TUTION

2

CO

NOIlflillSNI NVINOSHimS^

^ i \ ^

— ^ vi^osv^ „ - -

SMITHSONIAN INSTITUTION NOliniliSNl'^NVINOSHil WS ^S3 I MVH 8 11*
■ 2: r* 2 I

TUTION

± C/>

NOIlfliliSNI NVlNOSHillMS 1
z ^

B RAR I ES SMITHSONIAN

CO

C/) 2 to *.. 2 (/)

; SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIWS 1

<0 5 CO — (/)

• • ^ ><^ASOa7N.. LiJ ><d<VAr>v V III

I 2

NVINOSHIIIAIS S3 I dVBB 11
z

“ r~\

m ^ i

CO " £ (

ES SMITHSONIAN INSTITUTION
^ ^ ^ 2 » <

^ .^<ZTiT7r>v ^

NVINOSHilt^S^'

\ (O

\dh. LiJ

if)

NVINOSHill/MS !

in

i'^SMITHSONIAN

CO

I vj Ns4^ostA^ __

SMITHSONIAN INSTITUTION NOliniUSNI “‘nVINOSHJ.IWS

: >“ ^ H I-

TUTION

Z • ' “

NOIinXIXSNl~NVINOSHiIWS !

z: w

<C ^

vda n

B RAR I ES^SMITHSONIAN

CO
UJ

I

» \

SMITHSONI

)

INSTITUTION NOlinillSNI

CO

‘.rfaC'ffi’BjYj'l-X — /o'tjpWlpVV

roa W Jb^ O

Uk\

— V

5^ XjVASV^ ^

NVIN0SHimS^S3
(/)

/

Z1 'x
> '

dVdan^LIBRAR

>N \

>

in 2

ES PiMlTHCinNIAN I N.^TITIJTION

^NOliniliSNI^NVINOS
(/)

S SMITHSONIAN
in

INSTITUTION NOlinillSNI

in

NVINOSHIIIAIS

saiyvaan librar

ES SMITHS

in

UJ

ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 13

SPRING 1985

NUMBER 1

THE JOURNAL OF ARACHNOLOGY

EDITOR: Oscar F. Francke, Texas Tech University.

ASSOCIATE EDITORS: B. J. Kaston, San Diego State University.

William B. Peck, Central Missouri State University.

ASSISTANT EDITOR: James C. Cokendolpher, Texas Tech University.

EDITORIAL ASSISTANT: Scott A. Stockv^ell, Texas Tech University.

TYPESETTER: Sharon L. Robertson, Texas Tech University.

EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada.

William G. Eberhard, Universidad de Costa Rica.

Maria E. Galiano, Museo Argentino de Ciencias Naturales.

Willis J. Gertsch, American Museum of Natural History.

Neil F. Hadley, Arizona State University.

Vincent F. Lee, California Academy of Sciences.

Herbert W. Levi, Harvard University.

Emilio A. Maury, Museo Argentino de Ciencias Naturales.

William B. Muchmore, University of Rochester.

Martin H. Muma, Western New Mexico University.

Norman I. Platnick, American Museum of Natural History.

Gary A. Polis, Vanderbilt University.

Susan E. Riechert, University of Tennessee.

Michael E. Robinson, National Zoological Park
Vincent D. Roth, Southwestern Research Station.

Jerome S. Rovner, Ohio University.

William A. Shear, Hampden-Sydney College.

William J. Tietjen, Lindenwood College.

George B. Uetz, University of Cincinnati
Carlos E. Valerio, Universidad de Costa Rica.

Stanley C. Williams, San Francisco State University.

THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in Spring,
Summer, and Fall by The American Arachnological Society in cooperation with The

Graduate School, Texas Tech University.

Individual subscriptions, which include membership in the Society, are $25.00 for
regular members, $15.00 for student members. Institutional subscriptions to The Journal
are $35.00. Correspondence concerning subscriptions and memberships should be addres-
sed to the Membership Secretary (see back inside cover). Back issues of The Journal 2LYe
available from Dr. Susan E. Riechert, Department of Zoology, University of Tennessee,
Knoxville, TN 37916, U.S.A., at $10.00 for each number; the Index to volumes 1-10 is
available for $10.00. Remittances should be made payable to The American Arachnolog-
ical Society. Correspondence concerning undelivered issues should be addressed to the
Assistant Editor, Department of Biological Sciences, Texas Tech University, Lubbock,
Texas 79409, U.S.A.

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

Cutler, B. and D. T. Jennings. 1985. A revision of the Metaphidippus arizonensis group (Araneae,
Salticidae). J. ArachnoL, 13:1-8.

A REVISION OF THE METAPHIDIPPUS ARIZONENSIS
GROUP (ARANEAE, SALTICIDAE)

Bruce Cutler

1747 Eustis Street
St. Paul, Minnesota 55113

and

Daniel T, Jennings

Northeastern Forest Experiment Station
USDA Building, University of Maine
Orono, Maine 04469

ABSTRACT

The Metaphidippus arizonensis group contains two species, Metaphidippus arizonensis (Peckham
and Peckham) and M. helenae (Banks). The two species share similar genitalic morphology. M. ari-
zonensis occurs in both tall and shortgrass habitats in the central grasslands from southern Alberta to
eastern Minnesota south to Arizona and New Mexico. M. helenae occurs in the interior basins of North
Dakota, Utah, and Wyoming, and in saline marshlands of central California. Both species are rede-
scribed and M. glacialis (Scheffer) is synonymized with M. arizonensis.

INTRODUCTION

During the 1970’s we were independently engaged in biological studies oi Metaphidip-
pus arizonensis (Peckham and Peckham) at the southern and northern ends of its range
(these studies to be published separately). It became apparent that there was some
confusion in the taxonomy of the species, particularly concerning the status of M. glacial-
is (Scheffer). As a result, we decided to determine if indeed two species were represented.
In a search for similar species we found that only one other North American species is
closely related, M. helenae (Banks). It has been suggested that M. tillandsiae Kaston may
belong in this species group, but the genitalia are of a different type. This small species
group bears epigynal resemblances to two Siberian and Mongolian species described by
Prozynski (1979). He noted an external epigynal similarity to M. glacialis in hisDendry-
phantes biankii, but internally they are different. However, the internal epigynal structure
of his D. czekanowskii bears a close resemblance to the internal epigynal structures seen
in the M. arizonensis group. Both of these Palearctic species are represented by only three
female specimens; if males become available, they would be of considerable interest.

2

THE JOURNAL OF ARACHNOLOGY

METHODS

Certain measurements for statistical purposes were standardized. These were all done
on field collected females, males being relatively rarer in collections. Three populations of
M. arizonensis were tabulated separately. Fewer specimens of M. helenae were available,
so these measurements were pooled. All measurements throughout this paper are in
millimeters.

SYSTEMATICS

Metaphidippus F. O. P. - Cambridge 1901 .

Metaphidippus arizonensis group.

This small group may be readily distinguished from other Metaphidippus by genitalic
characters. Males have a retrolateral palpal tibial apophysis which bifurcates near the tip
forming a small hook. This is an oddity in North American members of the genus, the
others have the tibial apophysis simple. The unexpanded bulb of the palpus is large
compared to the cymbium. Females have a large external epigynum, which has a peculiar
general appearance, resembling that of an insect face (Figs. 8 and 16). The internal
genitalia are simple consisting of spermathecae with a loop (Figs. 9-11 and 17). Scale
ultrastructure as viewed by scanning electron microscopy (Figs. 1 and 2) exhibits the
predominantly three shafted morphology typical of “dendryphantine” salticids (Cutler
1981, Hill 1979).

KEY TO SPECIES

1. Males 2

Females 3

2. Embolus in ventral view broadly spatulate, tip not twisted and acuminate

helenae

Embolus in ventral view not broadly spatulate, tip twisted and acuminate

arizonensis

Figs. \ -2. —Metaphidippus arizonensis, Cedar Creek male: 1, scales between row III eyes; 2, opis-
thosoma, lateral. Both markers = 25 pm.

CUTLER AND JENNINGS-REVISION OF METAPHIDIPPUS ARIZONENSIS

3

3. External epigynum with sclerotized rims of openings perpendicular to the long axis of
the opisthosoma (Fig. 16); interior epigynum with loop of spermathecae posterior to

epigynal openings (Fig. 17)..... helenae

External epigynum with sclerotized rims of openings parallel to the long axis of the
opisthosoma (Fig. 8); interior epigynum with loop of spermathecae at level of epigynal
openings (Figs. 9-11) arizonensis

Metaphidippus arizonensis (Peckham and Peckham)

Figs. 1-11, Map 1

Dendryphantes arizonensis Peckham and Peckham 1901:326, pi. 28, f. 2, 1909:463, pi. 36, f. 7.
Metaphidippus arizonensis: Petrunkevitch 1911:622.

Dendryphantes glacialis Scheffer 1905:7, 1906:124, f. 3, 4, 8; Peckham and Peckham 1909:463, pi.

37, f. 7. NEW SYNONYMY.

Metaphidippus glacialis: Bonnet 1957:2814.

Notes.— The Peckham’s type was compared by Dr. H. W. Levi, and agrees with the
other specimens discussed here; it is a male from an unknown locality in Arizona. Schef-
fer’s specimens are unavailable or lost; letters to Kansas brought no response, and the
specimens are not at the AMNH, CAS, MCZ, or the U.S. National Museum. [Apparently
Scheffer described a number of species in 1905 in the Industrialist. These descriptions
were repeated (as new) the next year in the Transactions of the Kansas Academy of
Sciences. The latter publication is much more readily available in libraries, and contains
illustrations not in the original description.]

Male (Arizona, Chevelon Ranger District).— Total length 4.6, carapace 2.22 long, 1.67
wide. Eyerow I width 1.18, eyerow III width 1.19, eyefield length 0.89. Eye diameters:

Map 1.— Ranges of Metaphidippus arizonensis (closed circles = examined specimens, L = literature
records) andAf. helenae (open circles).

4

THE JOURNAL OF ARACHNOLOGY

Table 1.- Analysis of variation in populations of arizonensis. Femur = length of femur I, epigy-
num = minimum distance between sclerotized rims of epigynum, carapace = width of eyerow III.
Column means followed by the same letter are not significantly different, one-way ANOVA, P < 0.05.

FEMUR EPIGYNUM CARAPACE

Population

n

X±S.D.

Range

X±S.D.

Range

X ±S.D.

Range

Cedar Creek

35

1.23±0.08®

1.04-1.42

0.19±0.02^

0.13-0.23

1.23±0.07^

1.13-1.40

Kellogg

12

1.39±0.08^

1.18-1.48

0.20±0.02^

0.15-0.23

1.34±0.10^

1.12-1.42

Chevelon

31

1.26±0.08^

1.09-1.50

0.18±0.02^

0.12-0.23

1.25±0.07^

1.07-1.50

AME 0.28, ALE 0.17, PME 0.06, PEE 0.14. Distance ALE-PME 0.24, PLE-PME 0.30.
Femora lengths I 1.37, II 1.14, III 1.08, IV 1.31. Leg formula I, IV, II, III. Spines Leg I,
dorsal femoral 5, tibia 3-3, metatarsus 2-2. Color pattern see figure 3.

Female (Arizona, Chevelon Ranger District).— Total length 5.7, carapace 2.41 long,
1.81 wide. Eyerow I width 1.27, eyerow III width 1.29, eyefield length 0.95. Eye diame-
ters: AME 0.27, ALE 0.16, PME 0.07, PLE 0.15. Distance ALE-PME 0.25, PLE-PME
0.32. Femora lengths I 1.34, II 1.12, III 1.09, IV 1.40. Leg formula IV, I, II, III. Spines as
in male. Color pattern see figure 7.

Three populations of females were chosen for numerical comparison. Two were from
Minnesota, Allison Savanna — Cedar Creek, Anoka Co. (these sites are adjacent, separated
only by a county road) and Kellogg, Wabasha Co.; and, one from Arizona, Chevelon
Ranger District, Coconino Co. Three measurements were obtained from specimens: from
the carapace, the width of eyerow III; from an appendage, the length of right femur I;
from the opisthosoma, the narrowest point between the sclerotized rims of the epigynum.
Table I summarizes the data. Note that one of the Minnesota populations resembles the
Arizona population, whereas the other Minnesota population had members that were
significantly larger. Chi-square tests for independence of characters within populations
were nonsignificant. In addition to the identity in genitalic morphology, it appears that
morphometrically there are no great differences in northern and southern populations of
this species, other than local population differences.

Distribution.— From Alberta through Montana and North Dakota to southeast Minne-
sota, south through Kansas to New Mexico and Arizona. Does not occur west of the
Rocky Mountains.

Material examined.-CANADA: Alberta, Medicine Hat (Carr), female (AMNH). UNITED STATES:
Arizona; Coconino Co., Sitgreaves National Forest, T13N, R13E, sections 17-20, 30, 35, west of Barfs
Crossing (7000 feet) summer months 1969 to 1973, on forbs and seedling pine trees (D. T. Jennings),
one male, numerous females (AMNH); Iowa; Woodbury Co., four miles ENE of Hornick, 14 June
1970, sweeping upland prairie on bluff (B. Cutler), female (FSCA); Minnesota; Anoka Co., Helen
Allison Savanna, Nature Conservancy Area, E. of E. Bethel, April to October 1978 to 1980, on forbs
in and sweeping sand prairie (B. Cutler), numerous males and females (AMNH and BC); Anoka and
Isanti Cos., Cedar Creek Natural History Area, E. of E. Bethel, April to October late 1970’s to 1981,
on forbs in and sweeping sand prairie (B. Cutler), numerous males and females (BC); Wabasha Co., 3
miles SE of Kellogg, spring and summer months from 1974 to 1978, sweeping sand prairie (B. Cutler
and R. Huber), 3 males, numerous females (BC); Winona Co., Whitewater Game Refuge 1 mile E. of
Beaver, 31 July 1982, in retreats on heads of Lespedeza (B, Cutler), numerous females (BC); Montana;
Petroleum Co., 1.5 miles S., 5 miles W. of Winnett, May 1971, sweeping disturbed short grass plains
(N. E. Rees), male, female (BC); New Mexico ;5'ocorw Co., 21 miles E. of San Antonio, 28 June 1975,
on roadside table in desert grassland (R. Carter), male (BC); North YyaWota; McHenry Co., Denbigh
Sand Dune Reserve, 27 June 1970, on brush (P. D. Tobin), female (FSCA); 14 miles SW of Towner,

CUTLER AND JENNINGS-REVISION 0¥ METAPHIDIPPUS ARIZONENSIS

5

Figs. 3-6.-Metaphidippus arizonensis males: 3, Chevelon specimen, dorsal view; 4, Cedar Creek
specimen, apical view of palpal tibia after removal of tarsus; 5, Chevelon specimen, ventral view of
palpus; 6, Cedar Creek specimen, retrolateral view. Scale line in mm does not pertain to dorsal body
view,

15 June 1971, brushing grass (P. D. Tobin), female (FSCA); McLean Co., Garrison, near Douglas
Creek Bay, 4 July 1970, on weeds (P. D. Tobin), male (FSCA); Williams Co., Williston, 13 June 1973,
male on plant in field, females in nests in dry plants in field (D. Maddison), male, three females (WM).

In addition, the following literature records are believed to be reliable: Scheffer’s specimens were
from Kansas: Pottawatomie Co., St. George, and Riley Co., Manhattan; Jung and Roth (1974) Ari-
zona: Cochise Co., Chiracahua Mountains; Jennings (1973) Coconino Co., Sitgreaves National Forest,
Chevelon Ranger District, Dudley Burn, sec, 20, T13N, R14E (7100 feet), 24 July 1970, female and
egg retreat in dead stem of Tragopogon pratensis L. (AMNH). Although Worley and Pickwell (1927)
have been listed as recording this species (as D. glacialis) from Nebraska, they only noted that it was
possibly present.

Metaphidippus helenae (Banks)

Figs. 12-17, Map 1

Dendryphantes helenae (Banks) 1921:101-102, f. 5.

Metaphidippus helenae: Gertsch 1934:18.

Dendryphantes sausalitanus Chamberlin 1925:137, f. 57-58, Gertsch 1934:18 (synonymy with M.
helenae).

6

THE JOURNAL OF ARACHNOLOGY

Notes.— In the vial with the two female paratypes in the MCZ is a left palpus of a male,
this is probably the left palpus of the holotype which is missing from that specimen.

Male holotype.— Total length 4.3, carapace 1.85 long, 1.42 wide. Eyerow I width 1.00,
eyerow III width 1.00, eyefield length 0.73. Eye diameters: AME 0.28, ALE 0.18, PME
0.07, PEE 0.15. Distance ALE-PME 0.18, PLE-PME 0.23. Femora lengths: I 1.19, II 0.97,
III 0.92, IV 1.15. Leg formula I, IV, II, III. Spines, leg I dorsal femoral 4, tibia 3-3,
metatarsus 2-2. Range of total length in five males 3. 4-4 .4. Color pattern faded in this
specimen, see Figure 12 for unfaded appearance.

Female (paratype in CAS).— Total length 5.2, carapace 2.01 long, 1.54 wide. Eyerow I
width 1.16, eyerow III width 1.15, eyefield length 0.95. Eye diameters: AME 0.30, ALE

Figs. 1 -\ \ .-Metaphidippus arizonensis females: 7, Chevelon, dorsal view; 8, Chevelon, epigynum.
9-11. Variation in internal copulatory tubes. 9, Cedar Creek specimen; 10-11, Chevelon specimens.
Scale in mm does not pertain to dorsal body view.

CUTLER AND JENNINGS-REVISION OV METAPHIDIPPUS ARIZONENSIS

1

0.17, PME 0.07, PLE 0.15. Distance ALE-PME 0.18, PLE-PME 0.25. Femora lengths: I
1.20, II 1.00, III 0.99, IV 1.22. Leg order I, IV, II, III. Spination and color pattern as in
male, see figure 13 for unfaded specimen. In eight female specimens measured: the width
of eyerow III, mean 1.19, range 1.10-1.37; the length of right femur I, mean 1.15, range
1.00-1.25; the minimum width between the epigynal rims, mean 0.09, range 0.07-0.13.

Distribution.— Northern California, western North Dakota, southern Utah, north
central Wyoming.

Material examined.- UNITED STATES: California, Lassen Co., 13 mi S. of Ravendale, 5 June
1970, ex Sesymbrium (P. Rude), male {CW), Marin Co., 4 mi N. of Novato, 10 April \912, Salicornia
marsh, Devac (E. Schlinger), male (CB); San Francisco Co., San Francisco, 7 April 1918 (Helen van
Duzee), male holotype, female para type (CAS), San Francisco, 2 female paratypes (MCZ); North
Dakota; Co., Theodore Roosevelt National Park, North Unit, 11 July 1970, sweeping herbs

and shrubs in wooded gulley (K. V. Stone), female (ESC A); \}t2A\'.,Kane Co., Coral Pink Sand Dunes,
near Kanab, 19 June 1974, on sagebrush with eggs in retreat (D. T. Jennings), female {BC) \ Sevier Co.,
Richfield, 25 May 1930 (W. J, Gertsch), 2 females (AMNH); \Syom\ng\ Bighorn Co., 6 mi E. of Shell
on Highway 14, sweeping grasses and shrubby bushes in area of low shrubby vegetation and many
rocks (W. Maddison), male, 2 females (WM); Shell, 23 June 1965, saltbush (W. D. Fronk), male (BC).

The distribution of these two species overlaps in western North Dakota. M. helenae is
found in interior basins and along the Pacific coast, whereas M arizonensis is a grassland
species found from the eastern edge of the tall grass prairie, through the short grass plains,
and into mountain meadows. Undoubtedly both species will be found in more sites,

Figs, \2-\l .—Metaphidippus helenae: 12, Male, Wyoming, dorsal view (California coastal specimens
have the chevrons reduced, and the stripes extend to the posterior of the opisthosoma); 13, Female,
Kane County, Utah, dorsal view (Dark spots may be less prominent in other specimens); 14, Male
palpus, Wyoming, ventral view, 15, retrolateral view; 16, Female, Kane County, Utah, epigynum, 17,
internal copulatory tubes. Scale in mm does not pertain to dorsal body views.

8

THE JOURNAL OF ARACHNOLOGY

because the interior of the continent where they live is poorly collected relative to the
eastern grassland fringes and the coastal and desert areas. They may have different habitat
preferences, but this is not clearly established by the label data.

ACKNOWLEDGMENTS

We thank the following for their help in providing information and specimen access.
Acronyms in parentheses are used to designate collections in the taxonomic descriptions;
specimen in the collection of the first author are designated (BC). Dr. R. E. Crabill, Jr.,
United States National Museum, Washington, D. C.; Dr. G. B. Edwards, Florida State
Collection of Arthropods, Gainesville, Florida (FSCA); Mr. C. Griswold, Essig Entomo-
logical Museum, University of California, Berkeley, California (CB); Dr. B. J. Kaston, San
Diego State University, San Diego, California; Dr. H. W. Levi and Mr. W. Maddison,
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts (MCZ
and WM); Dr. N. 1. Platnick, American Museum of Natural History, New York, New York
(AMNH); Dr. W. J. Pulawski and Mr. V. F. Lee, California Academy of Sciences, San
Francisco, California (CAS).

