle & W.
Icenogle pers. comm.), all door-building cteni-
zids (B. Y. Main pers. comm.; Decae et al. 1982),
and a burrow-dwelling theraphosid (Kotzman
1990) apparently maintain the same burrow
throughout life.

The assumption that burrows which were
maintained contained spiders (deteriorated bur-
rows rarely did) was based in part on a 5 August
1978 evening census in which 97.3% of the main-
tained burrows {n = 263) were occupied by spi-
ders. Those burrows in which a spider was not
observed at the entrance may still have contained
a spider at the bottom, perhaps feeding or re-
pelled by my dim flashlight.

The last assumption equating missing burrows
with spider mortality may not be entirely accu-

32

THE JOURNAL OF ARACHNOLOGY

*D

Q>

W

O

3

n

E

Burrow size classes and closed burrows

Figure 2.— The number of occupied and closed burrows found for each size class September 1976, July 1977,
and August 1978.

rate. In all censuses a few previously unrecorded
larger burrows were found. This is most likely
indicative of overlooked burrows in the previous
census or, less likely, immigration of spiders
which had somehow been evicted from their for-
mer burrow into the plot. (Some unrecorded new
burrows may have been excavated over smaller
burrows thus accounting for some of the extreme
jumps in size class, as discussed earlier.) If there
is some immigration, there may also be come
emigration of spiders not detected by pitfall traps
or observation. In any case, however, the number
of new, larger burrows was small.

Demographics.— Many species of araneo-
morphs experience a marked seasonality, with
certain size classes restricted to specific times of
year. All size classes of^. riversi, however, occur
simultaneously throughout the year. Remark-
ably high numbers (538, 635, 596) were present
for the three years censused (Fig. 2) in densities
as high as ten burrows per 20 cm square. Most

of the burrows belonged to sub-adult spiders.
(Adult females belonged to size classes 1 1 and
above, the only size classes found with eggs or
spiderlings; most adult males emerged from size
10 burrows (Vincent 1980)). The July 1977 fre-
quency of burrows in size classes 4-11, 13 and
1 4 was similar to the September 1976 population
(x" = 16.92, df= 9,P < 0.05), yet close to half
the September 1976 burrows increased or de-
creased in size, and 99 (1 8.4%) were missing (Ta-
ble 1). Size classes 5-11, 13 and 14 in August
1978 were similar in frequency (x^ = 26.30, df
= \6,P < 0.05) to the September 1976 and July
1977 censuses (Fig. 2). Notably there were fewer
size class 3 burrows than size class 4 burrows
and fewer size class 9 burrows than size class 1 0
burrows in 1976 and 1978 (Fig. 2). The large
number of size class 3 and/or 4 spiders relative
to the smaller number of spiders in the larger
size classes indicates heavy early instar mortal-
ity, typical of arthropods, for all three years. The

VINCENT = NATURAL HISTORY OF ATYPOIDES RIVERSI

33

3 4 5 6 7 8 9 10 11 12 13 14 C

Burrow Size Classes and Closed Burrows

Figure 3. —The number of burrows for each size class occupied by spiders and the number of closed burrows
during the May 1978 census.

total numbers of individuals for size classes 3
and 4 varied with time of year, as well as year
to year (compare Figs. 2 and 3). However, the
large number of new size class 3 and 4 burrows
found in April 1977 and May 1978, 205 and 124
respectively, indicate a late fall and/or early spring
emergence for spiderlings. This finding is con-
sistent with Coyle’s (1971) data. Fewer new bur-
rows of size classes 3 and 4 combined were found
July 1977 and August 1978 (40 each) (Vincent
1980).

The drastic decrease in number of size class 3
burrows, the increase in number of size class 4
burrows from May 1978 (Fig. 3) to August 1978
(Fig. 2), and the relatively low number of indi-
viduals in size class 3 found in September 1976
(Fig. 2) suggest mortality for size class 3 and/or
a transition from size class 3 to a larger size class
from the spring to fall. Indeed, of the 187 size
class 3 burrows first found April 1 977, 59 (3 1 .6%)
were missing and 23 (12.3%) grew one size class

by July 1977. Of the 87 size class 3 burrows first
found May 1978, 10 (11.5%) were missing by
August 1978 and 46 (52.9%) grew one size class.
The relatively large number of size class 3 bur-
rows found July 1977 compared to August 1978
was a result of the larger number of new size
class 3 burrows (187) found April 1977 com-
pared to 87 found May 1978. The number of
new spiderlings can be expected to vary exten-
sively, however, as indicated by the range (21--
74; X = 46) in number of eggs produced by nine
females adjacent to the study site (Vincent 1980)
and a range of 43-80 {n = 1) found by Coyle
(1971) for a coastal population. Unfortunately,
the number and size of clutches within the study
site could not be determined. The drop off in
numbers of individuals larger than size class 4
(Fig. 2) is due to relatively high mortality (Fig.
4) and growth of size 4 spiders to larger size
classes (Tables 1, 2, 3). The more stable distri-
bution of size classes 5 and above appears to have

34

THE JOURNAL OF ARACHNOLOGY

Table 1.— The total number {n and percent (%) of burrows decreasing, not changing, and increasing in size,
and those burrows recorded closed or missing in the July 1977 census of burrows first recorded September 1976.
“Change” column reflects current burrow status. Other = old, abandoned, or destroyed.

Size classes as of September 1976

Change

n

%

3

4

5

6

7

8

9

10

11

12

13

14

-3

1

0.2

1

-2

3

0.6

1

1

1

-1

30

5.6

1

6

6

6

4

2

2

1

1

1

0

160

29.7

8

57

37

18

13

5

5

8

3

4

2

1

137

25.5

29

32

18

18

13

9

8

4

6

2

56

10.4

3

3

3

10

7

12

8

8

2

3

15

2.8

1

5

6

2

1

4

3

0.6

1

1

1

6

1

0.2

1

Closed

20

3.7

3

5

1

5

6

Missing

99

18.4

16

29

16

12

7

4

5

7

1

1

1

Other

13

2.4

4

4

1

2

1

1

Total

538

60

124

84

69

57

45

30

38

20

6

2

3

been maintained by a complex combination of
spiders growing at different rates over the same
time period and varying mortality rates for spi-
ders in each size class (Tables 1, 2, 3).

Growth rates.— Tables 1 and 2 list changes in
burrow sizes and conditions over time for the
population of burrows censused September 1976
and recensused July 1977 and the population of
burrows as of July 1977 and recensused August
1978. Considering all size classes combined (Fig.

5), the 1976 population had a higher proportion
of burrows that did not change size (30%), es-
pecially size classes 4 and 5 (Table 1), than the
1977 population (12%).

Most burrows increased one or two size classes
in the two year period from September 1976
through August 1978; however a few increased
as many as five size classes and some decreased
as many as four size classes (Table 3).

Feeding studies in the laboratory (Vincent

Table 2. —The total number {n) and percent (%) of burrows decreasing, not changing, and increasing in size,
and those burrows recorded closed or missing in the August 1978 census of all occupied burrows as of July
1977; * = other minor changes. “Change” column reflects current burrow status. Other = old, abandoned, or
destroyed.

Size classes as of July 1977

Change

n

%

3

4

5

6

7

8

9

10

11

12

13

14

-5

2

0.3

1

1

-3

2

0.3

1

1

-2

6

1.0

3

1

1

1

-1

24

3.9

4

4

1

1

1

4

4

4

1

0

74

12.5

4

25

13

7

3

3

2

5

7

2

1

2

1

157

25.6

51

39

20

14

5

2

5

7

7

6

1

2

67

10.9

7

10

16

8

5

9

2

7

3

3

27

4.4

3

5

5

5

6

1

2

4

6

1.0

1

3

1

1

8

1

0.2

1

Closed

17

2.8

2

1

2

3

2

2

1

2

1

1

Missing

203

33.1

74

51

23

12

16

5

9

7

1

5

Other

28

4.5

2

4

4

5

2

4

3

3

1

Total

614

141

137

92

54

41

32

26

37

22

24

5

3

VINCENT^NATURAL HISTORY OF ATYPOIDES RIVERSI

35

Table 3.— The total number («) and percent (%) of burrows decreasing, not changing, and increasing in size,
and those burrows recorded closed or missing in the August 1978 census of all occupied burrows first recorded
September 1976; “Change” column reflects current burrow status. Other = old, abandoned, or destroyed.

Change

n

%

Size classes as of September 1976

3

4

5

6

7

8

9

10

11

12

13

14

-4

2

0.4

2

-2

6

1.1

3

3

-1

13

2.4

3

1

2

3

1

1

1

1

0

21

3.9

1

5

5

1

2

1

1

2

1

2

1

87

16.2

13

35

12

9

2

4

5

4

2

2

71

13.2

9

11

16

8

6

6

5

6

3

1

3

48

8.9

4

8

4

7

8

9

5

2

1

4

19

3.5

1

2

5

5

5

1

5

4

0.7

1

1

1

1

Closed

15

2.8

2

3

4

3

1

1

1

Missing

229

42.6

33

58

37

24

21

15

10

20

8

1

2

Other

23

4.3

2

2

6

3

3

3

2

2

Total

538

100.0

60

124

84

69

57

45

30

38

20

6

2

3

1980) revealed that A. riversi do molt to smaller
or larger sizes depending on food intake. Assum-
ing similar abiotic conditions, the non-uniform
growth rate in the field suggests that prey avail-
ability or prey capture rates may not always be
optimum for some members of this densely
packed (up to 635 burrows ina2.0 x 3.2m area)
population. Perhaps ^4. riversi, like other spiders,
molts to a smaller size in the field as a response
to starvation to maintain abdominal hydrostatic
pressure (Anderson 1974), which is important
for locomotion and prey capture (Wilson, 1970).

A non-uniform growth rate may be advanta-
geous to a long-lived species such as A. riversi
living in dense aggregations. Assuming that mat-
uration of the same clutch is asynchronous, early
maturing males would be conducive to outbreed-
ing, especially since adult males appear to die
shortly after the mating season (Vincent pers.
obs.), whereas adult females live several years in
the laboratory and field (presumably with the
potential to mate). In this regard, B. Y. Main
(pers. comm.) has unpublished data which in-
dicate that males of the long-lived ctenizid An-
idiops mature a year earlier than females of the
same cohort and brood.

Compared to adult females, which continue to
molt after maturity (Coyle 1968, 1971), some-
times to larger size class, adult males of riversi
{n = 27) had little variability in sternum (2.00-
2.36 mm X = 2.16, 1 SD = 0.1 1) and carapace
(3.60-4.16 mm, X = 3.92, 1 SD = 0.15) mea-

surements, but a wide range of abdominal sizes

(as determined by casual observation) both in
the field and laboratory. It may be more advan-
tageous for a small penultimate male to sacrifice
some abdominal food reserve, which it could
have attained by growing another year, to be-
come sexually active sooner (perhaps for a short-
er period of time due to a smaller food reserve),
than to delay maturation and suffer more ex-
posure to mortality factors.

Survivorship. —Survivorship of^. riversi within
this study site varied from year to year and within
each size class (Fig. 4). Burrows found September
1976 and recensused July 1977 show a gradual
increase in survivorship from size class 3 through
size class 8 and for size classes 11 and 14. A
similar trend for the ctenizid Anidiops has been
recorded (B. Y. Main pers. comm.). The drop off
in survivors for burrows in size classes 9 and 1 0
probably reflects both male emergence in search
of mates, (see “abandoned” burrows (Vincent
1980)), as well as mortality within the burrow
(see “missing” burrows. Tables 1, 2, 3). Size class
1 0 burrows often contained males or were vacant
during and just after the mating season (pers.
obs.). Laboratory reared spiders of size 9, 10,
and occasionally, 11 often molted to mature
males. The emerged males probably die during
or shortly after the mating season. Adult males
were never found in the field after the mating
season (July-September) in this study or in Coyle’s
(1971). Unfortunately, size classes 12, 13, and

36

THE JOURNAL OF ARACHNOLOGY

1977 141 137 92 54 41 32 26 37 22 24 5 3

Burrow size classes and initial frequencies

Figure 4.— Survivorship curves, based on burrow censuses, for spiders in initial size classes 3-14 for two one-
year periods and one two-year period.

14 had too few members to suggest any trends,
yet it should be noted that none of the size class
14 individuals died during this study.

Survivorship was lower for spiders in most size
classes, especially size class 3, from July 1977 to
August 1978 than from September 1976 to July
1977 (Fig. 4). In the two year interval, 1976-
1978, over 50% of the spiders survived for all
size classes except 3, 10 and 13. The curve for
1976-1978 resembles the 1976-1977 survivor-
ship curve but is proportionally lower.

Several mortality agents (fungi, nematodes, and
acrocerid and tachinid flies) of A. rivers i were
reared in the laboratory from egg sacs or spiders
of the larger size classes (Vincent 1983, 1985).
Unfortunately, I was unable to rear all the par-

asites and parasitoids to maturity for complete
identification. In most cases the final size class
of the dead spider (presumably killed by the ac-
tion of the agent) was determined. Those mor-
tality agents isolated in the laboratory and the
pompilid Priocnemis oregona in the field are
probably responsible for some of the mortality
recorded in the study site. Pathogens or parasit-
oids were not isolated from spider sizes 3-8 (al-
though one egg sac did contain some dead eggs
contaminated with a fungus). This suggests that
the more mature spiders are the usual victims of
these mortality agents. It is not known at what
stage in the spider’s development it is first at-
tacked by the agent (except for those parasitized
by P. oregona).

VINCENT=NATURAL HISTORY 0¥ ATYPOIDES RIVERSI

37

O'

.£

O'

c

o

o

m TJ

^1

M- O

O' c/5
O cn

c °

05 O
^ 05

CL *cn

Type of change in burrow size or state

Figure 5.™ Change in burrow size or state for all burrows present September 1976 and recensused July 1977
and all burrows present July 1977 and recensused August 1978,

For the present, I suspect desiccation, canni-
balism, and starvation to be the most significant
mortality factors for spiderlings. On several oc-
casions spiderlings placed in non-moistened con-
tainers for transport to the laboratory from the
field were found dead and shriveled a few hours
later. Larger specimens of A. riversi kept under
similar conditions did not desiccate. Coyle (1971)
also noted that second instar spiderlings, the dis-
persal stage, desiccated quickly unless kept in
high humidity. If desiccation is a factor, it most
likely occurs during the dry summer months (see
July 1977 and August 1978 demographics for
burrows found in the previous spring); there was
1 cm precipitation from June through August
both in 1977 and 1978. Cannibalism may be
significant, especially in dense populations, since
positive geotropism might influence spiderlings
dispersing in search of burrow sites to enter oc-
cupied spider burrows. The proportionally large
number of missing burrows in the smaller size
classes during the April 1977 random sample and
the May 1978 census suggests that starvation

during the winter, when food is limited due to
snow cover and cold temperatures, may be a
significant mortality factor for spiderlings.

Longevity,— Mygalomorph spiders have been
known to live a long time. Baerg (1963) kept
certain theraphosids alive for at least 20 years in
the laboratory and believed that one specimen
lived 26 years (Baerg 1970). B. Y. Main (pers.
comm.) estimates that a ctenized, Anidiops vil-
losus (Rainbow), can live at least 23 years in the
field. In estimating longevity of A. riversi, it is
necessary to approximate several factors: date of
oviposition and eclosion, date of emergence and
burrow establishment, number of years to reach
a size 14 (the largest size class), and the tenure
of this size class. Since oviposition and eclosion
occur in the summer and early fall respectively,
and emergence in the following spring, the spi-
ders in the size 3 burrows in September were
about a year old. Based on extrapolations from
the growth rates observed for all size classes from
September 1976 to August 1978 (Table 3), the
time it could take a size 3 burrow to become a

38

THE JOURNAL OF ARACHNOLOGY

size 1 4 varied considerably. Size classes 3 and 4
most frequently grew one size class in two years,
size class five most frequently grew two size class-
es in two years, size class seven commonly grew
three size classes in two years, and sizes classes
10 and 12 could grow two size classes in two
years. Using these growth rates as an estimate,
it would take some burrows 13 years to reach
size 14. Other extrapolations, from Table 3, could
reasonably be used to estimate minimum to
maximum time to size class 14. Finally, in Sep-
tember of 1976 there were three size 14 spiders;
on 22 September 1979, three years later, one was
still alive. I estimate, therefore, that under sim-
ilar environmental conditions and growth rates
A. riversi can live in the field at least 16 years.

Most spiders (araneomorphs) live one to three
years depending on the species (Bonnet 1935).
Why do mygalomorphs live so long? Main (1976),
referring to arid adapted trapdoor spiders, sug-
gests it is advantageous for an adult female to be
able to wait out several continuous years of un-
favorable weather conditions that may disrupt
the emergence of reproductively active males. In
the fall, at Blodgett Forest, few females with eggs
or brood were collected, yet during the spring
many gravid females were collected. Lack of egg
deposition could be due to insufficient acquisi-
tion of food during the summer for complete egg
development by fall. Living several years would
increase the chances of obtaining enough food.
Unlike some araneomorphs (Turnbull 1964,
Riechert 1976), A. riversi does not appear to
change initial burrow locations to take advantage
of potentially more productive areas.

In addition to biotic mechanisms, A. riversi
may achieve a relatively long life because of its
sheltered microhabitat. It has been suggested that
the burrowing desert scorpion Paruroctonus me-
saensis (Vaejovidae) achieved its long life (ca.
five years ) as one result of its stable and pre-
dictable subterranean microhabitat (Polis & Far-
ley 1 980). The burrows of riversi protect them
from wind, to some extent rain and runoff, and
some potential predators. Additionally, an influx
of moisture from surrounding soil (Vogel 1978)
and regulating the turret entrance opening (Vin-
cent pers. obs.) may help prevent desiccation
during the dry California summers.

ACKNOWLEDGMENTS

This work was completed in partial fulfillment
of the requirements for the degree of Doctor of

Philosophy in the Department of Entomology of
the University of California at Berkeley. I thank
Drs. F. A. Coyle, O. F. Francke, M. H. Green-
stone, A. Kronk, B. Y. Main, J. M. Pound, and
E. 1. Schlinger for their helpful suggestions on
earlier versions of this manuscript. I am grateful
to Dr. 1. A. Boussey for his field assistance, Dr.
P. Rauch for his help with data management,
and Mr. K. Lindahl for statistical analyses. I also
thank Mr, R. Heald and the School of Forestry
of the University of California, Berkeley for their
cooperation and the use of the Blodgett Forest
Research Station.

LITERATURE CITED

Anderson, J. F. 1 974. Responses to starvation in Ly-
cosa lenta Hentz and Filistata hibernalis (Hentz).
Ecology, 55:576-585.

Atkinson, G. F, 1886a. A family of young trap-door
spiders. Ent. Americana, 2:87-92.

Atkinson, G. F. 1886b. Descriptions of some new
trap-door spiders; their notes (sic) and food habitats.
Ent. Americana, 2:109-117, 128-137.

Baerg, W. J. 1963. Tarantula life history records. J.

New York Ent. Soc., 71:233-238.

Baerg, W. J. 1970. A note of the longevity and molt
cycle of two tropical theraphosids. Bull. British Ar~
achnol. Soc., 1: 107-108.

Bickel, P. J. & K. A. Daksun. 1977. Mathematical
Statistics: Basic ideas and selected topics. Holden-
Day, Inc., San Francisco, California.

Bonnet, P. 1935. La Longevite chez les Araignees.

Bull. Soc. Ent. France. 40:272-277.

Cochran, W. 1977. Sampling Techniques. 3rd Ed., J.
Wiley & Sons. New York.

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. Bull. Mus. Comp. Zool. 141:269-
402.

Coyle, F. A. 1986. The role of silk in prey capture
by nonaraneomorph spiders. Pp. 269-305, In Spi-
ders: Webs, Behavior, and Evolution (W. A. Shear,
ed.).

Decae, A. E., Caranhac, G. & G. Thomas. 1982. The
supposedly unique case of Cyrtocarenum cunicu-
larium (Oliver, 1811) (Araneae, Ctenizidae). Bull.
British Arachnol. Soc. 5:410-419.

Humphreys, W. F. 1976. The population dynamics
of an Australian wolf spider, Geolycosa godeffroyi
(L. Koch 1865) (Araneae, Lycosidae). J. Anim. EcoL,
45:59-80.

Kotzman, M. 1990. Annual activity patterns of the
Australian tarantula Selenocosmia sterlingi (Ara-

VINCENT ^NATURAL HISTORY OF ATYPOIDES RIVE RSI

39

neae, Theraphosidae) in an arid area. J. ArachnoL,

18:123-130.

Main, B. Y. 1 976. Spiders. The Australian Naturalist

Library. William Collins Publishers. Sydney and

London.

Main, B. Y, 1978. Biology of the arid-adapted Aus-
tralian trap-door spider Anidiops villosus (Rain-
bow). Bull. British ArachnoL Soc., 4:161-175.

Marples, B. J. & M. J. Marples. 1972. Observations
of Cantuaria toddi and other trapdoor spiders (Ar-
anea: Mygalomorpha) in Central Otago, New Zea-
land. J. R. Soc. New Zealand, 2:179-185.

McQueen, D. J. 1978. Field studies of growth, re-
production, and mortality in the burrowing wolf
spider Geolycosa domifex (Hancock). Canadian J.
ZooL, 56:2037-2049.

McQueen, D. J. 1983. Mortality patterns for a pop-
ulation of burrowing wolf spiders, Geolycosa dom-
ifex (Hancock), living in southern Ontario. Cana-
dian J. Zool. 61:2758-2767.

Miller, G. L. & P. R. Miller. 1984. Correlations of
burrow characteristics and body size in burrowing
wolf spiders (Araneae: Lycosidae). Florida Ento-
moL, 67:314-317.

Miller, P. R. & G. L. Miller. 1991. Dispersal and
survivorship in a population of Geolycosa turricola
(Araneae, Lycosidae). J. ArachnoL, 19:49-54.

Polis, G. A. & R. D. Farley. 1 980. Population biology
of a desert scorpion: survivorship, microhabitat, and
the evolution of life history strategy. Ecology, 6:620-
629.

Riechert, S. E. 1976. Web-site selection in the desert
Agelenopis aperta. Oikos, 27:31 1-315.

Riechert, S. E. 1978. Games spiders play: behavioral
variability in territorial disputes. Behav. EcoL So-
ciobioL, 3:135-162.

Rivers, J. J. 1891. Description of the nest of the
California turret building spider, with some refer-
ences to allied species. Zoe, 2:318-320.

Smith, C. P. 1908. A preliminary study of the Ara-
neae Theraphosae of California. Ann. Ent. Soc.
America, 1:207-246.

Tumball, A. L. 1964, The search for prey by a web-
building spider Achaearanea tepidariorum (C. L.
Koch) (Araneae, Theridiidae). Canadian EntomoL,

96:568-579.

Vincent, L. S. 1980. The population biology of Aty-
poides riversi (Araneae, Antrodiaetidae) a fossorial
mygalomorph spider. Ph. D. dissertation, Univ. of
California, Berkeley. 149 pp.

Vincent, L. S. & Rack, G. 1982. Pseudopygmephorus
atypoides Rank n. sp. (Acari: Pygmephoridae) as-
sociated with the fossorial mygalomorph spider,
Atypoides riversi O. P.-Cambridge (Araneae: Antro-
diaetidae) in California. Pan-Pacific EntomoL, 58:
216-222.

Vincent, L. S. 1985. The first record of a tachinid fly
as an internal parasitoid of a spider. Pan-Pacific
EntomoL, 61:224-225.

Vincent, L. S. 1 986. Pathogens and parasitoids of the
fossorial mygalomorph spider Atypoides riversi O.
P.-Cambridge (Antrodiaetidae: Araneae) of various
size classes. Proc. IX Intemat. Congr. ArachnoL
Panama, 224-225.

Vogel, S. 1978. Organisms that capture currents. Sci.
American, 239: 128-139.

Wilson, R. S. 1970. Some comments on the hydro-
static system of spiders (Chelicerata, Araneae) Z.
Morph. Tiere, 68:308-322.

Manuscript received 10 March 1992, revised 30 No-
vember 1992.

1993. The Journal of Arachnology 21:40-49

ASPECTOS DE LA BIOLOGIa REPRODUCTIVA DE
LINOTHELE MEGATHELOIDES (ARANEAE: DIPLURIDAE)

Nicolas Paz S.: Departamento de Biologia, Universidad de Antioquia, Medellin,
Colombia

ABSTRACT. Of 50 specimens of Linothele megatheloides (Raven) growing under laboratory conditions, only
22 reached a sexually mature state after 10 and 1 1 ecdyses. The average number of eggs per sac was 161.5; the
average egg diameter was 22 mm. The spiders were monitored in the laboratory to determine the inter-instar
rate of growth. Eggs held in paraffin and carboxymethyl-cellulose developed only to gastrula. Under laboratory
conditions, spiders did not build complete egg sacs, and the eggs were eaten. Several egg sacs were destroyed
by fungus and parasitoids. The time of emergence of the spiderlings after oviposition was between 23 and 27
days. The reproduction period occurred between April and October and was apparently related to rain and
relative humidity. The spiders have little parental care of the eggs and spiderlings.

RESUMEN. De los 50 ejemplares de Linothele megatheloides (Raven), creciendo bajo condiciones de labo-
ratorio, solamente 22 llegaron a la madurez sexual, luego de entre 10 y 1 1 mudas. Las aranas se monitoriaron
en el laboratorio para determinar el incremento entre un instar y el proximo. Los huevos incluidos en parafina
liquida y en carboximetil celulosa, solamente llegaron a gastrula. Bajo condiciones de laboratorio, las aranas no
construyeron la ooteca completa y sus huevos eran normalmente comidos por ella. Varios sacos de huevos
fueron parasitados y destruidos por hongos y parasitoides. El tiempo de emergencia de las aranitas luego de ser
puesta la ooteca estuvo entre 23 y 27 dias. El periodo de reproduccion estuvo entre abril y octubre, situacion
aparentemente relacionada con el incremento en humedad relativa. Las aranitas no mostraron inversion parental
alta con sus huevos y crias.

En concordancia con los resultados obtenidos
por Paz (1988), al estudiar por primera vez en
el neotropico aspectos de la biologia de L. me-
gatheloides en bosques primarios de Panama y
Colombia relacionado con algunos patrones de
su conducta, se considero que otras situaciones
deberian ser investigadas y de manera especial
su biologia reproductiva. Asi, se diseno esta se-
gunda fase con los objetivos siguientes: verificar
si las aranas se reproducen a traves de todo el
ano, y si existe alguna relacion con las condicio-
nes climaticas reinantes en el area de estudio;
determinar el tiempo promedio de incubacion;
establecer la posible relacion entre el tamano de
la caparazon y algunos apendices (tarso uno; lar-
go quelicero y palpo) con la madurez sexual; el
peso y tamano de la ooteca y el numero promedio
de huevos por ooteca, numero posible de mudas
necesarias para alcanzar su madurez sexual;
tiempo y mecanismo de dispersion de las ara-
nitas y hasta donde fuese posible seguir los pri-
meros estadios embrionarios dentro del huevo y
la conducta maternal de la arana frente a su oo-
teca y aranitas.

METODOS

El area de estudio correspondio a la misma
descrita por Paz (1988). Alii se seleccionaron al-
gunos nidos al azar, se marcaron con tiras de
telas de color, con el objeto de hacer mas facil
su posterior localizacion, especialmente en boras
noctumas.

En el area se capturaron 50 aranas al azar desde
puber hasta adultas, segun metodo descrito por
Paz (1988). A las mismas se les midio largo y
ancho de la caparazon; largo del palpo, quelicero
(no incluida la una) y del tarso uno, con el fin de
establecer posibles correlaciones de su creci-
miento y la madurez sexual. A las aranas nor-
malmente se les regresaba a su nido luego de
medirlas o bien se seleccionaban algunas con ab-
domen muy redondeado (consideradas gravidas)
para traerlas al laboratorio en donde se les co-
locaban en cajas separadas de plexiglas.

Tambien en caso de encontrar nidos con ara-
nas gravidas o con ootecas, se les marcaba para
hacerles futuros seguimientos, o se colectaban
sus ootecas las que traidas al laboratorio se les

40

PAZ-REPRODUCCION DE LINOTHELE MEGATHELOIDES

41

media su diametro, se pesaban en una balanza
digital Quantum-Q-800, se abrian para contar
sus huevos, medirles su diametro e incluirlos en
un recipiente de plastico de 24 depresiones de 1
cm; 12 de los cuales contenian CMC y 12 con
parafina Hquida, para observar posibles cambios
embrionarios. Otras ootecas, se colocaban direc-
tamente en cajas de petri con algodon en una
incubadora de ICOPOR, con una fuente termica
de 60 W, temperatura entre 26-28 ®C y humedad
relativa entre 80-90%.

Con las ootecas producidas en el laboratorio,
se trabajo en igual forma y sirvieron para deter-
minar el tiempo de permanencia de las crias den-
tro de ellas, desde la oviposicion, hasta su eclo-
sion.

Aranitas inmaduras de pocos dias de haber
salido de la ooteca pero aiin asociadas con ella,
fueron coleccionadas, lo mismo que algunas que
ya la habian abandonado. A las primeras se les
colocaba en pequenos recipientes plasticos den-
tro de la incubadora, para hacerles seguimiento
post-embrionario de muda y de interaccion entre
ellas, para completar las observaciones de las
nacidas en el laboratorio. El seguimiento de muda
y de correlacion de las estructuras somaticas a
traves del crecimiento, se inicio con ejemplares
de segundo instar, correspondiente a aquellas
aranitas recien salidas de su ooteca al romperla.
Se tomaban de 6-8 ejemplares por cohors, se
pesaban cada una y luego se mataban en alcohol
al 20% para medirles las estructuras menciona-
das.

A partir de este estadio, las sucesivas medi-
ciones se hicieron sobre las respectivas exuvias
cada vez que mudaban y se pesaban 2 6 3 dias
despues de mudar. La pesada se hizo en una
balanza analitica hasta el instar 8, y a partir de
este estadio, en la digital Quantum en pequenos
recipientes de plastico. Los valores obtenidos para
ocho ejemplares de tres camadas diferentes para
cada instar se registraban y se determinaba el
tiempo entre una muda y la siguiente.

A las aranitas se les alimentd cada dos dias
inicialmente con Drosophila melanogaster y otras
especies de las cepas existentes en el laboratorio
de genetica y a partir del segundo instar con mos-
cas domesticas y pequenos hombpteros, cole6p-
teros, ort6pteros, hemipteros e is6podos, cuyos
tamanos se incrementaban en concordancia con
los instar. A medida que las aranitas crecian se
separanan en grupos menores para evitar pre-
dacidn entre ellas.

Con las adultas se hicieron observaciones, de

Tabla I.™ Valores promedios mensuales y anuales
de la humedad relativa (HR) con sus promedios, max-
imos y minimos, precipitacidn y el numero de dias que
llovid por mes en el area de estudio. Tornado del “Cal-
endario” Meteoroldgico del Himat, (Inst. Hidrologico,
Meteorologico y adecuacidn de tierras) para Colombia
1988.

°C °C Dias

Mes

HR

%

°C

(X)

(X)

max-

imo

(X)

min-

imo

Precip.

(mm)

de

Iluvia/

M

Enero

88

26

30

23

561

23

Febrero

86

26

30

23

480

20

Marzo

87

27

31

23

513

21

Abril

87

27

31

24

569

23

Mayo

87

27

31

23

705

26

Junio

87

27

32

23

760

24

Julio

86

27

31

23

784

26

Agosto

86

26

31

23

899

27

Septiembre

87

26

31

23

700

25

Octubre

88

26

31

23

611

26

Noviembre

87

26

30

23

692

25

Diciembre

88

26

30

23

667

25

X anual

87

26

31

23.1

661.7

24.25

fase precopulatoria, copulatoria y post-copula-
toria. Para la nominacidn de los instares se siguio
la nomenclatura de Vachon (Foelix 1982).

Los cambios meterol6gicos a traves del ano
con los registros promedios de humedad relativa,
temperatura y aquellas precipitacidn pluvial
mensual se obtuvieron del boletin mensual de la
estacidn del HIMAT para esta area del Choc6
(1988). Desde febrero 1988 hasta julio 1989 se
visit6 cada mes y medio el area de estudio, con
un periodo de permanencia entre ocho y cinco
dias.

RESULTADOS Y DISCUSION

De los valores de factores ambientales la pre-
cipitacibn suele incrementar manifiestamente a
partir del mes de abril hasta septiembre, alcan-
zando su maximo durante el mes de agosto. Exis-
te aparentemente una relaci6n entre la epoca en
que se reproducen estas aranas y el incremento
de los periodos de humedad ambiental. S61o a
partir de los liltimos dias del mes de abril del
primer semestre del 1988 y 1989, al incrementar
las Iluvias, se encontraron las primeras ootecas
sin reventar (tres en el 1988 y dos en el 1989).

Caparazon, ooteca y numero de huevos.— El
numero de huevos promedio de siete ootecas
abiertas al azar fue de 161.5 (cuatro procedentes

42

THE JOURNAL OF ARACHNOLOGY

Tabla 2.— Aqui se representan las ootecas, con sus correspondientes valores para ancho, largo, peso de cada
una y su numero correspondiente de huevos. Los asteriscos representan las ootecas abiertas para contar su
contenido.

N° ooteca

Ancho

caparazbn

(cm)

Largo

ooteca

(cm)

Peso

ooteca

(g)

N° de huevos

1

.8

2.4

1.5

No abiertos (incubacidn)

2

.9

2.1

No fue reiterada (oofagia)

_

3

.9

2.8

1.7

97

4

1.0

2.9

1.8

132

5

1.2

3.0

1.7

No abiertos (incubacidn)

6

1.2

3.0

No fue reiterada (oofagia)

—

7

1.2

2.9

1.6

184

8

1.1

2.8

No fue reiterada (oofagia)

9

1.1

2.8

No fue reiterada (oofagia)

—

*10

1.1

2.9

1.9

209

*11

1.0

2.1

1.9

183

12

.9

2.1

No fue reiterada (oofagia)

__

*13

1.3

2.8

1.9

207

*14

1.3

2.9

2.0

219

15

1.3

2.6

3.0

No abiertos (incubacion)

16

1.3

2.8

2.0

No abiertos (incubacidn)

del campo y tres del laboratorio), no se abrieron
para evitar alta perdida de aranitas. El numero
de estos por saco tendio a incrementar con los
mayores valores del ancho de la caparazon, largo
y peso de la ooteca, tabla 2. El diametro pro-
medio de los mismos para 100 tornados al azar
de cinco sacos (5 x 20) fue de 2.2 mm (rango
de 1.72-2.45).

Miyashita (1987) informa que Valerio (1976),
al trabajar con Achaearanea tepidariorum (Koch)
en Centro America (Costa Rica) encontro que el
potencial promedio de huevos puestos por una
hembra al ano es de 321 1.9 y 14.1 ootecas, po-
siblemente debido a una mayor presion de se-
leccion de predacion. Peaslee (1983) reporto para
Octonoba octonarius (Muna), un numero pro-
medio de huevos por ootecas de 78 (rango 45~
107). Galiano (1972, 1973) estudiando el desa-
rrollo post-embrionario de aranas de Therap-
hosidae y Dipluridae en la Argentina, encuentra
para Acanthoscurria sternalis (Pocock) un pro-
medio de huevos de 1 050-1 130 con diametros
de 1.3-1. 6 mm, para Avicularia avicularia (Lin-
naeus) de 70-1 12 huevos con diametros de 3.86-
4.06 mm, y para Ischonothele siemensi (Cam-
bridge) de 80-150 huevos con diametro de los
0.9-1 mm.

En cinco sacos (2, 6, 8, 9 y 10) puestos bajo
condiciones de laboratorio no retirados del re-
cipiente con la arana, fueron consumidos por

esta. Estos casos de oofagia estarian relacionados
con factores ad versos que bajo condiciones de
laboratorio, inducirian a no colocar las tres capas
de seda con las que suelen proteger y aislar sus
huevos, facilitando su consumo luego de ovi-
positardos.

Oviposicion vs. incubacion: Aranas gravidas
tejen en un area de su red una especie de tela
cuyo tejido es mucho mas fino y fusionado que
el resto de ella, semejando una verdadera “man-
ta”, de color bianco, en parte similar a la que
construye cuando va a mudar. Alii, la arana de-
posita primero un liquido claro bastante gelati-
noso, e inicia la deposicion de sus huevos alter-
nadamente, y luego comienza el tejido de las
capas de seda para aislar la ooteca a manera de
camara (Fig. 1). Bajo condiciones de cautiverio,
la colocacion de la ultima capa, no se observo,
quedando asi los huevos expuestos. Esta capa
poco elastica y altamente resistente, ademas de
aislar y proteger los huevos, servira por su puesto
para mantener la humedad en el interior del saco,
proveida por el liquido siruposo. El numero de
ootecas por epoca de reproduccion fue de una en
condiciones de laboratorio.

El tiempo de permanencia dentro del saco, se
estimo con tres ootecas (1, 5, y 16) las cuales no
se abrieron. La uno procedia del campo con fecha
de oviposicion junio 4, 1988 y las dos restantes
del laboratorio puestas en julio 26, 1988, agosto

PAZ=~REPRODUCdON DE LINOTHELE MEGATHELOIDES

43

Figura 1 .—Linothele megatheloides, al fina 1 izar una oviposicion. Observese la tupida red del saco y su abdomen

muy contraido.

6, 1988. La emergencia de las aranitas fue en su
orden a los 22, 27, y 23 dias, por lo cual el tiempo
promedio de permanencia se estimo en 24.5 dias,
con ranges de 22-27.

El verdadero tiempo de incubacion (periodo
entre la oviposicion y la prelarva intracorionica),
no fue determinado, ya que los huevos transfe-
ridos a CMC y parafina liquida no alcanzaron a
pasar de la fase de gastrula (la que se caracterizo
en ambos medios por el recogimiento de blas-
todermo alcanzado entre las 48-72 h).

Durante el desarrollo embrionario, observe que
buen nmero de huevos se ennegrecen o endu-
recen prontamente, lo que se puede atribuir a
que no estaban fecundados o haber side para-
sitados (Fig. 2).

De las ootecas abiertas, no fue posible deter-
minar con seguridad el nmero de huevos fecun-
dados, ya que al ser la oviposicion secuenciada
es normal esperar diferencias en el desarrollo em-
brionario y post-embrionario entre los huevos
fecundados, lo que se refleja en las aranitas al
emerger del saco.

Galiano (1972) encontro que la dipluridae
Ischnothele siemensi (Cambridge) tiene un pe-
riodo de incubacion de los 10-12 dias en Argen-
tina; Eason (1969), reporta que el periodo para

Pardosa lapidicina (Emerton), es de 23.4 dias con
rangos de 1 7-30, dependiendo de las condiciones
ambientales; para Cyrtophora moiuccensis (Do-
leschall), el tiempo estimado por Berry (1987)
fue de 24.4 dias con rangos de 24-28; Moore
(1977) reporta para Nephila clavipes (Linnaeus)
un periodo aproximado de un mes para el sur
de Norte America, siendo menor a nivel tropical,
posiblemente por las condiciones ambientales
mas favorables.

Estas aranas, de acuerdo a lo observado en el
campo y laboratorio, no suelen cuidar sus oo-
tecas, lo que fue normal encontrar en algunas
especies de Lycosidae, Pisauridae, Sparassidae
(Heteropodidae), Clubionidae, Araneidae, The-
ridiidae y otras familias que permanecian vigi-
lantes frente a ellos.

Emergencia desde la ooteca y dispersion: Las
aranitas al emerger del saco, no suelen abando-
narlo, sino que permanecen sobre el, en una pe-
quena red comunal. Alii permanecen hasta la
segunda muda (1-2) y a partir de este instar, co-
mienzan a tratar de construir pequenas redes in-
dividuates cada vez mas separadas, hasta alcan-
zar su tercer instar. En este estadio comienzan a
perseguirse y mantener una mayor distancia in-
dividual, conducta de agresion que se incrementa

44

THE JOURNAL OF ARACHNOLOGY

Figura 2.— Ooteca puesta en cauterio, con sus huevos no protegidos. Algunos han comenzado a ennegrecerse
y corresponden normalmente a no fecundados o haber sido parasitados.

con los sucesivos estadios. Fue necesario en este
estadio separarlas en grupos mas pequenos y en
recipientes de mayor espacio, para evitar el ca-
nibalismo. For esta razon, a partir del quinto
instar, cada recipiente solo contenia dos ejern-
plares 1-7. En el campo, es bastante posible que
a partir del 1-3 es cuando las aranitas comienzan
a dispersarse cada vez mas.

Relacion sexual. — Fue notorio el desfase ob-
servado entre hembras y machos, en los am-
bientes naturales y en cautiverio. Es asi que du-
rante los seis meses de trabajo en el Parque de
la Soberania de Panama, solo encontraron dos
machos sexualmente maduros y en casi dos anos
de salidas de campo al area del Choco, cinco. De
las aranas levantadas en condiciones de labora-
torio, hasta su madurez sexual {n = 1 9), solo una
emergida de una ooteca colocada en mayo 1988,
luego de la 10° muda, el extreme distal de sus
palpos evidencio transformacion en gonopidos.
Esta diferencia en la relacion de sexos se podria
explicar sobre el hecho que al llegar a su madurez
sexual (entre la 9-10 mudas), los machos aban-
donan sus redes (mas pequenas) e inician la fase
deambulatoria en busca de hembras receptivas
para copular, con lo cual se expondrian mas a
los depredadores.

Desarrollo embrionario y post-embrionario.—

Con los huevos colocados en parafina Hquida,
para seguir su desarrollo embrionario de acuerdo
al metodo de Holm/Galiano (1972, 1973a,
1973b) y los incluidos en CMC, con igual objeto,
no fue posible hacer este seguimiento. A pesar
que la membrana corionica era lisada y la ger-
minal o embrionaria permanecia normal, el de-
sarrollo del embrion solo alcanzo a llegar a “gas-
trula”. Su detencion en CMC, posiblemente se
debe a que esta sustancia se endurece poco a poco
y a las 72 h, estaba practicamente semisolificado.
En la parafina a pesar que esta permanecio nor-
mal, los estadios tampoco continuaron, ignoran-
dose su causa.

Entre las 8-2 horas despues dela oviposicion,
se podia observar al microscopio y al stereo, las
primeras etapas de division del disco germinal
blanquesino, y a las 24 h, este ya habia dado
origen al blastodermo, con aparatentes celulas
poligonales (Fig. 3-6).

La “prelarva”, se obtuvo de ootecas abiertas,
cuyos huevos contenian la aranita protegida por
la cuticula embrionaria transparente, sobre la cual
no fue facil detectar el area donde estarian los
dientes de eclosion y responsables de la ruptura
de la membrana del corion. Esta etapa se carac-

PAZ-~REPRODUCdON DE LINOTHELE MEGATHELOIDES

45

3

4

5 6

Figuras 3-6.~Primeras etapas del desarrollo embrionario de huevos de L. megatheloides. 3, huevo con corion
reventado y el germoplasma con su disco germinal dentro de la membrana; 4, capa de blastodermo con sus
celulas poligonales; 5, 6, contraccion del blastodermo y exposicion del espacio perivitelino.

teriza porque el cefalotorax esta manifiestamente
inclinado hacia abajo, las patas plegadas sobre
los flancos pleurales, las espinneretas posteriores
son mas desarrolladas que las anteriores y no hay
cerdas, ni tricobotrias ni pigmentacion somatica
(Fig. 7, 8).

La “larva” (deutovium), equivale al primer
instar postembrionario (I-l). Aqui la pre-larva
se ha liberado de la membrana embrionaria o
cuticula (primera exuvia), al mudar por primera
vez; sus miembros y cuerpos presentan cerdillas,
se evidencia podomerizacion pero no tricobotria
en sus patas, las que aparecen no plegados al
abdomen; queliceros separados, unas visibles,
pigmentacion foliar y mancha ocular solo lige-
ramente perceptibles. Esta fase es dentro de la
ooteca.

El tercer estadio (1-2) post-embrionario co-
rresponde a la “ninfa”, en donde los procesos

dentarios de las unas tarsales y queliceros son
bien detectables, lo mismo que los ojos, el patron
de pigmentacion del folium y cefalotorax, espin-
nereta posterior muy elongada, placas epigastri-
cas y cerdas de sus miembros y cuerpo visibles.
En este estado salen de la ooteca y son excelentes
constructoras de redes adyacentes al saco (Fig.
9).

El tiempo entre el I-l y el 1-2 estuvo de los
20-26 dias, para n= \9 ejemplares. A partir del
I-l, se procedio a determinar el posible factor de
progresion de crecimiento (FPC) relacionado con
el incremento promedio en el tamano de algunas
de sus partes y peso entre un instar y el proximo.
Los valores promedios tornados de las medicio-
nes directas de cada estructura utilizada y del
peso somatico, su respectivo estadio y FPC, se
representan en las tablas 3 y 4, lo mismo que el
tiempo entre una muda y la siguiente. Notese

46

THE JOURNAL OF ARACHNOLOGY

T abla 3 . — Relacion de valores promedio de las variables utilizadas y sus respect! vos estadios post-embrionarios
de Linothele megotheloides, luego de emerger de su ooteca y alcanzar su madurez sexual, n = 22; PS = Peso
somatico; LQ = Largo queliceros; LP = Largo palpal; AC = Ancho caparazon (cm); Tj = Largo tarso uno (cm);
xFPC = FPC de las medias; I = Instar; DTFPC = Desviacion tipica del FPC; FPC = Factor de progresion de
crecimiento; T = Tiempo promedio intermuda en dias.

PS

Instar

(g)

FPC

AC

FPC

T,

FPC

LP

FPC

LQ

FPC

Tiempo

Muda-2

0.02

0.19

0.14

0.31

0.08

LI

Muda-3

0.05

2.23

0.23

1.28

0.20

1.44

0.61

1.99

0.11

1.23

23

L3

Muda-4

0.07

1.38

0.23

1.07

0.23

1.12

0.71

1.16

0.14

1.33

29

L4

Muda-5

0.15

2.27

0.31

1.23

0.27

1.17

0.92

1.30

0.19

1.36

35

L5

Muda-6

0.38

2.48

0.39

1.26

0.31

1.15

1.14

1.23

0.23

1.19

42

L6

Muda-7

0.67

1.76

0.48

1.24

0.39

1.29

1.42

1.25

0.28

1.22

47

L7

Muda-8

1.25

1.86

0.58

1.22

0.46

1.17

1.66

1.17

0.32

1.16

49

L8

Muda-9

2.23

1.79

0.78

1.34

0.59

1.27

2.04

1.23

0.39

1.20

72

L9

Muda- 1 0

2.86

1.28

1.08

1.37

0.76

1.28

2.35

1.15

0.43

1.12

91

LIO

Mudaril

3.56

1.24

1.28

1.19

1.04

1.37

3.43

1.46

0.54

1.26

109

XFPC

1.81

1.25

1.25

1.33

1.24

DTFPC

0.44

0.10

0.08

0.12

0.17

Tabla 4.— Valores de la x, su DT y el respective factor de progresion de crecimiento (FPC) para cada uno de
los instares obtenido en condiciones de laboratorio hasta llegar a su madurez sexual, n = 22; PS = Peso somatico;
AC = Ancho caparazon; LP = Largo palpal; T, = Largo tarsal uno; LQ = Largo quelicero.

Instar

PS

AC

LP

T,

LQ

X

0.02

± 0.0

0.2

±

0.0

0.3

± 0.0

0.1

±

0.0

0.1 ± 0.0

L2

FPC

1.0

± 0.1

1.0

±

0.1

0.9

± 0.1

1.0

+

0.2

0.9 ± 0.1

X

0.1

± 0.0

0.2

0.0

0.6

± 0.1

0.2

±

0.0

0.1 ± 0.1

L3

FPC

1.4

± 0.1

1.1

+

0.1

1.1

± 0.3

1.1

±

0.3

1.0 ± 0.5

X

0.1

± 0.0

0.3

0.0

0.7

± 0.0

0.2

±

0.0

0.1 ± 0.0

L4

FPC

1.1

± 0.0

1.1

±

0.2

1.0

± 0.2

1.0

+

0.13

1.1 ± 0.4

X

0.2

± 0.2

0.3

±

0.0

0.9

± 0.0

0.3

0.0

0.2 ± 0.0

1-5

FPC

0.7

± 0.2

0.9

+

0.2

1.0

± 0.0

1.0

±

0.2

0.9 ± 0.1

X

0.4

± 0.2

0.4

±

0.1

1.1

± 0.1

0.3

+

0.0

0.2 ± 0.0

L6

FPC

1.6

± 0.7

1.2

±

0.2

1.0

± 0.2

1.0

+

0.1

1.1 ± 0.2

X

0.7

± 0.2

0.5

±

0.0

1.4

± 0.1

0.4

±

0.1

0.3 ± 0.0

L7

FPC

0.7

± 0.0

0.9

±

0.0

1.0

± 0.1

1.0

±

0.1

0.9 ± 0.1

X

1.3

± 0.1

0.6

±

0.0

1.7

± 0.0

0.5

+

0.0

0.3 ± 0.0

L8

FPC

1.1

± 0.2

1.0

±

0.0

1.0

± 0.1

1.0

±

0.1

1.0 ± 0.1

X

2.2

± 0.1

0.8

±

0.0

2.0

± 0.0

0.6

0.1

0.4 ± 0.0

1-9

FPC

0.7

± 0.5

0.8

±

0.3

1.0

± 0.0

1.0

+

0.0

0.9 ± 0.0

X

2.9

± 0.2

1.1

±

0.1

2.3

± 0.0

0.8

±

0.1

0.4 ± 0.1

1-10

FPC

0.9

± 0.1

1.1

±

0.1

1.1

± 0.4

1.0

±

0.1

1.1 ± 0.2

PAZ^REPRODUCCION DE LINOTHELE MEGATHELOIDES

47

8

Figuras 7, S.—Estadio de la “pre-larva” dentro de
la membrana embrionaria. 7, vista lateral; 8, vista fron-
tal.

que el valor del FPC suele normalmente decrecer
en concordancia con el incremento en muda al
irse alcanzando la madurez sexual. Estos valores
asi obtenidos, se utilizaron con la formula del
metodo teorico de correlacion de crecimiento de
Franke & Sisson (1984).

Cuatro aranas que ovipositaron en cautiverio,
mudaron entre los 45-57 dias despues eviden-
ciando que siguen mudando luego de llegar a su
madurez sexual. A las 50 aranitas tomadas al
azar en el campo desde joven hasta adultas se
les determine los valores correspondientes al tar-
so uno (Ti), largo palpo (LP); largo quelicero (LQ);
ancho caparazon (AC) y largo caparazon (LC),
para establecer su posible correlacion. Encon-
trandose que el mayor valor correspondio al val-

Tabla 5,— Valores en cm de las variables anatomicas
para « = 50 y un nivel de confidencia del 95% (0.05).
LC;^. = Limites de confianza de la x.

Maximo Minimo

X

DT

LC^

LC

1.5

0.5

1.1

0.3

1.2

1.0

AC

1.3

0.4

1.0

0.2

o o

T,

1.0

0.3

0.7

0.2

1.0

0.9

LQ

0.9

0.3

0.6

0.1

0.6

0.5

LP

3.6

1.0

3.0

0.6

3.0

2.7

or entre el ancho y el largo de caparazon (r =
0.935) y el menor a la relacion entre el largo del
quelicero y el tarso uno (r = 0.065). Los valores
maximos, minimos, su media, d.t. y limites de
confianza de las medidas se dan en la tabla 5.

Stratton & Lowrie (1984) encuentran que estas
relaciones son mas positivas en hembras que en
machos de Schizocosa mccooki (Montgomery);
Austin (1984) reporta que el valor de la corre-
lacion entre tamano de la caparazon y el numero
de huevos para Clubiona robusta (Koch) fue de
r = 0.81 y el numero de mudas requeridas para
la madurez sexual de 9-10 a para hembras y 1-
9 para machos.

Baerg (1928) encontro que machos de Eury-
pelma californica (Ausserer) (tarantula), llegaron
a su madurez sexual leugo de 10-11 anos, con
un numero de mudas de 22 y una longitud ce-
falotoraxica de 47.3 mm y que eran poco abun-
dantes.

Parasitismo.— Se encontraron acaros asocia-
dos extemamente con las ootecas, las que al pa-
recer no se comportan como verdaderos para-
sitos, al no ver ninguno de ellos romperla o
devorar sus huevos. Cuatro ootecas tomadas en
el campo, presentaban huevos posiblemente de
Hymenoptera o Diptera. Tres aranas se encon-
traron en el campo dentro de la cueva, muertas
y cubiertas de un hongo filamentoso. Igual cosa
sucedio con dos en cuativerio. El hongo corres-
pondio a un deuteromicete, de acuerdo a su pa-
tron de crecimiento.

Algunos investigadores han reportado que buen
numero de muertes de aranas son debidas a var-
ias fuentes de parasitos o parasitoides, tal es el
caso de Poinar & Thomas (1985), McQueen
(1978), Raymond (1984), y Nentwig (1985).

48

THE JOURNAL OF ARACHNOLOGY

Figura 9.— Tercer estadio post-embrionario (ninfa), correspondiente al segundo instar (U2), luego de emerger

de la ooteca.

Conducta pre-copulatoria^—Una hembra que
habia ovipositado en julio, se le agrego un macho
capturado en el campo en septiembre 1988 el
que carecia de espermatoforos en sus gonopo-
dios. Al hacer contacto con la hembra, la toco
sigilosamente con los extremos de sus patas de-
lanteras, ante lo cual esta se retiraba varios cen-
timetros; esta conducta se repitio por tres oca-
siones, pero en la cuarta la hembra reacciono
persiguiendolo hasta una distancia de unos 12
cm. En el sexto encuentro ambos se enfrentan
inicialmente tocandose con los extremos del tar-
so. Esta actitud al parecer de reconocimiento, se
repitio con un acercamiento mayor entre las pa-
rejas, sin que la hembra tratara de perseguir o

lesionar al macho. El enfrentamiento fue de du-
racion variable en cuanto a tiempo desde 2.5
hasta los 13 min. El macho en ocasiones trato
de tocar el area ventral de la hembra con sus
palpos, sin obtener reacciones negativas por par-
te de esta. Las observaciones de esta actividad
precopulatoria fueron realizadas en tres ocasio-
nes con periodos de dos horas. Hembras recep-
tivas al introducirle el macho no mostraron con-
ducta agonistica, lo que si se evidencio con
hembras gravidas (no receptivas).

AGRADECIMIENTOS

Deseo dejar expreso mi gran reconocimiento
al ICFES, al Centro de Investigaciones de nuestra

PAZ==-REPRODUCCION DE LINOTHELE MEGATHELOIDES

49

Facultad, y al Comite del Centro de Investiga-
ciones por su gran colaboracion tanto en lo eco-
nomico como en lo administrative, sin lo cual
no se hubiese realizado satisfactoriamente este
trabajo, al no tener inconveniente en ninguna de
las salidas de campo y de compra de material.
Tambien la colaboracion del profesor Abel Diaz
del Centro de Asesoria Estadistica del Departa-
mento de Matematicas por su oportuna orien-
tacion en el proceso de los datos. Al senor Victor
Garrido Paz por su invaluable asistencia en el
trabajo de campo, aun bajo tiempos muy ad-
versos.

BIBLIOCRAFIA CITADA

Austin, A. D. 1984. Life history of Clubiona robusta
(L. Koch) and related species (Araneae, Clubioni-
dae) in south Australia. J. Arachno!., 12:87-104.
Baerg, W. J. 1928. The life cycle and mating habits
of the male tarantula. Quart. Rev. Biol, 3:109-1 16.
Berry, J. W. 1987. Notes on the life history and be-
havior of the communal spider Cyrtophora moluc-
censis (Doleschall) (Araneae, Araneidae) in Yap,
Caroline Islands. J. Arachnol., 15:309-319.

Eason, R. R. 1969. Life history and behaviour of
Pardosa lapidicina Emerton (Araneae: Lycosidae).
J. Kansas Ent. Soc., 42:339-360.

Foelix, R. F. 1982. Biology of spiders. Harvard Uni-
versity Press, Cambridge, 306 pp.

Francke, O. F. & W. D. Sisson. 1984. Comparative
review of the methods used to determine the num-
ber of molts to maturity in scorpions (Arachnida),
with analysis of the post-birth development of Vae-
jovis coahuiiae Williams (Vaejovidae), J. Arachnol.,
12:1-20.

Galiano, M. E. 1972. El desarrollo postembrionario
larval

de Ischnothele siemensi Cambridge, (Araneae; Diplur-
idae). Physis (Buenos Aires), 31:169-177.

Galiano, M. E. 1973a. El desarrollo postembrionario
larval en Theraphosidae (Araneae). Physis (Buenos
Aires), 32:37-46.

Galiano, M. E. 1973b. El desarrollo postembrionario
larval de Avicuiaria avicularia (Linnaeus) (Araneae;
Theraphosidae). Physis (Buenos Aires), 32:315-327.

McQueen, D. J. 1978. Field studies of growth, re-
production and mortality in the burrowing wolf spi-
ders Geoiycosa domifex (Hancock). Canadian J.
ZooL, 56:2037-2049.

Miyashita, K. 1987. Development and egg sac pro-
duction of Achaearanea tepidariorum (C.L. Koch)
(Araneae; Theridiidae) under long and short pho-
toperiods. J. Arachnol, 15:51-58.

Moore, C. W. 1977. The life cycle, habitat and vari-
ation in selected web parameters in the spider, Ne-
phila clavipes Koch (Araneidae). American Midi
Nat., 98:95-106.

Nentwig, W. 1985. Parasitic fungi as a mortality fac-
tor of spiders. J. Arachnol, 13:272.

Paz, S. N. 1988. Ecologia y aspectos del comporta-
miento en Linothele sp. (Araneae; Dipluridae). J.
Arachnol, 16:5-22.

Peaslee, J. E. & W. B. Peck. 1983. The biology of
Octonoba octonarius (Muma) (Araneae; Ulobori-
dae). J. Arachnol, 11:51-67.

Poinar, O. G. Jr. & G. M. Thomas. 1985. Laboratory
infection of spiders and harvestmen (Arachnida:
Araneae and Opiliones) with Neoaplectana and HeP
erorhabditis Nematodes (Rhabditoidea). J. Arach-
nol., 13:297-302.

Raymond, L. M. 1984. The egg sac of Pityohyphantes
costatus (Hentz) (Araneae, Linyphiidae) and its
phorid parasite. J. Arachnol, 12:371-372.

Stratton, G. E. & D. C. Lowrie. 1984. Courtship be-
havior and life cycle of the wolf spider Schizocosa
mccooki (Araneae; Lycosidae). J. Arachnol., 12:223-
228.

Manuscript received 9 October 1990, revised 31 July
1992,

1993. The Journal of Arachnology 21:50-54

SURVIVABILITY OF OVERWINTERING ARGIOPE AURANTIA
(ARANEIDAE) EGG CASES, WITH AN ANNOTATED LIST
OE ASSOCIATED ARTHROPODS

T. C. Lockley' and O. P. Young^: Southern Field Crop Insect Management
Laboratory, USDA, Agricultural Research Service, P.O. Box 346; Stoneville,
Mississippi 38776 USA

ABSTRACT. Overwintering egg cases of the black and yellow garden spider, Argiope aurantia Lucas (Araneae:
Araneidae), were observed during the late winter and early spring of 1985, 1986, and 1987 in Washington
County, Mississippi. Of 120 egg cases monitored in the field in 1985, only three remained undamaged by the
period of peak spiderling emergence in May. An additional 1 1 5 field-collected egg cases were observed in the
laboratory in 1985. A total of 23,840 A. aurantia spiderlings emerged from the lab egg cases (mean = 341), with
1212 spiderlings emerging from one undamaged egg case. Adults or pupae of either the parasitic ichneumonid
wasp, Tromatopia rufopectus (Cr.), or the parasitic chloropid fly, Pseudogaurax signatus (Loew), emerged from
56% of the field-collected egg cases. Nineteen species of insects, representing 19 genera, 15 families and 5 orders
were collected from lab-reared egg cases in 1985. In addition, 11 species of spiders were recovered from A.
aurantia egg cases. In 1985, 97% of the egg cases observed in the field showed evidence of bird damage. In both
1986 and 1987, 100% of the egg cases were damaged by birds.

The black and yellow garden spider, Argiope
aurantia Lucas, is a common orb-weaving spider
found throughout the eastern part of the United
States and along the west coast of North America
into Central America (Levi 1968). It has been
reported from a variety of habitats, including
dense perennial vegetation, dry grassy hillsides,
vegetable gardens, roadside and deciduous woods
margins, and areas adjacent to streams, ponds,
and swamps (Gertsch 1979). Observations on the
general life habits, systematics, and distribution
of A. aurantia and related species were sum-
marized by Levi (1968). Other workers have re-
ported on the biology of this species, including
overwintering behavior and ecology (Enders
1974, 1977; Riddle & Markezich 1981; Howell
& Ellender 1984; Heiber 1985). Minimal infor-
mation is available, however, concerning the na-
ture and degree of overwintering mortality. Adults
and juveniles of^. aurantia do not typically sur-
vive the winter, even in the southern United
States. Adult females of this species produce egg
cases containing many hundreds of eggs in late
summer and fall. The eggs hatch during winter

'Present address: USDA-APHIS-S&T-IFAS, 3505 25th
Avenue, Gulfport, Mississippi 39501.

^Present address: USDA-APHIS-BBEP-EAD, 6505
Belcrest Road, Hyattsville, Maryland 20782.

and the spiderlings remain in the egg case until
spring (Tolbert 1976). The present study exam-
ines the overwintering survivability of auran-
tia egg cases in old fields and roadside margins
of Washington County, Mississippi.

METHODS

In January of 1985, nine sites were selected
within the Delta Experimental Forest (DEE) lo-
cated 3.0 km north of Stoneville, Washington
County, Mississippi. Four of the sites (Sites 1, 6,
8 and 9) were roadside margins that averaged
1. 0=2.0 m in width and had varied plant com-
munities. Sites 2, 3, 4, 5, and 7 were old field
successional habitats that ranged from a rela-
tively small field (10 x 100 m, site 4) to an area
2.0 km long and 100 m wide (site 7). Mixed tall
forbs (e.g., Solidago sp. and Aster pilosus) pre-
dominated, All sites were within an area (3.5 km
X 1.0 km) of the DEE bounded on three sides
by soybean, cotton, or fallow fields.

At each site, egg cases of A. aurantia were de-
tected by walking parallel linear routes and vi-
sually searching the vegetation. Each egg-case lo-
cation was marked by a 0.5 m strip of red and
white flagging tape, with a unique alpha-numeric
code written on the tape in indelible ink. To
minimize attraction of birds to the colored tape
and its associated egg case, the tape was attached

50

LOCKLEY & YOUNG-OVERWINTERING ARGIOPE

51

to vegetation O.S-l .0 m distant from the egg case.

Data were recorded for each egg case and in-
cluded condition, height above ground, vegeta-
tion substrate, degree of exposure, and number
of adjacent egg cases within 2.0 m. Marked egg
cases were monitored at 30-day intervals in Feb-
ruary and March and at 1 5 -day intervals in April
and May. At each monitoring, the condition and
possible mortality cause(s) were deter-
mined. Tolbert (1976) concluded that damage
exceeding 1 0% of the surface area or subsequent
disappearance of the Argiope egg case was at-
tributable to bird predation. The same criterion
was used in our field evaluations. There is no
indication, from a search of the literature or from
our own observations, that mammal predation
is a significant mortality factor in this type of
situation. Damage that involved less than 10%
of the egg case surface was attributed to insects.
This category also was designated when insect
emergence holes were detected.

In January 1985, 235 egg cases of A. aurantia
were located; 120 were marked for future field
observations and 1 1 5 were removed for labo-
ratory observation. Each collected egg case was
placed in a plastic 8 oz.(235 ml) cup with an
organdy screen cover held in place by lids from
which a 5.0 cm diameter circle had been re-
moved. Cups were then placed outside the lab-
oratory in a screened enclosure with a rain cover.
These conditions approximated the temperature
and humidity regimens experienced by egg cases
that remained in the field. In January 1986 and
1987, field surveys for egg cases were conducted
at the same sites as in 1985; however, egg cases
were not collected.

RESULTS

Field Observations: —Eighteen percent of the
120 egg cases marked in January 1985 showed
previous damage, apparently caused by birds. By
mid-February, the percent of damage caused by
birds to these egg cases had increased to 64%. In
late May, 97% of the egg cases either had been
extensively damaged by birds or had disappeared
altogether. The remaining 3% of field egg cases
in 1985 showed evidence of either insect para-
sitization or predation.

On one occasion in 1985, we observed actual
bird damage to an egg case. During the morning
of 19 February, a male House Sparrow, Passer
domesticus domesticus, was seen removing the
contents of a previously undamaged egg case.

The bird was startled by our approach and took
flight with the eviscerated egg case clasped in its
beak and strings of material trailing behind as it
disappeared from view.

In mid- January 1986, 27% of the 143 detected
egg cases showed evidence of bird damage. In
mid-January 1987, all of the 1 3 detected egg cases
showed evidence of bird damage. Over all sites,
egg case density also declined during the three
survey years. In 1985, density averaged > 5 egg
cases per 30 square m. In 1986, this had de-
creased to slightly > 1.0 egg case per 30 sq. m.
In 1987, only 13 egg cases were located within
all 9 sites and averaged <0.1 egg case per 30
sq. m.

Laboratory Observations.—^, aurantia spi-
derlings emerged from all but one of 115 egg
cases retained in enclosures from January to June
of 1985. Total emerged spiderlings from egg cases
(« = 114) was 23,840; range per egg case: 1 (ex-
tensively damaged egg case) to 1212 (completely
undamaged egg case). The mean emergence from
the 114 egg cases was 341 (±81 SE) spiderlings.
Thirty-five of the 115 field-collected egg cases
were initially damaged by birds. Eventual emer-
gence of spiderlings from these egg cases (x = 55
± 1 7 SE) was considerably less than from insect-
damaged egg cases (x = 134 ±32 SE) and from
undamaged egg cases (x = 456 ± 148 SE).

During enclosure observations, more than 4700
non-host arthropods also emerged from the 115
egg cases (Table 1). Two species of wasps (Hy-
menoptera) comprised 83.7% of all emerging non-
host arthropods, and one species of fly (Diptera)
comprised an additional 14.5%. These three spe-
cies were: the ichneumonid wasp, Tromatobia
rufopectus (Say), the eulophid wasp, Pediobius
brachycerus (Thomson), and the chloropid fly,
Pseudogaurax signatus (Loew).

Thirty-eight (33%) of the 115 egg cases reared
in enclosures were parasitized by T. rufopectus
(x = 6.8 ±1.9 SE T. rufopectus pupae per egg
case). Only one adult of this species emerged,
however, because 258 of the 259 T. rufopectus
pupae were parasitized by P. brachycerus. This
hyperparasite produced more than 3700 indi-
viduals from the 258 host pupae (x = 14.4 ±5.1
SE per pupa). Seven additional species of Hy-
menoptera also were found in the examined egg
cases (Table 1). Of these, only the eupelmid, Ar-
achnophaga scutata Gahan, and the eulophid,
Tetrastichus sp., are known parasites of spider
eggs (Eason et al. 1967).

The chloropid fly, P. signatus, is an obligate

52

THE JOURNAL OF ARACHNOLOGY

Table 1.— Arthropods associated with 115 egg cases
of Argiope aurantia in 1985 in Washington County,
Mississippi. * Less than 1%.

Taxon

Percent

occur-
rence
in egg
sacs

Number

of

indivi-

duals

Psocoptera

*

1

Coleoptera

Carabidae

Calleida decora Fab.

*

1

Stenolophus dissimilis DeJ.

*

1

Hydrophilidae

Cercyon sp.

2

2

Lathridiidae

Corticaria sp.

3

3

Mycetophagidae

Litargus sp.

♦

1

Rhyncophoridae

Lixus concavus Say

*

1

Diptera

Chloropidae

Pseudogaurax signatus
(Loew)

43

687

Lepidoptera

Arctiidae

Lithosiinae

2

2

Noctuidae

Palthis asopialis Guenee

*

1

Hymenoptera

Braconidae

A gat his sp.

*

1

Eulophidae

Pediobius brachycerus
(Thomson)

30

3713

Pnigalio sp.

*

1

Tetrastichus sp.

*

1

Eupelmidae

Arachnophaga scut at a

Gahan

*

1

Formicidae

Tapinoma sessile (Say)

*

20

Ichneumonidae

Itoplectus conquistor (Say)

2

2

Tromatobia rufopectus Cr.

33

259

Pteromalidae

Pteromalus sp.

*

1

Araneae

Araneidae

Eustala cepina (Walck)

*

1

Dictynidae

Dictyna sp.

*

1

Table 1.— Continued.

Taxon

Percent

occur-
rence
in egg

sacs

Number

of

indivi-

duals

Dictyna hentzi Kaston

2

2

Philodromidae

Philodromus sp.

*

1

Salticidae

Eris marginata (Walck)

♦

1

Hentzia sp.

2

2

Maevia sp.

*

1

Metaphidippus sp.

5

6

Metaphidippus galathea

(Walck)

4

4

Phidippus audax Hentz

*

1

Phidippus clarus Keys

6

27

Tutelina sp.

*

1

Total

4746

predator of spider eggs (Heiber 1984). In 1985,
43% of the enclosure egg cases produced adult
F. signatus flies. However, only 26 egg cases were
attacked singly by this fly; an additional 23 egg
cases were attacked by both P. signatus and T.
rufopectus.

Other workers have indicated possible pre-
dation of A. aurantia eggs by lepidopterous lar-
vae (Heiber 1984, Austin 1985), In our study,
two arctiid larvae (subfamily Lithosiinae) were
found within the confines of damaged egg cases
(Table 1). This subfamily is known to feed only
on lichens (Holland 1968). One noctuid moth,
Palthis asopialis Guenee, emerged from an ex-
tensively damaged egg case. Its pupal case and
numerous fecal pellets were recovered from
within the egg case, suggesting that spider eggs
or egg case material had served as food for the
larva. Six species of Coleoptera were found in
association withy4. aurantia egg cases. Only one,
the carabid beetle Calleida decora Fab,, is a known
predator; however, it was not observed feeding
on spider eggs or spiderlings.

Eleven species of spiders, representing nine
genera and four families, were obtained from
field-collected A. aurantia egg cases in the lab-
oratory. These spiders probably were secondary
invaders that entered holes made by insects or
birds. In one instance, an egg case and dead fe-
male of the salticid, Phidippus clarus Keyserling,

LOCKLEY & YOUNG --OVER WINTERING ARGIOPE

53

were found in a damaged A. aurantia egg case.
Subsequently, 20 P. dams spiderlings emerged
on 2 April, followed on 3 May by 4 1 2 aurantia
spiderlings. Other spiders have been observed
feeding on aurantia spiderlings in the egg case
(Tolbert 1976). We, however, observed no such
interspecific predation by spiders.

DISCUSSION

Sources of mortality to overwintering spider
eggs can be partitioned into abiotic and biotic
parameters. In the southeastern United States,
abiotic factors (e.g., weather) are considered of
minor importance to winter survival of A. au-
rantia spiderlings inside egg cases (Tolbert 1979;
Riddle 1980). Biotic factors (e.g., predation and
parasitization) are postulated to have a more pro-
found affect on survival (Auten 1925; Eason et
al. 1967).

Birds have been recorded as a major group of
predators of orb-weaving spiders on their webs
(Marples 1969; Robinson & Robinson 1970;
Blanke 1 972; Waide & Mailman 1 977), including
A. aurantia (Horton 1983). Birds have also been
implicated as a major source of mortality for
overwintering arboreal spiders (Gunnarson
1983). Bird predation on spider egg cases and
their contents, however, has been mostly docu-
mented by anecdotal or circumstantial evidence.
Several studies estimated rates of bird predation
on orb-weaving spider egg cases that ranged from
7-42%(Enders 1974; Tolbert 1976; Heiber 1984).
These studies, however, were conducted only in
the fall of each observation year; continued ob-
servations into the spring probably would have
produced higher incidences of bird predation,
perhaps approximating the 100% values ob-
served during our study. During the January to
May period of our investigation, birds were for-
aging both for food and for nesting materials.
The local bird density was also increasing during
this period, as summer residents were returning
and migrants were passing through on the way
north. These factors suggest that the level of bird
predation on egg cases that we observed may be
both typical for such situations and comparable
with the results of other investigations.

Egg cases of^. aurantia that are damaged by
birds provide nesting sites and sheltered habitats
for many arthropod species, including other spi-
ders. Many of these associated species are pred-
ators or scavengers and may consume host spider

eggs or spiderlings. Conversely, they also may
consume other predators or parasites of spiders
and consequently reduce the overall impact of
such organisms on A. aurantia eggs and spider-
lings. Our data does not allow a determination
of the net effect of predator/scavenger arthropods
associated with egg cases on the population dy-
namics of A. aurantia.

The level of egg-case parasitization demon-
stated by r. rufopectus in 1 985 3 3% —is in gen-

eral agreement with values found in previous
studies (e.g., 21.5%, Enders 1974; 26.3%, Tolbert
1976; 36.1%, Heiber 1984). T. rufopectus is a
well known parasitoid of spider eggs and was first
described by Cresson (1870) from A. aurantia
egg cases. It attacks y4. aurantia eggs by inserting
its long ovipositor through the outer cover of the
egg case into the flocculent layer. Wasp eggs are
deposited on or near the host egg mass and the
emerging wasp larvae make their way to the host
eggs and burrow into the mass to feed. P. bra-
chycerus, a parasitoid of T. rufopectus, does no
known damage to spider eggs or spiderlings (Peck
1985).

Previous studies have shown the chloropid fly,
P. signatus, to be a fairly common parasitoid of
A. aurantia eggs (Enders 1974; Tolbert 1976;
Heiber 1984, 1985). However, parasitization
values observed during our study were 3-4 times
greater than those found previously (Tolbert
1976; Heiber 1984). P. signatus oviposits on egg
case surfaces, whereupon after fly egg hatch the
larvae force their way through the outer covering
into the host egg mass (Kessel & Kessel 1937;
Hickman 1970). Heiber (1984) found that the
level of successful parasitization of P. signatus
increased significantly when it attacked egg cases
already damaged by other parasitoids or preda-
tors. These data suggest that prior egg case dam-
age may be an important factor in successful par-
asitization by this chloropid fly.

The egg cases of A. aurantia are assumed to
have evolved to protect their contents from one
or more mortality factors. It is apparent from
our study, however, that these structures have
not prevented considerable mortality to their
contents caused by bird damage or by parasit-
ization. These two mortality factors, when added
to subsequent mortality of spiderlings in the egg
case caused by other agents, may be the major
determinants of A. aurantia population density
in old field and margin habitats. On the other
hand, bird predation or parasitization typically

54

THE JOURNAL OF ARACHNOLOGY

does not cause complete mortality in affected egg
cases. The number of survivors from such egg
cases, combined with those from unaffected egg
cases, may be sufficient to maintain population
levels of A. aurantia in a particular area. Because
of the large-scale removal of egg cases from our
field sites for laboratory study, the subsequent
decline in A. aurantia populations cannot be de-
finitively ascribed to either natural or investi-
gator-associated parameters. Determining the
relative importance of various mortality factors
associated with A. aurantia egg cases awaits fur-
ther experimentation.

ACKNOWLEDGMENTS

Identification of insects was provided by the
following members of the USD A Systematic En-
tomology Laboratory, Beltsville, MD: R. W.
Carlson, E. E. Grissell, P. M. Marsh, R. W. Poole,
C. W. Sabrosky, M. E. Schauff, and D. R. Smith.
Spiders were identified by G. B. Edwards, Div.
Plant Industry, Florida Dept. Agric. & Cons.
Serv., Gainesville, FL. Voucher specimens are
located at these institutions and in the personal
collection of the senior author. An exceptionally
thorough manuscript review was provided by D.
T. Jennings, with additional review by M. H.
Greenstone.

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.

Auten, M. 1 925. Insects associated with spider nests.

Ann. Entomol. Soc. America, 18:240-250.

Blanke, B. 1972. Untersuchungen zur Oekophsiolo-
gie und Oekethologic von Cyrtophora citricola For-
skal (Araneae: Araneidae) in Andulusien. Forma et
Functio., 5:125-160.

Cresson, E. T. 1870. Trans. American Ent. Soc.,
3:145-148.

Eason, R. R., W. B. Peck & W. H. Whitcomb. 1967.
Notes on spider parasites including a reference list.
J. Kansas Entomol. Soc., 40:422-434.

Enders, F. 1974. Vertical stratification in orb- web
spiders (Araneidae, Araneae) and a consideration of
other methods of coexistence. Ecology, 55:31 7-328.
Enders, F. 1977. Web-site selection by orb- web spi-
ders, particularly Argiope aurantia Lucas. Anim. Be-
hav., 25:694-712.

Gertsch, W. J. 1 979. American Spiders, 2nd Ed. Van
Nostrand Reinhold Co., New York.

Gunnarson, B. 1983. Winter mortality of spruce-liv-

ing spiders: effect of spider interactions and bird
predation. Oikos, 40:226-233.

Heiber, C, S. 1984. The role of the cocoons of orb-
weaving spiders. Ph.D, Dissertation, Univ. of Flor-
ida, Gainesville.

Heiber, C. S. 1985. The “insulation” layer in the
cocoons of Argiope aurantia (Araneae: Araneidae),
J. Therm. Biol, 10:171-175.

Hickman, V. V. 1970. The biology of Tasmanian
Chloropidae (Diptera) whose larvae feed on spider’s
eggs. J. Entomol Soc. Australia (N. S. W.), 7:8-33.

Holland, W. J, 1968. The Moth Book. Dover Publ,
New York. Horton, C. C. 1983. Predators of two
orb-web spiders (Araneae, Araneidae). J. Arachnol,
11:447.^48.

Howell, F. G. & R. D. Ellender. 1984. Observations
on growth and diet of Argiope aurantia Lucas (Ar-
aneidae) in a successional habitat. J. Arachnol, 12:
29-36.

Kessel, E. L. & B. B. Kessel. 1937. The life history
of Gaurax araneae Coq., an egg predator of the black
widow spider, Latrodectus mactans (Fabr.). Pan-Pa-
cific Entomol, 13:58-60.

Levi, H. 1968. The spider genera Gea and Argiope
in America (Araneae: Araneidae), Bull. Mus. Comp.
Zool, 136:319-352.

Marples, B. J. 1 969. Observations on decorated webs.
Bull. British Arachnol. Soc., 1:13-18.

Peck, O. 1985. The taxonomy of the Nearctic species
of Pediobius (Hymenoptera: Eulophidae), especially
Canadian and Alaskan forms. Canadian Ent., 1 17:
647-704.

Riddle, W. A. 1 980. Cold survival of Argiope auran-
tia spiderlings (Araneae, Araneidae). J. Arachnol,
9:343-345.

Riddle, W. A. & A. L. Markezich. 1981. Thermal
regulation of respiration in the garden spider, Ar-
giope aurantia, during early development and over-
wintering. Comp. Biochem. Physiol, 69(A):759-765.

Robinson, M. H. & B. Robinson. 1970. The stabi-
limentum of the orb web spider, Argiope argent at a:
an improbable defense against predators. Canadian
Entomol, 102:641-655.

Tolbert, W. W. 1976. Population dynamics of the
orb- weaving spiders Argiope trifasciata and Argiope
aurantia (Araneae: Araneidae): density changes as-
sociated with mortality, natality and migration. Ph.D.
Dissertation., Univ. of Tennessee, Knoxville.

T olbert, W. W. 1979. Thermal stress of the orb-weav-
ing spider Argiope aurantia (Araneae). Oikos, 32:
386-392.

Waide, R. B. & J. P. Hailman. 1977. Birds of five
families feeding from spider webs. Wilson Bull, 89:
345-346.

Manuscript received 28 July 1992, revised 8 January
1993.

1993. The Journal of Arachnology 21:55-59

THE EFFECT OF THE COPULATORY PLUG
IN THE FUNNEL- WEB SPIDER, AGELENA LI MB AT A
(ARANEAE: AGELENIDAE)

Toshiya Masumoto: Laboratory of Ecology, Department of Biology, Faculty of
Science, Kyushu University 6-10-1; Hakozaki, Higashi-ku, Fukuoka, 812 Japan.

ABSTRACT. Some females of the funnel-web spider, Agelena limbata multiply mate. After copulation, males
make a visible copulatory plug which covers the female’s genital opening. I assessed the effect of copulatory
plugs on the fertilization success of second males, by conducting double mating experiments, using fertile and
sterilized males in sequence. When females copulated with only one fertile male, more than 90% of their eggs
were fertilized. Some males deposited complete plugs and others incomplete plugs. The relative size of the males
to females and absolute male size affected the completeness of plugs. Complete plugs prevented another male’s
insemination completely, but incomplete plugs allowed insemination by second males. In general, first males
had higher fertilization success than second or later males, and copulatory plugs enhanced the first male’s
advantage.

In many taxa, after mating, the male deposits
a copulatory plug that is thought to prevent in-
tromission by other males. This is known in in-
sects (Drummond 1984; Matsumoto & Suzuki
1992), mammals (Martan & Shepherd 1976),
snakes (Devine 1975), ticks (Oliver 1974) and
spiders (Levi 1959; Jackson 1980; Robinson
1982) .

In spiders, there are several reports on the pres-
ence of an amorphous secretion-like material
blocking the epigynum of just-mated females
functioning to prevent intromission by a second
male. Other mechanisms for preventing fertil-
ization by second males include the breaking off
of the embolus tip in the female’s genitalia (Tr-
giope, Nephila), or the sticking of the cap that
normally cloaks the virgin male’s embolus tip in
some Araneidae (Lopez 1987). Copulatory plugs
are considered an adaptive strategy in relation
to paternity assurance in spiders (Austad 1982).
However, plugs are generally not 100% effective
in preventing further mating (Eberhard 1985).
Jackson (1980) presented evidence that some
secondary mates of females could remove the
plug deposited by the first mate.

The funnel-web spider Agelena limbata is an
annual species with a conduit spermatheca in
which there are two separate genital openings:
one for copulation and one for oviposition. Some
females o^A. limbata are polyandrous, and males
secrete an amorphous material blocking the epi-
gynum of just-mated females.

To understand the evolution of copulatory
plugs, their effects must be assessed. In this pa-
per, the effect of copulatory plugs on fertilization
success in the funnel-web spider, A. limbata was
examined by conducting double fertilization ex-
periments, in which one male was sterilized by
exposure to 7-rays of C06O.

METHODS

The spiders.— The funnel web spider A. lim-
bata is an annual spider commonly distributed
in Japan. Females deposit one or two egg sacs
under their webs in September after the males
disappear, and they protect the sacs for a few
weeks before death. The web of this species con-
sists of a flat sheet with an attached funnel ex-
tending into twigs of the surrounding vegetation.
The sheet has no adhesive properties. After the
final molt, the male spider leaves his web and
starts his search for mates. Courtship behavior
and copulation are performed in the daytime;
the duration of copulation is about thee hours
(193 ±63 min; mean ±S.D.) (Masumoto 1991).

Field observations.— Observation of the mat-
ing behavior of A. limbata was conducted from
July to September, 1988 and 1989 at the Kyushu
University campus, located east of Fukuoka city,
Japan. Webs in the study area, located on trees
less than 2 m above the ground, were marked
and observed daily. In 1988, each web was vis-
ited daily in a predetermined order at 3 h inter-
vals from 0900 to 2100 h. In 1989, each web

55

56

THE JOURNAL OF ARACHNOLOGY

was visited daily at 1 h intervals from 0900 to
1500 h. Unmarked spiders found in the study
area were captured whenever possible. Each spi-
der was brought to the laboratory, anesthetized
with CO2 gas, and individually marked on the
dorsal surface of the abdomen with a color mark-
ing pen. Within the day, the marked spiders were
brought back to the entrance of their original web
after another anesthetization. Spiders cohabiting
with mates were not disturbed so as not to dis-
rupt their mating behaviors. After the mating
season from 1 988 to 1 992, females were collected
to determine the condition of any copulatory
plugs.

Laboratory experiments* -Sub-adult spiders
were collected in June and July in “Aburayama
Observation Park of Nature”, Fukuoka, Japan.
They were reared in plastic boxes (32 cm x 19
cm or 1 9 cm X 11 cm). Day length and tem-
perature were almost similar to field conditions.
Bees were given to females as a food every four
days, a mealworm was supplied to males every
week, and water was sprayed on both sexes every
four days.

After spiders matured, I measured the ceph-
alothorax width of individuals. Mating experi-
ments were conducted in the morning because
in this species mating is diurnal (Masumoto
1991). All experiments were started at between
0800-1000 h and lasted for at least six hours.
Each individual was used only one time a day.
Individual males chosen at random were re-
leased into a box containing a virgin female. Af-
ter copulation, the genitalia of the females were
examined to determine the condition of the cop-
ulatory plug. For some females, new males were
introduced every day until a secondary copula-
tion occurred.

To determine the fertilization success of sec-
ondary mates, double fertilization experiments
were conducted using sequential matings of ster-
ilized and fertile (non-treated) males. Males were
collected in the field while immature and, after
maturation, sterilized by exposure to 7 krad 7-ray
of C06O. The behavior of sterile males did not
seem to differ from that of unsterile males. Ster-
lized males were allowed to mate with virgin
females, and after copulation occurred, the shape
of copulatory plug was recorded. In some cases,
after successful mating with a sterile male, a fer-
tile (untreated) male randomly chosen was al-
lowed to mate with the same female. In these
cases, males were replaced everyday until the

0 1 2 3 4 5

No. of copulations

Figure L— The distribution of the total number of
copulations of marked females through the mating sea-
son in 1989 and 1990.

second copulation occurred. The reverse exper-
iment was also conducted with sequential mat-
ings of fertile (non-treated) and sterilized males.
Females started to deposit their eggs in late Au-
gust. In late September, the number of eggs and
juveniles was counted. Values are represented as
means ± SE.

RESULTS

Field observations.— To assess the number of
males which courted or copulated with a female,
I used data on 4 1 females which were monitored
from virginity to the end of mating season in
each year. The end of mating season means the
date when the last copulation in field was ob-
served in each year (23 August in 1988 and 21
August in 1989). Females were courted an av-
erage of 3.76 ± 2.12 times [4.10 ±0.37 {n = 29)
in 1988, 2.92 ±0.66 {n = 12) in 1989] and cop-
ulated an average of 1.34 ±0.73 times [1.24
±0.13 {n = 29) in 1988, 1.58 ±0.23 in 1989 (n
= 12)]. Of 39 females which copulated, 12 (30.8%)
copulated with an additional male (Fig. 1).

Variation of copulatory plugs. -= During the first
half of copulation, the male used only one of the
palps for insemination. After this, he changed
the position of the female which was laying on
the web, and he used another palp to inseminate
during the last half of copulation. After copula-
tion, all males made a copulatory plug which was
secreted from the palps. Some plugs were clas-
sified as complete; these covered the female gen-
ital opening completely. Others were classified
as incomplete; these covered only a portion of

MASUMOTO-EFFECT OF THE COPULATORY PLUG

57

Table L— Mean cephalothorax width of males and females, mean ratio of male cephalothorax width to female
cephalothorax width, and mean age of males when he first copulated with a virgin female according to condition
of copulatory plugs. Values are means ± SE, * Mann- Whitney U test.

Complete plug

Incomplete plug

z*

P*

n

31

19

%

62.0%

38.0%

Mean cephalothorax width of males (mm)
Mean cephalothorax width of

4,87 ± 0.06

4.67 ± 0.09

2.08

0.038

females (mm)

Mean ratio of male cephalothorax width/

4.79 ± 0.08

4.91 ± 0.09

1.02

0.307

female cephalothorax width

1.02 ± 0.02

0.96 ± 0.02

2.06

0.039

Mean age of males (days)

15.4 ±0.7

14.0 ± 1.1

1.53

0.126

the female genital opening. Among 50 untreated
virgin males, 31 (62.0%) made complete plugs
and 19 (38.0%) of them made incomplete plugs
on the first copulation with virgin females.

To assess the cause of variation of the copu-
latory plugs, I analyzed the relation between the
classification of copulatory plugs and relative
male size to female size (male cephalothorax
width/female cephalothorax width), maximum
cephalothorax width of males and females, and
male age (Table 1). The mean cephalothorax
width of males, and the ratio of male to female
cephalothorax width, was greater in males that
made complete plugs than in males that made
incomplete plugs.

To evaluate the variation of plugs in field, I
collected 20 females who were protecting their
egg sacs at the end of the mating season. Of 20
females collected, 13 (65%) had complete cop-
ulatory plugs in their genitalia and the other 7
(35%) had incomplete plugs.

Sperm usage pattern.— Most eggs of singly-
mated female hatched, and the proportion of eggs
hatched was not different between females with
complete and incomplete plugs. Most sperm were

sterilized by the radiation treatment because no
eggs hatched when females copulated with sterile
males. I could not obtain data about fertile male/
sterile male double copulations. When plugs were
complete, only the first male’s sperm were used
in fertilizing eggs. But when plugs were incom-
plete, the second male’s sperm were used for fer-
tilizing 62.9 % of eggs (Table 2).

From the observation of 1 9 cases of pre-cop-
ulatory behavior with non-virgin females (in 1 0
cases, complete plugs; in 9 cases, incomplete
plugs), I found that before insemination of fe-
males with incomplete plugs, five second males
removed the copulatory plug deposited by pre-
vious males, but that four second males did not
remove the copulatory plug which was too small
to cover the epyginum. Males hooked incom-
plete plugs with the palp and pulled the plug out.
But no male could pull out any complete plug.
After removal of an incomplete copulatory plug,
males started insemination and covered the gen-
ital opening of the female with a new plug after
insemination. In seven females which initially
had incomplete plugs and then copulated with
another male, five of them received complete

Table 2.— Proportion of hatched eggs per female in the experiments with sequential mating of sterilized and
fertile males in the spider Agelena limbata. Sample number is indicated in parentheses. Values are means ±
SE. * Mann-Whitney U test.

Mating regime

Proportion hatched

male

male

Complete plug

Incomplete plug

U*

P*

Fertile

—

94.0 ± 5.3% (14)

90.0 ± 9.1% (5)

37

ns

Sterile

—

0.0 ± 0.0% (3)

0.0 ± 0.0% (3)

Fertile

fertile

88.1 ± 11.9% (4)

93.0 ± 7.1% (4)

8.5

ns

Sterile

fertile

0.0 ± 0.0% (4)

62.9 ± 11.4% (6)

24

<0.01

Fertile

sterile

100.0 ± 0.0% (4)

—

58

THE JOURNAL OF ARACHNOLOGY

plugs in their genitalia after the second mating.
Plugs did not come out by themselves even after
oviposition.

DISCUSSION

From data obtained in field, about 70% of fe-
males copulated with a male only once. So first
males had a mating advantage compared to the
second or later males. But 30.8% of females re-
mated with the next courting male in the field.
The reason why some females copulate more than
once is unknown in^. limbata. Watson (1991b)
hypothesized that in the sierra dome spider, Lin-
yphia litigiosa, the second mating is important
for females as bet-hedging against a cryptic or
unexpressed deleterious character present in the
first male’s genes. Re-mating with another male
may be advantageous for some females in A.
limbata.

There are no externally visible plugs in Liny-
phia litigiosa (Watson 1991a) and in Fwntinella
pyramitela (Austad 1982). In L. litigiosa, Wat-
son (1991a) suggested that internal plugs would
have to be quite subtle and sections of epigyna
have not revealed any plugs, and that fertiliza-
tion success of some secondary mates is high,
although on average first mate sperm priority
exists.

Females of A. limbata have a conduit sper-
matheca in which there are separate openings for
entry and departure of sperm on opposite sites
of the spermatheca. A conduit system has been
thought to encourage a ’first-in/first-out’ bias in
sperm precedence favoring first males (Austad
1984; Watson 1991a). But when second copu-
lations of females with incomplete plugs were
successful, the proportion of eggs fertilized by
the first male was reduced: second mates were
able to fertilize 62.9% of eggs in the double fer-
tilization experiments. Thus in^. limbata, though
the mechanism of sperm competition of this spi-
der is unknown, first male sperm precedence does
not occur despite females having a conduit sper-
matheca; and gross spermathecal morphology is
inadequate to explain sperm priority patterns as
Watson (1991a) suggested.

When first mates do not make a complete cop-
ulatory plug, their fertilization success is re-
duced. Thus in A. limbata, the copulatory plug
is very important in assuring first male sperm
advantage. Incomplete plugs tended to be made
by relatively small males, though the statistic was
marginally significant. Smaller males may be un-

able to fill the genital opening of females with a
plug secretion, but the precise cause of incom-
pleteness of the copulatory plug in A. limbata is
unclear.

ACKNOWLEDGMENTS

I am grateful to M. Murai, F. Singer and par-
ticularly W. G. Eberhard for constructive criti-
cisms on an earlier draft of this manuscript.
Thanks are also due to Y. Ono for giving advice
and encouragement. Members of the Laboratory
of Ecology, Department of Biology, Faculty of
Science, Kyushu University helped in the field
survey and gave valuable advice. The irradiation
was carried out in the Co60 irradiation labora-
tory, Kyushu University.

LITERATURE CITED

Austad, S. N. 1982. First male sperm priority in the
bowl and doily spider, Fwntinella pyramitela. Evo-
lution, 36:777-785.

Austad, S. N. 1984. Evolution of sperm priority pat-
terns in spiders. Pp. 223-249, In Sperm competition
and the evolution of animal mating systems (R. L.
Smith, ed.). Academic Press, London and New York.
Devine, M. C. 1975. Copulatory plug in snakes: en-
forced chastity. Science, 187:844-845.

Drummond III, B. A. 1984. Multiple mating and
sperm competition in the Lepidoptera. Pp. 291-
370, In Sperm competition and the evolution of
animal mating systems (R. L. Smith, ed.). Academic
Press, London and New York.

Eberhard, W. G. 1985. Sexual selection and animal
genitalia. Harvard Univ. Press, Cam-
bridge,Massachusetts.

Jackson, R. R. 1980. The mating strategy of Phidip-
pus johnsoni (Araneae, Salticidae), II: Sperm com-
petition and the function of copulation. J. Arach-
nol., 8:217-240.

Levi, H. W. 1959. The spider genera Achaearanea,
Theridion and Sphyrotinus from Mexico, Central
America and the West Indies (Araneae, Theridi-
idae). Bull. Mus. Comp. ZooL, 121:57-163.

Lopez, A. 1987. Glandular aspects of sexual biology.
Pp. 121-132, In Ecophysiology of spiders (N. Nen-
twig, ed.). Springer- Verlag.

Marten, J. & B. A. Shepherd. 1976. The role of the
copulatory plug in reproduction of the guinea pig.
J. Exp. Zook, 196:79-84.

Masumoto, T. 1991. Male’s visits to females’ webs
and female mating receptivity in the spider, Agelena
limbata (Araneae: Agelenidae). J. EthoL, 9:1-7,
Matsumoto, K. & N. Suzuki. 1992. Effectiveness of
the mating plug in Atrophaneura alcinous (Lepi-
doptera: Papilonidae). Behav. EcoL SociobioL, 30:
157-163.

MASUMOTO- EFFECT OF THE COPULATORY PLUG

59

Oliver, J. H. 1974. Reproduction in ticks. 3. copu-
lation in Dermacenter occidentalis and Haema-
physalis leporispalustris. J. ParasitoL, 60:499”-506.

Robinson, M. H. 1982. Courtship and mating be-
havior in spiders. Ann. Rev. Entomol., 27:l--20.

Watson, P. J. 1991 a. Multiple paternity and first mate
sperm precedence in the sierra dome spider, Liny-
phia litigiosa (Linyphiidae). Anim. Behav., 41:135-
148.

Watson, P. J. 1991b. Multiple paternity as genetic
bet-hedging in female sierra dome spiders, Linyphia

litigiosa (Linyphiidae). Anim. Behav., 41:343-360.

Manuscript received 17 September 1992, revised 11 No-
vember 1992.

1993. The Journal of Arachnology 21:60-63

STING USE IN TWO SPECIES OF
PARABUTHUS SCORPIONS (BUTHIDAE)

Jan Ove Rein: Dept, of Zoology, University of Trondheim, N-7055 Dragvoll,
Norway

ABSTRACT. Scorpions sometimes capture and crush prey with their pedipalps and do not use their sting to
inject venom. Experiments were conducted to test the hypothesis that sting use is selective, resulting in conser-
vation of venom. Sting use in relation to prey size and activity was studied in two African scorpions, Parabuthus
liosoma and P. pallidus. Restrictive use of the sting was observed in both species. Decreased use of the sting
occurred with decreasing size/resistance of the prey. Also, prey were not stung immediately after being seized,
but only after resisting capture. The scorpions did not sting non-resistant prey. These results support the notion
that sting use depends upon the size, morphology and resistance of the prey as determined during initial
interactions with the scorpion.

Scorpions are notorious for their stinging be-
havior and powerful venoms. Sting use plays an
important role in prey capture and defense (Va-
chon 1953; Cloudsley-Thompson 1958; Stahnke
1966). As yet, there have been no controlled and
quantitative studies of sting use, but investiga-
tors have suggested a variety of factors that may
be correlated with sting use. It appears that scor-
pions with large, powerful pedipalps seldom use
the sting, while species with small, slender pedi-
palps readily sting their prey (Stahnke 1 966; Baerg
1961; McCormick & Polls 1990). Casper (1985)
proposed an ontogenetic change in sting use by
Pandinus imperator Koch. Young individuals
stung prey readily, while older and adult indi-
viduals were never observed to employ the sting.
Similar results were reported by Cushing &
Matheme (1980) for Paruroctonus boreus Gi-
rard. Le Berre (1979) noted decreased sting use
with smaller prey in Buthus occitanus Amor., and
similar observations were reported for other spe-
cies (Pocock 1893; Vachon 1953; Cloudsley-
Thompson 1958; Baerg 1961; Bucherl 1971; Po-
lls 1979).

The purpose of this study is to examine sting
use during prey capture by two East African
buthids, Parabuthus liosoma Hemprich & Eh-
renberg and Parabuthus pallidus Pocock. Both
species used their stings selectively, depending
upon the size, morphology and resisting behavior
of the prey. Results are discussed in terms of the
costs and benefits of venom injection during prey
capture.

METHODS

Natural history.-- Parabuthus liosoma and P.
pallidus are found in several countries in East
Africa (Probst 1973). Adults of the former spe-
cies are of medium size for scorpions and have
a yellow to yellowish-red body, except for part
of the cauda and telson which are dark red/brown.
They have small, slender pedipalps and a thick,
powerful cauda. Similar coloration and mor-
phology is present in P. pallidus, but these are
slightly smaller and lack the darkened distal part
of the cauda. There are no previous reports on
the life history or behavior of these species.

Materials. —Individuals of P. liosoma and P.
pallidus were collected in the vicinity of Isiolo,
Kenya in May and June, 1988. The animals were
found in the same semi-arid area under stones
along roadsides, but no more than one scorpion
was ever found beneath a single stone. The sub-
strate consisted of compacted sand with occa-
sional grass and bushes.

The scorpions were taken to Norway, where
1 1 individuals of P. liosoma and 1 2 individuals
of P. pallidus were used in the experiments. The
specimens were of unknown age and ranged in
length (pro- and mesosoma) from 1 8-32 mm (x
= 25.1 mm, P. liosoma) and 13-31 mm (x =
21.3 mm, P. pallidus).

Specimens were kept individually in terraria
(32 X 20 cm), with a substrate of sand and some
stones. The temperature was held at 24-30 °C,
and the light/dark period was 10:14 hr. Water
was provided weekly by misting. Animals were

60

REIN===STING USE IN SCORPIONS

61

(16-18 mm) (24-28 mm) (30« mm)

Figure 1 . -- Sting use against three different prey types
in Parabuthus liosoma. The whole columns represent
total sting use, whereas the dark shaded areas show the
percentage successful sting use (see text for further ex-
planations). “N” denotes the number of trials.

not fed except when tested. Only animals active
on the surface in the dark period were selected
for experiments. This appeared to be a useful
indication of hunger, since they usually respond-
ed rapidly when prey were offered.

For testing, the scorpions were transferred to
an observation terrarium (25 x 25 cm) with a
sand floor. They were given one hr for accli-
mation before prey was introduced. Data on all
activities were collected by direct observations
under low intensity red light that is apparently
not visible to scorpions (Machan 1968). All ob-
servations were made during the fall 1988 and
spring 1989. Results were tested using a sign test
(Lehner 1979).

Experiment 1,— Sting use was compared after
presentation of three different types of prey which
differed in size and morphology. These were small
(10-1 8 mm.) and large (24-32 mm) larvae of Te-
nebrio moiitor Linne and a centipede, Lithobius
forfkatus Verhoeff* (26-35 mm). Insect larvae and
centipedes were seen in the scorpions’ habitat in
Kenya, and thus are probably natural prey for
the two Parabuthus species.

After the acclimation period, a live prey was
introduced to the test scorpion, and if accepted,
observations were continued until ingestion was
started. The scorpions were allowed to complete
ingestion before they were transferred back to
their terrarium. If the prey was not accepted by
a scorpion, the test was discontinued, and the
animal was returned to its terrarium.

Experiment 2. — Sting use against non-resistant

(10-15 mm) (24-28 mm) (2&30 mm)

Figure 2. — Sting use against three different prey types
in Parabuthus paliidus. The whole columns represent
total sting use, whereas the dark shaded areas show the
percentage successful sting use (see text for further ex-
planations). “N” denotes the number of trials.

prey was investigated by introducing freshly killed
Tenebrio larvae (29-35 mm) to the scorpions.
The larvae were presented by moving them with
forceps on the substrate near the scorpion ped-
ipalps.

RESULTS

Prey were subdued in two ways. In 43.3% of
the trials {n = 138), scorpions grasped the prey
with one or both pedipalps and then pulled the
prey to the chelicera and began ingestion without
use of the sting. In the remaining trials, the scor-
pions used the sting to subdue the prey. In some
of the latter trials, scorpions did not succeed in
penetrating the prey integument; these scorpions
either attempted to sting again or stopped sting-
ing and devoured the prey alive. These cases were
recorded as sting use, whereas cases with pene-
tration of the integument were recorded as suc-
cessful sting use.

Sting use in P. liosoma. --In this species, the
sting was used significantly less (P < 0.001)
against small larvae than with the two prey of
larger size (Fig. 1). There were no significant dif-
ferences in sting use against the large larvae and
the centipedes. Attempts were made to sting both
of the large prey types in about 85% of the trials,
and the sting use was successful in 58.8% (larvae)
and 69.6% (centipedes) of the trials.

Sting use in P. Individuals of this

species attempted to sting the small larvae sig-
nificantly less {P < 0.005) than the two large prey
types (Fig. 2). Small prey were stung in 20. 1% of

62

THE JOURNAL OF ARACHNOLOGY

the trials, whereas the use of the sting against
large larvae and centipedes was observed in all
trials. Sting use was successful in 13.8% of the
trials with small larvae, 95.7% with large larvae
and 78.6% with centipedes.

Assessment of prey. —Prey were usually not
stung immediately after being seized by the pedi-
palps. Immediate sting occurred in 14.7% (P.
liosoma) and 26.3% (P. pallidus) of the trials in
which the sting was used. In most trials, the sting
was used only after the prey struggled and re-
sisted capture. In several trials, the scorpions at-
tempted to subdue the prey with the pedipalps
for several minutes before finally using the sting.

Sting use against non-resistant prey.— The
scorpions quickly grasped large, dead Tenebrio
larvae which were moved on the substrate near
the pedipalps. Sting use were never observed in
any of these cases. This is significantly different
from sting use with live prey of the same size (P.
liosoma, P < Q. 0^5, n = 9\P. pallidus, P < 0.001,
n = 13).

DISCUSSION

The results provide evidence that scorpions
restrict use of the sting and thereby conserve ven-
om. This is supported by the observations that
they displayed decreasing sting use with decreas-
ing size/resistance of prey (Figs. 1, 2). In most
trials when the prey were stung, scorpions did
not sting the prey immediately after seizing it (a
period of prey assessment occurred before use of
the sting). Moreover, no scorpion stung non-re-
sistant prey (dead larvae), even though they were
large in size. This also supports the notion that
the scorpion evaluates the struggle and resistance
activity of the prey before stinging it.

The possibility of restrictive sting use was sug-
gested from earlier observations of several scor-
pion species (Pocock 1893; Rosin & Shulov 1963;
Le Berre 1979; Polls 1979; Cushing & Matheme
1980), but experimental evidence was lacking
before the present investigation. Williams (1 987)
suggested that scorpions more commonly eat their
prey alive or crush them by pedipalps than inject
venom. A similar pattern of restrictive venom
use was reported for some other predators. The
ant, Camponotus maculatus, uses the venom
spray differently for large and small prey (Dejan
1988), and some snakes reportedly vary the
quantity of venom used for different prey (Gen-
naro et al. 1961; Allon & Kochva 1974).

Sting use in P. liosoma and P. pallidus prob-

ably depends upon the size, morphology and re-
sistance of the prey. Large prey (large larvae) and
prey with powerful mouthparts (centipedes) were
stung frequently by both Parabuthus species,
whereas small prey (small larvae) and non-resis-
tant prey (dead larvae) were seldom stung. The
size and resistance activity of the prey was eval-
uated by the Parabuthus in an assessment period
shortly after capture.

A restrictive sting use in P. liosoma and P.
pallidus is probably advantageous because the
use of the sting and the following venom renewal
is expensive from an energetic point of view. This
was not examined, but it is a reasonable hy-
pothesis since the venom contains a mixture of
water, salt, proteins and other complex mole-
cules (Simard & Watt 1990).

ACKNOWLEDGMENTS

I wish to express appreciation to my thesis
supervisors, Yngve Espmark and Karl Erik Za-
chariassen. I’m also grateful to Roger Farley,
Dieter Mahsberg, David Sissom, and Gary Polis
for valuable comments and criticism of various
drafts of the manuscript. I thank Inger Andresen
for assistance with the figures.

LITERATURE CITED

Allon, N. & E. Kochva. 1974. The quantities of ven-
om injected into prey of different size by Viper a
palaestinae in a single bite. J. Exp. ZooL, 188:71-
76.

Baerg, W. J. 1961. A survey of the biology of scor-
pions of South Africa. African Wildl., 13:99=-! 06.
Bucherl, W. 1971, Classification, biology, and venom
extraction of scorpions, Pp. 317-347, In Venomous
animals and their venoms. (W. Bucherl & E. Buck-
ley, eds.). Academic Press, New York.

Casper, C. S. 1985. Prey capture and stinging behav-
ior in the emperor scorpion, Pandinus imperator
(Koch) (Scorpiones, Scorpionidae). J. Arachnol., 1 3:
277-283.

Cloudsley-Thompson, J. L. 1958. Spiders, Scorpions,
Centipedes and Mites. Pergamon Press, London.
Cushing, B. S. & A. Matheme. 1980. Stinger utili-
zation and predation in the scorpion Paruroctonus
boreus. Great Basin Nat., 40:193-195.

Dejean, A. 1988. Prey capture by Camponotus ma-
culatus (Formicidae - Formicinae). Biol. Behav., 13:
97-115.

Gennaro, J. F., R, S. Leopold & T. W. Merriam. 1961.
Observations on the actual quantities of venom in-
troduced by several species of crotalid snakes in
their bite. Anatom. Rec., 139:303.

REIN -STING USE IN SCORPIONS

63

Le Berre, M. 1979. Analyse sequentielle du com-
portement alimentaire du scorpion Buthus occitan-
us (Amor.) (Arach. Scorp. Buth.). Biol. Behav., 4:97-
122.

Lehner, P. N. 1979. Handbook of Ethological Meth-
ods. Garland STPM Press, New York.

Machan, L. 1968. Spectral sensitivity of scorpion eyes
as possible roles of shielding pigment effect. J. Exp.
Biol., 49:95--105.

McCormick, S. J. & G. A. Polis. 1990. Prey, pred-
ators, and parasites, Pp. 294-320, In The Biology
of Scorpions. (G. A. Polis, ed.). Stanford University
Press, Palo Alto, California.

Pocock, R. I. 1893. Notes upon the habits of some
living scorpions. Nature, 48:104-107.

Polis, G. A. 1 979. Prey and feeding phenology of the
desert sand scorpion Paruwctonus mesaensis (Scor-
pionidae: Vaejovidae). J. Zool. London, 188:33-346.

Probst, P. J. 1973. A review of the scorpions of East

Africa with special regard to Kenya and Tanzania.
Acta Tropica, 30:312-335.

Rosin, R. & A. Shulov. 1 963. Studies on the scorpion
Nebo hierochonticus. Proc. Soc. London, 140:547-
575.

Simard, J. M. & D. D. Watt. 1990 Venoms and tox-
ins, Pp. 414-444, In The Biology of Scorpions. (G.
A. Polis, ed.). Stanford Univ. Press, Palo Alto, Cal-
ifornia.

Stahnke, H. L. 1966. Some aspects of scorpion be-
havior. Bull. South. California Acad, Sci., 65:65-
80.

Vachon, M. 1953. The biology of scorpions. En-
deavour, 12:80-89.

Williams, S. C. 1987. Scorpion bionomics. Ann. Rev.
EntomoL, 32:275-295.

Manuscript received 1 December 1991, revised 16 July
1992.

1993. The Journal of Arachnology 21:64-68

A NEW SPECIES OE VAEJOVIS (SCORPIONES, VAEJOVIDAE)
FROM WESTERN ARIZONA, WITH SUPPLEMENTAL NOTES ON
THE MALE OF VAEJOVIS SPICATUS HARADON

W. David Sissom: Department of Biology and Geosciences, West Texas State
University, Box 808 WT Station, Canyon, Texas 79016 USA

ABSTRACT. A new species of Vaejovis is described from two localities along the Colorado River in western
Arizona. The species is related to Vaejovis spicatus Haradon, to which it is compared. The first known male
specimen of V. spicatus is briefly described, and comments on its hemispermatophore are provided. Hemi-
spermatophore morphology of V. spicatus suggests a close phylogenetic relationship of these two species with
members of the genus Sermdigitus, although they lack several features considered diagnostic for that genus.

In 1974 a peculiar species of Vaejovis was de-
scribed from the Little San Bernadino Mountains
in southern California. Haradon (1974) named
this species Vaejovis spicatus because it was the
only vaejovid known to possess a distinct, spi-
noid subaculear tubercle. In the early 1980s, Dr.
Oscar Francke brought several interesting spec-
imens to my attention that were collected along
the Colorado River in western Arizona. Like V.
spicatus, these specimens bore the strong, spi-
noid subaculear tubercle. After studying them,
however, it became apparent that they differed
from V. spicatus in some significant features. It
is part of the purpose here to describe the Arizona
specimens as a new species.

More recently, the first known male specimen
of V. spicatus was made available for study. Be-
cause males of both forms were unknown, it seems
appropriate to provide some brief descriptive
notes on its morphology. Although the specimen
was in very poor condition, dissection of the
hemispermatophore revealed some interesting
characters that shed new light on the relation-
ships of these two species to other members of
the family.

Vaejovis mumai, new species
(Figs. 1-7)

Type data.— Adult holotype female from Gold
Road, Black Mountain, Mohave Co., Arizona on
17 May 1969 (M. A. Cazier, et al.). Deposited
in the American Museum of Natural History (0.
F. Francke Collection).

Etymology.— This species is dedicated to Dr.
Martin Muma for his many contributions to
American arachnology.

Distribution.— Known from several localities
in western Arizona.

Dmgnosis.— Vaejovis mumai is most similar
to V. spicatus. Vaejovis mumai and V. spicatus
are the only two vaejovid species possessing a
distinct, spinoid subaculear tooth on the telson
vesicle {Sermdigitus joshuaensis has a conspic-
uous tubercle, but not a spinoid tooth). Vaejovis
mumai may be easily distinguished from V. spi-
catus because the pedipalp chela fixed finger has
only five subrows of denticles along the cutting
margin (in V. spicatus, the fixed finger has six
subrows). The lateral inframedian carinae are
more highly developed in V. mumai, being more
or less complete on both I and II and extending
over the posterior Vi of segment III. The ventro-
lateral and ventral submedian carinae in V. mu-
mai are also stronger and more coarsely dentic-
ulate than in V. spicatus. The carinae of the
pedipalp chelae are somewhat stronger in V. mu-
mai. There are also some distinct morphometric
differences, as V. mumai is a larger species (fe-
males 24.5 mm V5. 16-17.5 mm) and has more
robust pedipalps and metasomal segments. The
following ratios demonstrate the differences in
the latter features (values for the holotype and
paratype females of V. spicatus given in paren-
theses; based on Haradon’s measurements):
Pedipalp femur length/width, 3.16 (3.33-3.45);
pedipalp patella length/width, 2.95 (3.29-3.31);
pedipalp chela length/width, 3.45 (3.50-3.61);
pedipalp chela fixed finger length/carapace length,
0.70 (0.76-0.77); metasoma III length/width, 0.86
(0.96-1.0); and metasoma V length/width, 1.27
(1.65-1.67).

Vaejovis mumai may be easily distinguished

64

SISSOM-NEW VAEJOVIS ¥KOM ARIZONA

65

from V, jonesi Stahnke, another small yellowish
Vaejovis in northern and western Arizona that it
superficially resembles, by possessing the sub-
aculear tooth on the telson, by having metasomal
segments I-III wider than or as wide as long (not
with II-III distinctly longer than wide), and by
having trichobothria ib and it of the chela fixed
finger subbasal (rather than at the extreme base
of the fixed finger). Vaejovis mumai also has only
five subrows on the pedipalp chela fingers,
whereas V. jonesi always has six subrows.

Description. —Adult (female) 24.5 mm in
length. Base color yellow to golden brown, with-
out contrasting dusky markings; metasoma and
pedipalps with orange tinge. Carapace moder-
ately coarsely granular. Tergites more finely
granular. Stemite VII with pair of weak, crenu-
late lateral keels. Pectinal tooth count 1 3 in males,
1 1 in females. Proximal pectinal tooth on each
side ovoid in shape and lacking sensilla.

Metasoma: segments I-III distinctly wider
than long; V 1.27 times longer than wide. Seg-
ments I-IV: Dorsolateral carinae strong, crenu-
late; terminal denticles enlarged, spinoid. Lateral
supramedian carinae strong, crenulate; terminal
denticles on I-III enlarged spinoid, on IV widely
flared. Lateral inframedian carina on I complete,
strong, irregularly crenulate; on II almost com-
plete, weak and granular anteriorly, moderate
and crenulate posteriorly; on III present on pos-
terior one-half, moderate, crenulate; on IV ab-
sent. Ventrolateral carinae moderate to strong,
serratocrenulate; ventral submedian carinae on
I weak, granular; on II-IV moderate, serrato-
crenulate. Setation of dorsolateral carinae 0:1:1:
2; ventral submedian carinae 3:3:3:3. Dorsal and
lateral intercarinal spaces with scattered coarse
granules. Segment V (Fig. 1): Dorsolateral cari-
nae strong, irregularly crenulate; lateromedian
carinae moderate, granulose; ventrolateral and
ventromedian carinae strong, crenulate to ser-
ratocrenulate; all surfaces moderately, coarsely
granular. Telson vesicle slightly granular with
distinct, pointed, subaculear tooth (Fig. 1).

Pedipalps: Trichobothrial pattern (Figs. 2-7)
Type C, orthobothriotaxic (Vachon 1974). Fe-
mur (Fig. 2) tetracarinate, with dorsal surface
lightly granular. Patella (Figs. 3-4) dorsointemal,
internal, and ventrointemal carina strong, cren-
ulate; dorsoextemal and ventroextemal carinae
moderate, unevenly granular. Chela (Figs. 5-7)
with dorsal marginal carina strong, granulose;
dorsointemal carina strong, crenulate; dorsal
secondary, digital, and ventroextemal carinae

moderate, smooth; fixed finger (Fig. 7) with pri-
mary denticle row divided into five subrows,
movable finger with six such subrows; tricho-
bothria ib and it of fixed finger situated between
base of finger and the sixth inner accessory gran-
ule. Ratio of pedipalp chela length/width 3.45;
of fixed finger length/carapace 0.70; of movable
finger length/metasoma V length 1.21.

Measurements of Holotype (in mm to nearest
0.05 mm): Total length, 24.5; carapace length,
3.50; mesosoma length, 8.50; metasoma length,
9.15 (I length/width, 1.35/1.80; II length/width,
1.50/1.80; III length/width, 1.60/1.85; IV length/
width, 2.20/2. 10; V length/width, 2.60/2.05); tel-
son length, 3.35 (vesicle length/width/depth, 2.55/

I. 85/1.30; aculeus length, 0.80); pedipalp length

II. 60 (femur length/width, 3.00/0.95; patella
length/width, 3.25/1.10; chela length/width/
depth, 5.35/1.55/1.65; fixed finger length, 2.45;
movable finger length, 3.15).

Variation. —Only a single adult, the holotype
female, was available for study. The juvenile
specimens (middle instars) differ primarily in
coloration, being very pale yellow, and in having
the cuticle more weakly sclerotized.

Comments.— Several attempts to re-collect this
species by myself and colleagues at “P” Moun-
tain have met with failure. The species is prob-
ably very uncommon and/or exhibits infrequent
surface activity during the year.

Specimens examined.— U.S. A.: Arizona: Mojave Co.:
Gold Road, Black Mountain, 17 May 1969 (M. A.
Cazier, et. al.), 1 holotype female, 2 juvenile paratypes
(AMNH-OFF); Gold Road (under rock), 15 March
1976 (M. A. Cazier, O. F. Francke), 1 juvenile paratype
(AMNH-OFF); “P” Mountain, near Parker, 14 March
1976 (M. A. Cazier, O. F. Francke), 1 juvenile paratype
(AMNH-OFF).

COMMENTS ON THE MALE OF
FTE'/OF/5'*SP/C^rC/5'HARADON, 1974

The original description of Vaejovis spicatus
Haradon was based on five specimens (two of
which were adult females) collected from Berdoo
Canyon in the Little San Bemadino Mountains
of Riverside Co., California. In sorting through
material on loan from the California State Uni-
versity at Long Beach, I found a male specimen
of V. spicatus from Pleasant Valley, Joshua Tree
National Monument, Riverside Co., California,
collected in a pitfall trap on August 27, 1966 by
E. L. Sleeper and S. L. Jenkins. Because the male
of this species is previously unknown and its

66

THE JOURNAL OF ARACHNOLOGY

Figures 1-7.— Morphology of holotype female of Vaejovis mumai, new species: 1, left lateral aspect of meta-
somal segments IV and V and telson; 2, dorsal aspect of pedipalp femur; 3, dorsal aspect of pedipalp patella;
4, external aspect of pedipalp patella; 5, external aspect of pedipalp chela; 6, ventral aspect of pedipalp chela;
7, cutting margin of pedipalp chela fixed finger, showing dentition and placement of trichobothria ib and it.

morphology proved quite interesting, it is im-
portant to add some descriptive notes here.

The male compares to the female as follows:
granulation of the carapace, tergites, and meta-
soma as well as the carination of the pedipalps
and metasoma are similar to that of the female.
A few morphometric differences are as follows:
metasoma V is considerably wider than in fe-

males (V length/width = 1 .44); the pedipalp fe-
mur (Fig. 8) and patella (Fig. 9) are slightly more
slender than in the female (femur length/width
= 3.53 V5. 3.33-3.45; patella length/ width = 3.50
V5. 3.29-3.31); but the pedipalp chela (Fig. 10)
is slightly more robust, with a chela length/width
ratio of 3.36 rather than 3.50-3.65. The pectinal
tooth count of the male is 12-12. The pedipalp

SISSOM--NEW VAEJOVIS ¥KOM ARIZONA

67

Figures 8- 14,— Morphology of male of Vaejovis spicatus Haradon: 8, dorsal aspect of pedipalp femur; 9,
dorsal aspect of pedipalp patella; 10, external aspect of pedipalp chela (note subtle scallop at base of fixed finger);

1 1 , cutting margin of pedipalp chela fixed finger, showing dentition and placement of trichobothria ib and it,

12, dorsal aspect of right hemispermatophore; 13, ental aspect of lamellar flange; 14, ventral aspect of “sperm
plug” of hemispermatophoric capsule (note the smooth margin at the arrow), lam = distal lamina; fl = flange;
tr = trunk; dtr = dorsal trough of distal lamina.

chela fixed finger has a slight basal scallop (Fig.
10), but there is no corresponding lobe on the
movable finger; this leaves a space between the
fingers when they are closed. The fixed finger,
with its six subrows of denticles, is shown in Fig.
11.

Measurements of the specimen are as follows
(in mm, to nearest 0.05 mm): Total length, 1 5.90;
carapace length, 2.20; mesosoma length, 4.95;
metasoma length, 6.65 (I length/width, 0.95/1.10;
II length/width, 1.05/1.10; III length/width, 1.10/

1.15; IV length/width, 1.50/?; V length/width,
2.05/1.40); telson length, 2.10 (vesicle length/
width/depth, 1.50/1.05/0.75; aculeus length,
0.55); pedipalp length, 6.95 (femur length/width,
1,95/0.55; patella length/width, 2.15/0.60; chela
length/ width/depth, 2,85/0.85/0.95; fixed finger
length, 1.60; movable finger length, 2.05).

The hemispermatophore is illustrated in Figs.
12-14. The specimen and its hemispermato-
phores were in very poor condition, so both
hemispermatophores were prepared for study as

68

THE JOURNAL OF ARACHNOLOGY

described by Sissom et al. (1990) in order to
obtain a composite drawing. Once the entire
hemispermatophore was drawn, attempts were
made to dissect the capsular region to discern its
fine structure. These attempts proved futile, as
the capsular structures fragmented. However, it
was still possible to make some important ob-
servations. The hemispermatophore is very slen-
der, with the distal lamina noticeably longer than
the trunk (Fig. 12; the ental margin of the distal
lamina bears a broad flange that terminates some
distance away from the base of the distal lamina
(Fig. 1 2); the flange (Fig. 1 3) is distally bilobed;
and the ental process of the inner lobe of the
capsule does not bear booklets (Fig. 14).

In light of the structure of the hemispermato-
phore, the earlier interpretation of V. spicatus as
a member of the Vaejovis nitidulus group (Sissom
&Francke 1985) now seems inappropriate. Vae-
jovis spicatus and V. mumai seem more properly
allied to Serradigitus (but not included therein)
based on the following evidence. First, the pres-
ence of the flange along the ental margin of the
distal lamina bearing a distally-positioned bi-
lobed termination is shared between V. spicatus
and Serradigitus, as well as with several other
vaejovid groups {Syntropis macrura and species
of the Vaejovis eusthenura, punctipalpi, and in-
trepidus groups; Sissom 1991). The presence of
the flange, the distal position of the bilobed ter-
mination, and the shape of that termination are
all hypothesized to be apomorphic. This con-
dition does not occur in other vaejovids. Second,
the proximal pectinal tooth on each side in the
female (of V. mumai, at least) is ovoid and lacks
peg sensilla, a feature previously thought to occur
only in Serradigitus (Sissom & Stockwell 1991).
And third, although fixed finger trichobothria ib
and it are not positioned at the sixth inner ac-
cessory denticle or beyond (a character uniting
all Serradigitus spp.), they occupy a subbasal po-
sition midway between the extreme base of the
finger and the sixth inner accessory denticle. In
this respect, they diflTer from members of the V.
nitidulus and mexicanus groups, in which the
trichobothria are at the extreme base of the fin-
ger. Lastly, it should be noted that placement of
V. spicatus and V. mumai within the genus Ser-
radigitus does not seem appropriate because the
terminal denticles on the pedipalp chela fingers

in these species are not enlarged and clawlike
and the primary denticle row is not conspicu-
ously serrate. Both of these features are regarded
as diagnostic of Serradigitus.

ACKNOWLEDGMENTS

I am grateful to Dr. Oscar Francke for allowing
me to examine the type specimens of Vaejovis
mumai and to Dr. Wojciech J. Pulawski of the
California Academy of Sciences for allowing me
to examine a paratype specimen of V. spicatus.
Dr. E. L. Sleeper kindly loaned me a large num-
ber of scorpion specimens from collection of the
California State University at Long Beach, one
of which turned out to be the male specimen of
V. spicatus described herein. Marshall Hedin and
T. Yamashita attempted to collect this species at
“P” Mountain at my request, and I am grateful
for their eflbrts. Page charges for this article were
paid by the Department of Biology and Geosci-
ences at West Texas State University. Finally, I
wish to thank Drs. Victor Fet and John T. Hjelle
for their reviews of the manuscript.

LITERATURE CITED

Haradon, R. M. 1 974. Vaejovis spicatus: A new scor-
pion from California (Scorpionida: Vaejovidae). Pan-
Pacific Entomol., 50:23-27.

Sissom, W. D. 1991. Systematic studies on the ni-
tidulus group of the genus Vaejovis, with descrip-
tions of seven new species (Scorpiones, Vaejovidae).
J. ArachnoL, 19:4-28.

Sissom, W. D. & O. F Francke. 1985. Redescriptions
of some poorly known species of the nitidulus group
of the genus Vaejovis (Scorpiones, Vaejovidae). J.
Arachnol. 13:243-266.

Sissom, W.D.,G. A. Polis,&D.D. Watt. 1990. Field
and laboratory methods. In The Biology of Scor-
pions. (G. A. Polis, ed.). Stanford University Press,
Stanford, California.

Sissom, W. D. & S. A. Stockwell. 1991 [1992]. The
genus Serradigitus in Sonora, Mexico, with descrip-
tion of four new species (Scorpiones, Vaejovidae).
Insecta Mundi, 5:197-214.

Vachon, M. 1974. Etude des caracteres utilises pour
classer les families et les genres de scorpions. Bull.
Mus. Nat. Hist. Nat., Paris, Ser. 3, No. 140, ZooL,
104:857-958.

Manuscript received 10 January 1993, revised 2 March
1993.

1993. The Journal of Arachnology 21:69“72

ON THE IDENTITY OF IDEOBISIUM TIBIALE BANKS
(NEOBISIIDAE: PSEUDOSCORPIONES: ARACHNIDA)

Bondar P. M. Curcic: Institute of Zoology, Faculty of Science (Biology), University of
Belgrade, Studentski Trg 16, 11000 Belgrade, Yugoslavia.

ABSTRACT. The holotype of Ideobisium tibiale Banks, from Colorado, USA, is redescribed. This species is
transferred from Microcreagris Balzan to Cryptocreagris Curcic (Neobisiidae). The diagnosis of the genus Cryp-
tocreagris is emended.

The pseudoscorpions originally assigned to the
genus Microcreagris Balzan 1892 and inhabiting
North America north of Mexico have been par-
tially revised by Curcic (1984, 1989). While un-
dertaking a further revision of this genus, it be-
came clear that Ideobisium tibiale Banks, 1909,
was erroneously assigned to the genus Micro-
creagris by subsequent researchers (Hoff 1958;
Harvey 1991). The results of the study of the
unique holotype of M. tibialis are presented here.
In addition, this study should further stimulate
an analysis of the taxonomic rank of all other
North American pseudoscorpions currently as-
signed to ""Microcreagris"'.

METHODS

The holotype specimen was borrowed from
the Museum of Comparative Zoology (MCZ),
Harvard University, Cambridge, Massachusetts,
USA. The specimen, mounted on a slide, was
thoroughly examined. Terminology basically fol-
lows that used by Curcic (1984, 1989).

Family Neobisiidae Chamberlin, 1930
GQnm Cryptocreagris Cnrcit, 1984

Diagnosis (emendations italicized). — Galea
with apical branchlets. Abdominal stemites VI
and VII each with 2 anterior discal setae. Sternite
VIII with 2 median setae only slightly anterior
to other marginal setae (not typical discal setae!).
Male genital area: sternite II with a group of
median and posterior setae, sternite III with a
group of anterior, some intermediate, and a se-
ries of posterior setae. Female genital area: ster-
nite II with a group of small setae on each side
of the midline, sternite III with a row of posterior
setae.

Manducatory process with A or 5 {occasionally
3 or 6) setae. Femur and chelal palm of pedipalps

smooth or with inconspicuous granulations. Tri-
chobothriotaxy: esb distal to et, ist-isb-ib clus-
tered on finger base; it and et located distally on
finger tip; esb nearer to it than to ist\ st slightly
closer to t than to sb {or equidistant from these);
sb slightly closer to b than to st {or equidistant
from these).

Leg IV: tibia, basitarsus, and telotarsus with 1
tactile seta each.

Type Microcreagris laudabilis Hoff.

Subordinate Cryptocreagris laudabilis

(Hoff), C. magna (Banks), and C. tibialis (Banks).

Cryptocreagris tibialis (Banks),
new combination (Figures 1-6)

Ideobisium tibiale Banks, 1909:306
Microcreagris tibialis (Banks): Harvey, 1991:345 (full

synonymy)

Description.— Epistome low and rounded api-
cally, carapace with 4 + 4 + 4 + 6 + 6 = 24
setae. Anterior eyes with ffattened lenses, pos-
teriors spot-like (Fig. 3). Galea with terminal
branchlets (Fig. 5). Flagellum with 8 anteriorly
pinnate blades.

Tergites I-X with 6-»9-9-12-13-?-=?--12--12-l 1
setae. Male genital area: unknown. Female gen-
ital area (Fig. 4): sternite II with a group of 6 or
7 setae on either side of midline, sternite III with
23 posterior setae and 5-1 setae along each stig-
ma. Sternite IV with 1 9 marginal setae and 4-6
setae along each stigma, sternite V with 14 setae.
Stemites VI and VII each with 16 or 17 setae
and 2 anterior discal setae each. Sternite VIII
with 15 setae and 2 setae only slightly anterior
to other marginal setae (these two setae are not
the typical discal setae!).

Pedipalps (Fig. 2): manducatory process with
4 long setae. Fixed chelal finger with asymmet-
rical distal teeth, gradually becoming square-

69

70

THE JOURNAL OF ARACHNOLOGY

Figures \~2>.—Cryptocreagris tibialis, holotype female. 1, pedipalpal chela (trichobothrial pattern); 2, pedipalp
(trichobothria omitted); 3, carapace. Scale in mm.

topped and eventually slightly asymmetrical.
Movable chelal finger with teeth similar in form
and size to those on the fixed finger; only a few
distal teeth asymmetrical. Trichobothriotaxy as
illustrated (Fig. 1).

Leg IV: tibia, basitarsus, and telotarsus with 1
tactile seta each (Fig. 6).

Measurements (mm). “Body length 4.80. Car-

apace 1.41/1.365. Chelicera 0.84/0.41, movable
finger length 0.53, galea 0.13. Pedipalps: coxa
1.04, trochanter 0.88, femur 1.71/0.38, tibia 1.60/
0.43, chela 2.98/0.69, chelal palm 1 .44, fixed fin-
ger 1.54. Leg IV: total length 5.515, coxa 0.78,
trochanter 0.60/0.29, femur 1.495/0.35, tibia
1.47/0.185, basitarsus 0.49/0.14, telotarsus 0.68/
0.14.

Figure A. — Cryptocreagris tibialis, holotype female. Genital area. Scale in mm.

CURCIC- IDENTITY OF IDEOBISIUM TIBI ALE

71

Figures 5, 6. — Cryptocreagris tibialis, holotype female. 5, cheliceral fingers; 6, leg IV. Scales in mm.

Tactile seta position/podomere length ratios:
tibia IV 0.345, basitarsus IV 0.20, telotarsus IV
0.29.

Distribution. “Colorado, USA.

Remarks.— Hoff (1956) found that the type
specimens of C. laudabilis showed “considerable
agreement with the original description” of C.
tibialis, which favored his assignment of these
two species to the same genus. Interestingly, Hoff
(1961) studied three more nymphs (two proto-
nymphs and one deutonymph) from Florissant
and from near Gothic, Colorado, USA, respec-
tively. He found that they resembled both C.
laudabilis and C. tibialis, and he pointed out that
the similarity “may be only in generic charac-
teristics”. Both conclusions of Hoff (1956, 1961)
support the assumption that C. laudabilis and C.
tibialis are congeneric.

The material of C. laudabilis (Hoff) and C.
magna (Banks) has been described elsewhere
(Curcic 1984, 1989).

Specimen examined.— Holotype female (WM
1213.01001), from USA: Colorado: Florissant; 8,000
ft., July- August (year and collector lacking on label).

ACKNOWLEDGMENTS

I am indebted to H. W. Levi (Museum of Com-
parative Zoology, Harvard University, Cam-
bridge) for the loan of the type specimen consid-
ered herein. My gratitude is also due to C. W.
Aitchison-Benell (University of Manitoba, Win-
nipeg), M. S. Harvey (Western Australian Mu-
seum, Perth), V. F. Lee (California Academy of
Sciences, San Francisco), and S. Nelson, Jr. (State
University of New York, College at Oswego, Os-
wego) for constructive comments on the manu-
script; their help is greatly appreciated.

This work was supported by the “Beobanka”-
Belgrade, the Serbian Academy of Sciences and
Arts, and by the Serbian Ministry of Science and
Technology grant 0324.

LITERATURE CITED

Banks, N. 1909. New Pseudoscorpionida. Canadian
Ent., 41:303-307.

Curcic, B. P. M. 1984. A revision of some North
American species of Microcreagris Balzan, 1892
(Arachnida: Pseudoscorpiones: Neobisiidae). Bull.
British Arachnol. Soc., 6:149-166.

72

THE JOURNAL OF ARACHNOLOGY

Curcic, B. P. M. 1 989. Further revision of some North
American false scorpions originally assigned to Mi-
crocreagris Balzan (Pseudoscorpiones, Neobisi-
idae). J. ArachnoL, 17:351-362.

Harvey, M. S. 1991. Catalogue of the Pseudoscor-
pionida. Manchester Univ. Press, Manchester & New
York, vi + 726 p.

Hoff, C. C. 1956. Diplosphyronid pseudoscorpions
from New Mexico. American Mus. Novit, 1780:1-
49.

Hoff, C. C. 1958. List of the pseudoscorpions of North
America north of Mexico. American Mus. Novit.,
1875:1-50.

Hoff, C. C. 1961. Pseudoscorpions from Colorado.
Bull. American Mus. Nat. Hist, 122:409-464.

Manuscript received 1 April 1992, revised 26 October
1992.

1993. The Journal of Arachnology 21:73“78

THE GENUS CHILEOGOVEA
(OPILIONES, CYPHOPHTHALMI, PETALLIDAE)

William A. Shear: Department of Biology, Hampden-Sydney College, Hampden-
Sydney, Virginia 23943 USA

ABSTRACT. The opilionid genus Chileogovea is reviewed and a new species, Chileogovea jocasta, from
Malleco Province, Chile, described. Some supplementary illustrations and new records from mainland Chile of
the type species, C oedipus Roewer, are provided.

The opilionid genus Chileogovea was estab-
lished by Roewer in 1961 for a new species, Chi-
leogovea oedipus, from Chepu, Isla de Chiloe,
Chile. Roewer’s description was sufficient to es-
tablish the validity of the new genus, which he
placed in the Family Sironidae (the only named
family of cyphophthalmids at the time), but
omitted important details, such as the form of
the male genitalia. In 1970, Juberthie and Mu-
noz-Cuevas produced a new description and new
illustrations, providing information on these
characters, and giving a new record from a main-
land locality, Nahuelbuta. They noted the resem-
blance of Chileogovea to the New Zealand genus
Rakaia Forster, and assigned both to the Family
Sironidae Simon. In 1980, in a study of the high-
er classification of cyphophthalmids, I placed the
genus Chileogovea in the new Family Pettalidae,
together with Rakaia and the other southern
hemisphere cyphophthalmids previously in Si-
roninae (Pettalus Thorell, Purcellia Hansen and
Sorensen, Speleosiro Lawrence, Parapurcellia
Rosas Costa, and Neopurcellia Forster; Juberthie
[1988] has named a new genus, Austropurcellia,
which also belongs in this family). The family is
distibuted in South Africa, Sri Lanka, New Zea-
land and Australia, and Chile.

In 1981, 1985-86, and 1992, Norman Plat-
nick, Oscar Francke and Randall Schuh of the
American Museum of Natural History made col-
lecting trips to Chile, as did A. Newton and M.
Thayer in 1982. Among their material were ex-
amples of Chileogovea, including specimens of a
new species, and near topotypes and several new
mainland records of C. oedipus. I thank Dr. Plat-
nick for the opportunity to study these speci-
mens, all of which have been deposited in the
American Museum of Natural History (AMNH).

James Cokendolpher and Emilio Maury provid-
ed helpful reviews of the manuscript.

Specimens were observed, measured, and
drawn using a dissecting microscope. The right
chelicera, pedipalp, first and fourth legs, and pe-
nis were then mounted in glycerine on a micro-
scope slide and examined with a compound mi-
croscope outfitted with Nomarski Interference
Contrast optics, and measured with an ocular
micrometer. All measurements are in millime-
ters; in the description, measurements of ap-
pendage segments are given in order from basal
to distal (beginning with trochanter for pedi-
palps, femora for legs), lengths first, separated
from widths by a diagonal stroke, and L/W ra-
tios, if significant, follow in parentheses.

Family Petallidae Shear
Genus Chileogovea Roewer

Chileogovea Roewer, 1961:99; type species C. oedipus

Roewer. Juberthie & Munoz-Cuevas, 1970:109.

Shear, 1980:25. Cekalovic, 1985:8.

Cekalovic (1985) evidently was unaware of the
Family Pettalidae, and repeated Roewer’s original
assignment of the genus to the Family Sironidae,
Subfamily Stylocellinae, already recognized as
incorrect by Juberthie & Munoz-Cuevas in 1 970.

A redescription of the genus was given in 1970
by Juberthie & Munoz-Cuevas, based entirely
upon the characters of C. oedipus, the only spe-
cies known at that time. The discovery of a sec-
ond species, C. jocasta, requires further emen-
dation of the generic diagnosis as follows.

Coxae 1, 2 free, 3, 4 fused. Eyes absent. Ozop-
hores type 3. Chelicerae (Fig. 3) robust, dorsally
crested. Cheliceral fingers with both large and
small teeth (Figs. 4, 12). Abdominal stemites 8,

73

74

THE JOURNAL OF ARACHNOLOGY

9 free, tergite 9 free. Tarsus 4 entire. Male sec-
ondary sexual modifications: adenostyle lamel-
lar, sharply curved (Figs. 8, 14); stemites 5-8
shallowly depressed in midline, or stemites 7, 8
each with pair of paramedian tubercles (mistak-
enly stated by Shear [1980] to be on 6, 7); anal
operculum with or without prominent median
ridge; tergite 9 evenly rounded posteriorly or
shallowly excavate. Penis (Figs. 9, 10, 15, 16) of
Siro type.

Distribution.— Central Chile, Concepcion
south to Chiloe.

Notes. —I cleared the posterior ends of males
of both species of Chileogovea in trypsin, mount-
ed the cleared parts on microscope slides, and
examined them under high magnification for
gland pores. Sternal gland pores occur in the males
of the genera Huitaca and Ogovea (Family Ogov-
eidae), and Troglosiro (family unknown), while
anal gland pores are found in tergite 9 of male
Sironidae. Because of the modifications of the
sterna I expected to find glands, but none could
be detected in either species, and anal glands
were likewise missing. Thus the function of the
sternal modifications in male Chileogovea re-
mains unclear.

The two species of the genus can be separated
by means of the diagnosis given below under the
description of C.jocasta.

Chileogovea Jocasta, new species
Figs. 1-10

Type data. — Holotype male, paratype female,
seven additional paratype males, and nine ad-
ditional paratype females (AMNH) from Berlese
sample of forest litter and moss, montane forest
zone, 300 m elevation, Monumento Nacional
Contulmo, Malleco Prov., Region IX (de la Ar-
aucafiia), Chile, collected 31 January 1986 by N.
I. Platnick and R. T. Schuh. Additional paratype
female (AMNH) from the same locality, but from
425 m elevation, collected 23 January 1985 by
N. I. Platnick and O. F. Francke; 32 male and
25 female paratypes (AMNH) from litter Berlese,
560 m elevation, Pata de Gallina, Arauco Prov.,
Region VIII (Bio Bio), collected 1 1 February 1992
by N. I. Platnick, P. Golobolf, and M. Ramirez.

Etymology.— Roewer was probably referring
to the well known Greek myth in naming his
species oedipus, though he did not explain his
reasons for doing so. I follow suit by naming this
new species for another figure from the same
myth (the name used as a noun in apposition).

Distribution.— Known only from the type lo-
calities.

Diagnosis.— Distinct in numerous characters
from C. oedipus, the only other known species
of the genus. Chileogovea jocasta males and fe-
males are less than 2 mm long, the males lack
conspicuous secondary sexual modifications of
the posterior stemites, the adenostyles of males
are markedly more slender (Fig. 8), the fourth
tibia (Fig. 7) has a L/W ratio of 1.77, and the
setation of the penis is reduced (Figs. 9, 10); C.
oedipus ranges from 2.5 to 3.28 mm long, the
males have strongly modified posterior stemites,
broad adenostyles (Fig. 14), the fourth tibia (Fig.
1 3) has a L/W ratio of 1 .58, and the penial setae
are more numerous (Figs. 15, 16).

Description.— Total length, 1.7, greatest
width, 1.05, L/W = 1.62. Body (Fig. 1) generally
egg-shaped, widest at posterior part of cephalo-
thorax, not dorsoventrally arched as in C. oed-
ipus. Dorsum shining, with heavily pebbled mi-
crosculpture. Ozophores well removed from
cephalothorax margin, directed straight up-
wards; seen laterally, paler in color than rest of
dorsum. Cephalothoracic sulcus distinct later-
ally, less so near midline; abdominal sulci pro-
nounced. Posterior end of body evenly rounded.
Ventral thoracic complex as in C. oedipus. Ab-
dominal stemites without conspicious modifi-
cations, stemites 6-8 somewhat depressed in
midline. Anal operculum without crest. First
cheliceral segment (Fig. 3) 0.83 long, 0.26 wide,
strong dorsal crest present, heavily pebbled. Sec-
ond cheliceral segment 0.77 long, 0.14 wide,
straight, evenly tapered, fixed finger 0.29 long,
38% length of second cheliceral segment. Chel-
iceral teeth (Fig. 4) irregular, large and small teeth
mixed. Palpal segments (Fig. 5) 0.22, 0.31/0.08,
0.18, 0.20/0.08, 0.25. Legs robust, with heavily
pebbled ornamentation. Leg 1 (Fig. 6) segments
0.48/0. 1 5 (3.2), 0.24/0. 1 6, 0.36/0. 1 6 (2.25), 0. 1 6/
0.14, 0.39/0.19. Leg 4 (Fig. 7) segments 0.41/
0.15 (2.73), 0.22/0.19, 0.32/0.18 (1.77), 0.17/
0.17, 0.37/0.18. Adenostyle (Fig. 8) sharply
curved, with long, acute tip reflexed to nearly
touch dorsal surface of tarsus. Penis in ventral
view (Fig. 9) with four ventral setae closely
grouped on distinct tubercle; in dorsal view (Fig.

1 0) with five apical setae slightly removed ven-
trally from apical margin of dorsal plate, lateral
setae single, well separated from dorsal plate,
four dorsal setae closely grouped, with bulbous
bases; gonopore margins with two long, acute.

SHEAR --OPILIONID GENUS CHILEOGOVEA

75

Figures 1-12.=(1-=10, Chileogovea jocasta, n. sp.), 1, dorsum of male; 2, dorsum of female; 3, chelicera of
male; 4, cheliceral teeth of male; 5, palpus of male; 6, leg 1 of male; 7, leg 4 of male; 8, adenostyle, 9, penis,
ventral view; 10, penis, dorsal view. (11, 12, male Chileogovea oedipus Roewer), 11, dorsum; 12, cheliceral
teeth. Scales = 1.85 mm for Figs. 1, 2, 1 1; 0.5 mm for Figs. 3, 5, 6, 7; 0.25 mm for Fig. 4; 0.13 mm for Figs.
8--10, 12.

76

THE JOURNAL OF ARACHNOLOGY

Figures 13-16. —Male Chileogovea oedipus: 13, leg 4; 14, adenostyle; 15, penis, ventral view; 16, penis, dorsal
view. Scales = 0.5 mm for Fig. 13; 0.13 mm for Figs. 14-16.

curved fingers with lateral lobes at bases. Color
dark blackish brown, dorsally with irregular black
streaks in older specimens; legs bright orange-
brown.

Female: (Fig. 2). Total length, 1.85 mm. Close-
ly resembling male in all nonsexual characters.

Notes.— One additional male whose penis was
dissected had four, rather than five, apical setae.

SHEAR^OPILIONID GENUS CHILEOGOVEA

At all three known localities, this species is syn-
topic with C. oedipus, taken in the same Berlese
sample.

Chileogovea oedipus Roewer
Figs. 11--16

Chileogovea oedipus Roewer, 1961:100 (male holo-
type, male paratype, two female paratypes from Che-
pu, Isla Chiloe, 850 ft. elevation, mixed evergreen
forest; in Senckenberg Museum, Hamburg, not ex-
amined); Juberthie & Munoz-Cuevas, 1970:1 10.

The excellent and detailed description by Jub-
erthie & Munoz-Cuevas needs little supplement.
However, they did not emphasize the distinc-
tiveness of the posterior paramedian sternal tu-
bercles, which, in most populations, on their me-
dian faces are nearly perpendicular to the stemite
surface. This face is bordered by a semicircle of
small, regular tubercles about one-third the size
of the tubercles ornamenting the body. They did
not mention at all a single, very much enlarged
tubercle situated in the midline of the fourth ster-
nite. This tubercle is in the form of an equilateral
triangle with its apex pointing posterior; each
side of the triangle is about five times as long as
an ordinary body tubercle.

A figure of the dorsum (Fig. 1 1) of a specimen
from a new locality (Rio Negro), and of the chel-
iceral teeth (Fig. 12), leg 4 (Fig. 13), and aden-
ostyle (Fig. 14) are presented here for compar-
ative purposes. The illustrations of the penises
of males from two localities given by these au-
thors were reproduced at small size, and I pro-
vide larger figures of the penis of a male from
Rio Negro (Figs. 15, 16).

The important differences between this species
and the foregoing new one are enumerated above.
Probably due to its rather wide distribution (about
750 north-south km), C. oedipus shows some
variation in both size and penial setation. Both
Roewer (1961) and Juberthie & Munoz-Cuevas
(1970) give the length of a male specimen from
Chepu as 2.5 mm and the latter gave the length
of a female from Chepu as 3.0 mm; I measured
a single Chepu male as 2.65 mm long; two fe-
males were 2.75 and 2.8 mm long. Juberthie &
Munoz-Cuevas (1970) had a male from Na-
huelbuta available but did not give its length; a
male from Nahuelbuta examined by me was 3.28
mm long, and males from the Rio Negro region
averaged 2.88 mm long. Juberthie and Munoz-
Cuevas (1970) illustrated the penises of males
from Chepu and from Nahuelbuta; the Chepu

77

male had three ventral and six apical setae, while
the Nahuelbuta male had four ventral and four
apical setae. A male from Rio Negro (Figs. 15,

1 6) shows either five ventral or four (rather than
the usual three) lateral setae on one side, and six
apical setae. The Estero Nonquen population
seems the most divergent. Males differ from those
in other populations in having the sternal lobes
and the crest on the anal plate reduced; the legs
of both sexes are somewhat more slender than
in individuals of the same body length from Rio
Negro. The penis has four ventral and six apical
setae, and the ventral plate has an irregular distal
margin. It is possible this population represents
a third species, but for now I consider it within
the range of variation of oedipus.

The penises illustrated by Juberthie & Munoz-
Cuevas, together with the size differences I ob-
served, led me at first to suspect that the Na-
huelbuta and Estero Nonquen populations were
distinct species, but more careful examination of
these and other specimens, as well as compari-
sons with the new species C. jocasta, caused me
to conclude that the differences were simply vari-
ations in a geographically widespread species.

Specimens examined.— CHILE: Region VIII (Bio Bio)
Amuco Prov., Pata de Gallina, 560 m elevation, litter
in forest, 11 February 1992, N. I. Platnick (NIP), P.
Goloboff, M. Ramirez, 1 1 males, 4 females; Region IX
(de la Araucania), Concepcion Prov., Estero Nonquen,
90 m elevation, litter berlese in modified forest, 16
November 1981, N. 1. Platnick, R. T. Schuh (RTS), 4
males, 5 females; Malleco Prov., Parque Nacional Na-
huelbuta, 1250 m elevation, mossy forest floor litter
{Nothofagus, Auracaria), 1 9 November 1981, NIP, RTS,
male; Monumento Nacional Contulmo, 300 m, wet
forest, 31 January 1986, NIP, RTS, 6 males; Region
X (de los Lagos), Llanquihue Prov., 1 3 km west of Rio
Negro, 20 m elevation, Berlese of litter from edge of
disturbed forest, 24 January 1986, NIP, RTS, 1 0 males,
9 females; 35 km northwest of Rio Negro, 240 m el-
evation, Berlese of litter from edge of disturbed forest,
24 January 1986, NIP, RTS, 9 males, 7 females; Lago
Chapo, 1 1.7 km east of Correntoso (site 657), berlese
of forest leaf and log litter, 320 m elevation, 16-27
December 1982, A. Newton (AN), M. Thayer (MT),
mate; 13.5 km east of Correntoso (site 656), window
trap in Valdivian rainforest, 310m elevation, AN, MT,
male; berlese of forest leaf and log litter, male, female.
Osorno Prov., hills south of Maicolpue, 75 m elevation,
wet disturbed forest, 26 January 1986, NIP, RTS, 2
males, 4 females; 1 0 km east of Bahia Mansa ,15m
elevation, disturbed forest, 30 January 1985, NIP, O.
F. Francke (OFF), male, female; Volcan Osorno, 610
m elevation, mature forest, 12 February 1985, NIP,

78

THE JOURNAL OF ARACHNOLOGY

OFF, male, female; Chincay, 10 km east of Bahia Men-
sa, 50 m elevation, berlese forest leaf and log litter in
secondary Valdivian forest, AN, MT, 3 males, female.
Chiloe Prov., Isla de Chiloe, Chepu, elev. 15 m, wet
forest, 2 February 1985, NIP, OFF, male, 2 females.

LITERATURE CITED

Cekalovic K., T. 1985. Catalogo de los Opiliones de
Chile (Arachnida). Bol. Soc. Biol. Concepcion, 56:

7-29.

Juberthie, C. 1988. Un nouvel opilion cypho-
phthalme aveugle d'Australie: Austropurcellia gen.
no\.,scoparian. sp. Mem. BiospeleoL, 15:133-140.

Juberthie, C. & A, Munoz-Cuevas. 1970. Revision
de Chileogovea oedipus Roewer (Opiliones: Cypho-
phthalmi: Sironinae). Senckenb. biol., 51:109-1 18.

Roewer, C. F. 1961. Opiliones aus Sud-Chile. Senck-
enb. biol., 42:99-105.

Shear, W. A. 1980. A review of the Cyphophthalmi
of the United States and Mexico, with a proposed
reclassification of the suborder (Arachnida, Opi-
liones). American Mus. Nov., 2705:1-34.

Manuscript received 28 July 1 992, revised 10 December
1992.

i993. The Journal of Arachnology 21:79-80

RESEARCH NOTE

ON THE FEMALE OF CRYPTOCELLUS GOODNIGHTI
(ARACHNIDA: RICINULEI)

The New World ricinuleid fauna includes two
genera, Cryptocellus and Pseudoceilus, occurring
in South and North America, respectively, but
showing broad sympatry within Central Amer-
ica. The 1 1 known Central American species of
Cryptocellus, reviewed by Platnick & Shadab
(1981a, b), belong to the centralis group, which
also extends into Colombia. Of the 1 1 Central
American species, four have been known only
from males.

Among a shipment of ricinuleids collected in
Costa Rica by Dr. Allen M. Young and recently
sent to me for study by Dr. Joan P. Jass of the
Milwaukee Public Museum (MPW) were adults
of two species. One male of C. fagei Cooke &
Shadab was taken in a rotten banana stem in
cacao at Finca La Lola, near Siquirres (10®06'N,
83®22'W), Limon, Costa Rica on August 9, 1 984,
and is the first record of that species from Limon
Province. Three other males, belonging to C.
goodnighti Platnick & Shadab, were taken in rot-
ten banana stem slices at Finca La Tigra, near
La Virgen (10°24'N, 84^07' W), Heredia, Costa
Rica from September 26-28, 1979 and on Sep-
tember 3, 1 989. A single female taken at the same
locality on September 15, 1978 appears to be the
first known female of the latter species, and is
described below. I thank Dr. Mohammad U.
Shadab of the American Museum of Natural
History for providing the illustrations.

Cryptocellus goodnighti Platnick & Shadab
Figs. 1, 2

Cryptocellus goodnighti Platnick & Shadab, 1981a: 10.

Diagnosis. —Females can easily be distin-
guished from those of the other known centralis
group species by their elongated, tripartite sper-
mathecae (compare Figs. 1, 2 with the illustra-
tions in Platnick & Shadab 1981a, b).

Female.— Total length, excluding pygidium,
5.36 mm. Carapace 1.97 mm long, 2.14 wide
near middle of coxae III, where widest, dark red,
lateral margins darkest, with small yellow trans-
lucent areas at margins opposite front of coxae
II; surface coated with strong white setae, rela-
tively uniform in length, with relatively few tu-
bercles largely confined to longitudinal median
depression, pair of oblique paramedian depres-
sions occupying about one-fifth of carapace
length, and posterior margin. Cucullus 0.94 mm
long, 1.17 mm wide, dark red medially with
slightly paler margins, bearing long white setae
sparsest proximally, with tubercles largely re-
stricted to distal margin; lateral lobes only very
slightly protuberant. Left chelicera: movable fin-
ger concave posteriorly, not widened transverse-
ly, armed with 1 3 teeth, of which most proximal
three, fifth, and tenth reduced to denticles, distal
three slightly enlarged; fixed finger armed with
five teeth of which most distal is much enlarged,
three most proximal reduced to denticles. Sternal
region with coxae I not meeting tritostemum;
coxae II meeting for almost their entire length,
their suture line almost three times as long as
that of coxae III; coxae IV meeting along their
median surfaces.

Abdomen 3.81 mm long, 1.39 wide near front
of tergite 12, where widest, coloration as in car-
apace except for light orange articular mem-
branes, white setae shorter than on carapace; tu-
bercles restricted to transverse band on tergite 9,
anterolateral depressions of median plates, cor-
responding depressions of stemites 11-13, pos-
teromedian surface of median plate of tergite 1 1 ,
and throughout length of median one-third of
median plate of tergite 12; median plates of ter-
gites 11-13 much wider than long. Pygidium with
notch in posterior dorsal margin of basal seg-
ment, without notch in ventral margin.

79

80

THE JOURNAL OF ARACHNOLOGY

Figures 1, 2.~Cryptocellus goodnighti Platnick and Shadab, female, posterior genital lip and spermathecae:
1 , anterior view; 2, posterior view.

Palp orange, with first trochanter and tibia
lightest; few tubercles on coxae and trochanters,
base of femora with cluster of tubercles at base
on retromargin; coxae each with two thick white
setae posteriorly along inner margin. Leg formula
243 1 . Legs dark reddish brown with tarsi lightest,
coated with thin, long, white setae, with few tu-
bercles concentrated on ventral ridges of tibiae
and dorsal ridges of metatarsi and tarsi. Leg mea-
surements are given in Table 1. Second legs

Table 1.— Leg measurements.

Leg

I

II

III

IV

Palp

Coxa

0.66

1.13

0.96

0.90

0.40

Trochanter I

0.53

0.85

0.62

0.75

0.49

Trochanter II

—

0.57

0.56

0.41

Femur

1.20

1.98

1.39

1.53

1.01

Patella

0.75

1.21

0.90

0.87

—

Tibia

0.97

1.61

0.85

1.00

1.54

Metatarsus

1.09

1.73

1.09

1.18

—

Tarsus

0.49

1.60

0.77

0.90

0.23

Total

5.69

10.11

7.15

7.69

4.08

slightly widened; femur I about twice, femur II
about three times as long as wide. Tarsal claws
large, evenly curved. Posterior genital lip and
spermathecae as in Figs. 1, 2.

Material Examined.— Only specimens men-
tioned above (MPW).

Distribution.— Known only from northeastern
Costa Rica.

LITERATURE CITED

Platnick, N. L & M. U. Shadab. 1981a. On Central
American Cryptocellus (Arachnida, Ricinulei).
American Mus. Novit., 2711:1-13.

Platnick, N. 1. & M. U. Shadab. 1981b. On the Cryp-
tocellus centralis group (Arachnida, Ricinulei). Bull.
American Mus. Nat. Hist., 170:18-22.

Norman I. Platnick: Department of Entomol-
ogy, American Museum of Natural History,
Central Park West at 79th Street, New York,
New York 10024 USA.

Manuscript received 19 January 1993, revised 1 March
1993.

INSTRUCTIONS TO AUTHORS

Manuscripts are preferred in English but may be accepted
in Spanish, French or Portuguese subject to availability of
appropriate reviewers. Authors whose primary language is not
English, may consult the Associate Editor for assistance in
obtaining help with English manuscript preparation. All man-
uscripts should be prepared in general accordance with the
current edition of the Council of Biological Editors (CBE) Style
Manual unless instructed otherwise below. Authors are ad-
vised to consult a recent issue of the Journal of Arachnology
for additional points of style. Send three identical copies of
the typed material together with copies of illustrations to:

Gary L. Miller, Associate Editor
The Journal of Arachnology
Department of Biology
The University of Mississippi
University, Mississippi 38677 USA
Telephone: (601) 232-7495
FAX: (601) 232-5144
Bitnet: BYMILLER@UMSVM
Correspondence relating to manuscripts should also be di-
rected to the Associate Editor.

Manuscripts longer than 1500 words should be prepared as
Feature Articles, shorter papers as Research Notes.

FEATURE ARTICLES

Title page.— Place in the upper left comer the complete name,
address, and telephone number of the author with whom proofs
and correspondence should be exchanged. Include FAX num-
bers and Bitnet addresses if available. Four lines below, type
the title, centered, all in capital letters. Another four lines be-
low, beginning with the left margin, type each author’s name
on a new line followed by the address.

Abstract.— The heading should be placed at the beginning
of the first paragraph set off by a period. In articles written in
English, a second abstract in an acceptable language may be
included pertinent to the nationality of the author(s) or geo-
graphic region(s) emphasized in the article. A second abstract
in English must be included in articles not written in the latter
language.

Text.— Double-space text, tables, legends, etc. throughout.
Three levels of heads are used. The first level (METHODS,
RESULTS, etc.) is typed in capitals and on a separate line.
The second level head begins a paragraph with an indent and
is separated from the text by a period and a dash. The third
level may or may not begin a paragraph but is underscored
and separated from the text by a colon. Use only the underline
method to indicate italics; DO NOT USE WORD PROCES-
SOR ITALICS CODES. Use the metric system unless quoting
text or referencing collection data. All decimal fractions are
indicated by the period regardless of language of the text.

Citation of references in the text.— Include within parenthe-
ses the surnames of the author(s) followed by the date of pub-
lication. A comma separates multiple citations by the same
author(s) and a semicolon separates citations by different au-
thors, e.g., (Smith 1970), (Jones 1988; Smith 1989), (Smith
1986, 1987; Smith & Jones 1989; Jones et al. 1989).

Literature cited section.— Use the following style:
Lombardi, S. J. & D. L. Kaplan. 1990. The amino acid
composition of major ampullate gland silk (dragline) of Ne-
phila clavipes (Araneae, Tetragnathidae). J. Arachnol., 18:
297-306.

Kraffit, B. 1982. The significance and complexity of com-
munication in spiders. Pp. 1 5-66, In Spider Communica-
tions: Mechanisms and Ecological Significance. (P. N. Witt
& J. S. Rovner, eds.). PrincetonUniversity Press, Princeton.
Footnotes.— Footnotes are only permitted on the first printed
page to indicate current address or other information con-
cerning the author(s). All footnotes are placed together on a
separate manuscript page.

Running head.— Author sumame(s) and an abbreviated title
must be typed all in caps and must not exceed 60 characters
and spaces. The running head is placed by itself on a separate
manuscript page.

Taxonomic articles.— Consult a recent taxonomic article in
the Journal of Arachnology for style or contact the Associate
Editor.

Tables.— Each table, with the legend above, should be placed
on a separate manuscript page. Only the appropriate horizontal
lines should be included. Use no footnotes; instead, include
all information in the legend. Make notations in the text mar-
gins to indicate the preferred location of tables in the printed
text.

Illustrations.— Original art work should be kept by the au-
thor(s) until final acceptance for publication. Alt art work must
be mounted on stiffboard ready for reproduction. In line draw-
ings, pay particular attention to width of lines and size of
lettering when reductions are to be made by the printer. Mul-
tiple photos assembled on a single plate should be mounted
with only a minimum of space separating them. In the case of
multiple illustrations mounted together, each illustration must
be numbered sequentially rather than given an alphabetic se-
quence. Written on the back should be the name(s) of author(s)
and an indication of top edge. The author should indicate
whether the figure should be one column or two columns in
width. The overall dimensions should be no more than 1 1
inches (28 cm) x 14 inches (36 cm). Larger drawings present
greater difficulty in shipping and greater risks of damage for
which the JOA assumes no responsibility. The author is en-
couraged to consider PMT (PhotoMechanical Transfer) repro-
ductions for large drawings and to consult an Editor for details
on appropriate reductions. Make notations in the text margins
to indicate the preferred position of illustrations in the printed
text.

Legends for illustrations should be placed together on the
same page(s) and separate from the illustrations. Adhere to the
following styles:

Figures \-A.—A-us x-us, male from Timbuktu; 1, left leg;
2, right chelicera; 3, dorsal aspect of genitalia; 4, ventral aspect
of abdomen.

Figures 27-34.— Right chelicerae of species of A-us from
Timbuktu: 27, 29, 3 1 , 33, dorsal views; 28, 30, 32, 34, prolater-
al views of moveable finger; 27, 28, A-us x-us, holotype male;
29, 30, A-us, w-us, holotype male; 31, 32, A-us z-us, holotype
male; 33, 34, A-us y-us, male. Scales = 1.0 mm.

Assemble manuscript for mailing.— Assemble the separate
sections or pages in the following sequence; title page, abstract,
text, figure legends, footnotes, running head, tables with leg-
ends, figures.

Page charges and reprints.— The current charge per journal
page will be assessed as follows: $50— nonmembers, $45 —
members with grant, institutional, or other support for page
costs; $25 —regular members who do not have support for page
costs; $20— student members who do not author with full
members and who do not have support. Reprints are available
only from the printer at current costs.

Waiver of Page Charges.- The Editor will consider requests
for a waiver of page charges for manuscripts which have been
accepted for publication. Requests should be made in writing
and should be submitted to the Associate Editor along with
the final revision of the manuscript. The author must certify
that neither grant nor institutional funds are available to cover
all of the publication costs. The Society will normally agree to
waive up to a maximum of 80% of the page charges, contingent
on the availability of funds; the author is expected to find
alternative sources for the remainder.

RESEARCH NOTES

Instructions above pertaining to feature articles apply also
to research notes except that abstracts and most headings are
not used and the byline follows the Literature Cited section.

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 21 Feature Articles NUMBER 1

Visual brightness discrimination of the jumping Menemerus bivittatus

(Araneae, Salticidae), Klaus Tiedemann 1

Circadian rhythmicity and other patterns of spontaneous motor activity in
Frontinella pyramitela (Linyphiidae) and Argyrodes trigonum (Theri-
diidae), Robert B. Suter 6

Predation by spiders on ground-released screwworm flies, Cochliomyia ho-
minivorax (Diptera: Calliphoridae) in a mountainous area of southern
Mexico, John B. Welch 23

The natural history of the California turret %yi\dtr Atypoides riversi (Araneae,
Antrodiaetidae): demographics, growth rates, survivorship, and longev-
ity, Leonard S. Vincent 29

Aspectos de la biologia reproductiva de Linothele megatheloides (Araneae:

Dipluridae), Nicolas Paz S 40

Survivability of overwintering Argiope aurantia (Araneidae) egg cases, with
an annotated list of associated arthropods, T. C. Lockley and O. P.
Young 50

The effect of the copulatory plug in the funnel-web spider, Agelena limbata

(Araneae: Agelenidae), Toshiya Masumoto 55

Sting use in two species of Parabuthus scorpions (Buthidae), Jan Ove Rein 60

A new species of Vaejovis (Scorpiones, Vaejovidae) from western Arizona,
with supplemental notes on the male of Vaejovis spicatus Haradon,

W. David Sissom 64

On the identity of Ideobisium tibiale Banks (Neobisiidae: Pseudoscorpiones:

Arachnida), Bozidar P. M. Curdc 69

The genus Chileogovea (Opiliones, Cyphophthalmi, Petallidae), William A.

Shear 73

Research Note

On the female of Cryptocellus goodnighti (Arachnida: Ricinulei), Norman

L Plat nick 79

q>L

The Journal of
ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 21

1993

NUMBER 2

THE JOURNAL OF ARACHNOLOGY

EDITOR: James W. Berry, Butler University

ASSOCIATE EDITOR: Gary L. Miller, The University of Mississippi

EDITORIAL BOARD: A. Cady, Miami (Ohio) Univ. at Middletown; J. E.
Carrel, Univ. Missouri; J. A. Coddington, National Mus. Natural Hist.; J. C.
Cokendolpher, Lubbock, Texas; F. A. Coyle, Western Carolina Univ.; C. D. Don-
dale, Agriculture Canada; W. G. Eberhard, Univ. Costa Rica; M. E. Galiano,
Mus. Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia,
Missouri; C. Griswold, Calif. Acad. Sci.; N. V. Horner, Midwestern State Univ.;
D. T. Jennings, Garland, Maine; V. F. Lee, California Acad. Sci.; H. W. Levi,
Harvard Univ.; E. A. Maury, Mus. Argentino de Ciencias Naturales; N. I. Plat-
nick, American Mus. Natural Hist.; G. A. Polis, Vanderbilt Univ.; S. E. Riechert,
Univ. Tennessee; A. L. Rypstra, Miami Univ., Ohio; M. H. Robinson, U.S.
National Zool. Park; W. A. Shear, Hampden-Sydney Coll.; G. W. Uetz, Univ.
Cincinnati; C. E. Valerio, Univ. Costa Rica.

The Journal of Arachnology (ISSN 0160-8202), a publication devoted to the
study of Arachnida, is published three times each year by The American Arach-
nological Society. Memberships (yearly): Membership is open to all those inter-
ested in Arachnida. Subscriptions to The Journal of Arachnology and American
Arachnology (the newsletter), and annual meeting notices, are included with mem-
bership in the Society. Regular, $30; Students, $20; Institutional, $80 (USA) or
$90 (all other countries). Inquiries should be directed to the Membership Secretary
(see below). Back Issues: Susan E. Riechert, Department of Zoology, Univ. of
Tennessee, Knoxville, TN 37916 USA. Undelivered Issues: Allen Press, Inc.,
1041 New Hampshire Street, P.O. Box 368, Lawrence, Kansas 66044 USA.

THE AMERICAN ARACHNOLOGICAL SOCIETY

PRESIDENT: James E. Carico (1993-1995), Dept, of Biology, Lynchburg, Vir-
ginia, 24501 USA.

PRESIDENT-ELECT: Matthew H. Greenstone (1993-1995), USDA-BCIRL,
Columbia, Missouri 65203 USA.

MEMBERSHIP SECRETARY: Norman I. Platnick (appointed), American
Museum of Natural History, Central Park West at 79th St., New York, New
York 10024 USA.

TREASURER: Gail E. Stratton (1993-1995), Department of Biology, Albion
College, Albion, Michigan 49224 USA.

BUSINESS MANAGER: Robert Suter, Dept, of Biology, Vassar College, Pough-
keepsie, New York 12601 USA.

SECRETARY: Alan Cady (1993-1995), Dept, of Zoology, Miami Univ., Mid-
dleton, Ohio 45042 USA.

ARCHIVIST: Vincent D. Roth, Box 136, Portal, Arizona 85632 USA.
DIRECTORS: Allen R. Brady (1993-1995), Jackie Palmer (1992-1994), Ann
Rypstra (1993-1995).

HONORARY MEMBERS: C. D. Dondale, W. J. Gertsch, H. Homann, H. W.
Levi, A. F. Millidge, T. Yaginuma.

Cover illustration: A male Tetragnatha extensa from Carlisle, Massachusetts. Original color photo
by Joe Warfel of Arlington, Mass. Photograph made with a handheld Olympus OM-1 35mm camera,
macro lens, telescoping extension tube and manual flash.

Publication date: 17 September 1993
THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER.

1993. The Journal of Arachnology 21:81-90

THE GENUS TROGLOSIRO
AND THE NEW FAMILY TROGLOSIRONIDAE
(OPILIONES, CYPHOPHTHALMI)

William A. Shear: Department of Biology, Hampden-Sydney College; Hampden-
Sydney, Virginia 23943 USA

ABSTRACT. The cyphophthalmid genus Troglosiro Juberthie, known only from New Caledonia, is made the
type of a monobasic new family Troglosironidae, the plesiomorphic sister group of [Pettalidae + Sironidae].
Five new species, raveni, juberthei, ninqua, tillierorum and platnicki, are described.

Juberthie (1979) described Troglosiro as a new
genus of cyphophthalmid based on the single spe-
cies Troglosiro aelleni Juberthie, from d’Adio
Cave (also known as Grotte de Ninrin-Reu) on
the island of New Caledonia. Despite its generic
name and the characterization of the species as
“cave-dwelling” in the paper’s title, T. aelleni
has no detectable morphological adaptations for
a troglobitic existance. Juberthie (1979, 1989)
and Shear (1980, 1985) were unable to place the
genus in the classification of cyphophthalmids
but agreed that it was related to the clade Siron-
idae + Pettalidae, and while zoogeographically
allied to the latter, had more characters in com-
mon with the former. In addition, T. aelleni has
at least three autapomorphies: some of the male
abdominal sterna have small, median exocrine
gland orifices, the apical setae of the penis are
greatly enlarged and basally fused, and the mov-
able fingers of the penis are very large, rough-
ened, and have fimbriate outer margins.

Recent collecting for soil animals on New Cal-
edonia by A. and S. Tillier, and by Norman Plat-
nick and Robert Raven, resulted in the discovery
of five new species sharing these apomorphies.
Study of this new material has convinced me that
Troglosiro constitutes the sister-group of Siron-
idae + Pettalidae, and thus should be placed in
its own family, named and diagnosed below.

I am grateful to Drs. Platnick and Tillier for
allowing me to study their material, and to Dr.
B. Hauser, Natural History Museum, Geneva,
Switzerland, for the loan of type material of Tro-
golosiro aelleni. All primary types have been de-
posited in the Musee National d’Histoire Na-
turelle (MNHN), Paris. Secondary types, where
available, have been deposited in the American
Museum of Natural History (AMNH), New York.

Specimens were observed, measured, and
drawn using a dissecting microscope. The right
chelicera, pedipalp, first and fourth legs, and pe-
nis were then mounted in glycerine on a micro-
scope slide and examined with a compound mi-
croscope outfitted with Nomarski Interference
Contrast optics, and measured with an ocular
micrometer. All measurements are in millime-
ters; in the descriptions, measurements of ap-
pendage segments are given in order from basal
to distal (beginning with trochanter for pedi-
palps, femora for legs), lengths first, separated
from widths by a diagonal stroke. Length/Width
ratios, if significant, follow in parentheses.

Family Troglosironidae, new

Diagnosis.— Distinct from all other cypho-
phthalmids in the following combination of
characters. Penis with apical setae enlarged and
fused, movable fingers of penis enlarged and with
dentate/fimbriate lateral margins (Figs. 15, 16,
24-28, 43), and sterna of males with 2-4 median
exocrine gland pores (Fig. 30).

Type gQnm,~Troglosiro Juberthie 1979, by
present designation and monotypy.

Distribution.— New Caledonia.

Remarks.— The new family is named because
the genus it contains cannot be placed in any of
the existing monophyletic families of cypho-
phthalmids, and because an integration of the
characters of its type genus into the cladistic anal-
ysis by Shear (1980) causes it to appear in the
cladogram as the sister group of the two families
Pettalidae and Sironidae, thus indicating at least
a family-level rank for the taxon. The new family
is supported by the autapomorphies given in the
diagnosis.

81

82

THE JOURNAL OF ARACHNOLOGY

Genus Troglosiro Juberthie

Juberthie, 1979:222; type species T. aelleni
Juberthie.

Description.— Coxae 1, 2 free, 3, 4 fused. Eyes
absent. Ozophores type 2. Chelicerae (Figs. 2,
10, 18, 31) robust, basal article with (Fig. 18) or
without (Fig. 10) dorsal crest. Cheliceral fingers
with regular or irregular teeth (Figs. 18, 38). Ab-
dominal stemites 8 and 9, and tergite 9 fused as
corona analis. Tarsus 4 entire. Tarsal claws 1, 3,
4 smooth, 2 toothed (Fig. 34). Male secondary

sexual modifications: adenostyle lamellar, not
curved, acute-triangular, at base of tarsus 4 (Figs.
5, 6, 37, 42). Stemites with 2, 3, or 4 small,
median exocrine gland pores (Fig. 30) variously
located; anteriormost pore often bilaterally
paired; stemites sometimes deeply depressed in
midline. Anal opercula of males unmodified, anal
glands absent, tergite 9 not modified. Penis (Figs.
7, 8, 15, 16, 24-28, 43) distinctive, with four
apical setae fused in pairs and their bases much
thickened, movable fingers enlarged, middle pair
of dorsal setae sometimes reduced or absent.

Key to Species

la. Dorsum with a distinct color pattern of black and brown (Fig.l); males with 4 sternal pores

raveni, n. sp.

lb. Dorsum uniformly colored 2.

2a. Body length about 2.5 mm 3.

2b. Body length about 2.0 mm, usually less 4.

3a. Males with 4 sternal pores; penis (Fig. 16) with 4 ventral setae, their bases contiguous . tillierorum, n. sp.

3b. Males with 2 sternal pores; penis (Fig. 26) with 2 ventral setae, their bases widely separated

aelleni Juberthie.

4a. Body length about 1.75 mm; males with 2 sternal pores (a third pore present in a minority of specimens
from one locale), stemites deeply depressed; penis (Fig. 24) with 1 ventral seta, median dorsal setae

(Fig. 25) as large as other dorsal setae juberthiei, n. sp.

4b. Body length 2-2.15 mm; males with 3 sternal pores, stemites deeply depressed (Fig. 30) or not, penis

with more than 1 ventral seta, median dorsal setae reduced in size or absent 5.

5a. Male stemites deeply depressed (Fig. 30); penis (Fig. 28) with 6 dorsal setae, the median pair reduced

platnkki, n. sp.

5b. Male stemites not depressed; penis (Fig. 43) with 4 dorsal setae, the median pair absent . . . ninqua, n. sp.

Troglosiro raveni, new species
Figs. 1-8

Type data.” Holotype male and paratype fe-
male (MNHN) from Berlese sample of dry forest
litter. Col des Rousettes, 490 m elevation. New
Caledonia, collected 29 May 1987 by Robert Ra-
ven and Norman Platnick.

Etymology.— The name honors one of the col-
lectors, a noted Australian arachnologist.

Distribution.— Known only from the type lo-
cality.

Diagnosis.— Distinct from its congeners in the
color pattern.

Description.— Mafe- Total length 2.03, width
across ozophores 1.13, greatest width (gW) 1.2,
L/gW = 1 .69. Body (Fig. 1) generally egg-shaped,
widest at posterior part of cephalothorax. Dor-
sum shining, with pebbled microsculpture.
Ozophores close to cephalothorax margin, di-
rected laterally. Cephalothoracic sulcus indis-

tinct; abdominal sulci scarcely visible. Posterior
end of body evenly rounded. Abdominal ster-
nites with 4 gland pores in midline; anteriormost
in posterior margin of stemite 2 + 3, appears as
pair of pores at high magnification; following 3
pores single, at anterior margins of stemites 4,
5,6. Pebbled ornamentation absent from sternal
midline, sterna not depressed. First cheliceral
segment (Fig. 2) 0.83 long, 0.19 wide, low dorsal
crest present. Second cheliceral segment 0.7 1 long,
0.14 wide, straight, evenly tapered, fixed finger
0.23 long, 32% length of second cheliceral seg-
ment. Cheliceral teeth regular. Palpal segments
(Fig. 3) 0.21, 0.33/0.08 (4.13), 0.22, 0.26/0.08
(3.25), 0.26. Legs robust, with heavily pebbled
ornamentation. Leg 1 (Fig. 4) segments 0.5 5/0. 19
(2.9), 0.34/0. 1 7, 0.32/0. 19(1 .68), 0. 1 4/0.23, 0.49/
0.18. Leg 4 (Fig. 5) segments 0.45/0.19 (2.37),
0.31/0.19, 0.32/0.21 (1.52), 0.26/0.16,0.38/0.14.
Adenostyle (Fig. 6) slightly curved, acutely tri-

SHEAR --OPILIONID GENUS TROGLOSIRO

83

Figures \-%. — Troglosiro raveni, new species, male: 1, dorsum; 2, chelicera; 3, pedipalp; 4, first leg; 5, fourth
leg; 6, adenostyle; 7, penis, ventral view, tip of ventral plate broken off; 8, penis, dorsal view, tip of ventral plate
broken off. Scale line: 1.5 mm for 1; 0.6 mm for 2--5; 0.3 mm for 6; 0.15 mm for 7, 8.

angular. Penis in ventral view (Fig. 7) with three
ventral setae; in dorsal view (Fig. 8) with three
lateral setae on each side and three pairs of dorsal
setae, median dorsal setae much reduced, lateral
two bladelike. Apical setae broken off in holotype
(and only) male, probably typical for genus. Gon-
opore structures: ventral plate large, with toothed
semicircular margin; movable fingers with very
large basal lobes, fingers with toothed lateral
margins; gonopore lip with fine teeth. Color pat-
tern as illustated (Fig. 1).

Female: Total length, 2.00 mm. Closely re-
sembling male in all nonsexual characters.

Remarks. —Though the characteristic large
apical setae are broken off in the only male, they
were clearly present at one time, and the other
characters of this species argue for its inclusion
in Troglosiro. Color patterns are rare in cypho-
phthalmids; usually the dorsum is evenly colored
black to light yellowish tan, with differences in
surface texture marking segmental limits. Often
the legs are a lighter color than the dorsum, or

84

THE JOURNAL OF ARACHNOLOGY

Figures 9~\lf.~-Troglosiro tilUerorum, new species, male: 9, dorsum; 10, chelicera; 11, cheliceral teeth; 12,
pedipalp; 13, first leg; 14, fourth leg. Scale line: 1.5 mm for 9; 0.6 mm for 10, 12, 13, 14; 0.15 for 11.

have light-colored distal segments. Forster (1 948),
however, described several species of the New
Zealand genus Rakaia Forster with distinctive
color patterns not unlike that of Troglosiro rav-
eni. All the known species of Rakaia lack sternal
glands in the males and have very well-devel-
oped modifications of the anal plate and poste-
rior tergites.

Troglosiro tilUerorum, new species
Figs. 9-16

Type data.— Flolotype male (MNHN) from
Berlese sample from humid forest, Bobeitio (Til-
lier station 16a; 165°01'01"E, 20"57'13"S), 350
m elevation. New Caledonia, collected 17 No-
vember 1988 by A. and S. Tillier.

Etymology.— The name honors the collectors,
diligent students of the New Caledonian fauna.

Distribution.— Known only from the type lo-
cality.

Diagnosis.— Closest in size and appendage
proportions to T. aelleni, but with 4, rather than
2, ventral penial setae, and 4, rather than 2, ster-
nal pores. Distinct from the other species of the
genus in its larger size.

Description.— Total length 2.5, width
across ozophores 1.33, greatest width (gW) 1.33,
L/gW = 1.88. Body (Fig. 9) generally egg-shaped,
widest at posterior part of cephalothorax. Dor-
sum shining, with pebbled microsculpture.
Ozophores close to cephalothorax margin, di-
rected laterally, slightly constricted apically. Ce-
phalothoracic sulcus distinct; abdominal sulci less
so. Posterior end of body evenly rounded. Ab-
dominal stemites with 4 gland pores in midline;
anteriormost in posterior margin of stemite 2 + 3,

SHEAR --OPILIONID GENUS TROGLOSIRO

Figures 15, \6.~Troglosim tillieromm, new species,
Scale line: 0. 1 5 mm.

second near midlength of stemite 4, third near
midlength of stemite 5, fourth in sulcus between
stemites 6 and 7. Pebbled ornamentation absent
from sternal midline, sterna not depressed. First
cheliceral segment (Fig. 10) 1.08 long, 0.21 wide,
dorsal crest absent. Second cheliceral segment
0.99 long, 0.13 wide, straight, evenly tapered,
fixed finger 0.2 1 long, 21% length of second chel-
iceral segment. Cheliceral teeth (Fig. 11) some-
what irregular, perhaps due to wear. Palpal seg-
ments (Fig. 12) 0.26, 0.48/0.07 (6.8), 0.26, 0.32/
0.07 (4.5), 0.30. Legs robust, with heavily peb-
bled ornamentation. Leg 1 (Fig. 13) segments
0.63/0.19 (3.1), 0.37/0.18, 0.40/0.19 (2.1), 0.32/
0.17, 0.48/0.19. Leg 4 (Fig. 14) segments 0.53/
0.19 (2.8), 0.37/0.20, 0.34/0.20 (1.7), 0.31/0.21,
0.41/0.17. Adenostyle not curved, acutely tri-
angular, Penis in ventral view (Fig. 15) with 4
ventral setae; in dorsal view (Fig. 1 6) with 3 lat-
eral setae on each side and 6 dorsal setae, median
dorsal setae much reduced. Apical setae typical
for genus. Gonopore stmctures: ventral plate
large, with toothed semicircular margin; mov-
able fingers with very large, laterally protmding
basal lobes, fingers with toothed lateral margins;
gonopore lip with small, blunt teeth.

85

male: 15, penis, ventral view; 16, penis, dorsal view.
Female: not collected.

Remarks.— This species is nearly identical in
size and appendage measurements to the type
specimen of T. aelleni, but the penis and the
sternal pores are quite different. With only the
single male of T. tillieromm available, and only
two males of T. aelleni known, it is difficult to
assess the range of variation in either population.
However, previous experience in other genera
suggests that differences of this magnitude con-
stitute species distinctions.

Troglosiro aelleni Juberthie

Fig. 26

Troglosiro aelleni Juberthie, 1979:222 (male holotype
and male paratype from Grotte d’Adio (Ninrin-Reu),
near Poya, Mt. Adio, 200 m altitude, collected by
Aellen and Strinati, 2 April 1977; in Museum d’His-
toire naturelle de Geneve, Switzerland, examined).

I examined the holotype slides and specimens
and found Juberthie’s 1979 description entirely
accurate. Juberthie did not illustrate a ventral
view of the penis, supplied here as Fig. 26. There
are 2 ventral setae.

Surprisingly, no additional specimens of this
species turned up in the Tillier and Platnick-

86

THE JOURNAL OF ARACHNOLOGY

Figures 11-25. — Twglosiw juberthiei, new species, male: 17, dorsum; 18, chelicera; 19, cheliceral teeth; 20,
pedipalp; 21, first leg; 22, fourth leg; 23, adenostyle; 24, penis, ventral view; 25, penis, dorsal view. Scale line:
1.5 mm for 17; 0.6 mm for 18, 20-22; 0.3 mm for 23; 0.15 mm for 19, 14, 25.

Raven collections. It is possible that the species
is limited to the cave at the type locality.

Troglosiro juberthiei, new species
Figs. 17-25

Type data.—Holotype male, paratype female
(MNHN) and six additional male and one ad-
ditonal female paratypes (AMNH) from Berlese
sample of montane forest litter, Riviere Bleue,
280 m elevation, collected 21 May 1987, by N.
L Platnick and R. J. Raven.

Etymology. —The name honors Dr. C. Juber-
thie, Laboratoire souterrain du C. N. R. S., Mou-

lis, St. Girons, France, who described the genus
Troglosiro, and who has contributed more than
anyone else to our understanding of cypho-
phthalmid Opiliones.

Distribution.— In addition to the type locality:
NEW CALEDONIA: Plot VI I, Station 250 d.
Riviere Bleue (166®39T6''E, 22°06'13"S), moist
forest Berlese, 4 December 1986, A. and S. Til-
lier, 1 male.

Diagnosis.— Closest in size and appendage
proportions to T. platnicki and T. ninqua, but
with 1 , rather than 3 or 4, ventral penial setae,
and usually with 2, rather than 3, sternal pores.
The median dorsal penial setae are of normal

SHEAR --OPILIONID GENUS TROGLOSIRO

87

Figures 26“"28.--- Penes “26, Troglosiroaelleni Juberthie, ventral view; 27, 28, Twglosiro platnicki, new species:
27, ventral view; 28, dorsal view. Scale line: 0.15 mm.

size, rather than reduced. Distinct from the other
species of the genus in its smaller size.

Description. —Affl/e.' Total length 1.77, width
across ozophores 1 .00, greatest width (gW) 1.10,
L/gW =1.61. Body (Fig. 1 7) generally egg-shaped,
widest at posterior part of cephalothorax. Dor-
sum shining, with pebbled microsculpture.
Ozophores close to cephalothorax margin, di-
rected laterally, slightly constricted apically. Ce-
phalothoracic sulcus rather indistinct; abdomi-
nal sulci even less so. Posterior end of body evenly
rounded. Abdominal stemites with 2 gland pores
in midline; anteriormost in posterior margin of
stemite 2 + 3, second near midlength of stemite
4; in two of seven males from the type series a
third near midlength of stemite 5. Pebbled or-
namentation absent from sternal midline, sterna
4 and 5 depressed. First cheliceral segment (Fig.
1 8) 0.82 long, 0. 1 8 wide, with pronounced dorsal
crest. Second cheliceral segment 0.75 long, 0.12
wide, straight, scarcely tapered, fixed finger 0.24
long, 32% length of second cheliceral segment.
Cheliceral teeth (Fig. 19) irregular. Palpal seg-
ments (Fig. 20) 0.22, 0.36/0.07 (5. 1 4), 0. 1 9, 0.25/

0.06 (4.2), 0.23. Legs robust, with heavily peb-
bled ornamentation. Leg 1 (Fig. 21) segments
0.50/0.16 (3.1), 0.28/0.15, 0.30/0.15 (2.0), 0.24/
0.13, 0.40/0.17. Leg 4 (Fig. 22) segments 0.40/
0.15 (2.7), 0.29/0.17, 0.28/0.17 (1.65), 0.23/0.15,
0,34/0.15. Adenostyle (Fig. 23) curved, acumi-
nate. Penis in ventral view (Fig. 24) with 1 ven-
tral setae, 3 lateral setae on each side; in dorsal
view (Fig. 25) with 6 dorsal setae, median dorsal
setae not reduced. Apical setae somewhat more
gracile than typical for genus. Gonopore struc-
tures: ventral plate large, with toothed semicir-
cular margin; movable fingers with very large,
laterally protruding basal lobes, fingers with
toothed lateral margins; gonopore lip with small,
blunt teeth.

Female: Slightly larger (length of paratype 1.83),
otherwise similar to male in nonsexual charac-
ters.

Troglosiro platnicki, new species
Figs. 27-35

Type data.— Holotype male (MNHN) from
Berlese sample from humid forest. Riviere Bleue

88

THE JOURNAL OF ARACHNOLOGY

Figures 29-35. — Tmglosim piatnicki, new species^ male: 29, dorsum; 30, ventral view of anterior abdominal
segments; 31, chelicera; 32, pedipalp; 33, first leg; 34, claw of second leg; 35, fourth leg. Scale line: 1.5 mm for
29; 0.75 mm for 30; 0.6 mm for 31-33, 35; 0,15 mm for 34.

(Tillier station 250k, plot VI X; 1 66*^39' 16"E,
22°06'13"S), 160 m elevation, New Caledonia,
collected 7 July 1987 by A. Tillier.

Etymology.— The name honors Dr. Norman
I. Platnick, internationally known authority on
arachnids.

Distribution.— In addition to the type locality,
NEW CALEDONIA: from the following Berlese
samples of moist forest litter along Riviere Bleue,
same coordinates and altitude as type collection.
Tillier Sta. 250c, plot VI 0, A. & S. Tillier, 3
November 1986, 2 males, female (MNHN); Sta.

250h, plot VI 0, A. & S. Tillier, 6 April 1987, 4
males, 2 females (MNHN). Riviere Bleue, Berlese
sample from humid forest (Tillier station 25 Id,
plot VII 0; 166°40'01"E, 22®05'59"S), 170 m eh
evation, A. & S. Tillier, 1 1 December 1986, male,
female (MNHN); wet forest along Riviere Bleue,
280 m elevation, N, Platnick & R. Raven, 21
May 1987, male (AMNH); Berlese of rainforest
litter, Mt. Dzumac, N. Platnick & R. Raven, 28
May 1987, male (AMNH).

Diagnosis. —Distinct from others species in
having 4 ventral setae on the penis, with their

SHEAR --OPILIONID GENUS TROGLOSIRO

89

Figures — Twglosiro ninqua, new species, male: 36, dorsum; 37, chelicera; 38, cheliceral teeth; 39,

pedipalp; 40, first leg; 41, fourth leg; 42, adeonstyle; 43, penis, dorsal view. Scale line: 1.5 mm for 36; 0.6 mm
for 37, 39-41; 0.3 mm for 42; 0.15 mm for 38, 43.

bases strongly toothed (Fig. 27); the sterna of
males are deeply depressed, the depression with
pronounced lateral rims (Fig. 30).

Description. Total length 2.0, width
across ozophores 1.08, greatest width (gW) 1.23,
L/gW =1.63. Body (Fig. 29) generally egg-shaped,
widest at posterior part of cephalothorax. Dor-
sum shining, with pebbled microsculpture.
Ozophores close to cephalothorax margin, di-
rected laterally, slightly constricted apically. Ce-
phalothoracic sulcus indistinct; abdominal sulci
less so. Posterior end of body evenly rounded.
Abdominal stemites with 3 gland pores in mid-
lines of stemites 2, 3, and 4 (Fig. 30); these ster-

nites deeply depressed, with few scattered setae,
depression with distinct lateral rims. First chel-
iceral segment (Fig. 31) 0.97 long, 0.19 wide,
dorsal crest very low. Second cheliceral segment
0.92 long, 0.13 wide, straight, scarcely tapered,
fixed finger 0.30 long, 33% length of second chel-
iceral segment; cheliceral teeth regular. Palpal
segments (Fig. 32) 0.23, 0.43/0.075 (6.14), 0.23,
0.33/0.06 (5.5), 0.29. Legs robust, with heavily
pebbled ornamentation. Leg 1 (Fig. 33) segments
0.60/0. 1 5 (4.0), 0.32/0. 1 6, 0.39/0. 1 6 (2.4), 0.24/
0.14, 0.55/0.21. Leg 4 (Fig. 35) segments 0.52/
0.16 (3.3), 0.32/0.19, 0.33/0.20 (1.7), 0.26/0.13,
0.43/0.15. Adenostyle slightly curved, acutely

90

THE JOURNAL OF ARACHNOLOGY

triangular. Penis in ventral view (Fig. 27) with 4
ventral setae, each toothed at base; in dorsal view
(Fig. 28) with 3 lateral setae on each side and 6
dorsal setae, median dorsal setae much reduced.
Apical setae typical for genus, but with more
coarse teeth basally. Gonopore structures: ven-
tral plate large, with toothed semicircular mar-
gin; movable fingers with very large, laterally
protruding basal lobes, fingers with lateral mar-
gins bearing small rounded teeth; gonopore lip
with small, acute teeth.

Females: Somewhat larger (paratype 2. 14 long),
similar to males in nonsexual characters.

Remarks.— This species is nearly identical in
size and appendage measurements to T. ninqua,
but the toothed ventral setae of the penis and the
deeply depressed male sterna of the present spe-
cies distinguish the two.

Troglosiro ninqua, new species
Figs. 36-43

Type data.— Holotype male and female para-
type (MNHN) from Berlese sample from humid
forest, Mt. Ninqua (Tillier station 288,
166°09'03"E, 21°44'24"S), 1000 m elevation. New
Caledonia, collected 28 October 1986 by A. and
S. Tillier.

Etymology.— The name, a noun in apposition,
is after the type locality.

Distribution.— Known only from the type lo-
cality.

Diagnosis.— Closely related to T. platnicki, but
differing in lacking median dorsal setae of the
penis, and having smooth, not toothed, ventral
setae; the sterna of males are only slightly de-
pressed rather than having a deep, rimmed de-
pression as in T. platnicki.

Description.— Total length 2.12, width
across ozophores 1.13, greatest width (gW) 1.3,
L/gW =1.6. Body (Fig. 36) generally egg-shaped,
widest at posterior part of cephalothorax. Dor-
sum shining, with pebbled microsculpture.
Ozophores close to cephalothorax margin, di-
rected laterally, slightly constricted apically. Ce-
phalothoracic and abdominal sulci nearly ob-
solete. Posterior end of body evenly rounded.
Abdominal stemites with 3 gland pores in mid-
lines of stemites 2, 3, and 4; these stemites slight-
ly depressed, lacking usual pebbled microsculp-

ture. First cheliceral segment (Fig. 37) 1.02 long,
0.19 wide, dorsal crest very low. Second cheli-
ceral segment 0.91 long, 0.14 wide, straight,
scarcely tapered, fixed finger 0.23 long, 25% length
of second cheliceral segment; cheliceral teeth
complex but regular, each consisting of a main
blade and two smaller points (Fig. 38). Palpal
segments (Fig. 39) 0.23, 0.46/0.08 (6.13), 0.24,
0.31/0.05 (6.2), 0.33. Legs robust, with heavily
pebbled ornamentation. Leg 1 (Fig. 40) segments
0.66/0.14 (4.71), 0.32/0.15, 0.41/0.16 (2.6), 0.29/
0.14, 0.50/0.19. Leg 4 (Fig. 41) segments 0.50/
0.17 (2.9), 0.26/0.17, 0.38/0.18 (2.1), 0.29/0.14,
0.42/0.15. Adenostyle (Fig. 42) slightly curved,
acutely triangular. Penis in ventral view with 4
ventral setae, basally bladelike, set on raised
sockets; in dorsal view (Fig. 43) with 2 lateral
setae on each side and 4 dorsal setae, median
dorsal setae absent. Apical setae typical for ge-
nus, bases somewhat more elongate, smoother.
Gonopore structures: ventral plate large, with
vaguely toothed semicircular margin; movable
fingers with moderate, laterally protmding basal
lobes, fingers with lateral margins bearing small
rounded teeth; gonopore lip narrow, with small,
acute teeth.

Female: Somewhat larger (paratype 2.13 long),
similar to males in nonsexual characters.

LITERATURE CITED

Forster, R. R. 1948. The Sub-order Cyphophthalmi
Simon in New Zealand. Dom, Mus. Rec. Entomol.,
1:79-119.

J uberthie, C. 1979. Un cyphophthalme nouveau d’une
grotte de Novelle-Caledonie: Troglosiro aelleni n.
gen., n. sp. (Opilion Sironinae). Rev. Suisse Zool.,
86:221-231.

Juberthie, C. 1989. Rakaia daviesi sp. nov. (Opi-
hones, Cyphophthalmi, Pettalidae) from Australia.
Mem. Queensland Mus., 27:499-507.

Shear, W. A. 1980. A review of the Cyphophthalmi
of the United States and Mexico, with a proposed
reclassification of the suborder (Arachnida, Opi-
liones). American Mus. Nov., 2705:1-34,

Shear, W. A. 1985. Marwe coarctata, a remarkable
new cyphophthalmid from a limestone cave in Ke-
nya. American Mus. Nov., 2830:1-6.

Manuscript received 25 January 1993. revised 5 March
1993.

1993. The Journal of Arachnology 21:91-106

THE INFLUENCE OF PREY AVAILABILITY AND HABITAT
ON ACTIVITY PATTERNS AND ABUNDANCE OF
ARGIOPE KEYSERLINGI (ARANEAE: ARANEIDAE)

Richard A. Bradley': School of Biological Sciences, University of Sydney, Sydney
NSW, Australia.

ABSTRACT. I examined habitat relationships and prey abundance to determine which (if either) of these
factors was more important in determining the local density patterns of the Saint Andrew’s Cross spider, Argiope
keyserlingi Karsch. Focusing on the relationship between a predator and its prey distribution presupposes that
prey capture rate is crucial to the biology of the predator. I also studied the influence of prey capture on survival,
reproduction and behavior of A. keyserlingi as a test of this assumption. Reproduction of females was influenced
by food availability under laboratory conditions. Survival was higher among individual females provided with
supplemental food in a field experiment. Adult female A. keyserlingi moved less frequently when they were
provided with supplemental food. Features of vegetation were correlated with patterns of spatial distribution of
this spider. There was a highly significant correlation between spider density on the study plots and the density
of the understory shrubs that were favored as web sites. On a broad scale, seasonal phenology of activity in A.
keyserlingi was positively correlated with potential prey abundance. At the scale of individual study plots, there
is evidence that prey distribution was unpredictable in both time and space and that neither the activity patterns
nor local density of Argiope keyserlingi tracked these fluctuations.

Recent experimental studies of arachnid ecol-
ogy have examined the influence of prey avail-
ability and the presence of competitors on the
distribution and abundance of these predators
(Wise 1979; Greenstone 1978; Schaefer 1978;
Horton & Wise 1983; Janetos 1983; Rypstra
1 983; Riechert& Cady 1983; Spiller 1984, 1986;
Miyashita 1986; Bradley 1989). Of these studies
the experimental work of Spiller (1984) provides
the only direct evidence of exploitation compe-
tition acting on unrestrained spiders in the field.

Riechert & Cady (1983) suggest that interfer-
ence and intraspecific exploitation competition
are more important than interspecific exploita-
tion competition among spiders. For a variety
of arachnids interference competition and/or
cannibalism influences the number and distri-
bution of individuals (Riechert 1974; Turner &
Polis 1979; Riechert & Cady 1983; Wise 1983,
1984; Polis & McCormick 1986a, 1986b; Rub-
enstein 1987). Despite the paucity of evidence
for exploitation competition, several studies have
shown that individual arachnids experience
shortages of food that limit reproduction (Wise
1975, 1979; Gillespie & Caraco 1987; Morse &
Fritz 1982; Fritz & Morse 1985; Suter 1985).

‘Current Address: Ohio State University, 1465 Mt.
Vernon Ave., Marion, Ohio 43302 USA

These apparently contradictory results may be
reconciled if there is no clear density-dependent
relationship between arachnids and their prey
(Riechert & Lockley 1984). This situation has
also been demonstrated in the predatory beetle
Hyphydrus ovatus L. (Juliano & Lawton 1990).

Wise ( 1 984) suggests that food-limitation does
not necessarily imply competition when re-
sources can neither be predicted nor dominated.
This situation is possible because individual spi-
ders may encounter insufficient numbers of prey
even though prey populations are not regulated
by spider density. In contrast, two studies of
agroecosystems indicate that spiders do regulate
prey populations in these relatively simple en-
vironments (Graze & Gigarick 1989, Riechert &
Bishop 1990).

Most arachnids are generalist predators and
because prey populations vary unpredictably in
both space and time, the foraging success of an
individual spider may have little impact on its
neighbors. Spiders which capture insufficient prey
suffer a “relative shortage of food” (Andrewartha
& Birch 1954). Relative shortage occurs when
some individuals do not obtain sufficient food
yet food is available in the environment. This is
often the result of the inability of the predator
to locate food, rather than its absence (Andre-
wartha & Birch 1984). Because polyphagous spi-

91

92

THE JOURNAL OF ARACHNOLOGY

ders probably do not regulate populations of their
prey and live among many other insectivores,
Andrewartha & Birch (1984, p. 49) would refer
to this situation as a case of extrinsic relative
food shortage. Relationships in which the prey
(donor) controls predator (recipient) density but
not the reverse are referred to as donor-con-
trolled systems (Pimm 1982). I will examine this
idea in the context of the omnivorous predator
Argiope keyserlingi.

It has been shown that habitat structure strong-
ly influences distribution and abundance of orb-
web building spiders (Coleboum 1974; Schaefer
1978; Rypstra 1983, 1986) and other arachnids
(Riechert 1977, 1979, 1981; Bradley 1986). There
is some evidence that suitable foraging or retreat
sites may even limit population density and de-
termine the pattern of dispersion of individuals
among ground foraging spiders (Riechert 1976)
and scorpions (Bradley 1986). Riechert & Gil-
lespie (1986) provided a summary table of the
evidence for habitat choice by spiders which in-
dicated that both vegetation structure and prey
were important factors, but that very few studies
compared these factors. Janetos (1 986) suggested
that prey encounter rates have a direct impact
on web-site occupancy because spiders abandon
unproductive sites but also stated that variability
of prey encounter at a particular site had been
little studied.

The question thus arises, are spider abundance
and activity responsive to prey variability, to
habitat characteristics, or to some combination
of both factors? I examined this question in an
empirical study of temporal and spatial relation-
ships between a generalist predator, the orb-
weaving spider Argiope keyserlingi Karsch, and
its habitat and arthropod prey. My study had
two primary goals: 1) to assess the importance
of variation in prey availability on activity pat-
terns of A. keyserlingi and 2) to compare the
relative influence of prey abundance and habitat
features on the pattern of local distribution of A.
keyserlingi individuals. A central assumption of
community ecology has been that the abundance
of food resources is crucial in determining pat-
terns of predator distribution and abundance
(Wiens 1989, p. 16). In this study I tested the
assumption that food availability is important
to A. keyserlingi by assessing the influence of
foraging success on survival and reproduction in
female A. keyserlingi.

Argiope keyserlingi is common in a variety of
habitats along the east coast of Australia from

NE Queensland south to NE Victoria. In Aus-
tralia this spider is commonly but incorrectly
known as A. aetheria (Levi 1983). It builds orb
webs in low vegetation in open habitats, includ-
ing heathland and salt marsh, as well as the un-
derstory of evergreen sclerophyl woodlands and
forests. In the Hawkesbury Sandstone plateau of
New South Wales they seem most abundant in
the understory of dry open forest (Benson &
Fallding 1981). This study focused on one open
forest population of this species. After emergence
from the egg case, second-instar A. keyserlingi
disperse (often by ballooning). Fresh egg cases
collected in the field and kept in the laboratory
hatched 14-25 days after they were laid (x = 19,
SEM = 1.4, « = 9). Juveniles build their first
webs in late summer (February and March). They
over-winter as immatures and emerge in spring
(November). Data from spider censuses on the
study area indicate a brief synchronous activity
period (Fig. 1 A). In addition to these census data,
qualitative observations for the previous (1983/
84) and subsequent (1986/87) summers conform
to the same restricted activity period. Individ-
uals found in other habitats of the Sydney region
are active for much longer periods during the
year. Adult female^, keyserlingi usually die after
laying eggs, and I often found their carcasses
hanging in or lying on the ground below the web
in January and February. A few A. keyserlingi
females survive the winter and become active
again during the following spring. These females
may either represent late-maturing or truly bi-
ennial individuals.

METHODS

General methods. —The study site was located
in Brisbane Water National Park, near the Uni-
versity of Sydney’s Crommelin Biological Re-
search Station at Pearl Beach NSW (33° 33' S,
151° 18' E). The habitat is dominated by Casu-
arina torulosa Ait. (70% of trees), Angophora
costata (Gaertn.) J. Britt, Eucalyptus spp. and
Syncarpia glomulifera (Sm.) Niedenzu. The un-
derstory is relatively open, with many shrubs,
principally Xanthorrhoea resinosa Pers., Dodon-
aea triquetra Wendl., Livistona australis (R. Br.)
Mart., and Lasiopetalum ferrugineum Sm. The
site is on a north-facing hillside with a few ex-
posed rock outcrops and a dense mat of Casu-
arina litter (5-20 cm deep). Sixteen 0.023 ha (1 5
m X 15 m) square sampling plots were estab-
lished and marked with wooden stakes. The plots
were separated by a minimum of 5 m.

BRADLEY --PREY, HABITAT AND ARGIOPE

93

P

UJ

tL

3

<

OC

m

a.

UJ

E 400
E

200 -

Mi

‘ A S O N d' J F ^ A M J ' J A. S 'o'n' dI J ' F 'm‘ a' '

1 984

1985

1986

Figure I.— Activity phenology and weather statistics for Argiope keyserlingi during the 1984/85 and 1985/86
seasons. A. Argiope activity in mean number of spiders per plot {n = 16) for each sampling date. The vertical
bars indicate the 95% confidence limits on this mean. B. Mean monthly temperature at the study area (°C). The
solid dots are the mean high temperatures for the sampling month, and the open dots are the mean low
temperatures for the sampling month. The lines connect the normal mean high temperature and normal mean
low temperature patterns based on the previous 10 years of weather records {n = 3647 sample dates). Stars
indicate monthly averages that differ from the 10-year average (z-test, P < 0.05). C. Monthly rainfall in mm,
the solid bars are means based on the previous 1 0 years of weather records, the open bars are the actual rainfall
totals for the sampling month. Stars indicate rainfall values that differ from the 10-year average (z-test, P <
0.05).

The number, DBH (diameter at breast height)
and identity of all trees and the number of pe-
rennial shrubs were counted for each study plot.
Twelve summary habitat variables (Table 1) were
subjected to principal components analysis. The
principal component scores for each plot along
the first three axes were compared to measures

of spider density using the Pearson product-mo-
ment correlation (r^). Mean elevation was cal-
culated for each plot. Features of the perennial
vegetation changed little during this study and
were measured only once (spring 1986). Three
descriptive variables (top thread length, sticky
orb diameter, and height above ground at orb

94

THE JOURNAL OF ARACHNOLOGY

Table l.—Argiope keyserlingi density and habitat features on the 16 study plots. Variables: Argiope = x no.
spiders/plot, Ang# = no. of Angophom costata and A. floribunda, Euc# = no. of trees in genus Eucalyptus (8
species), Cas# = no. of Casuarina torulosa, Bank# = no. of Banksia serrata, Syn# = no. of Syncarpia glomulifera,
Totlg = total no. of trees, Diam = x diameter of trees (breast height), Totar = cross-sectional area of trees at
breast height (m^), Xanth = no. of Xanthorrhoea resinosa, Macz = no. of Macrozamia communis, Palm = no.
of Livistona australis, Totsm = total no. of small shrubs, Totvg = Totlg + Totsm, Elev = x elevation of plot
(m).

Argiope

numbers
(x spiders/plot)

# 84/85 85/86

Habitat features

Elev

Trees & large shrubs

Small shrubs

Ang-

#

Euc-

#

Cas- Bank-

# #

Syn-

#

Totlg Diam Totar

Xanth Macz

Palm

Tot-

sm

Totvg

1

7.5

3.7

0

2

11

9

0

22

18.2

0.85

7

3

31

41

63

35.1

2

10.8

8.3

0

3

32

0

0

35

19.0

1.61

25

4

30

59

94

35.1

3

4.0

0.3

1

1

24

3

0

30

17.0

1.13

2

0

24

26

56

36.3

4

4.0

5.0

2

2

26

3

0

32

15.6

1.47

6

2

9

17

49

36.3

5

4.3

3.7

1

7

3

5

0

11

21.0

1.37

4

1

6

11

22

37.5

6

2.3

1.3

6

0

5

2

0

21

12.1

0.31

3

0

0

3

24

37.5

7

1.0

0.0

0

8

29

1

2

32

11.0

0.40

3

0

1

4

36

41.5

8

1.5

2.0

0

3

12

2

7

29

16.7

0.95

1

0

0

1

30

37.5

9

1.5

2.0

0

4

24

0

0

27

18.8

1.03

0

0

17

17

44

37.5

10

2.0

0.7

0

4

37

1

5

47

14.0

1.03

0

1

4

5

52

37.8

11

2.3

1.3

0

0

35

0

3

38

13.7

0.68

2

0

2

4

42

38.7

12

5.3

1.7

0

6

41

0

5

52

13.8

1.04

4

1

6

11

63

37.5

13

3.0

1.7

1

2

27

1

9

40

15.3

1.39

0

7

46

53

93

37.2

14

3.3

1.7

2

8

15

3

8

36

14.3

0.81

1

4

9

14

50

36.9

15

1.0

0.3

1

17

14

0

0

32

13.8

0.67

0

0

1

1

27

38.4

16

1.5

2.0

0

4

21

1

0

26

13.2

0.47

0

2

13

15

47

36.3

center) were measured on 1 18 webs of A. key-
serlingi mature females and 57 webs of immature
females (total body length [tbl] < 7 mm). Total
body length [tbl] was measured from the anterior
end of the median ocular area to the tip of the
opisthosoma. Mature male Argiope keyserlingi
often inhabit the webs of females, and may act
as kleptoparasites as they do in other species
(Robinson & Robinson 1978; Suter 1985). Oc-
casionally males are found in small webs alone;
1 5 such webs were measured and attributed to
males. A sample of 26 A. keyserlingi egg cases
was collected from areas adjacent to the main
study area during early February 1985. These
cases were weighed to the nearest 0. 1 mg (Mettler
balance), dissected and the number of eggs count-
ed. A second sample of 1 9 egg cases collected in
late February 1985 was weighed and maintained
in the laboratory until the spiderlings emerged,
and these were counted.