We also thank The Nature Conservancy (Allison Savanna, Minnesota) and the Cedar
Creek Natural History Area, University of Minnesota (Cedar Creek, Minnesota) for
permission to collect at sites under their jurisdiction.

Portions of this research were done while the junior author was stationed at the Rocky
Mountain Forest and Range Experiment Station, Albuquerque, New Mexico.

LITERATURE CITED

Banks, N. 1921. New Californian spiders. Proc. California Acad. Sci. 4th ser., 1 1:99-102.

Bonnet, P. 1957. BibUographia Araneorum, v. 3, G-M, pp. 1927-3026.

Cambridge, F. O. Pickard-. 1901. Arachnida-Araneida. Pp. 173-312, Biologja Centrali-Americana (F.
D. Godman and O. Salvin, eds.). London, vol. 2.

Chamberlin, R. V. 1925. New North American spiders. Proc. California Acad. Sci., 4th ser., 14:105-
142.

Cutler, B. 1981. A revision of the spider genus Paradamoetas (Araneae, Salticidae). Bull. Amer. Mus.
Nat. Hist., 170:207-215.

Gertsch, W. J. 1934. Further notes on American spiders. Amer. Mus. Novitates, No. 726, 26 pp.
Hill, D. E. 1979. The scales of salticid spiders. Zool. J. Linnean Soc., 65:193-218.

Jennings, D. T. 1973. Egg retreat of Me taphidippus arizonensis (Peckham) (Araneae: Salticidae) in a
hollow stem. Entomol. News, 84:317-320.

Jung, A. K. S. and V. D. Roth. 1974. Spiders of the Chiracahua Mountain Area, Cochise Co., Arizona.
J. Arizona Acad. Sci., 9:29-34.

Peckham, G. W. and E. G. Peckham. 1901. Spiders of the Phidippus group of the family Attidae.
Trans. Wisconsin Acad. Sci., 23:238-359.

Peckham, G. W. and E. G. Peckham. 1909. Revision of the Attidae of North America. Trans. Wiscon-
sin Acad. Sci., 16: 355-646.

Petrunkevitch, A. 1911. A synonymic index-catalogue of spiders of North, Central and South Amer-
ica. Bull. Amer. Mus. Nat. Hist., 29:1-809.

Prozynski, J. 1979. Systematic studies on East Palearctic Salticidae. III. Remarks on Salticidae of the
USSR. Ann. Zool. (Warsaw), 34:299-369.

Scheffer, T. H. 1905. Addition to the list of Kansas spiders. Industrialist (Kansas), 31:435444.
Scheffer, T. H. 1906. Additions to the list of Kansas Arachnida. Trans. Kansas Acad. Sci., 20:121-130.
Worley, L. G. and G. B. Pickwell. 1927. [1931]. Spiders of Nebraska. Univ. Stud., Nebraska, 27:1-
129.

Manuscript received January 1984, revised March 1984.

Galiano, M. E. 1985. Revision del genero Hurius Simon, 1901 (Araneae, Salticidae). J. ArachnoL,
13:9-18.

REVISION DEL GENERO SIMON, 1901

(ARANEAE, SALTICIDAE)

Marfa Elena Galiano

Museo Argentine de Ciencias Naturales “Bernardino Rivadavia”

Av. Angel Gallardo 470
1405 Buenos Aires, R. Argentina

ABSTRACT

The genus Hurius is redefined to include the male diagnostic characters. Males of Hurius can be
distinguished by the unidentate chelicerae with four or five teeth on promargin and by the presence of
two big retrolateral apophyses on the palpal tibia. Spinurius Mello-Leitao, 1941 is newly synonymized
and H. aeneus (Mello-Leitao, 1941); a new combination is estabhshed. Males of H vulpinus and
females of H. aeneus are described for the first time. Two new species are described: Hurius petrohue
from Chile and/f. pisac from Peru.

INTRODUCCION

El genero Hurius Simon, 1901 fue descripto originalmente con una sola especie, H.
vulpinus Simon, 1901, basada sobre ejemplares femeninos. A1 estudiar colecciones de
material indeterminado pertenecientes al Museum of Comparative Zoology, se hallo
un lote con hembras y machos de H. vulpinus, lo que permite ampliar la diagnosis del
genero con los caracteres del palpo del macho, hasta ahora desconocidos, que son muy
particulares.

En 1944, Mello-Leitao incorporo dos especies al genero, H. costulatus y H. tristis. La
primera de estas especies fue transferida (Galiano, 1970) al genero Tullgrenella como
un sinonimo de T. quadripunctata. En cuanto a H. tristis debe ser exclufda del genero
y transferida a otro taxon que se publicara proximamente.

En el presente estudio se incorporan al genero dos especies, H. petrohue n. sp. y
H. pisac n. sp. y se considera que Spinurius Mello-Leitao, 1941 es un sinonimo de Hurius,
por lo que se establece //t/nws aeneus (MeHo-Leitao, 1941) nueva combinacion. Se descri-
ben por primera vez las hembras de esta especie.

Nada se conoce sobre la biologia de las especies de Hurius, pero todos los ejemplares
provienen de areas montanosas ubicadas en la Cordillera de los Andes o en la Precordil-
lera. Es probable que vivan bajo piedras.

Material y metodos.— Las medidas se expresan en milimetros y fueron tomadas segun
se explica en una publicacion anterior (Galiano, 1963). La quetotaxia se menciona segun
el sistema de Platnick y Shadab (1975) con ligeras modificaciones. Las abreviaturas son
las siguientes: OMA, OLA, OMP y OLP, ojos medios anteriores, laterales anteriores.

10

THE JOURNAL OF ARACHNOLOGY

medios posteriores y laterales posteriores, respectivamente; p, prolateral, r, retrolateral, v,
ventral, d, dorsal, ap, apical; MACN, Museo Argentino de Ciencias Naturales “Bernardino
Rivadavia”:MCZ, Museum of Comparative Zoology; MNHN, Museum National d’Histoire
Naturelle, Paris; MLP, Museo de La Plata.

Hurius Simon, 1901

Hurius Simon 1901a: 583, 585 (n. gen.) Petrunkevitch 1928:202. Neave 1939:702. Roewer 1954:

1184. Bonnet 1957:2237. Brignoli 1983:627.

Spinurius Mello-Leitao 1941:187 (n. gen.). Roewer 1954:1185. Brignoli 1983:628, 749 (Spinutius,

lapsus). NUEVA SINONIMIA

Diagnosis (revisada).— Queliceros con tres o mas dientes en el promargen como en
Sitticus Simon, 1901, del cual se diferencia por tener un diente en el retromargen, muy
cercano a la base de la una, y por presentar dos apofisis tibiales en el palpo. Se distingue
de Scoturius y Atelurius cuyos queliceros son semejantes, por la forma del palpo.

Descripcion.— Pro soma robusto (ancho 44-47% del largo), mitad anterior de la region
toracica en el mismo nivel que la cefalica. Estria toracica pequena, algo alejada de los ojos
posteriores. Area ocular mas ancha que larga, mas ancha atras que adelante y con los OMP
mas proximos a los OLA que a OLP. El largo del area ocular representa aproximadamente
el 40% del largo del prosoma. Clipeo bajo, menor que el radio de OMA. Esternon relativa-
mente angosto, ancho del extremo anterior igual a la base del labio. Laminas maxilares
redondeadas. Queliceros paralelos, verticales, surco ungueal muy breve; promargen con
tres dientes, a menudo los dos mayores sobre una base comun y uno a tres pequenos
dientes suplementarios; retromargen con un diente, situado muy proximo a la base
de la una. Patas espinosas. Pata III siempre mas corta que la IV. Palpo del macho con dos
apofisis tibiales retrolaterales muy desarrolladas, de las cuales la inferior esta recubierta
por gruesas cerdas. El embolo es breve y nace en el borde prolateral superior del bulbo.
Epigino con mecanismo de anclaje formado por uno o dos bolsillos de ubicacion variable.
Las aberturas de entrada se continuan por un ancho conducto mas o menos cilmdrico,
que desemboca en una camara esferoidal de la cual parte por un lado un delgado con-
ducto ciego (con glandulas en el extremo) y por otro las espermatecas, que son esfericas y
ubicadas a cada lado de la linea media.

Especie tipo— Hurius vulpinus Simon, 1901b.

Hurius vulpinus Simon, 1901
Figs. 1-7

Hurius vulpinus Simon 1901b :1 54 (1 hembra Lectotypus y 1 hembra Paralectotypus, de Ecuador:

Quito, en MNHN, examinado); 1901a:583, 585 fig. 708. Petrunkevitch 1911:658; 1928:202.

Roewer 1954:1184. Bonnet 1957:2237. Galiano 1963:363, lam. 19, fig. 16.

Descripcion de la hembra N- 7923 MACN.— Largo total 3.93. Prosoma: largo 1.70,
ancho 1.30, alto 0.70. Clipeo, alto 0.06. Area ocular: largo 0.73; ancho de hilera anterior
1.08; de hilera posterior 1.20; distancia OLA-OMP 0.16; OMP-OLP 0.18; diametro OMA
0.36. Estria toracica, situada 0.26 mas atras del borde posterior de OLP. Queliceros con
surco ungueal muy breve; retromargen con un diente muy proximo a la base de la una;
promargen con cuatro dientes (dos anguiares mayores) sobre una base comun. Patas
IV-I-IIMI. Quetotaxia: Femures I, II d 1-1-1, p ap 1 ; III d 1-1-1, p ap 2, r ap 1 ; IV d 1-1-

GALIANO-REVISION DEL GENERO/ZC/Z^/t/^-

11

1, p ap 1 , r ap 1 . Patellas III, IV r 1. Tibias I v 2-2-1 p; II v Ir-lr-lp, p 1 ; III v Ip-lp, p 1-1 , r
1-1-1 ; IV V lp-2, p 1-1-1 , r 1-1-1 . Metatarsos I, II v 2-2; III v ap 2, p 1-2, r 1-2; IV v lp-2, p
1-2, r 1-1-2. Epigino: El horde posterior grueso y esclerosado, con dos profundos bolsillos
infundibuliformes de anclaje, en cuyos hordes y parte interna se implantan largas cerdas
plumosas de funcion desconocida. Los orificios de entrada a los conductos de las esper-
matecas se abren en el campo medio del epigino, seguidos por anchos conductos que
desembocan en una camara esferoidal, que a su vez se conecta por medio de un conducto
espiral con la espermateca, esferica, mediana y relativamente pequena. (Figs. 5-7). As-
pecto y color en alcohol: prosoma pardo anaranjado, con la region cefalica amarilla, con

Figs. 1-1 —Hurius vulpinus Simon, macho: 1 palpo, retrolateral; 2, palpo, ventral; 3, cuerpo; 4,
quelicero, cara posterior. Hem bra: 5, epigino; 6, epigino clarificado, vista ventral (se dibujan los pelos
de un solo lado); 7, el mismo, vista dorsal. Escala lOO/u, salvo indicacion.

12

THE JOURNAL OF ARACHNOLOGY

manchas marmoradas de guanina blanca que se ven por transparencia. Sobre la region
toracica, lineas radiantes pardas a partir de la estria. Pelos pardos y blancos, entremez-
clados; una mancha de pelos blancos sobre la estria y dos manchas en el medio del area
ocular; pelos blancos en el margen anterior, alrededor de los ojos anteriores y en el
espacio entre OLP y OMP; pelos blancos en los costados y en el declive toracico. El
espacio bajo los OLA ocupado por pelos blancos. Desde el borde externo de cada OMA
salen largos pelos blancos dirigidos oblicuamente, convergentes en la Imea media. Queli-
ceros amarillentos, con pelos blancos en la cara anterior. Patas pardo amarillento, con
abundantes pelos blancos. Palpos amarillos, con tarsos pardo claro con abundantes pelos
blancos. Opistosoma amarillo pardusco, con manchas de guanina blanca que se ven por
transparencia y bandas pardas con pelos pardo rojizo que forman un diseno de bandas en

V invertidas. Vientre amarillo, con manchas de guanina blanca; en el medio, una ban da
pardusca.

Descripcion del macho N- 7923 MACN.— Largo total 3.60. Prosoma: largo 1.80,
ancho 1.33, alto 0.83. Clipeo, alto 0.11. Area ocular: largo 0.80;ancho de hilera anterior
1.13; de hilera posterior 1.21 ; distancia OLA-OMP 0.18; OMP-OLP 0.23; diametro OMA
0.36. Estria toracica, situada 0.25 mas atras del borde posterior de OLP. Laminas maxi-
lares con angulo redondeado. Esternon algo estrechado adelante, levemente mas angosto
que la base del labio. Queliceros paralelos, verticales. Surco ungueal muy breve, retromar-
gen con un diente muy cercano a la base de la una; promargen con cuatro dientes, los dos
angulares sobre una base comun y dos mas pequenos hacia la base de la una. Patas IV-I-
III-II. Quetotaxia: Femures Id 1-1-1 , p ap 2, r ap 1 ; II d 1-1-1 , p ap 2, r 1-2; III d 1-1-1 , p
1-2, r ap 2; IV d 1-1-1 , p ap 2, r ap 2. Patellas I-IV p 1 , r 1. Tibias I v 2-2-2, p 1-1 , r 1-1 ; II

V lr-2-2, p 1-1, r 1-1 ; III, IV d 1, v 1-2, p 1-1-1, r 1-1-1 . Metatarsos I, II v 2-2, p ap 1 ; III,
IV V ap 2, p 1-2, r 1-1-2. Palpos: (Figs. 1 y 2). Aspecto y color en alcohol: esencialmente
como en la hembra. Largos pelos blancos dirigidos hacia adelante forman una banda que
baja desde el espacio OMA-OLA hasta el borde del clipeo. Tambien hay pelos blancos
bordeando los ojos y en la cara anterior de los queliceros. Patas pardo claro amarillento,
con abundantes pelos blancos. Tarso I pardo con pelos pardos abundantes. Palpo pardo
claro con femur amarillo y pelos al tono. Los pelos de la apofisis tibial inferior son pardo
negruzco. Opistosoma como el de la hembra, con un scutum pequeno, basal dorsal,
anaranjado (Fig. 3).

Observaciones.— Variacion en la quetotaxia: hembras, femur IV p ap 2, r ap. 1 Tibias I

V 2-2-lr; II p 1-1 , r 1-1-1 ; III v lp-2. Metatarso III v lp-2. Machos, femures I p ap 1. Patella
I p 1 . Tibia II v lr-lr-2 . Metatarsos I, II p 0 , r 0 ; III, IV v lp-2 .

Material estudiado.- ECUADOR: Tungurahua, Banos (1800 m elev.), 10 April 1939 (col. W. C.
Macintyre), 1 macho, 3 hembras (MCZ), 1 macho, 1 hembra N- 7923 (MACN), July 1938, 1 hembra
N5- 7924 (MACN).

Hurius petrohue, nueva especie
Figs. 8-14

Etimologia.— El nombre especifico es un sustantivo en aposicion; corresponde a la
denominacibn de la localidad tipica.

Diagnosis.— Se diferencia de H. vulpinus por presentar los bolsillos de anclaje en
la parte anterior del epigino y porque la apofisis tibial inferior del palpo es mucho mas
larga.

Descripcion del Holotypus hembra.— Largo total 4.57. Prosoma: largo 1.83, ancho
1.40, alto 0.80. Clipeo, alto 0.08. Area ocular: largo 0.80; ancho de hilera anterior 1.18;

GALIANO-REVISION DEL GENERO HURIUS

13

de hilera posterior 1.30; distancia OLA-OMP 0.18; OMP-OLP 0.25; diametro OMA0.40.
Estria toracica situada 0.25 mas atras del horde posterior de OLP. Queliceros pequenos,
verticales, paralelos. Surco ungueal muy breve. Retromargen con un diente muy cercano a

Figs. S-14.-Hurius petrohue sp. n., paratipo macho: 8, palpo, ventral; 9, palpo, prolateral; 10,
palpo, retrolateral. Holotipo hembra: 11, opistosoma, dorsal; 12, epigino; 13, epigino clarificado, vista
ventral; 14, el mismo, vista dorsal. Escala 100 p, salvo indicacion.

14

THE JOURNAL OF ARACHNOLOGY

la base de la una; promargen con cinco dientes, los dos angulares mayores y sob re una
base comun. Patas IV-I-III-II. Quetotaxia: Femures I, II d 1-1, p ap 1 ; III d 1-1, pap l,r
ap 1 ; IV d 1-1-1, p ap 1, r ap 1. Tibias I, v 2-2; II v Ir-lr; III p 1, r 1 ; IV v lp-2, p 1, r 1.
Metatarsos I, II v 2-2; III v 1-2, p 1-2, r ap 2; IV v 1-2, p ap 2, r 1-2. Epigino: dos grandes
bolsillos de anclaje situados contiguos y en el extremo anterior del epigino. Borde poste-
rior engrosado. Entrada de los conductos proximos al reborde. Espermatecas contiguas,
medianas. (Figs. 12-14). Aspecto y color en alcohol: prosoma pardo rojizo oscuro, con el
borde anterior de la region cefalica algo mas claro. Pelos blancos y pardo rojizos, distri-
buidos formando manchas de la siguiente manera: pelos blancos en el margen anterior y
entre OMA y OLA y detras de cada OLP. Pelos blancos bajo los OLA; en el clipeo,
largos pelos blancos formando una barba orientada hacia adelante y cubriendo en parte la
base de los queliceros. Clipeo amarillo. Pelos blancos en los costados y en el declive
toracico. El resto del prosoma con pelos pardo rojizos. Queliceros amarillentos con pelos
blancos. Opistosoma amarillo con grandes manchas pardas (Fig. 11). Patas pardas, las
primeras mas oscuras; la mitad basal de los femures y patellas y la parte media de tibias y
metatarsos, amarillentos. Palpos amarillos, con anillos pardos en base de patella, tibia y
tarso, con abundantes pelos blancos. Esternon pardo negruzco.

Descripcion del Paratypus macho N- 7922 MACN.— Largo total 3.66. Prosoma:
largo 1.63, ancho 1.22, alto 0.70. Clipeo, alto 0.10. Area ocular: largo 0.68; ancho de
hilera anterior 1.03; de hilera posterior 1.11; distancia OLA-OMP 0.16; OMP-OLP 0.23;
diametro OMA 0.33. Estria toracica pequena, situada 0.23 mas atras del borde posterior
de OLP. Queliceros verticales, paralelos. Surco ungueal muy breve; retromargen con un
diente cercano a la base de la una; promargen con tres dientes, dos de ellos sobre una base
comun. Es posible que exista un cuarto diente pequeno, que se ha desprendido. Patas
IV-I-II-III. Quetotaxia : Femures I, II, III d 1 -1 -1 , p ap 2 ; IV d 1 -1 -1 , p ap 1 , r ap 1 ; Patella
IV r 1. Tibias I v lr-2-lp, p 1-1; II v Ir-lr, p 1 ; III p 1-1, r 1-1; IV v 1-2, p 1-1-1 , r 1-1-1 .
Metatarsos I, II v 2-2; III v ap 2, p ap 2, r 1-1 ; IV v 1-2, p 1-2, r 1-1-2. Palpo: las dos
ap6fisis tibiales muy largas; la retrolateral con el borde superior interno dentado; la
ventral con gruesos, largos y densos pelos negros. El embolo nace en el extremo inferior
prolateral del bulbo; grueso en la mayor parte del trayecto se afina apicalmente. (Figs.
8-10). Aspecto y color en alcohol: prosoma pardo oscuro, con el margen anterior leve-
mente mas claro, con algunos pelitos blancos entre los ojos. El resto cubierto por pelos
pardo negruzcos. Clipeo con una banda amarilla de guanina que abarca el especio bajo
los cuatro ojos; sin barba. Queliceros rojizos, esternon pardo. Opistosoma pardo y amari-
llo, con diseno semejante al de al hembra, excepto que en el dorso basal hay un scutum
pardo oscuro. Patas pardas, las primeras mas oscuras que las otras, con tarsos negruzcos
con abundantes pelos negros. Las otras patas anilladas de pardo oscuro negruzco, con
anillos poco evidentes.

Observacion.— Variacibn en la quetotaxia, hembras: femures I, II d 1-1-1. Metatarso III
p ap 2, r ap 2.

Material estudiado.-CHILE: Llanquihue; Petrohue, 19-20 marzo 1965 (col. H. W. Levi), 1 hembra
Holotypus, 1 hembra Paratypus (MCZ), Volc^ Oscorno, E. slope above Petrohue (400-1000 m elev.),
21 de marzo 1965 (col. H. W. Levi), 1 hembra, 1 macho Paratypi N- 7922 (MACN).

Hurius aeneus (Mello-Leitao, 1941) nueva combinacion

Figs. 15-21

Spinurius aeneus Mello-Leitao 1941:187, fig. 80, t. xi fig. 53 [macho Holotypus, de R. Argentina:
Tucuman; Bahado, (col. Biraben), N- 14.989 (MPL), examinado]. Roewer 1954:1184.