Spider censuses. —Visual censuses were con-
ducted early in the morning on each of 1 4 dates
between August and March 1984/85 and 9 dates
between July and March 1985/86. Each plot was

censused by walking slowly and looking in and
under vegetation to detect Argiope webs. Repeat
censuses were conducted by a second observer
for the first two dates to verify the efficacy of the
method. Each spider found was classified into
one of 3 length categories (0-5 mm tbl, > 5-7
mm tbl, > 7 mm tbl).

Prey abundance.— Potential prey abundance
was assessed using sticky-boards. Brown Mason-
ite© boards 25.5 cm x 30.0 cm were placed on
wood posts with the center of the board 1.2 m
above the ground surface. The size of a sticky-
board is similar to that of an adult Argiope web.
Prey samples were collected once per month for
each month when spiders were active (Septem-
ber—February). The sample for October 1985
was lost. On each sampling date a clear tight-
fitting plastic bag was slipped over the board and
both sides of the bag were coated with Tangle-
foot© insect trapping adhesive. The plastic bag
covered with Tanglefoot refiected the colors of
the surrounding vegetation. Four boards were
used on each of the 1 6 plots, with two oriented
in a N-S direction and two oriented in an E-W

BRADLEY PREY, HABITAT AND ARGIOPE

95

direction. On each sampling date the bags were
left out for 24 h. To collect the bags, a larger
plastic bag was inverted over the sticky-bag and
both were removed. This left a clear plastic coat-
ing over the specimens, and all subsequent iden-
tifications and measurements were made through
the plastic. These traps were used because they
were successful in pre-sampling tests at capturing
examples of the known prey items in the diet of
Argiope keyserlingi (see below). Because sticky
traps do not behave like spiders (Robinson &
Robinson 1973; Rypstra 1982; Castillo & Eber-
hard 1983), I treated these data as an index of
prey abundance rather than a measure of actual
prey availability.

Each arthropod captured on the sticky-boards
was measured to the nearest 1 mm (body length).
Dry-weight biomass was estimated using regres-
sion equations appropriate for each taxon (Rog-
ers et al. 1976, 1977). In cases where no appro-
priate regression equation was available, I
calculated one from specimens captured in the
study area. Any arthropods that were captured
on the sticky traps which were not taken by Ar-
giope keyserlingi when fed to spiders in the lab-
oratory were eliminated from the sample. Most
insects were identified to the ordinal level; large
insects (> 10 mm body length) were identified
to the family level. Three summary variables
were tabulated for each sample of potential prey;
1) total number (NUMB), 2) number of large
prey [> 5 mm body length, NBIGS], and 3) total
biomass (BIOM). Analysis of spatial and tem-
poral patterns of these potential prey variables
were analyzed using a randomization test (Sokal
& Rohlf 1981). A randomization test (repeated-
measures ANOVA) was used because these data
violated assumptions of traditional ANOVA
(normality, heteroscedasticity) even after trans-
formation. The randomization test (T’-ratio used
as test statistic) provided a robust, ANOVA-based
way to examine variation among dates and across
plots. A model that incorporated the repeated
measures (boards on plots, dates) was used in the
randomization-ANOVA. The SAS GLM (SAS
1988) procedure was used to calculate the SSQ
values. A SAS data statement procedure was
written to conduct the random re-assignments.
Actual values were compared to 1000 random-
ized trials for estimation of significance. Para-
metric ANOVA (SAS GLM) was applied to
web-characteristic data because these data met
requisite assumptions. Correlation analyses were
conducted to compare both temporal and spatial

variation in the prey variables to spider census

data. All data were tested for normality using the
SAS univariate procedure. Pearson’s product-
moment correlation coefficient (r^) was used when
data met parametric assumptions. Spearman
Rank correlation {R,) was applied where data
were not normally distributed. Autocorrelation
with a lag of 1 was used to assess temporal vari-
ation in the spatial patterns of prey distribution.
For this test the data for all four sticky boards
were combined to produce one mean value of
each prey variable for each date/plot combina-
tion.

Prey captured by A. keyserlingi in the field were
also identified and measured. The frequency dis-
tribution of insects taken from Argiope webs was
compared to that collected on the sticky-boards
for both size and taxonomic grouping using a
goodness-of-fit test.

Stepwise multiple regression.— I compared the
relative importance of vegetation and prey as
independent variables for their ability to explain
variation in A. keyserlingi density. The mean val-
ues of six habitat variables (number of Xanthor-
rhoea, number of Macrozamia, number of Liv-
istona, number of large shrubs, plot elevation
and the first Principal Component Score for each
plot) and mean values of three prey variables
from the sticky board sampling (NUMB, BIOM,
NBIGS) were compared to the dependent vari-
able A. keyserlingi density across the 1 6 sampling
plots. The analysis was done separately for each
of the two years of this study. While the distri-
bution of individual variates for these variables
was skewed (see above), the mean values used
in this analysis were approximately normally dis-
tributed (Shapiro-Wilk statistic, P > 0.05), and
their variances were homogenous (Bartlett-Box
F and Cochran’s C tests P > 0.05). I used SAS
REG procedure for these analyses, with the for-
ward selection option (SAS 1988).

Laboratory experiment.— I captured 29 female
A. keyserlingi (penultimate instar) near the
Crommelin Biological Research Station on 10
November 1984. These spiders were weighed to
the nearest 0. 1 mg on a Mettler balance and in-
troduced into individual (30 x 30 x 7 cm) clear
perspex (Plexiglass) containers. Two wood dow-
els were fixed vertically in each container with a
piece of cotton thread strung between them about
2 cm from the top of the container. The spiders
readily built orb webs parallel to the long axis of
the containers, usually within hours of installa-
tion. A ball of moist cotton (re-wetted daily) was

96

THE JOURNAL OF ARACHNOLOGY

placed in each container to maintain humidity.
Each container had a 10 x 10 cm door on the
center of one side for feeding. After the spider
had constructed a web, this door could be opened
and potential prey placed in the web. Any prey
that was not consumed was removed after 24 h.
The spiders were randomly divided into two
groups. One group (low food) was fed 1 adult D.
tryoni once every other day (x = 1 2 mg/feeding,
SEM = 0.1 mg, n = 19). The second group (high
food) was fed 4 adult Queensland fruit flies {Da~
cus tryoni (Froggatt)) once every other day (x =
48 mg/feeding). For comparison, the median
biomass of prey captured on one sticky-board is
1 1,6 mg/day (no comparative data for captures
in the natural webs of A. keyserlingi are avail-
able). After a female molted into the final instar,
a freshly-captured adult male A. keyserlingi with
fully expanded palps was introduced to the con-
tainer, Any egg sacs were removed and weighed,
and the number of eggs counted. Females that
died during the experiment were removed and
weighed as soon as they were discovered (usually
within 12 h). At the end of the experiment (25
February 1985) the surviving females were re-
moved and weighed.

Field experiment.— A food manipulation ex-
periment was conducted to investigate the influ-
ence of supplemental prey on the behavior, sur-
vival and reproduction of adult female A.
keyserlingi. On 10 December 1985, 80 adult fe-
male A. keyserlingi were located and their webs
were mapped and marked in a comer with in-
conspicuous paper tags. The spiders were ran-
domly divided into four groups: fed and marked
{n = 30), fed and unmarked {n = 10), unfed and
marked {n = 30), and unfed and unmarked {n =
10). The unmarked spiders in both treatment
groups were included as a control for the marking
procedure.

Spiders in the fed group were supplied with
one meal worm {Tenebrio molitor Linnaeus) lar-
va (x = 0.16 g, SEM = 0.01 g, « = 19) twice per
day for four days. The web was watched until
the spider had captured and wrapped the sup-
plemental prey to confirm that prey did not es-
cape. Spiders that were to be marked were cap-
tured in a plastic vial and anesthetized with CO^
gas. They were then marked with four colored
non-toxic paint dots in a unique combination. I
judged that the small paint dots did not increase
the conspicuousness of these brightly colored spi-
ders. The spider was then released back onto its
web. All spiders were found each day and if they

had moved their new position was mapped and
marked. A team of four observers was used to
search for spiders. Spiders that disappeared were
scored as missing. Any marked spiders that molt-
ed (but remained in the same web) were re-
marked. Freshly molted A. keyserlingi were eas-
ily recognizable. Female A. keyserlingi are qui-
escent at the time of molting and marked exuviae
were found below the freshly molted individuals.
Nevertheless, some individuals may have moved
and molted, and these would have been scored
as missing. Such movement was relatively rare
in Argiope trifasciata Forskal and A. aurantia
Lucas and the rate increased after molting (En-
ders 1975). Although web-invasion was possible
(Riechert & Gillespie 1986; Hofimaster 1986),
many web movements of marked spiders were
observed and no marked spiders were ever re-
located in a web site that was previously occupied
by another marked individual. The experiment
was divided into three periods: pre-treatment pe-
riod (4 d; 10-13 December), treatment period (4
d; 14™ 17 December), post-treatment (5 d; 18-22
December), G-tests of independence (2 x 2;
Model II) were used to evaluate movement/mor-
tality data from this experiment (Sokal & Rohlf
1981). Individuals were checked on seven sub-
sequent dates (4 Jan to 10 March) and any egg
cases that were found in marked webs were col-
lected. These cases were maintained in the lab-
oratory and the number of spiderlings which
emerged from these cases was counted.

RESULTS

Natural reproduction.— I detected significant
variation in reproductive output among individ-
ual female Argiope in the field. This was ex-
pressed by increasing the clutch size rather than
egg size. Egg cases from 26 natural (no food sup-
plementation) female A. keyserlingi had a mean
mass of 0.095 g (SEM = 0.01 g), equivalent to
44% of a female’s mass before laying (x = 0.21
g, SEM = 0.02 g, n = 29). These egg cases con-
tained 4-750 (x = 298, SEM = 46, ^ = 26) eggs.
For a second sample of 1 9 egg-cases, mass was
strongly correlated with the number of juveniles
that emerged (r^ = 0.99, P < 0.001). Thus the
variation in egg case mass is almost completely
explained by variation in numbers of eggs; vari-
ation in egg size is relatively unimportant. Fe-
males laid from 1-4 egg cases; the maximum
reproductive output for any single unmanipu-

BRADLEY-PREY, HABITAT AND ARGIOPE

97

Table 2.— Web characteristics of Argiope keyserlingi. Mean (x), standard error of the mean (SEM), 95%
confidence interval about the mean (conf int.).

Age/sex class

n

Top thread length (mm)
and

X (SEM, conf int.)

Orb diameter (mm)
and

X (SEM, conf int.)

Orb height (mm)
and

X (SEM, conf int.)

Immature female

57

201 (14, 173-229)

118 (8, 101-134)

737 (40, 656-818)

Mature female

118

278 (12, 255-301)

185 (7, 172-199)

809 (34, 741-876)

Mature male

15

129 (29, 66-192)

65 (21, 20-111)

793 (101, 576-1009)

Classes combined

190

243 (8, 225-261)

155 (11, 143-167)

786 (25, 736-836)

lated female was 850 spiderlings (from 3 cases;

207, 258, 385).

Natural webs and prey. —Some web charac-
teristics differed among the age/sex classes of A.
keyserlingi (Table 2). Webs of mature females
were significantly larger than those of immature
females (top thread length ANOVA P < 0.001,
orb diameter ANOVA P < 0.001). Webs of ma-
ture males were significantly smaller than either
class of females (ANOVA P < 0.001). This is
not surprising considering the fact that males of
this species are tiny (mean mass = 0.005 g, SEM
0.001 g, « = 15) and rarely feed. The sticky orbs
of all classes were at similar heights (ANOVA,
ns).

The mean mass of 3 1 natural prey items was
0.04 g (SEM = 0.01 g), ranging from small gnats
and mosquitos (<0.001 g) to a large predatory
fly (Diptera: Asilidae, 0.12 g). The size distri- ^
bution (six categories) of these prey was com- gc
pared to that of potential prey captured on the u.
sticky boards {n = 1783, Fig. 2). The spiders ^
captured significantly more large prey (x^ = 365; o
df = 5; P < 0.001). Large prey (>0.05 g) con- P
stituted 26% of Argiope prey but only 0.9% of O
potentially available insects. Prey were also com- q

pared to taxonomic assignment (by order, 1 0 cat-
egories); significantly fewer flies (all Diptera com-
bined) and more wasps (Hymenoptera combined)
were captured by the Argiope than the sticky
boards (x^ = 254; df= 9;P < 0.001). Flies made
up 19% of A. keyserlingi prey but 81% of poten-
tial prey. Hymenoptera made up 29% of Argiope
prey but only 4% of potential insect prey.

Spider density. —I calculated density for both
1984/85 and 1985/86 on each of the 16 study
plots. This estimate is based on the mean number
of spiders/plot for all censuses when spiders were
active. The mean density in 1984/85 was 150
spiders/ha (SEM = 29, « = 16) and in 1985/86
it was 97 spiders/ha (SEM = 24, « = 16). There
was considerable variation across the 16 study

plots; 1984/85 coefficient of variation (CV) =
76%, 1985/86 CV 94% (Table 1). Density pat-
terns across plots were consistent between years
(r, = +0.81; P < 0.01).

Argiope keyserlingi density was not related to
any of the tree or large shrub variables, although
it was significantly correlated with total numbers
of small shrubs (1984 = +0.74, P < 0.001;

1985 Tp = +0.62, P < 0.01). This relationship
is primarily due to a strong correlation with Xan-
thorrhoea resinosa numbers (1984 r^ = +0.88,
P < 0.001; 1985 r^ = +0.86, P < 0.001). Xan~
thorrhoea plants are a favored web site for Ar-
giope. Principal component analysis generally

SIZE CATEGORY OF PREY

Figure 2. — Relative proportions of prey in six bio-
mass categories. The vertical scale is the percentage of
the sample which belongs in the size category. The six
categories (horizontal scale) are: 1 . 0-0.00 1 g; 2. >0.00 1-
0.005 g; 3. >0.005-0.01 g; 4. >0.01-0.05 g; 5. >0.05-
0. 1 g; 6. > 0. 1 g. The open bars represent the propor-
tions of potential prey in the sticky-board samples {n
= 1783), the solid bars represent the proportions of
prey captured and consumed by Argiope keyserlingi
observed in the field {n = 31).

98

THE JOURNAL OF ARACHNOLOGY

Table 3.— Results of autocorrelation analysis for prey
sampling variables. The values in the body of the table
are the mean correlation for lag = 1 among all dates
(1984/1985 n = 5, 1985/1986 n = 4). None of the
individual correlation or mean correlation values are
statistically significant. Large prey have a total body
length >5 mm.

Number

Year

Numbers

large prey

Biomass

1984/1985

-0.04

+0.002

+0.08

1985/1986

+ 0.07

+0.08

+0.04

failed to clarify the relationship between vege-
tation and spider density. A. keyserlingi density
was correlated with principal component axis 3
of the vegetation analysis (1984 = +0.69, P

< 0.01; 1985 Tp = +0.71, P < 0.01). This axis
was positively weighted on numbers of larger
broadleaf trees (higher canopy) and Macrozamia
density, both factors are indicative of a mesic
microenvironment. This axis, however, explains
only about 1 2% of the variation in the vegetation
data. Neither the first nor second principal com-
ponent axis was significantly correlated to spider
density. Spider density was negatively correlated
with plot elevation (1984 = -0.67, P < 0.01;

1985 Tp = -0.65, P < 0.01). This feature cov-
aried with vegetation characteristics, probably
because the lower plots were nearer a small creek
and supported lush shrub growth.

Relationship to prey.— Temporal patterns:
Appearance of foraging spiders on the study area
was slightly delayed during the spring of 1985/
86, possibly due to an unusually cool fall (Fig.

I B) . Despite this, the date of first appearance and
general phenology of Argiope during the four
summers from 1983/84 through 1986/87 were
all quite similar despite unusual patterns of rain-
fall. There was a drought prior to the summer of
1983/84 and unusually heavy rains during the
spring (November and December) of 1984 (Fig.

IC) .

Seasonal abundance patterns of potential prey,
as revealed by the sticky board samples (NUMB,

NBIGS), exhibited significant variation among
sampling dates (Repeated-measures ANOVA
randomization test P < 0.001). This seasonal
pattern of variation was consistent for corre-
sponding sampling dates during the activity pe-
riod of A keyserlingi between years (NUMB R,
= +0.90, NBIGS R, = +0.80). Although the
pattern of seasonal variation in potential prey
biomass (BIOM) was also significant (Repeated-
measures ANOVA randomization test P <
0.00 1), it was not consistent between years (BIOM
R^ = -0.40, Fig. 3). The pattern of A keyserlingi
seasonal abundance (averaged across plots) was
positively correlated with all three prey variables
for 1984/85 (NUMB R, = +0.90, BIOM R, =
+0.60, NBIGS R3 = +0.50). This comparison
for the 1985/86 season revealed a positive cor-
relation between spider and prey numbers
(NUMB R3 = +0.90, NBIGS R, = +1.0), but
not between spider numbers and prey biomass
(BIOM R3 = -0.10). Overall, the general sea-
sonal phenology of A. keyserlingi was related to
the seasonal pattern of abundance of potential
prey.

Relationship to Spatial patterns: Spa-

tial patterns of distribution of potential prey
numbers (NUMB, NBIGS) across the 16 study
plots exhibited significant variation (Repeated-
measures ANOVA randomization test P < 0.05).
Spatial variation in biomass (BIOM) across plots
was not significant (Repeated-measures ANOVA
randomization test P = 0.08). The spatial pat-
terns were not consistent between years (NUMB
R3 = +0.26, BIOM R3 = -0.26, NBIGS R^ =
+0.16; all ns). Autocorrelation of the three prey
variables using a lag of 1 (comparing a month
with the preceding month) with the 16 plots as
replicates, detected no significant correlations
(Table 3). Thus the spatial patterns of variation
in prey across plots were not even consistent from
month to month (Fig. 3). As concluded above A.
keyserlingi density on the plots was very consis-
tent, and it is therefore not surprising that there
was no relationship between A. keyserlingi on
the plots and any of the potential prey variables
when all dates are combined. A correlation ma-

Figure 3. — Summary of biomass of potential prey captured on sticky boards during the two sampling years
(1984/1985, 1985/1986). Each dot represents the mean biomass captured on the four sticky boards on that plot
for that sampling date (horizontal scale). The vertical scale is biomass in mg plotted on a log scale. The 16
graphs correspond to the 16 field sampling plots. Note that there is no consistent pattern across dates or plots,
and that no plot had consistently high or low potential prey captures.

Biomass captured (mg)

Sampling Date

100

THE JOURNAL OF ARACHNOLOGY

Table 4. —Results of stepwise multiple correlation
analysis of habitat and prey variables against the den-
sity of Argiope keyserlingi for the two study years.

Variable entering model

Partial

r2

Cu-

mu-

lative

Signifi-

cance

P <

1984/1985 season

Xanthorrhoea density

0.77

0.77

0.0001

Plot elevation

0.10

0.87

0.01

Prey biomass

0.05

0.92

0.05

Macrozamia density

0.02

0.94

ns

1985/1986 season

Xanthorrhoea density

0.75

0.75

0.0001

Plot elevation

0.08

0.83

0.05

Total large shrub

density

0.01

0.84

ns

Macrozamia density

0.01

0.85

ns

trix between spider density and potential prey
variables for all date and plot samples (data not
combined for rows or columns, thus all combi-
nations calculated separately) revealed no sig-
nificant correlations.

Stepwise regression analysis. —This analysis
simultaneously compares the relationships be-
tween habitat and potential prey patterns and
their correlation with variation in A. keyserlingi
numbers on the study plots. The analysis re-
vealed that habitat features were far more im-
portant predictors of spider distribution (Table
4). For both years the number of Xanthorrhoea
entered the model first, this factor alone ex-
plained 77% and 75% of the variation in spider
density for 1984/85 and 1985/86. Again for both
years plot elevation was the next factor to enter
explaining an additional 10% and 8% of the vari-
ation in spider numbers. All three of the potential

prey measures combined explained only 6.7% of
the variation in spider density in 1984/85 and
<1% in 1985/86. The clear conclusion from this
analysis is that habitat characteristics were a bet-
ter predictor of spider density than measures of
potential prey on the same plots.

Laboratory experiment. —All laboratory fe-
males mated with the males presented to them.
Females maintained under the high-food treat-
ment increased by an average of 49% of their
initial mass, while those maintained under the
low-food treatment did not significantly change
in mass (Table 5). Of the 1 5 high-food treatment
females, 12 produced egg cases; these females
had a total reproductive output similar to that
measured in the field (Table 5; /-test, ns). Only
7 of 1 4 females maintained under the low-food
treatment produced egg cases. These females had
a reproductive output significantly lower than
the high-food treatment group (Table 5; /-test, P
< 0.05).

Field experiment.— There were no differences
in web-site movement or disappearance of
marked and unmarked females, so these two cat-
egories were combined for subsequent analyses.
There were no differences in the proportion of
females moving between the fed and unfed groups
during the pre-feeding (control) period. Signifi-
cantly fewer females supplied with supplemen-
tary food moved during the feeding and post-
feeding periods (G-test, P < 0.001; Fig. 4). Of
those females that moved, there was no differ-
ence in distance moved between treatment groups
(combined x = 1.3 m, SEM = 0.2 m, « = 65).

There was no significant difference between
survival (as estimated by the disappearance of
individuals) of the fed and unfed groups during
the pre-feeding (control) period. There was a dif-
ference in the proportion missing during the
feeding period: more unfed individuals van-

Table 5. —Reproduction of Argiope keyserlingi in the laboratory and field. The “high food” treatment averaged
48 mg/feeding and the “low food” group averaged 12 mg/feeding. A sample of females observed in the field is
included for comparison. For these females the number of juveniles was estimated (=number eggs counted).

Treatment group

n

X change in
mass (SEM)

X number of
egg cases
per female

Total number
of juveniles
emerged per
female (SEM)

High food

15

+ 0.103 (0.04)

1.2

293 (57)

Low food

14

-0.031 (0.01)

0.6

120 (39)

Field

26

no data

1.4

367 (46)

BRADLEY~PREY, HABITAT AND ARGIOPE

101

P

m

>

o

2

H

O

o

0.40

0.20

0.00

PRE-FEEDING FEEDING POST FEEDING

Figure 4. — Influence of the field food-supplementation experiment on Argiope keyserlingi movement. The
data are partitioned into three periods: 1 . the pre-feeding control period of four days, spiders were marked and
observed but not manipulated; 2. the feeding period of four days where V2 of the individuals (all adult females)
were provided with supplementary food; 3. the post-feeding period of five days where the manipulation ceased
but the spiders were monitored. Data are the totals at the end of each period. The solid bars represent the spiders
that were provided with supplemental food, the open bars represent control spiders. Each treatment group began
with 40 spiders; sample sizes declined between periods because of spider mortality (Fig. 5).

ished, but this difference was not statistically sig-
nificant (G-test, ns; Fig. 5). This difference was
more apparent during the post-feeding period
(Fig, 5) and it was statistically significant (G-test,

P < 0.001).

Of the spiders surviving at the end of the sea-
son, 41% of the fed group produced at least one
egg case whereas only 25% of the unfed group
did so (Table 6), There was no significant differ-
ence between treatment groups in either the
number of spiderlings per egg case or the total
number of spiderlings per female (Table 6). Sev-
eral of the females from each treatment group
captured large prey items (in addition to the sup-
plemental food). One of the no-supplementation
females captured an unusually large prey item
(large Asilid fly). This fly represents the largest
single prey item recorded during this study (0.12
g), and this spider also had the highest repro-
ductive output recorded from a female A. key-
seriingi during this study (850 spiderlings from
three egg cases). If this exceptional female is re-
moved from the analysis, the reproductive out-
put per female among the experimental females
that were provided with supplementary food is
greater than the reproductive output of those
which were not fed (Table 6).

DISCUSSION

Three lines of evidence indicate that relative
foraging success largely determines differences in
survival and reproduction patterns among fe-
male A. keyserlingi. First, individual females kept
in the laboratory showed a direct response to
feeding treatment. The high food treatment fe-
males grew larger and were more fecund than
those females maintained on the low food diet.
Second, twice as many females provided with
supplementary food in the field experiment suc-
ceeded in completing at least one egg case. In
addition, if the single unfed female which hap-
pened to capture a very large prey item is dis-
counted, there would have also been a signifi-
cantly greater reproduction among fed females.
In one sense, the exception proves the rule; this
individual captured the largest prey item ob-
served in the study and exhibited the record high-
est reproductive output. Foraging success is un-
predictable but crucial to female A. keyserlingi.
Another conclusion from the field experiment is
that mere survival is not sufficient to insure suc-
cessful production of eggs or juveniles. Only 25%
of the surviving females that were exposed to
natural prey abundance actually produced egg

102

THE JOURNAL OF ARACHNOLOGY

O

2

£

<

m

Z

o

H

O

Pm

O

pe^

Pm

1.00

0.75

0.50

0.25

0.00

PRE-FEEDING FEEDING POST FEEDING

Figure 5. --Influence of the field food-supplementation experiment on Argiope keyserlingi survival. The data
are partitioned into three periods: 1. the pre-feeding control period of four days, spiders were marked and
observed but not manipulated; 2. the feeding period of 4 days where Vi of the individuals (all adult females)
were provided with supplementary food; 3. the post-feeding period of five days where the manipulation ceased
but the spiders were monitored. Data are the totals at the end of each period. The solid bars represent the spiders
that were provided with supplemental food, the open bars represent control spiders. Each treatment group began
with 40 spiders. Because of the intensive search effort by five observers and the relatively small area of suitable
habitat, disappearance is probably a good measure of survivorship. Individuals classified as “absent” were
assumed to be dead.

cases. Third, the mortality rate among Argiope
keyserlingi was lower for individuals provided
with supplementary prey in the field. Mortality
was assessed by disappearance; it is possible that
if some spiders moved and were not re-located
they would have been misclassified as dead. Be-
cause there were four observers searching a lim-
ited patch of relatively sparse habitat, I believe
that few individuals of this conspicuous spider
were missed.

What could cause increased mortality? It is
possible that individuals with poor foraging suc-

cess starved to death, but this seems unlikely.
Individuals maintained in the laboratory sur-
vived on low food for a period well beyond the
scope of this experiment. When individual Ar-
giope die from starvation, they are found hanging
from their webs or in the vegetation below. Dead
spiders were rarely found. It seems much more
likely that they suffered increased risk of pre-
dation. Visually hunting predators, including di-
urnal birds and wasps {e.g., Cryptocheilus sp.),
are more likely to notice moving spiders. Voll-
rath (1985) suggested that movement to a new

Table 6. —Fecundity of Argiope keyserlingi females from field manipulation experiment. Figures based on the
number of females that remained and produced at least one egg case (number that laid any eggs). The number
of juveniles was based on a total count of active juveniles that emerged (all cases combined). Unfed (subset)
treatment group recalculated excluding the single female that captured the record largest natural prey item.

Treatment group

Original

n

Number
at end of
experiment

Number
that laid
any eggs

Number of
egg cases
per fecund
female
(x)

Number of
spiderlings
per case

X (SEM)

Number of
juveniles
per female

X (SEM)

Food added

40

22

9

1.2

293 (34)

147 (66)

Unfed

40

16

4

2.0

290 (43)

145 (137)

Unfed (subset)

39

15

3

0.6

294(122)

98(118)

BRADLEY --PREY, HABITAT AND ARGIOPE

web site increased the risk of predation for Ne~
phila clavipes (Linnaeus), and that individuals
provided with supplementary prey moved less
and suffered a lower rate of mortality. As adult
female Argiope keyserlingi become sated with
food their behavior changed; they ceased re-
building the sticky-orb each morning and they
moved to a retreat in a curled leaf near the upper
attachment point of the web. This shift probably
made them far less obvious to a predator hunting
for spiders in a web. Several of the adult females
that were in the food supplementation treatment
group quit foraging after only two days of extra
food. These spiders remained in their retreat un-
til they laid eggs. Increased foraging success would
thus reduce the apparent exposure of a female to
predators which search webs, and might account
for the lower mortality in the fed group that was
evident only towards the end of the experiment
because the mortality effect would be cumulative
(Fig. 5).

Do patterns of activity and habitat selection
in A. keyserlingi reflect prey abundance or hab-
itat, or both? There is some evidence from this
study that the annual phenology of Argiope key-
serlingi is related to seasonal variation in prey
abundance. Spatial distribution of spider density
was, however, not related to prey abundance pat-
terns. If the data are analyzed treating each date
and plot as a separate sample, there seems to be
little relationship between A. keyserlingi and po-
tential prey. Argiope are most active when prey
are most abundant, but not necessarily in sites
with highest prey numbers. There was no serial-
autocorrelation in the sticky-board insect sam-
ples in this study, suggesting that prey were un-
predictable in time and space. The inconsistency
in prey numbers between years across plots
apparently obscures any more general seasonal
relationship between spider density and prey
abundance. In a comparison of Argiope aurantia
Lucas and A. trifasciata (Forskal), McReynolds
& Polis (1987) concluded that differences in hab-
itat and prey handling abilities explained the small
dietary differences between these two species. Dif-
ferences in diet reflected both seasonal change in
the prey available as well as the size relationships
between growing spiders and the prey that they
were capable of handling (McReynolds & Polis
1987). Few of the measured web characteristics
were correlated with the taxa of prey which were
captured (McReynolds & Polis 1987). My results
with A. keyserlingi appear consistent with those
of McReynolds and Polis insofar as the fact that

103

prey captured seem to reflect seasonal prey avail-
ability. I do not have any comparable informa-
tion on individual prey capture rates, or the re-
lationship between spider size and prey captured.

Although^, keyserlingi reproduction seems to
be closely tied to the biomass of prey captured,
there was no evidence indicating that A. keyser-
lingi exerted control over prey density. This sit-
uation resembles a donor-controlled system
(Pimm 1982). Pimm (1982) suggested that this
sort of relationship should be rare (with the ex-
ception of detritivores). The fact that A. keyser-
lingi abundance appeared to have had little effect
on prey density might be because this orb-weav-
ing spider was only one of many predators that
influenced insect abundance in the study area.
Alternatively, this may be related to the fact that
the density of Argiope keyserlingi was low com-
pared to estimates for oXhQx Argiope species which
average 40 to 50 times higher than those mea-
sured in this study (Olive 1980; Brown 1981;
Horton & Wise 1983). Thus there may be too
few Argiope keyserlingi present to have exerted
control over insect abundance.

Argiope keyserlingi density was related to gen-
eral habitat features, especially those associated
with web-site availability {e.g., Xanthorrhoea
density). The importance of Xanthorrhoea shrubs
as web sites may be related to the fact their struc-
ture, with a brush of long (1 m) narrow (1-6 mm)
leaf blades spreading radially from a central trunk,
provides an infinite gradation of gap sizes which
can accommodate webs of many sizes and ori-
entations. There was no correlation between these
preferred sites and potential insect prey abun-
dance. If prey are not predictable and there is
high mortality among adult females in their webs,
it is possible that Argiope choose web-sites as
much to avoid predation as to maximize prey
capture rate. In light of this, it would be inter-
esting to investigate the influence of web position
on the risk of predation.

Overall, these data support the idea that veg-
etation structure is the chief determinant of web-
site choice for orb-weavers, rather than prey
availability (Enders 1973; Coleboum 1974).
Three of four orb-weaver species studied by Pas-
quet (1984) exhibited a clear relationship be-
tween density and habitat structure, while only
two species built webs where prey abundance was
highest. Furthermore, vegetation structure but
not prey availability was found to be a very im-
portant predictor of spider community structure
(Greenstone 1984). In a study comparing the

104

THE JOURNAL OF ARACHNOLOGY

spider assemblages on three continents, Rypstra
(1986) found that vegetation structure was the
best predictor of spider activity. Prey abundance
was also significantly correlated with spider ac-
tivity at each locality (Rypstra 1986). Riechert
& Gillespie (1986) reviewed the basis for web-
site selection taken from the literature including
data from 14 species of araneids. Vegetation was
a web-site selection criterion for 12 of these 14,
while prey abundance was important for only 5
of the 1 4 species. In the sheet-web building age-
lenid Agelenopsis aperta (Gertsch), web sites are
the subject of intense intraspecific competition;
and there was a clear positive correlation of fa-
vorable web sites with both physical and prey-
capture criteria (Riechert 1974, 1976, 1977, 1979,
1981). It is clear that in cases where vegetation
and potential prey covary, inference about their
relative importance is difficult. Rypstra (1983)
demonstrated that both web-substrate complex-
ity and prey abundance are important to equi-
librium spider density within enclosures. In this
case, prey appear to be a more important deter-
minant, but this result is partially explicable by
reduced interspecific predation and cannibalism
among the spiders in the enclosures maintained
under the high food regimes.

My results reinforce the general conclusion that
vegetation structure is an important predictor of
orb-weaving spider abundance. Prey abundance
appears to be of lesser significance in relation to
spider density, but this may not indicate that it
is less important to the spiders. As Rypstra (1986)
points out, vegetation is easier to quantify; and
I would add that it is probably less variable in
space and time than insect abundance. Perhaps
the difficulties involved in precise quantification
of prey availability are an important confound-
ing factor in broad-scale community analyses.
Prey variability is real and it is possible that
spiders are constrained to use a more reliable
factor (vegetation) in their efforts to select prof-
itable foraging sites. An indication that prey en-
counters influence A. keyserlingi behavior is that
web-site movement was related to success in
capturing prey. Individuals provided with sup-
plementary prey were more sedentary (Fig. 4).
Similar results have been observed in other orb-
weaving spiders in the field (Olive 1982; Janetos
1982; Vollrath 1985). In contrast, there was no
relationship between dietary experience and
movement in laboratory experiments on Nephila
clavipes (Vollrath & Houston 1986).

Gillespie & Caraco (1987) found that individ-
uals of Tetragnatha elongata Walckenaer in a

prey-rich environment actually moved more than
those inhabiting a relatively depauperate area.
Their results appear consistent with the predic-
tions of a risk-sensitive foraging model where
movement will increase as prey availability ex-
ceeds an appropriate physiological requirement.
The behavior of Tetragnatha elongata appar-
ently does not match a second model described
in their paper, which predicts that spider mo-
bility would be inversely related to prey avail-
ability. The key difference between these two
models was whether a foraging spider used cap-
ture success information to predict the best strat-
egy. According to these authors the first model
assumes that temporal variation in prey abun-
dance makes it difficult for a forager to predict
spatial prey distribution. Hence if prey avail-
ability is high, spiders will benefit by sampling
several localities. Results for A. keyserlingi ap-
pear to conflict with Gillespie and Caraco’s re-
sult. Temporal variation in prey abundance ap-
peared to mask spatial predictability, and prey
abundance was limiting to female Argiope, Nev-
ertheless, individual female^, keyserlingi moved
more often when their foraging success was poor.

Interpretation of the present study depends
upon the scale of observation. On a broad scale,
there was a positive relationship between
keyserlingi seasonal phenology to temporal prey-
abundance patterns. On a finer scale, there was
little relationship between prey abundance and
the number of active foraging spiders on indi-
vidual sampling plots. At this scale, spider den-
sity seems to be related to the availability of
preferred sites for the construction of webs. At
the scale of individuals, the history of foraging
success predicted both survival and reproduc-
tion, and had a dramatic infiuence on behavior.

ACKNOWLEDGMENTS

I thank John Clark, Greg Wallis, Rebecca Bla-
don and Jill Smith for their assistance with field
data collection and laboratory analysis. I thank
Jessica Yuille, Merilyn Lean, Ann Parsons, and
Peter Higgins for assistance with the field exper-
iment. I thank Eva Odlander for conducting field
observations. I thank Basil Panayotakos and Sam
Ruggeri for construction of lab enclosures. I thank
Dave Bradley for writing the program for the
randomization test and advice on statistical anal-
yses. Aub Bartlett, Sat and Rabia Murphy pro-
vided excellent facilities and hospitality at the
Crommelin Biological Station. Roger Carolyn and
Belinda Pellow of the School of Biological Sci-

BRADLEY^PREY, HABITAT AND ARGIOPE

105

ences at Sydney University made these facilities
available for my use. I thank the New South
Wales National Parks and Wildlife Service for
permission to conduct research on lands under
their care. I thank Paul Adam, Charles Birch,
Lynn Day, Barry and Marylin Fox, Graham Pyke,
Mark Westoby for valuable discussions of re-
source ecology. I thank Alan Cady, Matthew
Greenstone, Susan Riechert, David Spiller, Amy
Tovar, David Wise and John Wiens for reading
an earlier draft of this manuscript and providing
many helpful suggestions. Amy Tovar assisted
in the field and provided much needed encour-
agement during the long gestation of this paper.
This research was supported by a University of
Sydney Special Projects Grant. This paper is Syd-
ney University LU.S. contribution #2.

LITERATURE CITED

Andrewartha, H. G. & L. C. Birch. 1954. The Dis-
tribution and Abundance of Animals. Univ. Chi-
cago Press, Chicago.

Andrewartha, H. G. & L. C. Birch. 1984. The Eco-
logical Web. Univ. of Chicago Press, Chicago.
Benson, J. S. & H. Fallding. 1981. Vegetation survey
of Brisbane Water National Park and environs.
Cunninghamia, 1 : 7 9™ 1 1 3 .

Bradley, R. A. 1986. The relationship between pop-
ulation density of Paruroctonus utahensis (Scor-
pionida: Vaejovidae) and characteristics of its hab-
itat. J. Arid Environ., 11:165-171.

Bradley, R. A. 1989. Are populations of the desert
grassland scorpion, Paruroctonus utahensis (Vae-
jovidae), limited by food abundance? Southwestern
Nat, 34:46-53.

Brown, K. 1981. Foraging ecology and niche parti-
tioning in orb-weaving spiders. Oecologia, 50:380-
385.

Castillo L., J. A. & W. G. Eberhard. 1983. Use of
artificial webs to determine prey available to orb
weaving spiders. Ecology, 64:1655-1658.
Coleboum, P. H. 1974. The influence of habitat
structure on the distribution of Araneus diadematus
Clerk. J. Anim. EcoL, 43:401-409.

Enders, F. 1973, Selection of habitat by the spider
Argiope aurantia Lucas (Araneidae). American Mid.
Nat., 90:47-55.

Enders, F. 1975. Change of web site in Argiope spi-
ders (Araneidae). American Mid. Nat., 94:484-490.
Fritz, R. S. & D. H. Morse. 1985. Reproductive suc-
cess and foraging of the crab spider Misumena vatia.
Oecologia, 65:194-200.

Gillespie, R. G. & T. Caraco. 1987, Risk-sensitive
foraging strategies of two spider populations. Ecol-
ogy, 68:887-899.

Greenstone, M. H. 1978. The 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, 43:181-193.

Greenstone, M. H. 1 984. Determinants of web spider
species diversity: vegetation structural diversity vs.
prey availability. Oecologia, 62:299-304.

Horton, C. C. & D, H. Wise. 1983. The experimental
analysis of competition between two syntopic spe-
cies of orb-web spiders (Araneae: Araneidae), Ecol-
ogy, 64:929-945.

Hoffmaster, D. K. 1986. Aggression in tropical orb-
weaving spiders a quest for food? Ethology, 72:265-
276.

Janetos, A. C. 1 982. Foraging tactics of two guilds of
web-spinning spiders. Behav, Ecol. Sociobiol,, 10:
19-27.

Janetos, A. C. 1983. Comparative ecology two lin-
yphiid spiders (Araneae, Linyphiidae). J. Arachnol.,
11:315-322.

Juliano, S. A. & J, H. Lawton. 1990. Extrinsic vs.
intrinsic food shortage and the strength of feeding
links: effects of density and food availability on feed-
ing rate of Hyphydrus ovatus. Oecologia, 83:535-
540.

Levi, H. W. 1983. The orb-weaver genera Argiope,
Gea, and Neogea from the western Pacific region
(Araneae: Araneidae, Argiopinae). Bull Mus. Comp.
Zool, 150:247-338.

McReynolds, C. N. & G. A. Polls. 1987. Ecomor-
phological factors influencing prey use by two sym-
patric species of orb- weaving spiders, Argiope au~
rantia and Argiope trifasciata. J. Arachnol, 15:371™
384.

Miyashita, T, 1986. Growth, egg production, and
population density of the spider, Nephila clavata in
relation to food conditions in the field. Res. Popu-
lation Ecol, 28:135-149.

Morse, D. H. & R. S. Fritz. 1982. Experimental and
observational studies of patch choice at different
scales by the crab spider Misumena vatia. Ecology,
63:172-182.

Graze, M. J, & A. A. Gigarick. 1989. Biological con-
trol of aster leafhopper (Homoptera: Cicadellidae)
and midges (Diptera: Chironomidae) by Pardosa
ramulosa (Araneae: Lycosidae) in California Rice
Fields. J. Econ. Entomol, 82:745-749.

Olive, C. W. 1980. Foraging specializations in orb-
weaving spiders. Ecology, 61:1 133-1 144.

Olive, C. W. 1982. Behavioral response of a sit-and-
wait predator to spatial variation in foraging gain.
Ecology, 63:912-920.

Pasquet, A. 1984. Predatory-site selection and ad-
aptation of the trap in four species of orb-weaving
spiders. Biol Behav., 9:3-19.

Pimm, S. L. 1982. Food Webs. Chapman and Hall,
London.

Polis, G. A. & S. J. McCormick. 1986a. Patterns of
resource use and age structure among species of des-
ert scorpion. J. Anim. Ecol, 55:59-73.

Polis, G. A. & S. J. McCormick. 1986b. Scorpions,
spiders and solpugids: predation and competition

106

THE JOURNAL OF ARACHNOLOGY

among distantly related taxa. Oecologia, 71:111-
116.

Riechert, S. E. 1974. The pattern of local web dis-
tribution in a desert spider: mechanisms and sea-
sonal variation. J. Anim. Ecol., 43:733-746.

Riechert, S. E. 1976. Web-site selection in the desert
Agelenopsis aperta. Oikos, 27:311-315.

Riechert, S. E. 1977. Games spiders play: behavioral
variability in territorial disputes. Behav. Ecol. So-
ciobiol., 4:2-39.

Riechert, S. E. 1979. Games spiders play. II. Re-
source assessment strategies. Behav. Ecol. Socio-

biol., 6:121-128.

Riechert, S. E. 1981. The consequences of being ter-
ritorial: spiders, a case study. American Nat., 1 17:
871-892.

Riechert, S. E. & L. Bishop. 1990. Prey control by
an assemblage of generalist predators: spiders in gar-
den test systems. Ecology, 71:1441-1450.

Riechert, S. E. & A. B. Cady. 1983. Patterns of re-
source use and tests for competitive release in a
spider community. Ecology, 64:899-913.

Riechert, S. E. & R. G. Gillespie. 1 986. 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, Cali-
fornia.

Riechert, S. E. & T. Lockley. 1984. Spiders as bio-
logical control agents. Ann. Rev. Entomol., 29:299-
320.

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

Robinson, B., & M. H. Robinson. 1978. Develop-
mental studies of Argiope argent at a (Fabricius) and
Argiope aemula (Walckenaer). Sym. Zool. Soc. Lon-
don, 42:31-40.

Rogers, L. E., Hinds, W. T. & R. L. Buschbom. 1976.
A general weight vs. length relationship for insects.
Ann. Entomol. Soc. America, 69:387-389.

Rogers, L. E., Buschbom, R. L. & C. R. Watson. 1977.
Length-weight relationships of shrub-steppe inver-
tebrates. Ann. Entomol. Soc. America, 70:51-53.

Rubenstein, D. 1. 1987. Alternative reproductive tac-
tics in the spider Meta segmentata. Behav. Ecol.
SociobioL, 20:229-237.

Rypstra, A. L. 1982. Building a better insect trap; an
experimental investigation of prey capture in a va-
riety of spider webs. Oecologia, 52:31-36.

Rypstra, A. 1983. The importance of food and space
in limiting web-spider densities; a test using field
enclosures. Oecologia, 59:312-316.

Rypstra, A. L. 1986. Web spiders in temperate and

tropical forests: relative abundance and environ-
mental correlates. American Mid. Nat., 1 15:42-51.

SAS Institute Inc. 1988. SAS/STAT User’s Guide,
Release 6.03 ed. SAS Institute Inc. Cary, NC. 1028

pp.

Schaefer, M. 1978. Some experiments on the regu-
lation of population density in the spider Floronia
bucculenta (Araneida: Linyphiidae). Sym. Zool. Soc.
London, 42:203-210.

Sokal, R. R. & F. J. Rohlf. 1981. Biometry. 2nd ed.
W. H. Freeman and Co., San Francisco.

Spiller, D. A. 1984. Competition between two spider
species: experimental field study. Ecology, 65:909-
919.

Spiller, D. A. 1986. Interspecific competition be-
tween spiders and its relevance to biological control
by general predators. Environ. Entomol., 15:177-

181.

Suter, R. B. 1985. Intersexual competition for food
in the bowl and doily spider, Frontinella pyramitela
(Araneae, Linyphiidae). J. Arachnol., 13:61-70.

Turner, M. & G. A. Polis. 1979. Patterns of co-ex-
istence in a guild of raptorial spiders. J. Anim. Ecol.,
48:509-520.

Vollrath, F. 1985. Web spider’s dilemma: a risky
move or site dependent growth. Oecologia, 68:69-
72.

Vollrath, F. & A. Houston. 1986. Previous experi-
ence and site tenacity in the orb spider Nephila (Ara-
neae, Araneidae). Oecologia, 70:305-308.

Wiens, J. A. 1 989. The ecology of bird communities.
Vol. 1, Foundations and patterns. Cambridge Univ.
Press, Cambridge, p 16.

Wise, D. H. 1975. Food limitation of the spider L/>7-
yphia marginata: experimental field studies. Ecol-
ogy, 56:637-646.

Wise, D. H. 1 979. Effects of an experimental increase
in prey abundance upon the reproductive rates of
two orb-weaving spider species (Araneae: Aranei-
dae). Oecologia, 41:289-300.

Wise, D. H. 1 983. Competitive mechanisms in a food-
limited species: relative importance of interference
and exploitative interactions among labyrinth spi-
ders (Araneae: Araneidae). Oecologia, 58:1-9.

Wise, D. H. 1984. The role of competition in spider
communities: Insights from field experiments with
a model organism. Pp. 42-53, In Ecological Com-
munities: conceptual issues and the evidence. (D.R.
Strong, D. Simberlof, L. G. Abele, A. B. Thistle,
eds.) Princeton University Press, Princeton, New
Jersey, USA.

Manuscript received 28 July 1992, revised 24 March
1993.

1993. The Journal of Arachnology 21:107-1 19

CONSTRAINTS AND PLASTICITY IN THE DEVELOPMENT
OF JUVENILE NEPHILA CLAVIPES IN MEXICO

Linden Higgins': Centro de Ecologia, Universidad Nacional Autonoma de Mexico,
Apartado Postal 70-275, Ciudad Universitaria, C. P. 01450, Mexico

ABSTRACT: The large, orb- weaving spider Nephila clavipes is found in a diversity of habitats within a narrow
latitudinal range in Mexico. This allowed nearly simultaneous study of post-embryonic development of six
disjunct populations in dissimilar environments. A common-garden laboratory study utilizing juveniles collected
in four sites reinforced the conclusions from the field. The developmental parameters influencing growth in size
at ecdysis did not vary within or among populations and may be genetically determined. Although very small
juveniles exhibit variation in the growth per ecdysis, larger juveniles exhibit very little variation. These data,
compared to data from field and laboratory studies of other tropical populations of N. clavipes, indicate that
growth per ecdysis is highly constrained. Thus, this developmental parameter establishes a developmental
trajectory that may be genetically determined and therefore subject to natural selection.

RESUMEN: La arana Nephila clavipes, tejedora de telas orbiculares, se encueritra en habitats diversos en
Mexico dentro de un limite angusto de latitud. Eso permitio estudios casi simultaneos de ontogenia de juveniles
en seis poblaciones desunidos en ambientes distintos. Un estudio del laboratorio utilizando juveniles colectados
en cuatro sitios fortalece las conclusiones del campo. Los parametros ontogeneticos determinando el crecimiento
por muda no vario dentro ni entre poblaciones, y posiblemente son geneticamente determinados. Aunque
juveniles muy pequenos mostraron variacion en el crecimiento por muda, juveniles mas grandes no mostraron
variacion en este parametro. Estos datos, combinados con datos de estudios de otras poblaciones en el laboratorio
y en el campo, indican que el crecimiento por muda esta muy coiistrenido. Asi, este parametro establece una
trayectoria ontogenetica que posiblemente sea geneticamente determinado y sujeto a seleccion natural.

Determining how environmental factors influ-
ence life history requires determination of the
developmental parameters at each life-history
stage and examination of whether the parameters
are phenotypically plastic, responding to the en-
vironment, or are genetically determined (Ca-
swell 1983, Via & Lande 1985, Pease & Bull
1988). The development of the large orb-weav-
ing spider Nephila clavipes (Linnaeus) (Araneae:
Tetragnathidae) can be expressed as a group of
interdependent parameters with varying degrees
of phenotypic plasticity (Higgins 1992a). Exper-
imental trials showed that weight gain and in-
termolt interval duration responded to shifts in
food availability, but growth per ecdysis did not
(Higgins pers. obs.). The duration of the inter-
molt interval was found to be correlated with the
size of the spider and apparently reflected the
length of time the individual required to achieve
the minimum weight necessary to molt to the
next instar. In field studies, whereas the rate of

' Current address: Dept, of Zoology, University of Tex-
as at Austin, Austin, Texas 78712 USA

weight gain and the number of juvenile molts
varied with habitat within and among popula-
tions, the growth per ecdysis did not vary within
a population or between two populations in the
tropics (Higgins 1992a). The constraints im-
posed upon development by the constant growth
per ecdysis were countered by phenotypic plas-
ticity in other developmental parameters, par-
ticularly the number of instars, generating vari-
ation in size and age at maturity (Higgins 1992a;
Higgins pers. obs.).

The previous studies utilized a range of sites
with widely differing physical and biological con-
ditions including photoperiod, making it difficult
to distinguish the relative influence of different
ecological factors on post-embryonic develop-
ment. In order to better understand the environ-
mental influence on development, a second field
Study was undertaken, utilizing the diversity of
habitats in which N. clavipes is found in Mexico.
Choosing sites within a U latitude range elimi-
nated variation in photoperiod among sites and
allowed nearly simultaneous study of popula-
tions experiencing very different environments.

107

108

THE JOURNAL OF ARACHNOLOGY

In order to more fully explore the plasticity and
constraints of development, juveniles from the
four most distinct habitats were brought into the
laboratory in a common-garden experiment.

Common-garden experiments, where individ-
uals from different environments are held in a
common environment, allow preliminary differ-
entiation between those parameters that are ap-
parently genetically determined and those that
are phenotypically plastic (Wise 1987). Param-
eters that are genetically determined and vary
among populations will express similar variation
in the laboratory among individuals from those
populations. In contrast, parameters that are
phenotypically plastic and vary among popula-
tions will not vary in the laboratory among in-
dividuals from different populations. The com-
bination of laboratory and field observations thus
allows distinction between constrained, poten-
tially genetically determined, parameters and
phenotypically plastic parameters.

METHODS

Populations studied.— Spiders were observed
in six sites spanning Mexico from the Veracruz
coast to the Jalisco coast. Three sites were in
Veracruz: Playa Escondida, Nanciyaga, and For-
tin de las Flores. One site was in the high altitude
desert valley of Tehuacan, Puebla. Two sites were
west of the central plateau: Arroyo Frio, Mi-
choacan, and Chamela, Jalisco. In 1989 I trav-
eled to Playa Escondida, Nanciyaga, Fortin de
las Flores, Arroyo Frio, and Chamela. In 1990,
I studied spiders in Playa Escondida, Nanciyaga,
Fortin de las Flores, Tehuacan, and Chamela.
Because these sites varied only 1® latitude, there
was no significant difference in photoperiod
among them. However, the sites varied in many
environmental parameters including type and
degree of seasonality, and prey capture rates (Ta-
ble 1).

Playa Escondida and Nanciyaga are privately
owned forest preserves about 13,5 km apart on
the Veracruz coast, separated by cattle ranches.
Both have wet climates with relatively cool win-
ters. Playa Escondida is approximately 1 km from
a previous study site, the biological station “Los
Tuxtlas” (Higgins 1992a, b). The N. davipes pop-
ulation at the latter site disappeared shortly after
dispersal of juveniles early in 1 989, and the study
was continued at Playa Escondida. The third Ve-
racruz site, Fortin de las Flores, is a mid-altitude
area of coffee plantations (Benton & Uetz 1986).
This site is cooler and experiences stronger win-

ters with minimum temperatures as low as 0 °C.
On the western, dry side of the Sierra Madre
Oriental in the valley of Tehuacan, I studied the
spiders at the Secretaria de Ecologia y Desarrollo
Urbano cactus garden near Zapotitlan Salinas,
Puebla. Annual rainfall in Tehuacan is very low
and there is a relatively cold winter. Climato-
logical data are not available for Arroyo Frio,
located near Perdenales in southwestern Mi-
choacan. The area is seasonally dry and, due to
the altitude, seasonally cool. However, there is
a permanent stream through the site that main-
tains high relative humidity within the arroyo
where the spiders are found. On the Pacific coast
of Jalisco, the spiders were studied at Chamela
field station, owned and run by the Institute de
Biologia, Universidad Nacional Autonoma de
Mexico (UNAM). This site is seasonally dry but
never cool. Spiders were found during the rainy
season at all sites, but are facultatively bivoltine
at both coastal Veracruz sites (Higgins in press).
There was no second generation at these sites in
1989 and 1990, so data are presented from only
the rainy season.

Rates of prey capture were estimated through
trap-line surveys (Turnbull 1960). The spiders
in Playa Escondida captured fewer prey than the
spiders at other sites (Table 1) (Higgins pers. obs.).
Predation load, estimated as the proportion of
juveniles less than 0.5 cm leg I tibia + patella
length that abandoned intact orb webs following
predator attack (as in Higgins 1 992b), was higher
in Tehuacan, Arroyo Frio, and Chamela, but the
differences were not significant (Higgins pers.
obs.).

I traveled alternately east and west from Mex-
ico City, visiting inland sites both on my way to
and returning from the coastal sites. The visits
at the coastal sites were slightly longer than visits
to the inland sites. The combination of longer
visits at the end points and repeated visits at
intermediate sites enhanced the probability of
observing molts by marked individuals.

Field ol)ser¥ations*~ Field observations of in-
dividuals utilized the methodology previously
described in detail (Higgins 1992a). At each site,
censuses of spider abundance and size were made
and web sites were flagged. Measurements, made
with Helios needle-tipped calipers, included spi-
der leg I tibia + patella length (TPF, cm), ab-
domen length, and abdomen width. Individuals
larger than 0. 5 cm TPL were marked with enamel
paint on their legs. From abdomen length and
width, the abdominal volume was estimated as

HIGGINS-- DEVELOPMENT OF JUVENILE NEPHILA CLAVIPES

109

Table L— Location, climate and relative prey capture rates at each site. Annual rainfall and mean temperature
were taken from the nearest weather station reported by Garcia (1973) for all sites except Chamela. Data from
Chamela come from Bullock (1986 and pers. commun.). Temperatures are for the growing season. No weather
data are available for Arroyo Frio. (§ Mean prey capture per 1 2 diurnal hours per spider (Higgins, pers. obs.),
determined as in Higgins & Buskirk 1992.)

Site

Coordinates

Altitude

(m)

Annual

rainfall

(m)

Mean

temp,

°C

Prey
capture §

Playa Escondida

18°35'N, 95°W

0

4.5

26

low

Nanciyaga

18°35'N, 95°W

100

4.5

26

high

Fortin de las Flores

18°50'N, 97°W

1000

2.5

22

high

Tehuacan

18°20'N, 97°30'W

1200

0.3

20

high

Arroyo Frio

19°10'N, 101°30'W

1200

Chamela

19°30'N, 105°W

50

0.7

—

high

a cylinder. For a given TPL, abdomen volume
is highly correlated with spider weight (Higgins
1992a). Each individual found on a non-viscid
silk platform was assumed to be pre- or post-
ecdysis (Higgins 1990). The size of the abdomen
relative to the legs distinguished between these
conditions: pre-ecdysis individuals have large,
distended abdomens whereas post-ecdysis indi-
viduals have much smaller abdomens relative to
leg length and carapace width. Collection of ex-
uviae provided additional data for the analysis
of growth per ecdysis. Post-ecdysis spiders often
hang the exuvium in the barrier webs near the
hub connection, and TPL of an exuvium is not
significantly different from the spider size in the
previous instar (Higgins 1992a). When several
exuviae were present, I only measured the larg-
est, from the most recent molt.

These data are used to compare the relation-
ships among abdomen volume, premolt TPL,
and postmolt TPL within and among the pop-
ulations over the entire life-cycle of the spiders.
These parameters all describe the growth per molt,
and as such are not strictly independent. How-
ever, because there was a chance that spiders gain
weight beyond that required to successfully com-
plete ecdysis (particularly relevant for penulti-
mate-instar individuals), premolt abdomen vol-
ume was compared as well as premolt and
postmolt TPL. To describe growth patterns of
the species in North America, data from previous
studies in Texas, USA, “Los Tuxtlas”, Veracruz
Mexico, and Panama (Higgins 1 992a) were com-
pared to those presented here.

Common garden experiment. —In 1990, juve-
nile spiders from two wet and two dry sites were
brought into the laboratory. The populations se-

lected for laboratory study were Nanciyaga, For-
tin de las Flores, Tehuacan and Chamela. The
spiders were maintained on three dimensional
frames made of two intersecting 30 cm circles
made of fiberglass strips, and were free to move
about the laboratory. From each site, spiders of
0.2-0. 4 cm TPL were collected. Twenty-three
spiders were used in the experiment, as follows:
Nanciyaga (3 females, 1 male), Fortin (3 females,
3 males), Tehuacan (2 females, 4 males), Cha-
mela (2 females, 5 males). The ratio of juvenile
males to females depended upon the exact dates
of collection. Later in the season, small spiders
are more likely to be males (pers. obs.). In ad-
dition, three spiders escaped prior to being
marked during a trip to Fortin and Tehuacan.
These animals were included in the description
of weight gain during the intermolt interval.

The spiders were maintained 3-4 days with
only water to increase the probability of spinning
when released, and were offered food immedi-
ately after a web was spun. Any spider that did
not spin an orb within three days of release in
the laboratory was not included in the experi-
ment. Throughout the experiment, each spider
was offered three Drosophila virens each day. For
a period of one week, only D. melanogaster was
available; two of these were substituted for each
of the larger D. virens (for a total of six flies). At
the initiation of the experiment, the spiders were
measured (TPL, abdomen length, abdomen
width), and these measures were repeated with
each molting. Abdomen volume was measured
every other day during the intermolt interval to
monitor weight gain. The majority of the spiders
were held for two molts; one individual from
Fortin failed to molt a second time in the lab-

110

THE JOURNAL OF ARACHNOLOGY

□

0.5

0.0

□

o

Playa Escondida

□

Nanciyaga

A

Fortin

▲

Tehuacan

■

Arroyo Frio

•

Chamela

0.2 0.4 0.6 0.8

premolt abdomen volume, cc

1.0

Figure 1.— Postmolt size (leg I tibia + patella length, TPL) as a function of premolt abdomen volume at all
sites. Only data from juveniles and females molting to maturity are included in the graph. The curve represents
a fit to the entire data set, y = 1.66 x (R2 = 0.99).

oratory. Observations also distinguished those
days when the spiders were foraging and had
partially or wholly renewed orbs from the days
immediately pre- or post-molt, when the spiders
were not actively foraging. Data presented here
concern the growth per ecdysis, length of the in-
termolt, days spent foraging during the intermolt
and pattern of abdomen volume gain during the
intermolt period.

Statistical analyses* —The developmental pa-
rameters examined in this study are dependent
upon the size, TPL, of the individual. Therefore,
all analyses tested for a significant regression be-
tween TPL and the measurement in question. If
the regression analysis was significant for each
population, further analysis tested for significant
variation in the slope of the regression lines among
populations. If there was a significant interaction
of population and TPL, indicating difference in
slope, then the analysis was halted (Sokal & Rohlf
1981). If the interaction terms were not signifi-
cant, a final analysis of covariance (ANCOVA)
with TPL as covariate tested for variation in the
altitude of the line (y intercept). Lastly, if no
difference was found due to population or treat-
ment in a biologically important variable, a pos-
teriori power tests were calculated to determine
the minimum percent difference in the slope or
intercept that could have been detected with these
data.

RESULTS

Field observations. volume and
molting: The relationships among premolt ab-
domen volume, premolt TPL and postmolt TPL
vary little within or among the six populations
studied. Postmolt TPL is related to the abdomen
volume by a concave function, approximately a
function of the cube root of abdomen volume
(Fig. 1). The data are insufficient to allow com-
parison among the sites because recording pre-
molt abdomen volume and postmolt TPL for the
same individual was unlikely in the field.

The premolt abdomen volume is a function of
premolt TPL, and males and females molting to
sexual maturity do so at a lower abdomen vol-
ume compared to juveniles molting to juvenile
instars. The data from Fortin included the great-
est number of observations of molts to sexual
maturity (Fig. 2). Penultimate instar males and
females were identified as follows: almost all fe-
males of TPL greater than 1 .0 cm are molting to
sexual maturity (pers. obs.) and penultimate males
have swollen palpi. After correcting for heter-
oscedasticity by taking the square-root of the de-
pendent variable (cube root of abdomen vol-
ume), ANCOVA of the data from Fortin revealed
that the differences among juveniles, penultimate
instar males, and penultimate instar females were
significant (TPL: F(,_52) = 1475.6, P < 0.001; sex/

HIGGINS-== DEVELOPMENT OF JUVENILE NEPHILA CLAVIPES

111

Figure 2.— The cube root of premolt abdomen volume as a function of premolt TPL (leg I tibia + patella
length) for spiders in Fortin. These data include males molting to sexual maturity (= A) and juveniles and
females (= o). The points above premolt TPL = 1.0 cm are females molting to sexual maturity.

age: F(2,52) = 8.85, P < 0.001; interaction: F(2, 52)
= 13.75, P < 0.001). Separate comparison of
males and females molting to sexual maturity
indicated that the function of premolt abdomen
volume on premolt TPL have the same slope
(F(i, 15) = 2.28, ns, power test: 4.1% detectable
difference in slope).

Data from the remaining populations included
few observations of spiders molting to maturity,
so the comparison of the function of premolt
abdomen volume on premolt TPL among sites
utilized only data from juvenile molts. Arroyo
Frio data were excluded from the final analysis
because few spiders were observed in the size
range of 0.5 cm-1.0 cm TPL. ANCOVA of pre-
molt abdomen volume (to the 0. 1 5 power) with
TPL as covariate revealed that there was no dif-
ference in slope among all five sites (F(4^ 107) ==
1.75, ns, power test: 1.8% detectable difference).
The interaction term was dropped from the final
ANCOVA, and this test showed that spiders in
Chamela molted at a slightly but significantly
lower premolt abdomen volume for their size
[ANCOVA. TPL: F^,, ,„) = 4283.4, P < 0.001;
site: F(4, = 7.63, P < 0.001; regressions: Cha-

mela: y = 0,30 + 0,54 (TPL); remaining sites: y
= 0.32 + 0.55 (TPL)].

Growth per ecdysis: Growth per ecdysis was
compared within and among populations using
regression analysis of postmolt TPL on premolt
TPL. The slope of the regression line is an in-
dication of the rate of size-specific growth. Molt-
ing to sexual maturity was presumed to affect the
rate of growth at ecdysis because postmolt TPL
is correlated with premolt abdomen volume, and
whether the spiders were molting to maturity
influenced the premolt abdomen volume. There-
fore, the observations of molting juveniles, males
molting to maturity, and molting females larger
than LO cm TPL were considered separately.
Within sites, preliminary regression analyses
showed that premolt and postmolt TPL were sig-
nificantly correlated for juvenile molts at all sites
(all P < 0.003) (Fig. 3). Individual ANCOVA
were run to check for differences in growth during
juvenile molts between years at the sites studied
in both years: Nanciyaga, Playa Escondida, For-
tin de las Flores, and Chamela. No significant
differences were found (all P > 0.12); therefore,
data from 1989 and 1990 were combined for the
remaining tests. Due to small numbers of ob-
servations, the data from both years were pooled
for the analyses of growth during molts to sexual
maturity. Preliminary regression analyses of

112

THE JOURNAL OF ARACHNOLOGY

2.0 1

1.5 -

1.0

0.5-

0.0

2.0'

1.5

1.0

Tehuacan
Arroyo Frio

Chamela

O

O Tehuacan
□ Arroyo Frio

AO

0.5 - A

/

O 1989

A 1990

0.0

Nanciyaga

OO

A A

o

/

0

1989

A

1990

■

males

0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5

Playa Escondida

<9

GD

_|

0 1989

A 1990
■ males

0.0 0.5 1.0 1.5

premolt TPL, cm

premolt TPL, cm

Figure 3.— Growth per ecdysis, determined as the relationship between premolt TPL (leg I tibia + patella
length) and postmolt TPL for all populations observed. Arroyo Frio (1989) and Tehuacan (1990) are plotted
together, all other plots contain data from two years. Males molting to maturity are indicated by solid squares
(■).

postmolt TPL on premolt TPL were significant
for males from Playa Escondida, Nanciyaga, and
Fortin, and for females from Nanciyaga {P <
0.02).

Travel precluded collecting complete data sets
for all populations, therefore comparison be-

Table 2.~ ANCOVA of growth per molt of juveniles
and males molting to maturity in Playa Escondida,
Nanciyaga, and Fortin de las Flores. (* P < 0.01, ** P
< 0.001)

Factor

df

F ratio

Premolt TPL

1

419.3**

Site

2

1.26

Maturity

1

3.28

Site X premolt TPL

2

2.98

Maturity x premolt TPL

1

7.81*

Site X maturity

2

0.26

Error

133

tween age and size classes Ouvenile, male or fe-
male) were restricted to a few sites. Data from
Nanciyaga indicated that there was no difference
in growth per ecdysis between juveniles and fe-
males molting to maturity (no interaction effect
= 0.01, ns; ANCOVA. TPL: F„.58) = 1567.1,
P < 0.001; maturity: F,, jg) = 2.01, ns). In all
three Veracruz sites, males were observed molt-
ing to maturity. ANCOVA showed significantly
lower growth per ecdysis (slope of the line) in
males molting to maturity than in molting ju-
veniles less than 1.0 cm premolt TPL (Table 2).

Comparisons made among sites for juvenile
molts revealed no difference among sites in
growth per ecdysis either in slope or in intercept
(slope: F(5 208) = 1-98, ns, power test 0.9% de-
tectable difference; intercept: 213) = 1-48, ns,

power test 1.9% detectable difference) (Table 3).
However, these regression analyses obscure a
slight non-linearity of the data. Closer exami-

HIGGINS-DEVELOPMENT OF JUVENILE NEPHILA CLAVIPES

113

1.0-

A

0.8-

CL

^•i

<M

0.6-

CL

H

s

0.4-

il

_

0

0.2-

O.oJ

0.0

o-

□

Playa Escondida

A

Tehuacan

0

Nanciyaga

•

Arroyo Frio

A

Fortin

■

Chamela

0.5 1.0

premolt TPL, cm

1.5

Figure 4. —The growth statistic C as a function of premolt TPL (leg I tibia + patella length). The value of C
equal to the average slope of the equations of growth per ecdysis (Table 3) is indicated by the dotted line.

nation of variation in the growth per ecdysis was
permitted by calculation of a growth statistic, C
= ln(TPL2/TPLl). This growth statistic is related
to Huxley’s growth equation (1972 p. 6) as C =
In(aG) when dt = 1 instar, but it is not the same
as k. Mean C is a function of the regression equa-
tion slope (In(slope) = mean C). This transfor-
mation revealed that although C was indepen-
dent of premolt TPL for larger spiders, for
individuals of the first and second instars (TPL
< 0.3) C was strongly dependent upon premolt
TPL (Fig. 4). The data from all populations ap-
pear to fall on the same curve. The high values
of C for small spiders reflect the large changes in
TPL at ecdysis. Spiderlings with premolt TPL of
0.05 cm often molt to TPL of 0.1 1 cm, an in-
crease of 100% or C = 0.8.m.

The data concerning juvenile growth per ec-
dysis can be compared to data collected in earlier

studies of populations at Barro Colorado Island,
Panama, at Los Tuxtlas, Veracruz, Mexico, and
in southeastern Texas, USA (Higgins 1992a). The
older data set did not distinguish males molting
to sexual maturity. Therefore, the ANCOVA was
run with juveniles of less than 1.0 cm premolt
TPL and males. There was no difference among
any of the tropical populations (no interaction
effect: F(7 355) = 1.67, ns, power test: 2.2% de-
tectable difference; ANCOVA. TPL: F(, 372) =
1 9,370.4, P <0.001; site: F(7 372) = 0.99, ns, power
test: 4.7% detectable difference). Inclusion of ob-
servations from the University of Houston
Coastal Center in Galveston County, Texas, re-
sulted in a significant site effect on slope, reflect-
ing the significantly lower slope of the growth per
ecdysis for the Texas population (ANCOVA.
TPL: F(, 425) = 5776.6, P < 0.001; site: F(8 425) =
1.14, ns; site x TPL: F(8,425) = 4.91, P < 0.001).

Table 3. — Regression equations for juvenile growth per ecdysis, where initial TPL is less than 1.0 cm. (**P
< 0.001)

Site

n

Regression

intercept

Regression

slope

P2

F (regression)

Playa Escondida

31

0.068

1.24

0.99

2671.1**

Nanciyaga

55

0.051

1.29

0.97

1959.7**

Fortin de las Rores

41

0.043

1.32

0.99

3150.0**

Tehuacan

24

0.053

1.19

0.96

591.0**

Arroyo Frio

17

0.078

1.25

0.99

1423.0**

Chamela

52

0.047

1.29

0.99

4743.5**

114

THE JOURNAL OF ARACHNOLOGY

0.8-1

0.7-

1 0.6H

- 0.5 H

E

o

CL 0.4-

0.3-

0.2

>•

o ■

•m

oA O

0.1 0.2 0.3 0.4 0.5 0.6

premolt TPL, cm

0.4n

0.3

0.1

> •

■ •

• 4*0 *

□ ® ° °

0.1 0.2 0.3 0.4 0.5

premolt TPL, cm

0.6

Figure 5,— Growth in the laboratory by spiders from four populations. Data from two molts are plotted, with
the first molt indicated by open symbols and the second indicated by closed symbols. Squares = Chamela;
diamonds = Tehuacan; triangles = Fortin; circles = Nanciyaga. The arrows indicate values for the male from
Nanciyaga that delayed molting. Graph a = Growth per ecdysis in the laboratory, plotted as postmolt TPL (leg
I tibia + patella length) vs. premolt TPL. Graph b = Premolt abdomen volume as a function of premolt TPL.

Common garden experiment.— Twenty-three
spiders with TPL of 0. 2-0.4 cm were brought to
the laboratory from Nanciyaga, Fortin, Tehu-
acan, and Chamela and held for one complete
intermolt cycle (two molts). The entire study last-
ed from June to October, and while all changes
in TPL were verified by myself, variation caused
by different persons making measurements pro-
duced increased error in the estimations of ab-
domen volume. In particular, data from two dates
had to be excluded from the analysis of increas-
ing abdomen volume, resulting in removal of five
observations.

A total of 1 3 males were included in the study
and five of them molted to maturity in the second
molt in the laboratory: one from Nanciyaga, two
from Fortin and two from Tehuacan. Therefore,
analyses included the parameter of juvenile vy.
maturation molt where appropriate.

Utilizing both observed molts for each indi-
vidual, regression analysis of growth per ecdysis
was significant for each population (all P < 0.003)
(Fig. 5a). ANCOVA of growth per ecdysis showed
no significant difference among these popula-
tions or between juvenile and maturation molts
(no interaction effects, P > 0. 1 ; ANCOVA. TPL:
F(, 38) = 230.9, P < 0.001; population: 33) =

0.55, ns; maturation: F(, 33) = 0.004, ns). These
data were compared to 37 observations of molts
in the field from the same sites and the same

premolt TPL (0.2-0. 5 cm). Preliminary analysis
revealed no significant difference in slope among
sites or between conditions (field or laboratory),
nor a significant interaction of site and condition
(ANCOVA: site x TPL: F(3 gg) = 0.43, ns; con-
dition X TPL: F(, 68) = 0.10, ns; site x condition:
F(3. 68) = 0-14, ns). Final ANCOVA testing for
primary effects showed no significant variation
due to site or condition (TPL: F(, 75) = 450.9, P
< 0.001; site: F(3 75) = 1.52, ns; condition: F(, 75)
= 0.65, ns).

Several parameters describing the intermolt
interval were collected from the laboratory ani-
mals (Table 4). ANOVA showed no significant
difference among populations in mean TPL fol-
lowing the first molt in the laboratory, although
Nanciyaga and Fortin individuals were slightly
larger. The total intermolt interval and the num-
ber of days foraging between molts were not af-
fected by these slight differences in size (regres-
sion of total intermolt duration: TPL: F(, ,5) =
1.35, ns; regression of days foraging: TPL: F(, ,4)
= 3.43, ns). The total intermolt interval and days
foraging in the laboratory varied among sites but
was not affected by whether the individual molt-
ed to sexual maturity (ANOVA of intermolt in-
terval. site: F(3 ,6) = 3.69, P = 0.03; maturity:
F(, ,6) = 1.64, ns; ANOVA of days foraging, site:
F(3, 15) = 4.69, P = 0.02; maturity: F(, ,5) = 0.002,
ns). Differences between Nanciyaga and the re-

HIGGINS-DEVELOPMENT OF JUVENILE NEPHILA CLAVIPES

115

Table 4. — Intermolt duration in the laboratory for spiders from four populations. The TPL (leg 1 tibia +
patella length) reported is the measurement following the first molt in the laboratory. Letters refer to statistically
similar values among sites.

Site

n

TPL ± 1 SD

Total days ± 1 SD

Days foraging ± 1 SD

Nanciyaga

4

0.48 ± 0.08

26.8 ± 8.0 (a)

24.3 ± 9.0 (c)

Fortin de las Flores

6

0.47 ± 0.04

18.8 ± 4.0 (b)

15.4 ± 2.5 (d)

Tehuacan

6

0.40 ±0.13

17.3 ± 4.3 (b)

14.8 ± 3.9 (d)

Chamela

7

0.40 ± 0.04

18.7 ± 3.1 (b)

13.2 ± 1.7 (d)

maining sites appear due to one male from Nan-
ciyaga that took over 30 days to complete the
intermolt interval and molt to maturity, twice
the usual intermolt duration for spiders of this
size.

The abdomen volume gain in the laboratory
was independent of site. In the first molt, spiders
from Tehuacan molted at a significantly higher
premolt abdomen volume (no interaction affects;
ANCOVA. site: F^, ,3) = 5.42, P = 0.01). All
spiders molted at the same relative abdomen vol-
ume in the second molt (no interaction affects;
ANCOVA. TPL: F„, ,6) = 43.4, P < 0.001; site:
F(3, ,6) = 2.61, ns; maturity: F(, ,6) = 1-3; ns) (Fig.
5b). Because there was no difference in the sec-
ond molt in premolt abdomen volume among
sites or between molts to maturity and juvenile
molts, data from all individuals held for a com-
plete intermolt cycle (including three of unknown
origin) were combined to describe the pattern of
abdomen volume increase over the intermolt.
The relative change in abdomen volume [ln(av(d)/
av (0)], where d = day and 0 = day of molt, was
plotted against time for spiders molting within
20 days, for spiders molting in 20-26 days, and
for the individual from Nanciyaga requiring 36
days (Fig. 6). The general trend was for the rate
of abdomen volume increase to slow as the spi-
ders approached the next molt. The individual
from Nanciyaga that took longer between molts
did achieve a greater premolt abdomen volume
and grew slightly more at ecdysis than the other
spiders, as indicated in Figs. 5a and 5b by arrows.

DISCUSSION

In order to interpret variation in phenology
and size at maturity, the proximal developmen-
tal causes of the variation must be identified. In
arthropods, variation in two developmental pa-
rameters can lead to differences in size at ma-
turity: there may be variation in the change in
size at each molt, or there may be variation in
the number of juvenile molts. Variation in either

parameter can result in the same adult size, but
the conditions under which each varies may be
distinct. Genetic variation or phenotypic plas-
ticity can lead to differences in development
within and among populations, but the evolu-
tionary consequences of each source of variation
are distinct (Pease & Bull 1988). Longitudinal
observations of juvenile growth are a first step
towards determining how environmental factors
generate differences in adult size, and whether
these differences are the result of phenotypic re-
sponse to the environment or genetic variation
among individuals within or among habitats. The
results of the studies of N. davipes imply that
some developmental parameters are highly plas-
tic while growth per ecdysis is constrained and
may be genetically determined (Higgins 1992a,
present study). Such information is not available
from the census data presented in past arach-
nological studies without making basic assump-
tions concerning developmental processes.

Traditionally, field measures of growth uti-
lized either frequency distributions of a single
measure, such as carapace width, or the regres-
sion of two allometric body parts of individuals.
The difficulties of determining growth and instar
number from the former measure have been rec-
ognized (Polis & Sisson 1990); however, the lat-
ter analysis also presents incomplete information
(Teissier 1960). In his formulation dx/dt = aGx,
Huxley assumes constant growth per unit time
if the environmental factors represented by G
are constant (1972, p. 6). In order to determine
the rate of growth or the growth per ecdysis (/ =
1 instar), one must assume that the individuals
are moving along the trajectory described by the
allometric relationships at a constant rate of
growth per ecdysis, an assumption that may be
invalid if the growth per ecdysis responds to en-
vironmental factors (G). This assumption is in-
valid for the linyphiid Linyphia triangularis
Clerck and may be invalid for the lycosid Lycosa
helluo (Turnbull 1962, Uetz et al. 1992). Turn-

In (abdomen volume / abdomen volumeO)

116

THE JOURNAL OF ARACHNOLOGY

0 5 ° 15 20 25

T — — T— — r I T —I I

0 5 10 15 20 25 30 35

Day

Figure 6, —Mean and SD of abdomen volume gain in the laboratory over time. Spiders from all sites were
grouped according to the duration of the intermolt.

HIGGINS-- DEVELOPMENT OF JUVENILE NEPHILA CLAVIPES

117

bull (1962) found that juvenile L. triangularis
raised on quantitatively different diets molted at
distinct, diet-dependent premolt weights. Im-
plicit in his data is that the spiders grew different
amounts in ecdysis, and due to this variation
achieved different sizes at sexual maturity. He
apparently did not observe variation in the num-
ber of juvenile instars. Uetz et al. (1992) found
that sibling groups of L. helluo reared to maturity
on qualitatively different diets varied in age and
size at maturity as well as in juvenile mortality.
It is unknown what developmental parameter
varied with diet. Variation in the number of ju-
venile instars has been found for spiders in di-
verse families (Levi 1970), but the relative im-
portance of variation in growth per ecdysis and
variation in the number of juvenile instars in
determining final adult size is unclear in previous
studies of spiders (Levi 1970, Edgar 1971, Mi-
yashita 1986, Wise 1987). Strikingly, despite the
temporal variation in food levels experienced by
many spiders (Riechert & Luczak 1982, Higgins
& Buskirk 1992), there is as yet no evidence in
spiders of the retrogressive or supemumary molts
characteristic of many holometabolous insects
subjected to poor diets or starvation (e. g., Beck
1972, Nijhout & Williams 1974). However, al-
lometric data collected without longitudinal
studies would not detect such molts if they did
occur.

In N. clavipes, there is surprisingly little vari-
ation in the developmental trajectory described
by growth per ecdysis. The data from field and
laboratory reinforce the previous data, revealing
stronger constraints in growth per ecdysis than
previously reported (Higgins 1992a). Only the
smallest individuals showed variation in growth
per ecdysis, whereas larger juveniles and pen-
ultimate-instar individuals molt at the minimum
premolt weight. I postulate that this shift from
variable to constant growth per ecdysis may re-
flect either the reduction in predation pressure
with increased size of the spiders or changes in
the benefit of delayed ecdysis.

Growth per ecdysis in this spider is highly cor-
related with the premolt abdomen volume, and
the premolt abdomen volume is size-specific and
does not vary among the Mexican populations.
Spiders from the isolated, desert population at
Tehuacan might be expected to differ in their
development, as they experience a short growing
season and low rainfall (Higgins pers. obs.). In
fact, the slope of growth per ecdysis appears to
be lower in Tehuacan (Table 3) and is the same

as that reported for the Texas, USA, population
(Higgins 1992a). However, perhaps due to the
small sample size from Tehuacan, the slope of
growth per ecdysis did not differ statistically from
the other five Mexican sites. Among these five
sites, there is no apparent or statistical difference
in growth per ecdysis, and these developmental
trajectories are equal to those observed in Pan-
ama and in another coastal Veracruz site, Los
Tuxtlas. Although I previously predicted that the
spiders under good conditions (such as warm,
moist coastal Veracruz) might accelerate devel-
opment by delaying each molt, surpassing the
minimum premolt abdomen volume and grow-
ing more at the next ecdysis (Higgins 1 992a), this
was observed only for spiders in the earliest in-
stars. These small spiders are in the most heavily
predated size class (Higgins 1992b) and may be
seeking to escape predation by rapidly increasing
their size (Wilbur & Collins 1973). Growth per
ecdysis declines and is less variable after the spi-
ders reach 0.3 cm TPL.

The shifts from variable to constant growth
per ecdysis could also reflect the allometric re-
lationship of postmolt TPL and premolt abdo-
men volume, described by a concave curve. For
very small spiders, slight changes in premolt ab-
domen volume greatly alter postmolt TPL, so
small delays in molting accompanied by weight
gain will have a large effect on growth. For larger
juveniles, small changes in abdomen volume have
little effect on postmolt TPL, so much longer
delays in molting are required for a significant
change in growth in ecdysis. This is seen in the
very slight increase in growth per molt for the
individual from Nanciyaga that delayed molting
during the common-garden experiment. This re-
duced benefit of delayed ecdysis could also ex-
plain the patterns seen in animals molting to
maturity, which are of sizes found on the as-
ymptote of the curve.

Spiders in the penultimate instar have partially
developed external genitalia and are committed
to becoming sexually mature in the next instar.
Therefore, delaying maturity through additional
juvenile molts is not an option. Males and fe-
males molting to sexual maturity molt at lower
relative premolt abdomen volumes than juve-
niles molting to a juvenile instar. This is pre-
sumably due to the high benefit of reaching sex-
ual maturity early compared to the slight increase
in size that may be achieved by delayed ecdysis
in the last instar. Both male and female repro-
ductive success increase with increased adult size

118

THE JOURNAL OF ARACHNOLOGY

(Christenson & Goist 1979, Vollrath 1980, Hig-
gins 1992a), but the importance of early matu-
ration is known only for females. Females ma-
turing earlier in seasonal environments have
greater likelihood of reproducing and greater
likelihood of producing several egg sacs than do
females maturing later (Higgins pers. obs.). There
is an increase in fecundity per egg sac with in-
creased TPL (Higgins 1992a), but increasing the
number of egg sacs is proportionally more im-
portant as each egg sac may contain over 1000
eggs.

Although N. clavipes exhibits wide variation
in size at sexual maturity, the developmental tra-
jectory of this species described by growth per
ecdysis is apparently highly constrained within
a population and may be genetically determined.
The spiders must achieve a given minimum pre-
molt weight (here presented as premolt abdomen
volume) prior to molting, and most spiders molt
soon after reaching this weight. Premolt weight
is size-dependent and independent of diet either
in the field (current study) or in the laboratory
(Higgins pers. obs.). That it is possible for the
spiders to surpass the minimum premolt weight
is apparent from the growth patterns of the small-
est juveniles, and from occasional observations
of spiders in the laboratory that surpass the min-
imum weight and grow more in the subsequent
molt (present study; Higgins pers. obs.). This may
reflect either phenotypic plasticity or genetic
variability in minimum premolt weight. If phe-
notypic plasticity exists for premolt weight, it is
rarely expressed after the fourth instar. If we as-
sume that the selection has operated to optimize
the developmental trajectory, these data indicate
that, during the later juvenile instars, the costs
of delayed molting outweigh the benefits of in-
creased growth per ecdysis. This may be partic-
ularly true if, as implied by the concave rela-
tionship between premolt weight and postmolt
size, increasingly long delays are required for sig-
nificant differences in growth. However, the
physiological and ecological costs of molting are
not well enough understood to describe this op-
timization function, and the presence of specific
constraints in premolt conditions must be de-
termined first. Only future studies of the physi-
ological and genetic controls of ecdysis in spiders
will clarify this optimization function.

ACKNOWLEDGMENTS

This study would not have been possible with-
out the voluntary laboratory and field assistance

of H. Macias C. and L. Martinez. Discussions
with R. Buskirk, C. Pease and E. Escurra, and
the researchers of the Centro de Ecologia, UN AM,
provided insight during collection and analysis
of the data. G. Uetz and A. Cady made many
helpful suggestions on the manuscript. Many of
the sites were privately owned, and access was
graciously granted by L. Forbes and S. Ayala
(Fortin), C. Rodriguez (Nanciyaga), F. Aguilar
(Arroyo Frio), and the management of Hotel Pla-
ya Escondida. Permission to work at Chamela
was granted by the Instituto de Biologia, UNAM,
and permission to work at Tehuacan cactus gar-
den was granted by the Secretaria de Ecologia y
Desarrollo Urbano. The staff of the Centro de
Ecologia, UNAM, provided logistical aid, and
financial support was provided through the Or-
ganization of American States and a Universidad
Nacional Autonoma de Mexico postdoctoral
grant.

LITERATURE CITED

Beck, S. D. 1 972. Growth and retrogression in larvae
of Trogoderma glabrum (Coleoptera: Dermestidae).
3. Ecdysis and form determination. Ann, Entomol.
Soc. America, 65:1319-1324.

Benton, M. J. & G. W. Uetz. 1986. Variation in life-
history characteristics over a clinal gradient in three
populations of a communal orb-weaving spider.
Oecologia 68:395-399.

Bullock, S. 1986. Climate of Chamela, Jalisco, and
trends in the south coastal region of Mexico. Ar-
chives for Meteorology, Geophysics and Biocli-
matology, Series B 36:297-316.

Caswell, H. 1983. Phenotypic plasticity in life-history
traits: Demographic effects and evolutionary con-
sequences. American ZooL, 23:35-46.

Christenson, T. E. & K. C, Goist. 1979. Costs and
benefits of male-male competition in the orb-weav-
ing spidQv Nephila clavipes. Behav. Ecol. SociobioL,
5:87-92.

Edgar, W. D. 1971. Aspects of the ecological ener-
getics of the wolf spiders Pardosa {Lycosa) lugubris
(Walckenaer). Oecologia, 7:136-154.

Garcia, E. 1973. Modificaciones al Sy sterna de Clas-
sificacion Climatica de K5ppen. Universidad Na-
cional Autonoma de Mexico Press, Mexico City.
Higgins, L. E. 1 990. Variation in foraging investment
during the intermolt and before egg-laying in the
spider iVe/?/2//« clavipes. J. Insect Behav., 3:773-783.
Higgins, L. E. 1992a. Developmental plasticity and
reproduction in the orb-weaving spider Nephila cla-
vipes. J. Arachnol., 20:94-106.

Higgins, L. E. 1992b. Ontogeny of barrier web struc-
ture in Nephila clavipes. J. Insect Behav., 5:635-
655.

Higgins, L. E. in press. Facultative bivoltinism in

HIGGINS DEVELOPMENT OF JUVENILE NEPHILA CLAVIPES

119

Nephila clavipes in coastal Veracruz. In The Ecology
of the Research Station “Los Tuxtlas.” (E. Gon-
zMez, R. Dirzo & R. Vogt, eds.). Univ. Nacional
Autonomo de Mexico Press, Mexico City.

Higgins, L. E. & R. Buskirk. 1992. A trap-building
predator exhibits different tactics for different as-
pects of foraging behavior. Anim. Behav., 44:485-
499.

Huxley, J. 1972. Problems of Relative Growth, 2nd
Ed. Dover Publications, Inc., 312 pp.

Levi, G. 1970. The life cycle of Thomisus onustus
(Thomisidae: Araneae) and outlines for the classi-
fication of lifeaahistories of spiders. J. Zook, Lon-
don, 160:523-536.

Miyashita, T. 1986. Growth, egg production and pop-
ulation density of the spider Nephila clavata in re-
lation to food conditions in the field. Res. Pop, EcoL,
28:135-149.

Nijhout, H. F. & C. W. Williams. 1974. Control of
moulting and metamorphosis in the tobacco horn-
worm, Manduca sexta (L.): Growth of the last-instar
larva and the decision to pupate. J. Exp. Biol., 61:
481-491.

Pease, C. M. & J. J. Bull. 1988. A critique of methods
for measuring life history trade-offs. J. Evol. Biol.,
1:293-303.

Polis, G. A & W. D. Sisson. 1990. Life History. Pp.
161-223, In The Biology of Scorpions. (G. A. Polis,
ed,). Stanford University Press, Stanford.

Reichert, S. E. & J. Luczak. 1982. Spider foraging:
behavioral responses to prey. Pp. 353-386, In Spider
communication: Mechanisms and Ecological Sig-
nificance. (P. N. Witt & J. S. Rovner, eds.). Prince-
ton University Press, Princeton.

Sokal, R. R. & F. J. Rohlf. 1981. Biometry, The
Principles and Practice of Statistics in Biological
Research. 2nd ed. W. H. Freeman and Co. San Fran-
cisco.

Teissier, G. 1960. Relative growth, Pp. 537-560, In
The Physiology of Crustacea. Vol. 1: Metabolism
and Growth. T. H. Waterman, ed.). Academic Press,
New York.

Turnbull, A. L. 1960. The prey of Linyphia trian-
gularis (Clerck) (Araneae: Linyphiidae). Canadian
J. Zook, 38:859-873.

Turnbull, A. L. 1 962. Quantitative studies of the food
of Linyphia triangularis Clerck (Araneae: Linyphi-
idae). Canadian Entom., 94:1233-1249.

Uetz, G. W., J. Bischoff & J. Raver. 1992. Survi-
vorship of wolf spiders (Lycosidae) reared on dif-
ferent diets. J. Arachnok, 20:207-21 1.

Via, S. & R. Lande. 1985. Genotype-environment
interaction and the evolution of phenotypic plastic-
ity. Evolution, 39:505-522.

Vollrath, F. 1980. Male body size and fitness in the
web-building spider Nephila clavipes. Z. Tierpsy-
chok, 53:61-78.

Wilbur, H. M. & J. P. Collins. 1973. Ecological as-
pects of amphibian metamorphosis. Science, 182:
1305-1314.

Wise, D. 1987. Rearing studies with a spider exhib-
iting a variable phenology: no evidence of substan-
tial genetic variation. Bulk British Arachnok Soc.,
7:107-110.

Manuscript received 22 October 1992, revised 29 Jan-
uary 1993.

1993. The Journal of Arachnology 21:120-146

PATHOGENS AND PARASITES OF OPILIONES
(ARTHROPODA: ARACHNIDA)

James C. Cokendolpher': Adjunct Professor, Department of Biology, Midwestern
State University, Wichita Falls, Texas 76308 USA.

ABSTRACT. This is the first paper to review the literature records on all pathogens and parasites of Opiliones
on a global level. These organisms (bacteria, fungi, protozoans, cestodes, trematodes, nematodes, arthropods)
are listed in phylogenetic order along with available information on hosts, collection localities, life history, and
taxonomic history. The opilion hosts are also listed (by their currently accepted names) along with the names
of their known pathogens and parasites. Diagnostic characters and some taxonomic keys are provided for taxa
which are relatively well know. Citations to other available keys are provided. Many new host and distribution
records are provided.

Two fungi [Engyodontium aranearum (Cavara), Torrubiella pulvinata Mains] are removed from the list of
pathogens of opilions and it is suggested that the original hosts were misidentified spiders.

Two new combinations are recorded in the Mermithidae: Agamomermis phalangii (Haldeman 1851), Aga-
momermis truncatula (Rudolphi 1819). Agamermis incerta Steiner in Stipperger 1928 is regarded as a nomen
nudum.

The type locality of the mite Leptus lomani (Oudemans 1903b) is restricted to Corral (39°53'S, 73°25'W),
Valdivia, Chile.

Unlike many arachnids, Opiliones or harvest-
men lack a pumping stomach and therefore they
chew their food and often consume oocysts and
spores. Examination of their feces reveals a va-
riety of chitinous fragments from their arthropod
prey as well as plant pieces. Some saprophytic
fungi and yeast spores can be observed as well
as gametocytes of internal parasites. The fre-
quent grooming of the legs by the harvestmen
may also lead to the ingestion of oocysts and
spores. While ingestion is the common entrance
pathway for some opilion pathogens, fungi infect
their host through penetration of the cuticle. Al-
though gregarines and mites are frequently en-
countered when observing harvestmen, relative-
ly few researchers have documented their
occurrences.

Harvestmen are unique among arthropods by
possessing bilateral exocrine glands which open
onto the dorsal surface of the cephalothorax near
the base of the second pair of legs. These glands
produce a variety of volatile secretions (Ekpa et
al 1984, 1985) that have been generally consid-
ered to be defensive in nature. The glands have
also been proposed to function in a variety of
other behaviors including protection from ex-

'Home address: 2007 29th Street, Lubbock, Texas
79411 USA.

temal pathogens and parasites (see Holmberg
1986, and citations therein). To date, only de-
fense against predators and harvestman aggre-
gation formation have been demonstrated.

While working with a South America har-
vestman, Estable et al. (1955) discovered that
the exocrine gland secretion was a remarkably
effective antibiotic, in vitro, against 18 genera of
bacteria (Gram positive and negative) and pro-
tozoa. Their work revealed that the secretion was
also active when given orally to mice infected
with intestinal parasites. The substance was tol-
erated perfectly by the mice but destroyed giar-
dias, trichomonas and hexamites. The compo-
nents of the secretion were later determined to
be a composed of a variety of quinones (Fieser
& Ardao 1956).

The major components of the exocrine secre-
tions of harvestmen differ between the two sub-
orders, Laniatores and Cyphopalpatores. Minor
components and ratios of components differ
among congeneric species (Ekpa et al. 1985). The
few chemical analyses thus far reported (see Ekpa
et al. 1984, and citations therein) from Lania-
tores reveal a variety of alkylated benzoqui-
nones, phenols, N,N-dimethyl-i(?-phenylethyl-
amine and bomyl esters. Only the Palpatores
section of the suborder Cyphopalpatores has been
chemically investigated. Those analyses reveal

120

COKENDOLPHER^PATHOGENS AND PARASITES OF OPILIONES

121

members of this group secrete short-chained acy-
clic ketones, alcohols and naphthoquinones (see
Ekpa et al. 1985, and citations therein).

Even though harvestmen are abundant in warm
moist situations, few records are available of fun-
gi infecting these animals. Because the major
components of harvestman exocrine secretions
are members of chemical classes known to be
fungicides (see Torgeson 1969; Cole et al. 1975),
it is likely these secretions are used to protect
harvestmen from infection. The use of these se-
cretions in defense and grooming needs further
study.

While there are several world-wide taxonomic
revisions of harvestmen, no similar treatment for
their parasites has been undertaken. This is in
part due to the incorrect view that harvestmen
are not of economic importance. Mounting ev-
idence demonstrates that harvestmen are bene-
ficial and that they consume considerable quan-
tities of pest insects and mites. Because of this
beneficial status, no one has investigated para-
sites for controlling opilions. Experiments in-
volving insect pathogens on harvestmen reveal
opilions are susceptible. Like conventional in-
secticides, insect pathogens and parasites could
have a severe impact on the beneficial harvest-
men.

Many of the records of parasites from har-
vestmen are incomplete. In some cases the host,
but not the parasite, is identified to species. In
other cases, the parasite but not the host is iden-
tified to species. The purpose of this contribution
is to bring together the limited information on
this topic so that a foundation can be built for
future research.

Because of the lack of good characters in some
groups (/. e., Microsporida and juveniles of Mer-
mithidae) collective groups have been named.
Such groups or genera often include species which
probably are not related. This group name is used
simply for “taxonomic convenience” and in-
cludes species not readily placed in known genera
(possibly because a particular life stage is un-
known) and species incertae sedis. Some taxo-
nomically convenient groups also occur at higher
levels in fungi. In fungi, the sexual stage (teleo-
morph of ascomycetes and basidiomycetes) and
their asexual stages (anamorphs or conidial stages)
are sometimes placed in separate genera and
classes. In some cases, two or more ascomycetes
may be identified as having the same form spe-
cies for an anamorph.

When one discusses parasites of opilions, the

topic of phoresy arises. Phoresy is not parasitism
but rather a form of symbiotic relationship in
which the smaller organism associates with the
harvestman in order to obtain transportation.
Phoresy as well as passive transport of fungal
and plant spores will not be examined here.

Superkingdom Prokaryotae
Kingdom Monera
Division Gracilicutes

Family Enterobacteriaceae

Xenorhabdus Thomas & Poinar contains five
described species and other undescribed species
(Akhurst & Boemare 1990). They typically in-
habit nematodes and their host arthropods (in-
sects and arachnids). See under Nematoda
(Rhabditoidea) for further details on this rela-
tionship. Pertinent taxonomic papers are cited
with a review of the taxonomic problems in Ak-
hurst & Boemare (1990).

Xenorhabdus luminescens Thomas & Poinar
(1979) is introduced into the arthropod host by
a Heterorhabditidae [Heterorhabditis bacterio-
phora Poinar]. Poinar & Thomas (1985) dem-
onstrated this bacterium could kill a Phalangi-
idae {Phalangium opilio Linn., reported as P. sp.)
if introduced by the correct nematode.

Xenorhabdus nematophilus (Poinar & Thomas
1965) was originally described in combination
with Achromobacter Bergey, Breed & Murray.
This bacterium is introduced into the arthropod
host by a Steinemematidae [Steinernema car-
pocapsae (Weiser)]. Poinar & Thomas (1985)
demonstrated that this bacterium could kill a
Phalangiidae, Phalangium opilio (reported as P.
sp.), if introduced by the proper nematode.

Superkingdom Eukaryotae
Kingdom Fungi
Division Eumycota

At least one species of fungus successfully kills
a Gonyleptidae (see under Torrubiella gonylep-
ticida and unidentified fungi). Gonyleptoidea are
known to have phenols which are antagonistic
to fungal growth in their exocrine secretions. Ei-
ther T. gonylepticida and another unidentified
fungus from Panama are not retarded by phenols,
or the hosts were unable to produce phenols in
sufficient quantity. The extent of phenol pro-
duction in various gonyleptid genera and its use
in controlling fungi have not been investigated.
Likewise, the effects of age and health of the har-

122

THE JOURNAL OF ARACHNOLOGY

vestman on phenol production have not been
examined.

Unidentified Fungi

Griffiths (1978) illustrated a harvestman [not
identified, but almost certainly Nelima paessleri
(Roewer)] covered by mycelia of soil microfungi.
The fungi are reported not to be pathogenic, but
simply use the harvestman corpse as a substrate.

Mora (1987, fig. 8) reported mortality in adult
males of a Gonyleptidae {Zygopachylus aibom-
arginis Chamberlin) by an unidentified fungus
on Barro Colorado Island, Panama. Males of this
nest-building harvestman eat all fungus appear-
ing in the nest, thus preventing the proliferation
of mycelia. Mora (1987) suggested the males in-
gested the fungi (spores) which eventually killed
them. This is probably incorrect because nearly
all other fungal pathogens of invertebrates infect
their host through the cuticle (Samson et al. 1988).
Ten fatalities were observed from 199 nest-
guarding males examined by Mora.

Subdivision Ascomycotina
Class Pyrenomycetes
Order Clavicipitales
Family Clavicitaceae

Torrubiella Boudier is a genus with primary
host affinities for spiders (Araneae), although
several species are also known from insects, es-
pecially Coccidae (Kobayasi & Shimizu 1982;
Humber & Rombach 1987). Two species have
been reported from harvestmen, but only one
report appears to be valid.

Torrubiella gonylepticida (Moller 1901) was
originally described in combination with Cor-
dyceps Fries. Fetch (1937) transferred the species
to its present combination and redescribed the
species. Moller (1901), when describing the host,
referred to it as a spider (‘Die Spinnen’, not ‘We-
berknechte’). Subsequent authors (Fetch 1937;
Koval 1974; Kobayasi & Shimizu 1982) have
continued to list the only host as a spider. For-
tunately, the specific name refers to the true type
host, a Gonyleptidae harvestman. Moller (1901,
taf. 6, fig. 89) clearly illustrated the gonyleptid
host, but not in sufficient detail to determine to
which genus it belongs. Kobayasi & Shimizu
(1982) reprinted MolleCs illustration and stated
the type locality was Brazil.

Fetch (1937) described the conidial stage as

Spicaria longipes; which is now recognized as
Paecilomyces farinosus (Holm ex S. F. Gray)
(Brown & Smith 1957), Fetch recorded T. gon-
yleptidda and the conidial stage from various
spiders from Trinidad. KovaF (1974) listed the
conidial stage from spiders collected on Mag-
nolia Linne leaves in Russia (formerly Russian
Soviet Federative Socialist Republic, USSR). In
the key to Torrubiella spp. by KovaF, two va-
rieties of r. gonylepticida are differentiated on
the basis of perithecia and ascus lengths. How-
ever, the two taxa should be attributed to another
species: the third taxa in the key should be T.
arachnophila var. pleiopus Mains and the fourth
should be T. arachnophila var. pulchra Mains.

Torrubiella pulvinata Mains (1949) was de-
scribed from “Opilionoidea” collected on Oahu,
Hawaii. Paecilomyces (reported as Spicaria) pul-
vinata (Mains 1949) was the name given to the
conidial stage. Samson (1974) listed S. pulvinata
as a synonym of P, farinosus, thus regarding the
anamorph for both T. gonylepticida and T, pul-
vinata to be the same species. Mains (1949, p.
303) stated “The hosts of this collection are so
severely parasitized that accurate determination
is difficult. They appear to be arachnids belong-
ing to the Opilionoidea.” Because opilions ap-
pear to be absent from the Hawaiian Islands (F.
G. Howarth pers. commun.), the host is more
likely a long-legged, pholcid spider. The setae-
spines on the legs illustrated by Mains (1949, fig.
lA) are long and unlike those on harvestmen.
They are similar to those found on spiders. There
are five adventive cosmopolitan Fholcidae (Ara-
neae) established in the islands that could be
confused as opilions by non-specialists. The op-
ilion host records are considered here to be in-
correct.

Subdivision Deuteromycotina
Class Hyphomycetes

The Hyphomycetes is an artificial class rep-
resenting the asexual states of Ascomycetes and
Basidiomycetes, or fungi for which sexual states
are unknown. Orders and families do not exist
in current classifications of these fungi.

Hymenostilbe Fetch is comprised of seven de-
scribed species. Species are known to infect a
variety of insect hosts, spiders (Mains 1950; Evans
& Samson 1987) and harvestmen. Mains (1950)
stated members of this genus are the conidial
(anamorph) state of Cordyceps spp.; whereas

COKENDOLPHER-™ PATHOGENS AND PARASITES OF OPILIONES

123

Evans & Samson (1987) reported the teleomorph
connection remains unproven. Specific identifi-
cations are best made by consulting the diagnoses
provided by Mains (1950). Hymenostilbe ver-
rucosa Mains (1950) was originally described
from spiders collected in Maine, USA. Other re-
cords are from spiders in England and a “Phal-
angiidae” in England (Leatherdale 1970).