GALIANO-REVISION DEL GENERO HURI US

15

Medidas del Holotypus macho.— Prosoma: largo 1.77, ancho 1.53, alto 0.93. Clipeo,
alto 0.10. Area ocular: largo 0.97; ancho de hilera anterior 1 .33; de hilera posterior 1.50;
distancia OLA-OMP 0.23; OMP-OLP 0.26; diametro OMA 0.43.

Descripcion del macho N- 5035 MACN.— Largo total 4.47, Prosoma: largo 2.03,
ancho 1 .72, alto 1 .10. Clipeo, alto 0.1 1 . Area ocular: largo 0.98; ancho de hilera anterior
1.43; de hilera posterior 1.57; distancia OLA-OMP 0.26; OMP-OLP 0.29; diametro OMA

Figs. 15-21 -Hurius aeneus (Mello-Leitao), macho: 15, palpo, retrolateral; 16, palpo, ventral;
17, cuerpo; 18, quelicero, cara posterior. Hembra: 19, epigino; 20, epigino clarificado, vista dorsal; 21,
el mismo, vista ventral. Escala 100 ju, salvo indicacidn.

16

THE JOURNAL OF ARACHNOLOGY

0.43. Estna toracica 0.15 mas atras del borde posterior de OLP. Queliceros paralelos,
verticales; surco ungueal corto; retromargen con un diente fuerte; promargen con dos
dientes sobre una base comun (el angular mayor) y dos o tres dientes pequenos, en hilera
orientada hacia el fondo del surco ungueal. (Fig. 18). Patas I-IV-III-II. Quetotaxia:
Femures Id 1-1 -1 , p ap 2; II d 1 -1-1 , p ap 2, r ap 2; III d 1-1-1, p 1-2, r ap 1 ; IV d 1-1-1, p
ap 2, r ap 2. Patellas I-IV p 1, r 1. Tibias I v 2-2-2, p 1-1, r 1-1 ; II v 1-1-2, p 1-1-1, r

1- 1 ; III-IV d 1-1, V 1-2, p 1-1-1, r 1-1-1. Metatarsos I v 2-2, p 1-1 ; II v 2-2, p 1-1, r 1 ; III v

2- 2, p 1-2, r 1-2; IV v 2-2, p 1-2, r 1-1-2. Palpos: (Figs. 15 y 16). Aspecto y color en
alcohol; prosoma alto y ancho, la regidn cefalica levemente deprimida entre los OLPque
son bastante salientes. Color pardo oscuro, con la region cefalica pardo claro y abun-
dantes pelos pardos y amarillos entremezclados. En la region cefalica los pelos se disponen
en un patron semejante al que se observa en algunas especies de Sitticus: en cada mitad
del area ocular, los pelos se orientan divergiendo de una imaginaria linea longitudinal de
modo que se orientan de cada lado hacia afuera (hacia los ojos laterales) y hacia la linea
media del cuerpo, donde se encuentran con los correspondientes del lado opuesto. Entre
los OLP y OMP y entre los OMA y OLP hay manchas de pelos blancos; tambien abundan
en la region toracica. Bajo los OLA hay largos pelos blancos, que se dirigen hacia adelante
y forman un fieco en el borde del clipeo, donde se confunden con los largos pelos blancos
que cubren la cara anterior de los queliceros. Esternon y piezas bucales pardo claro.
Opistosoma amarillento, con manchas pardas formando tres anchas bandas transversas,
vinculadas entre si por manchas con forma de V invertida, mas cortas (Fig. 17). Vientre
amarillo, con ancha banda media parda. En la mitad basal del dorso, un scutum anaran-
jado, brillante. Patas pardas, las primeras mas oscuras, con la mitad de tibias y metatarsos
mas Claras, formando una anillacion poco evidente. Tarsos I pardo oscuro, con abun-
dantes pelos largos de ese color, que cubren todo el artejo. Palpos pardo claro, con
abundantes pelos blancos; apofisis tibial ventral con pelos pardo oscuro cobrizo.

Descripcion de la hembra N- 7848 MACN.— Largo total 5.66. Prosoma: largo 2.30,
ancho 1 .98, alto 1 .13. Clipeo, alto 0.08. Area ocular: largo 1.11; ancho de hilera anterior
1.66; de hilera posterior 1.90; distancia OLA-OMP 0.26; OMP-OLP 0.33; diametro OMA
0.51. Estria toracica situada 0.06 mas atras del borde posterior de OLP. Queliceros como
los del macho. Patas IV-I-III-II. Quetotaxia: Femures Id 1-1-1 , p ap 2; II, III, IV d 1-1-1,
p ap 2, r ap 1 . Patellas III, IV r 1. Tibias I v 2-2-2; II v 1-2-2, p 1-1 ; III, IV lp-2, p 1-1-1, r
1-1-1. Metatarsos I, II v 2-2; III v lp-2, p 1-2, r 1-2; IV v 2-2, p 1-2, r 1-1-2. Epigino: dos
bolsillos de anclaje poco profundos en posici6n anterior. Conductos y espermatecas
parecidos a los de H. vulpinus. (Figs. 19-21). Aspecto y color en alcohol: esencialmente
como en el macho, pero con el prosoma algo mas ancho y alto. Color como en el macho,
pero con las manchas claras del opistosoma menos notables. Las patas con anillos pardo
oscuro mas marcados que en el macho, con parte basal de femures y patellas amarillos.
Palpos pardo claro, con apice de femur, base de tibia y de tarso, pardo oscuro.

Observaciones.— El ejemplar tipico es un macho que carece de palpos. El aspecto
general, el color y la presencia de un area dorsal endurecida en el opistosoma permite
afirmar que se trata de un especimen adulto. La pubUcacidn original muestra que en el
momento de tomarse la fotografia ya faltaban los palpos; la fig. 80 es un dibujo que me
ha permitido identificar a los ejemplares que aqui se describen. Seguramente por un error
de imprenta, al comienzo de la descripcion original de la especie aparece el simbolo de
hembra, aunque se esta describiendo el macho cuyo palpo se ilustra. Esta confusion llev6
a Roewer a mencionar que Mello-Leitao describid macho y hembra. Hay tambien inexac-
titudes en la descripcidn de los caracteres por parte del autor, p.e., solo menciona dos
dientes en promargen; dice que los ojos de la segunda hilera estan separados de los OLA

GALIANO-REVISION DEL GENERO HURIUS

17

por el doble del espacio que los separa de OLP, menciona un clipeo densamente barbado
y describe la patella I como sin espinas. Como se ha podido comprobar estudiando el tipo,
se trata de fallas de observacion.

En otros ejemplares estudiados se ha observado la siguiente variacion en la quetotaxia:
machos, tibias II d 1 ; III, IV v 2-2, p l-l-l-l, r 1-1 -1-1. Metatarsos III, IV r 2-1-2. Hem-
bras, tibias I p 1-1 ; II V lp-2-lr; III v 2-2, r 1-1-2; IV v 2-2. Metatarso IV v lp-2.

Material estudiado.— R. ARGENTINA: Cordoba: Santa Rosa de Calamuchita, marzo-abril 1958
(col. M. J. Viana), 3 machos N- 5035 (MACN); San Alberto, Las Calles, 10 junio 1972 (col. G. Willi-
ner), 2 hembras N- 7848 (MACN); La Falda, marzo 1951 (col. M. J. Viana), 1 hembra N- 7847
(MACN).

Hurius pisac, nueva especie
Figs. 22-24

Etimologia.— El nombre especifico es un sustantivo en aposicion; corresponde a la
denominacion de la localidad tipica.

Diagnosis.— Se distingue de las otras especies por tener un solo bolsillo de anclaje impar
y medio en el epigino.

Descripcion del Holotypus hembra.— Largo total 4.80. Prosoma: largo 2.00, ancho
1.57, alto 0.93. Clipeo, alto 0.07. Area ocular: largo 0.90; ancho de hilera anterior 1.40;
de hilera posterior 1.52; distancia OLA-OMP 0.21 ; OMP-OLP 0.26; diametro OMA 0.46.
Estria toracica situada 0.28 mas atras del horde posterior de OLP. Queliceros con un
diente en retromargen y cuatro en promargen. Quetotaxia: Femures Id 1-1-1, p ap 2; II,
III, IV d 1 -1 -1 , p ap 2, r ap 1 . Patellas II p 1 ; III, IV r 1 . Tibias I v 2-2-2 ; II v lr-lr-2, p 1 -1 ;
III vap 2, p 1-1, r 1-1-1 ; IV V lp-2, p 1-1-1 , r 1-1-1 . Metatarsos I, II v 2-2; III v 1-2, p 1-2
(r 1-2 ?); IV V lp-2, p 1-1-2, r 1-1-2 (muchas de las espinas se han caido y la quetotaxia se
establece por los puntos de insercidn). Epigino: un bolsillo de anclaje impar y medio en la
parte anterior del epigino. Espermatecas muy grandes. (Figs. 22-24). Aspecto y color en

Figs. 22-2A.— Hurius pisac sp. n., holotipo hembra: 22, epigino; 23, epigino clarificado, vista ven-
tral; 24, el mismo, vista dorsal. Escala 100 ju.

18

THE JOURNAL OF ARACHNOLOGY

alcohol: prosoma robusto, pardo, con la region cefalica negruzca. Hay una mancha con
forma de Y, cuyo extremo posterior apoya en la estria, de color amarillento, que trans-
parenta manchas de guanina blanca. El margen anterior de la region cefalica es amarillen-
to. Opistosoma amarillento con manchas pardo oscuro, que forman en la linea media una
sucesion de triangulos pardos, algunos con los brazos oblicuamente prolongados hacia los
costados. Lados amarillos con manchitas pardo negruzco. Patas amarillo pardusco, con
anillos pardo oscuro en apice de femur, base de patella, base y apice de tibia y base de
metatarso y tarso. Palpos amarillentos con una manchita parda en base de patella, tibia y
tarso.

Observacion.— El Holotypus es el unico ejemplar conocido hasta el momento.

Material estudiado.-PERU.’ Departamento Cusco; Pisac (3000 m elev.), 13 enero 1983 (col. A.
Roig), 1 hembra Holotypus N- 7927 (MACN).

AGRADECIMIENTOS

Expreso mi reconocimiento por el prestamo de ejemplares tipicos y colecciones de
material indeterminado, a las siguientes personas: Dr. Max Vachon (MNHN), Dr. Herbert
W. Levi (MCZ), Dra. Olga Blanco (MLP).

LITERATURA CITADA

Bonnet, P. 1955-1958. Bibliographia Araneorum. Toulouse, 2(24):919-4230.

Brignoli, P. M. 1983. A Catalogue of Araneae described between 1940 and 1981. Manchester Univ.
Press, 719 pp.

Galiano, M. E. 1963. Las especies americanas de arahas de la familia Salticidae descriptas por Eugene
Simon. Redescripciones basadas en los ejemplares tipicos. Physis (Buenos Aires), 23(66):273470.
Galiano, M. E. 1970. Revision del genero Tullgrenella Mello-Leitao, 1941 (Araneae, Salticidae).
Physis (Buenos Aires), 29(79):323-355.

Mello-Leitao, C. 1941. Las arahas de Cordoba, La Rioja, Catamarca, Tucuman, Salta y Jujuy. Rev.
Mus. La Plata (N. S.) Zool., 2(12):99-198.

Mello-Leitao, C. 1944. Arahas de la provincia de Buenos Aires. Rev. Mus. La Plata (N. S.) Zool.,
3(24):311-393.

Neave, S. A. 1939. Nomenclator Zoologicus. London, 2:1-1025.

Petrunkevitch, A. 1911. A Synonymic Index-Catalogue of Spiders of North, Central and South Ameri-
ca with all adjacent Islands, Greenland, Bermuda, West Indies, Terra del Fuego, Galapagos, etc.
Bull. Am. Mus. Nat. Hist., 29:1-791.

Petrunkevitch, A. 1928. Sy sterna Aranearum. Trans. Connecticut Acad. Arts Sci., 29:1-270.

Platnick, N. 1. y M. U. Shadab. 1975. A revision of the spider genus Gnaphosa (Araneae, Gnaphosidae)
in America. Bull. Am. Mus. Nat. Hist., 155(l):l-66.

Roewer, C. F. 1954. Katalog der Araneae, 2b:927-1751.

Simon, E. 1901a. Histoire Naturelle des Araignees, 2(3):381-668.

Simon, E. 1901b. Descriptions d’Arachnides nouveaux de la famille des Attidae (suite). Ann. Soc.
entomol. Belgique, 45:141-161.

Manuscript received March 1 984, accepted April 1 984.

Haradon, R. M. 1985. New groups and species belonging to the nominate subgenus Paruroctonus
(Scorpiones, Vaejovidae), J. Arachnol., 13:1942.

NEW GROUPS AND SPECIES BELONGING TO THE NOMINATE
PARUROCTONUS (SCORPIONES, VAEJOVIDAE)

Richard M. Haradon

9 High Street

Stoneham, Massachusetts 02180

ABSTRACT

The nominate subgenus of the North American gQrms Paruroctonus Werner, 1934, comprises three
infragroups, differentiated primarily by cheliceral and pectinal characters, and named after P. gracilior
(Hoffmann, 1931), P. boreus (Girard, 1854), and P. stahnkei (Gertsch and Soleglad, 1966). The
gracilior infragroup is monotypic. The boreus mfragroup comprises four microgroups (10 species),
named after P. boreus, P. becki (Gertsch and Allred, 1965), R xanthus (Gertsch and Soleglad, 1966),
and P. baergi (Williams and Hadley, 1967). The stahnkei infragroup comprises four microgroups (15
species), named after P. stahnkei, P. shulovi (Williams, 1970), P. borregoensis Wilhams, 1972, and P.
williamsi Sissom and Tranche, 1981. New synonyms include: P. boreus (= Vejovis auratus Gertsch and
Soleglad, 1966);/*. gracilior (= Vejovis pallidus Williams, 1968). New species and subspecies include: P.
bantai Saratoga, n. ssp. (southern Death Valley); P. shulovi nevadae, n. ssp. (southern Nevada); P.
simulatus, n. sp. (western Great Basin); P, coahuilanus, n. sp. (Cuatro Cienegas basin, Coahuila). The
gracilior and boreus infragroups are primarily allopatric, but both are sympatric with the stahnkei
infragroup. Species within an infragroup are primarily allopatric, exceptions involving essentially
allotopic species.

INTRODUCTION

The scorpion genus Paruroctonus Werner, 1934, is distributed throughout most of
western North America, and contains at least thirty species. Recently, two subgenera
were defined, including Smeringurus Haradon, 1983, and two species groups were de-
limited within the nominate subgenus (Haradon, 1984a, 1984b). The complexity of this
genus is revealed further by the descriptions herein of two new species and two new
subspecies, prompting an effort at this time also to put the various species and groups in
the nominate subgenus into a sharper hierarchical perspective. Where necessary, to clearly
differentiate or synonymize species, supplementary data are given for certain previously
described forms. A key complementing those in Haradon (1984a, 1984b) is also provided.

METHODS

Relatively new characters involving the tarsal and pedipalpal macrosetae are explained
in the accompanying illustrations, and in more detail by Haradon (1984a). For examples
of the superior setae and mid-retrosuperior (mrs) seta on the basitarsus see Figures 7-8,
and for examples of the retrosuperior and retroinferior terminal setae on the telotarsus
see Figures 15-16.

20

THE JOURNAL OF ARACHNOLOGY

Most of the measurements used herein are defined by Stahnke (1970). Pectinal mea-
surements are shown in Figures 17-18. The cheliceral fixed digit length is taken dorsally
in a straight line from the digit’s tip to the bicusp’s proximal base. The length of each
distal tine on the cheliceral movable digit is taken dorsally in a straight line from the tip
to the point of common divergence; the ratio, inferior tine length/superior tine length, is
primarily useful only when the superior tine is short and essentially triangular (when
elongate and curved the planes and the proximal limit are indefinite).

Statistical data include the observed range (sample mean ± one standard deviation, n =
sample size). Acronyms of specimen depositories are explained below in the acknowledg-
ments.

SVBGENVS Paruroc tonus Werner, 1934
Uroctonoides Hoffmann, 1931:405;not Chamberlin 1920:36.

Paruroctonus Werner, 1934:283 (Jan.); Gertsch and Allred 1965:9 (subgenus of Vaejovis Koch,

1836; in part); Gertsch and Soleglad 1966:3 (subgenus of Vaejovis; in part); Wilhams 1972:2 (in

part), 1974:15 (in part), 1980:31 (in part); Hjelle 1972:26 (in part); Soleglad 1973:353, 355 (in

part); Francke and Soleglad 1981:241, 243 (in part).

Hoffmanniellius Mello-Leitao, 1934:80 (June).

Type Uroctonoides gracilior Hoffmann, 1931.

Diagnosis.— Nominate subgenus, differentiated by: metasomal segments I-IV without
short, reddish, intercarinal setae ventrally; metasomal segment III length/width ratio in
adult males 1.8 or less, junveniles and adult females 1.7 or less, immatures 1.6 or less;
carapace length/pectine length ratio in adult females 1.2 or more; pectinal teeth in
females 22 or fewer (rarely 23 or 24); adult carapace length usually less than 6.5 mm in
males and 7.0 mm in females.

Comparisons: Subgenus Smeringurus has numerous short, reddish, intercarinal setae
ventrally on metasomal segments I-IV, and differs significantly in the other four charac-
ters above (see Haradon 1983).

Distribution.— Western North America, southern Canada southward into Aguas-
calientes and Baja California Sur in Mexico.

Subordinate taxa.— The subgenus Paruroctonus comprises three infragroups (defined
below): gracilior infragroup (one species), boreus infragroup (10 species), stahnkei
infragroup (15 species).

Vaejovis minckleyi Williams, 1968a, assigned to Paruroctonus by Stahnke (1974:138),
exhibits only some of the characteristics that in combination define Paruroctonus, and is
here excluded from this genus.

Remarks.— One of the more striking dichotomies in the nominate subgenus is that
between the generally large species with many pectinal teeth and the generally small
species with few pectinal teeth. Although there are exceptions in each group, the diver-
gent tendencies in size and pectinal tooth counts, as well as in several coincident charac-
ters, are quite conspicuous. This dichotomy is also supported by the observations that the
two groups are widely sympatric, whereas the species within each group are, with few
exceptions, allopatric. However, one of the large species, Paruroctonus gracilior (Hoff-
mann, 1931), is geographically removed and morphologically divergent from the other
large species, and in various characters tends to link the large and small species. The link is
completed by Paruroctonus becki (Gertsch and Allred, 1965) among the large species and
Paruroctonus stahnkei (Gertsch and Soleglad, 1966) among the small species, each of

HARADON-SCORPION SUBGENUS PARUROCTONUS

21

which is intermediate to P. gracilior and most or all of the other species in their respective
groups with respect to infragroup characters 1-3 (see infragroup diagnoses below) as well
as in the number of retroinferior terminal setae on the telotarsi, the number of primary
denticle rows on the pe dipalp movable finger, and the development of denticles on the
inferior carina of the cheliceral fixed digit. The various characteristics shared by P.
gracilior and the remaining large species (i.e., infragroup characters 5-10) all appear
plesiomorphic, relative to the outgroup subgenus Smeringurus, and thus do not neces-
sarily support a closer relationship than one between P. gracilior and the small species, a
relationship for which, likewise, no synapomorphies are known. Therefore, the classifica-
tion that, in my opinion, best describes the available observations is one involving three
infragroups; namely, gracilior, boreus (including P. becki), and stahnkei.

GRACILIOR INFRAGROUP

Diagnosis.— An infragroup of nominate subgenus Paruroctonus differentiated by: (1)
cheliceral fixed digit with inferior carina confined distally, does not extend proximally to
level of bicusp (see Gertsch and Soleglad 1966:fig. 33); (2) cheliceral movable digit with
superior distal tine essentially triangular, inferior distal tine length/superior distal tine
length ratio 3.1 -3.2 (see Gertsch and Soleglad 1966:fig. 35); (3) carapace length/cheliceral
fixed digit length ratio 3.3-4. 1 (rarely 4.2); (4) basitarsus III with five superior setae,
including three distal plus two proximal (3+2; rarely 4+2, and only among arenicolous
specimens); (5) pectinal teeth in males 23-32 (more than 95% with 24-32), females 15-21
(more than 95% with 18-21); (6) pedipalp movable finger length/palm length ratio in
adult males and females 1.1 -1.2; (7) carapace length/pectine length ratio in adult males
0.9-1. 0, adult females 1.2-1 .3; (8) humerus with three (occasionally four) inframedial
macrosetae on proximal 3/5 of internal surface; (9) pedipalp primary denticles, excluding
proximal row, total 42-61 on fixed finger, 54-75 on movable finger; (10) adult carapace
length in males 4. 1-6.6 mm, females 4. 8-7. 2 mm.

Comparisons: the boreus infragroup (below) differs in characters 1-4; the stahnkei
infragroup (below) differs primarily in characters 14, but also significantly in 5-10.

Distribution.— Southeastern Arizona, eastward to Big Bend region of Texas, southward
to Aguascalientes in Mexico.

Included species.— P. gracilior (Hoffmann, 1931).

Paruroctonus gracilior (Hoffmann)

Uroctonoides gracilior Yioifmmn, 1931:406, figs. 4243; Gertsch 1958:15, 17.

Paruroctonus gracilior: Werner 1934:283, fig. 363; Stahnke 1957:253, 1961:206, 1974:136, 137,
138, figs. lOA, 11 A, IIB; WilUams 1972:3, 1980:31, 32, figs. 35A-B, 36C-D; Soleglad 1972:73,
1973:355, tbl. 2; Sissom and Francke 1981:97-98, 102, 107, figs. 7-12, 33-35; Francke and
Soleglad 1981:242, fig. 22.

Vejovis (Paruroctonus) gracilior: Gertsch and Allred 1965 :9; Gertsch and Soleglad 1966:6, 26-30, figs.
13, 18, 21, 23, 33-35, tbl. 3; Williams 1968a:7.

Hoffmanniellus gracilior: Gertsch and Soleglad 1966:26 (in synonymy); Williams 1972:3 (in syno-
nymy). Misspelling of Ho ffmanniellius Mello-Feitao, 1934:80.

Vejovis (Paruroctonus) pallidus Williams 1968a:6-ll, figs. 4-6, tbl. 2; Diaz-Najera 1975:7, 20. NEW

SYNONYMY.

Paruroctonus pallidus: Williams, 1972:3; Soleglad 1972:73, 1973:355, tbl. 2; Stahnke 1974:138;

Sissom and Francke 1981:98, 102.

Uroctonus gracilior: Diaz-Najera 1975:2 (erratum).

Vaejovis gracilior: Diaz-Najera 1975:2, 6, 8, 20.

22

THE JOURNAL OF ARACHNOLOGY

Jyi^QS—Uroctonoides gracilior: Lectotype male (adult) from Mexico, Aguascalientes,
Tepezala (C. C. Hoffmann), tagged #1. Depository: American Museum of Natural His-
tory.

Vejovis pallidus: Holotype male (adult) from Mexico, Coahuila, 0.5 kilometer SW
Cuatro Cienegas (S. C. Williams, et al.). Depository: California Academy of Sciences,
Type No. 10174.

Diagnosis.— See infragroup diagnosis above.

Description.— Supplementing above diagnosis, Gertsch and Soleglad (1966:26), and
Sissom and Francke (1981:97). Basic fuscous pattern (see Gertsch and Soleglad 1966:
figs. 18, 21) varies from dark and distinct to obsolete. Cheliceral fixed digit without
denticles on inferior carina. Humeral macrosetae: internals include one supramedial, three
(occasionally four) inframedials on proximal 3/5; four dorsals; usually three external
medials, middle seta often small in immatures and juveniles. Brachial macrosetae: four
internals. Chela: palm with eight major carinae moderately to well developed and granular
in both sexes, intercarinal surfaces weakly to moderately concave; macrosetae include
two or three on internal carina, usually four on ventrointernal carina, usually eight or
nine flanking ventral carina, none on fixed fmger, one long internal proximal and some-
times one short internal at mid-length of movable fingers; fingers essentially unscalloped
in both sexes; primary denticles in seven rows on movable fmger, six on fixed fmger;
supernumerary denticles well developed, six on fixed finger, seven on movable fmger.
Basitarsi I-III: not conspicuously compressed laterally; superior setae on I-III irregularly
distributed, usually three distal plus two proximal on III; mrs seta on I at most only
slightly offset from superior setae, on II moderately offset, on III set well apart from
superior setae. Telotarsal setae I-IV: proinferiors 1, 2,2,2; two each promedials, prosuperi-
ors, retrosuperiors and retromedials; usually one retroinferior; one retroinferior terminal.
Ungues I-IV about 1/3 as long as telotarsus. Pectines extend to proximal margin of femur
IV in males, to 1/4 length of trochanter IV in females. Metasomal carinae: ventral and
ventrolaterals I-III in males essentially smooth to crenulate (often strongly so), in females
smooth to weakly crenulate, IV entirely crenulate in both sexes. Metasomal setae: counts
variable; ventrals I-IV primarily 34,5-6,5-6,5-7; ventrolaterals I-V primarily 2 -3 ,4 ,4-5 ,4-6,
7-1 5 ; dorsals I-IV 0,1 ,1 ,2 (fewer than 5% with 1,1,1 ,2).

Variation.— Some specimens from arenicolous populations had on basitarsus III, in
addition to the usual 3+2 superior setae, a sixth seta (variably developed) between and
prolateral to the proximal and distal groups, resulting in a 4+2 pattern; distinctly smaller
extraneous setae might also be present, particularly in arenicolous specimens.

The ventrolateral metasomal seta counts varied considerably, including on V; e.g.,
eight to 15 (80% with nine to 12) in Texas and New Mexico, seven to nine (68% with
eight) in southeastern Arizona, and seven to nine, normally eight, in Cuatro Cienegas
basin of Coahuila.

Adult carapace lengths varied considerably among the samples; e.g., 4.1 -5.0 mm
(Chiricahua Mts., Arizona), 4. 5-6. 5 mm (Big Bend region, Texas), and 6.0-7. 2 mm (Cua-
tro Cienegas basin, Coahuila).

Remarks.— Pamroctonus pallidus, distinguished from P. gracilior originally (Williams
1968a: 7) by apparent differences in pigmentation and in the metasomal carinae, and
further (Sissom and Francke 1981:98, 102) by apparent differences in metasomal seta
counts, is here considered an arenicolous pigmentation variant ofP. gracilior. The range in
variation in the development of the metasomal carinae and the metasomal seta counts in
P. gracilior subsumes that of P. pallidus. When detectable, vestigial traces of fuscosity in
P. pallidus specimens conform to the general pattern characteristic of P. gracilior. The

HARADON- SCORPION SUBGENUS PARUROCTONUS

23

Table 1.- Diagnostic characteristics of the four microgroups constituting the boreus infragroup
of the nominate subgenus Paruroctonus.

Character

X an thus

becki

boreus

baergi

Carapace length/cheliceral fixed

digit length

7.0-9.0

5.5-6.5

7.0-9.0

7.0-9.0

Pedipalp movable finger length/

palm length

1.4-1. 6

1. 1-1.3

1. 1-1.2

1. 1-1.2

Pedipalp primary denticles,
rows on movable finger

1

6-7

6

6

less proximal row, fixed finger

>80

<80

<80

<80

movable finger

>90

<90

<90

<90

Pedipalp fingers, scalloping:

(A) not, (B) proximally d

C

A

A,B

B

only, (C) multiscalloped 9

c

A

A

A

Basitarsus II mid-retrosuperior

seta (A) present (B) absent

B

A

A

B

Basitarsus III superior setae:

10-11

6

6

7-11

(A) in distal plus proximal

rows, (B) in single file

B

A

A

A,B

Telotarsi II-IV retroinferior

terminal setae

2

1

2

2

Telotarsi III retrosuperior setae

6-7

2

2

24

amount of fuscosity typical of a population is often correlated with the darkness of the
substrate, and in several Paruroctonus species considerable variation in pattern intensity
exists (Haradon 1983:253, 261).

The above synonymy is based on the examination of P. pallidus paratypes (CAS,
OFF); the lectotype and two cotypes of P. gracilior (AMNH); and approximately 120
other specimens of P. gracilior from previously reported material (Gertsch and Soleglad
1966; Sissom and Francke 1981) from Arizona, New Mexico, Texas and Chihuahua.

BOREUS INFRAGROUP

Diagnosis.— An infragroup of nominate subgenus Paruroctonus differentiated by
combination of: (1) cheliceral fixed digit with inferior carina extending proximally at
least to level of bicusp (see Gertsch and Soleglad 1966:fig. 39); (2) cheliceral movable
digit with superior distal tine elongate and curved, inferior distal tine length/superior
distal tine length ratio less than 3.0; (3) carapace length/cheliceral fixed digit length ratio
5. 5-9.0; (4) basitarsus III with six or more superior setae, arranged in distal plus proximal
rows (4+2 to 5+2) or in essentially single file; (5) pectinal teeth in males 24-39 (except
20-23 in some populations of P. bantai and P. baergi, or rarely 23 in several other spe-
cies), females 18-24 (except some populations 16-17 in P. bantai and 13-17 inP. baergi,
or rarely 17 in several other species); (6) pedipalp movable finger length/palm length ratio
in adult males and females 1.1-1 .6; (7) carapace length/pectine length ratio in adult males
0.8-1 .0, adult females 1.2-1. 4 (except 1.6-1. 8 in P. utahensis); (8) humerus with three
inframedial macrosetae on proximal 3/5 of internal surface (except two in P. baergi); (9)
pedipalp primary denticles, excluding proximal row, total 25-90 fixed finger, 35-103
movable finger (rarely 35 or 36); (10) adult carapace length generally 4. 5-6.0 mm in
males, 5. 0-7.0 mm in females.

24

THE JOURNAL OF ARACHNOLOGY

Comparisons: The gracilior infragroup (above) differs in characters 14; the stahnkei
infragroup (below) differs primarily in character 5, but also significantly in 6-10. Excep-
tional males with fewer than 24/24 and females with fewer than 17/18 pectinal teeth
differ from the stahnkei infragroup in having 35 or more primary denticles (less proximal
row) on pedipalp movable finger in combination with either dorsal metasomal setae I-IV
0,0,0, 1 (P. bantai), or mrs seta absent on basitarsus II and either three internal inframe-
dial macrosetae on humerus or one retromedial seta on telotarsus III (baergi group).

Distribution.— Western North America, southern Canada southward into northern
Baja California Norte, Sonora and Chihuahua in Mexico.

Subordinate taxa.— The boreus infragroup comprises four primarily allopatric ele-
ments, differentiated as follows and in Table 1.

BOREUS MICROGROUP. Diagnosis: combination of carapace length/cheliceral fixed
digit length ratio 7.0-9 .0; basitarsus II with mrs seta. Distribution: Rocky Mountains
region westward to Pacific coastal mountains (excluding northwest coast), southern
Canada southward to northern Arizona and northern Baja California Norte (excluding
Mojave and Sonoran Deserts). Included taxa: P. boreus (Girard, 1854); P. silvestrii (Bo-
relli, 1909); P. bantai (Gertsch and Soleglad, 1966);P. bantai Saratoga, n. ssp.;P. arnaudi
Williams, 1972.

BECKI MICROGROUP. Diagnosis: carapace length/ cheliceral fixed digit length ratio
5. 5-6. 5. Distribution: Western and southern Great Basin, southwestward through Mojave
Desert to San Jacinto Mountains and edge of Colorado Desert in California. Included
species: P. becki (Gertsch and Allred, 1965).

XANTHUS MICROGROUP. Diagnosis: pedipalp primary denticles, excluding proximal
row, total 82-90 on fixed finger, 98-103 on movable finger; pedipalp movable finger
length/palm length ratio 1.4-1. 6 in both sexes; pedipalp fingers multi-scalloped in adults
of both sexes (see Gertsch Soleglad 1966:fig. 32); telotarsus III with six or seven retro-
superior setae. Distribution: Sand dunes, southeastern California and extreme north-
western corner of Sonora. Included species: P. xanthus (Gertsch and Soleglad, 1966).

BAERGI MICROGROUP. Diagnosis: combination of carapace length/ cheliceral fixed
digit length ratio 7.0-9 .0; pedipalp primary denticles on movable finger in six rows;
basitarsus II without mrs seta. Distribution: Loose sandy soils, primarily dunes, associated
with Colorado River and Rio Grande drainages, from southern Utah to northern Mexico.
Included species: This microgroup, named after P. baergi (Williams and Hadley, 1967),
and including P. utahensis (Williams, 1968b), is discussed in detail by Haradon (1984a).

Remarks.— The boreus and gracilior microgroups, where sympatric with the arenico-
lous baergi microgroup, occupy more compact soils; the boreus microgroup, where
sympatric with the becki microgroup, tends to occupy higher elevations.

Paruroctonus boreus (Girard)

Figs. 1-2

Scorpio (Telegonus) boreus Girard, 1854:267-269, zool. pit. xvii, figs. 5-7 (in part, not record from
Eagle Pass, Texas).

Buthus boreus: Wood 1863a:110, 1863b:368.

Vejovis boreus: Marx 1888:91; Gertsch 1958:6 (in part, not synonymy with “Vejovis silvestrii”).
Vaejovis boreus: Ewing 1928:12; not Bugbee 1942:320 (= Paruroctonus utahensis; see Sissom and
Francke 1981:94); not Diaz-Najera 1975:13 {= Paruroctonus silvestrii).

Vejovis aquilonalis Stahnke, 1940:101.

Vejovis (Paruroctonus) boreus: Gertsch and Allred 1965:9 (in part, not synonymy with “Vejovis
silvestrii”) ^GevXsch and Soleglad 1966:7.

HARADON-SCORPION SUBGENUS PARUROCTONUS

25

Vejovis (Paruroctonus) aquilonalis: Gertsch and Allred 1965:9; Gertsch and Soleglad 1966:42 (in part,
see Sissom and Francke 1981 :94).

Vejovis (Paruroctonus) auratus Gertsch and Soleglad, 1966:7 (key), 34, 4447 (description), figs. 55,
58, tbl. 6 (in part, holotype only). NEW SYNONYMY.

Vaejovis (Paruroctonus) boreus: Hjelle 1972:22.

Paruroctonus aquilonalis: Williams 1972:3; Soleglad 1972:74, 1973:355, Stahnke 1974:138; Sissom
and Francke 1981:94.

Paruroctonus auratus: Williams 1972:3, 1976:2, 1980:47; Soleglad 1972:75 (in part), 1973:355 (in
part); Stahnke 1974:138; Sissom and Francke 1981:96 (in part).

Paruroctonus boreus: Williams 1972:3, 1976:2; Soleglad 1972:74, 1973:355; Stahnke 1974:138;
Sissom and Francke 1981 :93.

-Scorpio (Telegonus) boreus: Female (adult) from the “Valley of the Great Salt
Lake of Utah”, collected by “Capt. Howard Stansbury”. Depository: United States
National Museum.

Girard (1854) based his description of P. boreus on a single specimen, which was last
reported to have been examined by Marx (1888). The 18 pectinal teeth reported by
Girard (1854:267, 268, fig. 6) indicate the original specimen to be a female.

Figs. 14.- Right pedipalpal segments, internal views: \,P. boreus, chela; 2,P. boreus, brachium;
3, P. bantai, chela; 4, P. bantai, brachium. Key: dsm = distal supramedial seta; circle = trichobothrium.
Scale =1.0 mm.

Figs.5-6.-Metasomal segment V, ventral views: 5, P. bantai bantai; 6, P. bantai Saratoga, n. ssp.
Scale = 1.0 mm.

26

THE JOURNAL OF ARACHNOLOGY

Vejovis aquilonalis: Holotype male (adult) from U.S.A., Arizona, (Coconino County),
30 mi. S Grand Canyon on Hwy. 64; locality corrected by Sissom and Francke (1981:
94). Depository: California Academy of Sciences.

Vejovis auratus: Holotype female (adult) from U.S.A., California, San Bernardino
County (Death Valley Natl. Mon.), Saratoga Springs. Depository: American Museum of
Natural History.

Diagnosis.— A species of subgenus Paruroctonus, boreus infragroup (cheliceral fixed
digit with inferior carina extending proximally to level of bicusp; pectinal teeth 24-35 in
males, 18-23 in females, pedipalp primary denticles, excluding proximal row, 37-52 on
fixed finger, 44-68 on movable finger), and boreus microgroup (carapace length/cheliceral
fixed digit length ratio 7.0-9.0; basitarsus II with mrs seta), differentiated by combination
of: (1) pedipalp fingers in adult male deeply scalloped proximally, closed fingers form
wide gap, in adult female weakly scalloped, closed fingers form a narrow proximal gap
(see Gertsch and Soleglad 1966: figs. 24, 25); (2) fuscous markings generally absent
in interocular triangle and do not extend to posterior margin on tergites II-VI (see
Gertsch and Soleglad 1966:figs. 6, 8); (3) ventrolateral metasomal setae on IV 3, on V 6;
(4) ventral metasomal setae I- IV 3,3 ,3 ,4; (5) dorsal metasomal setae I-IV 0,1, 1,2; (6)
ventrolateral metasomal carinae Mil smooth with few posterior crenulations; (7) ventral
metasomal carinae I-III obsolete to smooth; (8) brachium with four long internal macro-
setae (Fig. 2); (9) chelal palm macrosetae include two on internal carina (both long) and
two on ventrointernal carina (proximal long, distal short) (Fig. 1).

Comparisons: P. silvestrii has (1) weakly scalloped pedipalp fingers in adult males and
essentially unscalloped fingers in adult females; P. silvestrii and P. arnaudi have (2)
extensive fuscous markings in interocular triangle and extending to posterior margin on
tergites II-VI (see Gertsch and Soleglad 1966:figs. 7, 9), (3) ventrolateral metasomal setae
on IV 4, on V 7, and (4) ventral metasomal setae I-IV 3,4 ,4, 5; P. bantai (see below)
differs primarily in characters 5-9.

Distribution.— Western North America, from southern Canada to northern Arizona and
to coastal mountains of southern California, essentially excluding northwest coastal
region and the Mojave and Sonoran Deserts.

Table 2.— Numbers of ventral and ventrolateral metasomal setae in subspecies of Paruroctonus
bantai.

P. b. bantai
Metasomal Segment

Carina Setae I II III IV V

P. b. Saratoga
Metasomal Segment

II III IV V

Ventral

2/2

3

2/3

13

3/3

26

42

42

36

3/4

5

4/4

1

Ventro-

lateral

2/2

42

33

17

2/3

6

11

3/3

3

14

42

3/4

4/4

4/5

5/5

5/6

6/6

41

1

30

11

62 60 49 7

2 10 17

3 62 38

62

62 62 60
2

1

62

HARADON-SCORPION SUBGENUS PARUROCTONUS

27

Remarks.~Pan/rac?of2ws boreus is the most widely mentioned Paruroctonus species in
the technical and popular literature (often combined with Vaejovis or Vejovis), and the
above synonymy includes only the first citation of each name, major taxonomic ac-
counts, new synonyms and misidentifications. The bibliographic data given herein for
Girard (1854), a report contained in a rare volume, corrects a recurring error that began
with Wood(1863b:369).

The above diagnosis of P. boreus is based primarily on specimens from the Great Basin,
including the Great Salt Lake Desert. The inter- and intrapopulation variability of many
other characters in P. boreus is considerable, and still being studied.

Examination of the holotype of the nominal species Paruroctonus auratus revealed no
significant differences between it and the Great Salt Lake Desert (topotypic) population
of P. boreus. The subtle differences in metasomal and telson proportions between P.
auratus and P. boreus reported by Gertsch and Soleglad (1966:45) were found to be
insignificant using large samples of the latter species. As in certain other congeners (see
Remarks to gracilior infragroup above), many arenicolous populations of P. boreus,
including the topotypic populations of P. boreus and P. auratus, have essentially lost the
basic fuscous pattern characteristic of this species (see Gertsch and Soleglad 1966:figs. 6,
8), and simply represent pigmentation variants.

Paruroctonus bantai (Gertsch and Soleglad)

Figs. 3-6

Vejovis (Paruroctonus) bantai Gertsch and Soleglad, 1966:6 (key), 20-23 (description), figs. 12, 22,

29, tbl. 2; Williams and Hadley 1967:1 12; Williams 1970:8.

Paruroctonus bantai: Williams 1972:3, 1976:2; Soleglad 1972:75, 1973:355, tbl. 2;Stahnke 1974:

138.

Type.— Fe/ovA bantai: Holotype female (adult) from U.S.A., California, Inyo County,
Saline Valley, Warm Springs Road, Station 94, 8 May 1960 (B. Banta). Depository;
California Academy of Sciences, Type No. 10193.

Diagnosis.— A species of subgenus Paruroctonus, boreus infragroup (cheliceral fixed
digit with inferior carina extending proximally to level of bicusp; pectinal teeth 20-28 in
males, 16-20 in females; pedipalp primary denticles, excluding proximal row, 28-35 on
fixed finger, 3749 on movable finger), and boreus microgroup (carapace length/cheliceral
fixed digit length ratio 7.0-9.0; basitarsus II with mrs seta), differentiated by combination
of: (1) pedipalp fingers in adult male deeply scalloped proximally, closed fingers form
wide gap, in adult female weakly scalloped, closed fingers form narrow gap; (2) fuscous
markings generally absent in interocular triangle and do not extend to posterior margin
on tergites II-VI; (3) ventrolateral metasomal setae on IV 3, on V 4 or 6; (4) ventral
metasomal setae I-IV 3,3,3,34; (5) dorsal metasomal setae I-IV 0,0,0, 1 ; (6) ventrolateral
metasomal carinae I-III granular; (7) ventral metasomal carinae I-III smooth to granular;
(8) brachium with three long macrosetae on internal surface, dsm seta inconspicuous (Fig.
4); (9) pedipalp chelal macrosetae inconspicuous or absent (Fig. 3), except occasionally
one short proximal on internal or ventrointernal carina, especially in females and juve-
niles.

Comparisons: P. silvestrii differs in characters 1-9, P. arnaudi differs in characters 2-9,
andP. boreus differs in characters 5-9 (see P. boreus diagnosis above).