Engyodontium de Hoog is comprised of seven
species, two of which are reported to infect spi-
ders and one on '‘opilionids”. A key to the spe-
cies is provided by Gams et ah 1984. Engyo-
dontium aranearum (Cavara) was originally
described in the genus Sporotrichum Link ex Fries
and was transferred to its present combination
by Gams et al. (1984). A redescription and syn-
onymy are provided by Gams et al. (1984). The
teleomorph state is unknown, but other members
of the genus have a Torruhiella teleomorph. Those
same authors reported hosts as a fly, spiders and
opilions. The specimen in their photograph (fig.
3), as well as those of Samson et al. (1988, pi.
68a,b), superficially resembles opilions, but judg-
ing from the dense placement and morphology
of the setae on the host legs (figs. 3, 68b) they
are not harvestmen. They are more likely pholcid
spiders (Araneae: Pholcidae). The opilion host
record for this species of fungus is considered
herein incorrect.

Nomuraea Maublanc is composed of three de-
scribed species (Ignoffb et al. 1989; Greenstone
et al. 1988). Nomuraea rileyi (Farlow) Samson
is a well-known pathogen of insects. Nomuraea
atypicola (Yasudo) Samson is reported to infect
spiders, harvestmen and insects. Nomuraea ane-
monoides Hocking was originally isolated from
soil and, in high doses in the laboratory, can
cause mortality in insects.

Nomuraea atypicola (Yasuda 1915) was orig-
inally described as a member of the genus Isaria
J. Hill ex E. M. Fries. It was found on an Atyp-
idae spider in Japan. It was transferred to its
present combination by Samson (1974). The te-
leomorph or sexual state is Cordyceps cylindrica
Fetch (1937). Greenstone et al. (1988) reported
the infection of a harvestmen by this fungus un-
der laboratory conditions. The infected Sclero-
somatidae, Leiobunum vittatum (Say), was col-
lected in Missouri, USA. This species of fungus
is commonly found infecting spiders (Green-
stone et al. 1988) and under laboratory condi-
tions was also found to be infective to Lepidop-
tera larvae (Ignoffb et al. 1989).

Subdivision Zygomycotina
Class Zygomycetes
Order Entomophthorales
Family Entomophthoraceae

Pandora Humber (1989) is comprised of 16
species of obligately pathogenic fungi. Hosts in-
clude members of insects and arachnids. A single
species is recorded from opilions. Pandora phal-
angicida (Lagerheim 1898) was originally de-
scribed from Phalangiidae collected in Sweden
as a species of Empusa Cohn (Entomophthora).
Batko (1966) transferred the species to Zoo-
phthora Batko 1964, and placed it in the sub-
genus Pandora Batko 1966. Humber (1989)
placed it in his new genus Pandora. Ellis (1956)
and Leatherdale (1958, 1970) recorded this fun-
gus from a Phalangiidae, Phalangium opilio, in
England.

Entomophaga Batko includes 10 described
species (Humber 1989). All are obligate patho-
gens of insects and arachnids. A single species is
recorded from opilions. A key for identification
of members of this genus is provided by Keller
(1987). Comparisons to original descriptions
(species and citations are listed in Humber 1989)
are required for positive identifications. Ento-
mophaga batkoi (Balazy 1 978) was originally de-
scribed in the genus Entomophthora Fresenius.
Later, Remaudiere & Keller (1980) transferred
the species to Conidiobolus Brefeld (Family An-
cylistaceae), but the current combination with
Entomophaga was made by Keller (1987). BaJ-
azy (1978) described this fungus from harvest-
men collected near Poznan, Poland. Phalangi-
idae [Oligolophus tridens (C. L. Koch)] and rarely
Sclerosomatidae {Leiobunum rotundum Latreille
and Leiobunum blackwalli Meade) were infected.
An epizootic (temporary increase in the inci-
dence of infections) was observed during late
summer.

Keller (1987) reported this species of fungus
was rather common and often caused epizootics
in open woods, along the borders of forests and
hedges. From late July to the middle of Septem-
ber it was collected from Oligolophus tridens in
Switzerland.

Kingdom Animalia
Subkingdom Protozoa

Although seldom reported. Protozoa are com-
mon parasites of Opiliones. To date, all records
of Protozoan parasites of Opiliones are from USA,

124

THE JOURNAL OF ARACHNOLOGY

Europe and India. Their reported absence from
other localities is likely due to lack of study. While
dissecting gonads for anatomical and chromo-
somal studies, I have often observed gregarines
from North and Middle America species (es-
pecially from Phalangiidae and Cosmetidae). El-
lis (1913, p. 280) reported that he was unable to
locate gregarines in the “alimentary canal of per-
haps two hundred Phalangidea” from Michigan
and Colorado. His failure to locate parasites may
have been caused by the time of year he exam-
ined the opilions or possibly the taxa he exam-
ined do not harbor gregarines (these taxa are un-
known, but probably are members of the
Sclerosomatidae: Leiobuninae as they are the
dominate forms in the two mentioned areas).
Only two studies have been published on opilion
hematocytes, one of which resulted in the dis-
covery of a blood parasite.

Phylum Microspora
Class Microsporea
Order Microsporida
Collective Group Microsporidium

Species that cannot be readily placed to genus,
as well as species incertae sedis, are lumped into
Microsporidium sensu Sprague (1977). Micro-
sporidium weiseri (Silhavy 1960) was originally
described in the genus Stempellia Leger & Hesse
(Family Thelohaniidae). Sprague (1977) trans-
ferred this species to its present combination with
Microsporidium because the species did not fit
any of the known genera. This parasite was found
in smear-preparations of hemolymph of a Phal-
angiidae, Opilio parietinus (De Geer). The har-
vestmen was collected in Treble, Czechoslova-
kia. The plasmodium have 2, 4, 8 and 16 spores
and are found in the hemolymph and hemocytes
(plasmatocytes) of its host.

Phylum Apicomplexa

All known Apicomplexa parasites of opilions
are septate eugregarines and as such have several
features in common. Both sexual and asexual
stages occur (gametogony and sporogony), but
merogony is absent. The mode of infection is
ingestion of oocysts. The trophozoites attach to
the lining of the gut and divide to form mero-
zoites and gamonts. Gametocytes are passed in

the feces, and no intermediate host or vector is
needed. Because most species are believed to at-
tach to intestinal epithelial cells, gregarines in
opilions probably are not pathogenic.

The gregarine genera and some species known
from harvestmen can be identified by the follow-
ing taxonomical key. Because some species are
inadequately described (some life-stages un-
known) identifications to species are difficult.
Useful keys or tables of characters are mentioned
under specific genera in the following account.

Class Sporozoasida
Subclass Gregarinasina

Tsurusaki (1986) found gregarines in Sclero-
somatidae harvestmen, Leiobunum manubria-
turn Karsch and Leiobunum globosum Suzuki,
from numerous localities in Japan. He also pro-
vided data on parasitism rates as related to spe-
cies, locality and season. His gregarines have not
been identified.

Hunt (1 979) found numerous gregarines in the
midgut diverticula of Triaenonychidae harvest-
men, Equitius doriae Simon, from southeastern
Australia. His gregarines were never identified.

Mitov reported (pers. commun.) that he had
discovered gregarines in preserved material of
the following harvestmen from Vitosha Moun-
tain and West Rodopy, Bulgaria: Nemastoma-
tidae [Carinostoma ornatum (Hadzi), Parane-
mastoma radewi (Roewer), Pyza bosnica
(Roewer)]; Sclerosomatidae [Leiobunum rume-
licum Silhavy]; Phalangiidae [Lacinius ephippia-
tus (C. L. Koch), L. horridus (Panzer), L. dentiger
(C. L. Koch), Lophopilio palpinalis (Herbst),
Mitopus morio (Fabricius), Odiellus lendli (So-
rensen), Opilio dinaricus Silhavy, O. ruzickai
Silhavy, O. saxatilis (C. L. Koch), Phalangium
opilio, Zacheus anatolicus (Kulczynski), Z. crista
(Brulle)].

Other new records include unidentified greg-
arines from a Phalangiidae, Odiellus pictus Wood,
collected in the West Virginia University Forest,
Chestnut Ridge, Preston County, West Virginia,
USA and an unidentified gregarine from a Scle-
rosomatidae, Leiobunum politum Weed, collect-
ed in Columbus, Ohio, USA. This latter series
is remarkable as the parasites were only discov-
ered after a hundred years of storage.

Key For Identification Of Gregarines Found In Harvestmen

1 a. Oocysts without spines or thickening at poles 2

1 b. Oocysts with spines or thickenings at poles, sometimes at equator and also along edges

(Family Actinocephalidae) 5

COKENDOLPHER-PATHOGENS AND PARASITES OF OPILIONES

125

Subfamily Acanthosporinae

2a. Epimerite simple, spherical; oocysts biconical, with truncate ends, released unchained by simple de-

hiscense of the gametocyst Family Hirmocystidae

Arachnocystis arachnoidea (Devdhar & Gourishankar)
2b. Epimerite complex and varied; oocysts biconical orcylindroconical, united as a string of beads

Family Actinocephalidae 3

Subfamily Actinocephalinae

3a. Epimerite sessile, with short neck having 8-10 simple digitform processes at apex; neck persists more
or less in sporont, but digitform processes (tentacles) disappear; gametocysts dehisce by formation of
hole in wall through which oocysts are extruded in a single thread; oocysts biconical or lemon-
shaped Actinocephalus megabuni Ormieres & Baudoin

3b. Epimerite without digitform process at apex, gametocysts rupture by simple dehiscence 4

4a. Epimerite a large, flattened and fluted disk, oocysts ovoid to biconical, in lateral chains. . . Anthorhynchus

Anthorhynchus longispora Ormieres & Baudoin
Anthorhynchus sophiae (Schneider)

4b. Epimerite a large flattened centrally indented papilla with crenulate border, lost early. Protomerite
with numerous vertical laminations, broadening to an umbrella in the mature sporont, each costule
curved to form a spine pointing backward; oocysts biconical or ovoid, united as a string of beads . . .

Sciadiophora

Sciadiophora caudata (Rossler)
Sciadiophora fissidens (Rossler)
Sciadiophora gagrellula Devdhar & Amoji
Sciadiophora geronowitschi (Johansen)
Sciadiophora phalangii (Leger)
Sciadiophora claviformis Ormieres & Baudoin

5a. Epimerite a conical knob, dentated at the base with a series (about 20) of vertical lamelle. Oocysts

cylindrical with pointed ends, a tuft of spines at each pole Contospora opalniae Devdhar & Amoji

5b. Epimerite simple, globular, without ornamentation 6

6a. Oocysts barrel-shaped, asymmetrical, without terminal tufts, with two equatorial (lateral) thickenings

on longitudinal cordons Doliospora

Doliospora repelini (Leger)
Doliospora troguli (Geus)

6b. Oocysts biconical and symmetrical ... 7

7a. Oocysts with 8 to 10 slender spines at each pole and released in chains of 2 to 3 or more from the

gametocyst Echinoocysta phalangii (Amoji & Devdhar)

7b. Oocysts with slender spines on poles and sides; released unattached from the gametocyst

Cosmetophilus vonones Cokendolpher

Order Eugregarinorida
Suborder Septatorina
Superfamily Gregarinicae
Family Hirmocystidae

Arachnocystis Levine is restricted to Oribatei
mite and opilion hosts. Four species are known,
of which one (the type species) occurs in opilions
(Levine 1979, \9%5). Arachnocystis arachnoidea
(Devdhar & Gourishankar 1971) was originally
described in the genus Sycia Leger (Family Le-
cudinidae). Levine (1979) transferred the species
to his new genus Arachnocystis, where it was
designated the type species. This species was
found in the intestinal ceca of an Assamidae,
Oppalnia sp. (reported as Opalnia sp., see Cok-
endolpher 1991), from Someshwar near Dhar-
war, Karnataka State, India (Devdhar 1 962).

Superfamily Stenophoricae
Family Actinocephalidae
Subfamily Actinocephalinae

Actinocephalus Stein is a relatively large genus
with about 40 described species in insects and
one in an opilion (Levine 1985). Actinocephalus
megabuni Ormieres & Baudoin (1973) was dis-
covered in the intestine of a Phalangiidae, Me-
gabunus diadema (Fabricius). It was collected in
Besse, France.

Anthorhynchus Labbe are found in a termite
(Kalavati & Narasimhamurti 1978) and opilion
hosts. Two species are described from harvest-
men. The type species, by monotypy, is Anthor-
hynchus sophiae Schneider; an opilion parasite.

Anthorhynchus longispora Ormieres & Bau-
doin (1973) was described from the guts of two

126

THE JOURNAL OF ARACHNOLOGY

families of harvestmen: Sclerosomatidae [Leiob-
unum (reported as Liobunum) rotundum] and
Phalangiidae [Mitopus morio, Opilio parietinus,
Platybunus bucephalus (C. L. Koch)]. All har-
vestmen were collected near Besse-en-Chan-
desse, France.

Anthorhynchus sophiae (Schneider 1887) was
originally described in the genus Anthocephalus
Schneider. Because that generic name is preoc-
cupied, Labbe (1899) provided the replacement
name and transferred the species to its present
combination. The original collection of this par-
asite was from the intestine of a Phalangiidae,
Phalangium opilio, captured in Poitiers, France.
This species is also reported by Pfeifer (1956,
from Germany) and Ormieres & Baudoin (1973,
from France) from the Sclerosomatidae [Leiob-
unum blackwalli (reported as L. hassiae Muller),
Leiobunum {=Liobunum) rotundum] and the
Phalangiidae [Lacinius ephippiatus, Mitopus
morio, Oligolophus tridens, Phalangium opilio,
Rilaena (reported as Platybunus) triangularis
(Herbst)].

Sciadiophora Labbe are restricted to opilion
hosts. There are five described species. Devdhar
& Amoji (1978a) provided a table of characters
which is useful in making identifications.

Sciadiophora caudata (Rossler 1882) was orig-
inally described in the genus Stylorhynchus Stein.
Because that name was preoccupied, Ellis (1912)
provided the new generic name Stylocephalus .
Watson Kamm (1922) transferred the species to
its present combination with Sciadiophora, This
species was originally found in the intestines of
Phalangiidae [Mitopus morio, Odiellus (reported
as Odius) spinosus (Bose), Phalangium opilio,
Phalangiidae gen. sp.] from Germany. Ormieres
& Baudoin (1973) reported collections from the
same three hosts from Besse and Tamarissiere,
France.

Sciadiophora fissidens (Rossler 1882) was first
described in the genus Actinocephalus and was
later transferred by Labbe (1899) to Sciadi-
ophora. This species was found in intestines of
Phalangiidae [Lophopilio (reported as Odiellus)
palpinalis, Phalangium opilio, Phalangiidae gen.
sp.] from Germany.

Sciadiophora gagrellula Devdhar & Amoji
(1978b) was described from a Sclerosomatidae,
Gagrellula saddlana (Roewer), which were col-
lected in Dharwar and Kumta, Karnataka State,
India (Devdhar 1962), This gregarine is found
in the intestine and intestinal ceca. Unlike most

described opilion parasites, this species is known
(and illustrated) by all life stages.

Sciadiophora geronowitschi (Johansen 1894)
was described in the genus Actinocephalus and
was transferred to its present combination by
Labbe (1899). This protozoa was discovered in
the intestines of a Phalangiidae, Phalangium op-
ilio, from Russia (formerly Russian Soviet Fed-
erative Socialist Republic, USSR).

Sciadiophora phalangii (Leger 1897) was first
described in combination with Lycosella Leger.
It was the type and only species in the genus.
Because Lycosella was preoccupied, Labbe (1899)
proposed the new name Sciadiophora, with S.
phalangiihtm% the type-species. This species has
been recorded, redescribed and illustrated by nu-
merous authors (Minchin 1903; Wellmer 1911;
Ellis 1913; Watson Kamm 1922; Stipperger 1928;
Pfeifer 1956; Silhavy 1961; Geus 1969; Kudo
1971; Ormieres & Baudoin 1973). Two families
of harvestmen have been reported as hosts (all
European): Sclerosomatidae [Leiobunum rotun-
dum] and Phalangiidae [Lacinius dentiger, Mi-
topus morio (reported as Opilio grossipes), Opilio
parietinus, Phalangium sp., Phalangium opilio
(reported as Phalangium cornutum), Platybunus
bucephalus, Platybunus pinetorum (C. L. Koch),
Rileana Platybunus) triangularis]. The origi-
nal collection of this species was recorded from
two hosts, Phalangium crassum Dufour and P.
cornutum from Vallee de la Loire (where it was
rare) and Provence (where it was common),
France, As noted above, the latter species is now
known as P. opilio but the identification of the
former species is uncertain (Roewer 1923), Re-
cords of this parasite are from France, Austria,
Germany and Czechoslovakia.

Sciadiophora claviformis Ormieres & Baudoin
(1973) was found in the intestine of a Phalan-
giidae, Mitopus sp. (based on the collection lo-
cality the species is probably M. morio). The
collection locality was Vallee de Chaudefour,
France.

Subfamily Acanthosporinae

Contospora Devdhar & Amoji are known only
from opilions. The single species, Contospora
opalniae Devdhar & Amoji (1978a), was de-
scribed from the midgut and cecum of an As-
samidae, Oppalnia sp. (reported as Opalnia sp.,
see Cokendolpher 1991). It is known from So-
meshwar and Kalghatgi, Dharwar District, India
(Devdhar 1962).

COKENDOLPHER-= PATHOGENS AND PARASITES OF OPILIONES

127

Cosmetophilus Cokendolpher is a monotypic

genus restricted to an opilion host. It is the only
genus of gregarines positively identified from
harvestmen in the New World. Cosmetophilus
vonones Cokendolpher (1991) was described from
the intestine and intestinal ceca of the cosmetid
harvestman Vonones sayi (Simon) from Texas,
USA. Other samples presumably of this species
were recorded from the same host collected in
Tennessee, USA. Unlike most described opilion
gregarines, this species is known (and illustrated)
by all life stages.

Doliospora Ormieres & Baudoin ( 1 969) are re-
stricted to opilion hosts. There are two described
species. Doliospora repelini (Leger 1897) was
originally described in the genus Acanthospora
Leger (1892). It was designated the type-species
of the new genus, Doliospora, by Ormieres &
Baudoin (1969), This species has been reported
by Leger (1897) and Ormieres & Baudoin (1973)
from France from the intestines of the Sclero-
somatidae [Leiobunum {=^Liobunum) rotundum]
and the Phalangiidae [Oligolophus tridens, Opilio
parietinus, Megabunus diadema, Mitopus morio
{=0. grossipes), Phalangium opilio {=P. cornu-
turn - type host), Platybunus bucephalus].

Doliospora troguli (Geus 1969) was originally
described in the genus Acanthospora and trans-
ferred to its current combination by Levine
(1980). It was found in the intestine of a Tro-
guloidae, Trogulus tricarinatus (Linne), in Raths-
berg, Germany. Neither its gametocysts nor its
oocysts are described.

Echinoocysta Levine (1984) is composed of a
single species that is restricted to an opilion host.
Echinoocysta phalangii (Amoji & Devdhar 1979)
was originally described in the genus E chinos-
pora Amoji & Devdhar (1979). Because that ge-
nus was preoccupied, Levine (1984) proposed
the new name Echinoocysta and transferred the
species to its current combination. This protozoa
is found in the intestine and intestinal ceca of an
Assamidae, Oppalnia sp. (reported as Opalnia
sp., see Cokendolpher 1991) from Someshwar,
near Dharwad, Karnataka State, India.

Subkingdom Eumetazoa
Phylum Platyhelminthes
Class Cestoda (Cestoidea)

Order Cyclophyllidea
Subclass Eucestoda
Family Hymenolepididae

Pseudhymenolepis Joyeux & Baer (1935) is
monotypic. Pseudhymenolepis redonica Joyeux

& Baer (1935) was described from the shrew Cro-
cidura russula Herm. (Insectivora: Soricidae). A
flea, Ctenophtalmus arvernus (Hystrichopsylli-
dae), is known to be an intermediate host of this
cestode. Gabrion (1977) reported finding cysti-
cercoides in a Phalangiidae, Phalangium opilio,
collected during early July. The harvestman was
found in a shrew nest (previously named species).
Shrews in the general area of the nest revealed
proglotids of this cestode as well. It has been
proposed that P. opilio will serve as the inter-
mediate host when fleas are absent.

Class Trematoda
Order Digenea

Creplin (1846, p. 156) reported finding an un-
identified larval fluke in a Phalangiidae {Phal-
angium opilio). The fluke was listed as ''Disto-
mum Cystidicola Cr. sp. n.” As no illustration
or description was provided this name must be
considered a nomen nudum.

Family Dicrocoeliidae

Brachylecithum Strom was originally de-
scribed as a subgenus of Lypersomum Looss.
Adult flukes of this genus are found in the liver
and gall bladder of birds and rarely in mammals.
Data are available on the life cycles of six (in-
complete data for five species) Brachylecithum
spp. (see Carney 1970, 1972). In a typical life-
cycle the eggs are passed in the feces of the de-
finitive host, a bird or mammal. The eggs are
eaten by a terrestrial snail, the intermediate mol-
luscan host, where they develop into miracidia
and sporocysts. The cecariae emerge from the
snail as a slime ball and are eaten by a second
intermediate host (usually an arthropod). The
cecariae encyst in the arthropod hemocoel and
infect the vertebrate host upon eating the inter-
mediate host. In some arthropod hosts, the me-
tacercariae lodge in or near the host brain causing
behavioral and morphological changes (Hohorst
& Graefe 1961; Carney 1969). These changes
appear to increase the chances of predation upon
the arthropod host (Carney 1969).

Brachylecithum sp. cysts and metacercaires
were reported from a Phalangiidae, Phalangium
opilio, by Gabrion & Ormieres (1973). The trem-
atodes were found in the muscles and adipose
tissue of the body. The infected harvestman was
collected in Sete and Montpellier, France. Bra-
chylecithum adults are known from Passeri-
formes birds in the south of France. Because
Brachlecithum spp. appear to be relatively host-

128

THE JOURNAL OF ARACHNOLOGY

specific in the arthropod stage of development
(Carney 1970), the record in Phalangium is prob-
ably of an undescribed species.

Phylum Nematoda (Nemata)
Unidentified Class

Laniatores (Triaenonychidae and/or Synthe-
tonychidae) from New Zealand are reported by
Forster (1 954) to be infested by unspecified nem-
atodes. Dr. V. Tood (in Sankey 1949a) recorded
nematodes from Rilaena {=Platybunus) trian-
gularis.

Class Secementia
Subclass Rhabditia
Order Rhabditida
Suborder Rhabditin
Superfamily Rhabditoidea
Family unidentified

Pfeifer (1956) reported “rhabditid” nematodes
from Phalangiidae {Lacinius horridus and Phal-
angium opilio) that were captured in Berlin, Ger-
many.

Family Steinemematidae

Steinernema Travassos is comprised of nine
distinct species (Poinar 1990). Until recently,
most species were referred to Neoaplectana
Steiner. Others referred to Neoaplectana are ei-
ther synonyms, misidentified or insufficiently de-
scribed (Poinar & Welch 1981). Keys and other
descriptive data needed for identification of the
various species can be located in Poinar (1990).
Only one species is known from a phalangid host.
All species thus far discovered carry a single spe-
cies of symbiotic bacterium in the alimentary
tract of the third-stage juvenile. The infective
stage nematodes occur on soil and have the abil-
ity to locate and enter arthropod hosts. To reach
the hemolymph of the host, the nematodes enter
via a natural opening and then penetrate through
the gut or tracheal walls. Once inside the host,
the nematode releases its associated bacterium
which kills the host within 48 hours. The nem-
atodes mature into males and females inside the
arthropod and the females release eggs within
the cadaver (Poinar 1983).

Steinernema carpocapsae (Weiser 1955) was
originally described in combination with Neo-
aplectana from codling moth larvae collected in
Czechoslovakia. Poinar & Thomas (1985) dem-
onstrated this nematode could infect and suc-
cessfully reproduce in a Phalangiidae, Phalan-
gium opilio (reported as P. sp.). Its symbiotic

bacterium Xenorhabdus nematophilus (Poinar &
Thomas) killed the above mentioned arthropod
host. This nematode has a wide host range of
insects and arachnids (Poinar 1979; Poinar &
Thomas 1985; Poinar et al. 1985). A thorough
description and review of this nematode and its
relationship with X. nematophilus are provided
by Poinar (1979).

Family Heterorhabditidae

Heterorhabditis Poinar is the only genus in the
family. It is known by three described species
(Poinar & Welch 1981; Poinar 1990), one of
which is known to infect harvestmen. Keys to
the infective juveniles of the three species is found
in Poinar (1990). Adults are identified by elec-
trophoretic analysis of enzymes (Akhurst 1987),
DNA fingerprinting and morphology (Poinar et
al. 1987). The mode of entry into the host and
general life cycle follow that listed under Stei-
nernema, except that Heterorhabditis have a het-
erogenic life cycle. Maturing females can either
be hermaphroditic or amphimictic. The first her-
maphroditic generation is usually followed by
one or more amphimictic generations in a single
cadaver. Juveniles of Heterorhabditis can enter
host via natural openings, or in smaller, more
fragile host by breaking the cuticle with a dorsal
(and sometimes ventral) hook.

Heterorhabditis bacteriophora Poinar 1975, is
a well-known insect parasite. Considerable lit-
erature on this species is listed under a synonym
Heterorhabditis heliothidis (Khan et al. 1976);
which was originally described in combination
with the new genus Chromonema (Khan et al.
1976). Poinar & Thomas (1985) demonstrated
this nematode could infect and successfully re-
produce in the Phalangiidae Phalangium opilio
(reported as P. sp.). Its symbiotic bacterium Xe-
norhabdus luminescens killed the above men-
tioned host. This nematode has a wide host range
of insects and arachnids (Poinar 1979; Poinar &
Thomas 1985; Poinar et al. 1985). A thorough
review of this nematode is provided by Poinar
(1979).

Class Adenophorea
Subclass Enoplia
Order Mermithida
Superfamily Mermithoidea
Family Mermithidae

All known mermithid records from harvest-
men are based on juvenile nematodes. Conse-
quently, none can be accurately assigned to a

COKENDOLPHER~= PATHOGENS AND PARASITES OF OPILIONES

129

genus (see below under Agamomermis). Re-
searchers fortunate enough to obtain adult ma-
terial should consult the key provided by Poinar
(1977).

Matthiesen (1974) reported the discovery of a
Gonyleptidae {Gonyleptes fragilis Mello-Leitao)
which was infested by a internal parasite. Pre-
liminary observations through the harvestman’s
body (the parasite was apparently not dissected
from the host) suggested the parasite to be either
a Nematomorpha or mermithid. Because there
are no other recorded cases of the former at-
tacking Opiliones, I assume the parasite was a
juvenile mermithid.

Unknown species were reported by Poinar
(1985) from a Sclerosomatidae [Togwoteeus (re-
ported as Homolophus) biceps (Thorell) from
western Canada], a Cosmetidae [Paecilaemana
quadripuncta Goodnight & Goodnight from
Costa Rica] and a Protolophidae [Protolophus sp.
from the southwestern USA]. Pfeifer (1956) also
reported an unknown species from a Phalangi-
idae, Phalangium opilio, from Berlin, Germany.

Tsurusaki (1986) found unidentified mermi-
thids in two species of Sclerosomatidae {Leiob-
unum globosum, Leiobunum manubriatum) in
Japan.

Mitov reported (pers. commun.) that he had
discovered larval mermithids in preserved ma-
terial of the following harvestmen from Vitosha
Mountain and West Rodopy, Bulgaria: Nemas-
tomatidae {Paranemastoma radewi), Phalangi-
idae {Lacinius ephippiatus, L. horridus, L. den-
tiger, Lophopilio palpinalis, Mitopus morio,
Phalangium opilio, Zacheus crista).

Agamomermis Stiles is a collective group
erected to receive species which were described
from larvae (which lack meaningful taxonomic
characters) [see Poinar & Welch (1981)]. When
diagnosing Agamomermis, Stiles (1903) stated
the group was artificial and therefore should have
no type species. All of the mermithids thus far
recorded from harvestmen are considered incer-
tae sedis and therefore those species that were
originally described from harvestmen should be
transferred to Agamomermis. This action was
indicated but not formally performed by Poinar
(1985).

Agamomermis phaiangii (Haldeman 1851),
new combination, was originally described in
combination with Filaria Mueller from a Phal-
angiidae, Phalangium opilio (reported as P. cor-
nutum). Agamomermis truncatula Rudolphi

(1819), new combination, was originally de-
scribed in combination with Filaria. Steiner
(1922) transferred the species to Mermis. The
original specimens were from the abdomens of
Phalangiidae, Phalangium opilio and Opilio (re-
ported as Phaiangii cornuti and Opilionis). Dies-
ing (1851) listed the species as ""Gordius trunc-
tulus Diesing,” but it is unclear if he had
additional material

Agamermis incerta was reported by Stipperger
(1928) from Mitopus morio collected in Tirol,
Austria. Stipperger (1928, p. 60) stated that he
had sent the specimen to Dr. G. Steiner for iden-
tification and that he had received an identifi-
cation as ""Agamermis incerta n. sp.” Pfeifer
(1956) and Poinar (1985) referred to this species
as Agamermis incerta (Steiner), indicating that it
had been described in some other genus. I have
been unable to locate the description of this spe-
cies (in combination with Agamermis Cobb,
Steiner & Christie; Hexamermis Steiner, or Mer-
mis Dujardin) in Zoological Record (191 8-1940)
and presume it is a nomen nudum. Apparently,
Poinar (1985) also was unable to locate Steiner’s
description of incerta (in combination with Aga-
mermis or otherwise) from a spider, as this spe-
cies of mermithid does not occur in his table
except associated with Stipperger’s 1928 paper.

Hexamermis sp., incertae sedis, juveniles were
reported (Unzicker & Rotramel 1970) from an
immature Phalangiidae harvestman {Opilio sp.-
only species in region is O. parietinus) from Il-
linois, USA. Because of the uncertainty of the
identification, this species is best retained as Aga-
momermis sp.

Mermis sp. was reported by Kastner (1928).
He stated Julius Riihm of Nemberg saw a “Mer-
mis” emerge from a “Phalangiidaen.” This rec-
ord was later cited as Phalangiidae, Opilio sp.,
by Poinar (1985). The only paper by Riihm cited
by Kastner was published in 1926 and contained
no mention of a Mermis. Apparently, there has
been some miscommunication regarding this
record. Probably, Ruhm verbally communicated
this observation to Kastner and used “Mermis”
as a general term for a mermithid nematode.
Furthermore, because the record is of a post-
parasitic juvenile, the record is correctly attrib-
uted to Agamomermis sp.

Phylum Nematomorpha

Hairworms are free-living as adults and par-
asitic as juveniles in insects, spiders and crus-
taceans. Some early records of mermithid nem-

130

THE JOURNAL OF ARACHNOLOGY

atodes were incorrectly assigned to two genera
{Filaria and Gordius Linne) belonging to this
phylum. Those species are listed herein as Aga~
momermis spp. (see this group under the Nem-
atoda).

Phylum Arthropoda
Class Insecta
Order Diptera
Suborder Cyclorrhapha

The Cyclorrhapha is comprised of many fam-
ilies of flies, each having a different life cycle -
most are not parasitoides. Without knowing the
identification of the fly, little can be written other
than a notice of the single reported occurrence.

Soares (1945) reported the discovery of a fly
pupa inside an adult of a Gonyleptidae {Disco-
cyrtus invalidus Piza). The gonyleptid was col-
lected at Porto Cabral, Estado de Sao Paulo, Bra-
zil.

Suborder Nematocera
Family Ceratopogonidae

Tsurusaki (pers. commun.) reported finding
adult flies of this family, subfamily Forcipomyi-
inae, settled on the leg femora of Nelima nigri-
coxa and Gagrellula ferruginea in Japan. When
he disturbed the flies they would hover around
the host. He suspected they were sucking blood
from the harvestmen.

Order Hymenoptera
Family Chalcidae

Laniatores (Triaenonychidae and/or Synthe-
tonychidae) from New Zealand are reported by
Forster (1954) to be infested by chalcid wasps.
No specific identifications were provided.

Family Pompilidae

Anoplius Dufour is a large, diverse group of
wasps which prey almost exclusively upon spi-
ders (Evans & Yoshimoto 1962). The female
wasps sting and paralyze spiders which are in-
dividually entombed with a wasp egg. The wasp
young will then devour the spider as it grows.
Some of the species permanently paralyze their
prey while others only paralyze them temporar-
ily. Some adult wasps feed upon spiders while
others feed upon nectar of flowers. Only a single
species has been recorded to prey upon a har-
vestman.

Anoplius (Pompilinus) marginatus (Say 1824)
is found over most of temperate North America
east of the Rocky Mountains. It is often common

and unlike other pompilids is not very selective
as to the prey it takes. Prey items include at least
22 species of spiders from seven different fami-
lies (Evans & Yoshimoto 1962). Evans (1948)
recorded a female wasp taking a Phalangiidae
{Odiellus pictus) in East Hartford, Connecticut,
USA. Because the harvestman was taken away
from the wasp before it dug a nest it is uncertain
if it would use the O. pictus to provision the nest.
Evans (pers. commun.) recalled that the wasp
was captured while it was dragging the opilion
across the ground but he could not determine
whether the opilion was used in provisioning the
wasp nest. Pompilids often take prey and then
abandon it, sometimes after feeding on it.

Class Arachnida
Order Acarina
Suborder Prostigmata

Mites known to be parasitic on harvestmen
belong to the families Thrombidiidae and Ery-
thraeidae. Only the larval forms are parasitic
(protelean parasites) while the nymphs and adults
are predaceous on small insects. Because the lar-
val and post-larval stages of these two families
are heteromorphic, systematists have long used
different scientific names for each (Southcott
1961). Only after the larval and post-larval stages
are associated by rearings can any meaningful
classifications be constructed.

Laniatores (Triaenonychidae and/or Synthe-
tonychidae) from New Zealand are reported by
Forster (1954) to be often heavily infested by
mites. Hunt (1979) found a species of parasitic
mite on Triaenonychidae harvestman, Equitius
doriae Simon, from southeastern Australia. Bur-
ton & Burton (1984, pp. 218 and 226) published
a color photograph of a harvestman with nu-
merous parasitic red mites. The harvestman is
clearly Mitopus morio. The mites are probably
members of the genus Leptus, although this can
not be stated for certain. Elliott & Reddell (1987)
reported that many of the Leiobunum townsendi
occurring in caves in central Texas carried red
chiggers on their legs. The mites are probably not
chiggers but more likely the larvae of Leptus.
Eaton (1985) stated in a report on some har-
vestmen from a cave in southeastern. New Mex-
ico that the “The [harvestmen] spiders all had
one or more small, shiny, bright red, oval bumps
on their legs which appeared to be some kind of
parasite.” These parasites are likewise probably
Leptus sp. and the hosts are almost certainly L.
townsendi.

COKENDOLPHER== PATHOGENS AND PARASITES OF OPILIONES

131

Other unidentified mites from my collection
include: Cosmetidae {Vonones sayi) from Sam
Houston National Forest, Lake Stubblefield,
Walker County, and Lake Kirby, Taylor County,
Texas, USA (mites found on dorsa of abdomens);
Phalangiidae [Zacheus hebraicus (Simon)] from
Beith Shemesh, Israel (mite from tibia I); Scle-
rosomatidae: Leiobuninae {Leiobunum townsen-
di Weed) from near Cloudcroft, New Mexico,
USA; [Leiobunum ventricosum (Wood)], West
Virginia University Forest, Chestnut Ridge,
Monongalia County, West Virginia, USA.; {Neb
ima paessleri Roewer) from Moose Creek Re-
search Station, Idaho, USA; Sclerosomatidae:
Gagrellinae {Gagreiiopsis nodulifera Sato & Su-
zuki) from Mt. Daisen, Tottori Prefi, Japan (mite
found on dorsum of abdomen); {Trachyrhinus
rectipalpus Cokendolpher) from 2 km W. of
Cue vitas, Starr County and Buffalo Gap, Taylor
County, Texas, USA; {Prionostemma panama
Goodnight & Goodnight) from Orillas de Rio
Mata Ahogado el Vallo de Anton, Prov. Code,
Panama (mite was found on the abdomen); Scle-
rosomatidae: Metopilio Group {Globipes sp.)
from near Cloudcroft, New Mexico, USA.

Family Thrombidiidae

Known as the velvet mites, adults of this fam-
ily are among the largest and most conspicuous

families of mites.

Allothrombium Berlese is a small genus with
seven described species. Its members are para-
sitic on harvestmen, spiders, several orders of
insects and isopods (Welboum 1983). Megnin
(1876) described the larva of a mite reared from
opilions. He identified the mite as either Trom~
bidium fuliginosum Herman or Trombidium
gymnopterorum Berlese. Based on the structure
of the tarsal claws, Southcott (1961) identified
Megnin’s specimen (which was illustrated) as an
Allothrombium sp.

Allothrombium chanaanense Feider (1 977) was
described from an “Opilionida” from Jerusalem,
Israel. This species of mite is only known from
the larval forms. Host records also include in-
sects: an Acrididae [Prionsosthenus galericulatus
(Stal)] and an unidentified Aphidae from Israel
(Feider 1977).

Allothrombium neapolitum Oudemans (1910a)
was described from a Phalangidae {Phalangium
sp.) from Portici, Campania, southern Italy.
Oudemans (1913) redescribed and illustrated this
species. Specimens identified from my collection
were found attached to the edges of the abdom-

inal spiracles of a Phalangiidae, Zacheus crista
Brulle. The collections were from Lindos, Rho-
dos.

Trombidium Fabricius is a relatively large ge-
nus of conspicuous mites with about 20 species.
Member species have been observed and re-
corded since the first record in about 300 B.C.
by Apollodorus. About half of the described spe-
cies are known only by adults. Juveniles are
known to feed on numerous orders of insects as
well as spiders, a pseudoscorpion and harvest-
men (Welboum 1983).

Yokogawa (1940) described and illustrated a
Sclerosomatidae {Nelima sp.) parasitized by a
mite from Japan. The mite was identified as a
"^Trombidinium'^ [5/c]. Trombidium hungarium
Kobulej (1957) is recorded from a Phalangiidae
{Egaenus convexus Koch) from Matraszentimre,
Hungary. Both the larva and the nymph of this
species were described by Kobulej (1957).

Family Erythraeidae

The first record of a harvestman parasite was
probably an erythraeid mite. Lister (1678) re-
ported scarlet-colored “bugs” attached and feed-
ing from what is now known to be Phalangiidae
Phalangium opilio in England. Sankey (1949b)
reported mites of this family from numerous spe-
cies of harvestmen collected in England. Specif-
ically, he recorded hosts as: Sclerosomatidae
[Leiobunum blackwalli, L. rotundum, Nelima
silvatica (Simon)] and Phalangiidae [Mitopus
morio, Oligolophus hansenii (Kraepelin), O. tri-
dens, Opilio parietinus. Par oligolophus agrestis
(Meade), Phalangium opilio, Rilaena triangu-
laris {^Platybunus triangularis)]. Sankey (1 949a)
stated that he had records of 10 species of har-
vestmen (presumably those listed above) being
used as carriers by the larvae of Erythraeus phal-
angioides (De Geer 1778). This identification is
probably incorrect as this species is not otherwise
known to feed on harvestmen and there is some
question regarding the true identity of larval E.
phalangiodes (see Southcott 1961). Possibly,
Sankey confused the names phalangii and phal-
angiodes’, both of which were described by De
Geer. Martinez Crespo & Morales Soto (1979)
reported that mites of the family Erythraeidae
were parasitic on Opiliones from Mexico.

There are over 50 species of Charletonia
Oudemans described as larvae and 22 species
described originally as adults. Species are re-
corded from every continent except Antarctica
(Southcott 1991). Larvae of two species are par-

132

THE JOURNAL OF ARACHNOLOGY

asitic on harvestmen (Kawashima 1961; South-
cott 1961, 1965, 1991). The other species are
common parasites as juveniles on locusts (Acrid-
idae) and less commonly encountered on jump-
ing plant lice (Psyllidae: Homoptera), true bugs
(Lygaeidae and Miridae: Hemiptera), wasps
(Braconidae: Hymenoptera), Lepidoptera, drag-
onflies (Libellulidae: Odonata), flies (Tabanidae,
Dolichopodidae and Bombylidae: Diptera),
mantis (Mantidae: Mantodea), walking sticks
(Phasmatidae: Phasmida), katydids (Tettigoni-
idae: Orthoptera), beetles (Curculionidae, Me-
lyridae, Tenebrionidae: Coleoptera), mites (Er-
ythraeidae: Acarina) and spiders (Theridiidae,
Philodromidae: Araneae). Keys to the species are
provided by Southcott (1991).

Charletonia enghoffi Southcott ( 1 99 1 ) is known
by four larvae recovered from the dorsum and
femora of the Phalangiidae, Bunochelis canar-
iana (Strand). The species were obtained in Feb-
ruary in Teno Barranco de las Cuevas, Tenerife,
Canary Islands.

Charletonia southcotti Kawashima (1961) is
recorded from a Sclerosomatidae, Metagagrella
tenuipes (L. Koch) (reported as Gagrella japonica
Roewer), that was collected at the seashore of
Kasumigaoka, Fukuoka City, Fukuoka-Prefec-
ture, Kyushu, Japan. This species of mite is only
known from the single collection on 12 July.
Thirty-five mites were recovered from 20 opi-
lions. It is known only by the larval stage, which
was redescribed by Southcott (1965).

Leptus Latreille is a large genus and its mem-
bers are widespread. Over 60 Leptus spp. have
been described from larvae. Many adults have
also been described, but only in a few cases have
correlations been made between larval and post-
larval forms. Only in a single case is a species
described from larval and all post-larval stages
(Welboum & Jennings 1991). Many species re-
main undescribed. Member species are generally
parasitic on spiders, scorpions, harvestmen, dip-
lopods, Collembola, and insects. Many of the
early reports and even some more recent are sus-
pect as the true identity of the mites identified
is uncertain. Southcott (1961, 1991, 1992) re-
viewed some of the problems regarding the Eu-
ropean mites (phalangii, ignotus, nemorum, coc-
cineus) which had been referred to various genera.
Southcott (1989) provided a key to the parasitic
larval forms that were recognizable (most early
descriptions are inadequate to recognize the spe-
cies) in the New World. Welboum & Jennings
(1991) added a new species (from Lepidoptera

host) from the USA and provided some addi-
tional comments on members of the genus.
Southcott (1992) described numerous new spe-
cies and provided a key to the taxa from North
America and Europe. He also resolved the iden-
tity of L. ignotus and found the type species,
Acarus phalangii, to be an illegitimate name. Ka-
washima (1958) and Haitlinger (1990) provided
keys to the parasitic larval forms from Japan and
northern Africa, respectively.

Leptus spp. have been reported from a variety
of hosts and localities. The mode of attachment
was described in Leptus sp, on two species of
Phalangiidae {Mitopus naorio, Phalangium opi-
Ho) by Abro (1988). Abro (1991) described un-
successful parasitic attachments of larval Leptus
spp. to the ocular tubercle of Phalangium. Evans
et al. (1961, fig. 211) illustrated a Phalangiidae
{Mitopus morio) infested with Leptus sp. larvae
from the British Isles. Welboum (1983) reported
Leptus spp. from unidentified Opiliones collect-
ed in Ohio and Arkansas.

Robert G. Holmberg reported (pers. com-
mun.) that he had found 39 vials of harvestmen
infested with 78 mites, all of which have been
identified as Leptus spp. by 1. M. Smith (Bio-
systematic Research Center, Ottawa, Canada).
Dr. Holmberg’s collections were from: ^^Tog-
woteeus biceps from Canada and the USA, Mi-
topus morio from England, Odiellus pictus from
Canada, Oligolophus tridens from Canada, Par-
oligolophus agrestis from Wales, Phalangium op-
ilio from Canada and England, and Leiobunum
townsendi from U.S.A.”

Mullen (1988) reported “opilionids” com-
monly serve as the host to Leptus mites. Savory
(1938) recorded Belaustium [sic] (Ritteria) ne-
morum (Koch) from harvestmen in England.
This observation was later cited on several oc-
casions in general works about arachnids by Sa-
vory and Cloudsley-Thompson. The original ob-
servations were most likely based on a
misidentification and probably were represen-
tatives of the genus Leptus. Not only is Leptus
widely known as a harvestman parasite, but
members of the Balaustiinae are generally con-
sidered not to be parasitic on arthropods (South-
cott 1961).

Cox et al. (1921) found an immature Ery-
thaeus [sic] sp. on a “phalangid” in California,
USA. This mite is probably a Leptus sp. Mites
from almost every genus of the Erythraeidae have
been misidentified as Erythraeus Latreille (see
Southcott 1961).

COKENDOLPHER-PATHOGENS AND PARASITES OF OPILIONES

133

Leptus phalangii (De Geer 1778) was origi-
nally described in combination with Acarus
Linne. When erecting the genus Leptus, Latreille
(1796) designated (by monotypy) Acarus phaL
angii as the type species. The type specimens
were from a Phalangiidae {Phalangium sp.) col-
lected in Sweden. Apparently, none of De Geer’s
specimens were presented. There has been con-
siderable confusion over the identity of this spe-
cies. Furthermore, as noted by Southcott (1992)
the specific name is not available under the In-
ternational Code of Zoological Nomenclature as
De Geer did not treat it consistently as a bino-
men. Only in a few cases can specimens that have
been previously referred to in the literature as
Leptus phalangii be assigned currently recog-
nized names.

Leptus ignotus (Oudemans 1903a) was origi-
nally described in combination with Erythraeus.
The type locality is Borkum, Holland. Southcott
(1991) redescribed the species and limited the
species diagnosis to specimens which had not
been recorded from opilion hosts. Therefore all
records of this species from opilion host can be
assumed to be misidentified and are referred to
in Table 1 as Leptus sp. Evans et aL (1961) re-
corded a mite (identified as L. ignotus) parasitic
on Opilio parietinus in the British Isles. Other
records are also from the Phalangiidae: Mitopus
morio from Tirol, Austria (Stipperger 1928) and
Bulgaria (Beron 1975); Opilio ruzickai from Bul-
garia (Beron 1975); and Phalangium opilio, R.
triangularis, Lophopilio (reported as Odiellus)
palpinalis fmm Poland (Haitlinger 1987). South-
cott (1992) suggested that some of the specimens
identified by Beron and Haitlinger were Leptus
holmiae Southcott.

Mites reported as Leptus phalangii have been
reported by Pfeifer (1956) and Evans (1910) on
Phalangiidae (Phalangium opilio) in Berlin, Ger-
many, and Midlothian, Scotland, respectively.
Spoek (1964) also recorded this mite to be par-
asitic on harvestmen from the Netherlands. None
of these mites can be accurately identified at pres-
ent and are best referred in Table 1 as Leptus sp.
Meade (1855) reported “harvest-men” from En-
gland were frequently infested by a bright red
parasitic mite, which he identified as TrombL
dium phalangii {^Leptus phalangii). He further
specified that the mite occurred on Leiohunum
rotundum.

Sellnick (1940) recorded both Achorolophus
ignotus and Leptus (reported elsewhere in the
paper as Erythraeus) phalangioides (De Geer)

from Phalangiidae (M. morio) on Iceland. He not
only recorded both mites from a single species
of harvestmen, but in two cases he recorded what
he felt were these species from individual har-
vestmen. Although these cannot be identified with
certainty at this time, they are probably L. hoP
miae. Until specimens can be studied I am re-
ferring to them in Table 1 as Leptus sp.

Numerous mites from opilions in my collec-
tion represent new records and include: Sdero-
somatidae: Eumesosoma roeweri (Goodnight &
Goodnight) from Alma, Nebraska, USA.; Krusa
sp. from 10 mi. W. Aquismon, San Luis Potosi,
Mexico; Leiohunum aldrichi Weed from Tish-
omingo State Park, Tishomingo County, Missis-
sippi, USA; Leiohunum flavum Banks from
Beaver’s Bend State Park, McCurtain County,
Oklahoma; Merrymount Campground, 1 8 miles
SW Nashville, Davidson, Tennessee, USA;
Leiohunum montanum montanum Suzuki from
Mt. Ischizuchi, 1 490- 1 745 m., Ehime Prefecture,
Japan; Leiohunum sp. from 2 km. N. Tasquillo,
Rio Tula, Hidalgo, Mexico; Leiohunum sp. nr.
depressum Davis from 7.5 miles S. George West,
Live Oak County, Texas, USA,; Leiohunum
townsendi from East Turkey Creek, Chiracahua
Mountains, Cochise County, Arizona; outside
Hidden Cave (reared to deutonymph) and Her-
mit Cave, Eddy County, New Mexico, USA.;
Leiohunum vittatum (Say) from Homesville,
Nebraska, USA.; Trachyrhinus marmoratus
Banks from Pecos River, east of Pecos, Pecos
County, and Indio Mountains, 25 km S Van
Horn, Hudspeth County, Texas, USA. Phalan-
giidae: Odiellus pictus from Garland, Penobscot
County, Maine, USA.; Phalangium opilio from
Bowdoinham, near Cathance River, Sagahahoc
County, Maine, USA.

Additional mites from my collection have been
identified by W. Calvin Welboum as Leptus spp.
1-11. They are as follows: Leptus sp. 1 is known
from a Sclerosomatidae (Leiohunum townsendi)
and a Protolophidae (Protolophus singularis
Banks) from Fort Bayard, Grant County, New
Mexico, USA. Leptus sp. 2 is known from several
species of Sclerosomatidae: Eumesosoma roew-
eri from 14 miles E. Burnet, Bumet County, Tex-
as; 7.5 miles ESE Bandera, Bandera County, Tex-
as; Texas Tech University Center, Junction,
Kimble County, Texas, USA; Leiohunum flavum
Banks from Graham Creek, 5 miles SSE Zavalla,
Angelina County, Texas, USA; Leiohunum
townsendi from Montague County, Texas, USA.
Leptus sp. 3 is known from two species of Scle-

134

THE JOURNAL OF ARACHNOLOGY

rosomatidae {Leiobunum aldrichi, L. nigripes
Weed) from the W, Bank of J. Percy Priest Lake,
Elm Hills Park, Davidson County, Tennessee,
USA. Leptus sp. 4 is known from a Phalangiidae
{Egaenus convexus) from Biirgenland, Ruster
Hugelland, Austria. Leptus sp. 5 is known from
a Sclerosomatidae {Trachyrhinus marmoratus
Banks) from 39.6 miles SW Marfa, Presidio
County, Texas, USA. Leptus sp. 6 is known from
a Sclerosomatidae [Marthana nigerrima (Mull-
er)] from Tuba Mountains, S. Palawan Cabar,
Palawan, Philippines. Leptus sp. 7 is known from
a Sclerosomatidae {Eumesosomal sp.) from Joya
de Juan Mesa (outside), near La Laguna, Ta-
maulipas, Mexico. Leptus sp. 8 is known from a
Sclerosomatidae {Leiobunum sp.) from km 1 20
marker on Highway 70, San Luis Potosi, Mexico.
Leptus sp. 9 is known from a Sclerosomatidae
{Leiobunum sp.) from roadcut at Gomez Farias,
Tamaulipas, Mexico. Leptus sp. 10 is known from
a Sclerosomatidae {Lacinius ephippiatus) from
Wr. Wald, Latisberg (Cobenze) E-Mg ca. 380-“
400 m, Wien XIX, Austria. Leptus sp. 1 1 is known
from a Phalangiidae {Mitopus morio) from Wr.
Wald, Rekawiokel, near Bldf., N. G., Austria.

Leptus bicristatus Fain & Elsen (1987) was de-
scribed from a larva on an “Opilion” from Cho-
wo Rocks, Plateau de Nyika, Malawi (6-8 De-
cember 1981). The host has now been identified
as a Phalangiidae, Cristina lettowi (Roewer)
(Kauri, pers. commun.).

Leptus gagreilae (Oudemans 1 9 1 Ob) was orig-
inally described in combination with Achorolo-
phus Berlese. It was described from a Scleroso-
matidae {Gagrella sp.) from Tjibodas, West Java
Prov., Indonesia. This species was redescribed
and illustrated by Oudemans (1913).

Leptus hidakai Kawashima (1958) was de-
scribed from larvae collected on a Clubionidae
spider {Chiracanthium sp.) and a Phalangiidae
{Opilio pentaspinulatus Suzuki). All specimens
were obtained on 24 June at Tachibana-yama,
Kasuya-gun, Fukuoka Prefecture, Japan. A har-
vestmen is illustrated in the original description
with eight mites attached to its legs and abdo-
men.

Leptus holmiae Southcott ( 1 992) is a wide-rang-
ing species in the Holarctic region. It is recorded
(Southcott 1992) from a free living-example col-
lected in the Burzyanskij region, Bashkir ASSR;
and as ectoparasites on Phalangiidae: Mitopus
morio from Denmark, Iceland, Poland, Sweden;
Opilio sp. from Sweden; Opilio canestrinii (Tho-
rell) from Denmark; Phalangium opilio from En-
gland; Rileana (reported as Platybunus) trian-

gularis from England. Southcott (1992) stated
that he felt some additional specimens reported
in the literature might be this species but that he
could not be certain because he had not studied
any samples of the series reported. These ques-
tionable records are Phalangiidae: Mitopus mo-
rio and Opilio ruzickai from Bulgaria (Beron
1975); and Phalangium opilio, R. triangularis,
Lophopilio (reported as Odiellus) palpinalis from
Poland (Haitlinger 1987).

Leptus indianensis Fain et al. (1987) was de-
scribed from larvae collected on several species
of Sclerosomatidae: Leiobunum aldrichi (report-
ed as L. longipes) and Leiobunum calcar (Wood)
from 2 miles northwest Brazil, Clay County, In-
diana, USA.; Leiobunum sp,, L. nigripes, L. spe-
ciosum Banks and L. ventricosum from 9 miles
southwest of Crawfordsville, Montgomery
County, Indiana, USA. New records from my
collection include L. nigripes Weed from 4 miles
ESE Morris on Pine Bluff Road, Grundy County,
Illinois, USA.; L. formosum (Wood) from Po-
tomac River and Chesapeake Bay junction,
Wakefield, Virginia, USA.

Leptus jocquei Fain & Elsen (1987) was de-
scribed from nine larvae taken from “Opilions”
collected in Dembo, Plateau Nyika, Malawi (5-
20 December 1981). The host has now been iden-
tified as a Phalangiidae, Cristina lettowi (Kauri,
pers. commun.),

Leptus kalaallus Southcott (1992) is thus far
known only from the Phalangiidae, Mitopus mo-
rio, collected in Greenland. The larval mites were
found attached to the opilion abdomens.

Leptus lomani (Oudemans 1903b) was de-
scribed from a Gonyleptidae, Lycomedicus (re-
ported Discocyrtus) funestus (Butler), from Chile.
This species was redescribed and illustrated by
Oudemans (1913). The original series of 1 0 mites
was reported to have been collected by J. C. C.
Loman in 1900. Other sources indicate that Jan
C. C. Loman, of Amsterdam, did not collect the
specimens himself. The harvestmen were prob-
ably collected by Prof. Dr. Ludwig Plate and for-
warded to Oudemans by Loman. The only ex-
amples of this harvestman reported in the
literature from Chile during the same time period
was by Loman (Cekalovic K. 1985). Loman
(1899) stated that there were several specimens
of L. funestus from Corral that were in the Plate
collection. Therefore, I am herein restricting the
type locality of L. lomani to Corral (39®53'S,
73"25'W), Valdivia, Chile.

Leptus nearcticus Fain et al. (1987) was de-
scribed from larvae collected off three species of

COKENDOLPHER-PATHOGENS AND PARASITES OF OPILIONES

135

Sclerosomatidae: Leiobunum aldrichi (reported
as L. longipes), L. nigripes and L. vittatum from
2 miles northwest Brazil, Clay Co., Indiana (1™
18 September 1986). Fain et ai. (1987) reported
other samples from the type locality from Leiob-
unum sp. (females). These have now been iden-
tified as L. aldrichi.

Leptus oudemansi (Karppinen 1958) was orig-
inal proposed as a replacement name in the genus
Achorolophus. This name was provided because
Achorolophus gracilipes Oudemans 1910a, was
preoccupied by Rhyncholophus gracilipes Kra-
mer 1897; both were considered by Karppinen
(1958) to belong to Achorolophus. Both are now
considered by Southcott (1 992) to belong to Lep-
tus and are thus still homonyms. Oudemans’
(1910a) original specimens were found on a Cos-
metidae {Cynorta sp.) from Surinam. This spe-
cies was redescribed and illustrated by Oude-
mans (1913).

Leptus puylaerti Fain & Elsen (1987) is known
by five larvae found attached to “Opilions” col-
lected at Chowo Rocks, Nyika Plateau, Malawi
(6-18 Dec. 1981). The host has now been iden-
tified as a Phalangiidae, Cristina lettowi (Kauri,
pers. commun.).

Leptus polythrix Fain & Elsen ( 1 987) is known
by eight larvae found attached to “Opilions” col-
lected at Dembo, Nyika Plateau, Malawi (5-20
December 1981). The host has now been iden-
tified as a Phalangiidae, Cristina lettowi (Kauri,
pers. commun.).

Leptus stieglmayri (Oudemans 1905) was de-
scribed from Opiliones collected in Santa Cruz,
Rio Grande do Sul, Brazil. Oudemans (1913)
redescribed this species and recorded a specimen
that was collected from a beetle (Cleridae) col-
lected in Brazil.

A probable new genus (near Leptus) is under
study by W. Calvin Welboum. Thus far, mem-
bers are only known from harvestmen from my
collection obtained in Chile. The new specimens
are known from two species of Neopilionidae
{Thrasychirus modestus Simon, Thrasychirus
denticheiis Simon) from Isla Deceit Caleta To-
leda, archipielago Cabo de Homos, Magallanes,
Chile. This is the southern most record for a
harvestman parasite. Other host records include
species of Gonyleptidae: Eubalta meridionalis
(Sorensen) from Reserva Forestal Magallanes,
8 km west Punta Arenas, Magallanes, Chile;
Metagyndes pulchella (Loman); Niebla, near
Valdivia, Chile; Acanthoprocta pustulata Loman
from Cerro Nielol, Temuco, Chile.

ACKNOWLEDGMENTS

Like any project which is world-wide in scope,
many individuals were called upon for their as-
sistance. My colleagues who sent reprints of their
publications are gratefully acknowledged. Dr. Jan
Buchar (University Karlovy, Praha), Dr. Ashok
A. Hooli (Kamatak Science College, Dharwad),
Dr. Plamen G. Mitov (University of Sofia “Kli-
ment Ohridski”, Sofia), Mr. Sergio Sanchez-Pena
(Texas Tech University, Lubbock), and Mr. Lou-
is M. Sorkin (American Museum of Natural His-
tory, New York) kindly supplied additional use-
ful literature. Dr. Howard E. Evans (Colorado
State University, Fort Collins) is thanked for his
useful comments on pompilid wasps. Dr. Robert
G. Holmberg (Athabasca University, Athabasca)
kindly provided me data on the numerous spec-
imens of Leptus spp. obtained from harvestmen
in his collection. Dr. Francis G. Howarth (Bishop
Museum, Honolulu) is thanked for his com-
ments on harvestmen-like spiders of Hawaii. Dr.
Richard A. Humber (USDA-ARS, Ithaca) is
thanked for his useful comments on the manu-
script (especially the fungi section) and for send-
ing copies of several difficult to obtain papers.
Dr. Hans Kauri (Museum of Zoology, University
of Bergen, Bergen) is thanked for his identifica-
tions of the opilion hosts of Leptus spp. from
Malawi. These harvestmen were loaned to Dr.
Kauri by the Musee Royal de FAfrique Centrale,
Tervuren, I thank Dr. Plamen G. Mitov for his
comments on parasites of harvestmen in Bul-
garia and for his permission to publish his many
new records. Dr. Nobuo Tsurasaki (Tottori Uni-
versity, Tottori) is thanked for his comments on
gregarine and fly parasites in Japan, his aid in
providing difficult to obtain literature and for
translating some papers in Japanese. Mr. W. Cal-
vin Welboum (Acarology Laboratory, The Ohio
State University, Columbus) is thanked for his
many identifications of mites. Dr. George O.
Poinar, Jr. (University of California, Berkeley)
is thanked for his useful suggestions during the
preparation of this manuscript and for his com-
ments on the manuscript. I thank Dr. Norman
V. Homer (Midwestern State University) for his
many kindnesses shown me during the prepa-
ration of the manuscript. He helped me obtain
some literature and assisted with the adminis-
trative details in obtaining funds for publication.
More importantly, almost two decades ago Dr.
Homer served as my professor of bacteriology,
parasitology and arachnology.

136

THE JOURNAL OF ARACHNOLOGY

Table l.—List of pathogens and parasites grouped by opilion host.

Host

Family? incertae sedis
harvestmen, England
harvestmen, Netherlands
Opliones, Brazil
Opiliones, Mexico

Opiliones, U.S.A
Opilionide, Israel

opilionids
phalangid, U.S.A
Phalangium crassum,

France

Suborder Laniatores
Family Triaenonychidae and/or
Synthetonychidae
gen. sp., New Zealand

Family Triaenonychidae
Equitius doriae

Family Assamidae
Oppainia sp., India

Family Gonyleptidae
gen. sp. Brazil

Acanthoprocta pustulata
Discocyrtus invalidus
Eubalta meridionaiis
Gonyleptes fragilis
Lycomedicus funestus
Metagyndes pukhella
Zygopachylus albomarginis

Family Cosmetidae
Cynorta sp.

Paeciiaemana guadripuncta
Vonones sayi

Suborder Cyphopalpatores
Superfamily Troguloidea
Family Troguloidae
Trogulus tricarinatus
Family Nemastomatidae
Carinostoma ornatum
Paranemastoma radewi

Parasite

ILeptus sp.

Leptus sp.

Leptus stieglmayri
Erythraeidae

Leptus spp.
Allothrombium
chanaanense
Leptus sp.

ILeptus sp.
Sciadiophora
phalangii

Acarina, gen. sp.
Chalcidae, gen. sp.
Nematoda, gen. sp.

Acarina, gen. sp.
Gregarinasina, gen. sp.

Arachnocystis
amchnoidea
Contospora opalniae

Echinoocysta phalangii

Torrubiella
gonylepticida
N. gn. nr. Leptus sp.
Cydorrhapha, gen. sp.
N. gn. nr. Leptus sp.
Agamomermis sp.
Leptus lomani
N. gn. nr. Leptus sp.
Eumycota, gen. sp.

Leptus oudemansi
Agamomermis sp.
Acarina, gen. sp.
Cosmetophilus vonones

Doliospora troguli

Gregarinasina, gen. sp.
Gregarinasina, gen. sp.

Source

Evans 1910; Savory 1938
Spoek 1964
Oudemans 1905
Martinez Crespo and
Morales Soto 1979
Welboum 1983
Feider 1977

Mullen 1988
Cox et al. 1921
Leger 1897

Forster 1954
Forster 1954
Forster 1954

Hunt 1979
Hunt 1979

Devdhar 1962; Devdhar
and Gourishankar 1971
Devdhar 1962; Devdhar
and Amoji 1978a
Amoji and Devdhar 1979

Moller 1901

herein
Soares 1945
herein

Mattheisen 1974
Oudemans 1903b
herein
Mora 1987

Oudemans 1910a
Poinar 1985
herein

Cokendolpher 1991

Geus 1969

herein

herein

COKENDOLPHER-PATHOGENS AND PARASITES OF OPILIONES

137

Table L— Continued.

Host

Parasite

Source

Agamomermis sp.

herein

Pyza bosnica

Gregarinasina, gen. sp.

herein

Superfamily Phalangioidea

Family Neopilionidae

Thrasychirus dentichelis

N. gn. nr. Leptus sp.

herein

Thrasychirus modestus

N. gn. nr. Leptus sp.

herein

Family Protolophidae

Protolophus sp., U.S.A.

Agamomermis sp.

Poinar 1985

Protolophus singular is

Leptus sp. # 1

herein

Family Sclerosomatidae

Metopilio group

Globipes sp.

Acarina, gen. sp.

herein

Subfamily Leiobuninae

Eumesosomal sp.

Leptus sp. #7

herein

Eumesosoma roeweri

Leptus sp.

herein

Leptus sp. #2

herein

Leiobunum sp., U.S.A.

Leptus indianensis

Fain et al. 1987

Leiobunum sp. near

Leptus sp.

herein

depressum, U.S.A.

Leiobunum sp., Hidalgo,

Leptus sp.

herein

Mexico

Leiobunum sp,, San Luis

Leptus sp. #8

herein

Potosi, Mexico

Leiobunum sp.,

Leptus sp. #9

herein

Tamaulipas, Mexico

Leiobunum aldrichi

Leptus sp.

herein

{=Leiobunum longipes)

Leptus sp. #3

herein

Leptus indianensis

Fain et al. 1987

Leptus nearcticus

Fain et al. 1987

Leiobunum blackwalli

Anthorhynchus

Pfeifer 1956

(=L. hassiae)

sophiae

Entomophaga batkoi

Balazy 1978

Erythraeidae, gen. sp.

Sankey 1949b

Leiobunum calcar

Leptus indianensis

Fain et al. 1987

Leiobunum globosum

Agamomermis sp.

Tsurusaki 1986

Gregarinasina, gen. sp.

Tsurusaki 1986

Leiobunum flavum

Leptus sp.

herein

Leptus sp. #2

herein

Leiobunum formosum

Leptus indianensis

herein

Leiobunum manubriatum

Agamomermis sp.

Tsurusaki 1986

Gregarinasina, gen. sp.

Tsurusaki 1986

Leiobunum montanum

Leptus sp.

herein

montanum

Leiobunum nigripes

Leptus sp. #3

herein

Leptus indianensis

Fain et al. 1987; herein

Leptus nearcticus

Fain et al. 1987

Leiobunum politum

Gregarinasina, gen. sp.

herein

Leiobunum rotundum

Anthorhynchus

Ormieres and Baudoin 1973

longispora

Anthorhynchus

Pfeifer 1956

sophiae

Doliospora repelini

Ormieres and Baudoin 1973

138

THE JOURNAL OF ARACHNOLOGY

Table l.~ Continued.

Host

Parasite

Source

Leiobunum rumelicum
Leiobunum speciosum
Leiobunum townsendi

Leiobunum ventricosum
Leiobunum vittatum

Nelima sp., Japan
Nelima nigricoxa

Nelima paessleri
Nelima silvatica
Togwoteeus biceps

Subfamily Gagrellinae
Gagrella sp., Indonesia
Gagrellopsis nodulifera
Gagrellula ferruginea

Gagrellula saddlana

Krusa sp., Mexico
Marthana nigerrima
Metagagrella tenuipes
Prionostemma panama
Trachyrhinus marmoratus

Trachyrhinus rectipalpus
Family Phalangiidae
gen. sp., England

gen. sp., Germany

gen. sp., Sweden
Subfamily Phalangiinae
Bunochelis canariana
Cristina lettowi

Phalangium sp,, Italy

Entomophaga batkoi
Erythraeidae, gen. sp.
Leptus sp.
Sciadiophora
phalangii

Gregarinasina, gen. sp.
Leptus indianensis
Acarina, gen. sp.

Leptus sp.

Leptus sp. # 1
Leptus sp. #2
Acarina, gen. sp.
Leptus indianensis
Leptus sp.

Leptus nearcticus
Nomuraea atypicola
Trombidium sp.
Forcipomyiinae,
gen. sp.

Acarina, gen. sp.
Erythraeidae, gen. sp.
Agamomermis sp.
Leptus sp.

Leptus gagrellae
Acarina, gen. sp.
Forcipomyiinae,
gen. sp.

Sciadiophora
gagrellula
Leptus sp.

Leptus sp. #6
Charletonia southcotti
Acarina, gen. sp,
Leptus sp.

Leptus sp. #5
Acarina, gen. sp.

Hymenostilbe

verrucosa

Pandora phalangicida
Sciadiophora caudata
Sciadiophora fissidens
Pandora phalangicida

Charletonia enghoffi
Leptus bicristatus
Leptus jocquei
Leptus polythrix
Leptus puylaerti
Allothrombium
neapolitum

Batazy 1978
Sankey 1949b
Meade 1855
Pfeifer 1956

herein

Fain et al. 1987
Elliott and Reddell 1987;

herein

herein

herein

herein

herein

Fain et al. 1987
herein

Fain et al. 1987
Greenstone et al. 1988
Yokogawa 1940
herein

herein

Sankey 1949b
Poinar 1985
herein

Oudemans 1910b

herein

herein

Devdhar 1962; Devdhar
and Amoji 1978b
herein
herein

Kawashima 1961

herein

herein

herein

herein

Leatherdale 1970

Leatherdale 1970
Rossler 1882
Rossler 1882
Lagerheim 1898

Southcott 1991
Fain & Elsen 1987
Fain & Elsen 1987
Fain & Elsen 1987
Fain and Elsen 1987
Oudemans 1910a

COKENDOLPHER-- PATHOGENS AND PARASITES OF OPILIONES

139

Table L— Continued.

Host

Parasite

Source

Phalangium sp., Sweden

Leptus sp.

De Geer 1778

Phalangium sp., Europe

Sciadiophora

phalangii

Geus 1969

Phalangium opilio

Agamomermis sp.

Pfeifer 1956; herein

{^Phalangium cornutum)

Agamomermis

phalangii

Haldeman 1851

Agamomermis

truncatula

Rudolph! 1819

Anthorhynchus

Schneider 1887; Ormieres

sophiae

and Baudoin 1973

Brachylecithum sp.

Gabrion and Ormieres 1973

Digenea, gen. sp.

Creplin 1846

Doliospora repelini

Leger 1897; Ormieres
and Baudoin 1973

Erythraeidae, gen. sp.

Sankey 1949b

Gregarinasina, gen. sp.

herein

Heterorhabditis

heliothidis

Poinar and Thomas 1985

Leptus sp.

Evans 1910; Pfeifer 1956;

Abro 1988, 1991; herein

Leptus holmiae

Southcott 1992

Leptus holmiael

Haitlinger 1987

Pandora phalangicida

Ellis 1956; Leatherdale

1958, 1970

Pseudhymenolepis

redonica

Gabrion 1977

Rhabditida, gen. sp.

Pfeifer 1956

Sciadiophora caudata

Rossler 1882; Ormieres
and Baudoin 1973

Sciadiophora fissidens

Rossler 1882

Sciadiophora

geronowitschi

Johansen 1894

Sciadiophora phalangii

Leger 1897; Geus 1969

Steinernema

carpocapsae

Poinar and Thomas 1985

Xenorhabdus

luminescens

Poinar and Thomas 1985

Xenorhabdus

nematophilus

Poinar and Thomas 1985

Rilaena triangularis

Anthorhynchus

Pfeifer 1956

(=Platybunus triangularis)

sophiae

Erythraeidae, gen. sp.

Sankey 1949b

Leptus holmiae

Southcott 1992

Leptus holmiael

Haitlinger 1987

Nematoda, gen. sp.

Sankey 1949a

Sciadiophora phalangii

Pfeifer 1956

Zacheus anatolicus

Gregarinasina, gen. sp.

herein

Zacheus crista

Agamomermis sp.

herein

Allothrombium

neapolitum

herein

Gregarinasina, gen. sp.

herein

Zacheus hebraicus

Acarina, gen. sp.

herein

Subfamily Oligolophinae

Lacinius ephippiatus

Agamomermis sp.

herein

140

THE JOURNAL OF ARACHNOLOGY

Table 1.— Continued.

Host Parasite Source

Lacinius dentiger

Lacinius horridus

Mitopus morio
(^Opilio gwssipes)

Mitopus sp., France

Odieilus lendli
Odiellus pictus

Odieilus spinosus
i^Odius spinosus)
Oligolophus hansenii
Oligolophus tridens

Paroligolophus agrestis

Subfamily Opilioninae
Egaenus convexus

Opilio sp,, Europe

Anthorhynchus

sophiae

Gregarinasina, gen. sp.
Leptus sp. #10
Agamomermis sp.
Gregarinasina, gen. sp.
Sciadiophora phalangii
Agamomermis sp.
Gregarinasina, gen. sp.
Rhabditida, gen. sp.
Acarina, gen. sp.
Agamomermis sp.
Anthorhynchus
longispora
Anthorhynchus
sophiae

Doliospora repelini
Erythraeidae, gen. sp.
Gregarinasina, gen. sp.
Leptus sp.

Leptus sp. #11
Leptus holmiae
Leptus holmiae?

Leptus kalaallus
Sciadiophora caudata

Sciadiophora phalangii

Sciadiophora

claviformis

Gregarinasina, gen. sp.
Anoplius marginatus
Gregarinasina, gen. sp.
Leptus sp.

Sciadiophora caudata

Erythraeidae, gen. sp.
Anthorhynchus
sophiae

Doliospora repelini
Entomophaga batkoi
Erythraeidae, gen. sp.
Leptus sp.

Erythraeidae, gen. sp.

Leptus sp.

Leptus sp. #4
Trombidium hungarium
Agamomermis sp.
Agamomermis
truncatula

Pfeifer 1956

herein

herein

herein

herein

Silhavy 1961

herein

herein

Pfeifer 1956

Burton and Burton 1984
Stipperger 1928; herein
Ormieres and Baudoin 1973

Pfeifer 1956

Ormieres and Baudoin 1973

Sankey 1949b

herein

Stipperger 1928; Evans
et al. 1961; Abro 1988;
herein
herein

Southcott 1992
Sellnick 1940; Beron 1975
Southcott 1992
Rossler 1882; Ormieres
and Baudoin 1973
Stipperger 1928; Pfeifer
1956; Ormieres and
Baudoin 1973

Ormieres and Baudoin 1973

herein
Evans 1948
herein
herein

Rossler 1882; Ormieres
and Baudoin 1973
Sankey 1949b

Ormieres and Baudoin 1973

Ormieres and Baudoin 1973
Batazy 1978; Keller 1987
Sankey 1949b
herein

Sankey 1949b
herein

herein

Kobulej 1957
Kastner 1928
Rudolphi 1819

COKENDOLPHER--^ PATHOGENS AND PARASITES OF OPILIONES

141

Table 1.™ Continued.

Host Parasite Source

Opilio canestrinii
Opilio dinaricus
Opilio parietinus

Opilio pentaspinulatus
Opilio ruzickai

Opilio saxatilis
Subfamily Platybuninae
Lophopilio palpinalis
i=Odiellus palpinalis)

Megabunus diadema

Platybunus bucephalus

Platybunus pinetorum

Leptus holmiae
Leptm holmiae
Gregarinasina, gen. sp.
Agamomermis sp.
Anthorhynchus
longispora
Doiiospora repelini
Erythraeidae, gen. sp.
Leptus sp.
Microsporidium
weiseri

Sciadiophora phalangii
Leptus hidakai
Gregarinasina^ gen. sp.
Leptus holmiael
Gregarinasina, gen. sp.

Agamomermis sp.

Gregarinasina, gen. sp.
Leptus sp.

Leptus holmiael
Sciadiophora fissidens
Actinocephalus
megabuni
Doiiospora repelini
Anthorhynchus
longispora
Doiiospora repelini
Sciadiophora phalangii
Sciadiophora phalangii

Southcott 1992
Southcott 1992
herein

Unzicker and Rotramel 1970
Ormieres and Baudoin 1973

Ormieres and Baudoin 1973
Sankey 1949b
Evans et al. 1961
Silhavy 1960

Pfeifer 1956
Kawashima 1958
herein
Beron 1975
herein

herein

herein

Haitlinger 1987
Haitlinger 1987
Rossler 1882

Ormieres and Baudoin 1973

Ormieres and Baudoin 1973
Ormieres and Baudoin 1973

Ormieres and Baudoin 1973
Ormieres and Baudoin 1973
Pfeifer 1956

LITERATURE CITED

Abro, A, 1988. The mode of attachment of mite lar=
vae {Leptus spp.) to harvestmen (Opiliones). J. Nat,
Hist, 22:123-130.

Abro, A. 1991. Unsuccessful parasitic association of
mite larvae {Leptus spp.) to harvestmen (Opiliones).
Fauna Norvegica, Ser. B, 38:43. (not seen)

Akhurst, R. J . 1987. Use of starch gel electrophoresis
in the taxonomy of the genus Heterorhabditis (Nem-
atoda: Heterorhabditidae). Nematologica, 33:1-9.
(not seen)

Akhurst, R. J. & N. E. Boemare, 1990. Biology and
taxonomy of Xenorhabdus. Pp. 75-90, In Ento-
mopathogenic nematodes in biological control.
(Gaugler, R. & H. K. Kaya, eds.). CRC Press, Boca
Raton, Florida.

Amoji, S. D. & M. J. Devdhar. 1979. Echinospora
phalangii gen. n., sp. n. an actinocephaid [sic] greg-
arine found in the intestine of phalangid host, Opal-
nia sp. Acta Protozool., 18:517-526.

Batko, A. 1964. On the new genera: Zoophthora gen.
nov., Triplosporium (Thaxter) gen. nov. and Ento-
mophaga gen. nov. (Phycomycetes: Entomophthor-
aceae). Bull. Acad, Polonaise Sci., CL II, Ser. Sci.,
Biol, 12:323-326.

Batko, A. 1966. On the subgenera of the fungus genus
Zoophthora Batko 1964 (Entomophthoraceae). Acta
MycoL, 2:15-21.

Balazy, S. 1978. A new species of Entomophthora-
ceae (Mycophyta: Entomophthorales) from Poland.
J. Invert. Pathol, 31:275-279.

Beron, P. 1975. Erythraeidae (Acariformes) larvaires
de Bulgarie. Acta ZooL Bulgarica, 1:45-75.

Brown, A. H. S. & G. Smith. 1957. The genus Pae-
cilomyces Bainier and its perfect stage Byssochlamys
Westling, Trans. British Mycol. Soc., 40: 17-89. (not
seen)

Burton, M, & R. Burton. 1984. Encyclopedia of in-
sects and arachnids. Finsbury Books and BPC Publ,
San Sebestian, Spain, 252 pp.

142

THE JOURNAL OF ARACHNOLOGY

Camey, W. P„ 1969. Behavioral and morphological
changes in carpenter ants harboring dicrocoeliid me-
tacercariae. American Midi. Nat., 82:605-6 11.

Camey, W. P. 1970. Brachyledthum mosquensis: in-
fections in vertebrate, molluscan and arthropod
hosts. Trans. American Micr. Soc., 89:233-250.

Camey, W. P. 1972. Studies on the life history of
Brachyledthum myadestis sp. n. (Trematoda: Di-
crocoeliidae). J. ParasitoL, 58:519-523.

Cekalovic K., T, 1985. Catalogo de los Opiliones de
Chile (Arachnida). Bol Soc. Biol. Concepcion, 56:
7-29.

Cokendolpher, J. C. 1991. Cosmetophilus vonones,
n. g., n. sp,, (Apicomplexa: Actinocephalidae) in the
harvestman Vonones sayi (Arachnida: Cosmetidae).
J, ProtozooL, 38:461-464.

Cole, L. K., M. S. Blum & R. W. Roncadori. 1975.
Antifungal properties of the insect alarm phero-
mones, citral, 2-heptanone, and 4-methyl-3”heptan"
one, Mycologia, 67:701-708.

Cox, F., P. Jahraus & W. Moore. 1921. A list of
California Arachnida. VI. Acarina or the mites and
ticks. J. Entomol. ZooL, 13:23-37.

Creplin, F. C. H. 1846. Nachtrage zu Gurlt’s Ver-
zeichmiss der Thiere, bei welchen Entozoen gefun-
den worden sind. Arch. Naturg., 12:129-160.

De Geer, C. 1778. Memoires pour servir a Thistoire
des insects. Tome 7. P. Hesselberg, Stockholm, xii
+950 pp. +49 pis,

Devdhar, M. J. 1 962. Studies on the Gregarines from
some arthropods of the northern Mysore Region.
Unpubl. Ph. D. thesis, Karnataka University, Dhar-
ward, India.

Devdhar, M. J. & S. D. Amoji. 1978a. On a new
actinocephalid gregarine, Contospora opalnia gen.
nov., sp. nov. from the gut of an arachnid Opalnia
sp. Archiv. Protist., 120:182-189.

Devdhar, M. J. & S. D. Amoji. 1978b. Sdadiophora
gagrellula sp. n. from the phalangid arthropod, Ga-
grellula saddlana (Roewer), Acta ProtozooL, 1 7:247-
253.

Devdhar, M. J. & G. Gourishankar. 1971. On a new
gregarine, Sycia arachnoidea, n. sp., from an arach-
nid host Opalnia sp. J. Anim. Morphol. Physiol,
18:182-187.

Diesing, C. M. 1851. Systema Helminthum. W.
Braumuller, Vienna, vol. 2, vi +588 pp,

Eaton, J. 1985. Trip reports: Cave of the Bell, Eddy
Co., NM. June 28, 1985. Southwest. Cavers, 23:52.

Ekpa, O., J. W. Wheeler, J. C. Cokendolpher & R. M.
Duffield, 1984. N, N-dimethyl-/3-phenylethyl-
amine and bomyl esters from the harvestman Scler-
obunus robustus (Arachnida: Opiliones). Tetrahe-
dron Lett., 25:1315-1318.

Ekpa, O., J. W. Wheeler, J, C. Cokendolpher & R. M.
Duffield. 1985. Ketones and alcohols in the de-
fensive secretion of Leiobunum townsendiW and

a review of the known exocrine secretions of Pal-

patores (Arachnida: Opiliones). Comp. Biochem.
Physiol, 81B:555-557.

Elliott, W. R. & J. R. Reddell 1987. Kickapoo project
biology report. Texas Caver, 32:133-134, 148.

Ellis, E. A. 1956. Entomogenous fungi in Norfolk.
Trans. Norfolk Norwich Nat. Soc., 18:23-28 +12
plates.

Ellis, M. M. 1912. A new species of polycystid greg-
arine from the United States. Zool Anz, 39:25-27.

Ellis, M. M. 1913. A descriptive list of the Cephaline
Gregarines of the New World. Trans. American Mi-
crosc. Soc., 32:259-296.

Estable, C., M. 1. Ardao, N, P. Brasil & L, F. Fieser.
1955. Gonyleptidine. J. American Chem, Soc., 77:
4942.

Evans, H. C. & R. A. Samson. 1987. Fungal patho-
gens of spiders. Mycologist, 21:152-159.

Evans, H. E. 1948. Biological notes on two species
of Anoplius (Hymenoptera: Pompilidae). Entomol
News, 59:180-184.

Evans, H. E. & C. M. Yoshimoto. 1962. The ecology
and nesting behavior of the Pompilidae (Hymenop-
tera) of the northeastern United States. Misc. Publ
Entomol. Soc. America, 3:67-119.

Evans, G. O., J. G. Sheals & D. MacFarlane. 1961.
The terrestrial Acari of the British Isles. London,
British Museum. Vol. 1. Introduction and biology,
219 pp.

Evans, W. 1910. Note on ""Leptus phalangir and
''Leptus autumnalis'\ and their parent earth-mites.
Proc. R. Phys. Soc. Edinburgh, 18:100-101.

Fain, A. & P. Elsen. 1987. Observations sur les larves
du genre Leptus Latreille, 1795 (Acari, Erythrae-
idae) d’Afrique centrale. Rev. Zool Africaine, 101:
103-123.

Fain, A., S. L. Gummer & J. O. Whitaker. 1987. Two
new species of Leptus Latreille (Acari, Erythraeidae)
from the U.S.A. Intern. J. Acarol, 13:135-140.

Feider, Z. 1977. Contribution a la connaissance des
larves d’acariens du bassin oriental de la Mediter-
ranee. Israel J. Zool, 26:100-1 13.

Fieser, L. F. & M. 1. Ardao. 1956. Investigation of
the chemical nature of gonyleptidine. J. American
Chem. Soc,, 78:774-781.

Forster, R. R. 1954. The New Zealand harvestmen
(Sub-order Laniatores). Canterbury Mus. Bull, no.
2, 329 pp.

Gabrion, C. 1977. Presence de Cysticercoides d'un
cestode cyclophyllide chez un arachnide, Phalan-
gium opilio. Ann, Parasitol, 52:229-234.

Gabrion, C. & R. Ormieres. 1973. Presence chez
Phalangium opilio L. (Arachnida, Phalangidae) de
metacercaires d'un Trematode du genre Brachyle-
dthum Strom, 1940 (Dicrocoeliidae). C. R, Acad.
Sci., Sen D, 276:2957-2959.

Gams, W., G. S. de Hoog & R. A. Samson. 1984. The
hyphomycete genus Engyodontium a link between
VerticillumsLndAphanocladium. Persoonia, Leiden,
12:135-147.

COKENDOLPHER^PATHOGENS AND PARASITES OF OPILIONES

143

Geus, A. 1969. Sporentierchen, Sporozoa. Die Gre-
garinida der land-und suBwasserbewohnenden Ar-
thropoden Mitteleuropas. Pp. 1-608, In Die Tier-
welt Deutschlands und der angrenzenden Meeresteile
nach ihren Merkmalen und nach ihrer Lebensweise.
(F. Dahl, M. Dahl & F. Peus, eds.). 57 Teil, Gustav
Fischer Verlag, Jena.

Greenstone, M. H., C. M. IgnofFo & R. A. Samson.
1988. Susceptibility of spider species to fungus No-
muraea atypicola. J. Arachnol., 15:266-268.

Griffiths, P. 1978. Collection, culture and identifi-
cation of soil microfungi from Florence Lake Cave
near Victoria, B. C. Canadian Caver, 10:35-38.

Haitlinger, R. 1987. The genus Leptus Latreille, 1796
and Charletonia Oudemans, 1910 (Acari, Prostig-
mata, Erythraeidae) in Poland (Larvae). Polskie Pis-
mo Entomol., 57:339-349.

Haitlinger, R. 1990. Two new species of La-

treille, 1796 (Acari, Prostigmata, Erythraeidae) from
Tenebrionidae (Coleoptera) with a key to European
and North African species. Polskie Pismo Entomol.,
60:45-58.

Haldeman, S. S. 1851. Invertebrates. Pp. 48-196, In

Outlines of general zoology. (S. Baird, ed.). R. Ga-
rrigue, Publ., New York.

Herman, J. F. 1804. Memoire apterologique. Stras-
bourg. (not seen)

Hohorst, W. & G. Graefe. 1961. Ameisen - obliga-
torische Zwischenwirte des Lanzettegels {Dicrocoe-
lium dendriticum). Naturwissenschaften, 48:229-
230.

Holmberg, R. G. 1986. The scent glands of Opiliones:
A review of their function. Pp. 131-133, In Proc.
Ninth Intern. Congr. Arachnol., Panama, 1983. (W.
G. Eberhard, Y. D. Lubin & B. C, Robinson, eds.).
Smithsonian Inst., Washington.

Humber, R. A. 1989. Synopsis of a revised classifi-
cation for Entomophthorales (Zygomycotina). My-
cotaxon, 34:441-460.

Humber, R. A. & M. C. Rombach. 1987. Torrubiella
ratticaudata sp. nov. (Pyrenomyctes: Clavicitales)
and other fungi from spiders on the Solomon Is-
lands. Mycologia, 79:375-382.

Hunt, G. S. 1979. Male dimorphism and geographic
variation in the genus Equitius Simon (Arachnida,
Opiliones). Unpubl. Ph. D. thesis, Univ. New South
Wales, Australia, xiv +312 pp.

Ignoffo, C. M., C. Garcia & R. A. Samson. 1989.
Relative virulence of Nomumea spp. {N. rileyi, N.
atypicola, N. anemonoides) originally isolated from
an insect, a spider and soil. J. Invert. Pathol., 54:
373-378.

Johansen, M. 1894. Actinocephalus goronowitschi,
eine anscheinend neu Gregarinenform. Zool. Anz.,

17:140-145.

Joyeux, Ch. & J.-G. Baer. 1935. Un Tenia hypera-
polytique chez un Mammifere. C. R. Soc. Biol., 120:
334-336.

Kalavati, C. & C. C. Narasimhamurti. 1978. Three

new species of gregarines from termites. Proc. In-
dian Acad. Sci., 87 B (Anim. Sci.-4), no. 12, pp.
397...404.

Karppinen, E. 1958. Beobachtungen liber das Vor-
kommen von Arten der Familie Erythraeidae (Acar.)
in Finnland sowie Verandemngen in deren No-
menklatur. Ann. Entomol. Fennici, 24:42-45.

Kastner, A. 1928. Opiliones (Weberknechte, Rank-
er). Pp. 1-51, In Die Tierwelt Deutschlands und der
angrenzenden Meeresteile nach ihren Merkmalen
und nach ihrer Lebensweise. (F. Dahl, ed.). Gustav
Fischer, Jena, vol. 8.

Kawashima, K. 1958. Studies on larval erythraeid
mites parasitic on arthropods from Japan (Acarina:
Erythraeidae). Kyushu J. Med. Sci., 9:190-21 1.

Kawashima, K. 1961. Notes on larval mites of the
genus Charletonia Oudemans, 1910 in Japan (Ac-
arina: Erythraeidae). Kyushu J. Med. Sci., 12:15-
19.

Keller, S. 1987. Arthropod-pathogenic Entomo-
phthorales of Switzerland. I. Conidiobolus, Ento-
mophaga and Entomophthora. Sydowia, 40:122-
167.

Khan, A., W. M. Brooks & H. Hirschmann. 1976.
Chromonema heliothidis n. gen., n. sp. (Steineme-
matidae, Nematoda), a parasite of Heliothis zea
(Noctuidae, Lepidoptera), and other insects. J. Ne-
matol., 8:159-168.

Kobayasi, Y. & D. Shimizu. 1982. Monograph of the
genus Torrubiella. Bull. Natn. Sci. Mus., Tokyo, Ser.
B, 8:43-78.

Kobulej, T. 1957. Beitrage zur Trombidiidenfauna
ungams. 1. Feststellung der Identitat der Trombi-
diumlarve. Acta Vet. Hungarica, 7:175-184.

Koval’, E. Z. 1 974. [Guidebook to entomophilic fun-
gi of the USSR.] Naukova Dumka, Kiev, 260 pp.
(in Russian)

Kramer, P. 1 897. Gronlandische Milben. Bibl. Zool.,
20:77-83. (not seen)

Kudo, R. R. 1971. Protozoology. Charles C. Thomas,
Publ., Springfield, Illinois, xi + 1 174 pp.

Labbe, A. 1899. Sporozoa. Pp. 1-180, In Das Tier-
reich. Eine Zusammenstellung und Kennzeichnung
der rezenten Tierformen, (F. E. Schulze & O. Biitschli,
eds.). W. de Gruyter, Berlin, vol. 5. (not seen)

Lagerheim, G. 1898. Mykologische Studien. 1. Bei-
trage zur Kenntnis der parasitischen Pilze, 1-3. Bi-
hang Till K. Sv. Vet. Akad. Handling., Bd. 24, Afd.
3, No. 4, 3-22, pi. 1-3.

Latreille, P. A. 1796. Precis de caracteres generiques

des Insectes, disposes dans un ordre naturel par le
Citoyen Latreille, F. Bourdeaux, Brive. An. 5, pp.
i-xiii +1-201 +[7 pp. tabs.].

Leatherdale, D. 1958. A host catalogue of British
entomogenous fungi. Entomol, Month. Mag., 94:
103-105.

Leatherdale, D. 1970. The arthropod hosts of ento-
mogenous fungi in Britain. Entomophaga, 15:419-
435.

144

THE JOURNAL OF ARACHNOLOGY

Leger, L. 1892. Recherches sur les gregarines. Tabl.
ZooL, 3:1-182 +pls. I-XXII.

Leger, L. 1897. Nouvelles recherches sur les poly-
cystidees parasites des arthropodes terrestres. Ann.
Fac. Sci. Marseille, 6:1-54 -hpls. II, III.

Levine, N. D. 1979. New genera and higher taxa of
Septate Gregarines (Protozoa, Apicomplexa). J.
Protozool., 26:532-536.

Levine, N. D. 1980. Some nomenclatural corrections
among septate gregarines (Protozoa, Apicomplexa).
Proc. Zool. Soc. (Calcutta) (1976), 29:21-28.

Levine, N. D. 1984. Nomenclatoral corrections and
new taxa in the Apicomplexan Protozoa. Trans.
American Microsc. Soc., 103:195-204.

Levine, N.D. 1985. Phylum 11. Apicomplexa Levine,
1970. Pp. 322-374, In An illustrated guide to the
Protozoa. (J. J. Lee, S. H. Hunter & E. C. Bovee,
eds.). Society of Protozoologists, Lawrence, Kansas.

Lister, M. 1678. Historiae animalium Angliae tres
tractatus. Unus de Araneis. Alter de Cochleis turn
terrestribus turn fluviatilibus. Tertius de Cochleis
marinis. London, 250 pp.

Loman, J. C. C. 1899. Die Opilioniden der Samm-
lung Plate. Zool. Jahr. Suppl. 5 (Fauna Chilensis),
2:1-14 +pl. 1.

Mains, E. B. 1949. New species of Torrubiella, Hir-
sutella and Gibellula. Mycologia, 41:303-310.

Mains, E. B. 1950. Entomogenous species of Akan-
thomyces, Hymenostilbe and Insecticola in North
America. Mycologia, 42:566-589.

Martinez Crespo, A. C. & M. Morales Soto. 1979.
Los opilliones [sic] (Arachnida) y sus ectoparasittos
[sic] eritreidos (Acarida, Erythraeidae). 14 Congr.
nac. entomol., Soc. Mexicana EntomoL, Marzo 25-
28, 1979. Monterrey, N. L., P. 44.

Matthiesen, F. A. 1974. Sobre um inimigo natural
de Opiloes brasileiros. An. IV Jornada Cien. Fac.
Cienc. Med. Biol. Botucatu, S.P., Sec. 2, p. 5.

Meade, R.H. 1855. Monograph on the British species
ofPhalangiidae or harvestmen. Ann. Mag. Nat. Hist.,
Ser. 2, 15:393-416.

Megnin, P. 1876. Memoires sur les metamorphoses
des Acariens en general, et en particular sur celles
des Trombidions. Ann. Sci. Nat., Ser. 6, 4: 1-20. (not
seen)

Minchin, E. A. 1903. The Protozoa (continued). Sec-
tion K.-The Sporozoa. Pp. 150-360, In A Treatise
on Zoology. (E. R. Lankester, ed.). Adam & Charles
Black, London.

Mora, G. 1987. Male egg-guarding behavior in the
Neotropical harvestman, Zygopachylus albomar-

ginis (Opiliones: Gonyleptidae), Unpubl. Masters
thesis, Univ. of Florida, Gainesville, 91 pp.

Moller, A. 1901. Phycomyceten und Ascomyceten.
Schimper’s Bot. Mit. Tropen. Gustav Fischer Ver-
lag, Jena, Heft 9, 319 pp.

Mullen, G. R. 1988. Mites parasitic on arachnid hosts.
American Arachnol., no. 38, p. 7.

Ormidres, R. & J. Baudoin. 1969. Proposition de Do-

liospora n. g. pour I’Eugregarine '"Acanthospom re-
pilinr Leger, 1897, parasite d’Opilions. Ann. Stn.
Biol. Besse-en-Chandesse, 4:271-279.

Ormieres, R. & J. Baudoin. 1973. Eugregarines par-
asites d’Opilions de la region de Besse. Ann. Stn.
Biol. Besse-en-Chandesse,7: 199-208 +Figs. 1-47.

Oudemans, A. C. 1 903a. Acarologische Aanteeken-
ingen, VIL Ent. Ber., Amsterdam, 1:91-92.

Oudemans, A. C. 1 903b. Notes on Acari. Fifth series.
Tijdschr. Ent, 45:123-150 + pis. 10-12.

Oudemans, A. C. 1905. Acarologische Aanteekeni-
gen XVIII. Ent Ber., Amsterdam, 1:236-241. (not
seen)

Oudemans, A. C. 191 Oa. Acarologische Aanteeken-
ingen XXXI. Ent. Ber., Amsterdam, 3:47-51.

Oudemans, A. C. 191 Ob. Acarologische Aanteeken-
ingen XXXII. Ent. Ber., Amsterdam, 3:67-74.

Oudemans, A, C. 1913. Die bis jetzt bekannten Lar-
ven von Thrombidiidae und Erythraeidae mit be-
sonderer Berkucksichtigung der fiir den Menschen
schadlichen Arten. Zool. Jb. Abt. 1, Suppl. XIV,
No. 1. 230 pp.

Fetch, T. 1937. Notes on entomogenous fungi. Trans.
British Mycol. Soc., 21:34-67.

Pfeifer, H. 1956. Zur Okologie und larvalsystematik
der Weberknechte. Mitt. Zool. Mus. Berlin, 32:59-
104.

Poinar, Jr., G. O. 1975. Description and biology of
a new insect parasitic rhabditoid, Hetewrhabditis
bacteriophora n. gen., n. sp. (Rhabditida; Heterhab-
ditidae n. fam.). Nematologica, 21:463-470.

Poinar, Jr., G. O. 1977. CIH key to groups and genera
of nematode parasites of invertebrates. Common-
wealth Agricultural Bureaux, Famham Royal, En-
gland, 43 pp,

Poinar, Jr., G. O. 1979. Nematodes for biological
control of insects. CRC Press, Inc., Boca Raton,
Florida, [xi] +277 pp.

Poinar, Jr., G. O. 1983. The natural history of Nem-
atodes. Prentice-Hall, Inc., Englewood Cliffs, New
Jersey, x +323 pp.

Poinar, Jr., G. O. 1985. Mermethid (Nematoda) par-
asites of spiders and harvestmen. J. Arachnol., 13:
121-128.

Poinar, Jr., G. O. 1990. Taxonomy and biology of
Steinemematidae and Heterorhabditidae. Pp. 23-
61, /« Entomopathogenic nematodes in biological
control. (Gaugler, R. & H. K. Kaya, eds.). CRC
Press, Boca Raton, Florida.

Poinar, Jr., G. O., T. Jackson & M. Klein. 1 987. Het-
erorhabditis megidis sp. n. (Heterorhabditidae:
Rhabditida), parasitic in the Japanese beetle, Po-
pillia japonica (Scarabaeidae: Coleoptera), in Ohio.
Proc. Helminthol. Soc. Washington, 54:53-59.

Poinar, Jr., G. O. & G. M. Thomas. 1965. A new
bacterium, Achromobacter nematophilus sp. nov.
(Achromobacteriaceae: Eubacteriales) associated
with a nematode. Int. Bull. Bacteriol. Nomencl.
Taxon., 15:249-252.

COKENDOLPHER^PATHOGENS AND PARASITES OF OPILIONES

145

Poinar, Jr., G. O. & G. M. Thomas. 1985. Laboratory
infection of spiders and harvestmen (Arachnida:
Araneae and Opiliones) with Neoaplectana and Met-
er or habditis nematodes (Rhabditoidea). J. Arach-
noL, 13:297-302.

Poinar, Jr., G. O., G. M. Thomas & V. F. Lee. 1985.
Laboratory infection of Garypus californicus (Pseu-
doscorpionida, Garypidae) with neoaplectanid and
herterorhabditid nematodes (Rhabditoidea). J.
ArachnoL, 13:400-402.

Poinar, Jr., G. O. & H. E. Welch. 1981. Parasites of

invertebrates in the terrestrial environment. Pp. 947-
954, In Review of Advances in Parasitology. (W.
Slusarski, ed.). Polish Scientific Publishers, Warsza-
wa.

Remaudiere, G. & S. Keller. 1980. Revision syste-

matique des genres d’Entomophthoraceae a poten-
tialite entomopathogdne. Mycotaxon, 11:323-338.
Roewer, C. F. 1923. Die Weberknechte der Erde.
Systematische Bearbeitung der bisher bekannten
Opiliones. Gustav Fischer, Jena. 1116 pp.

Rossler, R. 1882. Beitrage zur Anatomic der Phal-
angiden. Zeits. Wiss. Biol., Leipzig, 36:671-702
+Tafel XLI, XLII.

Rudolphi, C. A. 1819. Entozoorum Synopsis cui ac-
cedunt mantissa duplex et indices locupletissimi. A.
Rucker, Berlin. 811 pp.

Riihm, J. 1926. Uber die Nahrung von Phalangium

L. Zool Anz., 68:154-158.

Samson, R. A. 1974. Paecilomyces and some allied
Hyphomycetes. Stud. Mycol., no. 6. 1 13 pp.
Samson, R. A., H. C. Evans &J.-P. Latge. 1988. Atlas
of entomopathogenic fungi. Springer- Verlag, Berlin,
xi +187 pp.

Sankey, J. H. P. 1 949a. British harvest-spiders. Essex
Nat., 28:181-191.

Sankey, J. H. P. 1949b. Observations on food, ene-
mies and parasites of British harvest-spiders (Arach-
nida, Opiliones). Entomol. Mon. Mag. 85:246-247.
Say, T. 1824. Narrative of an Expedition to the source
of St. Peter’s river, &c., under the command of
Stephen H. Long, Major U.S.T.E. Philadelphia,
2:268-378.

Savory, T. H. 1938. Notes on the biology of har-
vestmen. Quekett Microsc. Club, 1:89-94.
Schneider, A. 1887. Gregarines nouvelles ou peu con-
nues. Tabl. zooL, Poitiers, 2:67-85, pis. X, Xbis
Sellnick, M. 1940. Die Milbenfauna Islands. Gote-
borgs Kungl. Vetensk.-Samh. Handl., Ser. 5, 6B:1-
129.

Silhavy, V. 1960. wmer/ n. sp., eine neue

Mikrosporidienart aus dem Weberknechte Opilio
parietinus (De Geer). Vest. Ceskoslovenske Zool.
SpoL, 24:50-53.

Silhavy, V. 1961. Vyskyt gregarinosy sekacu (Opi-

lionidea) u Tfebice. Vlastivedny Shorn. Vysociny,
5:135-146.

Soares, B. M. 1945. Opiloes de Porto Cabral. Pap.
Avulsos, Sao Paulo, 5:107-118.

Southcott, R. V. 1961. Studies on the systematics
and biology of the Erythraeoidea (Acarina), with a
critical revision of the genera and subfamilies. Aus-
tralian J. Zool., 9:367-610.

Southcott, R. V. 1965. Revision of the genus Char-
letonia Oudemans (Acarina: Erythraeidae). Austra-
lian J. Zool., 14:687-819.

Southcott, R. V. 1989. A larval mite (Acarina: Ery-
thraeidae) parasitizing the European honey bee in
Guatemala. Acarologia, 30:123-129.

Southcott, R. V. 1991. A further revision of Char-
letonia (Acarina: Erythraeidae) based on larvae, pro-
tonymphs and deutonymphs. Invertebr. Taxon.,
5:61-131.

Southcott, R. V. 1 992. Revision of the larvae of Lep-
tus Latreille (Acarina: Erythraeidae) of Europe and
North America, with descriptions of post-larval in-
stars. Zool. J. Linn. Soc., 105:1-153.

Spoek, G. L, 1964. Spinachtigen-Arachnida. III. De
Hooiwagens (Opilionida) van Nederland. Wet. Med.
Koninklijke Nederlandse Natuurhist. Vereniging, 50:
1-28.

Sprague, V. 1977. Classification and phylogeny of the
Microsporidia. Pp. 1-30. Annotated list of species
of Microsporidia. Pp. 31-334, In Comparative pa-
thobiology. Vol. 2. Systematics of the Microspori-
dia. (L. A. Bulla, Jr. & T. C. Cheng, eds.). Plenum
Press, New York.

Steiner, G. 1922. Beitrage zur Kenntnis der Mer-
mithiden. Iste Teil. Mermithiden von Neu-Meck-
lenburg und Revision einiger v. Linstowscher Arten
und Rudolphis Filaria truncatula = Mermis trun-
catula. Centralbl. Bakrt. Jena, Abt. 1, 87:451-465.
(not seen)

Stiles, C. W. 1903. A parasitic roundworm (Aga-
momermis culicis n. g., n, sp.) in American mos-
quitoes {Culex sollicitans). Bull. Hygienic Lab., 13:
15-17.

Stipperger, H. 1928. Biologic und Verbreitung der
Opilioniden Nordtirols. Arbt. Zoolog. Inst. Univ.
Innsbruck, 3:19-79.

Thomas, G. M. & G. O. Poinar, Jr. 1979. Xenor-
habdus gen. nov., a genus of entomopathogenic, ne-
matophilic bacteria of the family Enterobacteri-
aceae. Int. J. System. BacterioL, 29:352-360.

Torgeson, D. C. (ed.) 1969. Fungicides. An advanced
treatise. Volume II. Chemistry and Physiology. Ac-
ademic Press, New York, 742 pp.

Tsurusaki, N. 1986. Parthenogenesis and geographic
variation of sex ratio in two species of Leiobunum
(Arachnida, Opiliones). Zool. Sci., 3:517-532.