Description.— Supplementing above diagnosis and Gertsch and Soleglad (1966:20).
Adult carapace lengths in males 4. 9-6. 2 mm, in females 5. 5-7 .4 mm. Chelicera: fixed digit
with denticles on inferior carina; on movable digit, superior distal tine elongate and

28

THE JOURNAL OF ARACHNOLOGY

curved, about 1/2 as long as inferior distal tine. Trichobothria typical of genus in number
and distribution. Humeral macrosetae: internals include one supramedial, three inframe-
dials on proximal 3/5; four dorsals; usually three external medials, middle seta smallest
and occasionally absent. Chela: supernumerary denticles well developed, six on fixed
finger, seven on movable finger. Basitarsi I-III: not conspicuously compressed laterally;
superior setae on I 2+2 or 3+2, II 3+2 or 4+2, III 4+2; mrs seta on Mil stout, short and
distinctly offset from superior setae. Telotarsal setae I-IV: proinferiors 1,2,2,2; two each
promedials, prosuperiors, retrosuperiors, retromedials; retroinferiors 1,1,2,2; retroinferior
terminals l-2,2,2,2. Ungues about 3/5 as long as telotarsus. Pectines in adult males extend
to about 3/4 length of trochanter IV, in adult females to slightly beyond coxa IV or to
about 1/3 length of trochanter IV. Telson setae: two long ventroanteriorly, two long
at subaculear tubercle, others short or inconspicuous.

Distribution.— Saline Valley and southern Death Valley, California

Remarks.— Two subspecies of P. bantai are delimited as follows.

Paruroctonus bantai bantai (Gertsch and Soleglad)

Figs. 3-5

Synonymy.— Same as for species (see above).

Diagnosis.— A subspecies of P. bantai differentiated by combination of: ventrolateral
metasomal setae on segment II 2 (90% of specimens), on V 4 (Fig. 5) (fifth seta, if
present, smaller and offset from others); ventral metasomal setae on segment IV 3 (90%
of specimens), on V 3 (rarely 4); pectinal teeth in males (79%) 23 or fewer, females (75%)
17 or fewer.

Comparisons: P. bantai Saratoga, n. ssp., differs in all five characters (see diagnosis
below).

Table 3.— Numbers of pectinal teeth in subspecies of Paruroctonus bantai 2a\d Paruroctonus shu-
lovi, and in Paruroctonus simulatus, n. sp.

11

12

13

14

15

16

17

18

19 20

21

22

23

24

25 26

27

28

P. b. bantai

Males

Females

4

20

7

3

1

13

9

10

6

3

P. b. Saratoga

Males

Females

1

7

32

23 1

3

3

13

21 11

7

2

P. s. shulovi

Males

Females 1

4

33

44

11

6

12 15

6

P. s. nevadae

Males

Females

2

7

9

4

1

4

7

P. simulatus

Males

Females

1

4

8

4

7

2

1

2 10

9

4

1

1

P. coahuilanus
Males

2

6 11

6

1

HARADON-SCORPION SUBGENUS PARUROCTONUS

29

Variation.— Total adult length of males 37-50 mm, females 50 mm. Adult carapace
length of males 4. 9-6. 2 mm, females 5. 5-6.6 mm. Variation in the numbers of metasomal
setae is presented in Table 2. Pectinal tooth counts in males 20-25, females 16-19 (Table
3). Pedipalp primary denticles, excluding proximal row, total 28-35 (32.15 ± 1.91, n =
39) on fixed finger, 37-47 (41.77 ± 2.01, n = 39) on movable finger. The dsm seta on the
internal brachial surface, and the chelal internal setae, are generally inconspicuous from
the late immature state on.

Distribution.-Saline Valley, CaUfornia.

Specimens examined.— U.S. A.: CALIFORNIA County, Saline Valley, 21 September 1971 (D.
Giuliani), 1 male (CAS), Saline Valley, June 1959 (B. Banta), 7 males, 4 females (CAS), 4 July 1959
(B. Banta), 2 males, 3 females (CAS), 5 July 1959 (B. Banta), 3 males, 2 females (CAS), 22 November
1959 (B. Banta), 2 males (CAS), 27 November 1959 (B. Banta), 6 males (CAS), 28 November 1959
(B. Banta), 1 males, 1 female (CAS), 3 March 1960 (B. Banta), 1 male (CAS), 6 March 1960 (B.
Banta), 1 female (CAS), 3 April 1960 (B. Banta), 1 male, 2 females (CAS), 8 May 1960 (B. Banta), 1
male (CAS), 3 April 1962 (B. Banta), 1 male, 3 females (CAS).

Paruroctonus bantai Saratoga, new subspecies

Fig. 6

Type .—Paruroctonus bantai Saratoga: Holotype male (adult) from U.S. A., California
San Bernardino County, Death Valley Natl. Mon., Saratoga Springs, salt flats, 11 June
1970 (M. A. Cazier, L. Welch, O. F. Franc ke). Depository: California Academy of Sci-
ences, Type. No. 15057.

Diagnosis.— A subspecies of P. bantai differentiated by combination of: ventrolateral
metasomal setae on segment II 3, on V 6 (Fig. 6); ventral metasomal setae on segment IV
4, on V 4 (72% of specimens); pectinal teeth in males (90%) 24 or more, females (87%)
18 or more.

Comparisons: P. bantai bantai differs in all five characters (see diagnosis above).

Description of male holotype.— Measurements: Table 4. Pedipalp primary denticles on
fixed fingers 4,5,7-6,8,10-8,12-14, movable fingers 5,8-6,9,9-11,14,9-10. Metasomal
setae: dorsals 0, 0,0,1; dorsolaterals 0,1, 1,2; laterals 1,0,0,02; ventrolaterals 2,3,3,3,6;
ventrals 3,3 ,3 ,4 ,4.

Allotype.— Measurements in Table 4.

Variation.— Total adult length of males 50-60 mm, females 50-60 mm. Adult carapace
length of males 5.2-6. 1 mm, females 6. 0-7.4 mm. Variation in the numbers of metasomal
setae is presented in Table 2. Pectinal teeth in males 22-28, females 16-20 (Table 3).
Pedipalp primary denticles, excluding proximal mw, total 28-35 (32.26 ± 1.38, n = 62)
on fixed finger, 39-49 (43.81 ± 2.16, n = 62) on movable finger. The dsm seta on the
internal brachial surface, and the chelal internal setae, are generally inconspicuous or
absent in subadult and adult specimens.

Etymology.— The name “Saratoga” refers to the type locality.

Distribution.— Salt flats at Saratoga Springs, southern Death Valley, California.

Specimens examined.-Paratypes. U.S. A.: CALIFORNIA; San Bernardino County, Death Valley
Natl. Mon., Saratoga Springs, 11 June 1970 (M. A. Cazier, L. Welch, O. F. Francke), 29 males, 31
females (OFF), allotype (CAS).

STAHNKEI INFRAGROUP

Diagnosis.— An infragroup of nominate subgenus Paruroctonus differentiated by
combination of: (1) cheliceral fixed digit with inferior carina extending proximally at

30

THE JOURNAL OF ARACHNOLOGY

Table 4. -Measurements (in millimeters) of type specimens of new Paruroctonus species and
subspecies. L = length, W = width, D = depth.

P. bantai Saratoga

Holotype Allotype

male female

P. shulovi
nevadae

Holotype

female

P. simulatus

Holotype Allotype

male female

P. coahui-
lanus

Holotype

male

Total L

50.7

56.2

40.6

38.6

41.6

41.6

Carapace L

6.1

7.0

4.5

4.4

5.3

5.0

Mid-length W

5.0

5.5

3.6

3.6

4.0

4.0

Posterior W

5.8

6.5

4.6

4.0

4.8

4.6

Median eyes W

1.2

1.4

1.0

1.0

1.0

1.0

Mesosoma L

13.8

17.4

10.5

10.7

12.7

11.4

Metasoma I L/W

3.2/3. 1

3. 3/3.6

2.2/2.2

2.6/2.2

2.5/2.4

2.8/2.7

II L/W

3. 9/3.0

3.9/3.4

2.6/2.0

3.0/2.1

2.912.2

3. 2/2.6

III L/W

4.1/2.8

4.1/3. 2

2.8/2.0

3.212.0

3. 1/2.1

3.4/2.4

IV L/W

5. 1/2.8

5. 1/3.2

3.4/1.8

4.0/1. 8

3. 9/2.0

4. 2/2.2

VL/W

7.6/2.8

7. 7/3. 2

5. 2/1.8

5.8/1.8

6.0/2.0

6.1/1.9

Telson L/W

6.9/2.7

7.7/3.4

4. 7/1.9

5.0/1.7

5.212.2

5.4/1.6

Ampulla L/D

4.0/2.3

4. 3/2.6

2.8/1.6

3.0/1. 4

3.4/1. 7

3. 2/1.4

Chehcera palm L/W

1.6/1.3

2.4/1. 6

1.5/1. 2

l.O/l.O

1. 3/1.2

1.4/0.9

Fixed digit L

0.8

0.9

0.6

0.6

0.7

0.9

Movable digit L

1.5

1.6

1.2

1.0

1.4

1.5

Humerus L/W

5. 1/1.7

5.6/2.0

3.6/1. 2

4.0/1.2

4. 4/ 1.4

3.8/1.4

Brachium L/W

5.4/2.3

5.9/2. 3

4.0/1. 2

4. 0/1. 6

4.6/1.8

4.0/1. 7

Pedipalp palm L/W

5.3/4.8

6. 0/5.0

3. 9/2.6

4.0/2.6

4.4/2.9

4.7/2.9

Fixed finger L

4.6

5.1

3.2

3.0

3.7

2.9

Movable finger L

6.4

6.8

4.2

4.0

4.9

4.2

Pectine L

6.2

5.2

3.2

4.6

3.9

4.4

Dentate L

5.9

4.1

2.4

3.9

2.8

3.8

Basal L/W

1. 7/1.4

1.7/1. 2

1.0/0.6

1.0/0.9

1. 2/0.8

-

Pectinal teeth

25/25

19/18

15/15

21/21

16/16

19/19

least to level of bicusp (see Gertsch and Soleglad 1966: fig. 36); (2) cheliceral movable
digit with superior distal tine essentially triangular or elongate and curved, inferior distal
tine length/superior distal tine length ratio 3.0 or less; (3) carapace length/cheliceral fixed
digit length ratio 4.2-8 .0; (4) basitarsus III with six or more superior seta, arranged in
distal plus proximal rows (4+2 to 6+2) or in essentially single file; (5) pectinal teeth in
males 13-23 (except 26-27 in one new species; rarely 24 inP. stahnkei orP. simulatus, n.
sp.), females 8-17 (except 18 in one new species); (6) pedipalp movable finger length/
palm length ratio in adult males 0.8-1 .0 (except 1. 0-1.1 in shulovi microgroup), adult
females 1. 0-1.1; (7) carapace length/pectine length ratio in adult males 1.0-1. 2, adult
females 1.5-2. 2 (except 1.4-1. 5 inP. stahnkei, and 1.4 in shulovi microgroup); (8) humer-
us with two inframe dial macrosetae on proximal 3/5 of internal surface (except two to
three in shulovi microgroup); (9) pedipalp primary denticles, excluding proximal row,
total 17-47 on fixed finger, 22-57 on movable finger; (10) adult carapace length generally
3. 0-5.0 mm in males, 3. 5-5. 5 mm in females.

Comparisons: the gracilior infragroup (above) differs primarily in characters 14, but
also significantly in 5-10; the boreus infragroup (above) differs primarily in character 5,
but also significantly in 6-10. Specimens with pectinal tooth counts exceeding 23/24 in
males or 17/17 in females (see Haradon, 1984b) have 34 or fewer primary denticles on
pedipalp movable finger; all species in the boreus infragroup have 35 or more primary
denticles on the movable finger.

HARADON-SCORPION SUBGENUS PARUROCTONUS

31

Distribution.— Western and southern Great Basin, southward into Sonoran Desert and
Baja California Sur; also Chihuahuan Desert.

Subordinate taxa.— The stahnkei infragroup comprises four primarily allopatric ele-
ments, differentiated as follows and in Table 5.

STAHNKEI MiCROGROUP.Diagnosis: combination of carapace length/ cheliceral fixed
digit length ratio 4.2-5 .0; telotarsi II-IV with one retroinferior terminal seta (similar to
Fig. 16). Distribution: Northern Sonoran Desert. Included species: P. stahnkei (Gertsch
and Soleglad, 1966).

SHULO VI MICROGROUP. Diagnosis: combination of carapace length/ cheliceral fixed
digit length ratio 7.0-8.0; basitarsus II with mrs seta. Distribution: Western and southern
Great Basin. Included taxa: P. shulovi (Williams, 1970); P. shulovi nevadae, n. ssp.;P.
simulatus, n. sp.

BORREGOEN SIS MICROGROUP. Diagnosis: basitarsus II without mrs seta. Distribu-
tion: Southern Great Basin, southward into northwestern Sonora and northern Baja
California Sur. Included species: This microgroup, named after P. borregoensis Williams,
1972, is discussed in detail by Haradon (1984b).

WILLI AMSI MICROGROUP. Diagnosis: combination of carapace length/cheliceral
fixed digit length ratio 4.8-5 .8; telotarsi II-IV with two retroinferior terminal setae.
Distribution: Chihuahuan Desert. Included species: P. williamsi Sissom and Francke,
1981 ;P. pecos Sissom and Francke, 1981 ;P. coahuilanus, n. sp.

Pamroctonus shulovi (Williams)

Figs. 7-10, 15, 17-20

Vejovis (Pamroctonus) shulovi Williams, 1970:7-11, figs. 5, 6, tbl. 3.

Paruroctonus shulovi: Williams 1972:3, 1976:2, tbl. 1; Soleglad 1972:74, 1973:355, tbl. 2; Stahnke
1974:138.

Ty^Q.— Vejovis shulovi: Holotype female (adult) from U.S.A., California, Inyo County,
Death Valley Natl. Mon., Grapevine Spring, 4 miles E Ubehebe Crater, 12 April 1968 (S.
C. Williams, V. F. Lee, J. Bigelow). Depository: California Academy of Sciences.

Diagnosis.— A species of subgenus Paruroctonus, stahnkei infragroup (cheliceral fixed
digit with inferior carina extending proximally to level of bicusp ; pectinal teeth 1 8-22 in
males, 11-17 in females; pedipalp primary denticles, excluding proximal row, 24-31 on
fixed finger, 3442 on movable finger; basitarsus II with mrs seta; dorsal metasomal setae
I-IV 0,1, 1,2), and shulovi microgroup (carapace length/cheliceral fixed digit length ratio
7. 0-8.0; cheliceral fixed digit with denticles on inferior carina), differentiated by: (1)
telotarsi II-IV with two retroinferior terminal setae (Fig. 15); (2) basitarsus III with seven
(5+2) superior setae (Figs. 9-10); (3) pedipalp fingers in adult male deeply scalloped
proximally, closed fingers form wide gap (Fig. 19), in adult female weakly to moderately
scalloped proximally, closed fingers form narrow to moderate gap (Fig. 21); (4) pedipalp
palm length/ width ratio in adult males 1 .3-1 .4.

Comparisons: P. simulatus, n. sp., differs in characters 14 (see below).

Description.— Supplementing above diagnosis and Williams (1970:7). Total adult
length of male 3540 mm, females 3545 mm. Adult carapace length of males 3.84.8 mm,
females 4. 2-5. 6 mm. Trichobothria typical of genus in number and general distribution.
Humeral macrosetae: internals include three inframedials on proximal 3/5, one suprame-
dial; four dorsals; usually two external medials on distal 3/5, small middle seta occasional-
ly present. Brachial macrosetae: four internals. Chela: supernumerary denticles well

32

THE JOURNAL OF ARACHNOLOGY

developed, six on fixed finger, seven on movable finger; primary denticles in six rows on
both fingers; macrosetae include two on internal carina (both long), two on ventrointer-
nal Carina (proximal long, distal short), one short internal on fixed finger, two internal on
movable finger (proximal long, mid-length short). Basitarsus Mil: slightly compressed
laterally; superior setae on I 3+2 irregularly set, II 4+2, III 5+2;mrs seta on I long and in
line with proximal superior setae, on II about 2/3 as long as superior setae and generally
set well apart from superior setae. Telotarsi MV setae: proinferiors 1,2, 2,2; two each
promedials, prosuperiors, retrosuperiors and retromedials; retroinferiors and retroinferior
terminals 1, 2,2,2. Pectines in adult males extend to 2/3 to 3/4 length of trochanter IV,
adult females to or slightly beyond distal edge of coxa IV. Metasomal setae: dorsals
0,1, 1,2; dorsolaterals 0,1 ,1,2; laterals 1,0,0,0,2; ventrolaterals 3,4, 4, 5,6-8; ventrals 3,4,4,
5. Telson of adult male granular.

Variation.— Pedipalp palm length/width ratio in adult males 1.3-1. 4 (1.32 ± 0.02, n =
9), adult females 1.4-1 .6 (1.46 ± 0.05, n = 39). Variation in other characters is discussed
under subspecies.

Distribution.— Death Valley in California, and southern Nevada.

Paruroctonus shulovi shulovi (Williams)

Figs. 7-10, 15, 17, 19-20

Synonymy.— Same as for species (see above).

Diagnosis.— A subspecies of /I shulovi differentiated by combination of: pectinal
posterior basal length/maximum basal width ratio in adult females 1. 8-2.3 (1.96 ± 0.11, n

Table 5.- Distribution and diagnostic characteristics of the four microgroups constituting the
stahnkei infragroup of the nominate subgenus Paruroctonus.

Character

stahnkei

williamsi

shulovi

borregoensis

Desert region

Carapace length/ cheliceral fixed

Sonoran

Chihuahuan

Great Basin

Mojave to
Vizcaino

digit length

Cheliceral fixed digit, inferior

4.2-5.0

4.8-5.8

7.0-8.0

6.8-8.6

denticles (A) absent (B) present
Pedipalp movable finger length/

A

A

B

B

palm length, adult male

Pedipalp primary denticles.

1.0

0.9-1.0

1.0-1. 1

0.8-1.0

rows on movable finger

7

6-7

6

6

less proximal row, fixed finger

3747

21-34

24-34

17-30

movable finger

Pedipalp fingers, adult male;

43-57

33-43

3444

22-38

(A) unscalloped, (B) scalloped
Pedipalp palm carinae, adult

A

A

B

A,B

female: (A) granular, (B) smooth
Humeral internal inframedial

A

B

A

B

macrosetae

Basitarsus II mid-retrosuperior

2

2

2-3

2

seta (A) present, (B) absent
Telotarsi II-IV retroinferior

A

A

A

B

terminal setae

1

2

1-2

2

Carapace length/pectine length 9

1.4-1 .5

1.6

1.4

1.5-2.2

HARADON-SCORPION SUBGENUS PARUROCTONUS

33

= 49); pectine length/dentate margin length ratio in adult females 1.4-1 .6 (1.54 ± 0.05, n
= 49) (Fig. 17); pectinal teeth in males (85%) 18-20, females (88%) 1 1-14; metasomal
segment V with 7/7 or fewer ventrolateral setae (78% of specimens).

Comparisons: P. shulovi nevadae, n. ssp., differs significantly in all four characters (see
below).

Variation.— Pectinal tooth counts varied as in Table 3. Ventrolateral setae on meta-
somal segment V (sample n = 58) varied as follows: 6/6 (15), 6/7 (20), 7/7 (10), 7/8 (10),
8/8 (3); seventh and eighth setae were generally smaller than and offset from other six.
Pedipalp primary denticles, excluding proximal row, 24-31 (28.51 ± 1.63, n = 45) on
fixed finger, 34-42 (37.72 ± 2.12, n = 43) on movable finger.

Distribution.— Northern Death Valley, California.

Specimens examined.--U.S.A.,: CALIFORNIA; /«yo County, Death Valley Natl. Mon., Scotty’s
Ranch (3000 feet), 13 April 1968 (M. A. Cazier, et al.), 3 males 25 females (CAS, OFF), Grapevine
Spring, 4 mi. E Ubehebe Crater (2100 feet), 12 April 1968 (S. C. Williams, V. F. Lee), 2 males, 8
females (CAS), 1 mi. N Ubehebe Crater (2100 feet), 12 April 1968 (S. C. Williams, V. F. Lee), 5

Figs. 7-14. -Right basitarsi. Figs. 7-10, P. shulovi: 7, basitarsus II, retrolateral view; 8, basitarsus
II, superior view; 9, basitarsus III, retrolateral view; 10, basitarsus III, superior view. Figs. 11-14, P.
simulatus, n. sp.: 11, basitarsus II, retrolateral view; 12, basitarsus II, superior view; 13, basitarsus III,
retrolateral view; 14, basitarsus III, superior view. Key to setae: large solid circles = diagnostic superior
setae; small solid circle = mid-retrosuperior (mrs) seta; small open circles = landmark setae. Scale = 1.0
mm.