Unzicker, J. D. & G. L. Rotramel. 1970. A new par-
asite record for Opiliones (Arachnida). Trans. Illi-
nois State Acad. Sci., 63:223-224.

Watson Kamm, M. 1922. Studies on gregarines II.
Synopsis of the polycystid gregarines of the World,
excluding those from Myriapoda, Orthoptera, and
Coleoptera. Illinois Biol. Monogr., 7:1-104.

Weiser, J. 1955. Neoaplectana carpocapsae n. sp.

146

THE JOURNAL OF ARACHNOLOGY

(Anguillulata, Steinemematinae), novy cizopasnik
housenek obalece jablecneho, Carpocapsa pomo-
nellaL. Vest.Ceskoslovenske Zool. SpoL, 19:44-52.

Welboum, W. C. 1983. Potential use of trombidiod
and erythraeoid mites as biological control agents
of insect pests. Pp. 103-140, In Biological control
of pests by mites. (M. A. Hoy, G. L. Cunningham
& L. Knutson, eds.). Agric. Exper. Stn., Div. Agric.
Nat. Resources, Univ. California, Berkeley. Spec.
Publ. 3304.

Welboum, W. C. & D. T. Jennings. 1991. Two new
species of Erythraeidae (Acari: Prostigmata) asso-
ciated with the spmce budworm, Choristoneura fu-

miferana (Clemens) (Lepidoptera: Tortricidae), in
Maine. Canadian Entomol, 123:567-580.

Wellmer, L. 1911. Sporozoen ostpmssischer Arthro-
poden. Schr. phys.-5kon. Ges. Konigsberg, 52:103-
164.

Yasuda, A. 1915. [“Kumotake (=spider fungi)” is a
new species]. Bot. Mag., Tokyo, 19:117. (in Japa-
nese)

[Yokogawa, K.] 1940. [Two observations on arach-
nids]. Acta ArachnoL, 5:145-147. (in Japanese)

Manuscript received 4 August 1992, revised 10 March
1993.

1993.. The Journal of Arachnology 21:147-151

RESEARCH NOTES

FIRST SCORPION (BUTHIDAE: CENTRUROIDES) FROM
MEXICAN AMBER (LOWER MIOCENE TO UPPER OLIGOCENE)

A juvenile scorpion, moderately well-pre-
served in Chiapas (Mexico) amber collected in
February 1992 came to our attention (Figs. 1, 2).
The relatively elongated metasomal segments
suggest that the specimen is a male, probably a
Centruroides (Marx 1890) (Buthidae). The spec-
imen, which lacks most of its right pedipalp, is
the first scorpion reported from Chiapas amber.
Centruroides beynai Schawaller 1979 (see also
Schlee 1980), Microtityus ambarensis (Scha-
waller 1982) (see also Santiago-Blay, Schawaller
& Poinar 1990; Schawaller 1984), and Tityus

geratus (Santiago-Blay & Poinar 1988) are known
from Dominican amber.

The piece containing the fossil is believed to
have originated from mines near the village of
Simojovel (State of Chiapas). The amber in the
Simojovel mines is located in a sequence of pri-
marily marine calcareous sandstones and silt with
beds of lignite. The amber-bearing strata extend
from the Balumtun Sandstone of the lower Mio-
cene to the La Quinta formation of the upper
Oligocene. These deposits have been assigned to
the planktonic foraminiferal zones N3 and N4

Figures 1, 2. — Centruroidesl sp. from Chiapas (Mexico) amber: 1. dorsal, overall; 2. ventral, overall.

147

148

THE JOURNAL OF ARACHNOLOGY

Figures 3-S.~Centruroidesl from Chiapas (Mexico) amber: 3, chelicera (see also Fig. 9), dorsal; 4, chelicera
of C. gracilis nymph, right, dorsal; 5, chelicera of C. gracilis nymph, left, ventral; 6, left pedipalp patella of C
nitidus nymph, ventral. Note microsetae (m); 7, Centruroidesl sp. from Chiapas (Mexico) amber. Mesosomal

RESEARCH NOTES

149

Figures 9, \0.—Centruroides? sp, from Chiapas (Mexico) amber. 9, chelicera; 10, pedipalp patella venter, dots
indicate possible microsetal (m) insertions. Scale lines = 1 mm.

in the Cenozoic Planktonic Foraminiferal Zonal
Sequence and radiometrically dated from 22.5-
26 million years (Berggren & Van Couvering
1 974). It should be noted that the amber deposits
are secondary; thus, the above dates provide a
minimum age.

Owing to the scarcity of published data on the
systematics and ontogeny of extant, juvenile
Centruroides and to the problems of interpreting
several important structures, which are con-
founded by imperfect preservation of this spec-
imen, the authors prefer to document the find

Figures 1 1~14.— 1 1, distal end of fixed pedipalp finger of Rhopalurus princeps (Karsch 1879) nymph. (The
distal most fixed finger trichobothria are dt, db, et, and est). Note rows of denticles on movable finger and
suggestion of supernumerary granules, sg); 1 2, Centruroides! sp. from Chiapas (Mexico) amber, left pedipalp
chela, dorsal; 1 3, Centruroides! sp. from Chiapas (Mexico) amber, retrolateral; 14, Centruroides! sp. from Chiapas
(Mexico) amber, fifth metasomal segment and telson.

keel-like structures (arrow head), dorsal; 8, Centruroides! sp. from Chiapas (Mexico) amber. Stemopectinal area.
Note apodeme-like structure (arrow head) of problematic interpretation— -an artifact?

150

THE JOURNAL OF ARACHNOLOGY

rather than describing this scorpion as a new
taxon, pending acquisition of additional speci-
mens from Chiapas.

Centruroidesl sp.

(Figs. 1--3, 7-10, 12-14)

Possibly a fourth instar juvenile male, 17.1
mm long, pale-yellowish brown, with pedipalp
chela and V metasomal segment dark brown,
suggestion of two longitudinal bands on meso-
somal tergites 2-3; possibly 8 primary rows of
denticles on pedipalp fingers, supernumerary
granules obsolete (Figs. 1 2, 1 3); pectines with 1 8™
1 9 teeth (Fig. 8); metasomal segments I-V with
10, 8, 8, 8, and 5 complete, crenulated keels; very
well-developed subaculear tooth (Fig. 14). Mea-
surements as in Table 1.

The chelicerae (Fig. 3), typical of the Buthidae,
are also depicted on Fig. 9. They resemble those
of similar-sized nymphs of the widespread Neo-
tropical scorpion, C gracilis (Gervais 1841) (Figs.
4, 5). The pedipalp patella venter shows several
pits that are interpreted as sites of microsetae
(Figs. 6, 10). Examination of these two structures
definitely eliminated the possibility that the scor-
pion was a vaejovid. Centruroides beynai has
similar pits on the pedipalp patella dorsum whose
interpretation has varied (Armas 1982; Armas
& Marcano Fondeur 1987; Schawaller 1979;
Santiago-Blay 1990).

However, the presence of what appears to be
submedial mesosomal keels (Fig. 7) and an apo-
deme-like structure on the prosomal sternum (Fig.
8) is puzzling. Some buthids mature at relatively
earlier instars, therefore the possibility that this
specimen is a small adult, although unlikely, can-
not be disregarded (Williams 1987).

This Centruroides from Chiapas can be distin-
guished from C. beynai by the pectine teeth num-
ber and metasomal ventral keel sculpturation:
the latter has 21-24 pectine teeth and smooth
ventral keels on metasomal segments II-IV.
Rhopalurus, another common Neotropical buth-
id genus with supernumerary pedipalp finger
granules, has a different arrangement of tricho-
bothria db and et (Fig. 11), relatively longer
metasomal segments, and lacks a definite
A-shaped (= inverted V) sulcus flanking a slightly
raised portion of mesosomal stemite III. The
current knowledge of the Centruroides fauna of
the region precludes us from distinguishing this
specimen from many of its extant congenerics.

Since 1987, the authors have been accumu-
lating morphometric data on fossil scorpions

Table 1. — Measurements of fossil Centruroides! sp.
from Chiapas (Mexico) amber. All measurements in
mm. Some measurements could not be obtained be-
cause of positioning of the specimen in the piece.

Character

Measurements

(mm)

Prosoma

Carapace

Anterior, median,

posterior widths

1.0, 1.5, 2.3

Diad width

0.4

Median, diad-front

margin length

2.0, 0.8

Chelicera

Basal piece width,

fixed finger lengths

1.4, 0.2

Pedipalp

Femur length, width

1.9, 0.5

Patella length, width

2.5, 0.6

Palm length, width

1.1, 0.5, 2.3

Palm underhand, movable

finger lengths

0.9, 2.2

Sternum

Length, anterior.

posterior widths

0.6, 0.2, 0.5

Mesosoma

Terga

I length, width

0.3, 2.0

II length, width

0.4, 1.9

III length, width

0.5, 2.2

IV length, width

0.6, 2.4

V length, width

0.7, 2.3

VI length, width

0.6, 2.0

VII length, anterior.

posterior widths

1.0, 2.1, 1.0

Overall length

3.7

Metasoma

Segments

I length, width

1.5, 1.0

II length, width

1.8, 1.0

III length, width

1.9, 1.0

IV length, width

2.3, 1.0

V length, width, depth

1.9, 0.7, 0.8

Telson

Vesicle length, depth

0.9, 0.4

Aculeus length

0.1

Total length

17.1

preserved in amber and hope to create a data
base that will ease identifications, particularly
when only parts of specimens have been pre-
served. The present specimen is maintained in
the private collection of Dr. Rodolfo Molina.

RESEARCH NOTES

151

Author GOP can be contacted for further infor-
mation on the piece.

We thank Stanley C. Williams and Vince F.
Lee (Department of Entomology, California
Academy of Sciences, San Francisco) for offering
suggestions on an earlier version of the typescript
and for helpful discussions with author JASB.
Grateful appreciation is extended to Dr. R. Mo-
lina who brought the specimen to Berkeley for
study.

LITERATURE CITED

Armas, L. F. de. 1982. Adiciones a las escorpiofaunas
(Arachnida: Scorpiones) de Puerto Rico y de la Re-
publica Dominicana. Poeyana, 237:1-25.

Armas, L. F. de & E. J. Marcano Fondeur. 1987.
Nuevos escorpiones de Republica Dominicana.
Poeyana, 356:1-24.

Berggren, W. A. & J. A. Van Couvering. 1974. The

late Neogene. Paleogeogr. Paleoclimat. PaleoecoL,
16:1-216.

Marx, G. 1890. Arachnida. In The scientific results
of explorations by the U.S. Fish Commission Steamer
Albatross. (L. O. Howard, ed.). No. V. Annotated
catalogue of the insects collected in 1887-88. Proc.
United States Natl. Mus., 12:185-216.
Santiago-Blay, J. A. 1990. Systematics and some as-
pects of the biology of the scorpions (Arachnida:
Scorpiones) of Hispaniola (Dominican Republic and
Haiti), West Indies. Ph.D. Dissertation. Univ. Cal-
ifornia, Berkeley. 277 pp.

Santiago-Blay, J. A. & G. O. Poinar, Jr. 1 988. A fossil
scorpion, Tityus geratus new species (Scorpiones:

Buthidae) from Dominican amber. Hist. Biol.,
1:345-354.

Santiago-Blay, J. A., W. Schawaller & G. O. Poinar,
Jr. 1990. A new specimen of Microtityus ambar-
ensis (Schawaller 1982) (Scorpiones: Buthidae), fos-
sil scorpion from Hispaniola: evidence of the tax-
onomic status and possible biogeographic
implications. J. ArachnoL, 18:115-117.

Schawaller, W. 1979. Ersnachweis eines Skorpions
in Dominikanischem Bernstein (Stuttgarter Bem-
steinsammlung: Arachnida; Scorpionida). Stutt-
garter Beitr. Naturkd. Ser. B (Geol. Palaontol.), 45:
1-15.

Schawaller, W. 1982. Zwei weitere Skorpione in
Dominikanischem Bernstein (Stuttgarter Bemstein-
sammlung: Arachnida; Scorpionida). Stuttgarter
Beitr. Naturkd. Ser. B (Geol. Palaontol.), 82:1-14.

Schawaller, W. 1984. Spinnentiere (Arachnida) im
Dominiskanischen Bernstein. Pp. 72-78, In Bern-
stein-Neuigkeiten. Stuttgarter Beitr. Naturkd. Ser.
C, Nr. 18. 100 pp.

Schlee, D. 1980. Bemstein-Raritaten. Farben. Struk-
turen. Fossilen. Handwerk. Bernstein. Staat. Mus.
Naturk. Stutt. 88 pp.

Williams, S. C. 1987. Scorpion Bionomics. Ann. Rev.
Entomol., 32:275-295.

Jorge A. Santiago-Blay and George O. Poinar,
Jr.: Department of Entomological Sciences,
University of California, Berkeley, California
94720-0001 USA.

Manuscript received 24 September 1992, revised
1 April 1993.

1993. The Journal of Arachnology 21:152

THE FEMALE OF GALLIENIELLA BETROKA
(ARANEAE, GALLIENIELLIDAE)

The spider genus Gallieniella is known only 0.05, ALE-PLE 0.07; MOQ length 0.20, front

from Madagascar and the Comoro Islands. Four
species have been described (Platnick 1984), but
two of them, G. bland and G. betroka, have been
known only from males. Through the courtesy
of Dr. Hubert Hofer of the Landessammlungen
fiir Naturkunde de Karlsruhe, Germany (LNK),
I’ve recently been able to examine a fine series
of G. betroka that includes the first known fe-
males of the species, described below (in the for-
mat used in the revision). I thank Dr. Moham-
mad U. Shadab of the American Museum of
Natural History for providing the illustrations.

Gallieniella betroka Platnick
Figs. 1, 2

Gallieniella betroka Flatnick, 1984: 10.

Diagnosis. —Females can easily be distin-
guished from those of G, mygaloides Millot and
G. jocquei Platnick by having epigynal ducts sit-
uated anteriorly of the spermathecae (Figs. 1 , 2).

Male. —Described by Platnick (1984).

Female.— As in male, except for the following.
Total length, not including chelicerae, 4.91 mm.
Carapace 2.06 long, 1.69 wide. Femur II 1.49
wide. Eye sizes and interdistances: AME 0.07,
ALE 0.07, PME 0.06, PLE 0.06; AME- AME 0.09,
AME-ALE 0.02, PME-PME 0.15, PME-PLE

Figures 1, 2. — Gallieniella betroka Platnick,

width 0.23, back width 0.28. Clypeal height at
AME only slightly greater than their diameter.
Chelicerae extending forward distance about
three-fifths of carapace length, without ventral
tubercle on fang; white scales restricted to pars
thoracica. Leg spination: tibia IV v2-3-2. Femur
I with lateral stripes more obvious than on femur
IT Epigynal ducts relatively long (Figs. 1 , 2).

Material Examined.-MADAGASCAR: Tulear: Foret
de Kirindy, Morondava, March 5-19, 1990 (LNK), 23,
29; Trockenwald, Morondava, March 7~15, 1991, pit-
fall trap (Butterweck, Petzold, LNK), 23, 29.

Distribution. — Known only from southern
Madagascar.

LITERATURE CITED

Platnick, N. 1. 1984. Studies on Malagasy spiders, I.
The family Gallieniellidae (Araneae, Gnaphoso-
idea). American Mus. Novit., 2801:1-17.

Norman I. Platnick: Department of Entomol-
ogy, American Museum of Natural History,
Central Park West at 79th Street, New York,
New York 10024 USA.

Manuscript received 19 January 1993, revised 1 March
1993.

!, epigynum: 1, ventral view; 2, dorsal view.

152

1993. The Journal of Arachnology 21:153-155

NATURAL HISTORY NOTES ON THE HUNTSMAN SPIDER
HOLCONIA IMMANIS (ARANEAE, HETEROPODIDAE)

Our ecological knowledge of huntsman spiders
of the family Heteropodidae is very limited. Even
the taxonomy of this family has been worked out
poorly. The last complete revision dates back to
Hogg ( 1 903). The recently resurrected genus Hol-
conia ranges over most of mainland Australia
(Hirst 1990), and Holconia immanis (Koch) is
found in eastern Australia from Queensland to
Victoria. H. immanis is a large spider: males
attain a body length of 30 mm, and females one
of 47 mm (Mascord 1970).

New observations of this species were made
during a population ecological study of the ar-
boreal gecko Gehyra variegata (Dumeril & Bi-
bron) (Henle 1990) in Kinchega National Park
(32°28'S, 142®20'E), western New South Wales,
Australia, from September 1985 to May 1987.
Voucher specimens are deposited in the National
Insect Collection, Commonwealth Scientific and
Industrial Research Organization, Canberra. I
thank R. Moran and D. Russell for identification
of the spiders.

Gehyra variegata and H. immanis were found
primarily in black box {Eucalyptus largiflorens)
riverine woodland on heavy textured cracking
clay and in low numbers on a red sand dune
covered by hopbush {Dodonaea attenuata). In
contrast to the abundant gecko, three adult H.
immanis lived at the huts of Kinchega station in
1986-87. However, several specimens were found
on an old brick building at Mt. Wood Station,
Sturt National Park. H. immanis was not found
on river red gum {E. camaldulensis) in Kinchega
nor at Mt. Wood Station.

Detailed observations on adult females (adults
usually determined by size alone: body length >
30 mm) were made in a 1 50 x 100 m study plot
in riverine woodland with 4 1 widely spaced black
box trees. H. immanis and C. variegata use the
same microhabitat. They were found primarily
on the trunks of trees or on large branches. The
correlation of the number of adult females seen
per tree with eight microhabitat variables (height,
diameter, number of trunks, leaf area index,
number of potential retreat sides, distance to the

next tree, food availability, and number of G.
variegata seen - see Henle [1990] for details of
methods) was tested. None of the product-mo-
ment-correlations (-0.16 < r < 0.40) was sig-
nificant (all a > 0.05). However, in another more
elevated study site of black box riverine wood-
land, Henle (1990) found a significant correlation
of the number of adult female spiders with tree
diameter and a marginally significant correlation
with tree height and leaf area index. Thus, in this
study site, larger trees tend to harbor more spec-
imens.

Both species are typical sit-and-wait foragers.

Adult H. immanis seem to have 1--2 preferred
ambush sites where most individuals were ob-
served on many consecutive nights up to a period
of 6 months. The capture of ten food items was
witnessed: 1 Pauropoda, 1 Chilopoda (Fig. 1), 2
Lycosidae, 1 subadult H. immanis, 1 Phasmida,

1 Heteroptera, 2 Coleoptera, and 1 Lepidoptera.
The size of the prey ranged from approximately
0.75 cm to > 10 cm. One unsuccesful predation
attempt on a juvenile Gehyra variegata was ob-
served. The gecko was not pursued for more than
2-3 cm.

Active specimens were found in all months
between September and May. They were inactive
in July and August. The recapture of marked
specimens showed that H. immanis overwinters
as adults as is the case in Clubiona robusta (Koch)
of South Australia (Austin 1984) but contrasts
to spiders of colder climates in the Northern
Hemisphere which overwinter mainly as eggs
(Turnbull 1973).

In September and November 1986 and Jan-
uary 1987, a mark-recapture study was under-
taken in the 150 x 100 m study plot. Adult
females (body length >30 mm) were marked on
different legs with two colors of nail paint. The
marking was visible at least for three weeks,
sometimes for two months, and lasted through
hibernation (four months). The mark-recapture
data of 10 consecutive days were fitted to the
geometric and the Poisson distributions (Caugh-
ley 1980). The fit to the Poisson distribution was

153

154

THE JOURNAL OF ARACHNOLOGY

Figure \.—Holconia immanis preying upon a large centipede.

poor in all three cases (x^ > 1 .949; a < 0.2) while
the geometric distribution fitted the data well (x^
< 0.218; a > 0.5). Thus, the geometric distri-
bution was used to estimate population size.
Confidence intervals (Cl) were calculated ac-
cording to Henle (1983). The estimated number
of adult female H. immanis was 35 (95%“CI:
26-45), 22 (95%-CI: 14-30), and 19 (95%-CI:
11-27) for September 1986, November 1986,
and January 1987, respectively. Thus, there ap-
pears to have been considerable mortality of adult
females between September and November 1986.
The mortality of approximately 50% during four
months suggests that adult females live for ap-
proximately 4-16 months. Miller & Miller (1 99 1)
found a similar yearly survivorship of Geolycosa
turricola (Treat). Only two of the 47 marked in-
dividuals with a size of > 3 cm changed the tree

of original capture and moved 30 m and 15 m
within two and three days, respectively.

LITERATURE CITED

Austin, A. D. 1984. Life history of Clubiona robusta
L. Koch and related species (Araneae, Clubionidae)
in South Australia. J. ArachnoL, 12:87-104.

Caughley, G. 1980. Analysis of Vertebrate Popula-
tions. Wiley, New York.

Henle, K. 1983. Populationsbiologische und -dy-
namische Untersuchungen am Wiesenpieper {An-
thus pratensis) auf der Insel Mellum. Vogel warte,
32:57-76.

Henle, K. 1990. Population ecology and life history
of the arboreal gecko Gehyra variegata in arid Aus-
tralia. Herpetol. Monog., 4:30-60.

Hirst, D. B. 1990. A review of the genus Isopeda L.
Koch (Heteropodidae: Araneae) in Australasia, with

RESEARCH NOTES

155

descriptions of two new genera. Rec. South Austra-
lian Mus. (Adelaide), 24:11-26.

Hogg, H. R. 1903. On Australasian spiders of the
subfamily Sparassinae. Proc. ZooL Soc. London,
1902:414-466.

Mascord, R. 1970. Australian Spiders in Colour. Reed,
Wellington.

Miller, P. R. & G. L. Miller. 1991. Dispersal and
survivorship in a population of Geolycosa turrkola
(Araneae, Lycosidae), J. Arachnol., 19:49-54.

Turnbull, A. L. 1973. The ecology of true spiders
(Araneomorphae). Ann. Rev. EntomoL, 18:305-348.

Klaus Henle: Projektbereich Natumahe Land-
schaften, Umweltforschungszentram Leipzig-
Halle GmbH, Permosertstr. 15, I>-(0) 7050
Leipzig, Germany

Manuscript received 6 March 1992, revised 20 January
1993.

1993. The Journal of Arachnology 21:156-158

MATING BIOLOGY RESOLVES TRICHOTOMY
FOR CHELIFEROID PSEUDOSCORPIONS
(PSEUDOSCORPIONIDA, CHELIFEROIDEA)

Mating behavior and spermatophore mor-
phology have provided phylogenetically useful
information for both vertebrate and invertebrate
taxa (e. g,, Proctor 1992a; Alberti et al. 1991;
Prum 1990). However, because they are transi-

tory phenomena that cannot be observed in pre-
served specimens, behavior and fragile ejaculates
are seldom employed in phylogenetic studies.
This is unfortunate because they often contain
characters potentially helpful for resolving po-

Table 1 . —Spermatophore morphology, mating behavior and male morphology in cheliferoid pseudoscorpions.
(+) = character present; (-) = character absent.

Characters

Family

Species

6 pulls 9

Sperma- over
tophore sperma-
droplet tophore

Ram's

horn

organs

6 pushes
sperm in
9'S

genital

opening

Reference

Chemetidae

Epactiochernes tumidus
(Banks)

+

+

Weygoldt 1966a

Cherries cimicoides
(Fabricius)

+

+

_

_

Weygoldt 1966b

Dendrochernes morosus
(Banks)

+ +

Weygoldt 1970

Lustrochernes pennsylvanicus
(Ellingsen)

T

+

Weygoldt 1970

Americhernes oblongus
(Say)

+

+

Weygoldt 1970

Parachernes litoralis

Muchmore & Alteri

+

—

—

Weygoldt 1970

Cheliferidae

Dactylochelifer latreillei
(Leach)

+ -

+

+

Weygoldt 1966b

Chelifer cancroides
(Linnaeus)

4- -

+

4-

Weygoldt 1966b

Rhacocheiifer disjunctus
(Koch)

+

4-

4-

Weygoldt 1970

Hysterochelifer meridianus
(Koch)

+ -

4-

4-

Weygoldt 1970

Hysterochelifer tuberculatus
(Lucas)

+ -

4-

+

Weygoldt 1970

Paracheiifer superbus

Hoff

+ -

4-

4-

Weygoldt 1970

Atemnidae

Paratemnoides braunsi

(Tullgren)

+

Weygoldt 1970

Atemnus politus
(Simon)

- +

—

—

Weygoldt 1969a

Withiidae

Withius subpiger

Simon

- +

Weygoldt 1969b

156

RESEARCH NOTES

157

lychotomies that have proven intractable to tra-
ditional morphological approaches. As well, un-
like the usual alternatives of electrophoretic or
DNA analyses, characters resulting from studies
of mating biology are evolutionarily interesting
in themselves.

Because of the diversity of sperm transfer be-
havior and spermatophore morphology present
in pseudoscorpions (Weygoldt 1 969a), this group
is likely to respond well to phylogenetic resolu-
tion using mating characters. In his recent cla-
distic study, Harvey (1992) ascribed two repro-
ductive synapomorphies to the superfamily
Cheliferoidea: production of spermatophores with
complex rather than simple sperm masses and
transfer of sperm during mating dances rather
than without pairing between the sexes. Within
the Cheliferoidea, the families Cheliferidae,
Chemetidae and Atemnidae were differentiated
from the Withiidae by synapomorphies of leg
morphology; however, Harvey found no char-
acters to resolve the trichotomy formed by the
first three families. In recent literature reviews
(Proctor 1992b) I turned up several features of
cheliferoid mating biology that both help to re-
solve this trichotomy and suggest adaptive sce-
narios for the evolution of mating behavior in
this superfamily.

Table 1 lists characteristics of spermatophore
morphology, male anatomy and mating behavior
for species in the four families of the Chelifer-
oidea. Spermatophore stalks of the Chemetidae
and Cheliferidae apomorphically possess a large
droplet of apparently hypotonic liquid that caus-
es the sperm packet to swell and expel its con-
tents into the female genital atrium (Weygoldt
1975). This synapomorphy allows the Chelifer-
oidea to be resolved from Harvey’s (1992) ar-
rangement of [Withiidae (Cheliferidae + Cher-
netidae + Atemnidae)] to [Withiidae (Atemnidae
(Cheliferidae + Chemetidae))]. Other aspects of
mating biology provide phylogenetic and evo-
lutionary insight. Males of the Cheliferidae apo-
morphically possess genital sacs (ram’s horn or-
gans) that are everted after spermatophore
deposition to attract the females, presumably
through pheromones on their surface (Weygoldt
1 969a). Concomitant with the evolution of ram’s
horn organs is loss of the male behavior of pulling
females over spermatophores, which is present
in the other three cheliferoid families (Table 1),
This suggests that pheromonal attraction of the
female replaced physical manipulation in the
Cheliferidae and raises the possibility that chem-

ical guidance has some advantage over physical
contact for these males {e. g. , reduced likelihood
of palp damage, greater guarantee of female in-
terest in mating). Another apotypic behavior in
the Cheliferidae is the male’s use of his forelegs
to push sperm into the female genital opening
after she has mounted the spermatophore (Table
1); no other cheliferoids do this, although there
is often extended contact between male and fe-
male after the female takes up sperm {e. g., Wey-
goldt 1970). Adaptive explanations for these and
other reproductive characters will be possible only
after studying their effects on male fitness.

This project was supported by a Natural Sci-
ences and Engineering Research Council of Can-
ada Postdoctoral Fellowship. I am grateful to
Mark Harvey for his encouragement and gen-
erosity with his unpublished phylogeny. The
manuscript was improved by comments from M.
Harvey and V. F. Lee.

LITERATURE CITED

Alberti, G., N. A. Fernandez & G. Kummel. 1991.
Spermatophores and spermatozoa of oribatid mites
(Acari: Oribatida). Part II: Functional and syste-
matical considerations. Acarologia, 32:435-449.
Harvey, M. S. 1 992. The phylogeny and classification
of the Pseudoscorpionida (Chelicerata: Arachnida).
Invert. Taxonomy, 6:1373-1435.

Proctor, H. C. 1992a. Sensory exploitation and the
evolution of male mating behaviour: a cladistic test
using water mites (Acari: Parasitengona). Anim. Be-
hav., 44:745-752.

Proctor, H. C. 1992b. The evolution of sperm transfer
behaviour in water mites (Acari: Parasitengona). Ph.
D. dissertation; University of Toronto, Toronto,
Ontario, Canada.

Prum, R. O. 1990. Phylogenetic analysis of the evo-
lution of display behavior in the Neotropical man-
akins (Aves: Pipridae). Ethology, 84:202-231.
Weygoldt, P. 1966a. Mating behaviour and sper-
matophore morphology in the pseudoscorpion Din-
ocheirus tumidus Banks (Cheliferinea, Chemetidae).
Biol. Bull., 130:462-467.

Weygoldt, P. 1966b. Vergleichende Untersuchungen
zur Fortpflanzungsbiologie der Pseudoskorpione.
Beobachtungen iiber das Verhalten, die Sameniiber-
tragungsweisen und die Spermatophoren einiger
einheimischen Arten. Z, Morph. Okol. Tiere, 56:
39-92.

Weygoldt, P. 1969a, The Biology of Pseudoscor-
pions. Harvard University Press, Cambridge.
Weygoldt, P. 1969b. Paamngsverhalten und Samen-
iibertragung beim Pseudoskorpion Withius subruber
Simon (Cheliferidae). Z. Tierpsychol., 26:230-235.
Weygoldt, P. 1970. Vergleichende Untersuchungen

158

THE JOURNAL OF ARACHNOLOGY

zur Fortpflanzungsbiologie der Pseudoskorpione IF
Z. ZooL Syst. Evolutionsforsch., 8:241--259.
Weygoldt, P. 1975. Die indirekte Spermatophoren-
iibertragung bei Arachniden. Verb. Deutschland,
Zool. Ges., 1974:308-313.

Heather C. Proctor: Department of Biological
Sciences, University of Calgary, Calgary, Al-
berta T2N 1N4, Canada.

Manuscript received 20 November 1 992, revised 5 March
1993.

1993. The Journal of Arachnology 21:159-160

BOOK REVIEW

Harvey, M. S. 1991. Catalogue of the Pseu-
doscorpionida. (edited by V. Mahnert). Man-
chester University Press, Manchester Ml 3 9 PL
UK (distributed exclusively in the USA and Can-
ada by St. Martin’s Press Inc., 175 Fifth Avenue,
New York, New York 10010 USA), vi + 726
pages. Price $200.00.

It is a pleasure to write about a book which I
use almost every day, an indispensable tool for
the pseudoscorpion taxonomist. Both the author
and the editor deserve many thanks for a job
well done.

Harvey’s Catalogue is the first comprehensive
work on the pseudoscorpions of the world since
the monographs of Beier (1932a, b) and the lists
of Roewer (1937, 1940). Only its Bibliography
has a modern counterpart in the bibliography of
Scha waller (1980).

An Introduction in three languages (English,
French, and German) explains the organization
of the book, the systematic treatment of the in-
cluded taxa, and the limited number of taxo-
nomic changes that have been introduced. It is
revealed that 22 families, 434 genera, 3064 spe-
cies and 169 subspecies of pseudoscorpions are
recognized.

The first working section of the book is the
Bibliography of nearly 2700 entries. Most ref-
erences deal with systematic matters, but other
aspects of pseudoscorpion biology are included
as well, such as behavior, biogeography, ecology,
histology, morphology, reproduction, develop-
ment and life history. The list of publications
covers the years from 1758-1988, with a few
references to papers published in 1989-90. As
far as I have found, very few references to specific
pseudoscorpions have been omitted, none of great
importance. A valuable feature of the entries is
the citation of the names of journals in full -
none of the tricky abbreviations which often
prove difficult to track down.

The catalogue itself is in the form of a list of
all valid species (including fossils) through 1988
(plus a few later ones). The taxonomic arrange-
ment follows a recommendation by me (1982)
to forego subordinal groupings and treat only
superfamilies. The arrangement of the superfam-

ilies is essentially that of Chamberlin (1931) and
Beier (1932a, b), except that the Feaelloidea are
placed near the Chthonioidea because of per-
ceived relationships. Within each superfamily the
families are listed alphabetically, as are the gen-
era in each family and the species and subspecies
in each genus.

Each taxon is documented by a complete syn-
onymy, including reference to papers cited in the
Bibliography, with relevant pages and (for spe-
cies) figures. As the bibliography is essentially
complete, so the synonymies are essentially com-
plete. These synonymies can serve to change the
shape of papers on pseudoscorpion taxonomy
from this time on. No longer will long synony-
mies be needed for each known species treated
when a simple “for synonymy see Harvey 1991”
will do.

For each genus, the type species is noted; and
for each species, the type locality and reported
distribution are given. In a future edition, valu-
able additions would be the mention, where pos-
sible, of the depository of the type specimen(s)
of each species and the sex(es) known for each.
Some of this information will be difficult or im-
possible to obtain, but it would be good to begin
accumulating such data.

Following the list of those he considered valid
species, Harvey presents a list of nomina dubia
and nomina nuda. Included here are two genus-
group names and 45 species-group names. In ad-
dition, two forms and one variety are mentioned.
Then follows a summary of taxonomic changes
introduced in the body of the catalogue, includ-
ing five replacement names, six new type species,
five new synonymies, and 101 new combina-
tions.

The Index includes all names appearing in the
text. Family-group names and junior synonyms,
junior homonyms, nomina dubia, and nomina
nuda are distinguished by differences in type-
face, The genus in which a species was originally
described is clearly indicated.

Altogether, this is a very valuable book, com-
prehensive in content and easy to use. It is ab-
solutely required for the pseudoscorpion taxon-
omist and should be consulted by anyone dealing
with any aspect of the biology of these animals.
The price of $200 is a bit steep, but for one (or

159

160

THE JOURNAL OF ARACHNOLOGY

a group) who is seriously interested in pseudo-
scorpions, this catalogue will be worth every
penny.

Unfortunately, the Catalogue will not remain
current for long. Already, Harvey (1992, 1993)
has introduced changes in the systematics of the
Pseudoscorpionida which will necessitate major
rearrangement of some families and genera. Oth-
er workers will, undoubtedly, be stimulated into
action by Harvey’s ideas, and amendment of this
edition will soon be required.

LITERATURE CITED

Beier, M. 1932a. Pseudoscorpionidea 1. Subord.

Chthoniinea et Neobisiinea. Tierreich, 57:1-258.
Beier, M. 1932b. Pseudoscorpionidea 11. Subord. C.

Cheliferinea. Tierreich, 58:1-294.

Chamberlin, J. C. 1931. The arachnid order Chelo-
nethida. Stanford Univ. Publ. Biol. Sci., 7:1-284.
Harvey, M. S. 1 992. The phylogeny and classification
of the Pseudoscorpionida (Chelicerata: Arachnida).
Invertebr. Taxon., 6:1373-1435.

Harvey, M. S. 1993. The systematics of the Hyidae
(Pseudoscorpionida: Neobisioidea). Invertebr. Tax-
on., 7:1-32.

Muchmore, W. B. 1982. Pseudoscorpionida, Vol. 2,
pp. 96-102, In Synopsis and classification of liv-
ing organisms (S. P. Parker, ed.). McGraw-Hill Book
Co. New York.

Roewer, C. F. 1937, 1940. Chelonethi oder Pseu-
doskorpione. 5 (4) (6) (2, 3): 16 1-354, In Klassen
und Ordnungen des Tierreichs (H. G. Bronns, ed.),
Leipzig.

Schawaller, W. 1980. Bibliographie der rezenten und
fossilen Pseudoscorpionidea 1890-1979 (Arachni-
da). Stuttgarter Beitr. Naturk. (A), 338:1-61.

William B. Muchmore: Department of Biolo-
gy, University of Rochester, Rochester, New
York 14627 USA.

Manuscript received 26 April 1993.

INSTRUCTIONS TO AUTHORS

Manuscripts are preferred in English but may be ac-
cepted in Spanish, French or Portuguese subject to avail-
ability of appropriate reviewers. Authors whose primary
language is not English may consult the Associate Editor
for assistance in obtaining help with English manuscript
preparation. All manuscripts should be prepared in general
accordance with the current edition of the Council of Bi-
ological Editors Style Manual unless instructed otherwise
below. Authors are advised to consult a recent issue of the
Journal of Arachnology for additional points of style.
Manuscripts longer than 1500 words should be prepared
as Feature Articles, shorter papers as Research Notes. Send
three identical copies of the typed material together with
copies of illustrations to the Associate Editor of the Journal
of Arachnology:

Gary L. Miller, Associate Editor
Dept, of Biology
The University of Mississippi
University, Mississippi 38677 USA
[Telephone: (601)-232-7495; FAX: (601)-232-5144;
Electronic mail: BYMILLER@UM.CC.OLEMISS.EDU].
Correspondence relating to manuscripts should also be
directed to the Associate Editor. After the manuscript has
been accepted, the author will be asked to submit the
manuscript on a computer disc in MS DOS WordPerfect
(preferably) or ASCII text file.

FEATURE ARTICLES

Title page.— The title page will include the complete
name, address, and telephone number of the author with
whom proofs and correspondence should be exchanged, a
FAX number and electronic mail address if available, the
title in capital letters, and each author’s name and address,
and the running head (see below).

Abstract.— The heading in capital letters should be placed
at the the beginning of the first paragraph set off by a
period. In articles written in English, a second abstract in
an acceptable language may be included pertinent to the
nationality of the author(s) or geographic region(s) em-
phasized in the article. A second abstract in English must
be included in articles not written in the latter language.

Text,— Double-space text, tables, legends, etc. through-
out. Three levels of heads are used. The first level (METH-
ODS), RESULTS, etc.) is typed in capitals and on a sep-
arate line. The second level head begins a paragraph with
an indent and is separated from the text by a period and
a dash. The third level may or may not begin a paragraph
but is underlined and separated from the text by a colon.
Use only the metric system unless quoting text or refer-
encing collection data. All decimal fractions are indicated
by the period regardless of language of the text.

Citation of references in the text: Cite only papers already
published or in press. Include within parentheses the sur-
name of the author followed by the date of publication. A
comma separates multiple citations by the same author(s)
and a semicolon separates citations by different authors,
e.g., (Smith 1970), (Jones 1988; Smith 1993), (Smith 1986,
1987; Smith & Jones 1989; Jones et al. 1990).

Literature cited section.— Use the following style:
Lombardi, S. J. & D. L. Kaplan. 1990. The amino acid
composition of major ampullate gland silk (dragline) of
Nephila clavipes (Araneae, Tetragnathidae). J. Arach-
nol., 18:297-306.

Krafft, B. 1 982. The significance and complexity of com-
munication in spiders. Pp. 15-66, In Spider Commu-
nications: Mechanisms and Ecological Significance. (P.
N. Witt & J. S. Rovner, eds.). Princeton University
Press, Princeton, New Jersey.

Footnotes.— Footnotes are permitted only on the first
printed page to indicate current address or other infor-
mation concerning the author. All footnotes are placed
together on a separate manuscript page.

Running head.— The author’s surname(s) and an abbre-
viated title should be typed all in capital letters and must
not exceed 60 characters and spaces. The running head
should be placed near the top of the title page.

Taxonomic articles.— Consult a recent taxonomic article
in the Journal of Arachnology for style or contact the As-
sociate Editor.

Tables.— Each table, with the legend above, should be
placed on a separate manuscript page. Only horizontal
lines (usually three) should be included. Use no footnotes;
instead, include all information in the legend. Make no-
tations in the text margins to indicate the preferred loca-
tion of tables in the printed text.

Illustrations.— Address all questions concerning illus-
trations to the Editor of the Journal of Arachnology:

James W. Berry, Editor

Dept, of Biological Sciences

Butler University

Indianapolis, Indiana 46208 USA

[Telephone (317)-283-9344; FAX (317)-283-9519;

Electronic mail: BERRY@BUTLERU]

All art work must be camera-ready for reproduction. In
line drawings, pay particular attention to width of lines
and size of lettering when reductions are to be made by
the printer. Multiple photos assembled on a single plate
should be mounted with only a minimum of space sepa-
rating them. In the case of multiple illustrations mounted
together, each illustration must be numbered sequentially
rather than given an alphabetic sequence. Written on the
back should be the name(s) of author(s) and an indication
of top edge. The author should indicate whether the illus-
tration should be one column or two columns in width.
The overall dimensions should be no more than 1 1 inches
(28 cm) X 14 inches (36 cm). Larger drawings present
greater difficulty in shipping and greater risks of damage
for which the JOA assumes no responsibility. Make no-
tations in the text margins to indicate the preferred po-
sition of illustrations in the printed text.

Legends for illustrations should be placed together on
the same page(s) and separate from the illustrations. Each
plate must have only one legend, as indicated below:

Figures \-A.—A-us x-us, male from Timbuktu: 1. left
leg; 2. right chelicera; 3. dorsal aspect of genitalia; 4. ven-
tral aspect of abdomen.

Figures 27-34.— Right chelicerae of species oIA-us from
Timbuktu: 27, 29, 31, 33. dorsal views; 28, 30, 32, 34,
prolateral views of moveable finger; 27, 28. A- us x-us
holotype male; 33,34. A-us y-us male. Scale 1.0 mm.

Assemble manuscript for mailing.— Assemble the sep-
arate sections or pages in the following sequence; title page,
abstract, text, figure legends, footnotes, tables with legends,
figures.

Page charges and reprints.— The current charge per
journal page will be assessed as follows: Manuscripts in
the appropriate word processing program $ 20/page; others
$3 5/page. Reprints are available only from the printer and
should be ordered when the author receives the galley
proof pages. Waiver of page charges: In exceptional cases,
the Editor will consider requests for a waiver of page charg-
es for manuscripts which have been accepted for publi-
cation. Requests should be made in writing to the Editor.
The author must certify that neither grant nor institutional
funds are available to cover all of the publication costs.
The Society will normally waive no more than 80% of the
page charges.

RESEARCH NOTES

Instructions above pertaining to feature articles apply
also to research notes, except that abstracts and most head-
ings are not used and the author’s name and address follow
the Literature Cited section.

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 21 Feature Articles NUMBER 2

The genus Troglosiro and the new family Troglosironidae (Opiliones,

Cyphophthalmi), William A. Shear 81

The influence of prey availability and habitat on activity patterns and abun-
dance of Argiope keyserlingi (Araneae: Araneidae), Richard A. Bradley 9 1

Constraints and plasticity in the development of juvenile Nephila clavipes

in Mexico, Linden Higgins 107

Review Article

Pathogens and parasites of Opiliones (Arthropoda: Arachnida), James C.

Cokendolpher 120

Research Notes

First scorpion (Buthidae: Centruroides) from Mexican amber (Lower Mio-
cene to Upper Oligocene), Jorge A. Santiago- Blay and George O.
Poinar, Jr, 147

The female of Gallieniella betroka (Araneae, Gallieniellidae), Norman L

Platnick 152

Natural history notes on the huntsman spider Holconia immanis (Araneae,

Heteropodidae), Klaus Henle 153

Mating biology resolves trichotomy for cheliferoid pseudoscorpions (Pseu-

doscorpionida, Cheliferoidea), Heather C. Proctor 156

Book Review

Catalogue of the Pseudoscorpionida, William B. Muchmore 159

The Journal of
ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 21

1993

NUMBER 3

THE JOURNAL OF ARACHNOLOGY

EDITOR: James W. Berry, Butler University

ASSOCIATE EDITOR: Gary L. Miller, The University of Mississippi

EDITORIAL BOARD: A. Cady, Miami (Ohio) Univ. at Middletown; J. E.
Carrel, Univ. Missouri; J. A. Coddington, National Mus. Natural Hist.; J. C.
Cokendolpher, Lubbock, Texas; F. A. Coyle, Western Carolina Univ.; C. D. Don-
dale, Agriculture Canada; W. G. Eberhard, Univ. Costa Rica; M. E. Galiano,
Mus. Argentino de Ciencias Naturales; M. H. Greenstone, BCIRL, Columbia,
Missouri; C. Griswold, Calif. Acad. Sci.; N. V. Horner, Midwestern State Univ.;
D. T. Jennings, Garland, Maine; V. F. Lee, California Acad. Sci.; H. W. Levi,
Harvard Univ.; E. A. Maury, Mus. Argentino de Ciencias Naturales; N. I. Plat-
nick, American Mus. Natural Hist.; G. A. Polis, Vanderbilt Univ.; S. E. Riechert,
Univ. Tennessee; A. L. Rypstra, Miami Univ., Ohio; M. H. Robinson, U.S.
National Zool. Park; W. A. Shear, Hampden-Sydney Coll.; G. W. Uetz, Univ.
Cincinnati; C. E. Valerio, Univ. Costa Rica.

The Journal of Arachnology (ISSN 0160-8202), a publication devoted to the
study of Arachnida, is published three times each year by The American Arach-
nological Society. Memberships (yearly): Membership is open to all those inter-
ested in Arachnida. Subscriptions to The Journal of Arachnology and American
Arachnology (the newsletter), and annual meeting notices, are included with mem-
bership in the Society. Regular, $30; Students, $20; Institutional, $80 (USA) or
$90 (all other countries). Inquiries should be directed to the Membership Secretary
(see below). Back Issues: Susan E. Riechert, Department of Zoology, Univ. of
Tennessee, Knoxville, TN 37916 USA. Undelivered Issues: Allen Press, Inc.,
1041 New Hampshire Street, P.O. Box 368, Lawrence, Kansas 66044 USA.

THE AMERICAN ARACHNOLOGICAL SOCIETY

PRESIDENT: James E. Carico (1993-1995), Dept, of Biology, Lynchburg, Vir-
ginia, 24501 USA.

PRESIDENT-ELECT: Matthew H. Greenstone (1993-1995), USDA-BCIRL,
Columbia, Missouri 65203 USA.

MEMBERSHIP SECRETARY: Norman I. Platnick (appointed), American
Museum of Natural History, Central Park West at 79th St., New York, New
York 10024 USA.

TREASURER: Gail E. Stratton (1993-1995), Department of Biology, Albion
College, Albion, Michigan 49224 USA.

BUSINESS MANAGER: Robert Suter, Dept, of Biology, Vassar College, Pough-
keepsie, New York 12601 USA.

SECRETARY: Alan Cady (1993-1995), Dept, of Zoology, Miami Univ., Mid-
dleton, Ohio 45042 USA.

ARCHIVIST: Vincent D. Roth, Box 136, Portal, Arizona 85632 USA.
DIRECTORS: Allen R. Brady (1993-1995), Jackie Palmer (1992-1994), Ann
Rypstra (1993-1995).

HONORARY MEMBERS: C. D. Dondale, W. J. Gertsch, H. Homann, H. W.
Levi, A. F. Millidge, T. Yaginuma.

Cover illustration: A male Tetragnatha extensa from Carlisle, Massachusetts. Original color photo
by Joe Warfel of Arlington, Mass. Photograph made with a handheld Olympus OM-1 35mm camera,
macro lens, telescoping extension tube and manual flash.

Publication date: 29 December 1993

@ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

1993. The Journal of Arachnology 21:161-174

ULTRASTRUCTURE OF CRIBELLATE SILK OF NINE SPECIES IN
EIGHT FAMILIES AND POSSIBLE TAXONOMIC IMPLICATIONS
(ARANEAE: AMAUROBIIDAE, DEINOPIDAE, DESIDAE,
DICTYNIDAE, FILISTATIDAE, HYPOCHILIDAE, STIPHIDIIDAE,

TENGELLIDAE)

William Eberhard: Smithsonian Tropical Research Institute, and Escuela de
Biologia, Universidad de Costa Rica; Cuidad Universitaria, Costa Rica

Flory Pereira: Escuela de Biologia, Universidad de Costa Rica, Sede del Atlantico,
Costa Rica

ABSTRACT. The ultrastructure of cribellum silk and associated fibers is described for nine species in eight
families, and data from studies of 22 other species are summarized. Possible synapomorphies for filistatids
(flattened cribellum fibers), for all cribellates other than hypochilids + filistatids (nodules on cribellum fibers),
for deinopids + uloborids + dictynids, and for uloborids + dictynids (loss of reserve warp fibers) are described.
Filistatid silk is distinctive and especially complex, and the spatial arrangement of different components is
described for the first time.

RESUMEN. Se describe la ultraestmctura de la seda del cribelo y las fibras asociadas con ella de nueve especies
en ocho familias, y se resumen ademas los datos de 22 otras especies. Se destacan posibles sinapomorfias para
filistatidos, para todos los cribelados menos hypochilidos y filistatidos, para dinopidos + uloboridos + dictynidos,
y para uloboridos + dictynidos. La seda de los filistatidos es especialmente distinctiva, y la ubicacion espacial
de los diferentes componentes de ella se describe por primera vez.

Nonviscous adhesive silk is produced by cri-
bellate and some sicariid spiders. The ultrastruc-
ture of this silk and the lines associated with it
have been described in six different families (Fil-
istatidae, Uloboridae, Deinopidae, Eresidae,
Oecobiidae, and Amaurobiidae), using both the
light microscope and the scanning and trans-
mission electron microscopes (Comstock 1948;
Lehmensick & Kullmann 1957; Friedrich &
Langer 1969; Kullmann 1970, 1975; Zimmer-
mann 1975; Opell 1979, 1989a; Peters 1987,
1992a-c). This paper describes the cribellum silk
and associated lines of species in five additional
families, Desidae, Dictynidae, Hypochilidae,
Stiphidiidae, and Tengellidae, and from addi-
tional species of Amaurobiidae, Deinopidae, Fil-
istatidae and Uloboridae. We review data on the
distribution of several characteristics of cribel-
lum fibers and associated lines. Some characters
are apparently consistent within taxonomic
groups, and may be useful in systematic studies.

METHODS

No single technique is adequate for studying
the complex arrays of fine fibers and lines in
cribellate adhesive threads. The light microscope
is incapable of resolving finer fibers, while the
harsh preparation techniques and observation
conditions of both the transmission electron mi-
croscope (TEM) and the scanning electron mi-
croscope can seriously distort arrays of silk (Pe-
ters 1987, 1992a). Both light microscope and
TEM were used in the present study.

Silk was collected in the field from webs of
mature or nearly mature females, using micro-
scope slides to which three or four square plex-
iglass rods had been glued (Opell 1989b). The
upper surface of each rod was covered with dou-
ble-sided sticky tape. The web was pressed against
the tape, taking care to minimize stress on threads
between the rods, and scissors were used to cut
the threads connecting the sample to the rest of

161

162

THE JOURNAL OF ARACHNOLOGY

Figures 1-4.— Mature female Hypochilus thorelli: 1, Mass of cribellum fibrils (barely visible) and highly coiled
primary reserve warp (RWl) laid on a foundation line (F) (light microscope); 2, same, with arrow showing edge
of cloud of cribellum fibrils (light microscope); 3, one pair of straight axial lines (AX) and two highly curled
reserve warp lines of different diameters (RWl, RW2), with fibrils in the background (TEM); 4, cylindrical
cribellum fibrils lacking nodules (TEM). Note that fibril diameters do not vary along their lengths (compare
with Fig. 9 of Kukulcania). Scale lines are, respectively, 50p, lOp, 2.6)u., and 0.5 p.

the web. Most observations with the light mi-
croscope were made on these slides.

Samples of silk for TEM study were carefully
placed on untreated grids under a dissecting mi-
croscope, taking care to avoid stressing threads.
None of the silk samples were coated or treated
in any way before being examined. The silk of
Hyptiotes thorelli Marx had been stored in sealed
containers for 2-3 years; that of the other species
was fresher (less than about six weeks old).

The terms “fiber” and “fibril” are used for the
smallest units of silk (single cylinders); “line”
refers to a combination of fibers of the same type
running in parallel; “thread” and “band” refer
to combinations of fibers and lines of different
types. “Cribellum fibers” are presumed to emerge
from the cribellum, while “cribellate” lines and
threads (“calamistrated strands” of Peters 1987)
contain cribellum fibers as well as other lines that
presumably emerge from other spinnerets.

Terminology for different types of fibers and
lines follows that of Peters (1987), with the ex-

ception that we have used the earlier, function-
ally descriptive term “reserve warp fibers” of
Kullmann (1975) for the highly curled or un-
dulating thicker fibers often associated with cri-
bellum fibrils (“undulating fibers” and “U-fi-
bers” of Peters 1987, 1992a). Identifications of
different lines were based only on the morphol-
ogy and location of the lines, so homologies are
thus tentative. A straight or nearly straight fiber
running in a pair (except when two separate cri-
bellate threads were laid by a spider with a di-
vided cribellum) in the midst of a mat of cri-
bellum fibers was termed an axial fiber; curled
fibers always in the midst of cribellum fibers, also
generally in pairs, were termed reserve warp. In-
formation on the glandular origins of different
fibers, the spigots from which they emerge (e. g.,
Peters 1984, 1992a), and their chemical prop-
erties will be needed to establish more certain
homologies.

Voucher specimens of the spiders are depos-
ited in the Museum of Comparative Zoology,

EBERHARD & PEREIRA^FINE STRUCTURE CRIBELLUM SILK

163

Figures 5“9.— Mature female Kukulcania hibernalis: 5, a pair of highly coiled primary reserve warp lines
(RWl) and cribellum silk (barely visible) near a foundation line (F) to which they were attached (light microscope);
6, scalloped edge of mass of cribellum silk (arrow), primary reserve warp (RWl) and foundation line (F) composed
of multiple fibers (light microscope); 7, “crinkled” axial line (AX) and primary reserve warp (RWl) in mass of
cribellum fibrils (barely visible) (light microscope); 8, flattened primary reserve warp line (RWl) (note variation
in diameter) and thinner secondary reserve warp lines (RW2) in mass of cribellum fibrils (TEM); 9, flattened
cribellum fibrils (note that apparent diameters change where lines are folded (TEM). Scale lines are, respectively,
200^5 100|i, lOju , Six , and O.Sju.

Cambridge, Massachusetts 02138. Collection
sites for different genera were the following: Hy-
pochilus - near Cullowhee, North Carolina, USA;
Kulkania, Tengella, and Dictyna - near San An-
tonio de Escazu, Costa Rica; Badumna and Par-
amatachia - Lamington National Park, SW of
Brisbane, Queensland, Australia; Avella and Ma-
hura ” Cape Tribulation, N of Cairns, Queens-
land, Australia; and Stiphidium - Gilles Highway
W of Cairns near maximum elevation on way to
Atherton, Queensland, Australia.

RESULTS

Table 1 summarizes our observations and those
of other authors. More detailed descriptions of
the species we studied follow.

Hypochilus thorelli Marx (Hypochilidae)—
Observations with the light microscope revealed
a more or less cylindrical mass of cribellum silk
associated with a pair of linear axial fibers plus
a pair of moderately coiled reserve-warp fibers
(Figs. 1, 2). Additional, thinner secondary re-
serve warp fibers were revealed with the TEM
(Fig. 3). The fibrils of cribellum silk were ap-
parently cylindrical, and lacked nodules (Fig. 4).

Kukulcania hibernalis (Hentz) (Filistatidae)
The band of cribellate silk was laid along a thick
foundation line (Figs. 5, 6), to which it was at-
tached periodically. The foundation line had
multiple grooves (Fig. 6), suggesting it was com-
posed of many different strands. The silk of the
foundation line was unusual in being relatively
rigid: when cut, the line did not sag or fold. The

Table l.~ Characteristics of cribdlum silk and associated lines in 31 species of spider, (a) Type of microscope; LM = light microscope; TEM = transmission
electron microscope; SEM = scanning electron microscope, (b) Fibrils: C = cylindrical ibril; R = iattened, ribbonlike fibril; N = nodules along fibril, (c) Number
of pairs of lines, (cc) C = cylindrical lines; R = fiatteeed fibers, (d) Thick lines of two different diameters present in photo, but their arrangement (coiled, straight,
etc.) unknown, (e) Opell 1979. (f) Eberhard 1972 for building behavior, (h) Eberhard 1982 for building behavior, (i) No direct observations, but builds typical

164

THE JOURNAL OF ARACHNOLOGY

e

0

1

o

U

0© 2

6

r 6
t °

O S

I B

o «

t-i

a r®

s 1

a .s

o 8

a s

3 2

2 I

t -2

^ I

o 3

a ■'g

b °

‘-o S

o g
P: e

1 §

os 00

oj .2
..D .£3

.S o

2 c

.S

”3 o

cd S

^ s

a ^

■^5

OO 3

(N a
On fi
— < 3

II

j I

I EJJ

o S

s c--
•S u

« o

^ 1
^ &
« ed

d

.2

'a e
a

p

£

cd

o a

a 8

^ S

u

VO

vs

a\

d

g

3

^ t-
6®

o

'fc«

Oi 60

2 a

11

a $

X

ov 3
VO 5
OV W

S 3

ao

^ j -

s| §
s ^ s

•C O

« 'C O

J td U

u

E

Al Al

§

s

0

.N

•S

S

1

s

«

m

d

X

o

•c

£

» ^ s

ON ^ Ov

as -a ^

M _ Q>5 ■<-» M

Srt ” 60

« «
a 2* a

^ O Pm

«

VO

^ O
Os 0^

— Ov
d

S ^

I -s

p p

^ u

>=^ X 5

O ^ Ov

■ O <»

e ^ Os

W3 « „

B 3 § 3 3
■g G, 'g a a.

£ O J O O

2 ^ S ^ 2

»9C^O0C«»9tfl60«9t«Q06O

C«QOaO«960«9t«696O«

flj4s4>4s4i4>JW4)aw

c- O O c- O O O O O O e-- C-"

3' O

z z z z z

+ + A + +

u u u u u

s

s

S

s s

w

w

w

w

W H

w

H

H

H

H E

H

W S

w

^ w

nJ

H hJ

H

J

J

a e d

+ + A

U U U

S S

w w

m c/3

s s s s s

J J J J J

w w

i §

•5

^ c
.1 ^

»i2

a

S

K

&

S to

a

’V

Ci, S

tO

•C

Q

■8-8

i

^ a.

S W9 5 _

^ 60 F s o

^ S .h Ci, S

•- S O g

^ g . ^ s

^ a ? ^

^ ^

3

I

si

■2 ^
O 6,

S S

“s -a

s

~C)

I §

53 s:
> 53

^ c

2

Table 1.— Continued.

EBERHARD & PEREIRA- FINE STRUCTURE CRIBELLUM SILK

165

< I

w o

^ 3

<u b

^ I

t I
^ &
eS

C

.2

c5 0)
-o c
c

ir>

^ o

os fs|
OS

e 2

S 2

S3 <U
pL,

Vi Vi Vi

OJ O (D

Vi VI

(U (U (U
>, >.

U

o o o
c c e

^ 3

a^l <i

t-l

c«

<U

"5

in o

t~- m
os os ^

'—I os fs| rsi

e £2 os os

e § a 2 2

ed

- - - S

3 3 3 ^ -S

^ ^ pu a,

oo

OS

—1 fNl

g2t

Is"

b 3 c«
o o ‘S
U Pu 3

o

G

^ ^ m ^ ^
^ 4) ^ (U

2

o

>

o m cn

Vi

o o o
see

U

eN

O —'

+ +
u u

•S ^ ^
q q X

Co

S

“I 5

CS

Vi

s I

Co Co ^ Co Oo

w

D

c«

.3

3

o

o

(U

O

3

■S'

3

c

c

<u

.3

3

a

p

p

a

<

'a

p

s

d

Vi

S;

3

Vi

.a

a

&

2

(U

p

3

5

>S

a

a

S

d^

3

so

f-t

So

a

3

PJ

PJ

'3 ^

^ f*s

H

CO

2 8

s s

s

s

s

s

s

S

0* Vi

S w

u

UJ

pp

UP

UP

UP

^ o

CO H

H

CO

H

H

H

H

^ 2

S

u

s s

S

s

S

UP

UP

s'

S

S

CO

J J

3

J

H

3

H

3

o

H

a,

o

3 '3

'O c

'c -b

3 Q
Q

Stiphidiidae

Stiphidium sp. LM, TEM C + N no 1 C yes no/yes this study

Desidae

Badumna sp. LM, TEM C + N no 1 yes no this study

Paramatachia decorata LM, TEM C + N yes 0 no “yes” this study

166

THE JOURNAL OF ARACHNOLOGY

Figures lO™! L— Mature female Tengella radiata: 10, flat mass of cribellum fibrils (barely visible), with a pair
of straight axial lines (AX) and a pair of folded reserve warp lines (RW) (light microscope); 1 1 , cylindrical
cribellum fibrils with nodules (TEM). Scale lines are, respectively, IGOju, and O.S/x.

foundation line was laid as the spider moved
away from its retreat, and the cribellum silk and
associated fibers were laid during the return trip
(Eberhard 1988). In some places the band of cri-
bellate silk was more or less linear (Fig. 5), but
more often it was piled up or coiled on itself,
forming irregular loops.

The internal structure of the band was com-
plex. Under the light microscope a more or less
looped and folded pair of helical fibers was seen
(the helix is relatively extended in Fig. 5), with
the mass of cribellum fibrils visible as a faint
cloud (Figs. 5~1). Within each helix, a relatively
thick, smooth primary reserve warp fiber was
curled in a highly regular fashion that included
a series of short, more or less straight basal por-
tions alternating with longer loops (Figs. 5, 7).
Each loop was oriented in nearly the same di-
rection as the previous one. The axial line, which
was thinner and apparently somewhat crinkled,
ran near and approximately parallel to the straight
basal portions of the loops of the primary reserve
warp fiber (Fig. 7).

Under the TEM, the primary reserve warp
proved to be flattened and ribbon-like, rather
than cylindrical (Fig. 8). The axial line was seen

to consist of a pair of lines, with the “crinkles”
consisting of portions where one fiber was curled
helically around the other. Additional, finer sec-
ondary reserve warp fibers (number uncertain)
were folded loosely and irregularly in the area of
the loops of primary reserve warp (Fig. 8). The
cribellum fibrils were smooth and ribbon-like,
rather than cylindrical (Fig. 9). They lacked the
nodules seen in the silk of many other species
(Table 1).

Tengella radiata (Kulczynski) (Tengellidae)—
A more or less flat mat of cribellum silk lay on
or around a pair of axial fibers plus a pair of
kinked or somewhat curled reserve-warp fibers
(Fig. 1 0) which were produced at the same time
as the cribellum silk. The edges of the mat were
not regularly scalloped, and the reserve warp fi-
bers appeared to be cylindrical. The mat twisted
from side to side as a relatively rigid unit in weak
air currents under the light microscope. The fi-
brils of cribellate silk were apparently cylindrical,
with many small nodules scattered along their
lengths (Fig. 11).

Dictyna sp. (Dictynidae)“The cribellum silk
formed a relatively flat mat with regularly scal-
loped edges (Fig. 12). In some cases the mat was

EBERHARD & PEREIRA-FINE STRUCTURE CRIBELLUM SILK

167

Figures 12-14.— Mature female Dictyna sp.: 12, mat of cribellum silk with scalloped edges (arrow) laid along
a foundation line (light microscope); 13, cribellum fibrils clumped together in places to form cables (arrow)
(TEM); 14, cylindrical cribellum fibrils with nodules (TEM). Scale lines are, respectively, lOO/z, 1^, and lju.

laid on a relatively thick foundation line (Fig.

1 2), while in others there was no foundation line.
Careful searches using the TEM showed that there
were neither axial nor reserve warp lines. Under
the light microscope a pair of darker lines were
sometimes visible in the central portion of the
mat of cribellum fibers, but these presumably
corresponded to cables composed of accumula-
tions of cribellum fibrils (Fig. 1 3). Cribellum fi-
brils were cylindrical, with nodules along their
length (Fig. 14).

Stiphidium sp. (Stiphidiidae)— The non-pla-
nar mass of cribellum fibrils was not laid along
a foundation line (Fig. 1 5), and did not have a
regularly scalloped outline (Fig. 1 6). Associated
with the cribellum silk were a pair of straight,
apparently cylindrical axial fibers and a pair of
curled, cylindrical reserve warp fibers (Figs. 15,
16). The cylindrical reserve warp was curled
tightly for short stretches which alternated with
stretches of similar lengths in which it was rel-
atively uncurled (Figs. 15-17). The cribellum fi-
brils were cylindrical, with nodules (Fig. 1 8).

Badumna sp, (Desidae)— In places two mats
of cribellate silk ran in close parallel, presumably
the product of the divided cribellum; in other

places they were farther apart. The lateral out-
lines of mats were not regularly scalloped (Fig.
1 9). Each mat had a straight, relatively thin axial
fiber, and a cylindrical reserve warp fiber in which
the degree of coiling varied (Figs. 19-21). The
cribellum fibrils were cylindrical, with nodules
(Fig. 22).

Paramatachia decorata (Dalmas) (Desidae) —
The lateral outlines of mats of cribellum silk were
often regularly scalloped, although the thicken-
ings (“puffs”) often did not occur at the same
point on either side of the mat (Fig. 23). Mats
of cribellate silk were usually but not always as-
sociated with foundation lines (Figs. 24, 25).
When viewed with the light microscope a pair
of straight axial fibers seemed to be present (Fig.
23), but no reserve warp fibers were seen. In some
places the mat of cribellum fibrils was coiled on
the axial line. Neither axial nor reserve warp
fibers were found using the TEM, however. Over
short stretches, cribellum fibrils came together
to form cables which gave the false impression
of thicker fibers (Fig. 25), but these differed from
the axial fibers seen in the light microscope in
being only relatively short. It appears that axial
fibers were absent from some samples, but it is

168

THE JOURNAL OF ARACHNOLOGY

Figures 15“1 8. -“Mature female Stiphidium sp.: 15, straight axial line (AX) and reserve warp line (RW) with
alternating highly curled and straighter regions (light microscope); 16, same, showing non=scalloped edge of mat
of cribellum fibrils (arrow) (light microscope); 17, curled cylindrical reserve warp line (RW) with cribellum
fibrils, some of which clump together to form cables (arrow) (TEM); 1 8, cylindrical cribellum fibrils with nodules
(TEM). Scale lines are, respectively, 100^, IGOju, Sju, and 0.5^.

uncertain whether they were present in others.
The cribellum fibrils were cylindrical, with nod-
ules (Fig. 26),

Mahura sp. (AmaErobiidae)-“Many mats of
cribellate silk in the sheet of this spider’s web
were composed of parallel double bands, pre-
sumably due to the divided cribellum. In con-
trast, mats of cribellate silk in the mesh above
the sheet were usually single. Each cribellate mat
collected from the sheet had a single cylindrical
reserve warp fiber, which was alternately tightly
coiled and relatively uncoiled (Fig. 27). Axial
fibers were not clearly visible in the light micro-
scope (appearing to be present only in short
stretches), and no axial lines were seen with the
TEM. In places cribellum fibrils came together
to form cables, and presumably these were the
“axial fibers” seen in the light microscope, Cri-
bellum fibrils were cylindrical, with nodules (Fig.
28).

Avella sp. (Deinopidae)— The lateral margins

of each mat of cribellum silk were strongly scal-
loped (Fig. 29). A pair of linear axial fibers and
a pair of loosely coiled, cylindrical reserve warp
fibers ran through the central portion of the mass
(Figs. 29, 30), Cribellum fibrils were cylindrical,
with nodules (Fig. 31).

DISCUSSION

The data available to date suggest that some
ultrastmctural characteristics of cribellate cap-
ture silk are relatively constant within and be-
tween taxa (Table 1). The consistency is es-
pecially clear in the cylinder plus nodule structure
of cribellum fibrils, and the lack of reserve warp
fibers in the best studied family, Uloboridae, It
should be bom in mind that the changes in cri-
bellum fibril morphology in the TEM (electron
bombardment in a vacuum) are not known. Thus
the morphology of fibrils described here may dif-
fer from that of fibers under normal conditions.
The tentative nature of homologies of the lines

EBERHARD & PEREIRA-FINE STRUCTURE CRIBELLUM SILK

169

Figures 19-22.— Mature female Badumna sp.: 19, mat of cribellum fibrils with irregularly scalloped edges,
axial lines (AX), and curled reserve warp line (light microscope); 20, pair of axial lines (AX) and pair of alternately
curled and uncurled reserve warp lines (RW), with a more regularly scalloped mat of cribellum fibrils (light
microscope); 21, curled cylindrical reserve warp line (RW) and straight axial line (AX) with cribellum fibrils
(TEM); 22, cylindrical cribellum fibrils with nodules (TEM). Scale lines are, respectively, ISO/x, 100/u, 2ix, and
0.5m.

associated with cribellum silk should also be kept
in mind.

While much more data need to be gathered to
determine whether the patterns of distribution
will hold up, it may be useful to attempt a ten-
tative comparative analysis. If one superimposes
the data on silk ultrastructure on a recently pro-
posed pylogeny of cribellate spiders (Coddington
& Levi 1991), several hypotheses result (Fig. 32):
1. Ribbon-like cribellum fibrils are a derived
character of filistatids {Filistata, Kukulcania). 2.
Nodules on cribellum fibrils are a synapomorphy
linking all cribellates other than filistatids and
hypochilids; 3. Lack of ‘"reserve-warp” lines is a
derived character, present in the single dictynid,
one of the two desids, and all of the 1 2 uloborids.
Since several details of web construction behav-
ior link Uloboridae and Deinopidae (which has
reserve warp fibers), the loss either occurred in-
dependently in Uloboridae and Dictynidae (Fig.
32) (with subsequent reacquisition of both axial
and reserve warp lines in Badumna and loss of

axial fibers in Mahura), or dictynoids are the
sister group of uloborids + deinopids, and dei-
nopids and Badumna secondarily re-acquired re-
serve warp lines (with a loss of axial lines in
dictynoids and Mahura).

An additional character, noted by other au-
thors, is the scalloped outline of the mass of cri-
bellate silk (“puffs”), which may unite Ulobor-
idae, Deinopidae, Dictynidae and the desid
Paramatachia (in at least some uloborids, a puff
is actually shaped more nearly like a twisted to-
rus). This character may be somewhat less useful,
however, since: 1) intermediate degrees of “scal-
loping” occur (e. g., Figs. 12, 20, 23), and it is
not clear how regular scalloping must be to be
condsidered a puff; and 2) some uloborid mats
are only barely scalloped (Peters 1 984, 1987). We
were unable to confirm the presence of paracri-
bellar fibrils (Peters 1984, 1987) in any of our
species (unless they correspond to the “cables”
of cribellum fibrils seen in Dictyna, Paramata-
chia, and Mahura).

170

THE JOURNAL OF ARACHNOLOGY

Figures 23-26. —Mature female Paramatachia decorata: 23, mat of cribellum silk with scalloped edges (arrow)
(apparent axial lines are slightly out of focus except at right and left margens) (light microscope); 24, foundation
line (F) to which cribellum silk was attached (note multiple fibers) (TEM); 25, foundation line with cables formed
by multiple cribellum fibrils (arrow) (TEM); 26, cylindrical cribellum fibrils with nodules (TEM). Scale lines
are, respectively, lOO/u, 5^, Sju, and 0.5)u.

The positions of the fibers associated with cri-
bellum silk help clarify some details of combing
behavior. Assuming that spider silk is polymer-
ized by being pulled (e. g., Foelix 1982), the pres-
ence of highly curled reserve warp fibers, which
are presumably pulled out by strokes of the cal-
amistrum and then fold or coil upon themselves,
suggests that cribellum silk perse is piled on itself
in the sticky threads of all species with curled
reserve warp fibers. In some cases the tendency
of reserve warp fibers to curl up may even cause
clumping to occur. For instance, the secondary
helices of Filistata and Kukuicania may result
from curling of the axial fibers and/or the pri-
mary reserve warp lines. In species such as Ten-
gella radiata, where the reserve warp fibers curl
less, they appear to have little influence on the
shape of the mass of cribellum fibrils. In both
these groups (as well as in Stegodyphus - see Eber-
hard 1988), the spiders do not pull the cribellate
silk threads taut in their webs. Rather, silk ac-
cumulates and sags free behind the spider as it
is combed from the cribellum with the calam-
istrum. The thread is under no tension other than

that resulting from its own weight and friction
with air currents, and is actually often piled on
itself in Kukuicania and Stegodyphus webs. Pre-
sumably when cribellum fibrils accumulate in
this way, the force of adhesion is increased by
bringing more silk surface into contact with the
prey (Opell 1990). The effective length of the silk
is probably also increased, making escape more
difficult when the prey attempts to pull away.

Many authors have thought that each of the
puffs in a mass of cribellum fibrils is produced
by a single combing movement of the calamis-
trum (Eberhard & Langer 1969; Friedrich &
Langer 1969; Opell 1979; Peters 1992c), but Pe-
ters (1984) attributed puffs to rhythmic clamping
movements of the posterior spinnerets. The pres-
ence of many helical turns of reserve warp fibers
between each pair of puffs in the sticky threads
of Deinopus sp. and Deinopus subrufus (Kull-
mann 1975; Peters 1992c), and Avella sp. (this
study) indicates that the second hypothesis is
more likely. The combing movement necessary
to produce a puff would be too short to pull out
such lengths of reserve warp fiber.

EBERHARD & PEREIRA^ FINE STRUCTURE CRIBELLUM SILK

171

Figures 27-28. —Mature female Mahura sp.: 27, reserve warp line (RW) which is more tightly curled in some
places than others (TEM); 28, cylindrical cribellum fibrils with nodules (TEM). Scale lines are, respectively, 2tx
and 0.5/u.

Similar reasoning indicates that combing
movements of the calamistrum in many species
are not responsible for pulling out axial fibers.
The looped and tangled cribellar fibrils (presum-
ably pulled by the calamistrum) are substantially

longer than the axial fibers. Probably many axial
fibers are pulled out as the spider moves away
from the last attachment point. This mechanism
is not possible, however, in spiders such as K.
hibernalis and Stegodyphus gregalis, which do

Figures 29-31.— Mature female Avella sp.: 29, highly scalloped mat of cribellum silk with pair of axial lines
(AX) and reserve warp lines (RW) (light microscope); 30, axial line (AX) with pair of reserve warp lines (RW)
and cribellum fibrils (TEM); 3 1 , cylindrical cribellum fibrils with nodules (TEM). Scale lines are, respectively,
lOOju, 5fx, and O.Sjti.

172

THE JOURNAL OF ARACHNOLOGY

been studied (after Coddington & Levi 1991), with data on silk morphology (Table 1) superimposed to show

possible transitions.

not move forward during most of the time cri-
bellum silk is being combed (Eberhard 1988; see
also Opell 1990 on Miagrammopes). It is not
clear how axial fibers are pulled from the spin-
nerets in these species.

Comstock (1948) speculated that the helix of
threads (he saw them as loops) of Kukulcania
hibernalis (under the name Filistata) result from
movements of the spinnerets, while the very reg-
ular loops of the primary reserve warp are made
by combing movements of the calamistrum.
Given the much longer length of the secondary
reserve warp fibers, however, it seems more like-
ly that their irregular folding may be associated
with the combing movements of the calamis-
trum. The highly ordered folding of the primary
reserve warp and the helical coiling of the swath
itself is presumably due to their intrinsic curli-
ness (but see below), and the fact that spider
moves forward very little as it combs out silk,
so that cribellate silk “piles up” between attach-
ments to the foundation line.

Comstock also thought that the axial fibers of
K. hibernalis are highly elastic, stretching “to fifty
times their first length”. We were unable to con-
firm this. Instead, when a swath was pulled under
the light microscope, a process of sequential

breaking occurred (possibly of the axial fibers),
bringing the reserve warp fibers under tension as
described by Kullmann (1975) for Stegodyphus.
As the swath was slowly pulled, it extended: the
primary reserve warp began to unfold, but did
so unevenly, in little starts. It became completely
unfolded in some places before others. Eventu-
ally the primary reserve warp became completely
extended. If the tension was then relaxed, the
reserve warp remained extended, and did not
recoil to its original position (thus failing to show
the intrinsic curliness postulated above). Further
extension caused the primary reserve warp fiber
to break, and with that the entire thread usually
broke. Thus the finer, secondary reserve warp
fibers of K. hibernalis apparently serve in ad-
hesion (of the cribellum silk to the primary re-
serve warp? to the prey?) rather than to increase
the tensile strength and elongation of the array
of lines as do the secondary reserve warp fibers
of Stegodyphus (Kullmann 1975). Presumably
the extension Comstock observed was the exten-
sion of the entire array of cribellum silk and as-
sociated fibers.

The most complex and distinctive arrays of
cribellum silk and associated fibers are those of
filistatids. These may show intergeneric differ-

EBERHARD & PEREIRA FINE STRUCTURE CRIBELLUM SILK

173

ences. Lehmensick & Kullmann (1956) describe
a two-part mass of adhesive silk in Filistata in-
sidiatrix, laid in small accumulations on a pre-
viously built foundation line, just as in K. Mb-
ernalis. Although they did not mention that each
of the two parts has a helical form, this seems to
be the case in the light microscope photo of Pe-
ters (1987) of the same species. Lehmensick &
Kullmann also noted a pair of axial fibers, which
seem (in their light microscope photo, plate 2,
fig. 3) to be thicker and straighter than those of
K. Mbernalis. The fiber labelled axial line in their
TEM micrograph (plate 2, fig. 4) may, however,
may not correspond to the light microscope axial
fiber: it does not run through the mass of cri-
bellum fibers and curled reserve-warp fibers; and
a thinner fiber, which is more appropiately lo-
cated and which resembles the axial fiber of K.
Mbernalis, is unlabelled. Perhaps the line they
labelled as an axial line in their TEM micrograph
was a foundation line.

Also unique to filistatids is the non-cylindrical,
ribbon-like form of the primary reserve warp
fiber. Judging by the flattened tips of the para-
cribellar spigots on the posterior median spin-
nerets of K. Mbernalis (figs. 56-"58 in Platnick et
al. 1991), these spigots may be the source of pri-
mary reserve warp fibers. This speculation is sup-
ported by the existence of a somewhat similar,
slit-shaped opening of the “major ampullate gland
spigot” on the anterior lateral spinneret of Lox-
osceles rufescens and L. reclusa (Platnick et al.
1991), and the fact that L. rufescens also makes
a wide, ribbon-like band of silk (Lehmensick &
Kullmann 1956; Kullmann 1975). It is in ap-
parent conflict with the lack of paracribellar spig-
ots in hypochilids, eresids, and Tengella (Plat-
nick et al. 1991). Peters (1992a) has established
that reserve warp fibers are secreted from spigots
on the posterior median spinnerets in Stegody-
phus. Further work is needed to establish which
spigots produce these and other fibers.

The band-like cribellum fibrils of K. Mbernalis
may be associated with their bladder-shaped
“claviform” cribellar spigots (fig. 52 of Platnick
et al. 1991), which are quite different from the
more sharply-tipped “strobilate” spigots known
for other cribellates (Kullmann 1975;Opell 1979;
Peters 1984, 1987, 1992; Platnick et al. 1991).
A second possible silk-spigot association, be-
tween the presence of nodules on cribellum fibrils
and nodule-like expansions on the cribellum
spigots, is apparently ruled out, however, by the
presence of expansions on the spigots of Hypo-

cMlus pococki (Platnick et al. 1991), and the ab-
sence of nodules on the fibrils of H. thorelli.

Homologies of the filistatid fibers with those
of other species are somewhat uncertain. Com-
stock (1948) apparently also noted the axial fi-
bers, and both primary and secondary reserve
warp fibers in K. Mbernalis (calling them, re-
spectively, primary looped threads, secondary
looped threads, and irregular threads). We have
designated as “axial lines” the least folded lines
within the helices, but the double nature of these
lines is unique. If, instead, the wider, regularly
looped “primary reserve warp” fibers are ho-
mologous to the axial fibers of other species, the
characteristics of Kukulcania in Table 1 and the
position of filistatids in Fig. 32 would be little
altered.

ACKNOWLEDGMENTS

Fred Coyle kindly sent carefully packed sam-
ples of HypocMlus silk. Herb Levi helped with
identifications. Robert Raven facilitated work in
Australia. Brent Opell, M. J. West-Eberhard and
especially Charles Griswold made helpful com-
ments on preliminary drafts of the manuscript.
The Vicerrectoria de Investigacion of the Univ-
ersidad de Costa Rica provided financial sup-
port. We thank all for their help.

LITERATURE CITED

Coddington, J. A. 1 986. Orb webs in “non-orb-weav-
ing” ogre-faced spiders (Araneae: Dinopidae): a
question of genealogy. Cladistics, 2:53-67,
Coddington, J. A. & H. W. Levi 1991. Systematics
and evolution of spiders (Araneae). Ann, Rev. Ecol.
Syst, 22:565-592,

Comstock, J. 1 948. The Spider Book. Comstock Publ.

Assoc,, Ithaca, New York. 729 pp.

Eberhard, W. G. 1972. The web of Uloborus diversus
(Araneae: Uloboridae). J. Zool., London, 166:417-
465,

Eberhard, W. G. 1988. Combing and sticky silk at-
tachment behaviour by cribellate spiders and its tax-
onomic implications. Bull. British Arachnol. Soc.,
7:247-251.

Eberhard, W. G. & R. Langer. 1 969. Laboratory pho-
tography of spider silk. American Zool. 9:97-101.
Foelix, R. 1982. Biology of spiders. Cambridge, Har-
vard Univ. Press.

Friedrich, V. & R. Langer. 1969. Fine structure of
cribellate spider silk. American Zool., 9:91-96.
Kullmann, E. 1970, Bemerkenswerte Konvergenzen
im Verhalten cribellater und ecribellater Spinnen.
Freunde Kolner Zoo, 4:123-150.

Kullmann, E. 1975. Die Produktion und Funktion

174

THE JOURNAL OF ARACHNOLOGY

von Spinnenfaden und Spinnengeweben, Pp. 318-
378, In Netz in Natur und Technik. (Inst. Leichte
Flachentragwerke, ed.). Stuttgart- Vaihingen.

Lehmensick, R. & E. Kullmann, 1956. Uber den
Feinbau der Faden einiger Spinnen (Vergleich des
Aufbaues der Fangfaden cribellater und ecribellater
Spinnen). Zool. Anz., SuppL: 123-129.

Lehmensick, R. & & E. Kullmann. 1957. Uber den
Feinbau der Spinnenfaden. Proc. Stockholm Conf.
Electron Microscopy. Pp. 307-309.

Marples, M. J. & B. J. Marples. 1937. Notes on the
spiders Hyptiotes paradoxus and Cyclosa conica.
Proc. Zool. Soc. London, 107A:2 13-221.

Opell, B, D. 1979. Revision of the genera and tropical
American species of the spider family Uloboridae.
Bull. Mus. Comp. Zool., 148:443-549.

Opell, B. D. 1987. The influence of web monitoring
tactics on the tracheal systems of spiders in the fam-
ily Uloboridae (Arachnida, Araneida). Zoomorph.,
107:255-259.

Opell, B. D. 1989a. Functional associations between
the cribellum spinning plate and capture threads of
Miagrammopes animotus (Araneida, Uloboridae).
Zoomorph., 108:263-267.

Opell, B. D. 1 989b. Measuring the stickiness of spider
prey capture threads. J. Arachnol., 17:1 12-1 14.

Opell, B. D. 1 990. Material investment and prey cap-
ture potential of reduced spider webs. Behav. Ecol.
SociobioL, 26:375-381.

Opell, B. D., Roth, G., & P. Cushing. 1 990. The effect
of Hyptiotes cavatus (Uloboridae) web-manipula-
tion on the dimensions and stickiness of cribellar
silk puffs. J. Arachnol., 18:238-240.

Peters, H. 1983. Struktur und Herstellung der Fang-
faden cribellater Spinnen (Arachnida: Araneae).
Verb. Naturwiss. Ver. Hamburg, 26:241-253.

Peters, H. 1984. The spinning apparatus of Ulobor-
idae in relation to the structure and construction of
capture threads (Arachnida, Araneida). Zoomorph.,
104:96-104.

Peters, H. 1987. Fine structure and function of cap-
ture threads, Pp. 187-202, In Ecophysiology of spi-
ders. (W. Nentwig, ed.). Springer, New York.

Peters, H. 1 992a. Uber Struktur und Herstellung von
Fangfaden Cribellater Spinnen der Familie Eresidae
(Arachnida, Araneae). Verb. Naturwiss. Ver. Ham-
burg, 33:213-227.

Peters, H. 1992b. On the burrowing behaviour and
the production and use of silk in Seothyra, a sand-
inhabiting spider from the Namib Desert (Arach-
nida, Araneae, Eresidae). Verb. Naturwiss. Ver.
Hamburg, 33: 1 9 1-2 1 1 .

Peters, H. 1992c. On the spinning apparatus and the
structure of the capture threads of Deinopus subrufus
(Araneae, Deinopidae). Zoomorph., 112:27-37.

Platnick, N. L, Coddington, J. A., Forster, R. R., & C.
E. Griswold. 1991. Spinneret morphology and the
phylogeny of haplogyne spiders (Araneae, Araneo-
morphae). American Mus. Novitates, 3016:1-73.

Wiehle, H. 1928. Beitrage zur Biologic der Araneen,
insbesondere zur Kenntnis des Radnetzbaues. Z.
Morph. Okol. Tiere, 11:115-151.

Zimmerman, W. 1975. Biologische und rasterelek-
tronenmikroskopische Festellungen an Oecobiinae,
Uroecobiinae und Urocteinae (Araneae: Oecobi-
idae) als Beitrag zum “Cribellaten-Ecribellaten-
Problem”, Thesis, Univ. Bonn.

Manuscript received 7 May 1993, revised 19 July 1993.

1993. The Journal of Arachnoiogy 21:175-183

STUDIES ON SPECIES OF HOLARCTIC PARDOSA GROUPS
(ARANEAE, LYCOSIDAE).

V. REDESCRIPTION OF PARDOSA WASATCHENSIS GERTSCH
AND DESCRIPTION OF A NEW SPECIES FROM UTAH

Torbjorn Kronestedt; Department of Entomology, Swedish Museum of Natural
History, Box 50007, S-104 05 Stockholm, Sweden

ABSTRACT. Two North American Pardosa species assigned to the modica group are treated and illustrated.
Pardosa wasatchensis Gertsch (Montana, Wyoming, Colorado, Utah, Idaho, Washington; with Pardosa subra
Chamberlin & I vie from Oregon placed as junior synonym) is redescribed, and the male is described for the
first time. Pardosa vogelae, new species, is described on material from Leidy Peak and vicinity in the Uintah

Mountains (Utah).

Species of the modica group within the wolf
spider genus Pardosa have previously been treat-
ed by Kronestedt (1975, 1981, 1986, 1988) and
Dondale & Redner (1 990). This paper deals with
another two species assigned to this group of
species.

Pardosa wasatchensis was described from the
female by Gertsch (1933) and since then little
has been added to the knowledge of this species,
which seems to be restricted to western U. S.
(Fig. 22). P. subra, described by Chamberlin &
Ivie (1942), regrettably also from the female sex
only, seems to be conspecific with P. wasatch-
ensis. The male of P. wasatchensis is described
here for the first time.

The new species Pardosa vogelae has so far
been found only in the eastern parts of the Uintah
Mountains of Utah, Whether or not it has a re-
stricted distribution remains to be explored. [This
discovery is parallelled by the find at a single
locality in Colorado of a high altitude species in
the nigra group, P. gothicana Lowrie & Dondale
(1981), calling for more intense studies of Cor-
dilleran Pardosa.]

METHODS

Material is deposited in the following collec-
tions: AMNH-— American Museum of Natural
History, New York; BRV— private collection of
B. R. Vogel; CNC— Canadian National Collec-
tion of Insects and Arachnids, Ottawa; MCZ—
Museum of Comparative Zoology, Cambridge,
Massachusetts; NRS-- Swedish Museum of Nat-
ural History, Stockholm (material gratefully do-
nated by D. C. Lowrie and B. R. Vogel).

Terminology and methods of study follow
Kronestedt (1975, 1986). Measurements refer to
specified individuals. Eyepiece micrometer units
(as given for eyes) can be converted to mm by
dividing by 80.

Pardosa wasatchensis Gertsch
Figs. 1, 3, 5, 6, 9, 10, 12, 14, 15, 17, 19, 21a, 22;

Table 1

Pardosa wasatchensis Gertsch, 1933: 25, fig. 37 (female
holotype from United States: Utah, Sevier County,
Fish Lake, in AMNH, examined). Roewer 1954: 195.
Bonnet 1958: 3431.

Pardosa subra Chamberlin & Ivie, 1942: 30, fig. 71
(female holotype from United States: Oregon, Har-
ney County, Malheur Lake, in AMNH, examined).
Roewer 1954: 194. Syn. n.

Diagnosis. —Males may be distinguished by the
short, only slightly curved embolus, somewhat
widened toward its tip, the latter with a small
incision on its inner side (Fig. 5), as well as by
the configuration of the conductor (Fig. 3) and
the retrolateral grooved process of the terminal
apophysis (Fig. 10); females by the proportions
of the flask-shaped epigyne, with a narrow sep-
tum widened posteriorly like an inverted “T”,
and with anterior transverse pockets well sepa-
rated and extending more or less laterad (Figs.
14, 15).

Male. “(Utah, Sevier County, Fish Lake). To-
tal length 7.4 mm; carapace 3.60 mm long, 2.65
mm wide.

Carapace: Dusky brown, median band in tho-
racic part yellowish, lateral bands light brownish
to yellowish. Lateral bands broken into (2-)3 parts

175

1 2

Figures 1, 2. —Right male palp, ventral view. 1. Pardosa wasatchensis Gertsch from Fish Lake, Utah; 2. P.
vogelae sp, n. from Leidy Peak, Utah. Scale: 0.5 mm.

Figures 3, 4.— Terminal part of left palp with conductor {cond) and terminal apophysis {tl.ap). 3. Pardosa
wasatchensis Gertsch from 8 mi. N Fish Lake, Utah; 4. P. vogelae sp. n. from Leidy Peak, Utah. Scale: 0.5 mm.

KRONESTEDT-HOLARCTIC PARDOSA

177

5 6

7 8

Figures 5-8.— Embolus seen in frontal (5, 7) and ventral (6, 8) views. 5-6. Pardosa wasatchensis Gertsch from
8 mi. N Fish Lake, Utah; 7~8. P. vogelae sp. n. from Leidy Peak, Utah. Scale: 0.5 mm.

Table 1. — Leg I~IV measurements (mm) of Pardosa
wasatchensis Gertsch and Pardosa vogelae sp. n.
Fe = femur, Pa = patella, Ti = tibia, Mt = metatarsus,
Ta = tarsus.

Fe

Pt

Ti

Mt

Ta

Total

Pardosa wasatchensis

Male

I

2.80

1.30

2.35

2.40

1.65

10.50

II

2.70

1.25

2.15

2.40

1.60

10.10

III

2.65

1.20

2.10

2.80

1.55

10.30

IV

3.40

1.40

2.90

4.20

2.00

13.90

Female

I

2.70

1.30

2.05

2.05

1.45

9.55

II

2.65

1.30

1.90

2.10

1.45

9.40

III

2.55

1.20

1.90

2.55

1.40

9.60

IV 3.50

Pardosa vogelae

1.40

2.90

4.35

2.00

14.15

Male

I

2.30

1.10

1.85

2.10

1.55

8.90

II

2.30

1.10

1.75

2.05

1.50

8.70

III

2.25

1.00

1.70

2.45

1.45

8.85

IV

2.95

1.20

2.40

3.70

1.90

12.15

Female

I

2.25

l.iO

1.75

1.75

1.30

8.15

II

2.25

1.05

1.60

1.75

1.30

7.95

III

2.20

1.00

1.65

2.15

1.25

8.25

IV

2.95

1.20

2.45

3.55

1.75

11.90

by transverse brownish streaks. Postocular spots
in cephalic part brownish. Sides of thoracic part
with numerous short dark and recumbent gray-
ish hairs. Median band with whitish hairs, in
preserved specimens usually remaining behind
fovea; around fovea in addition with short dark
(and at midline with few long dark) erect hairs.
Lateral bands with dark and light hairs. Clypeus
yellowish, with long forwardly directed dark hairs
medially. Chelicerae yellowish to brownish with
grayish brown streaks, furnished with dark hairs.
Sternum dark brown, furnished with erect dark
and recumbent grayish hairs.

Eyes: Width of row 152 (slightly procurved as
seen from front), row II 7 1 , row III 94, row II-
III 70. Diameter of AME 12, ALE 10, PME 26,
PLE 2 1 . Distance between AME 8, between AME
and ALE 2.

Abdomen: Dorsally grayish brown, with light
brownish to yellowish, dark-bordered lanceolate
stripe in front. Posteriorly on each side of median
part a dark irregular line interrupted at intervals
by light dots (with white hair tufts). Median part
with two separate spots at end of lanceolate stripe,
rearwards followed by two close spots and more
posteriorly by a few bars, all brownish to yellow-
ish (pattern sometimes hardly discernible), each
spot with one dark dot, bars with two dark dots,
each carrying long dark hair. Dorsum and sides
with long erect and short dark as well as recum-
bent light hairs, venter brownish to yellowish
with recumbent white pubescence and numerous

178

THE JOURNAL OF ARACHNOLOGY

Figures 9-1 3. “-Terminal part (9, 1 1 in ventral view; 10 in retrolateral view) and tegulum with tegular apophysis
(12, 13 in ventral view), left male palp (for arrows see text). 9-10, 12. Pardosa wasatchensis Gertsch from 8 mi.
N Fish Lake, Utah; 11, 13. P. vogelae sp. n. from Leidy Peak, Utah. Scale: 200/im.

longer erect grayish to light hairs (latter seen only
in some males examined).

Legs: (Table 1) Yellowish. Femora with darker
blotches dorsally (“pseudoannulation”), outer
segments with very faint grayish tinge, on tibiae
and metatarsi arranged like the more distinct
annulation in female. Tibia I with two or only
distal retrolateral spine(s) present. Hairiness of
leg I as in rest of legs.

Palp: Patella 0.70 mm, tibia 0.65 mm, cym-
bium 1.35 mm. Femur, patella and tibia yellow-
ish; femur with dark markings, patella and tibia
less so, sometimes even unicolorous. Cymbium
brown to blackish brown, lighter apically. Fe-
mur, patella and tibia with dark and whitish hairs,
latter dominating on tibia; cymbium with dark
hairs except distally. Tegulum comparatively
protruding. Tegular apophysis as seen in ventral
view (Figs. 1,12) with subtriangular basal part
(including anteriorly directed branch); lateral

process basally almost as wide as length of basal
part and tapering toward ventrally directed
slightly hook-shaped tip; posterior (lower in fig-
ures) rim of lateral process with more or less
distinct denticle-like projection at some distance
from tip {arrow in Fig. 12). Terminal apophysis
(as seen in ventral view: Figs. 3, 9) with heavily
sclerotized tooth-like process protruding for-
ward, and with small, grooved retrolateral pro-
cess {arrow in Fig. 1 0; hidden below conductor
in ventral view). Conductor with distal part
curved dorsad (Fig. 1 0); its posterior rim folded
towards rounded tip, forming groove (Figs. 3, 9);
anterior rim more or less incised before rounded
tip. Embolus (Figs. 5, 6) short, in frontal view
slightly widening distally, incised at tip.

Female.— (Utah, Sevier County, Fish Lake).
Total length 6.7 mm (carried egg sac); carapace
3.70 mm long, 2.70 mm wide.

Similar to male in color pattern and hairiness.

KRONESTEDT-HOLARCTIC PARDOSA

179

14 15 16

Figures 14-16.— Epigyne, ventral view. 14. Pardosa wasatchensis Gertsch, holotype; 15. holotype of P. subra
Chamberlin & Ivie; 16. P. vogelae sp. n. from Leidy Peak, Utah. Scale: 0.5 mm.

Carapace lighter brownish than in male, and with
yellowish, sometimes unbroken, lateral bands
with more light hairs. Chelicerae usually lighter
than in male and furnished with light hairs in
addition to dark ones.

Legs: (Table 1) Femora yellowish to light
brownish with darker brownish “pseudoannu-
lation,” tibiae and metatarsi brownish, often with
very indistinct lighter brown to yellowish an-
nulation. Tibia I with two, only distal, or no
retrolateral spine(s) present.

Flask-shaped (Figs. 14, 15, 17; cleared
Fig. 19). Anterior transverse pockets well sepa-

rated, extending laterad. Septal ridge as wide as
or slightly wider than narrow septum. Septum
posteriorly widened. Lateral elevations some-
times sloping smoothly into anterior part of cav-
ities (i. e., without distinct rim). Receptacles
comparatively inflated (Fig. 1 9).

One female (carapace length 3.40 mm) carried
an egg sac with diameter of 4.7 mm and height
of 3.7 mm, containing 67 pulli (larvae).

Size variation.-— Carapace lengths of material
measured: males 3.00-3.60 mm {n = 1 5), females
2.75-3.70 mm {n = 1 5); tibia I length vs. carapace
length in Fig. 21a.

Figures 17, 18. — Epigyne. 17. Pardosa wasatchensis Gertsch from 8 mi. N Fish Lake, Utah; 18. P. vogelae
sp. n. from Leidy Peak, Utah (for arrow see text). Scale: 200Atni.

Figures 19, 20.--Epigyne (cleared), ventral view. 19.
Utah; 20. P. vogelae sp. n. from Leidy Peak, Utah.

Material examined.— UNITED STATES. Colorado.
Eagle County: 2 mi. S Bums, 27 June 1963 (B. Vogel,
AMNH), 1(3 29. Grand County: S Granby, 26 June
1940 (W. Ivie, AMNH), 23 39; 5 mi. S Parshall, 20
June 1963 (B. Vogel, AMNH), 13. Idaho. Bear Lake
County: Nounan, 9 August 1931 (W. J. Gertsch,
AMNH), 19. Payette County: Payette, 1956 (Evadina
Ivie, AMNH), 33 19. Teton County: Victor, 15 August
1940 (W. Ivie, AMNH), 29. Valley County: Cascade,

5 July 1943 (W. Ivie, CNC), 23 29; S Donnelly, 5 July
1943 (W. Ivie, AMNH), 19. Montana. Horse Prairie
(not located), 1 1 July 1935 (W. Ivie, AMNH), 29. Jef-
ferson County: S Butte, Toll Mt. Campground, 6 Au-
gust 1964 (D. C. Lowrie, AMNH), l9. Oregon. Harney
County: Malheur Lake, 18 June 1940 (L. W. Saylor,
AMNH), 19 (holotype of P. subra). Utah. Rich County:
Bear Lake (S end), 26 June 1962 (W. Ivie, AMNH),
19. Salt Lake County: Salt Lake City, September 1930
(W. J. Gertsch, AMNH), l9. Sevier County: Fish Lake,

4 September 1929 (Chamberlin & Gertsch, AMNH),
19, 22-23 June 1930 (W. J. Gertsch, AMNH), 39 (incl.
holotype), 13 July 1931 (W. J. Gertsch, AMNH), 13,

1 July 1940 (Gertsch & Hook, AMNH), 23 49; Seven-
mile Creek 8 mi. N Fish Lake, 9400 ft, 14 July 1973
(B. R. Vogel, BRV, NRS), 33 189. Summit County: W
Wasatch Station, 3 June 1933 (W. Ivie, AMNH), 13
19. Washington. Douglas County: 5-10 mi, E Bridge-
port, prairie community, in wet area, amongst Arte-
misia, 1 July 1964 (D. C. Lowrie, AMNH), 19. Wyo-
ming. Teton County: Grand Teton National Park, S
Jackson, 24 June 1938 (W. Ivie, AMNH), 13; Grand
Teton National Park, Moran area (Jackson Hole Bio-
logical Research Station, Uhl Hill &c.), July-August
1961-69 (D, C. Lowrie, AMNH), 23 179; Grand Teton

Pardosa wasatchensis Gertsch from 8 mi. N Fish Lake,

National Park, Lake Solitude, moist meadow, 3 August
1962 (D. C. Lowrie, NRS), 13; Teton National Forest
(all near Gros Ventre R.), Bridge Creek, in moist sedge-
grass field, 30 August 1964 (D. C. Lowrie, CNC), 39;
same, Lafferty Creek, in moist willows, 30 August 1964
(D. C. Lowrie, NRS), 19; same, nr Soda Lake, 30 Au-
gust 1964 (D. C. Lowrie, AMNH), 59. Uinta County:
10 mi. E Evanston, 18 July 1935 (W. Ivie, AMNH),
1 9; Yellowstone National Park: Bridge Bay, 9 July 1935
(W. Ivie, AMNH) 13 29, 20 June 1938 (W. Ivie,
AMNH), 213 169; Yellowstone Lake, 21 June 1938
(W. Ivie, AMNH), 19.

One 3 from Canada, Ontario, St. Thomas, 1928
(McBride, AMNH) is believed to be incorrectly re-
corded (cf. Dondale & Redner 1986:818 concerning
material of other Pardosa species incorrectly recorded
from St. Thomas).

Habitat.— According to Lowrie (1973:1 10) “a moist
meadow form. . .[but] further characterization of its
microhabitat is needed.”

Pardosa vogelae, new species

Figs. 2, 4, 7, 8, 11, 13, 16, 18, 20, 21b, 22;

Table 1

Type.— Male holotype from United States:
Utah, Daggett County, Leidy Peak, 1 1 ,500 ft (Be-
atrice R. Vogel & C. Durden), deposited in
AMNH.

Etymology.— Named for Dr, Beatrice R. Vo-
gel, one of the collectors, who has contributed
substantially to the exploration of the North
American lycosid fauna.

KRONESTEDT-HOLARCTIC PARDOSA

181

2.0

wasatchensis

2.5

• • o

CD O
O

b)

2.0-

o o

CD

1.5

vogelae

m •

• o

• O •CDO

3.0

3.5

3.0

3.5

Figure 21.-— Tibia I length (TiIL)/carapace length in adult males (closed circles) and females (open circles), a)
Pardosa wasatchensis Gertsch; b) P. vogelae sp. n.

Diagnosis.— Males may be distinguished by the
very prominent, curved and pointed tooth of the
terminal apophysis, and by the configuration of
the conductor (Fig. 4); females by the propor-
tions of the flask-shaped epigyne, with a very
wide septum and the rims of lateral elevations
characteristically curved (Fig. 16).

M^ailQ.—Holotype: Total length 6.6 mm; car-
apace 3.40 mm long, 2.45 mm wide.

Carapace: Dusky brownish; median band in
thoracic part yellowish; lateral bands light
brownish, uneven in width and with darker
breaks, not continuous to clypeus. Postocular
spots in cephalic part brownish. Sides of thoracic
part with numerous short dark hairs, fewer re-
cumbent grayish ones, and some longer erect,
somewhat wavy hairs. Median band with recum-
bent whitish hairs, around fovea additionally with
short and few long erect dark hairs. Clypeus yel-
lowish, in apparently old specimens now avail-
able devoid of hairs except long forwardly di-
rected dark ones. Chelicerae brownish with darker
streaks, distally yellowish on inner side, fur-
nished with dark hairs. Sternum blackish brown
with small light median stripe in front, furnished
with recumbent light and more erect dark hairs.

Eyes: Width of row I 48 (slightly procurved as
seen from front), row II 65, row III 85, row II™
III 62. Diameter of AME 10, ALE 10, PME 24,
PEE 19. Distance between AME 7, between AME
and ALE 2.

Abdomen: Dorsally grayish brown; posteriorly
on each side of median area a row of darker
patches with light dots between patches. Light
grayish brown, dark-bordered lanceolate stripe

in front. Dorsum with long erect and short dark
hairs as well as light hairs (latter numerous in
lanceolate stripe). Sides of abdomen mottled with
yellowish brown spots on grayish background.
Venter brownish with recumbent white pubes-
cence and scattered erect dark hairs.

Legs: (Table 1) Femora yellowish with sooty
brown dorsal markings, less sooty ventrally. Out-
er segments yellowish to light brownish. Hairi-
ness of leg I as in rest of legs. Tibia I with two,
only distal or no retrolateral spine(s).

Palp: Patella 0.65 mm, tibia 0.65 mm, cym-
bium 1.35 mm. Femur, patella, tibia and cym-
bium dusky brownish; femur with darker mark-
ings, cymbium lighter distally. Patella with dark
hairs, tibia with numerous long and short dark
hairs, cymbium with dark hairs except distally.
Tegular apophysis in ventral view (Figs. 2, 13)
shaped like bird’s head, with bulky, rounded bas-
al part (including anteriorly directed branch); lat-
eral process shorter than width of basal part, not
considerably narrowing before slightly hook-
shaped tip. Terminal apophysis (in ventral view;
Figs. 4, 1 1) with curved, heavily sclerotized tooth-
like process protruding forwards, tip acute; scler-
otized retrolateral process, end pointed; heavily
sclerotized rounded portion protruding posteri-
ad. Conductor (as seen in Figs. 4, 11) distally
bifurcate, upper branch longest, evenly tapering
to pointed tip; lower branch short, unsclerotized.
Embolus as in Figs. 7, 8.