34

THE JOURNAL OF ARACHNOLOGY

Figs. 15-16. -Right telotarsi, retrolateral views: 15, P. shulovi; 16, P. simulatus, n. sp. Key to
setae: solid circles = diagnostic setae; open circles = landmark setae; RI = retroinferior; RIT = retroin-
ferior terminal; RM = retromedial; RS = retrosuperior; ST = super oterminal.

males, 4 females (CAS), N side Ubehebe Crater Rd., 1.8 mi. W jet. Hwy. 72, 14 October 1977 (J.
Hjelle, W. Savary), 6 males, 4 females (CAS), Mesquite Springs Campground, 14 October 1977 (J.
Hjelle, W. Savary), 1 female (CAS), Mesquite Springs, 5 June 1970 (M. A. Cazier, et al.), 4 males, 4
females (OFF), Mesquite Springs, 10 June 1970 (M. A. Cazier, et al.), 2 males, 1 female (OFF).

Paruroctonus shulovi nevadae, new subspecies

Fig. 18

Type —Paruroctonus shulovi nevadae: Holotype female (adult) from U.S.A., Nevada,
Clark County, Corn Creek Field Station, 4-5 April 1971 (S. D. Slightham). Depository:
California Academy of Sciences, Type No. 15062.

Diagnosis.- Adult male unknown. A subspecies of P. shulovi differentiated by combi-
nation of: pectinal posterior basal length/maximum basal width ratio in adult females
1.5-1. 8 (1 .65 ± 0.10, n = 12);pectine length/dentate margin length ratio in adult females
1.3-1. 4 (1.39 ± 0.04, n = 12) (Fig. 18); pectinal teeth in males (92%) 21-22, females
(91%) 15-17; metasomal segment V with 7/8 or 8/8 ventrolateral setae (88% of speci-
mens).

Comparisons: P. shulovi shulovi differs significantly in all four characters (see above).

Figs. 17-1 8.- Right pectines: 17, P. shulovi shulovi; 18, P. shulovi nevadae, n. ssp. Key: BL =
basal length; DL = dentate length; PL = pectine length; W = basal width. Scale = 1.0 mm.

HARADON-SCORPION SUBGENUS PARUROCTONUS

35

Description of female holotype.-Measurements: Table 4. Pedipalp primary denticles
on fixed fingers 3,4,6-5,7,9-8,17, movable fingers 5,6-5,8,8,11,8. Metasomal setae: dorsals
0.1, 1,2; dorsolaterals 0,1, 1,2; laterals 1,0,0, 0,2; ventrolaterals 2,3 ,3 ,4,7-8; ventrals
3,4,4,5.

Variation.— Pectinal tooth counts varied as in Table 3. Ventrolateral setae on meta-
somal segment V (sample n = 17) varied as follows: 6/7 (1), 7/7 (1), 7/8 (9), 8/8 (6).
Pedipalpal primary denticles, excluding proximal row, total 26-31 (28.65 ± 1.40, n = 23)
on fixed finger, 3542 (38.12 ±1.68, n = 24) on movable finger.

Etymology.— The name “nevadae” refers to the state to which this subspecies is
largely restricted.

Distribution.— Southern Nevada and extreme southeastern Inyo County, California.

Specimens examined,— Paiatypes. U.S.A.: NEVADA; Clark County, Corn Creek Station, 4-5 april
1971 (S. D. Slightham), 4 males, 10 females (CAS); Vye County, 0.8 mi. N California-Nevada border,
along State Rt. 29, 12 August 1974 (R. M. Haradon, W. E. Savary), 1 female (CAS); CALIFORNIA;
Inyo County, 5 mi. N Tecopa (1400 feet), 1 February 1970 (V. Lee), 1 male (CAS).

Paruroctonus simulatus, new species
Figs. 11-14, 16,21-22

Type —Paruroctonus simulatus: Holotype male (adult) from U.S.A., Nevada, Mineral
County, 7 miles N Hawthorne, dunes SE Walker Lake, 15 August 1974 (R. M. Haradon,
W. E. Savary). Depository: California Academy of Sciences, Type No. 15063.

Diagnosis.— A species of subgenus Paruroctonus, stahnkei infragroup (cheliceral fixed
digit with inferior carina extending proximally to level of bicusp ; pectinal teeth 18-24 in
males, 12-17 in females; pedipalp primary denticles, excluding proximal row, 29-34 on
fixed finger, 3844 on movable finger; basitarsus II with mrs seta; dorsal metasomal setae

Figs. 19-22.-Right pedipalp fingers, adult state, external views: 19, P. shulovi, male; 20, P.
shulovi, female; 21, P. simulatus, n. sp., male 22, P. simulatus, n. sp., female. Scale = 1.0 mm.

36

THE JOURNAL OF ARACHNOLOGY

MV 0,1, 1,2), and shulovi microgroup (carapace length/cheliceral fixed digit length ratio
7.0-8. 0; cheliceral fixed digit with denticles on inferior carina), differentiated by: (1)
telotarsi II- IV with one retroinferior terminal seta (Fig. 16); (2) basitarsus III with six
(4+2) superior setae (Figs. 13-14); (3) pedipalp fingers in adult male moderately scalloped
proximally, closed fingers form moderate gap (Fig. 20), in adult female essentially unscal-
loped, closed fingers form at most a very narrow gap (Fig. 22); (4) pedipalp palm length/
width ratio in adult males 1 .5-1.6.

Comparisons: P. shulovi (see above) differs in characters 1-4.

Description of male holotype (allotype).— Measurements: Table 4. Pigmentation: pale
brownish yellow with fuscous markings on carapace, tergites, pedipalps, legs and ventral
surface of metasoma. Carapace: anterior margin straight; surface coarsely (moderately)
granular; furrows and carinae well developed. Tergites: I- VII anterior elevated area
smooth, posterior area finely granular in anterior half and coarsely (sparsely) granular in
posterior half; median carina MI weakly developed (obsolete), III-VII moderately deve-
loped, granular (weak, lightly granular); VII with two pairs granular lateral carinae.
Sternites: III-VI very finely granular (smooth), VII moderately (finely) granular with one
pair weak carinae. Chelicera: fixed digit with one denticle on inferior carina, subdistal
tine meets third superior tine of movable digit; movable digit with superior distal tine
elongate and curved, about 1/3 as long as inferior distal tine. Trichobothria typical of
genus in number and distribution. Humerus: all carinae well developed, granular; inter-
carinal surfaces lightly to moderately (lightly) granular; macrosetae include three internal
inframedials on proximal 3/5, one internal supramedial, four dorsals, two external
medials on distal 3/5. Brachium: all carinae well developed, granular ;intercarinal surfaces
finely granular; four internal macrosetae. Chela: eight major carinae well developed,
granular (lightly to moderately granular); intercarinal surfaces concave, finely granular;
macrosetae include two on internal carina (both long), two on ventrointernal carina
(proximal long, distal short), none on internal surface of fixed finger, one long internal
proximal on movable finger; supernumerary denticles well developed, six on fixed finger,
seven on movable finger; primary denticles on fixed fingers 4,6-5,8-7,7-6,7,13-11, mova-
ble fingers 5,7-8,8,8,10-11,8-9. Basitarsi Mil: not compressed laterally; superior setae on
I 2+2, or 2+3 including mrs seta, II and III 4+2; mrs seta only partially differentiated
from superior setae on I, distinctly offset from superior setae on II and III. Telotarsal
setae MV: proinferiors 1,2,2, 2; two each promedials, prosuperiors, retrosuperiors, retro-
medials; one each retroinferior; one each retroinferior terminal, one extraneous very
slender seta on III right and IV left. Ungues MV about half as long as telotarsus. Pectines
extend to distal margin of trochanter IV (slightly beyond coxa IV). Metasomal carinae:
dorsals MV serrate (crenulate); dorsolaterals MV serrate (crenulate), V granular; laterals I
crenulate to serrate, II granular posterior 1/3 (few posterior granules). III with few
posterior granules, IV obsolete, V granular anterior 1/2 (1/3); ventrolaterals well devel-
oped, MI granular posterior 1/3 (few posterior granules). III with few posterior granules,
IV irregularly crenulate to serrate posterior 1/2, V granular to dentate ; ventrals I moder-
ately (weakly) developed, smooth, IMII moderately to well developed, smooth, IV
irregularly granular to strongly granular, V dentate; intercarinal surfaces very finely
granular except V with scattered coarser granules ventrally. Metasomal setae: long; dorsals
0,1, 1,2; dorsolaterals 0,1, 1,2; laterals 1,0, 0,0,2; ventrolaterals 2,3,3,3,6; ventrals
3,4 ,4,4-5. Telson: ventral and lateral surfaces granular (with few vestigial granules); nine
pairs long ventral and lateral setae.

HARADON-SCORPION SUBGENUS PARUROCTONUS

37

Variation. —Total adult length of males 32-40 mm, females 36-50 mm. Adult carapace
length of males 3.44.6 mm, females 4.0-5 .6 mm (except one specimen 6.6 mm). Pedipalp
palm length/width ratio in adult males 1 .5-1 .6 (1 .51 ± 0.03, n = 13), adult females 1 .5-1 .6
(1.55 ± 0.04, n = 12). Pectinal tooth counts varied as in Table 3. Pedipalp primary denti-
cles, excluding proximal row, total 29-34 (31.53 ± 1.54, n = 19) on fixed finger, 38-44
(41.33 ± 1.97, n = 18) on movable finger. Ventral metasomal setae varied 3,4,4-5,4-5,
usually 3, 4 ,4, 5; ventrolateral setae on segment V (sample n = 27) varied as follows. 6/6
(20), 6/7 (6), 7/8 (1); seventh and eighth setae on V usually smaller than and offset
from other six.

Etymology.— The name “simulatus” refers to the close similarity of this species to its
apparent sister species, P. shulovi.

Distribution.” Western Nevada and northern Inyo County, California.

Specimens examined. -Para types. U.S.A.: NEVADA; Mineral County, 1 mi. N Hawthorne, sand
dunes SE Walker Lake, 15 August 1974 (R. M. Haradon, W. E. Savary), 6 males, 2 females, allotype
{C AS) Esmeralda County, 5 mi. NW Coaldale, 17 December 1972 (collector unknown), 1 male (CAS);
CALIFORNIA; /«yo County, Eureka Valley, sand dunes, 4 September 1975 (D. Giuliani), 1 male, 1

Figs. 23-30.— Right basitarsi. Figs. 23-26, P. pecos: 23, basitarsus II, retrolateral view; 24, basitar-
sus II, superior view; 25, basitarsus III, retrolateral view; 26, basitarsus III, superior view. Figs. 27-30,
P. coahuilanus, n. sp.: 27, basitarsus II, retrolateral view; 28, basitarsus II superior view; 29, basitarsus
III, retrolateral view; 30, basitarsus III, superior view. Key to setae: large sohd circles = diagnostic
superior setae; small solid circle = mid-retrosuperior (mrs) seta; small open circles = landmark setae.
Scale = 1.0 mm.

38

THE JOURNAL OF ARACHNOLOGY

female (CAS), Saline Valley, Racetrack Valley Rd. (1950-2100 feet), 27 November 1959 (B. Banta), 2
males, 1 female (CAS), Saline Valley, Grapevine Canyon Rd. (2300-3400 feet), 27 November 1959
(B. Banta), 1 male, 2 females (CAS), Death Valley Natl. Mon., along Grapevine Canyon Rd., 32 mi.
NWjct. Hwy. 190, 13 October 1977 (J. Hjelle, W. E. Savary), 3 males, 4 females (CAS).

Paruroctonus coahuilanus, new species
Figs. 27-30

Type .—Paruroctonus coahuilanus: Holotype male (adult) from Mexico, Coahuila,
Cuatro Cienegas basin, 14 August 1968 (S. C. Williams, M. A. Cazier, J. Bigelow). Deposi-
tory: California Academy of Sciences, Type No. 15059.

Diagnosis.— Female unknown. A species of subgenus Paruroctonus, stahnkei infragroup
(cheliceral fixed digit with inferior carina extending proximally to level of bicusp; pecti-
nal teeth 18-22 in males; pedipalp primary denticles excluding proximal row, 25-28 on
fixed finger, 32-38 on movable finger; basitarsus II with mrs seta; dorsal metasomal setae
I-IV 1,1, 1,2), and williamsi microgroup (carapace length/ cheliceral fixed digit length
ratio 4.8-5 .6; telotarsi II-IV with two retroinferior terminal setae; cheliceral fixed digit
without denticles on inferior carina), differentiated by: basitarsus III (Figs. 29-30) with
six distal plus two proximal (6+2) superior setae (occasionally 5+2 on one leg only).

Comparisons: P. williamsi and P. pecos differ in having four distal plus two proximal
(4+2) superior setae on basitarsus III (Figs. 25-26); P. williamsi differs further in having
1,3 ,3, 3 dorsal metasomal setae on I-IV.

Description of holotype male.— Measurements: Table 4. Pigmentation: uniformly
pale yellow, except fuscous markings about median ocular tubercle. Carapace: anterior
margin protrudes slightly medially; surface granular; furrows and carinae well developed.
Tergjtes: I-VII anterior elevated area smooth, posterior area finely granular in anterior
half and coarsely granular in posterior half; median carina I-II weak. III- VII moderately
developed, granular; VII with two pairs granular lateral carinae. Sternites: III-VI very
finely granular, VII finely granular with one pair weak lateral carinae. Chelicera: fixed
digit without denticles on inferior carina; similar to P. williamsi andP. pecos (see Sissom
and Francke 1981:figs. 27-28, 31-32). Trichobothria typical of genus in number and
distribution, as in P. williamsi and P. pecos (see Sissom and Francke 1981 : figs. 13-26).
Humerus: all carinae well developed, coarsely granular; intercarinal surfaces lightly
granular; macrosetae include one internal supramedial, two internal inframedials on
proximal 3/5 ; four dorsals; three external medials on distal 3/5. Brachium: all carinae well
developed, coarsely granular; intercarinal surfaces finely granular; four internal macro-
setae. Chela: eight major carinae moderately developed; external carina irregularly and
weakly granular, other carinae irregularly to moderately granular; intercarinal surfaces
finely granular, moderately concave ; supernumerary denticles well developed, six on fixed
finger, seven on movable finger; primary denticles on fixed finger 3-2,4,5,5-6,6-7,13-12,
movable finger 3,6-7,7,7,9-10,9-8; macrosetae include one long on internal carina, two on
ventrointernal carina (proximal long, distal short), one long internal proximal on movable
finger. Basitarsi I-III: moderately compressed laterally; superior setae on I six, irregularly
distributed, on II 4+2, on III 6+2; mrs seta on I not clearly differentiated from superior
setae, on II moderately off set from superior setae (Figs. 27-28), on III considerably off
set (Figs. 29-30). Telotarsal setae I-IV: proinferiors 1, 2,2,2; two each promedials, prosu-
periors, retrosuperiors, retromedials; retroinferiors 1 ,1 ,2,2; retroinferior terminals 1, 2,2,2.
Ungues about 3/5 as long as telotarsus. Pectines extend to about mid-length of trochanter
IV. Metasomal carinae: dorsals well developed, strongly crenulate; dorsolaterals I-IV

HARADON-SCORPION SUBGENUS PARUROCTONUS

39

Strongly crenulate, V coarsely to moderately granular; laterals I crenulate, II-III with few
posterior granules, IV obsolete, V present and granular anterior 1/4 only; ventrolaterals
I- IV well developed, granular, V dentate; ventrals MI moderately developed, lightly
granular, IIMV well developed, granular, V dentate; intercarinal surfaces finely granular,
except scattered coarser granules ventrally on V. Metasomal setae: all long; dorsals
1,1, 1,2; dorsolaterals 2,3,3 ,3,; laterals 2,0 ,0,0,2; ventrolaterals 2,3,4,4,8; ventrals 3,34,4,
5. Telson: smooth; 1 1 pairs ventral and lateral setae.

Variation.— Total adult length 3543 mm. Carapace length of adult males 4. 2-5.1 mm.
Pectinal teeth in males 18-22 (see Table 3). Three specimens (out of 13) each had on one
basitarsus III seven (five distal, two proximal) superior setae instead of the normal eight
(6+2). Pe dipalp primary denticles (three specimens only), excluding proximal row, total
on fixed finger 25-28, movable finger 32-38. Metasomal setae: dorsals 1,1, 1,2, except for
an occasional loss; ventrolaterals I-IV normally 2,3 ,3 ,4; ventrolaterals V with seven to 10
(95% with eight or nine); ventrals normally 3, 4 ,4, 5, except for loss or presence of individ-
ual extraneous seta.

Distribution.— Known only from the Cuatro Cienegas basin, Coahuila, Mexico.

Etymology.— The name “coahuilanus” refers to the state of Coahuila.

Remarks.— The difference in fuscous pigmentation between P. williamsi and P. pecos
reported by Sissom and Tranche (1981 :107) appears to reflect a difference in intensity of
a basically similar but highly variable pattern. The virtual absence of fuscosity in P.
coahuilanus is very likely only a local edaphic characteristic and not of fundamental
taxonomic importance (see P. gracilior Remarks above).

Some males of P. pecos (CAS) from southern New Mexico are very similar to P.
coahuilanus in the development of the ventral metasomal carinae, and therefore any
apparent distinction in this character that might be inferred from the description of P.
pecos by Sissom and Tranche (1981:103) seems, rather, to be obscure. This character is
subject to considerable sexual dimorphism in the williamsi and especially the borregoensis
microgroups (i.e., males have well developed granular carinae, and females have weakly
developed smooth carinae).

The three allopatric species constituting the williamsi microgroup are very similar,
differing significantly only in the two diagnostic characters used to separate them, and
possibly in the tendency of P. coahuilanus to have a longer mrs seta on basitarsus II
(compare Tigs. 23 and 27). Therefore, they might be regarded as subspecies of a single
species. That they are currently isolated from one another is indicated by the fact that
they have so far been found only in, and are probably restricted to, sand dunes. This
would explain, also, why the less restricted species, P. gracilior, though sympatric with
the williamsi microgroup and showing considerable local variation throughout that region,
has remained, in contrast, a single species.

The holotypes of P. williamsi and P, pecos (AMNH) and paratypes of each (OTT,

WDS) were examined, in addition to the following P. coahuilanus material.

Specimens examined.-Paratypes. MEXICO: COAHUILA; Cuatro Cienegas basin, 14 August 1968
(S. C Williams, et al.), 12 males (CAS).

40

THE JOURNAL OF ARACHNOLOGY

PARTIAL KEY TO THE GROUPS AND SPECIES IN
THE NOMINATE SUBGENUS

1. Cheliceral fixed digit inferior carina extends proximally at least to level of bicusp;

carapace length/cheliceral fixed digit length ratio 4.2 or more 2

Cheliceral fixed digit inferior carina does not extend proximally to level of bicusp;
carapace length/ cheliceral fixed digit length ratio 4.1 or less

gracilior infragroup, P. graaVzor

2. Pectinal teeth 24/24 or more in males and 17/18 or more in females and 37 or more
primary denticles (less proximal row) on pedipalp movable fingers, or if fewer pec-
tinal teeth then either (1) dorsal metasomal setae MV 0,0,0, 1, or (2) no mrs seta

on basitarsus II boreus infragroup, 3

Pectinal teeth 23/24 or fewer in males and 17/17 or fewer in females and 36 or fewer
primary denticles (less proximal row) on pedipalp movable fingers, or either (1) more
pectinal teeth, or (2) more primary denticles but dorsal metasomal setae I-IV 0,1, 1,1

or more and mrs seta on basitarsus II. stahnkei infragroup, 6

3 . Basitarsus II with mrs seta 4

Basitarsus II without mrs seta 5

4. Carapace length/cheliceral fixed digit length ratio 7.0 or more

boreus microgroup, 9

Carapace length/cheliceral fixed digit length ratio 6.5 or less

becki microgroup, P. becki

5. Telotarsus III with six or seven retrosuperior setae

xanthus microgroup, P. xanthus

Telotarsus III with two to four retrosuperior setae

baergi microgroup (see Haradon, 1984a)

6. Basitarsus II with mrs seta 7

Basitarsus II without mrs seta borregoensis microgroup (see Haradon, 1984b)

7. Carapace length/cheficeral fixed digit length ratio 7.0 or more; cheliceral fixed digit

with denticles on inferior carina shulovi microgroup, 13

Carapace length/cheliceral fixed digit length ratio 5.8 or less; cheliceral fixed digit
without denticles on inferior carina 8

8. Telotarsi II-IV with one retroinferior terminal seta; primary denticles (less proximal

row) 37 or more on pedipalp fixed finger stahnkei microgroup, P. stahnkei

Telotarsi IMV with two retroinferior terminal setae ; primary denticles (less proximal
row) 34 or fewer on pedipalp fixed finger williamsi microgroup, 15

9. Metasomal setae: ventrals MV 3, 4 ,4, 5 ; ventrolaterals 4 on IV and 7 on V 10

Metasomal setae: ventrals MV 3, 3 ,3 ,3 -4; ventrolaterals 3 on IV and 4 or 6 on V . 11

10. Pedipalp fingers in adult male deeply scalloped proximally, in adult female weakly

scalloped P. arnaudi

Pedipalp fingers in adult male weakly scalloped proximally, in adult female essential-
ly unscalloped P. silvestrii

HARADON-SCORPION SUBGENUS PARUROCTONUS

41

11. Metasomal dorsal setae on I-IV 0,1, 1,2; ventrolateral and ventral metasomal carinae

I-III smooth P. boreus

Metasomal dorsal setae on I-IV 0,0,0, 1; ventrolateral and ventral metasomal carinae
I-III granular P. bantai, 12

12. Metasomal ventrolateral setae 4 on V (Fig. 5) P. bantai bantai

Metasomal ventrolateral setae 6 on V (Fig. 6) P. bantai Saratoga, n. ssp.