Allotype: Total length 6.5 mm (car-
ried egg sac); carapace 3,35 mm long, 2.45 mm
wide.

Similar to male in color pattern and hairiness.

182

THE JOURNAL OF ARACHNOLOGY

Figure 22.— Collection localities of Pardosa was-
atchensis Gertsch (circles) and P. vogelae sp. n. (tri-
angle). Type localities of P. wasatchensis (1) and P.
subra Chamberlin & Ivie (2). Large circle refers to more
than three close localities.

Carapace with more distinct yellowish median
and broken lateral bands, latter with whitish hairs.
Lanceolate stripe on abdomen more clear yel-
lowish.

Legs: (Table 1) Brownish; femora laterally yel-
lowish, at least in distal half, sometimes with
traces of darker pseudoannulation; tibia I with
no or only distal retrolateral spine.

Epigyne: Flask-shaped (Figs. 16, 18; cleared
Fig. 20). Narrow indistinct (unsclerotized) septal
ridge continuing into very wide septum, evenly
rounded posteriorly. Sclerotized rim of lateral
elevations characteristically curved (Fig. 18, ar-
row). Lateral elevations coming close posteriorly,
separated by narrow slit. Two deep pockets cov-
ered by septum in front of extensive cavity scle-
rites. Receptacles comparatively narrow (Fig. 20).

Size variation. —Carapace lengths of material
measured: males 3.25-3.50 mm {n = 1 6), females
3.20-3.80 mm {n = 8); tibia I length vs. carapace
length in Fig. 21b.

Material examined. — UNITED STATES. Utah.
Daggett County: Uintah Mountains, Spirit Lake, Au-
gust 1935 (D. Cottam, AMNH), l9. Daggett & Uintah
Counties: Leidy Peak, 11,500 ft, 5-6 August 1964 (B.
R. Vogel & C. Durden, AMNH, BRV, CNC, MCZ,
NRS), 16(3 79 (inch holo- and allotype from 5 August).

ACKNOV^LEDGMENTS

I am indebted to the following persons for loan
and/or donation of material: Dr. C. D. Dondale
and Mr. J. H. Redner, Biosystematics Research
Centre, Agriculture Canada, Ottawa (including
tranference of material on loan to them); Prof.
D. C. Lowrie, Santa Fe (New Mexico); Dr. N. 1.
Platnick and Mr. L. Sorkin, American Museum
of Natural History, New York; and Dr. Beatrice
R. Vogel, Helena (Montana). I thank Dr. Don-
dale and Mr. Redner for reviewing the manu-
script.

LITERATURE CITED

Bonnet, P. 1958. Bibliographia araneorum, 2(4).
Toulouse.

Chamberlin, R. V, & W. Ivie. 1942. A hundred new
species of American spiders. Bull. Univ. Utah, 32(1 3)
(Biol, ser., 7(1)): 1-1 17.

Dondale, C. D. & J. H. Redner. 1986. The colora-
densis, xerampelina, lapponica, and tesquorum
groups of the genus Pardosa (Araneae: Lycosidae)
in North America. Canadian Entomol., 118:815-
835.

Dondale, C. D. & J. H. Redner. 1990. The insects
and arachnids of Canada. Part 1 7. The wolf spiders,
nurseryweb spiders, and lynx spiders of Canada and
Alaska (Araneae: Lycosidae, Pisauridae, and Oxy-
opidae). Publ. Dept. Agric. Canada, 1856:1-383.
Gertsch, W. J. 1933. New genera and species of North
American spiders. American Mus. Novitates, 636:
1-28.

Kronestedt, T. 1975. Studies on species of Holarctic
Pardosa groups (Araneae, Lycosidae). 1. Redescrip-
tion of Pardosa albomaculata Emerton and descrip-
tion of two new species from North America, with
comments on some taxonomic characters. Zool. Scr.,
4:217-228.

Kronestedt, T. 1981. Studies on species of Holarctic
Pardosa groups (Araneae, Lycosidae), 11. Redes-
criptions of Pardosa modica (Blackwall), Pardosa
labradorensis (Thorell), and Pardosa sinistra (Tho-
rell). Bull. American Mus. Nat. Hist., 170:1 1 1-124.
Kronestedt, T. 1986. Studies on species of Holarctic
Pardosa groups (Araneae, Lycosidae). III. Redes-
criptions of Pardosa algens (Kulczyhski), P. septen-
trionalis (Westring), and P. sodalis Holm. Entomol.
Scandinavica, 17:215-234.

Kronestedt, T. 1988. Studies on species of Holarctic
Pardosa groups (Araneae, Lycosidae). IV. Rede-
scription of Pardosa tetonensis Gertsch and descrip-

KRONESTEDT-HOLARCTIC PARDOSA

183

tion of two new species from the western United
States. Entomol. Scandinavica, 18:409--424.
Lowrie, D. C. 1973. The microhabitats of western
wolf spiders of the genus Pardosa. Entomol. News,

84:103==! 16.

Lowrie, D. C. & C. D. Dondale. 1981. A revision of
the nigra group of the genus Pardosa in North Amer-

ica (Araneae, Lycosidae). Bull. American Mus. Nat.

Hist, 170:125-139.

Roewer, C. F. 1954. Katalog der Araneae 2a. Bru-
xelles.

Manuscript received 18 May 1993, revised 20 July 1 993.

1993. The Journal of Arachnology 21:184-193

NEWLY-DISCOVERED SOCIALITY IN THE NEOTROPICAL
SPIDER AEBUTINA BINOTATA SIMON (DICTYNIDAE?)

Leticia Aviles': Museum of Comparative Zoology, Harvard University, Cambridge,
Massachusetts 02138 USA.

ABSTRACT. The neotropical spider Aebutina binotata Simon (Dictynidae?), previously known from a few
museum specimens, was discovered to live in colonies and to exhibit highly cooperative behaviors that would
classify it as non-territorial, permanently social. Colonies of this species, that contained from 14-106 adult
females plus their offspring, were observed in a tropical rainforest site in Eastern Ecuador. The spiders occupied
communal nests in which they cooperated in prey capture and fed communally on the prey. Large prey items
were moved to the feeding site by the coordinated effort of two or three individuals. The spiders periodically
carried out web maintenance activities; but when widespread damage to the nest occurred, they moved as a
group to a new location. Care of the brood appeared to be communal since the offspring from different mothers
intermixed in the colonies and were all cared for by a decreasing number of surviving females. Adult females
participated most heavily in all the activities of the colonies, with no apparent division of labor among them.
In particular, no reproductive division of labor was observed: all adult females in colonies observed throughout
the egg-laying period apparently laid a single egg sac each.

The most advanced form of social behavior
known for spiders involves cooperation among
members of a colony in building and maintaining
a communal nest, capturing prey on which to
feed communally, and taking care of the offspring
(Buskirk 1981; D’ Andrea 1987). These tasks are
performed by members of the same generation
without any apparent division of labor among
them (Darchen & Delange-Darchen 1986). In
particular, unlike what occurs in the most highly
social insects (Wilson 1971), no specialized re-
productive castes are present in social spiders:
most, if not all (Vollrath 1986), individuals in a
social spider colony apparently bear offspring.
This form of social behavior, labeled as non-
territorial permanently social (D’ Andrea 1987)
or quasisocial (Wilson 1971), has arisen inde-
pendently in at least six spider families. To date,
a total of 14 species in eight genera have been
described as possessing the traits that would de-
fine them as having attained this level of sociality
(for partial lists see Buskirk 1981; and D’ Andrea
1987; for species not included in these partial
lists, see Main 1988, Rypstra & Tirey 1989; Avi-
les in press 1, in press 2). Ten of these species
have been described or their sociality discovered
in the last 30 years, indicating that highly co-

' Present address: Dept, of Ecology and Evolutionary
Biology, Univ. of Arizona, Tucson, Arizona 85721
USA

operative behavior in spiders, although rare rel-
ative to the total number of spider species (about
40,000 described species), is perhaps more wide-
spread than previously believed.

Here, I report on newly discovered sociality
in the neotropical cribellate spider Aebutina bi-
notata Simon (Simon 1 892). This species was de-
scribed by Simon at the end of the last century
from a few female specimens collected in the
Brazilian Amazonas Province (Simon 1892).
From a systematic point of view, Aebutina bi-
notata has proven to be an enigmatic species
whose placement in any of the currently de-
scribed spider families is not fully resolved (Leh-
tinen 1967). The genus Aebutina, of which A.
binotata is the type species, is temporarily as-
signed to the Dictynidae (Petrunkevitch 1928;
Millot 1933), after having been originally placed
by Simon (1892) in the Uloboridae. Previous to
the present study, no information on the life his-
tory or behavior of A. binotata was available.
The observations that I report here indicate that
this species is colonial and that it exhibits the
strongly cooperative behaviors common to non-
territorial permanently social spiders.

In this paper I describe the structure of the
nests and colonies and report on behavioral as-
pects of the sociality of this species: observations
involving cooperation on web maintenance and
repair, colony relocation, prey capture and trans-
port, food sharing, communal brood care, and

184

AVILES~=SOCIALITY IN AEBUTINA BINOTATA SIMON

185

Figures 1, 2.~Aebutina binotata genitalia: 1. Male; 2
W. Maddison. Scale bars correspond to 0. 1 mm.

tolerance to members of other colonies. Even
though not directly related to sociality, I also
present observations on courtship and mating.

METHODS

The spider. — Live adult females of binotata,
which measure about 5 mm in length {x = 4.7,
n = 7), have a diamond-shaped, bright yellow
abdomen with a black spot on each side (there-
from the name binotata). The males (3.4 mm
when adult, n = 2) and early-instar individuals
are also yellow, of less intense coloration and
with somewhat less clearly marked spots. Males
are adult in the 1‘^ instar while females are adult
in the 8'^ instar (Aviles 1992). The egg sacs are
spherical and measure around 3.5 mm in di-
ameter. They consist of a mesh of white silk that
surrounds the yellow-colored embryos.

One of the reasons for the uncertainty in the
systematic placement of A. binotata has been the
lack of male specimens (Lehtinen 1967). In con-
nection with the present study I collected males
which have been deposited in the collections of
the Museum of Comparative Zoology, Harvard
University. In this study I do not address the
systematic placement of A. binotata. However,

I provide drawings of male and female genitalia
for future reference (Figs. 1, 2). Initial determi-
nation of the female specimens was done by J.
Hunter of the MCZ and later confirmed by com-
parison with the types by H. W. Levi.

The observations.— I studied colonies of A.

2

Female. Drawings from computer digitized images by

binotata in a tropical rainforest site by the Tar-
apuy River in Eastern Ecuador, Sucumbios (for-
merly Napo) Province (0° 08' S, 76° 16' W, 210
m above sea level). I first discovered two colonies
in January 1983 and then an additional one in
February 1984. From this date to September
1984, I visited the area on six occasions and
identified a total of 44 colonies (24 up to a July
visit and 20, either new or previously recorded
colonies that had relocated themselves, in the
September visit).

For each colony observed, I recorded its po-
sition, distance from the ground, and structure of
the nest, including the size of the leaf (or leaves)
supporting the colony and the percent of the leaf
(or leaves) occupied. I counted the total number
of adult females, egg sacs, and males present in
the colonies, and estimated the number of ju-
veniles of different size classes (i.e., instars).

I conducted behavioral observations on an op-
portunistic basis. Observations were conducted
during the day, usually between 900 and 1800
h, on one occasion from 700 h. Activities in-
volving nest maintenance and repair were ob-
served regularly in the colonies. I observed one
complete short-distance colony relocation event,
a portion of another, and a 1 3 -hour period ( 1 200-
1800 h and 700-1400 h of the day after) of one
long-distance relocation event. I also obtained
indirect evidence of six other relocation events.
I recorded complete sequences of prey capture,
including prey transportation and initiation of

186

THE JOURNAL OF ARACHNOLOGY

feeding, on 1 3 occasions, and on five others after
feeding had been initiated. Four cases of prey
rejection were observed. Other prey capture or
feeding events were observed on a more casual
basis. I observed courtship and mating for a pe-
riod of 1 00 min (1215-1 400, August 4, 1 984) in
one colony containing four adult males and 106
females. I observed a total of four copulations
and 1 6 unsuccessful mounting attempts. One ad-
ditional copulation was observed at an earlier
date in a different colony (colony 5, May 20, 1155
h). A pilot test of tolerance to conspecihcs in-
volved the introduction of one adult female into
a foreign nest. Additionally, for ten weeks I
maintained in the laboratory two colonies on
which I conducted casual behavioral observa-
tions.

RESULTS

The nest and colonies.— The nests of A. bi-
notata are basically two-dimensional structures
consisting of one or a few contiguous leaves and
their connecting branches covered on both sur-
faces by a continuous layer of silk (Fig. 3). The
outer surface of the web is covered by cribellate
silk that cause insects to get entangled when land-
ing on it. The sheet of silk on the underside of
the leaves (the lower web) is not attached to the
leaf blade, but separated by an open space that
is used by the spiders as refuge. Egg sacs and
spiderlings occupy this space, sitting on the inner
surface of this lower web and congregated to-
wards the center. Openings allow spiders to move
freely from one surface of the web to the other.
Adult spiders sit on the outer surface of the lower
web, lined up along the edges of the leaf (Fig. 4)
in a position that allows them rapid access to the
top of the leaf where insects usually get entangled.
Major perturbations to the nest, such as exper-
imental shaking, cause the larger individuals in
a colony to drop to the ground on silk draglines
along which they return once the disturbance has
stopped.

The size of the colonies, measured as the num-
ber of adult females present in a colony around
the time the eggs were being laid, ranged from
14-106 {n = 19, median = 40, mean = 46.7 ±

1 1 .8, 95% conf int., Fig. 5). Colonies with young
spiderlings could contain up to eight hundred
individuals; but, because of the smaller size of
the young, the total biomass was probably within
a similar range. The area occupied by a nest was
found to be proportional to the number of adult
females present (Fig. 6); it ranged from 74-200

cm^ and included either a portion of a large leaf
or several small leaves. Nests were found be-
tween 0.5 m and 4,5 m from the ground, although
nests occurring higher than this would have
probably been missed. The nests appeared in-
visible when seen from above and could only be
located by looking for the spiders underneath
leaves. From below, the appearance of the col-
onies was striking because of the bright yellow
coloration of the adult females and spiderlings
and the whiteness of the egg sacs.

Web construction and repair.— The nests were
constructed and maintained cooperatively. Web
maintenance activities involved: (a) periodically
adding cribellate silk to maintain the stickiness
of the web, (b) removing and replacing damaged
web following destruction by the rain, (c) re-
pairing holes left by ensnared prey, and (d)
throwing out debris. All these activities could be
simultaneously performed by several individuals
in different areas of the nest. For instance, in a
nest of 13 adults and around 190 juveniles, web
repair after a rain storm required about two hours
during which 16-26 spiders working at a time
removed the damaged silk, added new silk lines
across the surface of the leaves, and added a final
layer of cribellate silk.

Unlike other social spiders where regular web
reinforcement activities take place exclusively at
sunset (e. g., in Anelosimus eximius, Tapia & De
Vries 1980), A. binotata spiders added cribellate
silk periodically throughout the day. The spiders
were inactive in colonies seen early in the morn-
ing (around 700-730 h) and all activity seemed
to have ceased in two colonies observed at 1 800
h. At this time the spiders retreated to the un-
derside of the leaves and presumably carried out
only prey capture activities until the following
day.

Colony relocation.— Extensive destruction of
the web resulted in colonies abandoning their
original nest. One of the colonies, for instance,
moved to a leaf 1 2 cm from its original location
after the leaf that supported its nest dried out.
At later dates, this and three other colonies moved
from 0.3-4. 5 m after a heavy rain destroyed their
nests. Colony relocation involving much greater
distances (> > 5 m), independent of web destruc-
tion, apparently took place prior to mating and
egg laying (Aviles 1992).

Just prior to colony relocation, a fraction of
the spiders in a colony could be seen initiating
the production of airborne silk lines by hanging
down 2-5 cm from their nest (see Eberhard 1987

AVILES -SOCIALITY IN AEBUTINA BINOTATA SIMON

187

Figures 3, 4.-~Aebutina binotata colonies in Tarapuy, Ecuador: 3. Whole nest; 4. Adult females lined up along
the edge of their nest.

188

THE JOURNAL OF ARACHNOLOGY

number of adult females

Figures 5, 6.— -5. Number of adult females (and/or egg sacs, whichever is larger) present in Aebutina binotata
colonies seen sometime immediately before or during the egg laying period; 6. Correlation between the number
of adult females in a colony and the surface occupied by their nest (measured in cm^).

for method of airborne line production). Once a
dragline became attached to the nearby vegeta-
tion it was first followed by the spider that orig-
inated it and then by other spiders. Even if there
were draglines attaching in different directions,
all spiders eventually moved along a single drag-
line since the spiders following an isolated route
returned to join the majority. This method was
repeated from one stop to the next until the even-
tual settlement of the colony. In the cases ob-
served, all individuals in a nest moved to the
new location.

The relocation of a colony could be completed

in a few hours or in consecutive days. For in-
stance, a colony with eight adult females and
more than 200 juveniles (mostly 4th-instar)
moved to a location 30 cm away in the span of
5-”6 hours. Another colony that contained eight
adult females and approximately 350 3rd-"5th
instar juveniles moved first to an intermediate
stop 1.8 m from its original location, remained
there for a period of a day and then continued
on for an additional meter. Colonies migrating
previous to mating and egg laying appear to mi-
grate for longer periods. For instance, a colony
with adult males and females that I followed for

AVILES-SOCIALITY IN AEBUTINA BINOTATA SIMON

189

1 3 hours had not settled when I stopped the ob-
servations after a day and a half and 50 m of
group migration (Aviles 1992).

Prey capture and feeding, —Prey capture in-
volved the simultaneous participation of 1-6 in-
dividuals, apparently depending on the size of
the prey and the efforts it made to free itself. If
present in sufficient numbers, only the adult fe-
males participated in prey capture, though later-
instar spiderlings would participate when there
were relatively few adult females in the nest or
the prey were small. The females placed along
the edge of the nest closest to an struggling insect
would rush towards it and attack it by hrst biting
its appendages and then other parts of the body.
Once completely overcome, prey items were
moved to the underside of the leaves where feed-
ing took place. Adult spiders initiated prey di-
gestion (Fig. 7) and later left the prey to the ju-
veniles. Prey trapped and consumed included
wasps, mosquitoes, cockroaches, one large ant,
some beetles. A small coccinelid and another
small beetle were rejected. Very large insects,
such as a 4 cm long moth and a 2 cm cetonine
beetle, were ignored.

Group transport. —Transportation of items
around the nest, either prey to be consumed or
debris to be thrown out, were also among the
activities regularly carried out by the spiders.
Small items were handled by individual spiders.
The transportation of large objects, on the other
hand, required a group effort that was particu-
larly challenging given the sticky nature of the
entire surface of the web. Complete group trans-
port events were observed on three occasions. A
large ant moved from the upper to the lower
surface of the nest, for instance, involved the
participation of three individuals. While one in-
dividual cut pieces of web to release the ant, the
second one pulled and the third pushed the ant
in a given direction. Once released, the ant was
transported toward the edge of the nest. One in-
dividual lifted the ant from below so as to main-
tain it at a distance from the surface of the web.
As this individual walked towards the edge, the
other two individuals, one in front and one in
the back, helped by pulling and pushing in the
required direction. Once at the edge, the ant was
successfully moved to the other side by having
one spider hold it from above as it walked to-
wards the edge while the other two supported the
item and pulled it from below. It took the spiders
four minutes to disentangle and bring the ant to
the edge, three more minutes to bring it over the

edge and onto the other side of the nest, and
another three minutes to move it one cm into
its final position where feeding was initiated. Two
other group transport events observed required
two and three individuals to transport an ho-
mopteran and a fly, respectively.

Communal brood care.— While most of the
adult females lined up along the edges of the nest
ready to participate in prey capture, a number
of them mounted guard by the egg sacs and spi-
derlings. Besides sac guarding, parental care in-
cluded catching prey for the spiderlings and ini-
tiating the enzymatic digestion of the prey.
Regurgitation feeding was not observed during
the study period, though specific studies would
be required to confirm its absence.

Two lines of evidence suggest that parental
care is communal: (1) the spiderlings in a nest
intermix freely, in a way that it does not seem
possible for a mother to discriminate between
her own and other mother’s offspring; and, (2)
the number of adult females present in the col-
onies drops continuously (most likely due to
mortality) during the incubation and emergence
periods, while all the egg sacs and offspring pres-
ent in the colonies continue to be cared for. For
instance, half the adult females of one of the
colonies observed throughout most of its life cy-
cle were already gone by the time their offspring
had only reached their third or fourth instar. All
the offspring, however, continued to be cared for
by the remaining females whose number contin-
ued to decrease until none were left by the time
the offspring had reached their sixth instar. This
indicates that a large majority of the offspring
were raised by females other than their mother.

Division of labor.— Adult females tended to
participate disproportionately in all the activities
of the colony, though later-instar juveniles par-
ticipated to varying degrees. Later instar juve-
niles, for instance, handled small prey or partic-
ipated in group efforts when the relative number
of adults in the nest was low. Juveniles were
relatively more active in web maintenance ac-
tivities, particularly in laying down cribellate silk,
though the numbers in which they participated
were not representative of the proportion in which
they occurred in the colonies. For instance, after
a storm had destroyed the upper web of one of
the colonies under study, all adult females (a total
of eight) and 20% of the juveniles (out of a total
of 200) were seen removing the damaged web
and laying down new strands of silk. In a different
colony, which contained 13 adult females and

190

THE JOURNAL OF ARACHNOLOGY

Figures 7, S.—Aebutina binotata. 1. Group of adult females feeding on a prey; 8. Mating couple in Tarapuy,
Ecuador,

AVILES^-SOCIALITY IN AEBUTINA BINOTATA SIMON

191

around 190 juveniles, 30-100% of the females
were active at different times during a two-hour
period following a storm, while only 4-7% of the
juveniles were. The youngest juveniles that ap-
peared able to participate in activities such as
laying cribellate silk belonged to the 4th-instar.
Immature males in the colonies kept in the lab-
oratory were seen laying silk during web rein-
forcement, while it was not possible to determine
whether adult males participate in this or any
other activity since they were seen in the colonies
for only a short period of time.

There was no evidence of division of labor
among spiders of a given age group: a task started
by one individual was often completed by an-
other and the same individual could be seen car-
rying out different tasks. Marking experiments,
however, are needed to confirm these observa-
tions. Regarding reproductive division of labor,
in all the colonies observed throughout the egg
laying period the total number of eggs cases pres-
ent was the same as the number of adult females
in the nest (Table 1). Since all the egg sacs in a
colony were laid within a short time span (Aviles
1992); and thus it is unlikely that some fe-
males may have laid more than one case, it fol-
lows that all females in the colonies observed
reproduced. Reproductive division of labor,
therefore, appears absent in Aebutina binotata.

Social interactions and tolerance to conspecif-
ics.—The 14 adults and 74 juveniles kept in the
laboratory in a 40 x 30 x 30 cm terrarium re-
mained aggregated throughout a 1 0-week obser-
vation period. When they relocated their nest
within the terrarium, all the spiders moved to-
gether. Encounters between spiders, which were
common during the course of their daily activ-
ities, involved touching each other with the legs
and pedipalps.

Spiders in the field did not appear to discrim-
inate against members of other colonies. One
spider experimentally introduced into a foreign
nest was initially approached by other spiders,
probably in response to the vibrations produced,
but was soon treated as a member of the colony.
After its introduction, the spider resumed the
activity (adding cribellate silk) it had been per-
forming when removed from its native nest. A
month later, the spider was still in the colony,
and, aside from the experimental mark, was in-
distinguishable from other spiders in the nest.

Courtship and mating.— I observed courtship
and mating in a colony that contained 106 fe-
males, 4 males, and 1 egg sac. During the 100
min of observations, three of the males attempt-

Table 1 .—Number of females and egg sacs laid in A.
binotata colonies periodically observed throughout the
egg laying period. In addition to the 45 egg sacs shown,
colony 14 contained newly eclosed juveniles from
around four egg sacs.

Colony

9

22

23 14

25

Adult females

102

44

30 48-49

52

Egg sacs present

102

44

29 45 +

54

ed insistently to mount females. Most of the at-
tempts observed (16 out of 18) were rejected by
the females who either moved away or resisted.
Two resulted in copulations. Two additional
copulations had already been initiated when the
observations started. Copulations took place with
the male over the back of the female and both
facing in opposite directions (fig. 8; position “c”
in Foelix 1982, p. 195). In one of the matings
observed the female remained motionless all
through the mounting. In a second mating ob-
served, the male seemed to exert force over the
female. The two complete copulations observed
lasted around 4 min. Given the disparity in the
numbers of males and females, and the obser-
vation that most or all the females in a colony
lay eggs (see above), it follows that each male is
able to fertilize a large number of females.

DISCUSSION

The social behavior here described for A. bi-
notata has strong similarities with that of other
non-territorial permanently social spiders pres-
ent in the genera Achaearanea (Theridiidae),
Agelena (Agelenidae), Anelosimus (Theridiidae),
Diaea (Thomisidae), Mallos (Dictynidae), Ste-
godyphus (Eresidae) , Tapinillus (Oxyopidae), and
Theridion (Theridiidae) (Buskirk 1981; D’ An-
drea 1987; Aviles pers. obs.). These similarities,
which include cooperative web building and
maintenance, cooperative prey capture, com-
munal feeding, communal brood care, tolerance
to members of other colonies, and a lack of castes,
are particularly striking given the phylogeneti-
cally diverse set of species in which they have
evolved.

One feature common to this diverse set of spe-
cies, which might be to a large extent responsible
for these similarities is an irregular type of web,
present in all but the social thomisid (Main 1988;
Evans & Main 1 993), but absent in other colonial
but non-cooperative species such as Metabus

192

THE JOURNAL OF ARACHNOLOGY

gravidus or Philoponella republicana (Buskirk
1981; D’ Andrea 1987). An irregular web is
thought to constitute a preadaptation for coop-
erative behavior in spiders because it allows
communal habitation and the simultaneous in-
volvement of more than one individual in web
construction and repair as well as in prey capture
(Buskirk 1981). Cooperative web building, be-
cause it allows the construction of a relatively
large area or volume of entangling web, leads to
the capture of relatively large prey items, which,
in turn, require the concurrence of several in-
dividuals for their subjugation. Large prey items
can then be shared by several individuals in a
nest, leading to communal feeding, a trait com-
mon again to the non-territorial permanently-
social species studied (Buskirk 1981; see also
Main 1988; Rypstra & Tirey 1989), but absent
in the colonial orb weavers. A communal nest
also facilitates communal care of the brood be-
cause it renders discrimination among spider-
lings intermixed in a common space impractical.

A. binotata is typical among the non-territorial
permanently social spiders in having an irregular
type of web and in having developed cooperative
prey capture, communal feeding, and communal
brood care. As in these other species (Buskirk
1981), cooperation allows A, binotata spiders the
capture of prey items larger than those single
individuals could handle and facilitates the shar-
ing of prey among a larger proportion of colony
members than those participating in their cap-
ture. The advantages of cooperation in the care
of the brood become specially evident in A. bi-
notata, where the survival of orphaned offspring
is only possible because surviving females in-
discriminately care for all the young in a nest (see
also Christenson 1984; D’ Andrea 1987).

The ways in which the architecture of the nests
of A. binotata differs from that of other non-
territorial permanently social species may be re-
sponsible for some of the features that appear
unique to this species. One such feature is the
cooperative transport of prey items from the site
of their capture to their consumption site, a trait
that in this species is developed to a greater ex-
tent than in any other social spider. Three aspects
of the structure of the nests of A. binotata pose
a special challenge to prey transport in this spe-
cies: ( 1 ) the sticky nature of the web surface; (2)
the relatively small capture area; and, (3) the fact
that prey items need to be brought over the edge
of the leaf with the consequent danger of acci-
dentally dropping them out of the nest. In other
cooperative spiders, webbing usually surrounds

the prey in all directions during its transport and
the nests are considerably larger than the prey
being transported, so that their overall efficiency
is not critically affected by damage to a portion
of the web caused by dragging a prey item.

The instances of group transport that I ob-
served in A. binotata required a degree of coor-
dination among the individuals involved that
can only be explained if some sort of commu-
nication was taking place among them. Prey
transport in other species usually involves a
number of individuals pulling in the same gen-
eral direction or some form of, apparently un-
coordinated, relay activity (e.g., Ward & Enders
1985). In the genus Agelena the prey is either
carried by a single individual (Krafft 1971) or
eaten on the site of its capture when it is too large
for individual transport (Darchen 1967; Krafft
1971). In Achaearanea disparata (Darchen 1967)
some individuals pull while others aid in cutting
the threads that hinder the movement of the item,
suggesting, in this case, some degree of coordi-
nation. It must be noted that group transport is
an extraordinary task to be performed by an in-
vertebrate. Even among vertebrates, group trans-
port is only known among humans, dolphins,
whales, and some canids (Moffett in press).
Among invertebrates, the only other reported case
of group transport is represented by the ants (Mof-
fett in press).

Another feature apparently unique to A. bi-
notata which may also result from the special ar-
chitecture of its nests is its nomadic habit that
leads to the periodic relocation of its colonies.
The nests of other non-territorial permanently
social species, which are usually expanded and
occupied by more than one generation of spiders,
are three-dimensional structures that are prob-
ably expensive to build and whose prey capture
efficiency does not appear to depend on their
invisibility. The two-dimensional nests of A. bi-
notata, on the other hand, may be less expensive
to rebuild and, most importantly, they appear to
critically depend on their invisibility to trap prey.
Such invisibility can best be achieved by rebuild-
ing the nest in a new and debris-free locality.

The observation that egg laying is universal in
A. binotata (Table 1) illustrates perhaps to an
extreme a feature common to non-territorial,
permanently social spiders which critically dif-
ferentiates them from the most highly social in-
sects: a lack of reproductive castes (Buskirk 1981;
Darchen & Delange-Darchen 1986). Even though
competition over resources leading to differences
in reproductive success are not entirely absent

AVILES SOCIALITY IN AEBUTINA BINOTATA SIMON

193

in social spiders (Riechert 1985; Seibt & Wickler
1988; Vollrath 1986; Rypstra in press), social
spider colonies critically contain multiple re-
productives of both sexes that can mate among
themselves to produce subsequent generations.
This leads to mating within colonies and to the
highly subdivided population structure that
characterize non-territorial permanently social
species (Lubin & Crozier 1985; Smith 1986; Main
1988; Roelolfs & Riechert 1988; Aviles 1992).
In contrast, perpetual inbreeding is rare or absent
among the eusocial insects in which nuptial flights
result in the crossing of individuals from different
nests (Wilson 1971). Evidence that A. binotata
has followed the route of other non-territorial
permanently social spiders in developing strong
population subdivision leading to intercolony se-
lection and female-biased sex ratios will appear
elsewhere.

ACKNOWLEDGMENTS

I am grateful for the logistic support of City
Production Company (CEPCO), the hospitality
of Roberto and Maruja Aguirre, and the com-
ments on the manuscript by W. Maddison and
G. Uetz.

LITERATURE CITED

Aviles, L. 1992. Metapopulation biology, levels of
selection and sex ratio evolution in social spiders.
Ph. D. Thesis, Harvard University.

Aviles, L. in press 1 . Social behaviour in a web build-
ing lynx spider, Tapinillus sp. (Araneae, Oxyopi-
dae). Biol. J. Linnean Soc.

Aviles, L. in press 2. Aranas sociales de la Reserva
Faunistica Cuyabeno, In Ecologia de la Amazonia
del Ecuador: el noreste amazdnico y la Reserva
Faunistica Cuyabeno (E. Asanza & T. De Vries,
eds.), Publicaciones Abya-Yala, Quito.

Buskirk, R. 1981. Sociality in the Arachnida. Pp.
282-367, In Social Insects (H. R. Hermann, ed.),
vol. 4. Academic Press, New York.

Christenson, T. 1984. Behaviour of colonial and sol-
itary spiders of the Theridiid species Anelosimus
eximius. Anim. Behav., 32:725-734.

D’ Andrea, M. 1987. Social behaviour in spiders
(Arachnida, Araneae). Italian J. ZooL, N. S. Mon-
ogr., 3.

Darchen, R. 1967. Ethologie de quelques araignees
sociales: L’interatraction, la construction et la chasse.
Actes Vme Congress U.LE.I.S., Toulouse. Pp. 335-
345.

Darchen, R. & B. Delange-Darchen. 1986. Societies
of spiders compared to the societies of insects. J.
Arachnol., 14:227-238.

Evans, T. A. & B. Y, Main. 1 993. Attraction between
social crab spiders: silk pheromones in Diaea so-
cialis. Behav. Ecol. 4:99-105.

Eberhard, W. 1987. How spiders initiate airborne
lines. J. Arachnol. 15:1-9.

Foelix, R. F. 1982. Biology of Spiders. Harvard Uni-
versity Press, Cambridge, Massachusetts.

Krafft, B. 1971, Contribution a la biologie et a la
ethologie d'Agelena consociata Denis (Araignee so-
ciale du Gabon). Ill partie. Biol. Gabonica, 7:3-56.
Lehtinen, P. 1967. Classification of the Cribellate
spiders and some allied families, with notes on the
evolution of the suborder Araneomorpha. Ann. Zool.
Fennici, 4:209.

Lubin, Y. D. & R. H. Crozier. 1985. Electrophoretic
evidence for population differentiation in a social
spider Achaearanea wau (Theridiidae). Insect. Soc.,
32:297-304.

Main, B. Y. 1988. The biology of a social thomisid

spider. Australian Arachnol., 5:55-73.

Millot, J. 1933. Le genre Aebutina (Araneides). Bull.

Soc. Zool. France, 53:92-95.

Moffet, M. in press. Group transport. Sci. American.
Petrunkevitch, A. 1928. Systema Aranearum. Trans.

Connecticut Acad. Arts Sci., 29:1 12.

Riechert, S. E. 1985. Why do some spiders cooperate?
Agelena consociata, a case study. Florida Entomol.,
68:105-116.

Roeloffs, R. & S. E. Riechert. 1988. Dispersal and
population-genetic structure of the cooperative spi-
der, Agelena consociata, in West African rainforest.
Evolution 42:173-183.

Rypstra, A. in press. Prey size, social competition

and the development of reproductive division of
labor in social spider groups. American Nat.
Rypstra, A. & R. Tirey. 1989. Observations on the
social spider, Anelosimus domingo (Araneae, Ther-
idiidae), in southwestern Peru. J. Arachnol,, 17: 368-
371.

Seibt, U. & W. Wickler. 1988. Why do “family spi-
ders”, StegodyphusZZ (Eresidae), live in colonies?
J. Arachnol., 16:193-198.

Simon, E. 1892. Histoire naturelle des Araignees I.
Roret, Paris, Pp. 221-222.

Smith, D. R. 1986. Population genetics of Anelosi-
mus eximius (Araneae, Theridiidae). J. Arachnol.,
14:201-217.

Tapia, Y. & T. De Vries. 1980. Tolerancia y coop-
eracion en la arana social Anelosimus jucundus del
bosque tropical Rio Palenque, Ecuador. Rev. Univ.
Cat61ica, Quito, 8:51-74.

Vollrath, F. 1986. Eusociality and extraordinary sex
ratios in the spider Anelosimus eximius (Araneae:
Theridiidae). Behav. Ecol. Sociobiol., 18:283-287.
Ward , P. I. & M. M. Enders. 1985. Conflict and
cooperation in the group feeding of the social spider
Stegodyphus mimosarum. Behaviour, 94:167-182.
Wilson, E. O. 1971. The Insect Societies. Belknap
Press, Cambridge, Massachusetts.

Manuscript received 27 January 1993, revised 25 May
1993.

1993. The Journal of Arachnology 21:194-201

DNA SEQUENCE DATA INDICATES THE POLYPHYLY
OF THE FAMILY CTENIDAE (ARANEAE)

Kathrm C. Hubert Thomas S. Haider 2, Manfred W. Muller Bernhard A. Huber*,
Rudolf J. Schweyen^, and Friedrich G. Barth*: *Institut fur Zoologie, Althanstr. 14;
1090 Wien; and ^Institut fiir Mikrobiologie und Genetik; Dr. Bohrgasse 9; 1030
Wien (Vienna), Austria.

ABSTRACT. Mitochondrial DNA fragments comprising more than 400 bases of the 16S rDNA from nine
spider species have been sequenced: Cupiennius salei, C. getazi, C. coccineus and Phoneutria boliviensis (Ctenidae),
Pisaura mirabilis, Dolomedes fimbriatus (Pisauridae), Pardosa agrestis (Lycosidae), Clubiona pallidula (Clubi-
onidae) and Ryuthela nishihirai (syn. Heptathela nishihirai\ Heptathelidae: Mesothelae). Sequence divergence
ranges from 3-4% among Cupiennius species and up to 36% in pairwise comparisons of the more distantly
related spider DNAs. Maximally parsimonious gene trees based on these sequences indicate that Phoneutria
and Cupiennius are the most distantly related species of the examined Lycosoidea. The monophyly of the family
Ctenidae is therefore doubted; and a revision of the family, which should include DNA-data, is needed.

Cupiennius salei (Ctenidae) is one of the most
extensively studied species of spiders (see Lach-
muth et al. 1985). The phylogeny of the Ctenidae,
a mainly South and Central American family, is
poorly understood; and systematists propose
highly contradicting views on its classification
and phylogenetic placement (see e. g., Lehtinen
1967; Bucherl 1969). Coddington & Levi (1991)
have recently questioned the monophyly of the
Ctenidae, showing that the available information
about taxonomically useful characters is still
meager and fragmentary.

In recent years the value of DNA sequence
data for taxonomic and phylogenetic research
has become increasingly clear. DNA sequences
contain a nearly inexhaustable quantity of in-
formation and may provide valuable insight al-
lowing the evaluation of groups whose phylogeny
is largely unresolved by morphological and other
data (Kocher et al. 1989; Gatesy et al. 1992;
Cunningham et al. 1992; review: Femholm et al,
1989). Moreover, sequencing DNA—above all
mitochondrial DNA (mtDNA; mainly maternal
inheritance, lack of recombination) —has specific
advantages over other techniques of genetic com-
parisons such as DNA/DNA hybridization or
isoenzyme analysis: e. g., greater resolving power
over a hierarchical range of intraspecific to in-
tergeneric comparison and easy comparability
with sequences from other species (see also Wil-
son et al. 1985). The polymerase chain reaction
(PCR, Mullis & Faloona 1987; Saiki et al. 1988)
is a fast alternative to conventional cloning to

get a high copy number of the DNA segment of
interest. The PCR depends on the availability of
oligonucleotides that specifically bind to the
flanking sequences of this DNA segment. These
oligonucleotides serve as primers for a polymer-
ization reaction that copies the segment in vitro.
The PCR-product obtained is suitable for direct
sequencing.

The principal aim of the present study was to
elucidate the phylogenetic position of the family
Ctenidae within the Lycosoidea sensu Homann
(1971), using four ctenids, two pisaurids, one ly-
cosid, one clubionid and one liphistiomorph spi-
der. Specific PCR-products were obtained by us-
ing primers for the mitochondrial 1 6S ribosomal
DNA, designed according to those used by Cun-
ningham et al. (1992). The molecular data largely
agree with the conclusions drawn from morpho-
logical taxonomy. The most intriguing and sur-
prising result of our study is the indication of a
considerable phylogenetic distance between Cu-
piennius and Phoneutria.

METHODS

Animals and DNA extraction.— Table 1 lists
the animals investigated in this study. In the big-
ger spiders, muscle tissue was dissected out of
the femora; for the smaller ones, the complete
prosoma and legs were used to extract DNA.
Tissues were put into a digestive solution (70
mM NaCl, 10 mM Tris-HCl pH 7.4, 25 mM
EDTA pH 8.0, 0.9% SDS, 6 p,g/m\ Proteinase
K), and incubated for 2-8 h in a water bath at

194

HUBER ET AL.-DNA SEQUENCE DATA INDICATES CTENID POLYPHYLY

195

Table L— Systematic position and geographical ori-
gin of the spiders investigated. The number of indi-
viduals that were sequenced is given in parentheses.

Spider classification

Geograph-

ical

origin

Mesothelae

(1) Ryuthela nishihirai
(Haupt 1979)

Japan

Opisthothelae

Clubionidae

(1) Clubiona pallidula
(Clerck 1757)

Austria

Lycosoidea

Ctenidae

(1) Cupiennius salei
(Keyserling 1876)

Mexico

(2) Cupiennius getazi

Simon 1891

Costa Rica

(1) Cupiennius coccineus

F. Pickard-Cambridge 1901

Costa Rica

(2) Phoneutria boliviensis

(F. Pickard-Cambridge 1897)

Costa Rica

Lycosidae

(2) Pardosa agrestis
(Westring 1862)

Austria

Pisauridae

(2) Dolomedes fimbriatus
(Clerck 1757)

Austria

(1) Pisaura mirabilis
(Clerck 1757)

Austria

M 1 2 3 4

1632 bp

517bp__
506 bp—

396 bp

344 bp- m

Figure 1.™ Products of a radioactively labelled PCR
amplification of a mt 1 6S rDNA fragment. M = marker;
1 = Cupiennius getazv, 2 = Pardosa agrestis\ 3 = Clu-
biona pallidula; 4 = Dolomedes fimbriatus. Arrow in-
dicates the main products.

50 °C. Proteins were then precipitated with 5 M
potassium acetate and RNA was digested by
RNAse A (125 jug/ml; 15 min, 37 °C). DNA was
precipitated with 60% v/v isopropanol ( 1 0 min,
-20 ®C), pelleted by centrifugation at 1 5,000 rpm
for 20 min at 4 ®C, washed with 400 pi of 70%
ethanol, dried at 45 and resuspended in 20-
40 Ml TE buffer (10 mM Tris-HCl pH 8.0, 1 mM
EDTA pH 8.0). The DNA preparation was stored
at -20 ®C.

Amplification and sequencing of rDNA. —The

mtl 6S rDNA fragment was amplified using PCR
with mtl6S rDNA primers as designed by Cun-
ningham et al. (1992), but with terminal exten-
sions for Sac I restriction endonuclease (16sar:
5'-ATAGAGCTCCCATGGCGCCTGTTTAT
CAAAAACAT-3' and 16sbr: 5'-ATAGAGCT
CCCATGGCCGGTCTGAACTCAGATCACG
T-3'). For the amplification assay 90-700 ng DNA

and 2.5 units of DNA polymerase from Thermus
aquaticus (Taq, Stratagene) were incubated in 100
m1 of PCR buffer (Stratagene) with each of the
four deoxynucleoside triphosphates (50 pM) and
the primers (0.4 pM). For radioactive PCR 2-2.5
MCi alpha-^2p_^AXp ^^s added. The thermal
profile for 40 cycles was as follows: (1) DNA
melting for 1.5 min at 94 °C, (2) annealing for 2
min at 56 °C and (3) polymerization for 2 min
at 72 ®C. The product was electrophoresed on a
5% polyacrylamide gel (bisacrylamid : acrylamid
1:30; 8 M urea). Upon autoradiography of the
gels the band of expected size was excised and a
small piece was used for reamplification in 200
m1 of buffer (concentration as above, without al-
pha-^^p^ATP). DNA was purified by the Gene-
clean Il-kit (Bio 101) procedure according to
the manufacturer’s instruction. The template was
then sequenced by the dideoxy chain termination

196

THE JOURNAL OF ARACHNOLOGY

60

Ryuthela GTTGGTAATA AAAAATCTTA CCTGCTCCCT GCTATAAGTT AATAGCCGCA GTATTATGAC

Dolomedes AGAAA.T.*T ..T.G.AAA. T AA. .AA.ATT-.A -ATA-^ —

Pisaura AGAAA.T..T ..T.G.AAA. T AA. .A..A.T-.C .......... -AT.^ —

Pardosa AGAAT....T ..T...AAAT AA. .A-.A.T-.A -.T. —

C.salel AGAAA....T ..T.G.AAA. T. , . . , .AA. .A.T..T-.A -.TA^

C.getazl AGAAA....T ..T.G.AAA. T......AA. .A.T..T-.A -.T.-^

C,coccineus AGAAA....T ..T.G.AAA. T AA. .A.T..T-.A -.T. —

Phoneutria A-AAA.T.AT ..T.G.AAG. T .AA. .A-.A.T-.A .......... -ACA— ^ — —

Clublona AGATT.T.AT ..T.G.A.A. T AA. .AAT.-T~.A ........ .G A. T.

Psalmopoeus A.GCT.C.C. TTT.G.AAG. . . . . .A.AA. .A..C-T-.A T CAT-^ —

120

Ryuthela TGTGCTAAGG TAGCATAATC ATTTGTCTTT TAAATGAGGT CTGGAATGAA GGGTTTGATC

Dolomedes T.A.A.A .,A...CA.. A. A... A.-.

Pisaura T. . .A.A ..A...CA.. A. A... A.-.

Pardosa A T.A.A.A ..A...CA.. A A.-.

C.salel A T.A.A.A ..A...CA.. A. A... A.-.

C.getazl A T.A.A.A . .A. . . .A. . A. A... A.-.

C . cocclneus A T.A.A.A ..A...CA.. A.A. ..A.-.

Phoneutria C T A .C T.A.A.A .--AA..CA.. A. A... A.-.

Clublona A.... ...C T..,A.A ..A...CA.. A.A..AA.-.

Psalmopoeus C. . .A .A.A.C..A. ...T..TA.G A...C A...C.A.-.

180

Ryuthela GAAGAAAGTC CTGTCTCTTT ATTATTTGGT -GAATTAAAT TAGCTAGTAA AAAGGCTAGT

Dol omedes ATTT ...T.A A.T.T. TAAA TC . . A . . AT . TA TT . . . AA . - AA . ATT .

Pisaura TTTT...T.A A.A. T. TAAA T. . . .AATT. TA TC. ..AA...... . . , .A.ATT.

Pardosa ATCTC. . T. A A. A. T. TA. A .GA.AC.TT. TA TT. ..AT.- AA.A.T.

C.salel ATCTT..T.T A. T... TAAA T. . . .C.AT. TA TC. ..AA.- A.ATT.

C.getazl ATTTT..T.T A.T.T. TAAA T....A.AT. CA TC. ..AA...... ....A.ATT.

C. cocclneus ATCTT..T.T A.T.T. TAAA T....C.AT. TA TC. ..AA...... ....A.ATT.

Phoneutria ATTTT. . .A. A. . .AA-ACA ....CA.AT. TA..C..TT. ...T.TT.C. ...AAAA.C.

Clublona ATTTT.. T.A T.T.A.T..A .A.. A.. AT. AA. . . .TTCC ..AA.-..,. ....A.ATT.

Psalmopoeus ATGA.GCT.T ..T.A.TA.A .AA. .GAAT. GA. . .T.GCA .GAAA- A..TTA

240

Ryuthela ATAGGCCTGA AAGACGATAA GACCCTATTA AGCTTAATTT TTAAAATTTT ACTGGGGCGG

Dolomedes ...TTATA CG .A...TTAC. . . .G — ^..AA .TA.

Pisaura . , .ATTAA G .A...T-AC. ...G--...A TA.

Pardosa T.TAAAAA CG .A...T-AC. ...G— ...A AA.

C.salel ..CTAATA C. .A...T-AC. . . .G — . .AA A.

C.getazl ..CTAATA C. .A...T-AC. ...G— ..AA .... A.

C. cocclneus ...TAATA.. C. .A,..T-AC. . . .G— . .AA ....... .A.

Phoneutria ...CAAAAA C .G .A.... -AC. . . .G — ..CA A.

Clublona .C.ATATA C G .A -AC. A.TG— ..AA A.

Psalmopoeus

Figure 2.— Multiple alignment of the mtl6S rDNA sequences of the nine spider species investigated. Periods
represent nucleotide identity with the reference sequence Ryuthela nishihirai. Dashes indicate positions where
gaps were introduced to obtain maximal alignment. Highly conserved sequences in mtl6S rDNA (as marked
in Fig. 3) are underlined. Psalmopoeus sp., a theraphosid spider is included in this figure but was only partially
sequenced and not taken into further consideration in this paper.

method of Sanger etal. (197 7) using a Sequenase used for the PCR. Sequencing reactions were
kit (U. S. Biochemical) as described in the pro- electrophoresed on 7.5% polyacrylamide gels for
tocol of the manufacturer. Sequencing was per- 2-7 h (Sambrook et al. 1989).
formed in both directions with the primers also Data analysis. —Sequence data were aligned

HUBER ET AL.--DNA SEQUENCE DATA INDICATES CTENID POLYPHYLY

197

300

Ryuthela

TAGGATAAGA

TTATAATCTT

ATCCATAATG

GTTGATATTT

ATTGACCCAA

TTTTATTGAG

Dolomedes

« TAAT . .

, ATTTA- . AT

AAAATCT . AA

TAA T. .

.A. A. . .A.T

Plsaura

* TAAT , .

.ATA.A-.AT

AA.TT. .AAA

T.C. . . .T. .

Pardoma

. TAAT . .

.AT-. . . . .T

AAAT . . . CAA

T .T, .

.AC A.C

C.salei

.TAAT. .

.

. . .

.

.CT. .

. ATT . . T . . T

AAATTC. .AA

-.A. . . .T. .

.A. A. . .A.T

C.getazl

. TAAT . .

. .T. .

. ATT ..T.AT

AAATT.GAAA

T.A. . . .T. .

.A. A. . .A.T

C »cocaineu3

.T.AT. .

. .T. .

GATT ..T.AT

.AATT.CAAA

C.A T. .

.A. A. . .A.T

Phoneutria

.T.AT. .

AA. .

-

. . . . .

. ATT . C . T . T

TAA — C.AAA

TA T. .

.CCA. . .A.T

Clublona

.TAA. . .

.

. .AT

. .T. .

. . TT , CT . . T

AA.T.ATAAA

T .T.T. .

.A A.T

PsalmopoeuB

TTTA.C.

.C. .CGATTA

360

Ryuthela

GGTAAGATAA

AGCTACTATA

GGGATAACAG

CTTAATTTTC

CTTTGAAGAT

CTTATTTATT

Dolomedes

TTC . TA .

.C.

.T.

.CG. .

.G AAAA

T.C.T. ....

Plsaura

TAA.TA.

T . . . TT . . . ,

...... .GAA

Pardosa

AA. .TA.

T. .CT. ....

C.salei

TA . TTA .

T. . .A. . . . .

C.getazl

TACTTA.

T. . .A. . . . .

C .coccineus

TAATTA .

.C.

.T.

.G. .

.G AAAA

T. . .A

Phoneutria

TAA. .A.

A. C.A. . .C.

Clublona

AAA— —

A. . .A

AC.AA

Psalmopoeus

TAAT . C . <

cc.

•

.T.

•

.CGC.

.AC. . .C. .T

T . CAAG . . . C

CC.AA

420

Ryuthela

GGAAAGTTTG

AGACCTCGAT

GTTGAATTAA

AGTACCTTAT

AGGCGCAGTA

GGCTA-TAAA

Dolomedes

AT A

TAA-. . .A. .

TCA A. .

.TAA. . .T. .

Plsaura

ATTT. .A

C

T

TT.-. . .A. .

TTA. . . .A. .

. TT . . . . T . .

Pardosa

AT A

TAA-. . .A. .

TAA A. .

.TTA. . .T. .

C.salei

AT A

C

. . . . .T

TT.T. . .A. .

TAA. . . .A. .

.AAA. . .T. .

C . getazl

AT .... A

C

T. . . .

TT .T. . .A. .

TTT . . . . A. .

. AAA

C .coccineus

AT A

TT.T. . . A. .

TAA A. .

.AGA. . .T. .

Phoneutria

AA.T. .A

TAAT. -.A. .

TT A. .

.CAA. .A. . .

Clublona

TAT . . TA

. .A

T

TAA-. . .ATA

TTAT. . .A. .

ATTAT . A . . .

Psalmopoeus

AA A

. .A

T

-.ATT. .CC.

TAAA. . .AAG

CTTA.GA. . .

446

Ryuthela

GGAAGTCTGT

TCGACTTTTA

AATCTT

Dolomedes

Plsaura

Pardosa

C.salei

C.getazl

C .coccineus

. , .AA.

Phoneutria

.A

. .AAA.

Cl ublona

. .AAA.

Psalmopoeus

.A

by CLUSTAL V (Higgins et al. 1991). Pairwise
alignment and calculation of percent differences
was carried out by MICROGENIE (Queen &
Kom 1983). The data was subjected to DNA-
PARS and DNABOOT of PHYLIP 3.4 (Felsen^
stein 1991). Gaps comprising more than one site
were treated as missing data, and thus played no
role in phylogenetic reconstruction.

Aligned sequences were fitted into available
secondary-structure models from Gutell & Fox
(1988).

RESULTS

Initially we tried to use the conserved primers
of Kocher et al. (1989) to amplify a fragment of
the cytochrome b gene of spiders. These primers

198

THE JOURNAL OF ARACHNOLOGY

Table 2. — Percentage of base identities of the mtl6S rDNA between the investigated spiders. The values are
rounded off.

Ryu.

Phon.

Club.

Pard.

Pisa.

Dolo.

C.coc.

C.get.

Cupiennius salei

66

79

80

87

86

90

91

96

Cupiennius getazi

67

79

80

86

88

90

96

Cupiennius coccineus

65

79

80

87

88

89

Dolomedes

67

81

82

88

90

Pisaura

68

78

81

87

Pardosa

67

79

79

Clubiona

65

77

Phoneutria

64

proved to be unsuitable for our experimental an-
imals, however. In a second attempt we used
primers designed according to those used by
Cunningham et al. (1992) to amplify and study
mtl6S rDNA in crabs. This approach led to the
amplification of some fragments but suffered from
the poor reproducibility of the results. We there-
fore chose to apply a two step amplification pro-
cedure. In a first step, PCR amplification was
performed with simultaneous radioactive label-
ling of the polymerized DNA. The products were
separated on polyacrylamide gels and autoradio-
graphed (Fig. 1). In a second step DNA material
of the major amplification product (about 500
bp, see Fig.l) was gel-extracted and subjected to
a non-radioactive PCR amplification. A unique
distinct band was obtained which was subjected
to dideoxy sequencing after gel-extraction (see
Methods).

Some bands of the expected size were also elut-
ed from the first experiment. The DNA obtained
from them was partially sequenced after ream-
plification and the identity of its main band with
mtl6S rDNA confirmed.

The DNA sequences of the nine spider species
investigated in this study are shown in Fig. 2.
The length of the sequenced fragment varies be-
tween 421 {Clubiona pallidula) and 444 bases
{Ryuthela nishihirai). The percentage of identi-
ties in sequence is 64% or higher in all pairwise
comparisons (Table 2); and a full alignment was
reached, assuming a few small deletions only (Fig.
2). This clearly indicates the homology of the
sequences determined.

In four cases, two individuals of the same spe-
cies were examined {cf. Table 1) and as expected
from previous work on tetragnathid spiders
(Croom et al. 1 99 1 ) no intraspecific variation was
found. Interspecific variation between the three
Cupiennius species is low (3-4%), whereas in-

tergeneric differences vary conspicuously, rang-
ing from 10-36% (Table 2).

The percentage of A and T along the sequenced
DNA fragment is high (75.0-78.6%) in all species
investigated except the “primitive” liphistio-
morph spider Ryuthela nishihirai (66.7%). Similar
results were obtained by Cunningham et al. ( 1 992)
for Crustacea (the “primitive” Artemia salina:
63% AT; the highly evolved king crabs Pagurus
spp.: about 73% AT). In insects data is only avail-
able for highly evolved species such as Drosoph-
ila yakuba (Clary & Wolstenholme 1985) and
Aedes albopictus (HsuChen et al. 1984). Both of
these insects have a very high percentage of AT
(about 76%), too. Possibly, an increase in the
percentage of A and T is a general trend in ar-
thropod phylogeny {cf. Clary & Wolstenholme
1985).

Figure 3 presents an attempt to fit the partial
sequence of Cupiennius salei rDNA into a gen-
erally accepted secondary structure model for an-
imal mtl6S rRNA (Gutell & Fox 1988). The
resulting secondary structure is very similar to
that of the mtl6S rRNA of both Drosophila ya-
kuba and Aedes albopictus which are the only
arthropods studied in this regard. In C. salei gen-
erally well conserved sequences (marked in Fig.
3) take the same positions relative to the overall
secondary structure of mtl 6S rRNA. In addition
these sequences show no variation among the
nine spiders examined {cf Fig. 2). Thus we con-
clude that a fragment of mtl6S rDNA indeed
has been sequenced.

One of the several most parsimonious trees
constructed by DNAPARS is shown in Fig. 4.
When changing the order of the DNA sequences
in the input file the same result was obtained in
most cases. Other trees varied slightly regarding
the relationships between Cupiennius, Pardosa
and the pisaurids. In no case was Phoneutria

HUBER ET AL.--DNA SEQUENCE DATA INDICATES CTENID POLYPHYLY

199

aA-t/ ^actt/

T A~T
AyA

290

A

® " A A

CGTAATA

1 1 • n I •

GCGTTAG

AaA

340

I I I I I I I c
TAAAA^TTatT

360

400

440'

- A

a-t I

"'tcctaa TT^^CGC*

G 1 1 n t I I I • *

T - A
C-G

w

Figure 3.— Mitochondrial DNA of Cupiennius salei folded to show the secondary structure of the sequenced
16S ribosomal RNA for which it codes. Dashes represent Watson-Crick pairings, dots represent the weaker
hydrogen bonds between T and G or A and G. Grey beams indicate the supposed secondary structure of the
adjacent sequences, based on the model for Drosophila yakuba in Gutell & Fox (1988). 16sar, 16sbr (white
beams): priming sites. Black lines indicate extremely conserved regions known to be almost identical in verte-
brates as well as in Escherichia coli. Numbering as in Fig. 2.

interpreted as a sister group of Cupiennius. How-
ever, in a bootstrap analysis with 1 000 replicates
the three Cupiennius species were regarded
monophyletic in 94% of the bootstrap estimates.
This occurred only 19-54% in the other groups.

DISCUSSION

This communication presents the first extend-
ed analysis of spider mtl6S rDNAs for phylo-

genetic studies. It is based on the PCR amplifi-
cation of this gene fragment in nine spider species
representing five families.

The similarity {6A-91% identity) of the 421-
444 bp long sequences in all cases indicates that
homologous sequences have been determined
from all nine species. Furthermore, the derived
RNA sequences fit well into the conserved sec-
ondary structure of other animal mtl6S rRNAs

200

THE JOURNAL OF ARACHNOLOGY

Phoneutria b.

Clubiona p.

Pardosa a.
Dolomedes f.
Pisaura m.

C. getazi
C. coccinem
C. salei
Ryuthela n.

"Ctenidae"

Clubionidae

Lycosidae

I Hsauridae

"Ctenidae"

Heptathelidae

Figure 4.— Mtl6S rDNA gene-tree based on the most
parsimonious tree obtained in most cases when apply-
ing the DNAPARS program of PHYLIP 3.4 (Felsen-
stein 1991). Note the position of Phoneutria in relation
to Cupiennius.

and highly conserved signature sequences of
mtl6S rRNA can be identified unambiguously.
Finally, short sequences of this mtl6S rDNA
fragment and of a previously determined mt 1 2S
rDNA fragment (Groom et al. 1991) are suc-
cessfully used as primers for the PCR amplifi-
cation of large, adjacent parts of both rDNAs (R.
J. Felber, pers. comm.).

Sequence variation is low when the three spe-
cies of Cupiennius are compared (3-4%) whereas
it varies between 10-23% among the Opistho-
thelae studied here (Table 2). This indicates that
sequences of the mtl6S rDNA may be useful for
studies of spider phylogeny at the family and
higher taxonomic levels.

Although limited, the data presented here re-
veal an intriguing result: In all of the maximum
parsimonious gene trees Phoneutria was more
distantly related to Cupiennius than Pardosa and
the two pisaurids. This sheds doubt on the avail-
able classification of both gtnQm— Phoneutria and
Cupiennius— within the Ctenidae and indeed on
the monophyly of this family, which was estab-
lished by Keyserling in 1877. The monophyly of
the included genera has mainly been based on
the following morphological characters: (1) ecri-
bellate spiders with (2) an eye-formula quite pe-
culiar to them among (3) the two-clawed spiders
(Pickard-Cambridge 1897; Bucherl et al. 1964).
These are all characters which also occur in other
families and therefore cannot be considered as
strong synapomorphies. Depending on the po-
sition of the nominal genus Ctenus— which could
not be investigated in this project —either a new
family for Cupiennius {Ctenus closely related to
Phoneutria) or for Phoneutria {Ctenus closely re-
lated to Cupiennius) may well be appropriate.
One obvious next step is therefore the exami-
nation of the nominal species of the Ctenidae.

Clubiona p.

Pardosa a.
Dolomedes f.
Pisaura m.

Phoneutria b.
Cupiennius spp.
Ryuthela n.

Clubionidae

Lycosidae

I Hsauridae

I "Ctenidae"
Heptathelidae

Figure 5. —Current view on phylogenetic relation-
ships of the examined taxa, based on the cladistic hy-
pothesis of Coddington & Levi (1991).

The critical evaluation of this far-reaching im-
plication of our study asks for answers to several
questions. One of these concerns the identifica-
tion of Phoneutria species, which is still a prob-
lem. Many Neotropical taxa are still undescribed
(Coddington «fe Levi 1991) and there is no recent
revision of the genus. The two female individuals
studied by us were identified using the key of
Bucherl (1969). Their genitalia were dissected,
treated with KOH and then compared with il-
lustrations provided by Schiapelli & Gerschman
(1973) and by Valerio (1983). Given the species
specificity of spider genitalia, our morphological
study makes it highly probable that the spiders
in question belong to the species Phoneutria bo-
liviensis. Another question is the possibility of
contaminations. However, DNA was extracted
from both individuals independently. Separate
amplification and up to threefold sequencing of
some segments led to absolutely identical results.
Contaminations are therefore considered a very
unlikely reason for the surprising position of
Phoneutria in the gene tree derived from our
DNA analysis.

Except for the apparent polyphyly of the Cten-
idae, the maximum parsimony tree proposed in
Fig. 4 does not allow any further conclusions on
spider phylogeny. Details such as the relation-
ships between the three Cupiennius species or
between Pisaura and Dolomedes should not be
given too much weight. According to the low
values obtained by bootstrap analysis these re-
lationships are not significant (there is significant
evidence for the monophyly of a group if it occurs
in at least 95% of the bootstrap replicates: Fel-
senstein 1991). From a morphological point of
view Phoneutria belongs to the monophylum of
spiders characterized by at least one pair of sec-
ondary eyes with a grate-type tapetum (Homann
1971). Clubiona (with a canoe-shaped tapetum)
on the other hand does not belong to this mono-
phylum and should therefore branch off deeper

HUBER ET AL.-DNA SEQUENCE DATA INDICATES CTENID POLYPHYLY

201

in the tree (Fig. 5). There is not sufficient data
to resolve this discrepancy, however. We now
rather need a comprehensive examination of ad-
ditional genera, including both molecular and
morphological information.

ACKNOWLEDGMENTS

Financial support by the Austrian Science
Foundation (FWF; P7896 B) to F. G. B. is highly
acknowledged.

LITERATURE CITED

Biicherl, W., S. Lucas & V. Dessimoni. 1964. Ar-
anhas da familia Ctenidae, Subfamilia Ctenidae. 1.
Redescricao dos generos Ctenus Walckenaer 1805
e Phoneutria Perty 1833. Mem. Inst. Butantan, 31:

95-102.

Biicherl, W. 1969. Aranhas da familia Ctenidae. II
Phoneutriinae subfamilia nova. Mem. Inst. Butan-
tan, 34:25-31.

Clary, D. O. & D. R. Wolstenholme. 1985. The mi-
tochondrial DNA molecule of Drosophila yakuba:
Nucleotide sequence, gene organization, and genetic
code. J. Mol. Evol., 22:252-271.

Coddington, J. A. & H. W. Levi. 1991. Systematics
and evolution of spiders (Araneae). Annu. Rev. Ecol.
Syst., 22:565-592.

Croom, H. B., R. G. Gillespie & S. R. Palumbi. 1991.
Mitochondrial DNA sequences coding for a portion
of the RNA of the small ribosomal subunits of Te-
tragnatha mandibulata and Tetragnatha hawaiensis
(Araneae, Tetragnathidae). J. Arachnol., 19:21 0-2 1 4.
Cunningham, C. W., N. W. Blackstone & L. W. Buss.
1992. Evolution of king crabs from hermit crab
ancestors. Nature, 355:539-542,

Felsenstein, J. 1991. Manual to PHYLIP (Phytogeny
Inference Package) Version 3.4.

Femholm, B., K. Bremer & H. Jornvall. 1989. The
hierarchy of life. Molecules and morphology in phy-
logenetic analysis. Pp. 1-499. Excerpta Medica,
Amsterdam - New York - Oxford.

Gatesy, J., D. Yelon, R. DeSalle & E. S. Vrba. 1992.
Phytogeny of the Bovidae (Artiodactyla, Mamma-
lia), based on mitochondrial ribosomal DNA se-
quences. Mol. Biol. Evol, 9:433-446.

Gutell, R. R. & G. E. Fox. 1988. A compilation of
large subunit RNA sequences presented in a struc-
tural format. Nucleic Acids Res., 16 :R175-R313.
Higgins, D. G., A. J. Bleasby & R. Fuchs. 1992.
CLUSTAL V: improved software for multiple se-
quence alignment. CABIOS, 8:189-191.

Homann, H. 1971. Die Augen der Araneae. Anato-
mie Ontogenie und Bedeutung flir die Systematik
(Chelicerata, Arachnida). Z. Morph. Tiere, 69:201-
272.

HsuChen, C. C., R. M. Koten & D. T. Dubin. 1984.
Sequences of the coding and flanking regions of the

large ribosomal subunit RNA gene of the mosquito
mitochondria. Nucleic Acids Res., 12: 7771-7785.

Keyserling, E. 1877. fiber amerikanische Spinnen-
arten der Unterordnung Citigradae, Verb, zool.-bot.

Ges. Wien, 26:609-708.

Kocher, T. D., W. K. Thomas, A. Meyer, S. V. Ed-
wards, S. Paabo, F. X. Villablanca & A. C. Wilson.
1989. Dynamics of mitochondrial DNA evolution
in animals: Amplification and sequencing with con-
served primers. Proc. Natl. Acad. Sci. USA, 86:6 1 96-
6200.

Lachmuth, U,, M. GrasshofF & F. G. Barth. 1985.
Taxonomische Revision der Gattung Cupiennius
Simon 1891. Senck. Biol, 65:329-372.

Lehtinen, P. T. 1967. Classification of the cribellate
spiders and some allied families, with notes on the
evolution of the suborder Aranaeomorpha. Ann.
Zool. Fennici, 4:199-468.

Mullis, K. B. & F. A. Faloona. 1987. Specific syn-
thesis of DNA in vitro via a polymerase catalyzed
chain reaction. Methods in Enzymology, 155:335-
350.

Pickard-Cambridge, F. 1897-1905. Arachnida, Ar-
aneida and Opiliones. Biologia Centrali-Americana,
2:1-610. London.

Queen, C. & L. Korn. 1983. Manual to MICRO-
GENIE (TM) sequence analysis program.

Saiki, R. K., D. H. Gelfand, S. Stoffel, S. J. Scharf, R.
Higuchi, G. T. Horn, K. B. Mullis, & H. A. Erlich.
1988. Primer-directed enzymatic amplification of
DNA with a thermostable DNA polymerase. Sci-
ence, 239:487-491.

Sambrook, J., T. Fritsch & T. Maniatis. 1989. Poly-
acrylamide gel electrophoresis. Pp. 6.36-6.48, In
Molecular Cloning, 2nd ed., vol. 1 . Cold Spring Har-
bor Laboratory Press, Cold Spring Harbor, New
York.

Sanger, F., S. Nicklen & A. R. Coulson. 1977. DNA
sequencing with chain-terminating inhibitors. Proc.
Natl. Acad. Sci. USA, 74:5463-5467.

Schiapelli, R. D. & B. S. Gerschman de Pikelin. 1 972.
Diagnosis de Phoneutria reidyi (F. O. Pickard-Cam-
bridge, 1 897) y de Phoneutria boliviensis (F. O. Pick-
ard-Cambridge, 1897) (Araneae, Ctenidae). Rev. Soc.
Ent. Argentina, 34:31-38.

Valerio, C. E. 1983. Sobre la presencia de Phoneutria
boliviensis (F. O. Pickard-Cambridge) (Araneae,
Ctenidae) en Costa Rica. J. Arachnol., 1 1:101-102.

Wilson, A. C., R. L. Cann, S. M. Carr, M. George, U.
B. Gyllensten, K. M. Helm-Bychowski, R. G. Hig-
uchi, S. R. Palumbi, E. M. Prager, R. D. Sage & M.
Stoneking. 1985, Mitochondrial DNA and two
perspectives on evolutionary genetics. Biol. J. Linn.
Soc., 26:375-400.

Manuscript received 11 January 1993, revised 2 June
1993.

1993. The Journal of Arachnology 21:202--204

TAXONOMIC NOTES ON THE GENUS ARCHITIS
(ARANEAE, PISAURIDAE) AND STATUS OF THE
GENUS SISENNA SIMON

James E. Carico: Biology Department, Lynchburg College, Lynchburg, Virginia
24501 USA

ABSTRACT. Dyrines tenuipes (Simon) is a senior synonym of Architis vilhena Carico. Thanatidius proximatus
Mello-Leitao {=Thanatidius parahybensis Mello-Leitao) is a junior synonym of Architis tenuis Simon. Sisenna
Simon is a junior synonym of the genus Architis. Sisenna helveola Simon, the only species placed in Sisenna,

is redescribed and figured as Architis helveola (Simon),
a previous publication are noted.

During a survey of types of some lesser known
pisaurid genera in the New World, three mis-
placed species were found to belong to the pi-
saurid genus Architis. This paper reports on the
taxonomic status of these species.

Dyrines tenuipes (Simon).— Dyrines tenuipes
(Simon 1898*^:18) (transferred from Drances by
Petrunkevith 191 1:543) is a senior synonym of
Architis vilhena Carico (1981:150, figs. 1,10, 22,
23) based on the examination of six male syn-
types, #5408, deposited in the Museum National
d’Histoire Naturelle, Paris, collected from Ca-
meta, Para, Brazil by Mathan. The female was
described by Carico (1989:224, figs. 6, 7). The
correct name of this species is, therefore, Architis
tenuipes (Simon). NEW SYNONYMY.

Thanatidius proximatus Mello-Leitaao.—The
holotype of Thanatidius parahybensis Mello-
Leitao (1924) from Campino Grande, Paraiba,
Brazil, collected by Tranquilino Mello-Leitao
(specimen #88, labelled Pisaurina parahybensis
Mello-Leitaao), in the Museu Nacional do Rio
de Janeiro, is a penultimate female collected just
prior to ecdysis. The adult epigynum is clearly
visible through the old cuticle, and close exam-
ination reveals the characteristic features of Ar-
chitis tenuis Simon.

Mello-Leitao described this specimen first
(1920) as T proximatus and later (1924) as T.
parahybensis. Roewer (1954) first reported the
objective synonymy. Bonnet (1959) lists only T.
parahybensis. Both are junior synonyms of Ar-
chitis tenuis Simon. NEW SYNONYMY.

Errata in previous publication,™ In my revi-
sion of the genus Architis (Carico 1981), labels
for epigyna of two species were interchanged.

lectotype of the latter species is designated. Errata in

Figures 26, 27 should be labelled^, nitidopilosa,
and figs. 32, 33 should be labelled A. tenuis as
was reported by Sierwald (1989). Additionally,
fig. 6 should be labelled A. cymatilis.

Sisenna helveola.— type of the monotypic
genus, Sisenna, S. helveola Simon, is congeneric
with the genus Architis. Therefore, the genus Sis-
enna is a junior synonym (NEW SYNONYMY)
of the pisaurid genus Architis. This conclusion
differs from that of Sierwald (1990) who placed
Sisenna tentatively in her ''Trechalea genus-
group,” which I regard as the family Trechalei-
dae.

Below is the redescription of A. helveola, which
becomes the tenth known species in the Neo-
tropical genus Architis, four of which were de-
scribed by Simon. It is interesting to note that
Simon originally placed his four species into three
different pisaurid genera which he also described:
Architis (two species), Drances (one species) and
Sisenna (one species). The male genitalia of all
Architis species are quite similar while the eyes
and body shape show a greater range of vari-
ability. Apparently Simon placed more emphasis
on these non-genitalic characters in his generic
definitions.

Architis helveola (Simon)

Figures 1-4

Sisenna helveola Simon, 1 898'’: 1 2 (n. sp.). Simon, 1898^:

292, figs. 290, 291 (n. gen.).

Sisenna helvola, Roewer, 1954:123. Bonnet, 1955-

1959:4065.

Type, —Male syntype from Sao Paulo Oliven-
ga, Amazonas, Brazil, designated herein as lec-
totype. Female paralectotype from same locality.

202

CARICO-TAXONOMIC NOTES ON ARCHITIS

203

Figures 1-4.— Genitalia of Architis helveola: 1, 2, right male palpus; 1, ventral view; 2, retrolateral view; 3,

4, epigynum; 3, ventral view; 4, dorsal view. Scales are in mm.

This part of the Mathan Collection identified as
#6895 in the Museum National d’Histoire Na-
turelle, Paris. Examined.

Diagnosis.— This species resembles Architis
tenuis and A. nitidopilosa in the AE row config-
uration (see Carico 1981, fig. 9). It differs from
the latter two species in details of the genitalia,
particularly the tibial apophysis of the male and
the ventral view of the female epigynum, and by
the lack of short, stout spines on the ventral sur-
face of coxae I and II.

Description.— (Measurements in mm) Male
lectotype: Carapace low, no pattern on cream
ground color; each eye individually ringed in
black; length 2.9, width 2.4. Sternum length 1 .45,
width 1.50, unmarked; labium length 0.46, width
0.42, unmarked, rounded anteriorly. Eye mea-
surements: AE row 0.83, PE row 0.80; ocular
quadrangle height 0.42, width posterior 0.40,
width anterior 0.34; diameters PLE 0.16, PME
0.16, ALE 0.12, AME 0.14; interdistances PLE-
PME 0.18, PME-PME 0.12, ALE- AME 0.24,
AME-AME 0.08. Clypeus height 0.05 (ALE) or
0.22 (AME), width 1.02. Legs WI-IV-III, un-
marked with numerous long, dark macrosetae.
Measurements are given in Table 1.

Abdomen narrow, mostly covered with irreg-
ular white pattern on cream ground color except

for midventral area; patch of dark hairs on an-
terodorsal margin, length 5.1. Palpus (Figs. 1, 2)
with two tibial apophyses: ventral one curved
and pointed apically, retrolateral one ffat, curved
towards cymbium with small point at base ven-
trally.

Female paralectotype: Carapace color and pat-
tern as in male; length 3.0, width 2.5. Sternum

Table 1. — Leg measurements (in mm) for the male
lectotype and the female paralectotype of Architis hel-
veola (Simon).

I

II

Ill

IV

Male

Femur

5.45

5.2

3.5

4.4

Tibia-patella

lA

7.0

4.25

4.65

Metatarsus

6.6

6.1

3.5

4.45

Tarsus

2.1

1.95

1.1

1.6

Total

21.55

20.25

12.6

15.1

Female

Femur

4.9

3.4

4.1

Tibia-patella

6.75

4.0

4.55

Metatarsus

6.0

—

3.55

4.1

Tarsus

1.85

-

1.35

1.6

Total

19.5

-

12.3

14.35

204

THE JOURNAL OF ARACHNOLOGY

and labium as in male. Abdomen pattern and
hairs as in male, length 5,35. Sternum length
1.60, width 1.55, unmarked; labium length 0.55,
width 0.50, unmarked, rounded anteriorly. Eye
measurements: AE row 0.92, PE row 0.86; ocular
quadrangle height 0.40, width posterior 0.45,
width anterior 0.36; diameters PLE 0.17, PME
0. 1 7, ALE 0. 1 2, AME 0. 1 3; interdistances PLE-
PME 0.22, PME-PME 0.16, ALE- AME 0.33,
AME- AME 0.11. Legs (II missing), color and
macrosetae as in male, and the measurements
are in Table 1.

Epigynum with openings medially, each under
longitudinal ridge (Figs. 3, 4), pale and soft ex-
ternally.

Natural history.— Unknown.

Distribution and material examined.— These
male and female are the only known specimens.

Notes. — Bonnet (1955-1959:77) changed the
spelling of this species as well as other “helveola”
species of Simon to “helvola”, because the latter
is a more frequently used form. The “helveola”
spelling does indeed seem to be a rare form (H.
D. Cameron pers. comm.), but the spelling change
is not justified according to the current Inter-
national Code of Zoological Nomenclature.

ACKNOWLEDGMENTS

Help in obtaining specimens was through the
efforts of H. W. Levi and C. Rollard. Advice on
Latin was provided by H. D. Cameron. Appre-
ciation is extended to R. G. Bennett, N. L Plat-
nick and P. Sierwald for their reviews of the
manuscript.

LITERATURE CITED

Bonnet, P. 1955-1959, Bibliographia Araneorum.

Vol. 2, Toulouse, 5058 pp.

Carico, J. E. 1981. The Neotropical spider genera
Architis and Staberius (Pisauridae). Bull. American
Mus. Natur. Hist., 170(1):140-153.

Carico, J. E. 1986. Trechaleidae: A “new” American
spider family (Abstract). P. 305, In Proceedings of
the 9th International Congress of Arachnology. (W.
G. Eberhard, Y. D. Lubin & B. C. Robinson, eds.).
Smithsonian Inst. Press, 333 pp.

Carico, J. E. 1989. Descriptions of two new species
of the genus Architis (Araneae, Pisauridae) and the
female of vilhena. J. ArachnoL, 17:221-224.

Mello-Leitao, C. 1920. Um genero e quatro especies
novas de aranhas do Brasil. Rev. Sci. Soc. Brasileira
Sci., 4:179,180.

Mello-Leitao, C. 1924. Algumas aranhas novas do
Brasil. Bolet. Mus. Nac. Rio de Janeiro, 1(4):275-
281.

Petrunkevitch, A, 1911. An index-catalogue of spi-
ders. Bull. American Mus. Natur. Hist., 29:1-790.

Roewer, C. F. 1954. Katalog der Araneae. 2a, 923

pp.

Sierwald, P. 1989. Morphology and ontogeny of fe-
male copulatory organs in American Pisauridae, with
special reference to homologous features (Arachni-
da: Araneae). Smithsonian Contrib. Zool. no. 484,
24 pp.

Sierwald, P. 1 990. Morphology and homologous fea-
tures in the male palpal organ in Pisauridae and
other spider families, with notes on the taxonomy
of Pisauridae (Arachnida: Araneae). Occ. Pap. Del-
aware Mus. Natur. Hist., 35:1-59.

Simon, E. 1898^ Histoire Naturelle des Araignees.
Tome 2, Fascicule 2, Paris, Pp. 193-380.

Simon, E. 1898^ Descriptions d’Arachnides nou-
veaux des families des Agelenidae, Pisauridae, Ly-
cosidae et Oxyopidae. Ann. Soc. Entomol. Beige,
42:5-34.

Manuscript received 23 April 1 992, revised 7 May 1 993.

1993. The Journal of Arachnology 21:205-208

TWO NEW SPECIES OF THE GENUS LYSSOMANES (HENTZ)
FROM THE CAPE REGION, B.C.S., MEXICO

Maria Luisa Jimenez and Armando Tejas: Division de Biologia Terrestre, Centro de
Investigaciones Biologicas de Baja California. Sur, Apdo. Postal 128, La Paz, B.C.S.,
23000 Mexico.

ABSTRACT. Two new species of the genus Lyssomanes Hentz from the Cape Region, Baja California Sur,
are described and illustrated. Lyssomanes burrera n. sp. is similar to Lyssomanes jemineus Peckham, Peckham
& Wheeler, and Lyssomanes pescadem n. sp. is similar to Lyssomanes mandibulatus F. O. P. Cambridge.

Sixty-one species of spiders from the Americas
are included in the genus Lyssomanes Hentz (Ga-
liano 1980, 1984) and ten of these species are
present in Mexico: L, jemineus Peckham, Peck-
ham & Wheeler 1889, L. temperatus Galiano
1 980, L. diversus Galiano 1980, L. malinche Ga-
liano 1980, L. placidus Peckham, Peckham &
Wheeler 1889, L. (Taczanowski 1872),

L. deinognathus Cambridge 1900, L. mandibu-
latus F. O. P.-Cambridge 1900, L. elegans F. O.
Pickard-Cambridge 1901 and L. spiralis F. O.
Pickard-Cambridge 1901 (Richman & Cutler
1988).

The genus Lyssomanes is primarily tropical
with a high concentration of species in Brazil,
but it has not been well studied in Mexico; and
the possibility exists that there are many new
species there (Galiano 1980). Mexico, from a
biogeographical point of view, is considered to
be as a transitional zone occupied by elements
of a hybrid fauna of both Neotropical and Ne-
arctic origin. There are strong endemic charac-
teristics, but there are also affinities with the fau-
nas of both South America and temperate North
America (Halffter 1976).

This is the first record of the genus Lyssomanes
from the peninsula of Baja California. We de-
scribe two new species: Lyssomanes burrera n.
sp. is included in the jemineus group because of
fringes of black setae on the tibia and tarsus and
the long and divergent chelicerae of the males.
The bulb of the palp has a cylindrical embolus
ending in a sharp tip and its base has an apoph-
ysis. Lyssomanes pescadero n. sp. is included in
the viridis group because it has very long and
divergent chelicerae. The bulb of the palp is rel-
atively simple, with its basal part of the embolus

covered partial or wholly by a membranous
sheath.

Lyssomanes burrera, new species
(Figs. 1-8)

Type. —Male holotype from low deciduous
forest, 600 m elev.. Canon de la Burrera, Sierra
de la Laguna, Baja California Sur (25 June 1 992),
A. Tejas and G. Navarrete. Two male and one
female paratypes are from the type locality (25
June 1992, A. Tejas and M. Jimenez). One fe-
male paratype is from low deciduous forest, 753
m elev., Canon de la Zorra, Sierra de La Laguna,
(29 October 1987, M. Jimenez). The holotype
and a female paratype will be deposited in the
collection of the Instituto de Biologia, Univer-
sidad Autonoma de Mexico, and three paratypes
will be deposited in the arachnological collection
of the Centro de Investigaciones Biologicas de
Baja California Sur, A.C.

Etymology. —The specific name is derived from
the type locality.

Diagnosis.— Members of Lyssomanes burrera
n. sp. resemble L. jemineus Peckham, Peckham,
& Wheeler in coloration and body shape, but can
be separated from the other known similar spe-
cies by the shape of the embolus, the number of
teeth of the female chelicerae, leg spination, and
the shape of the epigynum.

Males.— (a2 = 2). Total length 5.58-8.33 mm;
prosoma 2 . 5 5-2 . 7 0 mm long, 1 . 9 5-2 .15mm wide
and 1.25-1.75 mm high (Fig. 1, 2). Clypeus0.20
high. Eye sizes and interdistances: AME 0.55,
ALE 0.26-0.29, PLE 0.23-0.26, ALE-PME 0. 1 8-
0.20, PME-PLE 0.34-0.36, ALE-PLE 0.52-0.58.
Eye row widths: first 1.08-1.14, second 1.00-
1.06, third 0.60-0.64, fourth 0.82-0.84. Chelic-

205

206

THE JOURNAL OF ARACHNOLOGY

Figures X-^.—Lyssomanes burrera n. sp. holotype male from Cape Region of Mexico: 1. dorsal view of the
carapace; 2. lateral view of the carapace; 3. right chelicera; 4. ventral view of palp; 5. retrolateral view of palp;
6. prolateral view of palp; 7. ventral view of epigynum; 8. female dorsal view of epigynum.

erae moderately long and divergent from the base.
Anterior surface with 2-3 median spines and 4-
7 dorsal spines. Promargin with four teeth, re-
tromargin with six teeth and a small apophysis
near the the fang (Fig. 3). Leg spination: Femora

1- III d 1-1-1, p 0-1-1, r 0-1-1; IV d 1-1-1, p
0-0- 1 , r 0- 1 - 1 ; Patella I-I V d 0-0- 1 ; Tibia I p 0- 1 -
1, r 0-1-1, V 2-2-2; II d 0-1-1, p 0-1-1, r 0-1-1,
V 2-2-2; III d 1-0-1, p 0-1-1, r 0-1-1, v 0-2-2; IV
d 1-0-1, p 0-1-1, r 0-1-1, V 0-1-0; Metatarsi I-II
p 0-0-1, r 0-0-1, V 2-2-2; III p 0-0-1, r 0-0-1, v

2- 2- 1 ; Palpi: Femur 1 . 1 4- 1 . 8 5 , d 0- 1 - 1 , p 0-0- 1 ,
r 0-0- 1 ; Patella 0.40-0.66, d 0-0- 1 , r 0-0- 1 ; Tibia
0.56-0.64, d 0-0-8, p 0- 1 - 1 ; Cymbium 1.50-1.76
(Figs. 4, 6).

Color in alcohol: Carapace light yellow with a
longitudinal dark band. Ocular quadrangle light
with the ALE, PME and PLE in black tubercles
and rounded with white setae, clypeus yellow
with lower edge orange red, setae on each side;
anterior surface of the chelicerae dark yellow,
inner side darker, sternum light yellow. Opis-
thosoma dorsally yellow with two green, longi-
tudinal dark brown bands; venter yellow, spin-
nerets darker. Legs yellow without marks and
stripes. First metatarsus curved and flattened on
the sides and a little dilated dorsoventrally,

blackish with iridescent shine and fringes of black
setae above and below. Tibia I with fringe of
black setae distally above and below. Palps yel-
low with distal part of the tibia orange yellow.

Living specimens are green in color, with white
and orange-red ocular setae.

Female.— (/t = 3). Total length 6.50-8.00 mm.
Carapace 2.50-2.60 long, 1.80-2.15 mm wide,
1.50-1.75 high. Clypeus 0.25. Eye sizes and in-
terdistances: AME 0.50-0.55, ALE 0.26-0.28,
PLE 0. 1 4-0.23, ALE-PME 0. 1 6-0.29, PME-PLE

0. 34-0.59, ALE-PLE 0.42-0.95. Eye row widths:
first 1.08-1.10, second 1.00, third 0.60-0.64,
fourth 0.82. Leg spination: Femora I-III d 1-1-

1, p 0-1-1, r 0-1-1; IV d 1-1-1, r 0-0-1; Patella
II-IV d 0-0-1; Tibia I p 0-1-1, r 0-1-1, v 2-2-2;
II d 0-1-1, p 0-1-1, r 0-1-1, V 2-2-2; III d 0-1-1,
p 0-1-1, r 0-1-1, V 0-2-2; IV d 1-0-1, p 1-0-1, r
0-1-1, V 0-0-1; Metatarsi I-II p 0-0-1, r 0-0-1, v
2-2-2; III p 1-1-1, r 0-0-1, v 1-1-1; IV p 0-M,
r 0-1-1, V 0-1-1. Epigynum (Fig.7, 8).

Color in alcohol: Light yellow, ocular quad-
rangle with lateral white and dorsal red setae,
and long whitish setae around AME and ALE.
Clypeus with iridescent setae and red orange se-
tae on the sides. Chelicerae with three promar-
ginal and five retrolateral teeth. Opisthosoma long

JIMENEZ AND TEJAS-NEW SPECIES OF LYSSOMANES

207

Figures 9-1 5. —Lyssomanes pescadero n. sp. holotype male from Cape Region of Mexico: 9. dorsal view of
the carapace; 10. lateral view of the carapace; 1 1. right chelicera; 12. ventral view of palp; 13. retrolateral view
of palp; 14. dorsal view of female epigynum; 15. ventral view of epigynum.

and slender, with two longitudinal marks as in
the males.

Habitat: Specimens were collected under leaves
of mango and underside of wide leaves of ripar-
ian shrubs.

Range. —Known only from the type locality.

Lyssomanes pescadero, new species
(Figs. 9-15)

Types.— Male holotype from Rancho San Si-
mon, Pescadero, Baja California Sur. (9 June
1992; A. Tejas, M. Jimenez and F. Cota). Eleven
paratypes are from the same type locality. The
holotype and a female paratype will be deposited
in the collection of Instituto de Biologia, Univ-
ersidad Autonoma de Mexico, and ten paratypes
which will be deposited in the arachnological col-
lection of the Centro de Investigaciones Biolo-
gicas de Baja California Sur.

Etymology. —The specific name is derived from
the type locality.

Diagnosis. —Male specimens of Lyssomanes
pescadero n. sp. resemble L. mandibulatus F. O.
Pickard-Cambridge in shape and coloration, but
can be separated from the other known similar
species by the shapes of the bulb in lateral view,
the median apophysis, and the embolus and by
the number of cheliceral teeth and leg spination.

Males.— (« = 6). Total length 6.07-8.52 mm
prosoma 2.94-3.62 mm long, 2.35-3. 1 3 mm wide
and 1.50-2.00 high (Figs. 9, 10). Clypeus 0.20-
0.25 high. Eye size and interdistances: AME 0.50-
0.60, ALE 0.23-0.26, PLE 0. 1 9-0.23, ALE-PME

0.14-0.20, PME-PLE 0.28-0.36, ALE-PLE 0.44-

0. 60. Eye row widths: first 1.04-1.22, second
1.04-1.18, third 0.64-0.76, fourth 0.78-0.96.
Chelicerae strong, very long and divergent from
the base. Anterior surface with three basal spines
and 19 distal spines. Promargin with one distal
apical tooth and three small teeth and retro-
margin with five teeth, the apical tooth bifur-
cated and a small apophysis near the base of the
fang (Fig. 1 1). Leg spination: Femora I-III d 1-1-

1, p 0-1-1, r 0-1-1; IV d 1-1-1, p 0-1-1, r 0-0-1;
Patella II-IV d 0-1-1; Tibia I p 0-1-1, r 0-0-1, v
2-2-2; II d 0-1-1, p 0-1-1, r 0-1-1, v 2-2-2; III d
0-1-1, p 0-1-1, r 0-1-1, V 0-2-0; IV d 0-1-1, p
0- 1 - 1 , r 0- 1 - 1 ; Metatarsi I p 0-0- 1 , r 0-0- 1 , v 2-2-
2; II p 0-0-1, r 0-0-1, v 2-2-2; III p 0-0-1, r 0-0-
1, V 2-2-1; IV p 1-1-1, r 0-1-1, v 0-0-1; Palps:
Femur 1.62-2.08, d 0-1-1, p 0-0-1, r 0-0-1; Pa-
tella 0.56-0.76, d 0-0-1; Tibia 0.69-0.89, d 0-0-
1, p 0-1-1. Cymbium 1.32-1.55 (Figs. 12,13).

Color in alcohol: Carapace robust and sides
prominent at the ocular area, dark yellow, with
a longitudinal dark band. The carapace is bor-
dered with a black line in darker specimens and
with white setae around the eyes; ALE, PME and
PLE on black tubercles, ocular quadrangle with
white dorsal and lateral setae and an inner curved
band of orange-red hairs; clypeus yellow with
orange-red hairs on each side and a dark line
under the median anterior eyes, border darker;
anterior surface of the chelicerae darker yellow,
fangs dark yellow, sternum and mouthparts shiny
yellow. Opisthosoma light yellow with two dor-

208

THE JOURNAL OF ARACHNOLOGY

sal longitudinal dark bands broken at the middle
and two pairs of median black spots. In light
specimens there are three pairs of black spots
only. Spinnerets darker, venter yellow without
marks. Legs yellow with gray lateral bands. First
metatarsus and tibia with small fringes of black
setae above and below. Palps yellow with the
cymbium darker. Living specimens are green or
yellow with shining orange red setae in the ocular
area.

Female.— (« = 6). Total length 5.58“9.01 mm.
Carapace 2.94-3.62 long, 2.35-3.13 wide and
1.30-1.75 high. Clypeus 0.2-0.25 high. Eye size
and interdistances: AME 0.55, ALE 0.23-0.26,
PLE 0. 1 9-0.23; ALE-PME 0.20-0. 1 4, PME-PLE
0.30-0.34, ALE-PLE 0.48-0.56. Eye row widths:
first 0.96-1 . 1 8, second 0.98-1 . 1 2, third 0.68-0.74
and fourth 0.76-0.90. Leg spination: Femora I-III
d Ul-1, p OG-1, r 0-1-1; IV d 1-1-1, r 0-0-1;
Patella II-IV d 0-0-1; Tibia I p 0-1-1, r 0-1-1, v
2-2-2; II d 0-1-1, p 0-1-1, r 0-1-1, v 2-2-2; III d
O-Ul, p 0-1-1, r 0-1-1, V 0-0-2; IV d 0-1-1, p
0-1-1, r 0-1-1; Metatarsus I-III p 0-0-1, r 0-0-1,
V 2-2-2; IV p 1-1-1, r 0-1-1, v 0-0-1. Epigynum
(Fig. 14, 15 ).

Color in alcohol: Light yellow with orange-red
setae between the eyes, dorsal and lateral sides
of the ocular area with white setae, eyes sur-
rounded by white setae. Clypeus with a band of
white setae and red setae on each side. Cheliceral
promargin with three teeth and retromargin with
five teeth and two dorsal inner spines. Opistho-
soma white with three pairs of green spots. Legs
yellow, the metatarsi and tarsi darker. Living

specimens are green or yellow, the ocular quad-
rangle with yellow setae.

Range. —Known from the type locality and
from Canon de la Burrera, Sierra de La Laguna,
B. C. S.

ACKNOWLEDGMENTS

We are very grateful to Dr. B. Cutler of the
University of Kansas and Dr. D. Richman of
New Mexico State University for their valuable
comments on the manuscript, and to Mr. Franco
Cota and Mr. Gabriel Navarrete for help in the
collection of the specimens. This work was sup-
ported by grants from the Centro de Investiga-
ciones Biologicas de Baja California Sur, A. C.,
the Consejo Nacional de Ciencia y Tecnologia
(CONACyT), and the Secretaiia de Programa-
cion y Presupuesto (SPP).

LITERATURE CITED

Galiano, M. E. 1980. Revisi6n del genero Lysso-
manes Hentz, 1845 (Araneae, Salticidae). Op. lil-
loana 30:1--104.

Galiano, M. E. 1984. New species of Lyssomanes
Hentz, 1845 (Araneae, Salticidae). Bull. British Ar-
achnol. Soc. 6:268-276.

Halffter, G. 1976. Distribucidn de los insectos en la
zona de transicidn Mexicana. Relaciones con la En-
tomofauna de Norteamerica. Fol. Entomol. Mexi-
cana 35:1-65.

Richman, D. B. & B. Cutler. 1 988. A list of the jump-
ing spiders of Mexico. Peckhamia 2:63-90.

Manuscript received 27 October 1992, revised 18 June
1993.

1993. The Journal of Arachnology 21:209-225

THE ORB- WEAVER GENUS KAIRA
(ARANEAE: ARANEIDAE)

Herbert W. Levi: Museum of Comparative Zoology, Harvard University; Cambridge,
Massachusetts 02138 USA

ABSTRACT. Adult specimens of Kaira are rarely collected and the females are difficult to separate. The few
specimens in collections represent 14 species, all American. Five species are new: Kaira cobimcha from southern
Brazil; K. dianae from southeastern Peru; K. erwini from Peru; K. shinguito from northern Peru; K. tulua from
Depto. Valle, Colombia. The female of K. hiteae is described and Araneus sexta is transferred to Kaira.

Haliger is a new synonym of Kaira, with H. corniferus a synonym of K. altiventer. Kaira obtusa and Wagneriana
minutissima are synonyms of K. gibberosa. Doubtful synonyms are Caira capra of K. altiventer, and Macpos

monstrosus of K. gibberosa.

In the past many authors named spiders with-
out adequately illustrating them, and without
comparison to other species in the same genus.
The current approach is to name new species
only as one aspect of a comprehensive revision
of the entire genus, including the examination
and comparison of old holotypes (vouchers for
species names), and adequate illustrations of both
sexes. While such revisions are required to make
it possible to determine spiders, only a few re-
visions of Neotropical spiders are available.

Kaira specimens are uncommon in collections.
According to Stowe (1986), the spiders spin small
webs, hanging upside down below the web and
attracting male moths that fly into a basket formed
by their legs (see below). The attractant, appar-
ently a moth pheromone, resembles that of the
Bolas spiders Mastophora. Mastophora and Kaira
both belong to the same orb-weaver family, Ar-
aneidae, but are not closely related within the
family.

METHODS

This revision is one of a series for American
orb weavers (Levi 1993). The procedures used
are similar to those described in previous revi-
sions (Levi 1993).

Eye measurements, as in previous papers, are
expressed as ratios of the diameter (with cornea
in profile) to those of the anterior median eyes
(Levi 1993, figs. 27, 28). Distances between eyes
of the anterior row are expressed as diameters of
the anterior median eyes (in profile); distances
between eyes of the posterior row are given as
diameters of the posterior median eyes (in pro-
file). The height of the clypeus, the distance be-

tween anterior median eyes and the edge of the
carapace, is given in diameters of an anterior
median eye and is measured below the eye (Levi
1993, fig. 28f). These measurements are approx-
imate.

The maximum length of the abdomen was
measured. In this revision “humps” refers to
paired protrusions on the abdomen, and “tuber-
cles” refers to small projections on the abdomen.

The collections used for this study came from
the following institutions: (AMNH)— American
Museum of Natural History, New York, United
States; N. Platnick, L. Sorkin. (BMNH)^The
Natural History Museum, London, England; P.
Hillyard, F. Wanless. (CAS)— California Acad-
emy of Sciences, San Francisco, United States;
W. J. Pulawski, D. Ubick. (MACN)-Museo Ar-
gentino de Ciencias Naturales, Buenos Aires, Ar-
gentina; E. A. Maury. (MCN)— Museu de Cien-
cias Naturais, Fundagao Zoobot^nica do Rio
Grande do Sul, Porto Alegre, Rio Grande do Sul,
Brazil; E. H. Buckup. (MCP)— Museu de Cien-
cias, Pontificia Universidade Catolica, Porto
Alegre, Rio Grande do Sul, Brazil; A. A. Lise.
(MCZ) — Museum of Comparative Zoology,
Cambridge, Massachusetts, United States.
(MLP)— Museo de Universidad Nacional, La
Plata, Argentina; R. F. Arrozpide. (MNHN)—
Museum National d’Histoire Naturelle, Paris,
France; J. Heurtault, J, Kovoor, C. Rollard.
(MNRJ)— Museu Nacional, Rio de Janeiro, Bra-
zil; A. Timotheo da Costa. (MUSM)— Museo de
Historia Natural, Universidad Nacional Mayor
de San Marcos, Lima, Peru; D. Silva D. (MZSP)—
Museu de Zoologia, Universidade de Sao Paulo,
Sao Paulo, Brazil; P. Vanzolini, L. Neme, J. L.

209

210

THE JOURNAL OF ARACHNOLOGY

M. Leme. (USNM)— National Museum of Nat-
ural History, Smithsonian Institution, Washing-
ton, D.C., United States; J. Coddington. I thank
the curators of these collections for loaning spec-
imens. The revision of Kaira was started with
National Science Foundation support grant no.
DEB 76-115568. I thank M. Stowe for infor-
mation. I am obliged to several readers, especial-
ly L. Leibensperger, L. R. Levi and E. H. Buckup
for finding many errors. I thank J. C. Coken-
dolpher and C. D. Dondale for reviewing the
manuscript and suggesting many improvements.

KAIRA SPECIES
Kaira O. P.-Cambridge

Kaira O. P.-Cambridge, 1889:56. Type species K. gib-
berosa O. P.-Cambridge, 1889, designated by F. P.-
Cambridge, 1904:522.

Caira Simon, 1895:894. Changed spelling for Kaira,
an invalid emendation.

Pronarachne Mello-Leitao, 1937:9. Type species by
monotypy P. aries Mello-Leitao, 1937 {=Kaira al-
tiventerO. P.-Cambridge). First synonymized by Levi,
1977.

Macpos Mello-Leitao, 1 940:59. Type species by mono-
typy M. monstrosus Mello-Leitao, 1 940 {= Kaira gib-
berosa O. P.-Cambridge). First synonymized by Levi,
1977.

Haliger Mello-Leitao, 1943:180. Type species by
monotypy H. corniferus Mello-Leitao, 1 943 {=Kaira
altiventer O. P.-Cambridge). NEW SYNONYMY.
Note on synonymy: Haliger corniferus is an early-
instar immature described as a theridiosomatid.

Diagnosis.— differs from other araneids
in having the abdomen attached close to the mid-
dle of its venter, rather than at its anterior end,
with the axis of the abdomen almost at a right
angle to the cephalothorax (Figs. 13, 56), except
in K hiteae (Fig. 85), K. cobimcha (Fig. 89), and
K. sexta. The female has tubercles on the surface
of the abdomen (Figs. 40, 55, 65, 78) or just on
the anterior humps (Figs. 12, 14), except in K
hiteae (Fig. 85), K cobimcha (Fig. 89), and K
sexta. The distal ends of the first to third tibiae
and the metatarsi and tarsi of females are armed
with many setae and macrosetae (Figs. 13, 18,
56, 60), indistinct in males. All species have dwarf
males (Figs. 12, 22, 47),

Unlike other araneids, Kaira females have a
small and lightly sclerotized epigynum, often with
a fiat keel-like scape (Figs. 3-5, 23-25, 48-50);
the epigynum is difficult to study.

Males lack macrosetae on the palpal patella.
(Males of other araneid genera have 1-3 such

setae.) The median apophysis (M in Fig. 27) bears
two fiagella, originating from the middle of the
apophysis, behind a row of teeth on its distal end
(Figs. 27, 28, 41, 42), in K sexta only one fia-
gellum. The median apophysis of K sexta ap-
pears turned on its long axis, having the single
flagellum pointed “down” (Levi 1991, fig. 342).
The distal articles of the anterior legs have only
an indistinct row of setae, most of one size.

Kaira females and immatures can be confused
with species of Pozonia (Levi, 1993). The geni-
talia separate Kaira species from species of Ocre-
peira and Pozonia, which may have a similarly
shaped abdomen and have setae on distal articles
of the first legs, but are not closely related (judg-
ing by the structure of the genitalia).

Description. “Pale yellow- white with scat-
tered, small, white, brown and black spots form-
ing no distinct pattern (Figs. 12, 38, 40, 60), or
transverse bands in K hiteae, K cobimcha and
K sexta (Figs. 85, 89). Carapace low, almost as
wide as long, with eye region about half the width
of the carapace (Figs. 1, 18, 1 9). Height of clypeus
about equal to diameter of an anterior median
eye. Eyes small and subequal in size; median
ocular quadrangle usually narrower behind than
in front; lateral eyes on a slight tubercle (Fig. 19).
Median ocular quadrangle square or narrower
behind than in front. Tibiae slightly sinuous; tar-
si, metatarsi, and distal portions of tibiae armed
with many setae (Figs. 13, 18, 22). Abdomen,
because of its relatively posterior attachment, al-
most perpendicular to carapace. Abdomen dif-
fers in shape in different species, usually having
tubercles, often having dorsal humps (Figs. 26,
36, 51, 60, 85).

Epigynum small, lightly sclerotized, often with
median keel and posterior median plate (Figs. 3-
5). Posterior median plate variable in shape in
different species (Figs. 24, 49, 58, 67). Tip of keel
perhaps tom by male when mating.

Shape of sternum may differ among species
(not illustrated) and among individuals of same
species.

Male Kaira less than half total length of females
and with less pigment (Fig. 1 2). Presumably due
to dwarf size, Kaira males lack the usual modi-
fications of male araneids: they lack patellar mac-
roseta, the endites are without teeth, the first coxae
without hook, and second tibiae not modified. In
males, the humps on the abdomen are smaller
than those of females, and usually without tuber-
cles (Figs. 22, 47). Unlike the epigynum, the pal-
pus is well developed. Median apophysis (M in

LEVI-THE ORB^WEAVER GENUS KAIRA

211

Map 1.— Distribution of Kaira altiventer and K. gibberosa. Circles = female records, squares = male records,
open circles = immature records.