13. Telotarsi II-IV with one retroinferior terminal seta (Fig. 16) .... P. simulatus, n. sp.

Telotarsi II-IV with two retroinferior terminal setae (Fig. 15) P. shulovi, 14

14. Pectinal teeth 18-20 in males and 11-14 in females; pectine length/dentate margin

length ratio in adult females 1 .5-1 .8 (Fig. 17) P. shulovi shulovi

Pectinal teeth 21-22 in males and 15-17 in females; pectine length/dentate margin
length ratio in adult females 1.3-1 .4 (Fig. 18) P. shulovi nevadae, n. ssp.

15. Basitarsus III with eight (6+2) superior setae (Figs. 29-30). . . . P. coahuilanus, n. sp.

Basitarsus III with six (4+2) superior setae (Figs. 25-26) 16

16. Metasomal dorsal setae on I-IV 1,3 ,3,3 P. williamsi

Metasomal dorsal setae on I-IV 1,1, 1,2 P. pecos

ACKNOWLEDGMENTS

The following persons made available specimens for study: Dr. 0. F. Franc ke, Texas
Tech University (OFF); Dr. D. H. Kavanaugh and Dr. W. J. Pulawski, California Academy
of Sciences (CAS); Dr. N. I. Platnick, American Museum of Natural History (AMNH); J.
T. Hjelle (specimens now at CAS); W. D. Sissom, Vanderbilt University (WDS); and Dr. S.
C. Williams, San Francisco State University (specimens now at CAS). Credit is also due J.
L. Marks, who contributed much time collecting specimens. Early drafts of several
reports, herein combined, were kindly reviewed by O. F. Francke and S. C. Williams.
Finally, S. C. Williams generously extended many other courtesies, making this study
possible.

LITERATURE CITED

Borelli, A. 1909. Scorpioni raccolti dal Prof. F. Silvestri, nelf America settentrionale e alle isole Ha-
waii. Boll. Lab. Zool. Agr., Portici, 3:222-227.

Chamberlin, R. V. 1920. South American Arachnida, chiefly from Guano Islands of Peru. Brooklyn
Mus. Sci. BuU., 3:3544.

Diaz-Najera, A. 1975. Listas y datos de distribucibn geografica de los alacranes de Mexico (Scorpio-
nida). Rev, Invest. Salud Pub., Mexico, 35:1-36,

Ewing, H. E. 1928. The scorpions of the western part of the United States, with notes on those
occurring in northern Mexico. Proc. United States Natl. Mus., 73(9): 1-25,

Francke, O. F. and M. E. Soleglad, 1981. The scorpion family luridae Thorell (Arachnida, Scorpiones).
J. Arachnol., 9:233-258.

Gertsch, W. J. 1958. Results of the Puritan- American Museum expedition to western Mexico. 4. The
scorpions. Amer. Mus. Novitates, No. 1903, 20 p.

Gertsch, W. J. and D. M. Allred. 1965. Scorpions of the Nevada Test Site. Brigham Young Univ. Sci.
BuU., Biol. Ser., 6(4):1-15.

Gertsch, W. J. and M. Soleglad. 1966. The scorpions of the Vejovis boreus group (subgenus Pzrwrac-
tonus) in North America (Scorpionida: Vejovidae). Amer. Mus. Novitates, No. 2278, 54 pp.

42

THE JOURNAL OF ARACHNOLOGY

Girard, C. 1854. Arachnidians. In: Marcy, R. B., Exploration of the Red River of Louisiana in the year
1852. Washington, 32nd Congress, 2nd Session, Executive Document No. 54, pp. 262-271, zool.
pits, xvi-xvii.

Haradon, R. M. 1983. Smeringums, a new subgenus of Paruroctonus Werner (Scorpiones, Vaejovidae).
J. Arachnol., 11:251-270.

Haradon, R. M. 1984a. New and redefined species belonging to the Paruroctonus baergi group (Scor-
piones, Vaejovidae). J. Arachnol., 12:205-221.

Haradon, R. M. 1984b. New and redefined species belonging to the Paruroctonus borregoensis group
(Scorpiones, Vaejovidae). J. Arachnol., 12:317-340.

Hjelle, J. T. 1972. Scorpions of the northern California Coast Ranges (Arachnida: Scorpionida). Occas.
Pap. California Acad. Sci., No. 92, 59 pp.

Hoffmann, C. 1931. Los Scorpiones de Mexico. Primera Parte: Diplocentridae, Chactidae, Vejovidae.
An. Inst. Biol., Mexico, 2:291-408.

Marx, G. 1888. Untitled (remarks on the type specimens of scorpions described by H. C. Wood in
1863). Proc. Entomol. Soc. Washington, 1:90-91.

Mello-Leitao, C. de 1934. A proposito de um novo Vejovida do Brasil. Ann. Acad. Brasileira Sci.,
6:75-82.

Sissom, W. D. and O. F. Francke. 1981. Scorpions of the genus Paruroctonus from New Mexico and
Texas (Scorpiones, Vaejovidae). J. Arachnol., 9:93-108.

Soleglad, M. E. 1972. Two new scorpions of the gQnus Paruroctonus from southern California (Scorp-
ionida: Vejovidae). Wasmann J. Biol., 30:71-86.

Soleglad, M. E. 1973. Scorpions of the mexicanus group of the genus Vejovis (Scorpionida, Vejovi-
dae). Wasmann J. Biol., 31:351-372.

Stahnke, H. L. 1940. The scorpions of Arizona. Iowa State Coll. J. Sci., 15:101-103.

Stahnke, H. L. 1957. A new species of scorpion of the Vejovidae: Paruroctonus mesaensis. Entomol.
News, 68:253-259.

Stahnke, H. L. 1961. A new species of scorpion of the Vejovidae: Paruroctonus vachoni. Entomol.
News, 72:206-212.

Stahnke, H. L. 1970. Scorpion nomenclature and mensuration. Entomol. News, 81:297-316.

Stahnke, H. L. 1974. Revision and keys to the higher categories of Vejovidae (Scorpionida). J. Ara-
chnol. 1:107-141.

Werner, F. 1934. Scorpiones. PP. 1-316. In: Klassen und Ordnungen des Tierreichs (H. G. Bronn, ed.),
Akad. Verlag., Leipzig, Bd. 5, Abt. 4, Buch 8, 512 pp.

Williams, S. C. 1968a. Scorpions from northern Mexico: Five new species of Vejovis from Coahuila,
Mexico. Occas. Pap. California Acad. Sci., No. 68, 24 pp.

Williams, S. C. 1968b. Two new scorpions from western North America (Scorpionida: Vejovidae).
Pan-Pacific Entomol., 44:313-321.

Wilhams, S. C. 1970. Three new species of Vejovis from Death Valley, California (Scorpionida: Vejovi-
dae). Pan-Pacific Entomol., 46:1-11.

Williams, S. C. 1972. Four new scorpion species belonging to the genus Paruroctonus. Occas. Pap.
California Acad. Sci., No. 94, 16 pp.

Williams, S. C. 1974. A new genus of North American scorpions with a key to the North American
genera of Vaejovidae (Scorpionida: Vaejovidae). Proc. California Acad. Sci., 4th ser., 40:1-16.

Williams, S. C. 1976. The scorpion fauna of California. Bull. Soc. Vector Ecol., 3:1-4.

Williams, S. C. 1980. Scorpions of Baja California, Mexico, and adjacent islands. Occas. Pap. California
Acad. Sci., No. 135, 127 pp.

Wilhams, S. C. and N. F. Hadley. 1967. Scorpions of the Puerto Pehasco area (Cholla Bay), Sonora,
Mexico, with description of Vejovis baergi, new species. Proc. California Acad. Sci., 4th ser.,
35:103-116.

Wood. H. C. 1863a. Descriptions of new species of North American Pedipalpi. Proc. Acad. Nat. Sci.,
Philadelphia, 1863:107-112.

Wood. H. C. 1863b. On the Pedipalpi of North America. J. Acad. Nat. Sci., Philadelphia, 2nd ser.,
5:357-376.

Manuscript received April 1984, accepted June 1984.

Larcher, S. F. and D. H. Wise. 1985. Experimental studies of the interactions between a web-invading
spider and two host species. J. Arachnol., 13:43-59.

EXPERIMENTAL STUDIES OE THE INTERACTIONS BETWEEN
A WEB-INVADING SPIDER AND TWO HOST SPECIES

Scott F. Larcher and David H. Wise^

Department of Biological Sciences
University of Maryland Baltimore County (UMBC)

Catonsville, Maryland 21228

ABSTRACT

Field experiments were conducted to uncover the effects of a web-invading spider, Argyrodes
trigonum (Hentz), on two spider species that serve as its host, Neriene radiata (Walckenaer) and
Metepeira labyrinthea (Walckenaer). A series of short-term experiments, each lasting one to three days,
investigated (1) the effect of \io^t-Argyrodes size differentials on the rate of host emigration and
mortality, (2) the effect of additional food on host Argyrodes emigration, (3) the rate of immigra-
tion to, and emigration from, host-occupied and host-unoccupied webs by Argyrodes, and (4) the use
of host webs by Argyrodes.

The presence of Argyrodes resulted in significant host emigration when \\osi-Argyrodes weight
ratios were below 10:1. In some \m2iS\om Argyrodes killed the resident spider. Additional prey did
not prevent the host from leaving webs containing adult Argyrodes, nor did added prey affect Argy-
rodes emigration from webs. Argyrodes invaded host-occupied and host-unoccupied webs with equal
frequency and captured prey when occupying both types of webs. These latter results suggest that A.
trigonum may often inhabit and use empty webs for prey capture, as well as webs occupied by the
original resident.

Thus, in its interactions with N. radiata andM labyrinthea, the web-invading ^4. trigonum behaves
perhaps as a commensal, and certainly as a predator, a thief of prey, a web-thief, and perhaps a web-
scavenger. The nature of the interaction between A. trigonum and its hosts appears to vary primarily
as a function of the relative size of host spider and A. trigonum.

INTRODUCTION

Spiders of the genus Argyrodes are often described as commensals that inhabit the
webs of other spiders, consuming prey neglected or undetected by the host [Exline 1945,
Archer 1946 (1947), Comstock 1948, Kaston 1978, Gertsch 1979]. Studies of Argyrodes
spp. have revealed that they also behave as kleptoparasites by stealing prey previously
captured by the host [Wiehle 1928, 1931, Thomas 1953, Kullman 1959 (all cited by
Kaston 1965); Robinson and Olazarri 1971, Robinson and Robinson 1973] . Kleptopara-
sitism can be detrimental to the host. Rypstra (1981) showed that prey consumption by
the host Nephila clavipes (Linnaeus) was significantly reduced with each additional
kleptoparasite in the web. Once prey consumption declined below a critical rate, N.
clavipes abandoned its web site.

*To whom reprint requests should be addressed.

44

THE JOURNAL OF ARACHNOLOGY

Argyrodes spp. can also prey on their host. Exline (Exline and Levi 1962) discovered
A. fictilium (Hentz) eating Araneus. Lamore (1958) observed ^4. trigonum (Hentz) eating
the basilica spider Mecynogea lemniscata (Walckenaer), and Archer [1946 (1947)]
reported that A, fictilium would kill and eat Frontinella pyramitela (Walckenaer).
Argyrodes spp. have also been found feeding on Neriene radiata (Walckenaer) and
peira labyrinthea Hentz (Wise 1982; J. Martyniuk, pers. comm.). A. baboquivari was
observed to feed on the eggs, juveniles, and adults of the uloborid Philoponella oweni
(Chamberlin) [Smith-Trail 1980 (1981)]. In an experimental study. Wise (1982) showed
that A. trigonum can cause significant mortality in populations of M. labyrinthea. He
also suggested that web invasions by Argyrodes may lead M. labyrinthea to abandon their
webs. This is consistent with the observation that the webs oi N. radiata andM labyrin-
thea are often found containing only Argyrodes (pers. ob.).

Argyrodes apparently can behave towards its host as a commensal, a kleptoparasite, or
a predator. Responses by the host range from apparent tolerance, through loss of prey,
loss of web as a result of emigration, to loss of life. The outcome of a particular interac-
tion between an Argyrodes species and its host may be determined by combinations of
variables including species and size of the host, species and size of Argyrodes, morphology
of the host web, food intake of the host and Argyrodes, and energy investment by the
host and Argyrodes. Wise (1982) has suggested that interactions between temperate
Argyrodes species and their hosts may vary from commensal to predatory as a function of
variables such as host feeding rate and relative size of host and Argyrodes.

Evidence suggests that the size ratio of host to Argyrodes may be particularly impor-
tant. In situations where the host is very large relative to the web-invading Argyrodes, the
latter may be incapable of injuring the host, or may risk injury to itself if it attacks the
host. In such cases Argyrodes may assume the role of commensal or kleptoparasite.
Commensal and kleptoparasitic interactions appear to be particularly common between
Argyrodes and Argiope and Nephila, hosts which are generally large relative to Argyrodes
(Robinson and Olazarri 1971, Robinson and Robinson 1973, Vollrath 1979a, Rypstra
1981). When host size approximates that of Argyrodes, predation on the host can occur
or the host may emigrate soon after Argyrodes has invaded its web [Smith-Trail 1980
(1981); Wise 1982].

The level of prey consumption by host or Argyrodes may also determine the type of
interaction. The host is likely to remain in the web if its food consumption is above a
certain level, but will leave the web site if stealing of prey by Argyrodes decreases the rate
of prey capture below a threshold value (Rypstra 1981). Prey densities might also influ-
ence Argyrodes' behavior. If the capture rate of small insects by Argyrodes is high, it may
be less likely to kleptoparasitize or attack its host, particularly if risk to the Argyrodes
exists.

Invasion of a web by Argyrodes may be a quest for a web as well as an opportunity to
capture the host. Since Argyrodes has been observed in webs without hosts, possibly
Argyrodes utilizes the web even in the absence of the host. Remaining in the web after
the host has emigrated may be beneficial to Argyrodes if it can use the empty web to
capture prey. The assumed danger of predation while moving to a new web and the
complications of finding a suitable host are reduced if more time can be spent in a given
web. Apparently to date no one has compared the growth, survival and reproduction of
Argyrodes in webs with a host to these parameters for Argyrodes in vacated webs.

The particular questions addressed by this research are the following: (1) Does com-
parison of the size of the host relative to that of Argyrodes reveal a critical size ratio

LARCHER AND mSE-ARGYRODES AND TWO HOST SPECIES

45

where one interaction is more likely than another? (2) Do combinations of particular
developmental stages (e.g., adult host with juvenile Argyrodes) result in a particular
interaction? (3) Does Argyrodes of a particular size or developmental stage exhibit
different behaviors towards different host species? (4) Does the presence or absence of
the host affect the rate of immigration, rate of emigration, and use of the web by Ar-
gyrodes? (5) Does additional prey affect the rate of host or Argyrodes emigration?

Most of the experiments focused on the questions relating to effects of size differences
between host and Argyrodes. A series of experiments was performed in which a range of
h.o%X-Argyrodes ratios was created. Argyrodes of a particular size class were introduced
into occupied host webs and observed at three-hour intervals over a 24 hour period. The
rate of emigration of those hosts was compared to that of host spiders whose webs were
kept free of Argyrodes. Two spider species which have only partially coincident life
histories (i.e., juveniles and adults of both species do not occur at the same time) were
used as hosts so that the behavioral flexibility of Argyrodes towards hosts of different
developmental stages could be observed.

Experiments to study rates of immigration and emigration involved allowing Argy-
rodes to enter or leave webs with the hosts present or absent over a period of several days.
In several experiments of similar design, food was added to some webs and Argyrodes
movement was compared between supplemented and unsupplemented webs. Argyrodes
behavior while in host webs was noted; thus, the results of the experiments can be inter-
preted in the context of specific behaviors of both host and the web-invader.

Argyrodes trigonum (Hentz) (Theridiidae) overwinter as juveniles and are commonly
found in antepenultimate or penultimate stages before June. During June and early July
mature spiders are present. By late June gravid females and females accompanying egg
sacs are abundant. At this time females are sometimes found in small tangles of their own
construction as well as webs of other species. In July the spiderlings emerge from the egg
sac and early-stage spiders are common. Later-instar Argyrodes are commonly found in
September, but numbers begin to decline in October.

The life histories of the two host species used in this study, the filmy dome spider
Neriene radiata (Walckenaer) (Linyphiidae) and the labyrinth spider Metepeira labyrin-
thea (Walckenaer) (Araneidae), contrast with that of A. trigonum. Evidence indicates
some N. radiata complete two generations per year (Wise 1976, in press). Juveniles
overwinter and appear in webs in late March or early April. They undergo several molts
and reach maturity in May- June. These adults reproduce and spiderlings appear in June-
July. Spiderlings that hatched early in the season reach maturity by August or September
and produce another generation of overwintering juveniles. Later-emerging spiderlings
may overwinter as late-stage juveniles before maturing the next spring. N. radiata abun-
dance declines in October and no mature spiders overwinter (Wise 1976). The web of the
filmy dome spider is a fine-meshed dome with an extensive tangle above. The spider hangs
inverted beneath the center of the dome and captures prey that has entered the tangle by
shaking the tangle and pulling the fallen prey through the dome.

Metepeira labyrinthea hatchlings overwinter in the egg sac and emerge in May. Matur-
ity is reached by late July - early August. Females produce egg sacs through October but
males disappear in September. The web of M. labyrinthea is a composite consisting of a
protective tangle and a typical, vertically oriented araneid orb. Within the tangle M.
labyrinthea builds a protective retreat that often includes web debris or an egg sac. A
signal line runs from the center of the orb to the retreat. M. labyrinthea generally remains
in the retreat and advances to the orb on the signal line to capture prey that has hit
the orb.

46

THE JOURNAL OF ARACHNOLOGY

METHODS

Experiments were conducted in mixed deciduous-pine woods on the Patuxent Wildlife
Research Center, Prince Georges Co., Maryland, USA, on natural vegetation and on
artificial web sites. The artificial units consisted of a wooden frame 4 m long, 2 m high,
and 1.6 m wide. Galvanized wire fencing (5.1 cm mesh chicken wire) composed two
1 -meter high rows of undulating waves. These rows were separated by a 1.6 x 4 m hori-
zontal piece of fencing and a similar piece was secured to the top of the unit. The units
were distributed approximately 10 m apart on a grid. Host spiders that had colonized or
had been introduced onto the artificial web sites were used for the experiments. Spiders
on these units were easier to locate and generally easier to remove from their webs than
were spiders found on natural vegetation. Removal of spiders found on natural vegetation
was difficult because movement of small branches often disturbed the web and webs were
often very close to the ground. Webs found on natural vegetation were used when webs
located on artificial web sites provided an inadequate sample size. For all experiments 1-8
spiders were added to each of several units. Only 50-60% of these spiders colonized the
units, so that at the start of an experiment each unit contained approximately three
spiders spaced .5 m to > 3 m apart. For all experiments each web was treated as an
individual replicate, since treatments were assigned at random to individual webs, not to
units.

Effect of Relative Size of Host and Argyrodes.—T)\xnng the summer of 1982, webs
containing adult females or juveniles of the host species were located on the units or,
when necessary, on natural vegetation. These spiders were removed from their webs,
measured and weighed, then returned to their original locations. These webs were ran-
domized into control (without Argyrodes) and experimental (Argyrodes added) treat-
ments. Argyrodes from host webs on the units or in the surrounding vegetation were then
collected, measured, weighed, and introduced into the host webs selected for the experi-
mental treatment.

This type of introduction was conducted several times during the season and encom-
passed a range of host-Argyrodes size combinations: (a) juvenile Argyrodes with both
juvenile and diAeXi Neriene and 2idM\t Metepeira, and (b) 2i&\x\X Argyrodes with juvenile and
adult Neriene and juvenile Metepeira (Tables 1 , 2). The different phenologies of Argy-
rodes and the host species prevented all possible permutations from being used.

The number of Argyrodes added per web depended upon its size relative to that of the
host spider at the time the experiment was conducted. Adult Argyrodes, which were
always females, were added one to a web for all stages of host spider that were used. Only

Table 1. -Outline of the life history phenologies of the spiders used in this study. The term “spider-
lings” is used for animals that have recently emerged from the egg sac. Additional details are found in
the text.