Fig. 27) has two flagella and a distal row of teeth;
distal hematodocha is present; embolus tip is hid-
den between terminal apophysis (A) and conduc-
tor (C in Figs. 27, 28). Shape of conductor sepa-
rates males of Kaira species (C in Fig. 28).

Relationship.— The two flagella of the male me-
dian apophysis (M in Fig. 27) are believed ho-
mologous to those found in Aculepeira, Amazo-
nepeira and Metepeira; the distal row of teeth on
the median apophysis is also found in some spe-
cies of Metepeira, Aculepeira and Amazonepeira.
While in these genera the presence of two median
apophysis flagellae in males usually correlates with
a tapering, pointed scape on a lightly sclerotized
(except Aculepeira) epigynum in females, the scape
is often flattened in Kaira.

Natural history. —Although a Kaira species was
known to Hentz (1850), nothing was known of
their habits until recently. In response to my 1977
paper, Karl T. Stone (5 March 1978) sent his re-
port of observations made on a Kaira alba female
in a wide-mouth jar. The spider remained on the
underside of the lid, without a web, until a fly was
introduced. The spider dropped on what seemed
a single thread, one-half (12 mm) inch long, and
hung there until the fly blundered into her, and
she clamped her legs around it, killing it.

More recently Mark Stowe (1986) reported on
Kaira alba. They do not make an orb and spe-
cialize in catching male moths. The spider builds
a small trapezoidal web, remade every 20 min,
containing two triangular zigzags of threads. The

spider hangs upside-down by the fourth leg on the
lower and shorter parallel edge of the trapezoid
spread by the other legs (Stowe 1986, fig. 5.7b).
When a moth flies into the basket formed by the
spider’s legs, the spider drops on a short line while
clasping and biting the moth. After the moth stops
struggling it is wrapped in the usual araneid fash-
ion. The wrapped moth is placed on a trapeze line
between the spider’s fourth legs and the hunting
posture is resumed. As many as eight moths are
wrapped together before the spider feeds on the
package. The moths caught are listed by Stowe
(1986). Since all moths caught are male, and these
present only a small proportion of the available
moth species, Stowe assumes that the spider uses
a pheromone as an attractant. Two young ob-
served had the same hunting posture as the adult.
Although the zigzag lines are minimally viscid,
they may be homologous to the viscid spiral in
the araneid orb; here they play no part in food
capture. The Kaira diurnal resting posture with
legs I and II extended forward resembles that of
tetragnathids. I agree with Stowe that the moth
catching behavior must be independently evolved
from that of Mastophora.

The egg sacs have an outside covering of fluffy
silk and are made on top of each other (Stowe
1986, fig. 5.9). (This Stowe illustration also shows
the top of the white spider, above a hanging moth.)

Matching sexes.— Immature specimens can be
determined with uncertainty by the shape of the
abdomen. Males have never been collected with

212

THE JOURNAL OF ARACHNOLOGY

females. When revising the North American spe-
cies (Levi 1977), I found males labeled K. alba by
W. Gertsch and A. Archer in the collections. This
appears to be correct because the male of K. alba
has the same distribution {e. g., Florida) as the
female and does not fit with any other female
araneid. Kaira alba is one of the two species whose
sexes are matched with some confidence. The oth-
ers are K. hiteae and K. sexta, in which the ab-
domen of the male is similar to that of the female.

Distribution.— species are known only

from the Americas (Maps 1, 2). Kaira sabino and
the male of K. hiteae are illustrated in Levi ( 1 977),
K. sexta in Levi (1991, figs. 339-342),

Misplaced species.— A'pe/ra electa Keyserling,
1883, placed in Kaira by Levi, 1991, is probably
an Araneus.

Kaira dromedaria O. P.-Cambridge, 1893, is a
Pozonia {IjQYi 1993).

Kaira granadensis Mello-Leitao, 1 94 1 a, is a Po~
zonia (Levi 1993).

KEY TO KAIRA FEMALES

1. Abdomen with pair of humps, without tubercles, shield-shaped (Figs. 85, 89) 2

Abdomen otherwise (Figs. 12, 36, 51, 78) 4

2(1). Abdomen wider than long (Fig. 89) 3

Abdomen longer than wide (Fig. 85); south-central United States hiteae

3(2). Epigynum in ventral view with two transverse bars (Fig. 86); southern Brazil (Map 2) .... cobimcha

Epigynum in ventral view with only comers of anterior bar showing (Levi 1991, fig. 339); Guatemala

to Amazon area (Map 2) sexta

4(1). Epigynum with a median notch in a posterior transverse bar (Levi 1977, fig. 141); Arizona

(Map 2) sabino

Epigynum otherwise; not in Arizona 5

5(4). Abdomen with a long median anterodorsal projection (Figs. 69, 73, 74) 6

Abdomen with a pair or more of humps or slight median projections (Figs. 12, 36, 60) 7

6(5). Abdomen drop-shaped (Fig. 69); northern Argentina (Map 2) candidissima

Abdomen a long cone (Figs. 73, 74); southeastern Brazil to northern Argentina (Map 2) conica

7(5). Posterior of abdomen without tubercles (Figs. 12-18, 26, rarely one pair of small humps, Levi

1977, fig. 134) 8

Posterior of abdomen with humps or tubercles (Figs. 40, 51, 65, 78) 9

8(7). Epigynum with longitudinal projection (Figs, 23-25); southeastern United States to northern Mex-
ico (Map 2) alba

Epigynum with transverse projection (Figs. 3-1 1); Texas to southern Brazil (Map 1) altiventer

9(7). Abdomen much longer (or higher) than wide (Fig. 35) 10

Abdomen as wide or wider than long (Figs. 65, 78) 11

10(9). Abdomen rounded anteriorly (Fig. 51); Colombia (Map 2) tulua

Abdomen with a pair of dorsal humps (Figs. 35-40); Mexico to southern Brazil (Map 1) . . .gibberosa

1 1(9). Abdomen rectangular (Figs. 55, 78) 12

Abdomen subcircular (Figs. 60, 65); Peruvian Amazon region 13

12(1 1), Epigynum with extended scape (Fig. 75-77); southeastern Brazil (Map 2) echinus

Epigynum with a short scape (Figs. 52-54); Peruvian Amazon region (Map 2) erwini

13(1 1). Epigynum as in Figures 61-64; Map 2 dianae

Epigynum as in Figures 57-59; Map 2 shinquito

KEY TO KAIRA MALES

(Males of K candidissima, K. conica, K dianae, K. erwini, K. sabino, K shinguito, K tulua are
unknown.)

1. Median apophysis twisted, with only one flagellum (Levi 1991, fig. 342); Guatemala to Amazon area

(Map 2) sexta

Median apophysis not twisted, with two flagella (Figs. 20, 21, 79, 80, 90, 91) 2

2(1), Abdomen wider than long (Fig. 81); conductor of palpus as in Figure 80; from Bahia State, Brazil to

northern Argentina (Map 2) echinus

Abdomen longer than wide (Figs. 22, 47, 92); conductor of palpus otherwise 3

LEVI -THE ORB- WEAVER GENUS KAIRA

213

3(2). Abdomen dorsum with an anterior dorsal shield (Fig. 92); terminal apophysis with “transverse”

sclerotized edge (Fig. 91); Mato Grosso, Brazil (Map 2) cobimcha

Abdomen without dorsal shield (Figs, 22, 47) 4

4(3). Conductor with a distal tooth on side of terminal apophysis (Figs. 42, 44, 46); Mexico to southern

Brazil (Map 1) gibberosa

Conductor without distal tooth on conductor (Figs. 21, C in 28) 5

5(4). Median apophysis of palpus with only 3 or 4 large teeth (Figs. 20, 21); Texas to southern Brazil

(Map 1) altiventer

Median apophysis with more than 6 teeth (M in Figs. 27, 28) 6

6(5). Conductor with dark distal swelling overhanging subdivided lateral pockets (Levi 1977, fig. 140);

terminal apophysis bluntly pointed (Levi 1977, fig. 139); south-central United States (Map 2) ... hiteae
Tip of conductor facing flagella (C in Fig. 28); terminal apophysis sharply pointed (A in Fig. 27);
southeastern United States to northern Mexico (Map 2) alba

Kaira altiventer O. P. -Cambridge
Figures 3--22; Map 1

Kaira altiventer O. P.-Cambridge, 1889:56, pi. 3, fig.
13, $. Female holotype from Veragua [Veraguas
Prov.], Panama, in BMNH, examined. Keyserling,
1892: 62, pi. 3, fig. 48, 9. F. P.-Cambridge, 1904:

522, pi. 51, fig. 10, 9 Levi, 1977:218, figs. 130-137,

9, 6.

? Caira spinosa Simon, 1897:478. Female lectotype
designated by Levi, 1977 and imm. paralectotype
from Sao Paulo de Olivenga, Amazonas State, Brazil,
and Pebas, Depto. Loreto, Peru in MNHN, exam-
ined. First synonymized by Levi, 1977.

Map 2.— Distribution of Kaira species. Circles of K. echinus and K. hiteae = female records, squares = male
records.

214

THE JOURNAL OF ARACHNOLOGY

? Caira capra Simon, 1897:479. Immature holotype
from Paraguay in the MNHN, examined. NEW
DOUBTFUL SYNONYMY.

Pronarachne aries Mello-Leitao, 1937:9, fig. 10, 9, Fe-
male holotype from Itatiaia, Rio Grande do Sul in
MNRJ, examined. First synonymized by Levi, 1 977.
Haliger corniferus Mello-Leitao, 1 943: 1 80, fig. 1 8, imm.
Immature holotype from Rio Grande do Sul in
MNRJ, lost. Brignoli, 1983:239. NEW SYNONY-
MY.

Synonymy.— The holotype of Caira spinosa is
a large, mature individual, 13.5 mm total length,
abdomen 10.3 mm high, with only one pair of
tubercular humps and tubercles between (Figs.
14, 15). The Caira capra holotype is immature
(5.8 mm total length) and lacks some of the tu-
bercles on the humps of the abdomen (Fig. 1 8).
The holotype of Pronarachne aries has a flat tri-
angular scape (Figs. 9--1 1) as in K. altiventer but
the abdominal humps are much thinner than in
other females (Figs. 16, 17, 19). The holotype of
Haliger coniferus, originally placed in Theridio-
somatidae, is only 2.5 mm total length, lacks
tubercles on the humps and on the posterior of
the abdomen, and has a thin white line going
from the tip of one hump to the tip of the other,
as do other specimens from this area (Fig. 12).
Coddington (1986), in his study of theridioso-
matids, considered Haliger unrecognizable, but
it had been misplaced in that family. None of
these specimens has posterior tubercles on the
abdomen.

Description. — from Parana State,
Brazil: Carapace yellowish, cephalic region with
tiny, irregular black spots. Chelicerae, labium,
endites spotted orange. Sternum orange with
brown line all around. Coxae orange, with dark
brown spots; legs orange with tiny black spots.
Dorsum of abdomen with posterior part darker
than anterior and with minute stipples, a light
transverse line between humps (Fig. 12), and
larger black spots on tubercles; sides and venter
spotted. Eyes subequal in size. Anterior median
eyes 1.5 diameters apart. Posterior median eyes
1 .2 diameters apart. Height of clypeus equals 1 .4
diameters of the anterior median eyes. Sternum
elongate, extending between fourth coxae. Ab-
domen with two humps, bearing asymetrical tu-
bercles (Figs. 12, 13). Total length 10 mm. Car-
apace 4,2 mm long, 3.8 wide, behind lateral eyes
1.8 wide. First femur 5.4 mm, patella and tibia
6.7, metatarsus 3.3, tarsus 1.2, Second patella
and tibia 5.4 mm, third 2.9, fourth 3.5. Abdomen
9.4 mm high.

Male from Hidalgo, Mexico: Color as in female

but legs with wide, dark rings, and sides of the

abdomen with irregular dusky spots having a col-
orless center (Fig. 22). Posterior median eyes 1 .2
diameters of anterior medians, laterals 1 diam-
eter. Anterior median eyes 1.2 diameters apart,
1 diameter from laterals. Posterior median eyes
their diameter apart, 2. 1 diameters from laterals.
Height of clypeus equals 0.9 diameter of the an-
terior median eyes. Abdomen with a pair of an-
terior humps (Fig. 22). Total length 2.0 mm. Car-
apace 1.00 mm long, 0.84 wide, behind lateral
eyes 0.53 wide. First femur 1 .06 mm, patella and
tibia 1.30, metatarsus 0.71, tarsus 0.47. Second
patella and tibia 1.00 mm, third 0.58, fourth
0.76. Abdomen 1.58 mm high.

Note: Males and females have not been col-
lected together but were matched by Levi, 1977,
because they have physical similarities and were
collected in Mexico and Central America.

Variation: Total length of females 7.0 to 13.5
mm, males 1.9 to 2.2. Illustrations (Figs. 3-5,
12, 13) were made from a female from Parana
State, Brazil and a male (Figs. 20-22) from Hi-
dalgo State, Mexico. Some females have the
humps curved, with their tips approaching and
almost touching, forming an “O”. The illustra-
tion (Levi 1977) fig. 134, was made of a female
from Edinburg, Texas; unlike all others, it had a
pair of posterior tubercles. The immature holo-
types of C. capra, H. corniferus and the immature
specimen from Montenegro, Rio Grande do Sul,
all lack tubercles on the humps of the abdomen
(Fig. 18).

Diagnosis. —The epigynum of the female, un-
like that of K. alba, has a flat, triangular, curved
scape, the tip projecting anteriorly (Figs. 3, 6)
and, like K. alba but unlike others, no humps or
tubercles on the posterior of the abdomen (Figs.
12-18). The tiny male can be separated from
others by having a palpus with only three or four
long, black teeth on the distal end of the median
apophysis and by the shape of the conductor,
pointed and longest on the side of the dark ter-
minal apophysis (center and 1 100 h of Fig. 21).

Natural History. —A male from Texas was col-
lected in low shrubs; another was the prey of a
Trypargilum nitidum wasp in Costa Rica. One
female in Peru was found hanging on a thread,
another was obtained at the same location by
fogging the canopy.

Distribution.— Southern Texas to southern
Brazil (Map 1).

Additional specimens examined. — MEXICO. Hi-
dalgo: El Salto, 22-23 April 1967, 3 (W. Peck, CAS).

LEVI -THE ORB- WEAVER GENUS KAIRA

215

Figures \~22.—Kaim species: K. alba carapace and chelicerae (1,2); 1, frontal; 2, lateral. K. altiventer (3-22);
3-17, 19, female; 3-1 1, epigynum; 3, 6, ventral; 9, anterior; 4, 7, 10, posterior; 5, 8, 11, lateral; 12, dorsal with
small male, same scale; 13, 15, 17, lateral; 14, 16, posterior; 19, carapace; 20-22, male; 20, 21, left palpus; 20,
mesal; 21, ventral; 22, lateral; 3-5, 12, 13, from Parana State, Brazil; 6-8, 14, 15, holotype of Caira spinosa;
9-11, 16, 17, 19, holotype of Pronarachne aries; 18, immature holotype of Caira capra\ 20-22, from Hidalgo
State, Mexico, Scale lines = 1.0 mm, of genitalia = 0.1 mm.

216

THE JOURNAL OF ARACHNOLOGY

EL SALVADOR. San Salvador: Un\x2iry, March 1 954,
<5 (J. B. Boursol, AMNH). COSTA RICA. San Jose:
San Antonio de Escazu, 1400 m, 9 October 1982, 9
(W. Eberhard SAI-73, MCZ), November 1988, 9 (W.
Eberhard, USNM). PERU. Loreto: Rio Samiria, 29
May 1990, 9 (D. Silva D, MUSM), 20 May 1990, imm.
(T. Erwin, D. Silva D, MUSM). BRAZIL. Sao Paulo:
Ilha Sao Sebastiao, 28 Jan. 1951, 9 (MZSP 6608). Pa-
rana: Rolandia, 1948, 9 (A. Mailer, AMNH); Almi-
rante Tamandare, 8 Aug. 1984, $ (C. C. Costa, MCN
12,500). R/o Grande do Sul: MontQnQgro, 1 September
1979, imm. (H. BischofF, MCN 6431); Ponta Grossa,
Porto Alegre, 7 May 1976, imm. (A. A. Lise, MCN
4241); Triunfo, 15 September 1977, imm. (A. A. Lise,
MCN 6492); Viamao, 22 October 1988, S (A. B. Bon-
aldo, MCN 17953). PARAGUAY. Concepcion: Apa,
Aug. 1909, 9 (AMNH).

Kaira alba (Hentz)

Figures 1,2, 2 3-2 8; Map 2

Epeira alba Hentz, 1850:20, pi. 3, fig. 7. Female from

North Carolina, destroyed.

Kaira alba: -LqVi, 1977:216, figs. 117-129,9,6. Stowe,

1986: 115, fig. 5, 7 (web).

Description. /r<9W Virginia: Cepha-
lothorax yellow-white with brown spots and
streaks; sternum spotted, legs with rings and spots.
Abdomen white with a dark patch between
humps; venter dusky, spotted. Eyes subequal in
size. Anterior median eyes 1.3 diameters apart,
1 .4 diameters from laterals. Posterior median eyes
1 diameter apart, 1.5 diameters from laterals.
Height of clypeus equals 0.7 diameter of anterior
median eye. Abdomen with a pair of dorsal
humps with tubercles (Fig. 26). Total length 4.8
mm. Carapace 2.1 mm long, 1.8 wide, behind
lateral eyes 0.9 wide. First femur 2.5 mm, patella
and tibia 3.1, metatarsus 1.7, tarsus 0.7. Second
patella and tibia 2.3 mm, third 1.2, fourth 1.8.
Abdomen 4. 1 mm high.

Male from North Carolina: Carapace yellow-
ish, with eye area dusky, with a median dusky
band, and with thoracic region having dusky
margin, Chelicerae yellowish with dusky patch.
Sternum with black marks. Legs with dusky rings
and spots. Dorsum of abdomen white with a
black patch between humps and scattered black
spots of various sizes; venter dusky. Posterior
median eyes 1.2 diameters of anterior medians,
laterals 1 diameter. Anterior median eyes their
diameter apart, their diameter from laterals. Pos-
terior median eyes 0.5 diameter apart, 1 diam-
eter from laterals. Height of clypeus equals 0.8
diameter of anterior median eye. Abdomen wid-

est in middle, humps slightly tubercular. Total
length 2.6 mm. Carapace 1.38 mm long, 1.11
wide, 0.65 wide behind lateral eyes. First femur
1.61 mm, patella and tibia 1 .96, metatarsus 1 .47,
tarsus 0.36. Second patella and tibia 1.54 mm,
third 0.87, fourth 1.01. Abdomen 1.87 mm high.

Note: Males and females were matched be-
cause of similarly shaped abdomens and because
both were collected in the same areas of the Unit-
ed States (Map 2).

Variation: ToX2i\ length of females 4. 8-7. 2 mm,
males 2. 6-2. 9. Female abdomens 4. 1-8.2 mm
high. Illustrations were made from a female from
Virginia and a male from North Carolina.

Diagnosis.— As in K. altiventer, females of K.
alba lack posterior humps or tubercles on the
abdomen (Fig. 26) but differ by the vertical keel
of the epigynum (Figs. 23-25). The median
apophysis of the male palpus differs from that of
K. alba in having six or more teeth (Figs. 27,
28), and the conductor differs from that of K.
altiventer in being longest on the side closest to
the median apophysis (C in Fig. 28).

Natural History.— Observations are given in
the introduction to the genus.

Distribution.— Virginia to Mexico (Map 2).

Additional specimen examined . — UNITED

STATES. Virginia: Black Pond [?], 14 Sept. 1913, 9
(USNM).

Kaira gibberosa O. P.-Cambridge
Figures 29-47; Map 1

Kaira gibberosa O. P.-Cambridge, 1890:57, pi. 3, fig.
12, 9. Female holotype from Veragua [Veraguas
Prov,;], Panama, in BMNH, examined. Keyserling,
1892:63, pi. 3, fig. 49, 9. F. O. P.-Cambridge, 1904:
522, pi. 51, fig. 9, 9. Roewer, 1942: 904.

Kaira obtusa Keyserling, 1892:66, pl.3, fig. 51, imm.
Immature holotype from Taquara, Rio Grande do
Sul, in BMNH, examined. Roewer, 1942:904. NEW
SYNONYMY.

? Macpos monstrosus Mello-Leitao, 1940:59, fig. 6, 9.
Female holotype from Jardim Botanico, Rio de Ja-
neiro, Brazil in MNRJ, lost. DOUBTFUL NEW
SYNONYMY.

Wagneriana minutissima Mello-Leitao, 19416:250.
Male holotype from Rio Negro, Parana State, Brazil,
in MNRJ, no. 58298, examined. Brignoli, 1983:28 1 .
NEW SYNONYMY.

Caira gibberosa: - Bonnet, 1956:925.

Caira obtusa: - Bonnet, 1956:925.

Kaira monstrosa: - Brignoli, 1983:271.

Note: The type locality of K. gibberosa is Ver-
agua, Panama. Veragua refers to Veraguas Prov-

LEVI~THE ORB-WEAVER GENUS KAIRA

111

Figures 23=47.— Kaira species: K. alba (23-28); 23-26, female; 23-25, epigynum; 23, ventral; 24, posterior;
25, lateral; 26, abdomen, dorsal; 27, 28, left male palpus; 27, mesal; 28, ventral. K. gibberosa (29-47); 29-36,

39, female; 29-34, epigynum; 29, 32, ventral; 30, 33, posterior; 31, 34, lateral; 35, 36, 39, abdomen; 37, 38,

40, immature abdomen; 35, 37, lateral; 36, 38-40, posterior; 41-47, male; 41-46, palpus; 41, 43, 45, mesal;
42, 44, 46, ventral; 47, dorsolateral; 29-31, 35, 36, holotype o^K. gibberosa; 39, holotype of Macpos monstrosus
(after Mello-Leitao); 37, 38, immature holotype of K. obtusa; 40, penultimate female from Parana State, Brazil;
45, 46, holotype of K. minutissima; 41, 42, from Michoacan State, Mexico; 43, 44, from Rio Grande do Sul
State, Brazil. Abbreviations: A, terminal apophysis; C, conductor; M, median apophysis; T, tegulum. Scale lines
= 1.0 mm, of genitalia = 0.1 mm.

218

THE JOURNAL OF ARACHNOLOGY

ince, Panama (Selander & Vaurie, 1962). The
fragmented holotype has insect-pin holes. It is
larger (total length 6.0 mm, abdomen 7. 0 mm
high) than the specimen from Paranaa State, Bra-
zil, but the measurements of the carapace and
legs are similar.

The holotype of Kaira obtusa, although im-
mature, has anterior and posterior humps on the
abdomen (Figs. 37, 38), as does K. gibberosa.
This species seems relatively common in south-
ern Brazil, the type locality of K. obtusa. The
specimen has a total length of 3.0 mm. Carapace
1.40 mm long, 1.35 wide, behind lateral eyes
0.70 wide. First femur 1.59 mm, patella and tibia
1.95, metatarsus 1.01, tarsus 0.48. Second pa-
tella and tibia 1.49 mm, third 0.87, fourth 1.09.
Abdomen (shrivelled) 3.0 mm high. The name
Kaira obtusa was erroneously synonymized with
K. altiventer hy Levi, 1977.

The holotype of Macpos monstrosus has a total
length of 6 mm, the first patella and tibia 5.5
mm (measurements from Mello-Leito 1 940) al-
most twice the length of the female of K. gib-
berosa illustrated. Mello-Leitao’s illustration of
the female from the side and the abdomen from
posterior (Fig. 39) suggest that it may be this
species, but the abdomen does not narrow dor-
sally and the anterior humps are almost as long
as the abdomen below (Fig. 39).

According to Mello-Leitao (1941 b), Wagner-
iana minutissima is described from a female, but
the holotype and description are of a male. The
holotype resembles the male illustrated (Figs. 45,
46) but is poorly preserved and the median
apophysis flagellae of the left palpus are broken
off.

Description.— from Parana State,
Brazil: The animal is heavily pigmented with
brown spots, some black and white streaks, ce-
phalic region darkest. Sternum dusky orange. Legs
spotted and streaked, distal ends of femora and
patellae dark brownish black with white streaks.
Dorsum of abdomen dark between humps and
posteriorly with a black chevron pointing ante-
riorly (Figs. 35, 36). Eyes subequal in size. An-
terior median eyes 1.3 diameters apart, 1.8 di-
ameters from laterals. Posterior median eyes 1.2
diameters apart, 2.5 diameters from laterals.
Height of clypeus 0.7 diameter of anterior me-
dian eye. Abdomen with two pairs of humps and
tubercles on sides of abdomen (Fig. 35--40). Total
length 5.2 mm. Carapace 2.5 mm long, 2, 1 wide,
behind lateral eyes 1 .2 wide. First femur 2.8 mm.

patella and tibia 3.6, metatarsus 1.7, tarsus 0.7.
Second patella and tibia 2.6 mm, third 1.5, fourth
2.1. Abdomen (shrivelled) 4.6 mm high.

Male from Vacaria, Rio Grande do Sul State:
Carapace pale yellowish white with white streaks,
a dusky patch covering cephalic region (Fig. 47).
Sternum pale light yellowish, appearing spotted.
Legs pale yellowish with distal halves of femora
and patellae dark dusky. Abdomen with white
line behind and around anterior protrusions,
darker patches on paired posterior swellings with
white mark behind; sides darker; venter pale (Fig.
47). Eyes subequal in size. Anterior median eyes
0.8 diameter apart, 0.7 diameter from laterals.
Posterior median eyes 0.9 diameter apart, 1.5
diameters from laterals. Height of clypeus equal
to 0.5 diameter of anterior median eye. Abdo-
men with a pair of diagonal, dorsal humps and
pair of smaller posterior humps (Fig. 47). Total
length 2.0 mm. Carapace 0.91 mm long, 0.83
wide, behind lateral eyes 0.47 wide. First femur
0.88 mm, patella and tibia 1.53, metatarsus 0.58,
tarsus 0.41. Second patella and tibia 0.87 mm,
third 0.52, fourth 0.69. Abdomen 1.46 mm high.

Note: Males and females were matched be-
cause they were collected in the same area and
both have a posterior pair of humps on the ab-
domen (Figs. 35-40, 47).

Variation: Total length of females 5.2 to 8.0
mm, abdomen 4.6 to 8.0 mm high. Illustrations
were made from a female from Parana State and
male from Vacaria, Rio Grande do Sul. In the
male from Michoacan State, Mexico, the con-
ductor of the palpus (Fig. 42) is shorter than in
males from southern Brazil (Figs. 43-46). I as-
sume they belong to one species.

Diagnosis. “The female differs from other
species of Kaira by having a longitudinal, keel-
shaped scape in the epigynum (Figs. 23-25, 29-
34) and by having a second pair of dorsal tu-
berculate humps on the abdomen, and tubercles
on the posterior humps and on the sides (Figs.
35-40). The male is separated from others by
having a distal black tooth on the conductor of
the palpus (Figs. 41-46).

Distribution. —Mexico to southern Brazil (Map

1).

Specimens examined. — MEXICO. Michoacan: 78
km SE Aquila, 13 July 1984, <?, doubtful determ. (J. B.
Woolley, MCZ.). PANAMA. Panama: Summit, Aug.
1950, 2 imm., 19 Aug. 1954, imm. (A. M. Chickering,
MCZ). BRAZIL. Parana: Rolandia, May 1947, 2 pe-
nult. $, 1948, 9 (A. Mailer, AMNH). Santa Catarina:

LEVI -THE ORB-^WEAVER GENUS KAIRA

219

Figures 48-74.— species; K. tulua, female (48-51); 48-50, epigynum; 48, ventral; 49, posterior; 50,
lateral; 5 1 , dorsal. K. erwini, female (52-56); 52-54, epigynum; 52, ventral; 53, posterior; 54, lateral; 55, abdomen,
posterior; 56, lateral. K. shinguito, female (57-60); 57-59, epigynum; 57, ventral; 58, posterior; 59, lateral; 60,
dorsal. K. dianae, female (61-65); 61-64, epigynum; 61, anterior; 62, ventral; 63, posterior; 64, lateral; 65,
abdomen, posterior. K. candidissima, female (66-69); 66-68, epigynum; 66, ventral; 67, posterior; 68, lateral;
69, abdomen, posterior. K. conica, female (70-74); 10-12, epigynum; 70, ventral; 71, posterior; 72, lateral; 73,
74, abdomen; 73, lateral; 74, posterior. Scale lines = 1.0 mm, of genitalia = 0.1 mm.

220

THE JOURNAL OF ARACHNOLOGY

Pinhal, 700 m, May 1947, 2 penult. 9 (A. Mailer,
AMNH). Rio Grande do Sul: Vacaria, 21-25 Apr. 1 982,
$ (A. A. Lise, MCN 10295); Porto Alegre, Morro San-
tana, 1 Sept. 1984, imm. (A. A. Lise, MCN).

Kaira tulua new species
Figures 48-51; Map 2

Holotype. — Female holotype from Rio Tulua,
1 100 m, Depto. Valle, Colombia, Aug. 1977 (W.
Eberhard E-231), in MCZ. The specific name is
a noun in apposition after the locality.

Description.— Female holotype. Carapace pale
yellow-white with white spots. Chelicerae, labi-
um, endites and sternum yellow-white. Legs pale
yellow-white with a black spot on the anterior
face of tibiae and tarsi. Dorsum of abdomen
whitish with a pair of posterior black marks (Fig.
51); venter whitish. Posterior median eyes 0.8
diameter of anterior medians, laterals 0.8 di-
ameter. Anterior median eyes 1 . 1 diameters apart,
1 .8 diameters from laterals. Posterior median eyes
1.1 diameters apart, 2.8 diameters from laterals.
Lateral eye tubercles distinct. Height of clypeus
equals 0.8 diameter of anterior median eye. Ab-
domen without large protrusions, but covered
with tubercles that are not symmetrical (Fig. 5 1).
Total length 7.0 mm. Carapace 2.9 mm long, 2.7
wide, 1.2 behind lateral eyes. First femur 2.8
mm, patella and tibia 3.9, metatarsus 2.1, tarsus
1.0. Second patella and tibia 3.0 mm, third 1.8,
fourth 2.5. Abdomen 5.8 mm high.

Diagnosis.— The round anterior profile of the
abdomen, posteriorly truncate, and covered with
tubercles (Fig. 51) distinguishes this species. The
epigynum has a small keel with a posterior me-
dian plate as wide as long (Figs. 48-50).

Kaira erwini new species
Fig. 52-56; Map 2

Holotype.— Female holotype from Rio Sa-
miria, Cocha Shinguito, Depto. Loreto, Peru,
June 1990 (T. Erwin), in MUSM. The species is
named after the collector, the entomologist T.
Erwin.

Description.— holotype: Cephalotho-
rax pale yellowish white with a brown stippled
patch on the carapace and a brown spot on the
posterior of second tibia. Dorsum of abdomen
with white pigment spots; white pigment on ven-
ter and underside of tubercles. Eyes subequal in
size. Anterior median eyes their diameter apart.

1 .4 diameters from laterals. Posterior median eyes
0.8 diameter apart, 1.8 diameters from laterals.
Height of clypeus equals diameter of anterior
median eye. Abdomen wider than long, with five
pairs of small humps anteriorly, three pairs pos-
teriorly (Figs. 55, 56). Total length 4.3 mm. Car-
apace 2.1 mm long, 2.1 wide, 0.9 behind lateral
eyes. First femur 3.1 mm, patella and tibia 3.6,
metatarsus 2.2, tarsus 1.1. Second patella and
tibia 2.5 mm, third 1.5, fourth 2.1. Abdomen
3.1 mm high.

Variation: ToX^iX length of females 4. 0-5.0 mm.

Diagnosis.— Ffl/ra erwini differs from Kaira
echinus by the placement of the humps on the
abdomen (Figs. 55, 56) and by having a long,
keel-shaped, pointed scape on the epigynum (Figs.
52-54).

Specimens examined. — PERU. Loreto: Rio Samiria,
fogging, 20 May 1 990, 9 paratype (T. Erwin, D. Silva,
MUSM). Madre de Dios: Tambopata, trocha del bam-
boo, 290 m, 7 June 1988, 9 (D. Silva, MUSM); AL
bergue Cuzco Amazonica, 12°33'S, 69°03'W, 6 Mar.
1990, night coll., 9 (D. Silva, MCZ).

Kaira shinguito new species
Figs. 57-60; Map 2

Holotype.— Female holotype from Rio Sa-
miria, Cocha Shinguito [ox-bow lake], Depto.
Loreta, Peru, 22 May 1990 (D. Silva D.), in
MUSM. The specific name is a noun in appo-
sition after the locality.

Description.— holotype: Carapace pale
yellow-white, with a pair of brown patches con-
taining darker veins surrounded by some white
pigment spots, and a dark spot posteriorly. Che-
licerae yellow-white with brown patches. Labi-
um, endites yellow- white. Sternum brown. Coxae
yellow-white with brown spot; legs yellow-white.
Dorsum of abdomen yellow-white with some
scattered small brown dots (Fig. 60); venter with
five indistinctly separated brown bands between
epigynum and spinnerets. Eyes subequal. Ante-
rior median eyes 0.9 diameter apart, 1.2 diam-
eters from laterals. Posterior median eyes 0.8
diameter apart, 2 diameters from laterals. Height
of clypeus equals 0.6 diameter of anterior median
eye. Abdomen subspherical with paired tubercles
(Fig. 60). Total length 5.1 mm. Carapace 2.2 mm
long, 2.1 wide, 1.1 behind lateral eyes. First fe-
mur 3.1 mm, patella and tibia 3.4, metatarsus
2.4, tarsus 1.1. Second patella and tibia 2.7 mm,
third 1.6, fourth 2.0. Abdomen 3.7 mm high.

LEVI -THE ORB^ WEAVER GENUS KAIRA

221

Diagnosis.— This female differs from those of
other species by the shape of the abdomen (Fig.
60) and by the shape of the scape of the epigynum
(Figs. 57-59).

Kaira dianae new species
Figs. 61-65; Map 2

Holotype.— Female holotype from Zona Re-
servada Pakitza, on low leaf of tree, 356 m, Dep-
to. Madre de Dios, 11°56'S, 7U17'W, Peru, 27
Sept. 1991 (D. Silva D.) in MUSM. The species
is named after the collector.

Description.— holotype: Carapace pale
yellowish white with brown and black speckles
in cephalic region, eye region with white pig-
ment, thoracic region with black border on sides.
Chelicerae, labium, endites yellowish white.
Sternum brown-black. Coxae, legs pale yellowish
white with brown and black speckles and some
white pigment. Dorsum of abdomen yellowish
white, speckled with black and tiny white spots;
a pair of black patches on anterior dorsal tuber-
cles (Fig. 65); venter speckled. Eyes subequal in
size. Anterior median eyes 0.8 diameter apart,

1 . 1 diameters from laterals. Posterior median eyes
0.6 diameter apart, 1.9 diameters from laterals.
Height of clypeus equals 0.8 diameter of anterior
median eye. Abdomen almost spherical, slightly
longer than wide with tubercles, none quite sym-
metrical (Fig. 65). Total length 5.5 mm. Carapace
2.4 mm long, 2.3 wide, 1.1 behind lateral eyes.
First femur 3.1 mm, patella and tibia 3.8, meta-
tarsus 2.3, tarsus 1.1. Second patella and tibia
2.8 mm, third 1.7, fourth 2.2. Abdomen 4.2 mm
high.

Variation: The immatures (of uncertain deter-
mination) have the sternum with light areas.

Diagnosis. —Kaira dianae differs from K. shin-
quito by having a curved tubular scape on the
epigynum (Figs. 61-64) and from other species
by the subspherical, tuberculate abdomen (Fig.
65).

Specimens examined.— PERU. Madre de Dios: 15

km E Puerto Maldonado, 12°33'S, 69°03'W, 23-25 June
1989, 2 imm. (D. Silva D., MUSM).

Kaira candidissima (Mello-Leitao)

Figs. 66-69; Map 2

Macpos candidissimus Mello-Leitao, 1941c:212, figs.

18, 19, $. Female holotype from El Rabdn, Santa Fe

Province, Argentina, in MLP, no. 15135, examined.
Kaira candidissima: - Brignoli, 1983:271.

Description.— holotype: Carapace,

sternum, legs pale yellow-white. Dorsum of ab-
domen white with sides and venter pale yellow-
white. Eyes subequal in size, very small. Anterior
median eyes slightly less than two diameters apart.
Posterior median eyes slightly more than one
diameter apart. Height of clypeus equals slightly
more than one diameter of anterior median eye.
Abdomen tapers dorsally to a single point (Fig.
69). Total length 6 mm. Carapace 2.6 mm long,

2.1 wide, behind lateral eyes 1.0 wide. First fe-
mur 2.6 mm, patella and tibia 3.3, metatarsus

1.7, tarsus 1.4. Second patella and tibia 2.7 mm,
third 1.7, fourth 2.0. Abdomen 6.2 mm high.

Diagnosis.— The female differs from K. conica
by having smaller eyes, a drop-shaped abdomen
(Fig. 69) and the sternum widest between second
and third coxa.

Kaira conica Gerschman & Schiapelli
Figs. 70-74; Map 2

Kaira conica Gerschman & Schiapelli, 1 948: 1 1 , fig. 11-
13, $. Female holotype from Santa Maria, Misiones
Prov., Argentina, in MACN, examined. Brignoli,
1983:271.

Description.— holotype: Cephalotho-
rax pale yellowish, cephalic region with some
orange meandering marks. Chelicerae dusky
proximally. Sternum light dusky. Femora with
small black spots and distal black ring; patellae
black ventrally. Abdomen white with anterior
median black line below tip; posterior protuber-
ances with black marks (Figs. 73, 74); venter
without marks. Posterior median eyes 1.1 di-
ameters of anterior medians, laterals 1 diameter.
Anterior median eyes 1.3 diameters apart, 2.1
diameters from laterals. Posterior median eyes

1.1 diameters apart, 3 diameters from laterals.
Ocular quadrangle square. Height of clypeus
equals 0.7 diameter of anterior median eye. Ab-
domen drawn out to a single point, with two
posterior bulges and tubercles (Figs. 73, 74). To-
tal length 5.8. Carapace 3.1 mm long, 2.5 wide,
1.3 wide behind lateral eyes. First femur 3.2 mm,
patella and tibia 4.1, metatarsus 2.0, tarsus 0.9.
Second patella and tibia 3. 1 mm, third 1.8, fourth

2.7. Abdomen 10 mm high.

Dmgnosis.— Kaira conica differs from K. can-
didissima by having larger eyes, a longer abdo-
men (Figs. 73, 74) and a narrower sternum, which
is as wide between the second and third coxa as
between the first and second.

222

THE JOURNAL OF ARACHNOLOGY

Specimens examined. —BRAZIL. Sdo Paulo: Pira-
cicaba, 9 (MNRJ). Rio Grande do Sul: General Camara,
19 Oct. 1982, imm. (E. H. Buckup, MCN 10902).

Kaira echinus Simon
Figs. 75-81; Map 2

Caira echinus Simon, 1895:478. Female holotype from

Rio Salobro, Prov. Bahia [Bahia State], Brazil, in

MNHN no. 8338, examined. Bonnet, 1956:925.

Kaira echinus: - Roewer, 1942:904.

Description. holotype: Carapace or-
ange-white with a dark reticulated patch on each
side of cephalic region. Chelicerae, labium, en-
dites orange-white. Sternum brownish black.
Coxae, legs orange-white with black spots and
patches. Dorsum of abdomen white with dusky
and black spots and a pair of anterior black
patches (Fig. 78); venter with black spots on white.
Eyes subequal. Anterior median eyes their di-
ameter apart. Posterior median eyes their di-
ameter apart. Height of clypeus equals 1.5 di-
ameters of anterior median eye. Abdomen with
humps and numerous paired tubercles of differ-
ent lengths (Fig. 78). Total length 6.5 mm. Car-
apace 2.7 mm long, 2.5 wide, 1.2 behind lateral
eyes. First femur 3.4 mm, patella and tibia 4.4,
metatarsus 2.7, tarsus 1.1. Second patella and
tibia 3.1 mm, third 2,0, fourth 2.5. Abdomen
(estimate) 5.3 mm high.

Male from Parana State, Brazil: Carapace
brown, spotted with black and some yellowish
patches. Sternum brown, black patches. Legs yel-
lowish with brownish black rings. Dorsum of
abdomen brown with black spots and patches
(Fig. 81). Posterior median eyes 1.3 diameters of
anterior medians, anterior laterals one diameter,
posterior laterals one. Anterior median eyes 2.5
diameters apart, 2 diameters from laterals. Pos-
terior median eyes 1.5 diameters apart, 1.5 di-
ameters from laterals. Ocular quadrangle almost
square, very slightly wider than long. Height of
clypeus equal to 1 diameter of anterior median
eye. Abdomen wider than long, with humps and
paired tubercles (Fig. 81). Total length 2.5 mm.
Carapace 1.40 mm long, 1.20 wide, 0.65 behind
lateral eyes. First femur 1.62 mm, patella and
tibia 1.89, metatarsus 1.14, tarsus 0.60. Second
patella and tibia 1.56 mm, third 0.88, fourth
1.06. Abdomen 1.98 mm high.

Note: Males were matched with females be-
cause both have an abdomen that is wider than
long.

Variation: Total length of males 2.3 to 2.6,

Illustrations were made from the female holo-
type and a male from Parana State.

Diagnosis.— The shape of the abdomen and
its dark coloration separate the female from oth-
ers. The eight teeth of the median apophysis and
the row of denticles of the conductor of the pal-
pus separate the male.

Specimens examined. — BRAZIL. Parana: Guarap-
uava, Estancia Santa Clara, 22 Nov. 1987, $ (A. D.
Brescovit, MCN 17122). Rio Grande do Sul: Viamao,
Morro do Coco, 9 Dec. 1982, S (A. A. Lise, MCN
11308). ARGENTINA. Misiones: Eldorado, 1 Sept.-
15 Nov. 1964, 3 (A. Kovacs, AMNH).

Kaira hiteae Levi
Figs. 82-85; Map 2

Kaira hiteae Levi, 1977:220, figs. 138-140, 6. Male

holotype from Cove Creek Valley, 9.3 km W of Prai-
rie Grove, Washington County, Arkansas, in MCZ.

Brignoli, 1983:271.

Description.— from southern Texas:
Carapace pale yellow-white with scattered dark
spots on cephalic region. Chelicerae pale yellow-
ish with dusky spots. Labium, endites pale yel-
low. Sternum yellow. Coxae pale dusky yellow,
legs yellow- white with brown rings and tiny brown
spots. Anteriorly dorsum of abdomen dark be-
tween humps, posteriorly light, becoming darker,
posteriorly except for a white transverse band
(Fig. 85). Venter gray with tiny dark spots and
white pigment spots. Eyes subequal. Anterior
median eyes 1.3 diameters apart. Posterior me-
dian eyes their diameter apart. Posterior median
eyes on slight swelling facing laterally and dor-
sally. Height of clypeus equals 1.5 diameters of
anterior median eye. Chelicerae with three long
teeth on anterior margin, three small teeth pos-
terior. Abdomen shield-shaped (Fig. 85). Total
length 6.2 mm. Carapace 2.8 mm long, 2.2 wide,
1 .2 wide behind lateral eyes. First femur 3.4 mm,
patella and tibia 4.3, metatarsus 2.5, tarsus 1.1.
Second patella and tibia 3. 1 mm, third 1.9, fourth
2.4. Abdomen 5.2 mm high.

Male: Description and illustration in Levi 1977:
220, figs. 138-140.

Note: Male and female have the abdomen sim-
ilarly shaped.

Diagnosis.— The female is separated from oth-
er Kaira species by the shield-shaped abdomen
(Fig. 85) and the triangular cross-section of the
scape of the epigynum (Figs. 82-84).

Distribution. — Southeastern United States
(Map 2).

LEVI -= THE ORB- WEAVER GENUS KAIRA

223

Figures 1 5--92. —Kaira species: K. echinus (75-81); 75-78, female; 75-77, epigynum; 75, ventral; 76, posterior;
77, lateral; 78, dorsal; 79-81, male; 79, 80, left palpus; 79, mesal; 80, ventral; 81, dorsal. K. hiteae, female (82-
85); 82-84, epigynum; 82, ventral; 83, posterior; 84, lateral; 85, dorsal. K. cobimcha (86-92); 86-89, female;
86-88, epigynum; 86, ventral; 87, posterior; 88, lateral; 89, dorsal; 90-92, male; 90, 91, palpus; 90, mesal; 91,
ventral; 92, dorsal. Scale lines = 1.0 mm, of genitalia = 0.1 mm.

Natural History.— Padre Island, the collecting
site of the female, is a xeric semibarren spit, windy
and hot (W. Peck, pers. comm.).

Specimen examined. —TEXAS. Cameron County:
South Padre Island, N Brazos Santiago Pass, sweeping
vegetation, 10 Nov. 1979, 9 (T. Allison, MCZ).

Kaira cobimcha new species
Figs. 86-92; Map 2

Holotype.— Male holotype from 260 km N of
Xavantina, 12M9'S, 51°46'W, 400 m, Mato
Grosso, Brazil (Xavantina-Cachimbo Expedi-

224

THE JOURNAL OF ARACHNOLOGY

tion), in MCN ex MCZ. The specific name is an
arbitrary combination of letters.

Description. Cephalothorax yellow-

ish. Abdomen with a transverse dark area an-
terior of two humps, posteriorly with six dusky
transverse lines (Fig. 89). Eyes subequal. Ante-
rior median eyes 1 diameter apart, 2.5 diameters
from laterals. Posterior median eyes 0.8 diameter
apart, 3.2 diameters from laterals. Ocular quad-
rangle slightly narrower behind. Height of clyp-
eus equals 1 diameter of anterior median eye.
Abdomen shield-shaped, wider than long (Fig.
89). Total length 8.0 mm. Carapace 3.3 mm long,
3. 1 wide, 1.3 behind lateral eyes. First femur 3.5,
patella and tibia 4.9, metatarsus 2.5, tarsus 1.1.
Second patella and tibia 3.7, third 2.1, fourth
2.9. Abdomen 5.5 mm high.

Male holotype: Carapace brown. Chelicerae,
labium, endites brown. Sternum brown. Legs
brown with white rings at proximal ends of third
and fourth tarsi. Dorsum of abdomen mostly
brown; anterior shield starting between humps,
area posterior of shield black (Fig. 92). Venter
black. Posterior median eyes 1.3 diameters of
anterior medians, laterals 0.8 diameter. Anterior
median eyes their diameter apart, 1.2 diameters
from laterals. Posterior median eyes 0.8 diameter
apart, 1.3 diameters from laterals. Ocular quad-
rangle square. Height of clypeus equals 0.7
diameter of anterior median eye. Abdomen
shield-shaped, completely covered by a scutum
anteriorly, with some tiny sclerotized spots pos-
teriorly (Fig. 92). Scutum with punctate texture.
Total length 1.8 mm. Carapace 0.92 mm long,
0.81 wide, 0.42 wide behind lateral eyes First
femur 1.08 mm (distal articles and second legs
lost). Third patella and tibia 0.62 mm, fourth
0.75. Abdomen 1.26 mm high.

Note: The female was matched to the male
because of similarities in the shape of the ab-
domen (Figs. 89, 92). The abdomen of both were
slightly lifted to make the illustrations.

Diagnosis. —The shape and markings of the
abdomen (Fig. 89) resembles that of K. sexta.
The epigynum differs by having two anterior
tranverse bars in ventral view (Fig. 86), K. sexta
shows only comers of the anterior bar (Levi 1991,
fig. 339). The scape is hidden by setae. The male
differs from others by the sclerotized plate on the
abdomen (Fig. 92), and the shape of the con-
ductor (center to 1 100 h in Fig. 91).

Natural history.— The holotype was collected
in campo-grassland.

Specimen examined. — BRAZIL. Rio Grande do Sul:
Santa Maria, 10 Nov. 1990, $ (Linck, MCP).

Kaira sexta (Chamberlin), new combination
Map 2

Aranea sexta Chamberlin, 1916:255, pi. 19, fig. 7, imm.
Immature female holotype from Panama, in MCZ,
examined. Roewer, 1942:852.

Ar aneus sextus: BormtX, 1955:598. Levi, 1991:259, figs.
339-342, 9, <3.

Note: The shape of the abdomen in females of
K, hiteae and of K. cobimcha called to mind the
strikingly similar Araneus sextus. On reexami-
nation, A. sextus was found to have the legs mod-
ified as in species of Kaira, and is now trans-
ferred. The abdomen of K. sexta is the same
shape as that of K. cobimcha (Fig. 89); the shape
of the genitalia separate the species.

Kaira sabino Levi
Map 2

Kaira sabino Levi, 1977:221, figs. 141-147, map 3.
Female holotype from Sabino [?Canyon, Pima
County], Arizona in MCZ.

Distribution.— Southern Arizona.

LITERATURE CITED

Bonnet, P. 1956. Bibliographia Araneorum. Tou-
louse 2 (2):9 19-1 925.

Brignoli, P. 1983. A catalogue of the Araneae de-
scribed between 1940 and 1981. Manchester Univ.
Press, Manchester, 755 pp.

Cambridge, F. P.- 1897-1905. Arachnida, Araneidea
and Opiliones. 2: 1-610. In Biologia Centrali-Amer-
icana, Zoologia, London.

Cambridge, O. P.- 1889-1902. Arachnida, Aranei-
dea. 1:1-317. In Biologia Centrali-Americana,
Zoologia, London.

Coddington, J. 1986. The genera of the spider family
Theridiosomatidae. Smithsonian Contrib. Zool.,
422:1-96.

Gerschman de Pikelin, B. S., & R. D. Schiapelli. 1 948.
Aranas Argentinas 11. Communicaciones del Museo
Argentine de Ciencias Naturales “Bernardino Ri-
vadavia”, Serie Ciencias Zoologicas, 1(4): 1-20.
Hentz, N. M. 1850. Descriptions and figures of the
Araneides of the United States. Boston J. Natur.
Hist, 6:18-35.

Keyserling, E. 1883. Neue Spinnen aus Amerika.
Verhandlungen der k. k. zoologisch-botanischen
Gesellsch. Wien, 32:195-226.

Keyserling, E. 1892-1893. Die Spinnen Amerikas,
Epeiridae. Numberg, 4:1-377.

LEVI-THE ORB=WEAVER GENUS KAIRA

225

Levi, H. W. 1977. The orb-weaver genera Metepeira,
Kaira, and Aculepeira in America north of Mexico
(Araneae: Araneidae). Bull. Mus. Comp. Zool., 148:
185-238.

Levi, H. W. 1991. The Neotropical and Mexican
species of the orb-weaver genera Araneus, Dubie-
peira and Aculepeira. Bull, Mus. Comp. ZooL, 152:
167-315.

Levi, H. W. 1993. The Neotropical orb-weaving spi-
ders of the genera Wixia, Pozonia and Ocrepeira
(Araneae: Araneidae). Bull, Mus. Comp. Zool., 153:
47-141.

Mello-Leitao, C. F. de 1937. Aranhas novas ou raras.

Annaes da Academia Brasileira de Sciencias, 9(1):

1-12.

Mello-Leitao, C. F. de 1940. Tres curiosos Argiopi-
dae do Brasil. Revista Chilena de Hist. Natur., 43:
57-62.

Mello-Leitao, C. F. de 1941a. Notas sobre a siste-
matica das aranhas con descrigoes de algumas novas
especies sud-americanas. Anais Acad. Brasileira de
Ciencias. 13:103-127.

Mello-Leitao, C. F. de 1941b. Aranhas do Parana.
Archivos Instituto Biologico Sao Paulo, 1 1(30):235-

257.

Mello-Leitao, C. F. de 1941 c, Aranas de la provincia

de Santa Fe coligidas por el Profesor Biraben. Re-
vista del Museo de la Plata (Nueva Serie) Zoologia,

2:199-225.

Mello-Leitao, C. F. de 1943. Catalogo das aranhas
do Rio Grande do Sul. Archivos do Museu nacional,
Rio de Janeiro, 37:147-245.

Roewer, C. F. 1942. Katalog der Araneae von 1758
bis 1940. Bremen, 1:1-1040.

Selander, R. B. & Vaurie, P. 1962. A Gazetteer to
accompany the “Insecta” Volumes of the “Biologia
Centrali- Americana.” American Museum Novi-
tates, no.2099:l-70.

Simon, E. 1895. Histoire Naturelle des Araignees.
Paris, 1:761-1084.

Simon, E. 1897. Etudes arachnologiques. 27e Me-
moire. XLIL Descriptions d’especes nouvelles de
I’ordre des Araneae. Annal. Soc. entomol. France,
65:465-510.

Stowe, M. 1986. Prey specialization in the Araneidae,
In Spiders: Webs, Behavior and Evolution. (Shear,
W. A., ed.) Stanford Univ. Press, Stanford, Cali-
fornia, pp. 101-131.

Manuscript received 10 January 1993, revised 1 July
1993.

1993. The Journal of Arachnology 21:226=257

REVISION OF THE GENUS TRECHALEA THORELL
(ARANEAE, TRECHALEIDAE) WITH A REVIEW OF THE
TAXONOMY OF THE TRECHALEIDAE AND PISAURIDAE
OF THE WESTERN HEMISPHERE

James E. Carico: Department of Biology, Lynchburg College, Lynchburg, Virginia
24501 USA

ABSTRACT. The spider family Trechaleidae, introduced by Simon in 1898 but abandoned immediately
thereafter, is defined. Characters proposed for trechaleids by recent authors for this family and some characters
used to separate it from the Pisauridae and Lycosidae are reviewed. Genera included in the family Trechaleidae,
all of which were previously placed in the Pisauridae, are: Trechalea, Hesydrus, Syntrechalea, Dossenus, Par-
adossenus, Dyrines and Enna. Familial placement of all other genera of the western hemisphere previously
placed in the Pisauridae are considered.

The genus Trechalea, with eleven species, is revised. Redescriptions of eight known species are given: T.
longitarsis (C. L. Koch), T. cezariana Mello-Leitao, T. macconnelli Pocock, T. paucispina di Caporiacco, T.
connexa (O. Pickard-Cambridge), T. extensa (O. Pickard-Cambridge), T. gertschi Carico & Minch and T.
amazonica F. Pickard-Cambridge. Descriptions of three new species: T. boliviensis, T. lomalinda and T. trin-
idadensis are presented. The holotype of the type species of Trechalea, T. longitarsis, is lost. The specimen
which was previously regarded as the holotype is mislabelled and belongs actually to another, unnamed, trechaleid
genus. All other relevant species names not synonymized with any of the above and not previously placed in
Trechalea are removed to other known genera or will be later included in new genera.

In the early 1 970’s I began a series of revisions
of the American Pisauridae with Pisaurina (Car-
ico 1 972) and Dolomedes (Carico 1 973). The im-
mediate goal was to clarify the taxonomy, at the
genus and species levels, which would establish
the required base for a later analysis at higher
levels. The traditional family Pisauridae was re-
tained as a matter of convenience with deferral
until later of the re-examination of the family
taxon itself.

While working on a group of primarily Neo-
tropical genera, including Trechalea, I became
aware that they represented a monophyletic clade
which challenged the limits of the traditional
family Pisauridae. This group of genera was
therefore set aside for special consideration and
was used as a basis for a study of the family-level
taxonomy.

Of the seven genera placed here in the family
Trechaleidae Simon, six were described around
the turn of the century by Simon and the Pickard-
Cambridges. These genera include Hesydrus
Simon (1898a), Syntrechalea F. Pickard-Cam-
bridge (1902), Dossenus Simon (1898b), Para-
dossenus F. Pickard-Cambridge (1903), Dyrines
Simon (1903) and Enna O. Pickard-Cambridge

(1897). Trechalea Thorell (1869) was described
approximately thirty years earlier.

Simon apparently realized that he was study-
ing a group of unique spiders when he, rather
cryptically, inserted a two-line notation of the
new family Trechaleidae in his summary of 1 890.
No description was offered and only the reference
to two genera, Dendrolycosa and the type genus
Trechalea, was given. He never again referred to
this family and continued (1898) to name new
species of the genus Trechalea in his old family
Pisauridae.

During a meeting of the International Congress
of Arachnology, I reintroduced the Trechaleidae
(Carico 1986), erroneously referring to this fam-
ily name as a nomen oblitum. At that time I
presented a provisional cladogram showing its
relationship to the families Pisauridae and Ly-
cosidae, with the Pisauridae as the probable sis-
ter group.

The complexity of the Trechaleidae combined
with the rather close relationship of the genera
made it highly desirable that all possible material
be examined before formal publication of a fam-
ily analysis was attempted. Only after examining
all available related type specimens and collec-

226

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

227

tions scattered in numerous museums was I able
to sort out the genera and species. It then seemed
necessary to revise the entire family at once be-
fore any generic revision could be published to
provide what was believed to be the proper foun-
dation for a secure set of family apomorphies.
As a result, much time was required to bring the
project to completion.

In the meantime, various authors have ex-
amined specimens of the family and have gen-
erally also concluded that the Trechaleidae rep-
resents a distinct group and is probably a valid
family. Sierwald (1990), in her efforts to work
out the higher taxonomy of the American Pi-
sauridae, has shown through an excellent study
of the male genitalia that the "'Trechalea genus-
group” represents a distinct taxon from the Pi-
sauridae and that the Lycosidae is the probable
sister group of her ""Trechalea genus-group.”
Griswold (1993) confirmed, by a preliminary cla-
distic analysis of the Lycosoidea, the distinctness
and monophyly of the Trechaleidae and placed
the family as the sister group of the Lycosidae.
Dondale (1986) listed family level synapomor-
phies for the Lycosidae, and further noted that
the family shared some of these characters with
Trechalea.

To date no complete treatment of the family
has been presented nor have any of its included
genera been revised. It is my intention in this
paper to revise the genus Trechalea and to review
the taxonomy of the family Trechaleidae in ref-
erence to the related families Pisauridae and Ly-
cosidae. Later, revisions of the remaining six de-
scribed genera and new genera will follow. A
summary of their phylogenetic relationships and
a key to their identification will also be presented
at the appropriate time.

METHODS

Careful dissections of unexpanded palps were
conducted to learn the relative positions of the
various components, as in the approach of Don-
dale (1986). This method develops a good un-
derstanding of palpal structure that makes pos-
sible comparisons among different genera because
all palpi in this unaltered condition are more
likely to be similar. Expanded palpi were also
studied, but expansion by its very nature, causes
distortion in positional relationships which in-
evitably results because of difficulty in control-
ling the degree of expansion. Comparisons be-
tween specimens of different genera were therefore
much more difficult on palpi in this state.

To study epigyna, the soft tissue was removed
by a combination of dissection and immersion
in KOH or a proteolytic enzyme. Since the in-
ternal parts were often dense and dark, they were
immersed in Chlorox (sodium hypochlorite so-
lution), as per the method of Griswold (1993),
until only the internal lining of the ducts was
uncleared. In addition, to trace better the tubules,
the cleared, alcohol-saturated structure was al-
lowed to air dry completely, which causes the
silvery trace of the lumen to appear when the
structure was reimmersed in alcohol.

Drawings were made with pen and ink, using
a Bausch & Lomb Stereozoom® Microscope
equipped with eyepiece drawing grid and mi-
crometer. The right palp was illustrated from the
ventral and retrolateral positions. SEM micro-
graphs of the median apophysis and tibial retro-
lateral apophysis at ventral view are provided
for comparison and diagnosis. Epigyna were
drawn from the ventral view with setae removed,
and from the dorsal view, uncleared, with the
soft tissue removed.

Measurements are in mm. As an index to the
size of the body, only the length of the carapace
is given because of variability in the condition
of the abdomen. Generally, the length of the ab-
domen is approximately equal to the carapace
length in males while the abdomen may be some-
what longer in females. Eyes or eye group mea-
surements were made with the surface plane per-
pendicular to the axis of sight. Abbreviations and
additional notes pertaining to eye group mea-
surements are in Table 1.

Museum abbreviations. — AMNH ~ American
Museum of Natural History; BMNH—The Nat-
ural History Museum, London; CAS-— Califor-
nia Academy of Sciences; EXPE— Exline-Peck
(now in CAS); FMNH-=- Field Museum of Nat-
ural History; INPA-— Institute National de Pes-
quisas Amazonia; JAK— J. A. Kochalka; JEC—
J. E. Carico; MACN— Museo Argentine de Cien-
cias Naturales, Buenos Aires; MCN— Museu de
Ciencias Naturals, Fundagao Zoobotanica do Rio
Grande do Sul; MECN— Museu Equitoriano de
Ciencias Naturales, Ecuador; MEG— M. E. Ga-
liano; MCZ— Museum of Comparative Zoology;
MZUCR— Museo de Zoologia Universitaria,
Costa Rica; REL— R. E. Leech; USNM — Na-
tional Museum of Natural History.

Abbreviations of anatomical terms for genita-
lia.—Many anatomical terms and abbreviations
of genitalia were adopted primarily from Sier-
wald (1989, 1990) while others were coined for

228

THE JOURNAL OF ARACHNOLOGY

purposes of this paper: <2/? —accessory bulb of
epigynum; (^/—anterior field of the epigynum;
bs— base of spermatheca; c— conductor; cd—
copulatory duct; cdd-~copu\aXoty duct divertic-
ulum; CO— copulatory opening; cy— cymbium;
o'o'— dorsal division of median apophysis; c—
embolus; cZ)— embolic base; ecd—ee\a\ division
of the retrolateral apophysis of the male palpal
tibia; cp/— epigynal fold; c^— embolic groove;
end-~en\a\ division of the retrolateral apophysis
of the male palpal tibia; c/?— epigynal p\aXe\fd—
fertilization duct; guide (part of mo); hs^
head of spermatheca; //o— internal fold of an-
terior field; // — lateral lobes; mo— median
apophysis; m/— middle field of the epigynum;
/? — petiolus; pmo — posterior margin of anterior
field of epigynum; r/o — retrolateral tibial apoph-
ysis; 5— spermatheca; sperm duct; 55— stalk

of spermatheca; 5/— subtegulum; /— tegulum; //—
tibia; vcm — ventral cymbio-tibial membrane of
the palpal tibia; v^/— ventral division of median
apophysis; vp— ventrodistal protuberance of male
palpal tibia; vr— ventrodistal rim of male palpal
tibia.

Family Trechaleidae Simon

Trechaleidae Simon, 1890:82, type genus Trechalea

Thorell, 1869. Bonnet, 1955-1959:4680; Carico,

1986:305; Coddington & Levi, 1991:22, (dado-

gram); Griswold, 1993:1-39.

Dolomedeae (in part), Simon, 1898a:301.
Dolomedidae (in part), Lehtinen, 1967:372.

Diagnosis.— The family Trechaleidae can be
distinguished from the Lycosidae by the presence
of: a retrolateral apophysis and a ventrodistal
refolded rim on the male palpal tibia (Fig. 7), the
posterior eye row (PE) in a single recurved row,
and eyes in two rows.

In only the Pisauridae, in comparison with
Lycosidae and Trechaleidae: the female produces
a nursery web (a matrix of irregular webbing sur-
rounding the egg sac and upon which the female
sits and defends against intruders), in some spe-
cies the male binds the female’s front two pairs
of legs with silk during copulation, and a unique
tubular retreat with associated resting behavior
is constructed by some species.

The Trechaleidae can be distinguished from
both the Lycosidae and Pisauridae by: the male
palpus which has a large distally situated median
apophysis equipped with a dorsal embolic groove
that extends distally into an apical guide (Fig. 7),
the presence of a discoid egg sac carried only on
the spinnerets, presence of a “skirt” on the seam

of the discoid egg sac (Fig. 6), transport of young
on the empty egg sac, and non-reattachment of
dislodged egg sac.

Description.— Very large (body length up to
21 mm, leg span up to 167 mm; Trechalea ce~
zariana) to small (body length down to 3.5 mm,
leg span down to 1 8 mm; undescribed genus and
species), entelegyne, araneomorph, ecribellate,
lycosoid. Carapace low to moderately low, about
as wide as long with cephalic area moderately
distinct to indistinct, longitudinal fovea distinct.
Eyes viewed from above in two rows, PE row
recurved, subequal in size, approximately equi-
distant and separated by about an eye diameter,
AE row straight or nearly so, always smaller than
PE, ALE smaller than AME; ocular quadrangle
wider above, about as high as wide. Clypeus
height variable. Sternum about as wide as long,
truncated anteriorly and acute posteriorly. La-
bium free, length-width ratio may be greater or
lesser than 1 .0. Chelicerae vertical, base robust,
often enlarged anteriorly on males, promarginal
teeth three, equidistant with middle one largest,
retromarginal teeth variable between three and
five in number varying in size and distribution.
Endites longer than wide, parallel.

Abdomen oval, low, somewhat flattened ven-
trally, moderately covered with setae, often patch
of setae present ventrally and anterior to spin-
nerets, no plumose setae found. Six spinnerets
and colulus present. Tracheal spiracle located just
anterior of colulus.

Male and female genitalia are described below.

Legs generally long, slender, often with flexible
tarsi (Fig. 1), and, in some species, flexible meta-
tarsi. Coxae notched ventrally. Third legs always
shortest but relative lengths of other legs vary.
Large macrosetae present on most segments,
found in varying numbers of pairs on ventral
surfaces of tibiae and metatarsi. Long setae, often
curved at tips (Fig. 1), present mainly on ventral
surface of legs. Trichobothria: tarsi with two rows
and in alternating position, metatarsi with one
row, tibia with cluster near proximal end with
few others scattered distally, femur none. Both-
rium (Fig. 2) with hood distinct, angled laterally,
not ridged or embedded. Tarsi and metatarsi may
be scopulate (Fig. 1) with one median and two
lateral claws, all usually dentate. Female palpus
with single, dentate claw.

Distribution.— The family is found entirely
within the region from the Gila River of Arizona,
United States southward to northern Argentina.

Natural history.— The included species have

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

229

Figures 1-5. "Scanning electron micrographs of structures of trechaleids: 1,2, Trechalea gertschi; 1, tarsus,
retrolateral view showing flexible integument and scopula; 2, bothrium; 3-5, Hesydrus habilis; 3, spinnerets (a
= anal pore, c = colulus); 4, enlargement of 3 with emphasis on posterior median spinnerets, mesally located
pairs of minor amputate gland spigots with silk emerging from three of them; 5, attachment of four “carrying
threads” to top valve of egg sac. Magnifications: 1, 60 x; 2, 2000 x. Scales: 3, lOO/u; 4, lOju; 5, 100^.

a wide size range, and one would expect a wide
divergence of habitat preferences to exist. Infor-
mation is available, however, mainly for the larg-
er species. The general impression from collec-
tion data and from personal observation is that
the preferred habitat is around the margins of
bodies of freshwater. Some moderate-sized spe-
cies have been taken from trees and other veg-
etation away from water.

No species is known to make a snare of any
kind and all are apparently entirely cursorial. The
most aquatic species are very adept at walking
on the water surface and crawling underwater in
the manner familiar to the aquatic species of the
pisaurid genus Dolomedes (Simon 1898a; pers.
obs.). The feature of flexible tarsi is common in
the family. The flexibility is due to a general
softness of the cuticle rather than the occurrence
of pseudosegmentation. This flexibility, which
may be quite extensive, could be an advantage
in the support and locomotion on the water sur-
face.

No nursery web is constructed. The female
carries the egg sac containing eggs or first instar
spiderlings by the spinnerets, specifically the
paired minor amputate gland spigots on each
posterior median spinneret (Figs. 3, 4, identified
by J. Coddington, pers. comm.) in the manner
of lycosids. The egg sac has a unique structure,
described as hemispherical by Simon (1898a),
with a generally flattened shape and with a dis-
tinct fringe or “skirt” at the seam between the
two valves (Fig. 6). As the eggs hatch and the
spiderlings occupy more of the internal space of
the egg sac, the “skirt” is obliterated by stretch-
ing. The young emerge from the weak junction
at this seam presumably without aid from the
mother. After emergence, the young are carried
upon the egg sac mostly on the upper valve but,
with crowding above, also sometimes below.
Young may also be found on the abdomen of the
mother, but apparently only as a consequence of
insufficient space on the egg sac and then only if
they are in contact with sac-bome spiderlings. I

230

THE JOURNAL OF ARACHNOLOGY

have not observed young only on the abdomen;
when only a few young are present, they are only
on the egg sac. The mother persists in carrying
the egg sac and has been observed carrying it well
after spiderlings have left. Only four strong lines
are attached from the posterior median spinner-
ets to the center or near center of the upper valve,
leaving a distinct single scar (Fig. 5). The female
will not reattach the egg sac at any stage if it is
dislodged. Both upper and lower valves are
smooth on the surface, the upper usually darker
and thicker while the lower is lighter, thinner,
and flatter (Simon 1898a; Sierwald 1990; pers.
obs.) with the eggs or young often visible.

Status of the type genus.— By reference to Tre-
chalea as the type genus of the family Trechal-
eidae (Simon 1890), it follows that the family
name therefore hinges on the type of its type
species, Trechalea longitarsis (C. L. Koch). How-
ever, the correct identity of this species presents
a problem.

C. L. Koch (1848) described a female from
what is now known as Colombia with a body
length of IVi lines (= 15.7 mm, H. Levi pers.
comm.) and gave it the name of Triclaria longi-
tarsis. Some important distinctive features list-
ed in his description included the pattern of the
eyes (which were figured) and the “sickle-shaped”
tarsi, both of which are important identifiers of
what has traditionally been considered Trechal-
ea. Trechalea was the name later given as a no-
men novum by Thorell (1869) to the genus be-
cause the name Triclaria was preoccupied by a
bird genus.

Later, Karsch (1879) found a dried spider in
the Berlin Museum, no. 2006, which he regarded
as the specimen Koch used when making the
description of Triclaria longitarsis because of
Koch’s reference to the “sickle-shaped” tarsi.
Karsch mentioned two labels with the specimen,
one of which states “Brasilien, Langsdorff” and
the other simply as “longitarsis?”. Noting the
discrepancy of the published locality as “Colom-
bia” and the label contained with the specimen
as “Brazil,” he dismissed it merely as a change
made presumably in curating. Kraus (1955)
agreed with Karsch’s conclusion.

I have examined the specimen no. 2006 men-
tioned by Karsch and found the following: (a)
The specimen is a juvenile with the body length
of 12.2 mm, presently preserved in alcohol, but
was apparently originally mounted dry. The
presence of a hole through the carapace and ster-
num indicates that it was probably perforated

Figure 6. — Diagram of a trechaleid egg sac with “car-
rying threads” attached.

with a pin. Only a single entire leg is still at-
tached; others are shattered, (b) The vial contains
a total of five labels which are as follows: “2006”,
“longitarsis Koch?”, “Brazilien Lgsof’, “Ty-
pus”, and a modem label which has the number,
species name, etc., and the notation, ^\sensu
Karsch)”. The first three of these labels have pin
perforations (two of which were mentioned by
Karsch) and were presumably pinned originally
with the specimen. The “Typus” and modem
label, which were not perforated, were appar-
ently added with the specimen after it was placed
into alcohol storage.

After careful consideration of the evidence, I
have concluded that this specimen, hereafter re-
ferred to as “2006,” is not the type specimen for
Trechalea longitarsis (Koch). My conclusion is
based on the following observations: (1) Koch
indicated that his specimen was a “female”, pre-
sumably meaning it was adult. “2006” is a ju-
venile. (However, these authors, including Simon,
often assumed that juveniles were females and
did not look for an epigynum [Levi, pers. comm.].)
(2) Koch stated that the length of his specimen
is “7V2'"”, (or 15.7 mm). “2006” is 12.2 mm,
which is significantly shorter. (3) An examination
of the external characters of “2006” shows it to
be a member of a new genus (which will be de-
scribed in a later publication) with a distinctive
carapace shape, and offset fifth retromarginal
cheliceral tooth. As will be shown later, this ge-
nus is found only in the southernmost states of
Brazil in the Rio de la Plata drainage system, far
from Colombia. Therefore, “2006” is not from
Colombia, and the specimen label which has the
locality “Brasilien” is probably correct. (4) The
label in the vial with “2006”, “longitarsis Koch?,”
placed after Koch’s description and before
Karsch’s discovery, indicates that the person who
pinned the label to the specimen had doubts about
it, as indicated by the question mark. Therefore,
uncertainty about the specimen apparently pre-

CARICO^THE GENUS TRECHALEA (TRECHALEIDAE)

231

dates Karsch. (5) The “Typus” label was added
after Koch, because it lacks the perforation and
does not add proof that Koch made this desig-
nation himself about this particular specimen.
The conclusion that this specimen was the type
could have been based on the assumptions of
Karsch and may have actually been added, in-
correctly, by him. (6) Karsch’s conclusion was
apparently promoted by the lack of other ma-
terial available to him for comparison combined
with an overemphasis on the distinct eye and
tarsus characters that led him to make the state-
ment that “2006” agrees exactly with the de-
scription.

The partially insect-eaten, dried, and presum-
ably faded specimen after 30 years must have
lost much of the detail, especially pattern, of the
fresh specimen. Thus, the conclusion of Karsch
may have been influenced by the combination
of a process of elimination, misleading labels,
and an overemphasis on the few distinct char-
acters remaining on the specimen.

The above evidence leads me to conclude that
“2006” is not the holotype specimen of Triclaria
longitarsis designated by Koch. With this con-
clusion, one is still left with the problem about
the status of the nomenclature of the genus and
thus of the family. In order to provide stability
of the names, I make certain assumptions based
on the examination of a considerable amount of
material.

I assume that the type specimen is lost and
that Colombia was the actual type locality as
stated by Koch. Furthermore, I assume that the
most probable part of the country where the spec-
imen was taken is that area of northern Colombia
extending from the Atlantic coast to Bogota, the
most likely area that a European collector might
visit. Thus, the presumed type locality is in the
coastal drainage region rather than in the Am-
azonian drainage area. As explained later, these
two drainages have distinct species of Trechalea
which fall into the size range stated by Koch.
Therefore, the Trechalea species found in the
coastal area, is assumed to be T. longitarsis.

Status of unrevised “pisaurid” genera in the
Western Hemisphere.— Since this work brings
about some reassignments of a major group of
genera from the Pisauridae (or Dolomedidae) to
the Trechaleidae, it is appropriate now to sum-
marize the familial placement of the remaining
unrevised “pisaurid” genera in the Western
Hemisphere. The remainder of this section is
devoted to my determination of the disposition

of these genera based on an examination of the
available types.

Tunabo Chamberlin (1 9 1 6) is based on an im-
mature male holotype and is a lycosid (confirmed
by H. Levi pers. comm.) but was regarded as a
zorid by Lehtinen (1967) and was therefore listed
in the latter family by Brignoli (1983) and Plat-
nick (1989). Sisenna Simon (1898a) is a junior
synonym of Architis and is treated elsewhere
(Carico 1993). Two species described by Key-
serling in the genus Tetragonopthalma Karsch
(1878), T. granadensis (1877) and T. obscura
(1891), are in the lycosid genus Porrimosa (Ca-
pocasale 1982). Aglaoctenus Tullgren 1905 is a
junior synonym of Porrimosa (NEW SYNON-
YMY), a decision based on an examination of
the holotype of Aglaoctenus bifasciatus Tullgren.
The genus Dyrinoides Badcock (1932) was de-
scribed for two species which are represented by
tiny, unidentifiable spiderlings; however, they
may be trechaleids. Thaumasia Perty (1833) is
a pisaurid.

The Mello-Leitao genera Xingusiella (1940a)
and Demolodos (1943) (Demelodos [5/c] Mello-
Leitao 1943; Demolodes [5ic] Roewer 1954; Brig-
noli 1983; Platnick 1989) are each represented
by a single species, and their types are lost in the
Rio de Janeiro museum and are not available
for examination.

A species ascribed to Nilus (O. Pickard-Cam-
bridge 1876), N. amazonicus, was described by
Simon (1898b) from the Amazon region of Bra-
zil. The type is a large, recently molted juvenile
of unknown sex which is unlike any known
American pisaurid and is not a trechaleid. Since
the genus is otherwise known only from Africa
eastward through Australia and Simon described
many spiders from other continents in his 1898b
publication, it is likely that this specimen was
from another area and its locality is in error.

Two species of a primarily African and Asian
genus, Hygropoda Thorell (1895), were described
by Simon (1898a) from South America. Exam-
ination of the types of these species, H. andina
and H. venezuelana, shows that they actually be-
long to the genus Paradossenus.

Ancylometes Bertkau (1 880) bears some ctenid
anatomical characteristics, but it is probably a
pisaurid because of its clearly pisaurid behavior
(Merrett 1988). A study of its relationships should
be conducted, but is beyond the scope of this
paper.

Status of various pisaurid species assigned to
revised genera of the Western Hemisphere.—

232

THE JOURNAL OF ARACHNOLOGY

During this study, no specimen of the widely
distributed genus Dolomedes has been found from
anywhere south of the Yucatan Peninsula. This
confirms the opinion of Carico (1973) that the
genus has only a Nearctic distribution in the
Western Hemisphere with the exception of areas
of southern Mexico. There are, however, a few
species in the older literature described from
South America which I have considered. Gie-
beTs (1863) description of the eyes of his Do-
lomedes intermedins from Colombia clearly sug-
gests that it is not a Dolomedes. I cannot
determine its genus, however, because the ho-
lotype has not been found. The holotype of Do-
lomedes albicoxa Bertkau (1880) is lost but a
reading of the description indicates that this spe-
cies is not a Dolomedes because of the strongly
procurved anterior eye row and relative sizes of
the eyes. It may be a Porrimosa or a species of
Architis. Although the holotype of Dolomedes
pullatus Nicolet (1849) is missing, the descrip-
tion of the color pattern suggests to me that this
species may be a Thaumasia. The holotype of
Dolomedes elegens Taczanowski (1873) is miss-
ing, but the description of the color pattern gives
very good indication that it is a Thaumasia.

Thanatidius spinipesF. O. Pickard-Cambridge
(1903: 1 56, 1 57) is a junior synonym of Staberius
spinipes (Taczanowski). NEW SYNONYMY. All
of the species of Thanatidius listed by Bonnet
(195 5“ 1959) have been synonymized. The genus
was synonymized earlier by Carico (1972) to the
entirely Nearctic genus Pisaurina.

Previous attempts at higher taxonomy. —Re-
moval of these seven genera from the Pisauridae
{sensu Simon) {Trechalea, Hesydrus, Syntre-
chalea, Dossenus, Paradossenus, Dyrines, Enna)
into a monophyletic family is accompanied be-
low by an historical examination of previous
schemes of higher taxonomy. At least three au-
thors have made a significant attempt to name
families or subfamilies in which these genera were
included.

Simon’s (1 898a) placement of Trechalea, Hes-
ydrus, Dossenus, Drances (= Dyrines), and Hy-
gropoda into ‘"Dolomedese” (along with other
genera), indicates his concept that these trechal-
eid genera are closely related. However, only the
American species of Hygropoda are trechaleids
as indicated above.

Roewer (1954) assigned the genera Dossenus,
Dyrines, Hesydrus, Paradossenus, Syntrechalea
and Trechalea to his Thaumasiinae. However,
he assigned Enna to his Pisaurinae.

Lehtinen’s (1967) attempt to place these gen-
era into subfamilies and into new families re-
sulted in the following: Trechalea (“..Lycosoidea
incertae sedis; probably related to Zoridae, es-
pecially Neoctenus.'"), Dossenus (Dolomedidae),
Dyrines (Dolomedidae), Hesydrus (Dolomedi-
dae), Paradossenus (Dolomedidae), Syntrechal-
ea (“Position obscure, but evidently a represen-
tative of Lycosoidea.”). Additionally, genera
which were previously assigned to Pisauridae are
assigned as follows: Xingusiella (Amaurobiidae:
Rhoicininae), Aglaoctenus (Dolomedidae) and
Dyrinoides (Lycosoidea incertae sedis).

STRUCTURE OF THE GENITALIA IN
TRECHALEA

Male palpus.— The basic structure of the male
palpus of Trechalea was described by Sierwald
(1990) in her survey of various pisaurid genera.
Many of her conclusions and abbreviations con-
cerning structure are used here; a more detailed
description of the anatomy is available in her
paper. Additional details were pointed out by
Dondale (pers. comm.).

Basically, there are four, major, articulating,
sclerotized elements which make up the bulb of
the male palpus, each separated by membrane(s)
(Fig. 7). Beginning at the base is the 1) subte-
gulum (st), attached to the cymbium (c) by a large
membrane, the basal hematodocha (bearing
within its wall the sclerotized petiolus). Next,
moving distally, one encounters the ring-like 2)
tegulum {t), described by Sierwald (1 990) as hav-
ing several loops or “switchbacks” of the sperm
duct. Attached to the membrane distal to the
tegulum is the large, conspicuous, 3) median
apophysis (ma), and the 4) embolus (e) which is
a thin, curved structure with a broad, subdivided
base {embolic base, eb; [perhaps incorporating
the terminal apophysis]). The embolic base is
partly subdivided by a narrow membrane. Of
these, only the embolus is not usually visible
when viewed ventrally in the unexpanded state.

The median apophysis is a large, complex
structure that appears to have a unique form in
the trechaleids (Sierwald 1990) and is located
distally on the bulb where it characteristically
occupies about a fourth to a third of the bulb
mass. On its dorsal side is a deep, narrow, lon-
gitudinal groove, the embolic groove (eg), in which
the thin part of the embolus rests and is appar-
ently comparable to that found in some lycosids
(Dondale & Redner 1983b; Roth 1985 [key]).
This groove (independently discovered by Don-

CARICO~THE GENUS TRECHALEA (TRECHALEIDAE)

233

Figure 7. -=Trechaleid palpus anatomy based on Tre-
chalea longitarsis, ventral view. Embolus with its large
embolic base in shading behind (dorsad) and showing
the narrow part contained in the embolic groove of the
median apophysis. Abbreviations in methods section.

dale, pers. comm.) continues distally into a curved
or hooked projection arising from the dorsal di-
vision {dd), which probably serves as a guide {g)
for the embolus during intromission (as is the
case in some lycosids [Dondale & Redner 1 978])
but seems not to support or protect the embolus
at rest as with, respectively, the fulcrum and con-
ductor found in, e. g., Dolomedes. In addition to
the dorsal division, there is the conspicuous sub-
division on the ventral side {ventral division, {vd)
which varies in size and shape. The shape and
position of the projections arising from these di-
visions are highly genus- and species-specific and
are given special attention in the description and
figures for each genus and species.

The palpal tibia {ti) bears a distal retrolateral
apophysis {rta), which in some genera may be
subdivided into two parts: the ental division {end)
which is surrounded by membrane (defined be-
low), and the ectal division {ecd) usually located
laterally and proximally to the ental division.

The size and shape of the retrolateral apophysis
is genus- and species-specific. In addition to the
retrolateral apophysis, there is a feature of the
ventrodistal margin of the tibia in which the bor-
der folds down upon itself forming a depression
down into the distal end of the tibia. This pit
thus formed has been used as a taxonomic char-
acter of some importance by some workers (Sier-
wald 1990; Griswold 1993). See the section be-
low on character analysis for further discussion
of this feature and additional features of the tibia,
ventrodistal rim (vr), and ventrodistal protuber-
ance {vp), and ventral cymbio-tibial membrane
{vcm).

Sierwald (1990) identified in the palpal bulbs
of species of Trechalea and Paradossenus addi-
tional structures such as a conductor, terminal
apophysis, etc., which are not used here for di-
agnostic purposes. Refer to her paper for a dis-
cussion of these details.

Female epigynum.— A thorough study of the
homologies of the structures of the female gen-
italia is beyond the scope of the present analysis
and the terminology used here is adapted from
the work of Sierwald (1989) on American pi-
saurid genera. Emphasis is here directed to iden-
tification of major components and unique fea-
tures of the genitalia of only Trechalea for
diagnostic purposes. A preliminary survey of
various trechaleid genera reveals a considerable
range of variation, and each one will be treated
individually later. Thus, broad generalizations
concerning the family or any assortment of gen-
era within it are left to work in progress. For
descriptive purposes, terms are coined for use in
Trechalea and may or may not be applicable to
other members of the family.

The epigynum is generally heavily sclerotized
{T. trinidadensis excepted), dark and opaque, with
some components fused together, which makes
detailed observation difficult without initial
clearing. The external epigynal plate {ep) (Fig. 8)
is composed of four regional elements, two lat-
eral lobes (//) which comprise the elevated por-
tions on either side, an anterior field {aj) which
occupies most of the anterior half of the epigyn-
um and is usually continuous with, but not al-
ways distinct from, the lateral lobes. The middle
field {mf) is a distinct, posterior, median com-
ponent set off from the other components by
furrows and/or the posterior margin of anterior
field {pma). (The “external, outer lateral margin
of the epigynal fold” described by Sierwald (1989)
for pisaurids appears in trechaleids to be round-

234

THE JOURNAL OF ARACHNOLOGY

ed, less distinct and to lie out of sight under
[dorsad to] the mf.) The relative shapes and di-
mensions of these components tend to be spe-
cies-specific and are illustrated.

Internally (Figs. 9, 10), on each side, a gen-
erally large, voluminous copulatory duct (cd) arises
from the anterior portion of the epigynal fold
{epf) at the copulatory opening {co) and curves
inward and posteriorly where it gives rise to a
spermathecum {s) composed of a terminal en-
largement {head of spermatheca, hs) at the end
of a stalk {stalk of spermatheca, ss), most of which
lies fused against the copulatory duct. A second
diverticulum, called the copulatory duct divertic-
ulum {cdd) arises secondarily from the copula-
tory duct. Although the function of this structure
is unknown, its presence has been noted by Gris-
wold (1993; ‘Tobate spermathecal base”, char-
acter #39). My interpretation of its relative po-
sition (between and fd), however, departs from
his. I have been unable to find this structure in
all trechaleids, except, e. g., a species of Dossenus.
A fertilization duct (fd) continues from this junc-
tion to meet the oviduct {o) on its ventral side.
An internal fold of the anterior field {ifa) in the
shape of a thin flap (called “wing” by Sierwald)
occurs in various shapes and is most conspicuous
laterally where it joins the cd.

CRITICAL REVIEW OF SOME

TAXONOMIC CHARACTERS USED IN
RELATED FAMILIES

Since I reintroduced the family Trechaleidae
(Carico 1986), it has been the independent opin-
ion of Sierwald (1989, 1990), and Griswold
(1993), that the Trechaleidae (or the "'Trechalea
genus-group”) deserves family status and has as
its nearest relatives the families Pisauridae and
Lycosidae, with Lycosidae the closer. Codding-
ton 8l Levi (1991) adopted the conclusions of
Griswold and used the family name Trechalei-
dae. Additionally, Dondale (1986) emphasized
in Trechalea its unique and lycosid-like features
in his description of the subfamilies of the Ly-
cosidae.

The Pisauridae, as currently constituted, ap-
pears to be a very complex group and may not
be monophyletic. Sierwald, in her excellent stud-
ies of the American pisaurid female copulatory
organs (1989) and male palpal organs (1990),
stated that the known genera are apparently poly-
phyletic, a conclusion with which I agree. In con-
trast, however, monophyly was assumed for Ly-
cosidae by Dondale (1986), who proposed a group

Figures S~10.— Trechalea epigynum anatomy based
on T. longitarsis: 8, epigynal plate, ventral view; 9,
internal structures, dorsal view; 1 0, diagram of relative
positions of internal structures. Abbreviations in meth-
ods section.

of synapomorphies for that family and proceed-
ed further to define a number of subfamilies.

Griswold (1993) presented a wide-ranging but
preliminary analysis of the Lycosoidea and in-
cluded as exemplars in his cladogram members
of the Pisauridae {Dolomedes tenebrosus Hentz,
Pisaura mirabilis (Clerck)), Lycosidae {Lycosa
helluo Walck., Sosippus placidus Brady), and the
Trechaleidae {Trechalea sp.). The controversial
genus Rhoicinus (Rhoicininae), considered a pos-
sible trechaleid by Sierwald ( 1 990), was included.
In his cladogram, a total of 20 apomorphies
emerged to distinguish the six exemplars of these
three families.

Both Sierwald (1990) and Griswold (1993)
considered rhoicinines to be related to the tre-
chaleids. The latter author included an unde-
scribed Rhoicinus in his cladogram which
emerged as a sister group to his genus Trechalea
sp. and therein implied the two could be consid-
ered members of the Trechaleidae. Because of
the problems historically in assigning a family
for rhoicinines (Platnick 1979), the perceived
weakness of Griswold’s synapomorphies (dis-
cussed below), and the diversity of characters
among the various genera currently assigned to
the group (Exline 1960), I do not include rhoi-
cinines in the Trechaleidae. However, I do not
exclude the possibility of the family being broad-
ened later to include rhoicinines as well as other
genera, e. g., Shinobius of Yaginuma (1991).

In any case, the conclusions here must be con-
sidered provisional pending the outcome of on-

CARICO--THE GENUS TRECHALEA (TRECHALEIDAE)

235

going studies by various workers who are con-
tinuing to revise the genera within these families.
Such work is required to reveal consistent char-
acters which contribute to greater confidence in
whatever conclusion is reached about systematic
relationships. Therefore, no cladistic analysis is
offered here.

Following is a discussion of the taxonomic
characters used to distinguish the families Tre-
chaleidae relative to Pisauridae and Lycosidae
used by recent workers (Dondale 1986; Sierwald
1989, 1990; Griswold 1993) including some of-
fered here for the first time.

PLE situated behind PME to form a third row
of eyes.— This character is a traditional one used
to separate lycosids from the pisaurids. It is men-
tioned by Dondale (1986) and Sierwald (1990),
and Griswold (1993).

Retrolateral apophysis on the male palpal tib-
ia.—This is also a long-standing character used
to separate pisaurids and lycosids and is men-
tioned by Dondale (1986), Sierwald (1990) and
Griswold (1993). Separately, Dondale (1986) as-
sumed that the loss of a retrolateral apophysis
in the lycosids is a derived state. It is present in
Trechaleidae and Pisauridae.

Median apophysis position on bulb.— This
character is offered here for the first time and
refers to the position of the median apophysis
on the palpal bulb as viewed from the ventral
side. The position is either somewhere on the
ventral face, e. g., Lycosidae and Pisauridae, or,
in the case of trechaleids, enlarged and occupying
the distal fourth or third portion of the bulb mass.

Folded ventrodistal rim of the male tibia.—
This is a reinterpretation of a so-called “pit”
located ventrally at the apical end of male palpal
tibia reported by others. Attention to a pit per-
haps began with a feature noted in Rhoicinus by
Exline (1960) who described it as an “unscler-
otized pit, surrounded by a fairly high, rebor-
dered, chitinous ring.” A “membranous pit that
accompanies the tibial apophysis” was noted lat-
er by Sierwald (1990) also in Trechalea, and she
proposed that this may be synapomorphous for
the two genera. Recently Yaginuma (1991, pp.
2, 4) noted a similar feature in a Japanese spider
Shinobius orientalis (Yaginuma) and thus as-
signed this spider to the Rhoicininae in the Pi-
sauridae with reservation. Griswold (1993) used
a similar character in his analysis of lycosoids
and named it as, “Male palpal tibia with retroap-
ical cuticle unsclerotized.” He used this as an
important character to link both Rhoicinus and

Trechalea together in his cladogram with the im-
plication that the former genus was a trechaleid.

An examination of Rhoicinus, however, re-
veals that there is indeed a “ring” present, but
the cuticle enclosed is actually sclerotized and
the whole structure is situated on the sclerotized
portion of the tibia. This ring and its enclosed
cuticle are not synonymous with the “membra-
nous pit” in the trechaleids as reported by these
latter workers and is actually homologous with
the retrolateral apophysis {rtd) of trechaleids and
pisaurids.

In the Trechaleidae (as here defined), another
type of pit (not found in Rhoicinus) is actually
formed from the ventral cymbio- tibial membrane
{vcm, Fig, 7) of the male palpal tibia. This mem-
brane, found in all spiders, may be broad and
flexible to permit a wide arc of flexion at the
articulation between the cymbium and tibia. It
is not uncommon to note a concavity of varying
degrees in this soft membrane in spiders of all
three families. Since this concavity results from
varying conditions of preservation, it is thus not
taxonomically significant. The pit of Trechalea
referred to by the previously-mentioned authors,
however, is due less to a condition of the mem-
brane itself, but rather to the morphology of the
adjoining sclerotized ventrodistal rim (vr) (Fig.
7) of the tibia. In the trechaleids and, to a lesser
degree, the pisaurids, there is a folding back
downward of the rim into the inside of the cy-
lindrical tibia, forming a noticeable depression
in the ventral and retroventral end of the tibia
because the vcm typically arises from this deep
recess. In these same two families, the rim is
usually also molded into a ventrodistal protuber-
ance (vp) and thus forms a characteristic shape
to the tibial rim from ventral view. No such rim
or protuberance is found in American lycosids.
What makes the so-called pit more noticeable in
the trechaleids is that the distal, gaping opening
of the cylindrical tibia is larger, and the dorso-
ventral elongation is accentuated by the vp.

Method of egg sac transport.— I depart from
other workers in the interpretation and use of
this character. The traditional approach has been
to state the choice between “carrying egg sac in
chelicerae” or “carrying egg sac on spinnerets”.
This is similar to Griswold’s (1993) character
#68 and is mentioned by Sierwald (1990) and
Dondale (1986). This seems an oversimplifica-
tion and inaccuracy because in all three families
the females carry the egg sac attached to the spin-
nerets. Therefore, the pisaurids, which alone em-

236

THE JOURNAL OF ARACHNOLOGY

ploy the chelicerae in transport of the egg sac,
also carry it simultaneously with the spinnerets;
this important fact is often overlooked. There
are examples of carrying the egg sac in the che-
licerae only in such diverse families as Pholcidae
(Kaston 1948, p. 67) and Synotaxidae {Mangua
gunni Forster et al. 1990, p. 76) and thus is not
equivalent to the case with pisaurids.

Egg sac transport by spinnerets alone is com-
monly associated with lycosids but is also re-
ported in trechaleids (Simon 1898a, Trechalea,
Hesydrus, etc.; Van Berkum 1982, Trechalea ex~
tensa’, Carico et al. 1985, Trechalea amazonica;
Carico pers. obs., Trechalea extensa, T. gertschi,
Hesydrus habilis).

Therefore use of the spinnerets to carry the egg
sac is apparently plesiomorphic for all three fam-
ilies. The apomorphy customarily stated as “egg
sac transported in chelicerae” in the context of
the current discussion is more accurately restated
by combining the two attachments, as is the case
in the pisaurids.

Structure of egg sac seam.— It is assumed that
during the construction of the egg sac in all three
families, a seam is made because of the typical
sequence of egg laying, i. e., lower sheet construc-
tion, followed by egg mass deposition and upper
sheet construction. Specihcally considered here,
however, is the seam appearance when the egg
sac is full of eggs (not hatchlings) after construc-
tion is complete. The appearances differ in all
three families: seam apparent on spherical egg
sac without “skirt” (lycosids), seam apparent as
a rim on discoid egg sac with a “skirt” (trechal-
eids) (Fig. 6), seam not apparent on spherical egg
sac (pisaurids).

Method of maternal care of young after emer-
gence from egg sac.— Some form of maternal care
is characteristic of all three families, but each
family has a distinct method of caring for young.
Trechaleids transport young on an empty egg sac.
The young congregate on the empty egg sac which
the female continues to carry until after the young
disperse. However, as has been noted by Carico
et al. (1985), transport of young on the abdomen
has been observed in the trechaleids. Caution
must be used in applying this character to the
trechaleids because, in my observations, the
presence of some spiderlings on the mother’s ab-
domen seems only to occur when they are crowd-
ed and pushed off the egg sac. Therefore, young
on the abdomen of trechaleids appears incidental
and not synonymous with young transport on
the abdomen in lycosids. Transport of young on

the mother’s abdomen is a character which is
typically associated with lycosids, and in Don-
dale’s (1986) list it is associated with transport
of the egg sac on the spinnerets. It is important
to note that Yaginuma (1 99 1) reported that three
Japanese lycosids, as well as his rhoicinine, Shi-
nobius, carry the young on the egg sac. The nurs-
ery web is one of the most traditional characters
of the family Pisauridae and has promoted the
term “nursery-web spiders” for this group. This
is mentioned by Carico (1973) and Sierwald
(1990). It is Griswold’s (1993) character #67.

Binding of female’s front two pairs of legs by
male during copulation,— Use of a bridal veil of
silk by the male during copulation is widespread
though apparently uncommon in spiders (Schmitt
1992). In the families here under consideration,
only the pisaurids Pisaurina mira (Bruce & Car-
ico \9%S),Ancylometesbogotensis{}AQrvQXX 1988),
and Dolomedes triton (Wojcicki 1990) are re-
ported to use this behavior. In these three genera
a unique type of veil is used, which has the struc-
ture spun only over the first two pairs of legs.

Web retreat. —Trechaleids are not known to
build webs or retreats of any kind. Some lycosids
are known to build webs with tubular retreats
(Brady 1962). A distinctive structure of the pi-
saurid retreat, when present, is the characteristic
short tube with the openings distinctively flared
that has been found in another disparate group
of pisaurid genera: Pisaura mirabilis (Lenler-Er-
ikson 1969), Pisaurina mira (Carico 1985), and
Architis nitidopilosa (Nentwig 1 985). This retreat
may be found only as a juvenile web in P. mi-
rabilis and P. mira or as a more extensive web
in the adult A. nitidopilosa. The spider rests with
its body at right angles to the axis of the tube
with its legs resting in the flared openings.

Reattachment of egg sac.— Female lycosids and
pisaurids are well-known to reattach the egg sac
if it is dislodged. In the trechaleids I have ob-
served, Trechalea gertschi, T. extensa, and Hes-
ydrus habilis, the egg sacs were never reattached
to the spinnerets. On examination of upper sur-
faces of several egg sacs of these and other spe-
cies, only a single, conspicuous attachment disc
is ever present, although a zig-zag pattern of these
“carrying threads” across the surface may be pro-
duced before final attachment.

Additional characters.— In Griswold’s (1993)
extensive character set, 14 additional new mor-
phological synapomorphies were utilized in his
cladistic analysis to distinguish these three fam-
ilies from each other and from the other families

CARICO=-THE GENUS TRECHALEA (TRECHALEIDAE)

237

of the Lycosoidea. Because their use has been
applied only to the six exemplars and their uni-
versalities have not been tested in other genera
of these families, they are not discussed further
in the present study.

Sierwald (1 990) referred to the ‘'small, reduced
conductor” (p. 51) as a synapomorphy for the
""Trechalea genus-group”. Because this structure
in pisaurids ranges in size from the “large and
strongly sclerotized. . .conductor in Tinus. . .”to
“a low hump or prominence, as the conductor
in Architis. . .” (p. 18) it is a difficult character
to use. Her observation of “a median apophysis
with two branches. . .” in this genus-group (p.
50) is difficult to apply because the median
apophyses of some trechaleids have more than
two while many lycosids have two branches, al-
though they may not be of the same form as in
trechaleids. She also observed that the “sperm
duct with several switchbacks. . .” (p. 50) is a
synapomorphy. If a switchback is defined as loop
wherein the tube reverses its direction, then it is
present in Trechalea and some other larger tre-
chaleids, but smaller trechaleids such as Dyrines
(p. 35) have only one or none in an undescribed
genus. There are a number of undulations of the
sperm duct which are also found in Staberius, a
pisaurid (1990, fig. 25) and some local lycosids.

Genus Trechalea Thorell

Triclaria C. L, Koch, 1848:101, (type species by orig-
inal designation, Triclaria longitarsis C. L. Koch).
Trechalea Thorell, 1869:37, {nomen novum for Tri-
claria, preoccupied). Simon, 1898a:279-281, 304=
312,315.Roewer, 1954, 2a:142. Bonnet, 1955=1959,
11:4678. Lehtinen, 1967:379 (genus incertae sedis).
Brignoli, 1983:461. Platnick, 1989:398.
Perissoblemma 0= Pick.-Cambridge, 1881:773. First
synonymized by Simon 1898a:31 1.

Diagnosis.— can be distinguished
from Syntrechalea by the fewer number of ven-
tral tibial macrosetae pairs (4-6) while the latter
has about twice as many. Also, Trechalea can be
distinguished from Hesydrus, its closest relative,
by having only the tarsi flexible while the latter
genus has also the metatarsi flexible. The middle
field of the epigynum in Trechalea is relatively
short and lobe-like while that of Hesydrus tends
to be longer and scape-like.

Description.— Carapace moderately low, ce-
phalic area relatively distinct, AE row straight or
slightly recurved when seen from above. Retro-
marginal cheliceral teeth ranging from 3-5, vari-
able in size and distance. Leg relative lengths

variable but III always shortest and IV almost
always longest, only tarsi flexible, scopula may
be present on metatarsus and tarsus, all claws
dentate, pairs of macrosetae on ventral side of
tibia I ranging from 4-6.

Male palpal bulb median apophysis with acute,
conspicuous guide, ventral division variable but
thickened, tibial retrolateral apophysis divided
with ental division distinct, often lobed and part-
ly surrounded by ventral cymbio-tibial mem-
brane, ectal division conspicuous and in various
forms.

Female epigynal plate with middle field about
as wide as long or only slightly longer than wide,
usually widest anteriorly (Fig. 8).

Distribution.— Found from the Gila River
drainage system of central Arizona, United States
southward to the state of Rio Grande do Sul,
Brazil.

Natural history. —Most members of this genus
are found on the margins of freshwater streams
and lakes. Many species are restricted to a par-
ticular river drainage system.

Disposition of nominal species of Trechalea.—
Trechalea ornata Mello-Leitao (1943), Trechal-
ea wygodzinski Soares and Carmargo (1948),
Trechalea keyserlingi F. Pickard-Cambridge
(1903), and Trechalea biocellata Mello-Leitao
(1926) are being transferred to new trechaleid
genera.

Trechalea reimoseri Caporiacco (1947) is rep-
resented by two syntypes, neither of which is a
Trechalea, each belonging to diflferent genus. The
female is a Syntrechalea and the male is in an
undescribed trechaleid genus. Trechalea protenta
Karsch (1879) is transferred to Paradossenus.
Trechalea thomisiformis (O. Pickard-Cam-
bridge, 1881) is a very small juvenile lycosoid
and appears not to be a trechaleid, but of doubt-
ful genus. Trechalea monticola Chamberlin
(1916) is transferred to Hesydrus.

Unsuccessful efforts were made to obtain cer-
tain holotypes of various species described by
Mello-Leitao. The provisional determination of
their status is made as follows: both Trechalea
syntrechaloides Mello-Leitao (1940b, type
#41476) and T. limai Mello-Leitao (1940b) ap-
pear from drawings and descriptions to belong
to a new unnamed trechaleid genus. T. aurantia
Mello-Leitao (1942) may be a Hesydrus. The ge-
neric status of T. numida Mello-Leitao (1943)
and T. langei Mello-Leitao (1947) cannot be de-
termined from their descriptions.

Note.— While working on the genus Hesydrus,

238

THE JOURNAL OF ARACHNOLOGY

Table 1. — Eye measurements for species of Trechalea. Measurements are dimensions with outer limits of
entities included. AE row == width of anterior eye row, PE row = width of posterior eye row, OQA = width of
ocular quadrangle anteriorly or width of anterior median eyes, OQP = width of ocular quadrangle posteriorly
or width of posterior median eyes, OQH = height of ocular quadrangle or height of anterior median eye and
posterior median eye, PLE = diameter of posterior lateral eye, PME = diameter of posterior median eye,
ALE = diameter of anterior lateral eye, AME = diameter of anterior median eye, PLE-PME = interdistance
between posterior lateral eye and posterior median eye, PME-PME = interdistance between posterior median
eyes, ALE-AME = interdistance between anterior lateral eye and anterior median eye, AME-AME = interdistance
between anterior median eyes. Ion = T. longitarsis, cez = T. cezariana, ama = T. amazonica, pau = T. paucispi-
na, con = T. connexa, ext = T. extensa, ger = T. gertschi, mac = T. macconnelli, bol = T. boliviensis, lorn = T.
lomalinda, tri = T. trinidadensis. Measurements in millimeters.

Species/

sex

AE

row

PE

row

OQA

OQP

OQH

PLE

PME

ALE

AME

PLE-

PME

PME-

PME

ALE-

AME

AME-

AME

lon5

1.88

3.56

1.08

1.78

1.70

0.76

0.74

0.32

0.49

0.58

0.27

0.15

0.25

Ion?

1.94

3.70

1.14

1.84

1.70

0.80

0.77

0.26

0.30

0.70

0.40

0.10

0.27

cez(3

1.99

3.70

1.16

1.78

1.60

0.88

0.76

0.32

0.51

0.67

0.35

0.12

0.27

cez9

2.10

4.08

1.25

1.89

1.75

0.90

0.82

0.34

0.55

0.74

0.40

0.13

0.29

mac<3

2.08

3.78

1.22

1.88

1.80

0.86

0.80

0.35

0.60

0.62

0.41

0.14

0.28

mac9

2.00

3.70

1.19

1.85

1.78

0.85

0.82

0.30

0.53

0.68

0.40

0.15

0.25

pau3

1.79

3.29

1.09

1.60

1.48

0.70

0.70

0.29

0.49

0.56

0.40

0.10

0.23

pau9

1.90

3.45

1.14

1.65

1.52

0.75

0.70

0.31

0.49

0.63

0.40

0.09

0.25

con6

1.40

2.58

0.83

1.25

1.11

0.56

0.50

0.22

0.34

0.45

0.30

0.07

0.22

con9

1.55

2.88

0.91

1.38

1.26

0.58

0.60

0.26

0.38

0.60

0.28

0.08

0.25

ext(5

1.71

3.25

1.01

1.61

1.42

0.78

0.70

0.28

0.40

0.50

0.32

0.10

0.23

ext9

1.70

3.26

1.02

1.60

1.44

0.74

0.68

0.30

0.44

0.55

0.36

0.10

0.20

ger5

1.48

2.80

0.85

1.32

1.15

0.55

0.50

0.25

0.34

0.52

0.30

0.09

0.21

ger9

1.61

2.98

0.96

1.40

1.25

0.65

0.50

0.25

0.40

0.58

0.33

0.10

0.25

amacJ

1.35

2.45

0.83

1.27

1.10

0.55

0.50

0.19

0.33

0.43

0.30

0.10

0.19

ama9

1.45

2.70

0.90

1.38

1.20

0.60

0.55

0.20

0.35

0.40

0.30

0.09

0.18

boU

1.07

2.13

0.57

1.15

0.94

0.47

0.48

0.18

0.23

0.32

0.27

0.08

0.16

bol9

1.05

2.18

0.58

1.15

0.96

0.50

0.47

0.18

0.23

0.37

0.27

0.07

0.16

lom(3

1.27

2.44

0.72

1.27

1.12

0.55

0.54

0.20

0.28

0.38

0.30

0.10

0.23

lom9

1.30

2.58

0.72

1.33

1.16

0.57

0.57

0.21

0.30

0.42

0.31

0.05

0.20

tri9

1.03

2.05

0.65

1.05

0.91

0.45

0.45

0.16

0.27

0.37

0.22

0.05

0.15

I was impressed with its similarity with Tre-
chalea and hold to the possiblity that later the
two genera may be found to be congeneric. Cur-
rently, however, because of lack of sufficient males
in Hesydrus, I will keep them as separate genera
with the main distinction based on the flexible
tarsi (i. e., flexible metatarsi only in Hesydrus)
mentioned in the diagnosis above.

Trechalea longitarsis (C. L. Koch)
Figures 7, 11-15, 25; Map 1

Triclaria longitarsis C. L. Koch, 1848:65. (Holotype is
a female from Colombia deposited in the Museum
fur Naturkunde der Humbolt-Universitat, presumed
lost. The Specimen #ZMB 2006, listed as the ho-
lotype, is misidentified and is not the original type
specified by Koch [see discussion above]. A neotype
male is hereby designated from Quebrada Docordo,
ab. 110 km N of Palestina, Rio San Juan, Choco,

Colombia, collected 20-25 January 1971 by B.
Malkin and P. Burchard, deposited in the Field Mu-
seum of Natural History)

Trechalea longitarsis, Karsch, 1879:450. Roewer, 1954:

143. Bonnet, 1955-1959:4679.

Trechalea urinatorSiman, 1 898b:20. (Male and female
syntypes from Guayaquil, Depto. Loja, Ecuador in
the Museum National d’Histoire Naturelle, exam-
ined.) Roewer, 1954:143; Bonnet, 1955-1959:4679.
NEW SYNONYMY.

Diagnosis. —This is the only species of Tre-
chalea with four retromarginal cheliceral teeth.
Both sexes are also distinguished by the details
of the genitalia. The median apophysis of the
palpal bulb bears a distinct tubercle between the
guide and ventral division (Figs. 1 1 , 1 5), a feature
shared only with T. extensa (Figs. 20, 53). T.
longitarsis differs from the latter species by the

CARICO~THE GENUS TRECHALEA (TRECHALEIDAE)

239

Table 2. — Leg measurements in male of Trechalea
longitarsis.

Leg segment

I

II

III

IV

Femur

18.5

21.2

16.0

21.0

Tibia-patella

25.2

26.6

20.0

25.4

Metatarsus

19.0

21.5

16.6

24.8

Tarsus

12.2

13.9

11.0

15.5

Total

74.9

83.2

63.6

86.7

length of the cymbium which is about twice the
length of the bulb (Fig. 1 1).

Description.— Ma/^; (Rio San Juan, Depto.
Choco, Colombia). Carapace low, cephalic area
not elevated, length 11.1, width 9.5, medium
brown with submarginal bands distinct posteri-
orly, less so anteriorly, dark at lateral margins
and in eye region. Sternum light, unmarked,
length 5.2, width 5.1; labium dark brown, lighter
at distal margin, length 2.2, width 2.0. Clypeus
height 1.36, width 5.0. Anterior eye row straight,
a cluster of bristles posterior to each PLE, eye
measurements in Table 1. Chelicerae face dark,
clothed with light hairs mostly in proximal two-
thirds, an oblique groove above fang and a lon-
gitudinal Carina laterally on distal two-thirds, four
retromarginal teeth equidistant, subequal in size
except distal one slightly smaller. Legs IV-II-I-
III, measurements in Table 2, ventral macrosetae
pairs on tibiae are 1-5, II-5, III-5, IV-5. Color of
legs medium brown, unmarked. Abdomen length
9.5, hairy above, dark above and without distinct
pattern except for reticulated and striated distri-
bution of pigment, light ventrally.

Palpus (Figs. 11, 12), median apophysis (Fig.
15) with ma guide winged apically, and with a
distinct distal tubercle situated between the dor-
sal and ventral subdivisions; cymbium length/
palpal bulb length ratio 2.14. Tibial apophysis
(Fig. 25) with ecd flattened and hooked distally
and serrated along inner margin.

Female: (Rio San Juan, Depto. Choco, Colom-
bia). Carapace shape and color as in male, length
9.9, width 9.0. Sternum light, unmarked, length

5.8, width 5.2; labium color as in male, length
2.00, width 1.96. Clypeus height 1.41, width 5.15.
Anterior eye row straight, eye measurements in
Table 1. Chelicerae color and hair as in male.
Legs IV-II-I-III, measurements in Table 3; ven-
tral macrosetae pairs on tibiae 1-5, II-4, III-4,
IV-3. Color of legs as in male. Abdomen length

8.8, hairy above, color as in male except for in-
distinct pairs of light spots. Epigynum (Figs. 13,

Figures 1 1-1 4. —Genitalia of Trechalea longitarsis
(Choco, Colombia): 11, 12, right palpus; 11, ventral
view; 12, retrolateral view; 13, 14, epigynum; 13, ven-
tral view; 14, dorsal view. Scales in mm.

1 4) with the lateral margins mf almost parallel;
5 mostly fused to cd.

Variation.— The average carapace length of
eight males is 11.24 (range 10.8-11.8), and the
average carapace length of nine females is 10.31
(range 9.5-11.0). The average cymbium/palpal
bulb length ratio is 1.96 (range 1.81-2.19, n =
8).

Natural history.— Four egg sacs were found in
three collections dated January, June, and Oc-
tober with an average diameter of 16.8 (range
1 5.0-19.8). All egg sacs were typically disc-shaped
with both the top and bottom valves opaque and
brown-colored. Two egg sacs were still attached
to the spinnerets by threads. Apparently the prin-
cipal habitat is around streams because most col-
lection records refer to a river while one collec-
tion note stated explicitly, “near water”. Another
collection note stated, “en la selva”.

Distribution.— From northern Peru northward
to northern Colombia along the eastern river

240

THE JOURNAL OF ARACHNOLOGY

Table 3. — Leg measurements in female of Trechalea
longitarsis.

Leg segment

I

II

III

IV

Femur

18.3

20.5

16.0

20.6

Tibia-patella

23.2

25.6

18.5

23.6

Metatarsus

18.6

21.1

17.1

24.6

Tarsus

12.8

13.0

10.8

15.9

Total

72.9

80.2

62.4

84.7

drainages. There are no records of collections
from the Amazon or Orinoco basins. (Map 1).

Specimens examined. —COLOMBIA: Choco: Que-
brada Dorcordo, 1 10 km N of Palestina, Rio San Juan,
20-25 Jan. 1971 (B. Malkin & P. Burchard), 5<3 39 5
juv. (FMNH); Pangola, 40 km N of Palestina, Rio San
Juan, 14-18 Jan. 1971 (B. Malkin & P. Burchard), 29,
6 juv. (FMNH); Cauca: Quebrada, Huangui, Rio Saija
area, 100 m (B. Malkin), 13 29 6 juv. (FMNH); Mag-
dalena: Serr. Nueva Granada, Sierra Nevada de Santa
Marta, 24 April 1975 (J. A. Kochalka), 13 19 (JAK);
Santander: Rio Suarez, 800-1000 m, 11-17 Aug. 1946
(collector unknown), 23 2 juv. (AMNH). ECUADOR:
Pichincha: 1 2 km SW of Santo Domingo de los Co-
lorados, 4-8 April 1971 (B. Malkin), 29 (FMNH); El
Oro: Rio Colorado Pasaje, 4 Nov. 1942 (Walls), 1 juv.
(EXPE); Imbabura: Lita, Sept. 1984 (D. Bastidos), 13
(MECN); Canar: Yanayacu, 22 Sept. 1984 (R. Navar-
rette), 19 (MECN). PERU: Piura: Cana Dulce, Oct.
1943 (C. & E. Ewing), 13 (CAS).

Trechalea cezariana Mello-Leitao
Figures 16, 26, 35-38; Map 1

Trechalea cezariana, Mello-Leitao, 1931:12, fig. 2 (The
holotype is a female from Rio Cruz, near Gramado,
Municipio de Taquara, Est. Rio Grando do Sul, Bra-
zil, collected by Cezar Pinto, deposited in the Museo
Nacional, Rio de Janeiro, Brazil, examined.) Roew-
er, 1954:142; Bonnet, 1955-1959:4678.

Diagnosis. —The median apophysis of the pal-
pus is notched anteriorly (Fig. 1 6) and the ectal
division of the tibial apophysis is small, flat and
curved (Fig. 26). The epigynum is characterized
by the very wide and flattened middle field with
a median groove (Fig. 37).

Description.— (Lassance, Minas Gerais,
Brazil). Carapace low, length 9.65, width 9.35,
no distinct pattern, dark in eye region, a pair of
small depressions with dark hairs anterior to tho-
racic groove; sternum light, unmarked, length
5.0, width 4.6; labium dark reddish brown, light
distally, length 2.25, width 1.75. Clypeus height
1.14, width 4.25. Anterior eye row straight, eye
measurements in Table 1. Chelicerae dark red-

Map 1 . —Distribution of species of Trechalea in South
America. • = T. macconnelli, M= T. paucispina, ♦
= T. cezariana, A = T. longitarsis. Dashed lines in-
dicate major river drainage basins and/or continental
divides.

dish brown, clothed with setae on anterior sur-
face, oblique depression above fang, a longitu-
dinal Carina along distal third of anterolateral
margin, three retromarginal teeth of equal size
with gap between proximal two. Legs IV-II-I-III,
light and unmarked, measurements in Table 4,
ventral tibial spine pairs 1-5, II-5, III-3, IV-4.
Abdomen length, 8.0, marked with indistinct
pattern above, light ventrally.

Palpus (Figs. 35, 36), ma (Fig. 16) notched, rta
(Fig. 26) with end thickened and rounded, ecd
small, thickened but acute distally and directed
mediad.

Female: (Lassance, Brazil). Carapace low,
length 9.6, width 9.8 marked as in male; sternum

Table 4. — Leg measurements in male of Trechalea
cezariana.

Leg segment

I

II

III

IV

Femur

14.1

16.6

13.8

16.4

Tibia-patella

19.3

22.0

16.6

20.3

Metatarsus

14.8

17.3

13.2

19.4

Tarsus

8.8

9.3

7.8

11.0

Total

57.0

65.2

51.4

67.1

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

241

Figures 1 5“24.=- Scanning electron micrographs of median apophyses from right palps of species of Trechalea,
ventral views; 15, T. longitarsis (dd = dorsal division, g = guide, vd = ventral division); 16, T. cezariana; 17,
T. macconneiU; 18, T. paudspina; 19, T, connexa; 20, T. extensa; 21, T. gertschi, 22, T. amazonica; 23, T.
boliviensis; 24, T. lomaiinda. Scales: lOO^ti.

242

THE JOURNAL OF ARACHNOLOGY

Figures 25-34. —Scanning electron micrographs of retrolateral tibial apophyses from right palps of species of
Trechalea, ventral views: 25, T. longitarsis (ecd = ectal division, end = ental division); 26, T. cezariana; 27,
T. macconnelli; 28, T. paucispina\ 29, T. connexa\ 30, T. extensa; 31,7", gertschi; 32, T. amazonica; 33, T.
boliviensis; 34, T. lomalinda. Scales: 100^.

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

243

Figures 35-38. —Genitalia of Trechalea cezariana
(Minas Gerais, Brazil): 35, 36, right palpus; 35, ventral
view; 36, retrolateral view; 37, 38, epigynum; 37, ven-
tral view; 38, dorsal view. Scales in mm.

light, unmarked, length 5.15, width 4.9; labium
color as in male, length 1.35, width 1.8. Clypeus
height 1 .20, width 4.78. Anterior eye row straight,
eye measurements in Table 1. Chelicerae as in
male but without oblique depression, cheliceral
teeth as in male. Legs IV-II-I-III, measurements
in Table 5, light, unmarked ventrally but with
faint oblique lateral gray maculae, pairs of macro-
setae on venter of tibia 1-4, II-4, III-4, IV~4.
Abdomen length 12.25, marked with indistinct
pattern above, light ventrally.

Epigynum (Figs. 37, 38) heavily sclerotized and
dark, mf very wide, flattened with a median Ion-

Table 5. —Leg measurements in female of Trechalea

cezariana.

Leg segment

I

II

III

IV

Femur

13.5

15.8

13.2

15.7

Tibia-patella

18.5

21.7

16.5

19.7

Metatarsus

13.6

16.0

13.3

18.8

Tarsus

8.3

9.3

8.0

11.2

Total

53.9

62.8

51.0

65.4

Figures 39, 40.— Dorsal patterns of species of Tre-
chalea: 39, T. macconnelli; 40, T. boliviensis. Scales

in mm.

gitudinal groove; internal parts heavily sclero-
tized and fused together.

Variation.— The average carapace length of six
males is 10.02 (range 8.3-1 1.0) and the average
carapace length of seven females is 9.66 (range
8.1-10.3).

Natural history.— None of the collection re-
cords provide information about the type of hab-
itat. Two quite flat egg sacs of typical construc-
tion were found with females collected 25 March
and 4 October and were 19.25 and 18.5 respec-
tively.

Distribution.— The rather sparse number of
collections of this species indicate that the main
area of distribution is in the various tributaries
of the Rio de la Plata in southern Brazil. There
are additional records from the coastal areas of
Brazil in the states of Rio Grande do Sul and
Rio de Janeiro, and a single collection from the
upper tributary of Rio Sao Francisco in the state
of Minas Gerais. (Map 1).

Specimens examined.— BRAZIL: Minas Gerais: Sao
Gongalo das Tobacas, Lassance, 25 March 1925 (D.
M. Cochran), 2s 3$, 3 imms. (USNM); Rio de Janeiro,
Thayer Expedition, la (MCZ); Rio Grande do Sul: Re-
serva Biol6gica do Ibicui-Mirim, Santa Maria, 4 Oct.
1989 (N. Silveira), 19 (MCN), Itauba, Arroio do Tigre,
7 April 1978 (A. A. Lise), 2s (MCN), 11 April 1978
(H. Bischoff), 19 (MCN), 11 April 1978 (A. A. Lise),
la 19 (MCN). ARGENTINA: Misiones: Parque Na-
cional Iguazu, Sept. 1963 (M. E. Galiano), la 19 (MEG).
BOLIVIA: Santa Cruz: Estacibn el Portbn, Serrania
de Santiago, 24 Sept. 1955 (F. Azambuya), la (CAS).

244

THE JOURNAL OF ARACHNOLOGY

Table 6. — Leg measurements in male of Trechalea
macconnelli.

Leg segment

I

II

III

IV

Femur

20.4

22.3

16.6

21.0

Tibia-patella

25.3

26.8

19.0

25.2

Metatarsus

19.5

21.8

16.1

24.3

Tarsus

12.5

13.6

11.2

16.0

Total

77.7

84.5

62.9

86.5

Trechalea macconnelli Pocock
Figures 17, 27, 39, 41-44; Map 1

Trechalea macconnelli Vocqc]l 1900:67, 68, fig. 2e (The
holotype is a male from Mount Roraima [base 3500
feet] Guyana, collected by F. V. McConnell and J.
J. Quelch, deposited in The Natural History Muse-
um, London, examined). F. Pick. -Cambridge, 1903:
159, 163. Petrunkevitch, 1911:548. Roewer, 1954:
143. Bonnet 1955-1959:4679.

Trechalea ellacombei F. Pickard-Cambridge 1 903: 161,
162, pi. XV, fig. 6 (The holotype is a female from
Bergen-Dal, Surinam, collected May 1892 by E. W.
Ellacombe, deposited in The Natural History Mu-
seum, London, examined). Petrunkevitch 191 1:548.
Roewer 1954:142. Bonnet 1955-1959:4679. NEW
SYNONYMY.

Diagnosis. —Both sexes are distinguished by
the transverse band of white hairs dorsally near
the posterior apex of the abdomen (Fig. 39). The
males are also distinguished by the shape of the
median apophysis (Fig. 17) and the hook-like
configuration of the ectal division of the tibial
apophysis (Fig. 27). Both sexes have a distinct
black tip on the dorsal abdominal apex (Fig. 39).

Description. —Ma/e.' (Cusuimi, Pastaza, Ec-
uador). Carapace low, cephalic area somewhat
elevated, length 10.7, width 10.0, medium brown,
darker at margin and black in eye region, light
hairs on clypeus; sternum light, unmarked, length
5.4, width 5.2; labium dark brown, lighter dis-
tally, length 2. 1 5, width 1 .92. Clypeus height 1 .20,
width 3.8. Anterior eye row straight, eye mea-
surements in Table 1. Chelicerae face dark,
clothed with light hairs clustered in longitudinal
patches, an oblique groove above each fang and
a short longitudinal carina laterally, three retro-
marginal teeth equal in size with distal two closer
together. Legs IV-II-I-III, measurements in Ta-
ble 6, ventral macrosetae pairs on tibiae are 1-5,
II-5, III-4, IV-4. Color of legs generally medium
brown, lighter on ventral side of femora and
without distinct pattern. Abdomen length 9.0,
hairy above with reticulated pattern of dark pig-

Figures 41 --44. —Genitalia of Trechalea macconnel-
li: 41, 42, right palpus (Pastaza, Ecuador); 41, ventral
view; 42, retrolateral view; 43, 44, epigynum (Ron-
donia, Brazil); 43, ventral view; 44, dorsal view. Scales
in mm.

ment except with a medium light cardiac area,
a transverse light band near the posterior apex
and a patch of dark hairs apically; light ventrally.
Palpus (Figs. 41, 42, second male from same
locality), ^ of ma (Fig. 1 7) divided into two blade-
like carinae and rugose on some surfaces; ecd of
rta (Fig. 27) flattened and hooked at tip.

Female: (Jamari, Rondonia, Brazil). Carapace
shape as in male, light submarginal bands and
radiating light lines on darker background (Fig.
39), dark around eye region, length 9.7, width
9.6; sternum light, unmarked, length 5.2, width
4.9; labium color as in male, length 2.05, width
1.80. Clypeus height 1.15, width 4.40. Anterior
eye row straight, eye measurements in Table 1.
Chelicerae face dark, clothed with light hairs
clustered into longitudinal patches with long dark
hairs between. Legs II-IV-I-III, measurements in
Table 7, ventral macrosetae pairs on tibiae 1-5,
II- 5, III-4, IV-4. Color of legs generally dark with
distinct but irregular pattern particularly on pro-
lateral surfaces. Abdomen length 10.8, hairy
above especially laterally, bold pattern (Fig. 39)
of light and dark including a transverse band of
light hairs near apex, apex with a dense patch of
black hairs; light ventrally. Epigynum (Figs. 43,

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

245

47 48

Figures 45-48. —Genitalia of Trechalea paucispina
(Guyana): 45, 46, right palpus; 45, ventral view; 46,
retrolateral view; 47, 48, epigynum; 47, ventral view;

48, dorsal view. Scales in mm.

44) with the mf dark and narrowed in the center;
5 mostly fused to cd.

Variation.— A considerable range of dorsal
patterns occurs from a bold pattern of light and
dark marks to a more uniform dark color with
only the subapical light transverse band present.
Correspondingly, the pattern on the legs varies
from a uniform color to a bolder pattern. Two
lobes of the pma of the epigynum may be en-
larged (Fig. 43) or inconspicuous.

These are rather large spiders with the average
carapace length of 12 males of 9.65 (range 8.5“
10.8) and mean carapace length of nine females
of 9.3 (range 8.0-10.0).

Natural history,— Egg sacs were found with
collections from Brazil (November, 15.3 diam.)
and Peru (June, 24.0 diam,). A note with a male
from Ecuador reads: “Spiders under bridge on
surface or under water on vertical rock surfaces.”

Distribution.— Found in the upper Amazon
River basin in eastern Ecuador, northeastern
Peru, and in Brazil from the State of Rondonia
northward and including the high altitude drain-
ages of coastal rivers in southern Surinam and
western Guyana. (Map 1).

Table 7. — Leg measurements in female of Trechalea
macconnelli.

Leg segment

I

II

III

IV

Femur

17.1

18.7

12.7

17.0

Tibia-patella

22.2

23.7

15.8

20.2

Metatarsus

16.2

17.7

13.9

18.7

Tarsus

10.4

11.9

8.5

11.5

Total

65.9

72.0

50.9

67.4

Specimens examined.— BRAZIL; Amazdnas: Ma-
naus, Reserva Ducke, on side of swimming pool, 30
Mar. 1986 (J. Adis), 2S (INPA); Rondonia: Porto Vie-
ho, Rio Jamari, 27 Dec. 1988 (Equipe Operagao Ja-
mari), 13 (MCN); 18 Nov. 1988 (Equipe Operagao Ja-
mari), 89 1 juv. (MCN). ECUADOR: Pastaza: Cusuimi,
on Rio Cusuimi, 320 m, 1-5 June 1971 (B. Malkin),
23 (FMNH); Cusuimi, on Rio Cusuimi, 150 km SE of
Puyo, 15-22 May 1971 (B. Malkin), 13 29 1 juv.
(FMNH); Morona- Santiago: Yanzatza, 33.4 km N, El
Pincho, 820 m (J. A. Anderson), 13 (USNM). PERU:
Loreto: Rio Yarapa, 80 km S Iquitos, June 1986 (J. &
K. Ribardo), 19 (CAS); Rio Ampiacu, 1 3 Nov.-l 9 Dec.
1961 (B. Malkin), 19 1 juv. (AMNH); Aquaitia, 170
m, 1-2 Sept. 1946 (F. Woytowski), 19 1 juv. (AMNH);
Huanuco: Tingo Maria, 670 m (Weyrauch), 13 (CAS);
Tingo Maria, 43 mi. E (E. I. Schlinger & E. S. Ross),
13 (CAS). SURINAM: Berg-en-del [dal?], May 1892
(E. W. Ellacombe), 19 (holotype of T. ellacombei F.
Pick.-Camb.)(BMNH); Litani, Fetibreek, 15 Sept. 1939
(Gercher), 13 (USNM); Keyserberg airstrip, E of Zuid
River (no date) (H. Baetty), 19 (FMNH).

Trechalea paucispina Caporiacco
Figures 18, 28, 45-48; Map 1

Trechalea paucispina Caporiacco, 1947:22 (The ho-
lotype is a female from Presso, Great Falls, Demer-
da, Guyana, collected September 1931 by Beccari
and Romiti, deposited in the Museo Zoologico della
Specola, Firenze, Italy, examined). Caporiacco, 1948:
633, figs. 24-26. Roewer, 1954:143.

Diagnosis.— Both sexes are distinguished by
details of their genitalia. The palpal bulb has a
distinctive shape of the median apophysis guide
distally (Fig. 1 8), and the tibial apophysis ectal
division truncated apically (Fig. 28). The epi-
gynum externally has the middle field flared along
the middle of its length over medial projections
from lateral lobes and the posterior margin of
the anterior field projecting posteriorly (Fig. 47).

Description.— Afa/e.’ (Canje Ikuruwa River,
Guyana). Carapace low, cephalic area somewhat
elevated, length 8.0, width 7.5, medium brown
medially with distinct submarginal bands, dark
at lateral margins and in eye region. Sternum

246

THE JOURNAL OF ARACHNOLOGY

Table 8.— Leg measurements in male of Trechalea
paucispina.

Leg segment

I

II

III

IV

Femur

13.3

14.2

11.3

14.7

Tibia-patella

16.9

18.1

13.2

17.2

Metatarsus

13.2

14.2

10.8

16.1

Tarsus

8.5

8.9

7.5

10.3

Total

51.9

55.4

42.8

58.3

light, unmarked, length 4.1, width 3.8; labium
dark brown, darker laterally at basal half, lighter
at distal margin, length 1.65, width 1.38. Clypeus
height 0.78, width 3.70. Anterior eye row straight,
a cluster of bristles posterior to each PLE, eye
measurements in Table 1. Chelicerae face dark,
clothed with light hairs and longer more erect
dark hairs, an oblique groove above fang and a
short longitudinal carina laterally, three retro-
marginal teeth equal in size with distal two slight-
ly closer together. Legs IV-II-I-III, measure-
ments in Table 8, ventral macrosetae pairs on
tibiae are 1-4, II-5, III-4, IV-3. Color of legs gen-
erally light with distinct markings on prolateral
surfaces of all femora and tibiae-patellae. Ab-
domen length 7.3, hairy above with a distinct
pattern including transverse marks in the pos-
terior third, light ventrally. Palpus (Figs. 45, 46),
ma (Fig. 1 8) with g winged apically, rta (Fig. 28)
with ecd straight, truncated apically.

Female: (Canje Ikuruwa River, Guyana). Car-
apace shape and color as in male, length 8.2,
width 8.2. Sternum light, unmarked, length 4.3,
width, 4.0; labium color as in male, length 1.76,
width 1.48. Clypeus height 0.80, width 4.05. An-
terior eye row straight, eye measurements in Ta-
ble 1. Chelicerae face dark, clothed with light
hairs and scattered longer more erect dark hairs.
Legs IV-II-I-III, measurements in Table 9, ven-
tral macrosetae pairs on tibiae 1-4, II-4, III-3,
IV-3. Color of legs as in male. Abdomen length
9.5, hairy above, color as in male. Epigynum
(Figs. 47, 48) with the mf narrowed centrally and
flared laterally over // and the pma projecting
posteriorly; 5 mostly fused to cd.

Variation.— Of eight females measured, the
average carapace length is 8.1 (range 7. 4-9. 2).
The only male available had the carapace length
of 8.0.

Natural history.— Most collection labels in-
dicate that the specimens were taken from rivers,
however a collection from Guyana also includes
the notation; “forest savanna”. Three egg sacs

Figures 49-52. — Genitalia Trechalea connexa: 49,
50, right palpus (Veracruz, Mexico); 49, ventral view;
50, retrolateral view; 5 1, 52, epigynum (Morelos, Mex-
ico); 51, ventral view; 52, dorsal view. Scales in mm.

found with females average 17.3 (1 3.7-24.0) and
are of the typical trechaleid structure.

Distribution.— Amazon River tributaries in
northwestern Brazil and central Peru northward
into the coastal river drainages of Guyana. (Map
1).

Specimens examined.— GUYANA: Canje Ikuruwa
River, 57.50W:5.70N, Aug.-Dee. 1961 (G. Bently), \S
29 (AMNH); Shudicar River, upper Essequibo River,
1 Jan. 1938 (W. G. Hassler), 1$ (AMNH). PERU: Lo-
reto: Aquaitia [Aguaytia R.], 170 m, 1-2 Sept. 1946
(F. Woytkowski), 19 (AMNH). BRAZIL: Amazonas:
Ica (Thayer Expedition), 19 (MCZ); Rondonia: Jamari,
#18568, 18 Nov. 1988 (Equipe Operagao Jamari), 19
(MCN); Acre: Rio Purus NW of Sena Madureira Ser-
ingal Santo Antonio (above Manuel Urbano), 15-18

Table 9. — Leg measurements in female of Trechalea

paucispina.

Leg segment

I

II

III

IV

Femur

12.6

13.8

11.0

14.3

Tibia-patella

16.3

17.3

13.0

17.3

Metatarsus

11.8

12.8

10.4

15.9

Tarsus

7.6

8.3

7.3

10.0

Total

48.3

52.2

41.7

57.5

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

247

Map 2.— Distribution of species of Trechalea in USA,
Mexico and Central America. # = 7". gertschi, M= T.

connexa, A = T. extensa.

Sept. 1973 (B. Patterson), 1$ (MCZ); Para: Rio Ma-

putra [Rio Mapuera?] 10 mi. S of Equator, 8-9 Feb.
1938 (W. G. Hassler), 1$ (AMNH).

Trechalea connexa (O. Pickard-Cambridge)
Figures 19, 29, 49^52; Map 2

Triclaria connexa O. Pick.=Camb., 1898:233 (The ho-
lotype is a male from Atoyac, Veracruz, Mexico,
collected by H. H. Smith, deposited in The Natural
History Museum, London, examined).

Trechalea connexa, F. O. Pick.-Camb., 1902:312, 313.
Petrunkevitch, 1911:548. Roewer, 1954:142. Bon-
net, 1955-1959:4679.

Diagnosis. —Both sexes are distinguished by
details of their genitalia. The median apophysis
(Fig. 19) differs from T, gertschi in details of the
guide and ventral division. The retrolateral tibial
apophysis also resembles that of T. gertschi but
is thinner (Fig. 29). In the female, the middle
field is narrowed centrally but is flared distally
(Fig. 51) while the same structure in T. gertschi
is broad throughout without distinct central nar-
rowing.

Description.— (Fortin, Veracruz, Mex-
ico). Carapace low, cephalic area not elevated,
length 6.5, width 6.2, medium brown with in-
distinct light areas laterally, dark at margin and
in eye region; sternum light, unmarked, length
3.5, width 3.3; labium dark brown, lighter dis-
tally, an irregular, longitudinal furrow at the bas-
al half on each side. Clypeus height 0.67, width
3.00. Anterior eye row straight, eye measure-
ments in Table 1 . Chelicerae face medium brown,
smooth, almost glabrous medially with hairs pe-
ripherally, an oblique groove above each fang

Table 10. — Leg measurements in male of Trechalea
connexa.

Leg segment

I

II

III

IV

Femur

9.5

11.0

9.1

10.7

Tibia-patella

12.5

14.0

10.9

13.2

Metatarsus

9.7

6.3

9.0

12.4

Tarsus

5.6

6.1

5.3

7.2

Total

37.3

37.4

34.3

43.5

and a longitudinal carina laterally on distal one-
third of its length, three retromarginal teeth equal
in size and equidistant. Legs IV-(II-I)-III, mea-
surements in Table 10, ventral macrosetae pairs
on tibiae are 1-4, II-4, III-4, IV-4. Color of legs
generally light with a faint pattern on prolateral
surface of femora and tibiae. Abdomen length
6.1, hairy above with reticulated pattern of dark
pigment, darker posteriorly and with light spots
anteriorly around muscle attachments and around
anterior margin; light ventrally. Palpus (Figs. 49,
50), ma (Fig. 19) with g winged apically; rta (Fig.
29) with ecd flattened, rounded apically.

Female: (Cuernavaca, Morelos, Mexico). Car-
apace shape and color as in male, length 7.2,
width 7.0; sternum light, unmarked, length 3.5
width 3.6; labium color as in male, length 1.45,
width 1.25. Clypeus height 0.75, width 3.36. An-
terior eye row straight, eye measurements in Ta-
ble 1 . Chelicerae face dark, clothed with scattered
short, light hairs and scattered, longer, more erect
hairs. Legs IV-II-I-III, measurements in Table
1 1 , ventral macrosetae pairs on tibia are 1-4, II-
4, III-3, IV-3. Color of legs as in male. Abdomen
length 8.0, hairy above, pattern in poor condition
but similar to male. Epigynum (Figs. 51,52) with
w/flared and grooved posteriorly; 5 mostly fused
to cd.

Variation.— In alcohol, the dorsum shows a
radiating group of dark lines on the carapace with
vague evidence of irregular submarginal bands,
while the abdomen is generally dark with varying
light marks near the anterior margin including
the cardiac area, and around the abdominal mus-
cle apodemes. The average carapace lengths of
14 males is 7.45 (range = 6. 1-9.3) and of 25
females is 7.63 (range 6.4-9.25).

Natural history.— Little biological data are
contained with the specimens and are limited to
two references to the vicinity of water. Egg sacs
are in four collections which are 12.7, 14.5, 10.8,
and 17.8 diameter from April, June, July, and
August respectively.

248

THE JOURNAL OF ARACHNOLOGY

Table 11. — Leg measurements in female of Tre-
chalea connexa.

Leg segment

I

II

III

IV

Femur

9.3

10.9

9.4

10.6

Tibia-patella

12.5

13.1

11.4

14.3

Metatarsus

8.9

10.2

8.9

11.7

Tarsus

5.5

6.0

5.5

6.7

Total

36.2

40.2

35.2

43.3

Distribution. —From the Isthmus of Tehuan-
tepec of Mexico northwestward through central
Veracruz on the Atlantic coast to southern Jal-
isco on the Pacific coast. (Map 2).

Specimens examined.— MEXICO: Veracruz: Fortin,
28 April- 1 May 1 944 (C. Bolivar & L Pina), 2S (AMNH);
Oaxaca: Tehuantepec, 22 Dec. 1947 (T. McDougall),
1(3 (AMNH); Morales: Cuernavaca, 1 April 1942, 19
(AMNH); Guerrero: Colotlipa, Rio Blanco, 1 Aug. 1941,
1$ (AMNH); Jalisco: 5 mi. N Pihuamo 2350 ft., 5 Aug.
1967 (R. E. Leech), 19 (REL); Sinaloa: Camino Real
de Piaxtla, 4 May 1949 (G. W. Bradt), 39 (AMNH);
Nayarit: Tepic, 2 Aug. 1947 (C. Goodnight), 19
(AMNH), 2-7 Aug. 1947 (C. & M. Goodnight & B.
Malkin), 5 mi. NW Tepic, 13 May 1963 (W. J. Gertsch
& W. Ivie), 56 39 2 juv. (AMNH), Mecatan 800 ft., 2
May 1949 (G. M, Bradt), IS 29 1 juv. (AMNH), Jesus
Maria, 25 June 1955 (B. Malkin), 46 89 (AMNH), 1-
15 July 1955 (B. Malkin), 16 39 (AMNH), 22-30 June
1955 (B. Malkin), 16 (AMNH), July 1955 (B. Malkin),
19 (AMNH), Arroyo Santiago, 3 mi. NW Jesus Maria,
4-6 July 1955 (B. Malkin), 19 (AMNH), 4 July 1955
(B. Malkin), 19 (AMNH).

Trechalea extensa (O. Pickard-Cambridge)
Figures 20, 30, 53--56; Map 2

Triclaria extensa O. Pickard-Cambridge, 1 896, 1 : 1 74-
175 (The holotype is a male from Rokminhi, Gua-
temala, collected by Sarg, deposited in The Natural
History Museum, London, examined).

Trechalea extensa, F. Pickard-Cambridge 1902, 2:313.
Petrunkevitch, 1911:549. Roewer, 1954, 2a:143.
Bonnet, 1955-1959:4679.

Trechalea magnijica Petrunkevitch, 1 925: 1 69-170 (The
syntypes are from Wilcox dam on San Lorenzo Riv-
er, Bocas Del Monte, La Mesa and Santiago, Pan-
ama, deposited in the Museum of Comparative Zo-
ology, examined). Roewer, 1954:143. Bonnet, 1955-
1959:4679. NEW SYNONYMY.

Diagnosis.— Both sexes are distinguished from
other species by details of the genitalia. The me-
dian apophysis of the male palpal bulb bears a
distinct tubercle between the guide and ventral
division (Fig. 20), a feature shared only with T.
longitarsis (Fig. 15). Trechalea extensa differs

Figures 5 3-56. —Genitalia oi Trechalea extensa (Ca-
nal region, Panama): 53, 54, right palpus; 53, ventral
view; 54, retrolateral view; 55, 56, epigynum; 55, ven-
tral view; 56, dorsal view. Scales in mm.

from the latter species by the length of the cym-
bium which is distinctly less than half the length
of the bulb (Fig. 53). The middle field of the
epigynum is short and usually with the sides al-
most parallel (Fig. 55). See diagnosis of T. Ion-
gitarsis for other comparisons.

Description.— Afflf/e.' (Barro Colorado Island,
Lago Gatun, Panama). Carapace low, cephalic
area not elevated, length 8.8, width 8.1, light
brown background color with lighter submargin-
al bands more distinct posteriorly, dark at lateral
margins, in eye region and a spot on each side
of clypeus. Sternum light, unmarked, length 4.5,
width 4.3; labium dark brown, lighter at distal
margin, length 1.80, width 1.62. Clypeus height
1.05, width 4.08. Anterior eye row straight, a
cluster of bristles posterior to each PLE, eye mea-
surements in Table 1. Chelicerae face dark,
clothed with light hairs mostly in proximal two-
thirds, an oblique groove above fang and a lon-
gitudinal Carina laterally on distal half, three re-
tromarginal teeth of equal size with distal two
closer together. Legs II-IV-I-III, measurements
in Table 12, ventral macrosetae pairs on tibiae
1-4, IL4, III-3, IIL3. Color of legs generally light
except for distinct dark marks on prolateral sur-

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

249

Table 12. — Leg measurements in male of Trechalea

extensa.

Leg segment

I

II

III

IV

Femur

14.3

16.2

12.5

16.1

Tibia-patella

19.4

20.4

14.7

19.7

Metatarsus

15.1

17.1

12.4

15.3

Tarsus

10.0

11.0

8.3

12.4

Total

58.8

64.7

48.0

63.5

face of all legs. Abdomen length 9.2, hairy above,
color a reticulated pattern of dark pigment with
distinct small, scattered, dark spots, and lighter
in cardiac area and a pair of indistinct spots in
the posterior third, light ventrally. Palpus (Figs.
53, 54), ma (Fig. 20) with ^winged apically, and
with distinct distal tubercle situated between g
and vd\ cymbium/palpal bulb length ratio 1.67.
The rta (Fig. 30) with ecd flattened and hooked
distally and serrated along inner margin.

Female: (Barro Colorado Island, Lake Gatun,
Panama). Carapace shape and color as in male
but with markings more distinct, length 8.8, width
8.5. Sternum light, unmarked, length 4.5, width
4.3; labium color as in male, length 1.74, width
1.52. Clypeus height 1.06, width 4.00. Anterior
eye row straight, eye measurements in Table 1.
Chelicerae color, teeth, and hair as in male but
with the addition of longer and more erect hairs
grouped mostly in longitudinal rows. Legs IV-
II-I-III, measurements in Table 13, ventral
macrosetae pairs on tibiae 1-5, II-4, III-4, IV-4.
Color of legs as in male. Abdomen length 1 1.6,
color and hair as in male. Epigynum (Figs. 55,
56), mf sides almost parallel; 5 mostly fused to
cd.

Variation.— The average carapace length of 19
males is 9.05 (range 8.3-10.0), and the average
carapace length of 32 females is 9.38 (range 8.2-
12.0). The average cymbium length/palpal bulb
length ratio of 1 9 males is 1 .72 (range 1 .60-1 .88).
The width of the middle field of the epigynum
is wider in specimens in the northern part of the
range.

The retromarginal cheliceral teeth number is
typically three, but two males from Panama had
four on only one side. One specimen had the
fourth a full-sized tooth while the second had
only a tiny added tooth at the base of another
tooth.

Natural history.— These spiders are a distinc-
tive feature of the streams of Panama and Costa
Rica. They seem to be completely restricted to

Table 13, — Leg measurements in female of Tre-
chalea extensa.

Leg segment

I

II

III

IV

Femur

13.1

15.0

11.9

15.3

Tibia-patella

17.2

19.0

13.9

18.7

Metatarsus

12.9

14.6

11.4

18.1

Tarsus

8.0

9.1

7.6

10.0

Total

51.2

57.7

44.8

62.1

the stream margins and bases of emergent rocks
and debris. They have been collected from large,
open-canopied rocky streams and from first-or-
der streams in vegetational thickets. They bite
readily the collector’s hand. Van Berkum (1982)
has reported an apparent preference for shrimps
as prey in Rincon de Osa in Panama.

Egg sacs are in five collections; the average of
four is 13.9 (range 10.0-15.5) collected during
the months of May and June.

Distribution. — In Central America ranging
from Central Panama in the south to the Mex-
ican state of Chiapas in the north. (Map 2).

Specimens examined.— PANAMA: (Central area
around Canal, summary of several collections), 56 1 39,
several juv.; El Valle de Anton, 1 April 1945 (C. D.
Miehener), 16, 19 (AMNH); El Valle, July 1936 (A, M.
Chickering), 19 (MCZ); river 10 km W of David, 8
Aug. 1983 (Carico, Coyle, Eberhard, Coddington), 96
29 several juv. (JEC). COSTA RICA: river 5.3 km N
of Las Canas on Rt. #19, 11 Aug. 1983 (Carico, Coyle,
Vogel), 26 19 (JEC); San Antonio de Escazu, near San
Jose, 14 Aug. 1983 (J. E. Carico), 19 (JEC), May 1984
(W. Eberhard), 19 (MCZ); Puntarenas, Esterillos, 20
June 1970 (D. C. Robinson & R. Saena), 19 (MZUCR);
18 km S of San Isidro del General, 2-3 June 1972 (J.
Baldridge), 19 (MZUCR); Butler’s Finca 9°18':
83“o47'W, 28 Jan. 1976 (Roth-Schroepfer), 19
(AMNH); Los Diamantes, Guapiles (C. E. Valerio), l9
(MZUCR); San Mateo (N. Banks), 19 (MCZ); Tilaran
(C. E. Valerio), 16 (MZUCR). HONDURAS: Copan
(R. V. Chamberlin), 29 5 juv. (AMNH), Lancetilla, July
1929 (A. M. Chickering), 39 3 juv. (MCZ). MEXICO:
Chiapas: 5.6 mi. SE Chiapa de Corzo, 2500 ft., 1 6 Aug.
1966 (D. E. Breedlove & J. Emmel), l9 (CAS), Ma-
pastepec, June-July, 1940 (H. Wagner), l9 (AMNH),
Rancho la Esperanza, 40 km Escuintla, 22 Jan. 1945
(T. C. Schneirla), 19 (CAS). NICARAGUA: Polv6n
(McNeill), 16 29 (MCZ).

Trechalea gertschi Carico & Minch
Figures 21, 31, 57-60; Map 2

Trechalea gertschi Carico & Minch, 1981:154-1 56, figs.
1-4 (Male holotype from 9 mi, S of Sunflower, Mar-
icopa County, Arizona, USA, collected by E. Minch,

250

THE JOURNAL OF ARACHNOLOGY

Figures 5 7=-60.— Genitalia of Trechalea gertschi (Ar-
izona, USA): 57, 58, right palpus; 57, ventral view; 58,
retrolateral view; 59, 60, epigynum; 59, ventral view;
60, dorsal view. Scales in mm.

9 June 1979, deposited in the American Museum of

Natural History, examined). Brignoli, 1983:700.

Platnick, 1989:389.

Diagnosis.— In the male, the median apoph-
ysis has the ventral division truncated apically
and a notch retrolaterally (Fig. 21). The ectal
division of the retrolateral tibial apophysis is a
rounded projection flattened ventrally and often
with a serrated surface (Fig. 31). The median field
of the epigynum is flattened and overlapping the
lateral lobes (Fig. 59).

Description. (Holotype). Carapace
length 7.0, width 7.4., moderately low with oc-
ular area dark, marginal band dusky, serrated
submarginal light band, median light band ex-
tending from thoracic groove to posterior margin
of carapace. Sternum light, unmarked, length 3.6,
width 3.8; labium dark, with transverse groove,
length 1.6, width 1.4. Clypeus height 0.75, width
3.60. Anterior eye row slightly recurved, eye
measurements in Table 1. Chelicerae face dark
with three equal sized retromarginal teeth, distal
two closer. Legs IV-II-I-III, measurements in Ta-
ble 1 4, ventral macrosetae pairs on tibiae are 1-4,
II-4, III-3, IV-3. Color of legs light with faint
band on dorsal side of femora, patellae-tibiae.

Table 14. — Leg measurements in male of Trechalea

gertschi.

Leg segment

I

II

III

IV

Femur

10.0

12.0

9.6

11.2

Tibia-patella

13.4

15.5

11.6

14.2

Metatarsus

10.0

11.8

9.4

13.5

Tarsus

5.5

6.5

5.5

7.6

Total

38.9

45.8

36.1

46.5

Abdomen length 6.0, dorsum has irregular darker
mottling clothed with several black, long setae,
venter light and unmarked except for black setae
in genital area. Palpus (Figs. 57, 58), ma (Fig.
21) with g winged on either side, vd truncated
distally with retrolateral notch and transverse
grooves, rta (Fig. 3 1) with ecd with uniform width,
rounded apically, ventral side flattened with ser-
rated surface. (Figs. 21, 31 from another male
from type locality).

Female: (Paratype). Carapace shape and color
as in male, length 7.9, width 7.4. Sternum light,
unmarked, length 4.0, width 4.O.; labium similar
to male, length 1.60, width 1.50. Clypeus height
0,80, width 3.70. Anterior eye row straight, eye
measurements in Table 1. Chelicerae as in male.
Legs IV-II-III-I(?), measurements in Table 15,
ventral macrosetae pairs on tibiae 1-4, II-2, III-
3, IV-3. Color of legs as in male. Abdomen length
8.5, color pattern as in male. Epigynum (Figs.
59, 60) has the m/ broad with an anterior con-
striction and overlapping // posteriorly.

Variation.— The average carapace length of 20
males is 7.59 (range 6. 8-8. 4), and the average
carapace length of 21 females is 7.57 (range 6.8-
8.9).

Natural history. —All instars of individuals are
found restricted to the margins of apparently per-
manent streams within the xeric regions. Typi-
cally they are found on the surfaces of rocks and
pebbles of varying sizes near the water margin.
They readily run across water and occasionally
crawl underwater by walking down the surface

Table 15. — Leg measurements in female of Tre-
chalea gertschi.

Leg segment

I

II

Ill

IV

Femur

10.5

12.5

11.0

12.5

Tibia-patella

13.9

16.5

13.3

14.6

Metatarsus

10.0

12.2

10.7

14.7

Tarsus

6.2

6.1

8.1

Total

40.6

-

41.1

49.9

CARICO-^THE GENUS TRECHALEA (TRECHALEIDAE)

251

Figures 61-64.— Genitalia of Trechalea amazonica
(Amazonas, Brazil): 61, 62, right palpus; 61, ventral
view; 62, retrolateral view; 63, 64, epigynum; 63, ven-
tral view; 64, dorsal view. Scales in mm.

of a partly submerged rock. Females carry egg
sacs by the spinnerets and show other typical
trechaleid behavior described above.

Five egg sacs were found with the collections.
Their dates and sizes (stated only if in a condition
to measure) are: 25-27 June (3 egg sacs, 18.8 &
19 mm), 6 July (1 egg sac, 15 mm), 17 July (2
egg sacs). Two apparently gravid females were
in collections made in January and July.

Distribution.— From Yavapai County of cen-
tral Arizona, USA southward into the Mexican
states of Sonora and Chihuahua. (Map 2).

Material examined. —USA: Arizona: Maricopa
County, 9 mi. S Sunflower, 3 June 1979, (E. Minch),
IS 22, (AMNH), 1 1 July 1979, (E. Minch), 2S, (AMNH);
Yavapai County, Clear Cr., 11 mi. E Camp Verde, 25-
27 June 1986, (J. E. Carico), 42, (JEC); Pinal County,
Sycamore Cr. 30 mi. NE Apache Junction, 28 June
1986, (J. E. & E. L. Carico, E. Minch), 33 12, (JEC);
Pima County., Sabino Canyon, Santa Catalina Mts. nr.
Tuscon, 1 Sept. 1939, (R. H. Crandall), \S 12, (AMNH),
5 Oct. 1937, (Crandall), 13, (AMNH), 6 July 1939, (R.
A. Rock), 12, (MCZ), 17 July 1971, (J. E. Carico), 33
62, (AMNH); Santa Cruz County, Santa Rita Mts., 12,
(AMNH); Gila County, Gisela, 5 Nov. 1977, (E. Minch),
13, (AMNH). MEXICO: Sonora: 6 mi. E Alamos, Rio
Cuchujachi, 22 June 1966, (V. Roth) 63 32 (AMNH);
SE Alamos on Rio Cuchujaqui, Jan. 1968, (V. Roth),
13 32, (AMNH); 57 mi. SE Aqua Prieta on bank of
Rio El Batista [Bavispe], 26 June 1972, (G. Dingerkus),

Table 16. — Leg measurements in male of Trechalea
amazonica.

Leg segment

I

II

III

IV

Femur

8.9

9.0

7.5

9.7

Tibia-patella

11.3

10.9

8.5

10.8

Metatarsus

7.7

7.8

6.8

10.4

Tarsus

4.5

4.7

4.5

6.0

Total

32.4

32.4

27.3

36.9

12, (AMNH); Chihuahua: Urique Rio, 21 April 1986,
(V. D. Roth), 13, (JEC).

Trechalea amazonica F. O. Pickard-Cambridge
Figures 22, 32, 61-64; Map 3

Trechalea amazonica F. O. Pickard-Cambridge, 1903:
163, plate 15, figs. 18-20 (The holotype is a male
from Santarem, Amazonia, Brazil, collected by F.
O. Pickard-Cambridge 1 895--1 896, deposited in The
Natural History Museum, London, examined; two
female paratypes with holotype from the same lo-
cality).

Trechalea manauensis Carico, in Carico et al., 1985,
6(7): 28 9-294, figs. 1-4 (The holotype is a male from
Ihla de Marchantaria, Rio Solimoes (near Manaus),
Amazonas, Brazil, collected by J. Adis, 28 June 1981,
deposited in the Systematic Entomology collection
of Instituto de Pesquisas da Amazonia, examined;
female paratype from the same locality). Platnick,
1989:398. NEW SYNONYMY.

Diagnosis.— The palpal bulb is distinguished
by the blade-like and dual rounded edge of the
median apophysis ventral division (Fig. 22) and
the relative size and shape of the components of
the retrolateral tibial apophysis (Fig. 32). The
epigynum is distinguished by the shape of the
middle field (Fig, 63) and unique shape of the
internal components (Fig. 64).

Description.— (Holotype of T. man-
auensis). Carapace length 5.5, width 5.3, mod-
erately low, light colored with indistinct mark-
ings and ocular area dark. Sternum light,
unmarked, length 2.80, width 0.98; labium dark
especially laterally, lighter at anterior margin,
length 1.18, width 0.98. Clypeus height 0.60,
width 2.70. Anterior eye row straight, eye mea-
surements in Table 1. Chelicerae face swollen,
dark, clothed with light hairs, an oblique groove
above fang and a longitudinal carina laterally on
distal one-half, three equal-sized retromarginal
teeth with distal two closest. Legs IV-(I-II)-III,
measurements in Table 16, ventral macrosetae
pairs on tibiae are 1-4, II-5, III-3, VI-3. Color of
legs light with faint bands on dorsal side of fern-

252

THE JOURNAL OF ARACHNOLOGY

Table 17. — Leg measurements in female of Tre-
chalea amazonica.

Leg segment

I

II

III

IV

Femur

8.7

8.7

7.5

9.7

Tibia-patella

10.7

10.8

8.3

11.2

Metatarsus

7.1

7.4

6.9

10.3

Tarsus

3.5

4.4

4.3

6.0

Total

30.0

31.3

27.0

37.2

ora, patellae-tibiae, and metatarsus. Abdomen
length 6.1, dorsum color a reticulated gray back-
ground, three pairs of diagonal gray maculae,
numerous erect, dark setae anteriorly, venter un-
marked on light background. Palpus (Figs. 61,
62), ma (Fig. 22) with vd flattened with two
rounded projections and g tapered and curved
at tip, rta (Fig. 32) with end large and pointed
distally, ecd smaller, tapered, curved, acute with
brush of long hairs separating the two divisions.

Female: (Paratype of T. manauensis). Cara-
pace length 5.-7, width 5.5, shape and color as in
male. Sternum light, unmarked, length 2.95,
width 2.75; labium similar to male, length 1.20,
width 1.05. Clypeus height 0.57, width 2.80. An-
terior eye row straight, eye measurements in Ta-
ble 1 . Chelicerae dark, clothed with light hairs,
teeth as in male. Legs IV-(I-II)-III, measure-
ments in Table 17, ventral macrosetae pairs on
tibiae 1-4, II-4, III-3, IV-3. Color of legs as in
male. Abdomen length 7.0, color patterns as in
male. Epigynum (Figs. 63, 64), pma extending
posteriorly around mf to form U-shaped frame
around mf. The posterior apex of the m/is black
and somewhat set off from the light-colored an-
terior portion by a constriction.

Variation.— Average carapace length of six
males is 5.45 (range 4.65-5.75) and average car-
apace length of 1 1 females is 5.9 (range 5. 4-6. 8).

Natural history.— Adis and Penny (in Carico
et al. 1986) provide much detailed information
on the natural history, behavior, and parasitism
on this species which inhabits the inundation
forests of the central Amazon River area. Six egg
sacs were found in the collections; February (1 2.0
diam.), 31 March (3 egg sacs, 8.35, 8.5, 8.5), 28
April (9.6, [infested with chalcid wasps]), 3 1 May
(9.8).

Distribution.— Known only from the main
channel, Rio Solimoes and Rio Amazonas in the
state of Amazonas, Brazil. (Map 3).

Material examined. — BRAZIL: Amazonas: Ihla
Marchantara, Solimoes, 28 April 1981 (J. Adis), 35 12,

Map 3.— Distribution of species of Trechalea in
northern South America. A = T. lomalinda, M = T.
trinidadensis, • = T. amazonica, ♦ = T. boliviensis.

16 Dec. 1987 (E. H. Buckup), 15, Rio Solimoes in
vdrzea forest 15 km from Manaus, 31 March 1976 (J.
Adis), 35 42, Lago Janauaca nr. Rio Solimoes, 50 km
from Manaus, no date (J. Adis), 32, Rio Taruma Mir-
iam, 20 km upstream from Manaus, 28 April 1976 (J.
Adis), 12, (same locality), 31 May 1976 (J. Adis), 12,
(previous specimens deposited variously in INPA,
AMNH, JEC).

Trechalea boliviensis, new species
Figures 23, 33, 40, 65-68; Map 3

Type.— The holotype is a male from Bolivia,
Dpto. Beni, Est. Biol. Beni, Zone 1, ca. 4°47'S:
66°15'W, ca. 225 m; collected 8-14 November
1989, by Coddington, Larcher, Penaranda, Gris-
wold, and Silva, deposited in the Institute de
Ecologia, La Paz, Bolivia. Female paratype from
the type locality deposited in the United States
National Museum.

Etymology.— The name means “from Boliv-
ia,” the country of origin.

Diagnosis. —Both sexes are distinguished from
those of all other species by the narrowed shape
of the posterior third of the abdomen (Fig. 40),
a patch of black hairs on the posterior apex of
the abdomen, and details of the shape of their
respective genitalia (Figs. 23, 33, 67, 68).

Description.— (Holotype). Carapace low,
cephalic area elevated, length 4.2, width 4.4, gen-
erally light but with dark on margin and black
around each eye; sternum light, unmarked, length
2.55, width 2.4; labium dark, light at distal mar-
gin, with a longitudinal darker band in the basal

CARICO=~THE GENUS TRECHALEA (TRECHALEIDAE)

253

Table 18.— Leg measurements in male of Trechaiea
boliviensis.

Leg segment

I

II

III

IV

Femur

8.8

8.3

6.4

9.0

Tibia-patella

11.7

10.0

6.9

9.7

Metatarsus

8.6

7.6

5.9

9.9

Tarsus

5.3

5.0

4.1

6.1

Total

34.4

30.9

23.3

34.7

half on each side, length 0.88, width 0.72. Clyp-
eus height 0.50, width 2.20. Anterior eye row
straight, eye measurements in Table 1. Chelic-
erae face dark, basal segments clothed with light
hair and a few larger, more erect dark bristles
medially, oblique depression above fang, and
longitudinal carina on distal one-third of lateral
margin; three retromarginal teeth equal in size
with distal two closer together. Legs IV-I-II-III,
measurements in Table 18, ventral macrosetae
pairs on tibiae are 1-6, II-6, III-4, IV-4; color
light ventrally with indistinct markings on other
surfaces. Abdomen length 5.5, median cleft at
anterior margin, narrowed posteriorly, irregular
dark band laterally, irregular dorsal pattern (Fig.
40), light ventrally, patch of dark hairs at pos-
terior apex. Palpus (Figs. 65, 66), ma (Fig. 23)
with g acute and curved ventrally; rta (Fig. 33)
with ecd with three lobes and ecd blade-like and
slightly curved medially.

Female: (Paratype). Carapace shape and color
as in male, length 4.5; width 4.5; sternum light,
unmarked, length 2.75, width 2.35; labium as in
male, length 0.98, width 0.80. Clypeus height
0.50, width 1.3. Anterior eye row straight, eye
measurements in Table 1. Chelicerae face gen-
erally dark, darker distally, basal segments clothed
with light hair and several more erect dark bris-
tles, teeth as in male. Legs IV-I-II-III, measure-
ments in Table 19, ventral macrosetae pairs on
tibia 1-6, II-6, III-4, IV-4; color as in male. Ab-
domen length 5.5, color as in male.

Epigynum (Figs. 67, 68) with mf triangular,
light anteriorly, with dark and narrowed poste-
rior apex; s fused to cd except distally, // wrinkled
posteriorly.

Variation.- Carapace length average of four
males is 3.9 1 (range 3. 7-4. 2) and the two females
have carapace lengths of 4.2 and 4.5.

Specimens examined and distribution.— BOLIVIA:
Beni: (Type collection), 26 2$ (USNM). PERU: Cuzco:
Quincemil, 750 m, August 1962 (Pena), 16 (MCZ).
(Map 3).

Figures 65=68.— Genitalia of Trechaiea boliviensis
(Beni, Bolivia): 65, 66, right palpus; 65, ventral view;
66, retrolateral view; 67, 68, epigynum; 67, ventral
view; 68, dorsal view. Scales in mm.

Trechaiea lomalinda, new species
Figures 24, 34, 69-72; Map 3

Type.— The holotype is a male from Lomal-
inda, Puerto Lleras, Meta, Colombia, 300 m el-
evation, collected 1 5 April 1986 by B. T. Carroll,
deposited in the California Academy of Sciences
Museum. Paratype female from the type locality,
collected March 1987 by B. T. Carrol, deposited
in the California Academy of Sciences Museum.

Etymology.— The name is a noun in apposi-
tion taken from the name of the type locality.

Diagnosis.— The ental division of male retro-
lateral tibial apophysis is more prominent than
ectal division (Fig. 34), The dorsal division of
the median apophysis has a unique prominent
projection other than the guide (Fig. 24). In the
female epigynum the middle field is triangular
and entirely white (Fig. 71); the copulatory duct
is distinctively narrow (Fig. 72).

The first leg pair is longer than the second. The
leg femora are slender and distinctly tapered.

Description.— (Holotype). Carapace very
low, cephalic area elevated, length 5.75, width
5.3, no distinct pattern and without black in the
ocular area; sternum light, unmarked, length 3.1,
width 2.7; labium dark reddish brown, light at
distal margin, length 1.2, width 1.06. Clypeus
height 0.70, width 2.87. Anterior eye row straight.

254

THE JOURNAL OF ARACHNOLOGY

Figures 69-72.— Genitalia of Trechalea lomalinda
(Meta, Colombia): 69, 70, right palpus; 69, ventral
view; 70, retrolateral view; 71, 72, epigynum; 71, ven-
tral view; 72, dorsal view. Scales in mm.

eye measurements in Table L Chelicerae face
dark, clothed with conspicuous light hair prox-
imally, and clothed sparsely distally with fine
hairs, oblique depression above fang, and lon-
gitudinal Carina on lateral margin; three retro-
marginal teeth equidistant and equal in size. Legs
IV-I-II-III, measurements in Table 20, ventral
macrosetae pairs on tibiae are 1-6, II-6, III-4, IV-
4, color light ventrally with indistinct markings
on other surfaces except for two oblique marks
on the distal part of prolateral surface of femur
III and a longitudinal mark on prolateral surface
of patella III. Abdomen length 5.1, shrivelled
and without apparent pattern. Palpus (Figs. 69,
70), ma (Fig. 24) with dd with a prominent pro-
jection besides the g. The vd relatively large and
acute and hooked distally. The rta (Fig. 34) with

,73 74

Figures 73, 74.— Epigynum of Trechalea trinidaden-
sis (Trinidad): 73, ventral view; 74, dorsal view. Scales
in mm.

end relatively prominent, flattened and rounded
apically, ect directed laterad with acute tip turned
mediad.

Female: (Paratype). Carapace shape as in male,
indistinct light pattern on lighter background,
dark at edge especially at edge of cephalic area,
and dark in eye region, length 6.0, width 5.7;
sternum light unmarked, length 3.15, width 2.8;
labium color as in male, length 1.25, width 1.10.
Clypeus height 0.73, width 3.0. Anterior eye row
straight, eye measurements in Table 1. Chelic-
erae face dark, clothed with conspicuous light
hair proximally, and clothed sparsely distally with
fine hair on a glossy integument; teeth as in male.
Legs IV-I-II-III, measurements in Table 2 1 , ven-
tral macrosetae pairs on tibiae are 1-5, II-5, III-
3, IV- 3; dark pattern on legs most distinct on
prolateral surfaces and missing from ventral sur-
face of femora. Abdomen length 7.3, light ven-
trally, color dorsally composed of reticulated pat-
tern of dark changing to fine lines posteriorly and
laterally, light spots around muscle apodeme
marks. Epigynum (Figs. 71, 72) with mf very
light and triangular, 5 mostly fused to cd\ cd nar-
row.

Variation. —The average carapace length of six
males is 5.36 (range 5.0--5.75). The carapace
lengths of two females are 5.9 and 6.0.

Natural history.— A note with one of the col-
lections states: “grasslands; patches of jungle,
woods, marsh. Indoors, daylight.”

Specimens examined and distribution. —Known only
from Colombia, Depto. Meta, Pto. Lleras, Lomalinda

Table 19. — Leg measurements in female of Tre-
chalea boliviensis.

Leg segment

I

II

III

IV

Femur

7.6

7.3

6.0

8.1

Tibia-patella

9.7

8.7

6.5

8.7

Metatarsus

6.4

6.0

5.3

8.5

Tarsus

4.0

4.2

4.1

5.6

Total

27.7

26.2

21.9

30.9

Table 20. — Leg measurements in male of Trechalea
lomalinda.

Leg segment

I

II

III

IV

Femur

10.3

9.6

7.9

11.0

Tibia-patella

13.9

12.2

8.9

11.9

Metatarsus

9.9

9.0

7.4

11.5

Tarsus

6.1

6.0

5.8

7.9

Total

40.2

36.8

30.0

42.3

CARICO^THE GENUS TRECHALEA (TRECHALEIDAE)

255

Table 21. — Leg measurements in female of Tre-

chalea lomalinda.

Leg segment

I

II

III

IV

Femur

9.6

9.8

8.1

10.8

Tibia-patella

12.3

11.7

9.0

11.9

Metatarsus

8.5

8.3

7.3

11.6

Tarsus

5.3

5.8

5.9

7.6

Total

35.7

35.6

30.3

41.9

(73°22'W:3®18'N), 300 m, from three collections by B.
T. Carroll: 1 5 April 1 986 (holotype) (CAS); March 1 989,
19 (CAS); 7 March 1986, 5<5 29 1 juv. (JEC). (Map 3).

Trechalea trinidadensis , new species
Figures 73, 74; Map 3

Type.— The holotype is an adult female from
Port-of-Spain, Trinidad, collected by Erik N.
Kjellesvig-Waering on 28 May 1968, deposited
in the American Museum of Natural History.

Etymology.— The name means ‘Trom Trini-
dad” taken from the name of the type locality.

Diagnosis.— This species is characterized by
the unique long spines on the pedipalps, the spi-
nation pattern on the ventral side of the tibiae
and details of the genitalia (Figs. 73, 74).

Description.— (Holotype). Carapace
moderately low, length 4.0, width 3.8, pattern
obscured with light marginal area evident, black
around each eye but none coalescing; sternum
light, unmarked, length 2.05, width 2.00; labium
moderately dark, light on apical margin, length
0.87, width 0.77. Clypeus height 0.4 1 , width 1 .90.
Anterior row straight, eye measurements in Ta-
ble 1 . Chelicerae moderately dark, rubbed of se-
tation, three retromarginal teeth, equidistant,
proximal one smallest, distal two equal in size.
Legs IV-II-I-III, ventral macrosetae pairs on tib-
iae are 1-5, II- 5, III- 3, IV-2; measurements in
Table 22, no pattern discernible. Pedipalp with
macrosetae longer than tibiae. Abdomen length
5.1, marked with a distinct but irregular pattern

Table 22. — Leg measurements in female of Tre-
chalea trinidadensis.

Leg segment

I

II

III

IV

Femur

6.0

6.1

5.1

7.0

Tibia-patella

7.6

7.5

5.9

7.7

Metatarsus

5.2

5.4

4.8

7.5

Tarsus

3.2

3.5

3.5

4.5

Total

22.0

22.5

19.3

26.7

dorsally, light ventrally. Epigynum (Figs. 73, 74)
with the pma extending around the mf to form
a squared U-shaped rim; the mf pale and bulbous
anteriad, narrowed and dark posteriad.

Specimens examined and distribution. — Known only
by the single type specimen. (Map 3).

ACKNOWLEDGMENTS
Collections were loaned from the following
sources (personal collections and museums) and
appreciation is extended to them; N. 1. Platnick
(American Museum of Natural History), H. W.
Levi (Museum of Comparative Zoology), J. A.
Coddington (National Museum of Natural His-
tory), C. Rollard (Museum National d'Histoire
Naturelle), K. H. Hyatt, F. R. Wanless, and P.

D. Hillyard (The Natural History Museum, Lon-
don), M. Moritz (Museum fur Naturkunde der
Humboldt-Universitat zu Berlin), D. Ubick
(California Academy of Sciences), E. Maury
(Museo Argentine de Ciencias Naturales), M. E.
Galiano, E. H. Buckup (Museo de Ciencias Na-
turais, Porto Alegre, Rio Grande do Sul), Z. A.
de Castellanos (Museo de la Plata), A. Martelli
(Museo Zoologico della Specola, Florence), H.
Reichardt (Museu de Zoologica Universidade de
Sao Paulo), H. Dybas, M. Prokop (Field Museum
of Natural History), C. E. Valerio (Museo Zool-
ogia Universitaria Costa Rica), J. Adis (Institute
National de Pesquisas Amazonia, L. Aviles (Mu-
seu Equitoriano de Ciencias Naturales, Ecuador),
Anna Timothea da Costa (Museu Nacional de
Rio de Janeiro).

Thanks are also extended to the following for
consultation and/or editorial review: J. A. Cod-
dington, C. D. Dondale, C. E. Griswold, H. W.
Levi, B. D. Opell, N. I. Platnick, P. Sierwald, F.

E. Walker.

Appreciation for translations: H. D. Cameron,
R. L. Germain, H. H. Lang, R. H. White.

Appreciation for help in collecting: F. C. Bap-
tista, N. A., E. L. and J. K. Carico, J. A. Cod-
dington, F. A. Coyle, W. G. Eberhard, E. W.
Minch.

Partial support was from Public Health Ser-
vice Grant AI-01944 to H. W. Levi, and Faculty
Development Grants from Lynchburg College.

LITERATURE CITED

Badcock, H. D. 1932. Reports of an expedition to
Paraguay and Brazil in 1926-1927, etc. J. Linn. Soc.
(Zoology), 38:1-48.

Bertkau, P. 1 880. Verzeichniss der von Prof Ed. Van

256

THE JOURNAL OF ARACHNOLOGY

Beneden auf seiner im Auftrage der Belgischen, . .
Mem. Cour. Acad. Beige, 43:1-129.

Bonnet, P. 1955-1959. Bibliographia Araneorum.
2:1-5058.

Brady, A. R. 1962. The spider genus Sosippus in
North America, Mexico, and Central America (Ara-
neae, Lycosidae). Psyche, 69:129-164.

Brignoli, P. M. 1983. A Catalogue of the Araneae
Described Between 1 940 and 1981. (P. Merrett, ed.).
Manchester Univ. Press. 755 pp.

Bruce, J. A. & J. E. Carico. 1988. Silk use during
mating in Pisaurina mira (Walckenaer) (Araneae,
Pisauridae). J. Arachnol., 16:1-4.

Cambridge, F. O. Pickard-. 1902. Arachnida. Ara-
neida. In Biologia Centralia-Americana, Zoologia.
London, 2:313-424.

Cambridge, F. O. Pickard-. 1903. On some new spe-
cies of spiders belonging to the families Pisauridae
and Senoculidae; with characters of a new genus.
Proc. Zook Soc. London. 1903:151-168.

Cambridge, O. Pickard-. 1876. Catalogue of a col-
lection of spiders made in Egypt, with descriptions
of new species and characters of a new genus. Proc.
Zool. Soc. London. 1876:541-630.

Cambridge, O. Pickard-. 1881. On some new genera
and species of Araneida. Proc. Zool. Soc. Lond.,
1881:765-775.

Cambridge, O. Pickard-. 1896. Arachnida. Araneida.
In Biologia Centralia-Americana, Zoologia. Lon-
don, 1:161-224.

Cambridge, O. Pickard-. 1897. Arachnida. Araneida.
In Biologia Centralia-Americana, Zoologia. Lon-
don, 1:225-232.

Cambridge, O. Pickard-. 1898. Arachnida. Araneida.
In Biologia Centralia-Americana, Zoologia. Lon-
don, 1:233-288.

Capocasale, R. M. 1 982. Las especies del genero Por-
rimosa Roewer, 1959 (Araneae, Hipposinae). J. Ar-
achnol., 10:145-156.

Caporiacco, L. di. 1947. Diagnosi preliminari di spe-
cie nuove di aracnidi della Guiana Britannica. Mon-
itore Zool. Italia, 56:20-24.

Caporiacco. L. di. 1 948. Arachnida of British Guiana
collected by Prof. Beccari. . . Proc. Zool. Soc. Lon-
don. 118:607-747.

Carico, J. E. 1972. The Nearctic spider genus Pi-
saurina (Pisauridae). Psyche, 79:295-310.

Carico, J. E. 1973. The Nearctic species of the genus
Dolomedes (Araneae: Pisauridae). Bull. Mus. Comp.
Zool., 144:435-488.

Carico, J. E. 1981. The Neotropical spider genera
Architis and Staberius (Pisauridae). Bull. American
Mus. Nat. Hist., 170:140-153.

Carico, J. E. 1985. Description and significance of
the juvenile web of Pisaurina mira (Walck.) (Ara-
neae: Pisauridae). Bull. British Arachnological Soc.,
6:295-296.

Carico, J. E. 1986. Trechaleidae: A “new” American
spider family (abstract), p. 305, In Proceedings of

the 9th International Congress of Arachnology. (W.
G. Eberhard, Y. D. Lubin and B. C. Robinson, eds.).
Smithsonian Inst. Press, 333 pp.

Carico, J. E. 1993. Taxonomic notes on the genus
Architis (Araneae, Pisauridae) and status of the ge-
nus Sisenna Simon. J. Arachnol., 21:202-204.

Carico, J. E., J. Adis & N. D. Penny. 1985. A new
species of Trechalea (Pisauridae: Araneae) from
Central Amazonian inundation forests and notes on
its natural history and ecology. Bull. British Arach-
nol. Soc., 6:289-294.

Carico, J. E. & E. W. Minch. 1981. A new species of
Trechalea (Pisauridae) from North America. Bull.
American Mus. Natur. Hist., 170:154-156.

Chamberlin, R. V. 1916. Results from the Yale Pe-
ruvian expedition of 1 9 1 1 . Bull. Mus. Comp. Zool.,
60:177-299.

Coddington, J. A. & H. W. Levi. 1991. Systematics
and evolution of spiders (Araneae). Annu. Rev. Ecol.
Syst., 22:565-592.

Davies, V. T. 1982. Inola nov. gen., a web-building
pisaurid (Araneae: Pisauridae) from northern Aus-
tralia with descriptions of three species. Mem.
Queensland Mus., 20:479-487.

Dondale, C. D. & J. H. Redner. 1978. Revision of
the Nearctic wolf spider genus Schizocosa (Aranei-
da: Lycosidae). Canadian Ent., 110:143-181.

Dondale, C. D. & J. H. Redner. 1983a. Revision of
the wolf spiders of the genus Arctosa C. L. Koch in
North And Central America (Araneae: Lycosidae).
J. Arachnol., 1 1:1-30.

Dondale, C. D. & J. H. Redner. 1983b. The wolf
spider genus Allocosa in North and Central America
(Araneae: Lycosidae). Canadian Ent., 1 15:933-964.

Dondale, C. D. 1986. The subfamilies of the wolf
spiders (Araneae: Lycosidae). Actas X Congr. Int.
Arachnol. Jaca/Espana., 327-332.

Exline, H. 1960. Rhoicinine spiders (Pisauridae) of
western South America. Proc. California Acad. Sci.,
IV series, 29:577-620.

Forster, R. R. et al. 1990. A proposal and review of
the spider family Synotaxidae (Araneae, Araneo-
idea), with notes on theridiid interrelationships. Bull.
American Mus. Nat. Hist., 193:1-116.

Giebel, C. G. 1863. Drei und zwanzig neue und einige
bekannte Spinnen der Hallischen Sammlung. Zeits.
Gesam. Naturw., 21:306-328.

Griswold, C. E. 1993. Investigations into the phy-
togeny of the lycosoid spiders and their kin (Arach-
nida, Araneae, Lycosoidea). Smithsonian Contri-
butions to Zoology, 539:1-39.

Karsch, F. 1878. Exotisch araneologisches. Zeits.
Gesam. Naturw., 51:323-333, 771-826.

Karsch, F. 1879. Arachnologische Beitrage. Zeits.
Gesam. Naturw., 52:534-562.

Kaston, B. J. 1938. Family names in the order Ara-
neae. American Midi. Natur., 20:638-646.

Kaston, B. J. 1 948. Spiders of Connecticut. Bull. Con-
necticut St. Geol. Nat. Hist. Surv., 70:1-874.

CARICO-THE GENUS TRECHALEA (TRECHALEIDAE)

257

Koch, C. L. 1 848. Die Arachniden. Fiintzehnter Band.
Numberg, 210 pp.

Keyserling, E. G. 1877. Ueber amerikanische Spin-
nenarten der Unterordnung Citigradae. Verb. zooL-
bot. Ges. Wien, 27:85-96.

Keyserling, E. G. 1891. Die Spinnen Amerikas. Bras-
ilianische Spinnen. Vol. 3. Numberg.

Kraus, O. 1955. Spinnen aus El Salvador (Arach-
noidea, Araneae). Abhand. der Senkenberg. Natur-
for. Gesel, 493:1-121.

Lehtinen, P. T. 1967. Classification of the Cribellate
spiders and some allied families, with notes on the
evolution of the suborder Araneomorpha. Ann. Zool.
Fennici, 4:199-468.

Lenler-Eriksen, P. 1969. The hunting- web of the young
Pisaura mirabiiis. J. ZooL, 157:391-398.

Mello-Leitao, C. 1926. Algumas aranhas do Brasil
meridional. Bol. Mus. Nac. Rio-de-Janeiro, 2:1-18.

Mello-Leitao, C. 1931. Arachnidos do Rio Grande
do Sul. Bol. Biol. Rio de Janeiro, Sao Paulo, 17:10-
14.

Mello-Leitao, C. 1 940a. Aranhas do Xingu colhidas
pello Dr. Henry Leonardos. Anais Acad. Brasil.
Cien., 12:2-32.

Mello-Leitao, C. 1940b. Aranhas do Parana. Arq.
Inst. Biol. Sao Paulo, 11:235-257.

Mello-Leitao, C. 1 943. Alguns pisauridas e tomisidas
do Brasil. Rev. Chilena Hist. Nat, 45:164-172.

Mello-Leitao, C. 1947. Aranhas do Parana e Santa
Catarina. . . Rio Archos Mus. Parananense, 6:231-
304.

Merrett, P. 1988. Notes on the biology of the neo-
tropical pisaurid, Ancylometes bogotensis (Keyser-
ling) (Araneae: Pisauridae). Bull. British Arachnol.

Soc., 7:197-201.

Nentwig, W. 1985. Architis nitidopilosa, a neotrop-
ical pisaurid with a permanent catching web (Ara-
neae, Pisauridae). Bull. British Arachnol. Soc., 6:297-
303.

Nicolet, H, 1849. Aracnidos, In Gay C., Historia
fisica y politica de Chile. Zoologia, III, Pp. 3 1 9-543.

Petty, M. 1833. Arachnides Brasilienses. Pp. 191-
209, In Delectus Animalium Articulatomm quae in
itinere per Braziliam ann. 1817 et 1820 colligemnt.
(J. B. deSpix & F. P. Martins, eds.). Monachii.

Petmnkevitch, A. 1911. A synonymic index-cata-
logue of spiders. . . Bull. American Mus. Nat. Hist,,
29:1-791.

Petmnkevitch, A. 1925. Arachnida from Panama.
Trans. Conneticut Acad. Arts Sci., 27:51-248.

Platnick, N. 1. 1979. A revision of the spider genus
Barrisca (Araneae, Rhoicininae). J. Arachnol, 6:2 1 3-
217.

Platnick, N. 1. 1989. Advances in Spider Taxonomy
198 1-1987, etc. (P. Merrett, ed.). Manchester Uni v.
Press, 673 pp.

Pocock, R. 1. 1900. Myriapoda and Arachnida. In
Report on a collection made by F. V. McConnell

and J. J. Quelch at Mount Roriama in British Guina.
Trans. Linn. Soc. London, 2:64-71.

Roewer, C. F. 1954. Katalog der Araneae. 2a, Pp. 1-
923.

Roth, V. D. 1985. Spider Genera of North America.
American Arach. Soc. 172 pp.

Schmitt, A. 1992. Conjectures on the origins and
functions of a bridal veil spun by the males of Cu~

piennius coccineus (Araneae, Ctenidae). J. Arach-
noL, 20:67, 68.

Sierwald, P. 1989. Morphology and ontogeny of fe-
male copulatory organs in American Pisauridae, with
special reference to homologous features (Arachni-
da: Araneae). Smithsonian Contrib. Zool, 484:1-
24.

Sierwald, P. 1990. Morphology and homologous fea-
tures in the male palpal organ in Pisauridae and
other spider families, with notes on the taxonomy
of Pisauridae (Arachnida: Araneae). Nemouria, Occ.
Pap. Delaware Mus. Natur. Hist, 35:1-59.

Simon, E. 1890. Etudes arachnologiques. 22^ Me-
moire. XXXIV. Etude sur les Arachnides de 1’ Ye-
men. Ann. Soc. Ent. France, 10:77-124.

Simon, E. 1898a. Histoire Naturelle des Araignees.
Tome 2, Fascicule 2, Paris, Pp. 193-380.

Simon, E. 1898b. Descriptions d’arachnides nou-
veaux des families des Agelenidae, Pisauridae, Ly-
cosidae et Oxyopidae. Ann. Soc. Ent. Beige, 42:5-
34.

Simon, E. 1903. Histoire Naturelle des Araignees.

Tome 2, Fascicule 4, Paris, Pp. 669-1080.

Soares, B. A. M. & H. F. Camargo. 1948. Aranhas
cologidas per la fundagao Brasil Central. Bolm Mus.
Para. Emilio Goeldi, 10:355-409.

Taczanowski, L. 1873. Les Araneides de la Guyane
frangaise. Horae Soc. ent. Ross., 10, Pp. 56-115, pi
11.

Thorell, T. 1869. On European spiders. Part 1. Re-
view of the European genera of spiders, preceded
by some observations on zoological nomenclature.
Nova Acta Reg. Soc. Sci. Upsaliensis, 7:1-242.

Thorell, T. 1895. Descriptive Catalogue of the Spi-
ders of Burma. British Museum. 406 pp.

Tullgren, A. 1905. Araneida from the Swedish ex-
pedition through the Gran Chaco and the Cordil-
leras. Ar. Zool, 2:1-81.

Van Berkum, F. H. 1982, Natural history of a trop-
ical, shrimp-eating spider (Pisauridae). J. Arachnol,
10:117-121.

Wojcicki, J. P. 1990. The reproductive biology of
Dolomedes triton (Walckenaer) (Araneae; Pisauri-
dae). American Arachnol, 42:14. (abstract).

Yaginuma, T. 1991. A new genus, Shinobius, of the
Japanese pisaurid spider (Araneae: Pisauridae). Acta
Arachnologica, 40:1-6.

Manuscript received 18 May 1993, revised 20 August
1993.

1993. The Journal of Arachnology 21:258-260

RESEARCH NOTE

COHABITATION AND COPULATION IN IXEUTICUS MARTIUS
(ARANEAE, AMAUROBIIDAE)

Ixeuticus martins (Simon 1899) is a solitary,
medium size cribellate spider. In spite of its very
high abundance its biology is poorly known.
However, I observed some cases of coexistence
and tolerance between males and juvenile fe-
males for this species in the field. Male-juvenile
female cohabitation has been long known in the
Gnaphosidae and Clubionidae (Bristowe 1941),
and it has shown to be common in the Salticidae
and Araneidae (Jackson 1986). Austad (1984)
and Jackson (1986) analysed this phenomenon
and suggested it is a mating tactic among spiders.
There are few reports on sexual behavior in the
Amaurobiidae: precopulatory communication
(KraflTt 1978; Leborgne 1984), copulation (Ger-
hardt 1923, 1924; Gregg 1961), and cohabitation
(Jackson 1986). I therefore thought it was im-
portant to communicate my observations on co-
habitation and copulation in this Neotropical
spider species.

Three male-subadult female pairs were ob-
served cohabiting in their webs during 1989: pair

1 in Marindia, Canelones (December 2) and pairs

2 and 3 in Prado, Montevideo (both December
8). The web of pair 2 was contiguous with an-
other web which contained another subadult fe-
male; the male was twice seen passing from one
web to the other. No attacks were seen, but an-
other smaller male was observed in the periphery
of both webs. With the exception of this male,
all individuals were captured and observed un-
der laboratory conditions. During observations
of copulation the room temperature was 27° C
and relative humidity 61%.

Both spiders of pair 2 remained in contact in-
side the refuge built by the female in the lab. Five
days after capture, at 1230 h (counted as minute
zero of the observation), the female molted dur-
ing a period of 1 1.4 min. At 23 min, the male
repeatedly tapped the female and the nearby
exuvia, and placed himself in various positions
in relation to the female (ventral-ventral, face to
face, etc.). Finally, the male placed himself on
the female’s side, both with the ventral zone up-
wards (Fig. 1), Palpal insertions started at 44.8

min, while the male vibrated his abdomen sag-
ittally. The copulatory pattern was as follows:
insertion with one palp— palpal withdrawal, and
chewing-like movements on it— new insertion
with the same palp and chewing-like movements
(1-8 successive insertions; mean 3.7)— the male
walked around the female through the silk threads
up to the other side (mean delay: 1 min)— in-
sertion with the other palp— disinsertion, chew-
ing-like movements, reinsertion and so on. The
male carried out approximately 140 palpal in-
sertions and changed sides 38 times during a 144
min period. Mean insertion duration and mean
period spent in one side were estimated to be 40
s and 3.3 min, respectively. Three times the male
appeared disoriented when walking around the
female (total delay: 18 min). The male also
adopted atypical positions (ventral - ventral,
ventral - dorsal, face to face).

After copulation the male immediately made
a sperm induction. The palps alternated four
times in contacting the sperm, during 10.5 min.
At 2 1 6.7 min the female placed herself in contact
with the male, and both remained immobile until
the following day. During days 2 and 4 the female
ate the male, and after some time after she had
offspring.

Females of pairs 2 and 3 molted during the
night (at days 1 and 9, respectively) and conse-
quently copulations were not seen. Thereafter,
both males and females of each pair remained
in contact 4 and 3 days, respectively. Female
attacks were later observed in both pairs. After-
wards, the male of pair 2 was taken out and
placed with the female captured beside pair 2.
This female had molted five days before (9 days
after capture). Copulation occurred on the first
day; although observation was incomplete, it
proved to be similar to the copulation of pair 1 .
The male and female remained in contact during
4 days following copulation. All the observed
females produced young.

Cohabitation in Ixeuticus martins would be
determined both by the male’s capacity to rec-
ognize subadult females and by tolerance toward

258

RESEARCH NOTES

259

Figure 1 .—’Copulatory position of Ixeuticus martius,
pair 1, ventral view„ The male (shaded spider) inserts
its left palp on the epigynum of the female (white spi-
der). The web’s silk threads are omitted.

males shown by females. Resident males prob-
ably keep other males away from females (female
guarding)— based on field observation of the be-
havior shown by the two males (pair 2), and the
presence of a single male in the other webs. The
end result of copulation suggests that cohabita-
tion is a mating tactic: the male safeguards his
first insemination and prevents the female from
copulations with other males. Austad (1984) an-
alyzed mating tactics and sperm priority patterns
in spiders. He hypothetized that males insemi-
nating first (“first male sperm priority”) should
have greater success than subsequent males in
spiders with “conduit spermathecae”. Support-
ing this prediction, Jackson (1986) reported 161
spider species showing precopulatory cohabita-
tion, including six Amaurobiidae species {sensu
Lehtinen 1967). Only seven of these 161 species
had “cul de sac spermathecae”. On the other
hand, postcopulatory cohabitation is rare and it
has not been reported previously in Amaurobi-
idae, in spite of observations of this sort in Die-
tynidae (Montgomery 1903; Starr 1988). A fe-
male that copulated five days after molting
suggested that the 2-4 day tolerance period would
be more related to copulation than to adult fe-
male age.

The copulatory position of Ixeuticus martius
is similar to that of Metaltella simoni (Prandi
pers. comm.) but differs from the typical Amau-

robiidae position: the male placed ventrally to
female, following a face-to-face encounter (Ger-
hardt 1923, 1924; Gregg 1961; Leech 1972). The
copulatory pattern of I. martius (both numerous
brief insertions and side changes) is also unusual
for the Amaurobiidae’s pattern (few and pro-
longed insertions, multiple ejaculations for each
insertion). These differences may not be attrib-
uted to the immediate postmolting female state
because the copulation with a mature female was
similar. The Australian amaurobiid spider Ixeu-
ticus longinuus makes an intermediate number
of palpal insertions with long pauses (Gregg 1961)
and would be the most similar to that of L mar-
tius. The literature was revised bearing in mind
the controversial placement of several species
among the families Amaurobiidae, Dictynidae
and Desidae (Lehtinen 1967; Forster 1970; For-
ster & Wilton 1973), and it showed similar po-
sition and general copulatory pattern in Dictyn-
idae (Jackson 1979; Starr 1988) as compared to
Amaurobiidae. Singularities showed by 1. mar-
tius could be useful in future (and urgently need-
ed) studies on the systematics of these families.

I wish to thank R. M, Capocasale for identi-
fying specimens. C. Viera, E. Gudynas, R. M.
Capocasale, R. R. Jackson and S. N. Austad made
critical readings of the manuscripts. 1. Trabal
prepared the English version.

LITERATURE CITED

Austad, S. N. 1984. Evolution of sperm priority pat-
terns in spiders. Pp. 223--249, In Sperm competition
and the evolution of animal mating systems. (R. L.
Smith, ed.). Academic Press, Orlando, Florida.
Bristowe, W.S. 1941, The comity of spiders. Ray
Society, London.

Forster, R. R. 1970. The spiders of New Zealand.
Part III. Desidae, Dictynidae, Hahniidae, Amau-
robiidae, Nicodomidae. Otago Mus. Bull, 3:1-184.
Forster, R. R. & C. L. Wilton. 1973. The spiders of
New Zealand. Part IV. Agelenidae, Stiphidiidae,
Amphinectidae, Amaurobiidae, Neolanidae, Cten-
idae, Psecridae. Otago Mus. Bull., 4:1-309.
Gerhardt, U. 1923. Weitere sexualbiologische Un-
tersuchung an Spinnen. Arch, Naturgesch., 89:1-
225.

Gerhardt, U. 1 924. Weitere Studien iiber die Biologic
der Spinnen. Arch. Naturgesch,, 90:85-192.

Gregg, M. 1961. The mating of Ixeuticus longinuus.
Proc. Roy. ZooL Soc. New South Wales, 1558-59:
85-86.

Jackson, R. R. 1979. Comparative studies of Dictyna
and Mallos (Araneae, Dictynidae). IL The relation-
ship between courtship, mating, aggression and can-

260

THE JOURNAL OF ARACHNOLOGY

nibalism in species with differing types of social or-
ganization. Rev. ArachnoL, 2(3):103“132.

Jackson, R. R. 1986. Cohabitation of males and ju-
venile females: a prevalent mating tactic of spiders.
J. Nat. Hist., 20:1193-1210.

Krafft, B. 1978. The recording of vibratory signals
performed by spiders during courtship. Symp. Zool.
Soc. London, 42:59-67.

Leborgne, R. 1984. Specificite d’organisation des
comportements de cour des males de trois especes
d" Amaurobius (Araneae, Dictynidae). Rev. Arach-

noL, 5:239-245.

Leech, R. 1972. A revision of the Neartic Amauro-
biidae (Arachnida: Araneida). Mem. Ent. Soc. Can-
ada, 84:1-182.

Lehtinen, P. T. 1967. Classification of the cribellate
spiders and some allied families with notes on the

evolution of the suborder Araneomorpha. Ann. Zool.
Fennici, 4:199-468.

Montgomery, T. H. Jr. 1903. Studies on the habits
of spiders, particularly those of the mating period.
Nat. Sci. Philadelphia, 55:59-149.

Starr, C. K. 1988. Sexual behavior in Dictyna volu-
cripes (Araneae, Dictynidae). J. Arachnol., 16:321-
330.

Fernando G. Costa: Division Zoologia Exper-
imental, Institute de Investigaciones Biologi-
cas Clemente Estable, Av. Italia 3318, Mon-
tevideo, Uruguay.

Manuscript received 4 January 1993, revised 18 May
1993.

INSTRUCTIONS TO AUTHORS

Manuscripts are preferred in English but may be ac-
cepted in Spanish, French or Portuguese subject to avail-
ability of appropriate reviewers. Authors whose primary
language is not English may consult the Associate Editor
for assistance in obtaining help with English manuscript
preparation. All manuscripts should be prepared in general
accordance with the current edition of the Council of Bi-
ological Editors Style Manual unless instructed otherwise
below. Authors are advised to consult a recent issue of the
Journal of Arachnology for additional points of style.
Manuscripts longer than 1 500 words should be prepared
as Feature Articles, shorter papers as Research Notes. Send
three identical copies of the typed material together with
copies of illustrations to the Associate Editor of the Journal
of Arachnology:

Gary L. Miller, Associate Editor
Dept, of Biology
The University of Mississippi
University, Mississippi 38677 USA
[Telephone: (601)-232-7495; FAX: (601)-232-5144;
Electronic mail: BYMILLER@UM.CC.OLEMISS.EDU].
Correspondence relating to manuscripts should also be
directed to the Associate Editor. After the manuscript has
been accepted, the author will be asked to submit the
manuscript on a computer disc in MS DOS WordPerfect
(preferably) or ASCII text file.

FEATURE ARTICLES

Title page.— The title page will include the complete
name, address, and telephone number of the author with
whom proofs and correspondence should be exchanged, a
FAX number and electronic mail address if available, the
title in capital letters, and each author’s name and address,
and the running head (see below).

Abstract.— The heading in capital letters should be placed
at the the beginning of the first paragraph set off by a
period. In articles written in English, a second abstract in
an acceptable language may be included pertinent to the
nationality of the author(s) or geographic region(s) em-
phasized in the article. A second abstract in English must
be included in articles not written in the latter language.

Text.— Double-space text, tables, legends, etc. through-
out. Three levels of heads are used. The first level (METH-
ODS), RESULTS, etc.) is typed in capitals and on a sep-
arate line. The second level head begins a paragraph with
an indent and is separated from the text by a period and
a dash. The third level may or may not begin a paragraph
but is underlined and separated from the text by a colon.
Use only the metric system unless quoting text or refer-
encing collection data. All decimal fractions are indicated
by the period regardless of language of the text.

Citation of references in the text: Cite only papers already
published or in press. Include within parentheses the sur-
name of the author followed by the date of publication. A
comma separates multiple citations by the same author(s)
and a semicolon separates citations by different authors,
e. g., (Smith 1 970), (Jones 1 988; Smith 1 993), (Smith 1 986,
1987; Smith & Jones 1989; Jones et al. 1990).

Literature cited section.— Use the following style:
Lombardi, S. J. & D. L. Kaplan. 1990. The amino acid
composition of major ampullate gland silk (dragline) of
Nephila clavipes (Araneae, Tetragnathidaei J. Arach-
nol., 18:297-306.

Krafft, B. 1 982. The significance and complexity of com-
munication in spiders. Pp. 15-66, In Spider Commu-
nications: Mechanisms and Ecological Significance. (P.
N. Witt & J. S. Rovner, eds.). Princeton University
Press, Princeton, New Jersey.

Footnotes.— Footnotes are permitted only on the first
printed page to indicate current address or other infor-
mation concerning the author. All footnotes are placed
together on a separate manuscript page.

Running head.— The author’s sumame(s) and an abbre-
viated title should be typed all in capital letters and must
not exceed 60 characters and spaces. The running head
should be placed near the top of the title page.

Taxonomic articles.- Consult a recent taxonomic article
in the Journal of Arachnology for style or contact the As-
sociate Editor.

Tables.— Each table, with the legend above, should be
placed on a separate manuscript page. Only horizontal
lines (usually three) should be included. Use no footnotes;
instead, include all information in the legend. Make no-
tations in the text margins to indicate the preferred loca-
tion of tables in the printed text.

Illustrations.— Address all questions concerning illus-
trations to the Editor of the Journal of Arachnology:

James W. Berry, Editor

Dept, of Biological Sciences

Butler University

Indianapolis, Indiana 46208 USA

[Telephone (317)-283-9344; FAX (317)-283-9519;

Electronic mail: BERRY@BUTLERU]

All art work must be camera-ready for reproduction. In
line drawings, pay particular attention to width of lines
and size of lettering when reductions are to be made by
the printer. Multiple photos assembled on a single plate
should be mounted with only a minimum of space sepa-
rating them. In the case of multiple illustrations mounted
together, each illustration must be numbered sequentially
rather than given an alphabetic sequence. Written on the
back should be the name(s) of author(s) and an indication
of top edge. The author should indicate whether the illus-
tration should be one column or two columns in width.
The overall dimensions should be no more than 1 1 inches
(28 cm) X 14 inches (36 cm). Larger drawings present
greater difficulty in shipping and greater risks of damage
for which the JOA assumes no responsibility. Make no-
tations in the text margins to indicate the preferred po-
sition of illustrations in the printed text.

Legends for illustrations should be placed together on
the same page(s) and separate from the illustrations. Each
plate must have only one legend, as indicated below:

Figures \-A.—A-us x-us, male from Timbuktu: 1. left
leg; 2. right chelicera; 3. dorsal aspect of genitalia; 4. ven-
tral aspect of abdomen.

Figures 27-34.— Right chelicerae of species oiA-us from
Timbuktu: 27, 29, 31, 33. dorsal views; 28, 30, 32, 34,
prolateral views of moveable finger; 27, 28. A-us x-us
holotype male; 33,34. A-us y-us male. Scale 1.0 mm.

Assemble manuscript for mailing.— Assemble the sep-
arate sections or pages in the following sequence; title page,
abstract, text, figure legends, footnotes, tables with legends,
figures.

Page charges and reprints.— The current charge per
journal page will be assessed as follows: Manuscripts in
the appropriate word processing program $20/page; others
$3 5/page. Reprints are available only from the printer and
should be ordered when the author receives the galley
proof pages. Waiver of page charges: In exceptional cases,
the Editor will consider requests for a waiver of page charg-
es for manuscripts which have been accepted for publi-
cation. Requests should be made in writing to the Editor.
The author must certify that neither grant nor institutional
funds are available to cover all of the publication costs.
The Society will normally waive no more than 80% of the
page charges.

RESEARCH NOTES

Instructions above pertaining to feature articles apply
also to research notes, except that abstracts and most head-
ings are not used and the author’s name and address follow
the Literature Cited section.

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 21 Feature Articles NUMBER 3

Ultrastructure of Cribellate Silk of Nine Species in Eight Families and Pos-
sible Taxonomic Implications (Araneae: Amaurobiidae, Deinopidae,
Desidae, Dictynidae, Filistatidae, Hypochilidae, Stiphidiidae, Tengel-
lidae) William Eberhard and Flory Pereira 161

Studies on Species of Holarctic Pardosa Groups (Araneae, Lycosidae). V.
Redescription of Pardosa wasatchensis Gertsch and Description of a
New Species from Utah Torbjorn Kronestedt 175

Newly-discovered Sociality in the Neotropical Spider Aebutina binotata

Simon (Dictynidae?) Leticia Aviles 184

DNA Sequence Data Indicates the Polyphyly of the Family Ctenidae (Ara-
neae) Kathrin C. Huber, Thomas S. Haider, Manfred W. Muller, Bern-
hard A. Huber, Rudolf J. Schweyen, and Friedrich G. Barth 194

Taxonomic Notes on the GtnmArchitis (Araneae, Pisauridae) and the Status

of the Genus Sisenna Simon James E. Carico 202

Two New Species of the Genus Lyssomanes (Hentz) from the Cape Region,

B.C.S., Mexico Maria Luisa Jimenez and Armando Tejas 205

The Orb-weaver Genus Kaira (Araneae: Araneidae) Herbert W. Levi .... 209

Revision of the Genus Trechalea Thorell (Araneae, Trechaleidae) with a
Review of the Taxonomy of the Trechaleidae and Pisauridae of the
Western Hemisphere James E. Carico 226

Research Note

Cohabitation and Copulation in Ixeuticus martius (Araneae, Amaurobiidae)

Fernando G. Costa 258

40 242 SI XL II

OB/30/94 \ 198115 t||

zz, »

>

2

/

>* 2 > XjVASt^^ 2

R I ES’ SMITHSONIAN_INSTITUTION “N0tinilXSNI_NVIN0SHilWs‘^S3 I W Va 8 n“LI B RAR I ES”'SMI'^

iisNi NvmosHims S3idvyan libraries Smithsonian institution NoiiniiiSNi nvin

^ . 2 r- ^ 5 C 2 r-

o ;;; 2 ..m 2

s^rvcu,

&

R I Es‘”sMITHSONiAN~INSTlTUTION NOtJ.niliSNl”NVINOSHillAIS S3 1 MVa 8 n“LI B RAR I ES*”sMn

2 .V'. ^ ^ ^ .cr.*'.- ^

C/) _ , . _

^ I i I I I I I Vc

liSNI NVIN0SHilWs“’s3iavaan^LIBRARIEs”sMITHS0NIAN INSTITUTION'^NOliniliSNI NVI^

^ Z ^ ^ ~

LiJ ><t^^V4rS. /ft y^ASp^JX W /ft y<^i2vJX ^

I?

q:

H

|/ C

ss^ 5 5 5 “ '<ai»52^ o

R I ES^SMITHSONIAN^INSTITUTION N0liniliSNl|;^NVIN0SHilIAIS^S3 I MVa 8 Il'^LI B R AR I ES^SMI

P S 2 5 1% . i ra °

</>

m _

IISNI NVIN0SH1IIMS^S3 I B VB 9 n"“L 1 B R AR 1 ES SMITHSONIAN~INSTITUTION^NOIinillSNl“NVII

C/> . 2 V, W 2 V </i 2

<

z vfc. H /:FMirx:i^\ z k:^. .w h 2

I
tn
O

^ y' ~ Wi

R I ES^SMITHSONIAN JNSTITUTI0N‘”N0liniliSNI_NVIN0SHilWs‘”S3 I U VB 8 tif LI B RAR I ES^SMII

2 «i Z _| 2

liSNI NVINOSHimS S31BV99n LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNl NVlh
n Z r- 2 r- 2 -

w

70

>

:o

R I ES^SMITHSONIAN INSTITUTION^ NOliniliSNl NVIN0SHilWS^S3 I H V9 9 n*"!! B R AR I ES^SMI'

</> z .V 2 2 2: .V. </>

2 2 .< 2 .<

IISNI NVINOSHilWs‘^S3iavaan LIBRARIEs“’sMITHSONIAN institution NOliniliSNI NVII

C/> 5 c/5 2 ^

^ iJj xC^^vXTx /ft cj4 fz yf^^oy^X y^^^^r

I i § ••)^'' 5’ g

VN INSTITUTION NOliflilXSNI NVINOSHillNS SaidVaan LIBRARIES SMITHSONIAN INSTITI
— c/) _ _ Z *'• ^

^ I

2 W

(/)

(/)

C

5 o “ xjMsss^ o “

2 -J ^ 2*

\is S3iyvdan libraries Smithsonian institution NoiiniiiSNi nvinoshii^vs saiav

2 r- z r* 2 *“ 2

o 2 ..5 2 S o

> (ip 331 C >

(/)

?0

^ m ^ X£0iay^ m

\N’"lNSTITUTION^ NOlinillSNrNVINOSHillAIS S3 I a VB 8 n~LI B RAR I ES^^SMITHSONIAN^INSTIT
2 .V. c/> z S 5 S 2:

H "•

S (ft 3l) S ^ ^

£ W^y i i

Vs"’s3iaVaan^l-IBRARIEs“sMITHS0NIAN INSTITUTION NOliniliSNI NVIN0SHilWs‘”s3 1 BV

(/> r; c/) ^ c/J “ c/5

O pc^ — O 'N^2LSi>^ «—> Q

\N~INSTITUTION N0liniliSNl“*NVIN0SHilWS^S3 1 B Va 3 n“'LI B RAR I ES^SMITHS0n7aN“INSTIT

1“ z z_r“ z «”

CD

?a

3^

CO
PO
i>

_ z x^iVAsji^' rn N2\orH^ ^ fT8

MS^ S3 I MVa a n“LI B RAR I ES SMITHSONIAN*’lNSTITUTION^NOIinillSNl““NVINOSHimS^ S3 1 n\

CD . Z •- CD - Z • CD Z CD

5>'

(N“|NSTITUTI0N ’ N0liniliSNI_NVIN0SHilWS”S3 1 B VB a n“LI BRAR 1 ES"SMITHSONIAN^INSTIT

= CD ^ H CD = en _

CD

z

CD

c

H

O

Q x^osv^i>^ v-» «. O X^OSH^ ” o

Z -J z J 2 z

IS S3 1 a VB an libraries SMITHSONIAN institution NOlinillSNI NVINOSHII&MS S3IB\i
2 r* z f z r- 2

^N“INSTITUTI0N^ NOIinillSNrNVINOSHilV\IS S3 I BVB a n~LI B RAR I ES^SMITHSONIAN INSTIT
Z .V. CD z w z .y, g 2

MS^^Sa I d VB a LI B RAR l Es‘^SMITHS0NIAN^INSTITUTI0N'^N0linillSNI^NVIN0SHlll^Is‘^S3 \U\

CD = CD cD_r: . cd