Argyrodes

Neriene

Metepeira

May

Juveniles

Juveniles & Adults

Spiderlings

June

Adults

Adults & Spiderlings

Juveniles

July

Adults & Spiderlings

Juveniles

Juveniles

August

Juveniles

Adults & Juveniles

Adults

September

Juveniles

Adults, Spiderlings
& Juveniles

Adults

October

Juveniles

Juveniles

Adults

ARGYRODES AND TWO HOST SPECIES

<U

lO

CO

CN

1-H

O

CO

-M

3

VO

vq

CN

MO

CN

'Cd;

q

f-H

(U

>

tUO

cd

(X

<5

O

CO

d

CN

d

•

4-J

£

+1

+1

+1

+1

+1

+1

+1

+1

cd

O

'w'

00

cn

00

VO

VO

CN

VO

CN

0>

,

fO

00

CN

CO

00

q

q

t"

d)

•

«

«

•

CN

N

CO

hH

00

CN

VO

*c<^

03

M

1-H

hH

hH

'd"

CO

a

4->

P

e

cd

c/3

03

£

(D

M

£

c

m

CN

CO

CO

CN

CN

CO

c

o

o

o

o

cq

o

o

o

P

CO

cd

o

o

CO

CO

cd

d

d

d

<

+1

+1

+1

+1

+1

+1

+1

+1

o

3

P-H

CM

00

Cf^

CN

1— H

VO

o

s

CO

cn

VO

ov

q

q

q

cn

(D

-(-»

H

CM

r4

O

o

t-H

CN

T— ^

t-H

N

o:

• 1^

c«

bO

• rH

03

o

T3

w

OD

CD

CD

03

V.

P

X

X

X

4-»

P

o

-(->

-»->

G

G

G

-!->

4-*

3

P

• ^H

X)

4-^

<4H

H

CT)

O

<

H

oo

3

T)

<

3

TS

<

(D
>
3
1— »

OD

>

G

>—>

CD

>

G

>—)

3

X

<;

3

<

CD

>

G

>—>

O

O

3C

1-1

o

c

o

03

}-M

£

Os

CN

cq

X

3

CN

hH

MO

s

C

•s

X

CN

CO

O

<0

iz:

00

CO

ocT

vn

VO

VO

'Td

<N

CN

CO

VO

4=^

■4->

OJ

-t-J

cu

o

•

c«

CD

4-»

o

l-i

a

c/3

.g

.g

C3

<D

.

u

CD

<D

CD

CD

•ri,»

CD

1— ^
03

4=»

CD

cn

JD

u

K

CD

,<D

ci.

s;

K

CD

a.

K

<D

e

_N

’C

c:

■£:

4h.a

’s:

• wH

4«-»

Vi

Si

X

cd

H

CLh

c/3

03

"u^

4^

e

c

e

• rH

O

4-#

S

nd

QD

Oi

(N

CN

lO

lo

CO

Tt-

03

P

a

(X

O

cq

vq

oo

q

q

q

c

03

ed

£

cj

CO

CO

d

d

d

d

03

Vi

V— ✓

+1

+1

+1

+1

+1

+1

+1

+1

03

3

•

CN

T— H

VO

ov

o

OV

COV

o

CO

O

CN

CO

vq

q

CN

q

t-H

l-H

03

X)

£

r-^

VO

»o

»o

d

d

hH

+-»

'C

0)

03

£

Cu

.§

3

X

C

a>

/~v

03

O

<D

C

Ph

X

03

X

c

c

m

CN

CN

CO

r— i

,— 1

H

P

c

V— ✓

o

O

o

CO

q

o

q

q

•

Co

c5

CD

CO

(d

d

d

d

d

TO

<

+1

+1

+1

+1

+1

+1

+1

+1

.1-H

OJ

v.)

03

l-H

00

lo

CO

hH

o

C7V

t''

r-H

CO

p

CO

1-H

CN

CN

CN

CN

'd;

q

0>

3

03

P

>.

O

QC

H

1-H

r-H

T— H

d

d

d

a

CO

IH

P

/-*s

pa

P

P

O

03

4-»

O

QC

CO

P

3

U

CD

CD

CD

x

ba

+1

CO

*H

03

O

<

-4H

4H

•M

H-'

4-»

X

G

X

G

•

(0

CO

T3

3

3

3

3

3

CD

CD

CD

s

CD

c

• rH

Pi-

co

H

CO

•x

<

t3

<

<

X

<

TJ

<

>

3

>

G

>—)

3

c

c

O

Qi

CO

X

l-H

<L)

N

• fM

CO

-M

XI

CD

g

VO

• rH

<u

X

P

00

hH

(Jv

CN

VO

OO

VO

z

Z

hH

CO

CN

'd"

CN

i-t

•

<D

G

4-»

>

O

t=!

a

a

3

<U

tH

,<1>

oT

bX)

CO

)-l

X

x

• rH

VO

00

VO

as

cd

-*->

C/3

CD

TS

tj

CD

oo

1

CN

1

8-9

t-H

CO

CN

1

>o

6-7

t-H

00

l_

P4

CO

e3

CN

CN

hH

<N

4h

>.

3

c3

o

CD

IH

c

X

-*->

CD

U

H

M

CD

X

H

CD

X

c

G

G

<

Q

c

3

•— s

"p

CD

CD

00

00

rH

MO

+1

VO

cn

VO

o

o

+l

lo

>0

t-H

3

T3

<

MO

.s

<4i

I

o

o

+1

oo

o

o

+1

CM

lO

o

(U

C

!L>

>

13

<N

CO

o\

VO

CM

I

VO

<N

48

THE JOURNAL OF ARACHNOLOGY

one experiment was performed with single juvenile Argyrodes and an adult host; this
experiment, with adult Neriene females, demonstrated the difficulty of working with
juvenile Argyrodes. Small Argyrodes were difficult to manipulate and appeared fre-
quently to move on and off the webs. By introducing more than om yxvQmiQ Argyrodes
the probability of having at least one remain in the web was increased. Non-experimental
webs often have more than one juvenile Argyrodes per web. Following the preliminary
experiment, three juvenile Argyrodes were introduced to adult Neriene webs to investi-
gate the effect of several small A. trigonum on a single mature host. Two juvenile
rodes were added to webs containing mature Metepeira. In the Neriene - Argyrodes
experiment (September 6-7) that involved juveniles of both species, only oxvq Argyrodes
was introduced per web because of the similarity in size of Argyrodes and juvenile filmy
dome spiders.

Argyrodes was added to the support tangle of Neriene webs and to the barrier tangle
of Metepeira webs. Following Argyrodes introduction each web was censused and the
presence or absence and the location of the host or Argyrodes was noted. Censuses were
conducted every three hours for 24 hours. Argyrodes found in control webs were re-
moved. Adult males of the host species that entered webs were ignored because males
visit different female webs and it is difficult to control their presence in the web.

Net Colonization of Host Webs by Argyrodes.— Dming September 1981, Argyrodes'
preference for webs with a host versus webs without the host was tested. Sixty-two
Neriene webs were divided into host-present and host-absent treatments. All Argyrodes
found in the webs were removed. Argyrodes from the surrounding vegetation were
allowed to invade these webs for the following three days, and the number of webs of
each treatment that contained Argyrodes was noted. The preference of Argyrodes for
host-occupied versus host-unoccupied Metepeira webs was tested in a similar manner in
two separate runs.

In August 1982, the effect of web occupancy by the host on the net colonization by
Argyrodes was tested by direct introduction of Argyrodes. Fifty Neriene webs were
marked, and the host spider was removed from 25 webs. Two juvenile Argyrodes were
introduced into all webs. The number of webs containing A. trigonum was noted in each
of 1-3 censuses conducted each day over the next 72 hours. Two runs of this same
experiment were conducted with mature female Metepeira and two juvenile A. trigonum.
Following the first run of the experiment, treatments were reversed so that all webs
previously designated as controls had the host removed and the original residents of the
removal treatment were returned to their webs.

Argyrodes Colonization of Food-Supplemented Neriene Webs.— Studies of the effect
of food supplementation on the net colonization by mature female Argyrodes of webs of
mature female Neriene were conducted with host-unoccupied and host-occupied webs in
separate experiments during May and June, 1982.

The use of host-unoccupied webs uncovers the effects of supplemental food on Argy-
rodes neglecting any effects due to the host. A single female Argyrodes was introduced
into each of 50 webs from which the host had been removed. Twenty-six of these webs
received supplemental prey. Each feeding round involved introducing a termite larva to
the tangle and noting whether the Argyrodes responded to the prey (defined as increased
activity when prey was introduced), captured the prey, or did not respond. Termite
nymphs, found within rotting logs in nature, are not natural prey of web-building spiders.
However, termite nymphs were used because they could be captured easily, remained in
the tangle after being introduced, continued to move for long periods of time once in the

LARCHER AND mSE-ARGYRODES AND TWO HOST SPECIES

49

tangle, and A. trigonum would feed on them. Feeding rounds were performed for three
days and a final census was taken on the fourth day. A single feeding round was con-
ducted on day one and two rounds spaced 2.5 hours apart were conducted on day two
and three. It was considered that only two feeding rounds were necessary on these days
because Argyrodes usually were feeding on termites from previous rounds or did not
respond to prey introduced in the second round. The experiment was terminated when
50% of the Argyrodes had abandoned the web. Data collected included presence/absence
of Argyrodes, response to the prey, and a subjective evaluation of web quality.

The experiment with host-occupied webs uncovered the effect of supplemented
prey on Argyrodes colonization with the host present, and also provided additional
information on the emigration rate of the host from an ^dr^rcxies-occupied web. Before
an Argyrodes was introduced, a single termite was added to each of 51 webs occupied by
a mature female Neriene. It was thought this preliminary addition of food might decrease
the tendency of Neriene to emigrate when an A. trigonum was introduced. A mature
female Argyrodes was added to each web, and prey were added to 26 webs in two feeding
rounds spaced two hours apart. Each feeding round consisted of adding two termites to
the tangle, separated by a 5-10 min. interval.

By the following day 43 of the hosts had abandoned their webs. Argyrodes had also
emigrated from six of these webs. The experiment was continued with the 37 webs that
contained only Argyrodes. Over the next three days 19 of the webs (all from the original
food-supplemented treatment) received four termite nymphs. The same data were col-
lected as in the experiment involving host-unoccupied webs.

Behavioral Observations of Argyrodes.— Jhrou^out the course of these studies,
behaviors of Argyrodes and its hosts were noted. These observations include capture of
the host spider by Argyrodes, the behavior of Argyrodes in the web with the host present,
kleptoparasitism by Argyrodes, and the behavior of Argyrodes in a web without a host.
These observations will be described in conjunction with the results of the experiments.

RESULTS

Effect of Relative Size of Host decid Argyrodes.— A significant proportion of adult filmy
dome spiders left their webs in response to adult Argyrodes (Table 3). Juvenile Neriene
also abandoned their webs when paired with adult Argyrodes, as did smaller Neriene in
response to juvenile Argyrodes. The only Neriene - Argyrodes combination that did not
result in significant host emigration is that in which very small Argyrodes were introduced
into adult filmy dome webs (Table 3). The additional experiment in which three small
Argyrodes were introduced to each host web also did not result in significant host emigra-
tion (x^ = 0.004, df = 1, p> 0.90; based on emigration of 3/21 control Neriene and 3/22
Neriene with A. trigonum).

Results of the introductions (Table 3) were also analyzed as a 2 x 2 x 5 contingency
table in order to determine whether the effect of Argyrodes on its host varied significant-
ly as a function of the stages paired. The three-way interaction term is significant (x^ =
14.19, df = 4, p = 0.013), confirming that Argyrodes had a varying effect on Neriene.

Some combinations of Argyrodes and Neriene were not studied or observed. Large
Argyrodes were not usually observed in the webs of extremely small filmy dome spiders,
perhaps because their webs are too small to support large Argyrodes. Combinations of
very small Argyrodes and the earliest instar Neriene were not studied because of the
difficulty of working with first and second instar Neriene. Removal of these stages from
their webs could not be accomplished without extensive damage to the web.

50

THE JOURNAL OF ARACHNOLOGY

Table 3. -Results of experiments testing outcome at different ]\o^X-Argyrodes size ratios. All
probabilities are presented for one-tailed statistics with d.f. = 1. (Rem. = remain; Prop. = proportion).

NERIENE

STAGE

no ARGYRODES

with ARGYRODES

Date

Neriene

Argyrodes

N

Rem.

Prop.

N

Rem.

Prop.

P

June 8-9

Adult

Adult

27

23

0.85

31

6

0.19

25.02

0.001

June 15-16

Adult

Adult

25

16

0.64

26

7

0.27

7.08

0.005

July 13-14

Juvenile

Adult

14

9

0.64

12

2

0.17

6.00

0.012

July 25-26

Adult

Juvenile

27

22

0.82

30

25

0.83

0.04

N.S.

Sept 18-19

Juvenile

Juvenile

25

16

0.64

23

4

0.17

10.71

0.002

METEPEIRA

STAGE

no ARGYRODES

ARGYRODES

Date

Metepeira

Argyrodes

N

Rem.

Prop.

N

Rem.

Prop.

P

June 27-28

Juvenile

Adult

21

19

0.90

21

3

0.14

24.44

0.001

July 8-9

Juvenile

Adult

19

16

0.84

18

1

0.06

23.03

0.001

Sept 6-7

Adult

Juvenile

24

24

1.00

24

24

1.00

0.0

N.S.

Sept 25-26

Adult

Juvenile

25

25

1.00

25

24

0.96

1.02

N.S.

Adult Argyrodes increased the emigration rate of juvenile Metepeira (Table 3). Small
and intermediate-sized juvenile Argyrodes had no effect on 2id\x\t Metepeira emigration. It
should be noted that in the latter experiments two svaAl Argyrodes were introduced into
the webs because it was assumed a priori, based on the previous results obtained from
adding single Argyrodes to mature Neriene webs, that single small Argyrodes would not
affect adult Metepeira. The differential response of Metepeira, which varied with the
relative size of host and Argyrodes, was statistically significant (x^ of three-way interac-
tion term from 2x2x4 table = 10.48, df = 3, p = 0.03). The relative size of the host and
invading Argyrodes was quantified by calculating the weight ratio of host to Argyrodes.
Low ratios correspond to a high emigration rate for both host species (Fig. 1).

Certain size combinations of these species do not occur. Adult Argyrodes do not
coincide with adult Metepeira. There may be some overlap between very small A. tri-
gonum and large juvenile Metepeira, but these combinations were not studied because
Metepeira populations were lower during 1982 than previous years (unpubl. data) and we
were unable to locate enough host spiders for an adequate sample size.

Predation on the host spider was observed only early in the season, when mature
Argyrodes were introduced to the host webs containing mature Neriene or juvenile
Metepeira (Fig. 2). Argyrodes that were feeding on a host spider at the first census the
predation was observed were still feeding on the host at the next census 93% (12/14) of
the time. Therefore it appears that most required a minimum of three hours to complete
feeding on a captured host. Because webs were censused every three hours, it is reason-
able to assume that predation by Argyrodes was rarely mistaken for emigration of the
host. Possibly a captured host may have been discarded or lost for some reason before the
capture had been recorded, but this seems to have been an infrequent event.

LARCHER AND mSE-ARGYRODES AND TWO HOST SPECIES

51

# - Ner iene radiata
O- Metepeira labyrinthea

0

10

O

20 ^

40

O

50 60

WEIGHT RATIO, HOST/ARGYRODES

Fig. l.-The effect of increasing hosX-Xo-Argyrodes weight ratio on host emigration. Host emigra-
tion due to the presence of Argyrodes was estimated by subtracting emigration rate in the control
from that in the experimental treatment.

Net Colonization of Host Webs by Argyrodes.-lv^o experiments conducted in 1981
indicate that web invasion by Argyrodes is not affected by the presence of the host.
Argyrodes were allowed to invade host-occupied and host-unoccupied Neriene or Mete-
peira webs. Argyrodes invaded 48% (15/31) of the occupied and 39% (12/31) of the
unoccupied Neriene webs (x^ = 0.59, df = 1, p > 0.25; 2x2 contingency table). Pooled
data for two runs of the same experiment using Metepeira hosts reveals that 35% (13/37)
of host-occupied webs and 43% (16/37) of host-unoccupied webs were invaded by A.
trigonum. As with the Neriene experiment, there is no statistically significant difference
between invasion rates of each type of web by Argyrodes (x^ = 0.51, df = 1, p > 0.25).

Argyrodes that were introduced into host webs during the experiments investigating
the effects of relative size of host and Argyrodes on host emigration did not necessarily
remain in the web (Fig. 3). The proportion of Argyrodes that abandoned the experi-
mental webs appears relatively constant for Neriene but somewhat erratic fox Metepeira.
It was hypothesized that Argyrodes would be more likely to abandon the host web when
the host is relatively large and Argyrodes has no effect on host emigration. Two juvenile
Argyrodes were introduced to each of 25 host-occupied and 25 host-unoccupied Neriene
webs. These Argyrodes were allowed to emigrate from the webs for approximately 72
hours. Presence of the host had no significant impact on the emigration rate of Argy-
rodes. Four left from 25 occupied webs, and none left the empty Neriene webs (p = 0.1 1 ,
df = 1 ; Fisher’s Exact Probability Test). Two runs of a similar experiment mih Metepeira
webs yielded similar results, though overall emigration rates were higher. Data for these
runs were pooled because there was no discernible difference between the size of Argy-
rodes or size of Metepeira used in each run. Thirty -six percent (17/47) of the Argyrodes

52

THE JOURNAL OF ARACHNOLOGY

left the Metepeira-occu^\Q& webs whereas 22% (11/50) left the host-unoccupied webs.
There is no indication of a preference by Argyrodes for either occupied or vacant webs
(x^ = 2.40, df=l,p>0.10;2x2 contingency table).

Argyrodes Colonization of Food-Supplemented Neriene Qhs— Argyrodes placed in
webs from which the host had been removed did not differentially abandon webs in
response to food supplementation (Fig. 4a; “ 0.002, df = 1, p > 0.90; 2x2 contin-

gency table). When food was introduced into webs containing mature Argyrodes and
mature Neriene, web abandonment by Neriene over the next 24 hours was high and did
not differ between treatments (Fig. 4b; x^ = 0.004, df = 1, p > 0.90). In the same time
period few Argyrodes left the webs, and their rate of emigration did not differ between
food treatments (Fig. 4c; x^ = 0.60, df = 1, p > 0.25). Feeding rounds were continued
using webs that contained only Argyrodes. Their disappearance was analyzed between
treatments. Following six feeding rounds (3 days), rates of web abandonment by Argy-
rodes did not differ between experimental and controls (Fig. 4d;x^ = 0.249, df = 1, p >
0.50).

Observations of Argyrodes Behavior in Host Webs.— (1) Occupation of Web Space
by Argyrodes. Argyrodes that occupy an abandoned Neriene web are often found in
the dome in the location previously occupied by the host. A sample of 52 webs derived
from the seventh or eighth census of the host- Argyrodes relative-size experiments that
contained only Argyrodes revealed that in 75% (39/52) of the webs, Argyrodes was in the
dome. Argyrodes was in the dome in only 35% (6/17) of the webs that contained the
host. Argyrodes always occupied the barrier tangle of Metepeira webs, regardless of host
presence (based on 36 host-occupied and 15 host-unoccupied webs).

(2) Prey Capture by Argyrodes in Host-Unoccupied Webs. Argyrodes captured intro-
duced prey when occupying a web abandoned by the host. Termites were introduced into
41 Neriene webs containing only Argyrodes. Fifty-four percent (22/41) captured the

Fig. 2. -Host mortality resulting from Argyrodes. Data are derived from the experimental treat-
ments of \vosX-Argyrodes relative size experiments.

LARCHER AND mSE-ARGYRODES AND TWO HOST SPECIES

53

Fig. 3. -Emigration of Argyrodes from host webs during Eost-Argyrodes relative size experiments.
Sample size is in parentheses.

prey. These data agree with a preliminary experiment in which Drosophila were intro-
duced to Argyrodes-ocm^iQ^ Neriene webs. Fifty-three percent (9/17) of the Drosophila
was captured by the Argyrodes.

Argyrodes that occupy webs containing the host appear to be aware of host move-
ments. During June 1981 Argyrodes were introduced into host-occupied webs and
observed for periods up to one hour. Argyrodes responded to movement in the dome by
outstretching their legs and rotating one leg of the first pair in an apparently searching
manner. Argyrodes sometimes moved towards the host when the host was wrapping prey
and moved away if approached by the host. On one occasion the Argyrodes dropped to a
lower portion of the web when chased by the host Neriene. In another instance a laby-
rinth spider approached the Argyrodes, made contact with the invader, and retreated
across the web while being followed by Argyrodes. Observations of ]\os>t-Argyrodes
combinations throughout this study produced several generalizations: (a) Argyrodes
movement in the tangle or dome of the Neriene web or near the retreat of the Metepeira
web caused the host to stop its own movement or to retreat from the area of the Argy-
rodes, and (b) all host spiders that moved away from an Argyrodes later abandoned
their webs.

Two observations of kleptoparasitism by mature Argyrodes were noted. In the first
example the host Neriene was wrapping a prey item in the dome. The A. trigonum moved
into the dome and approached within 3 cm of the host and prey. The Argyrodes was

54

THE JOURNAL OF ARACHNOLOGY

chased by the host and dropped to a lower portion of the dome. The host retreated to the
opposite side of the dome, leaving the wrapped prey in the upper portion of the dome.
The Argyrodes returned to the prey and began to wrap and feed on the prey. The Neriene
did not return to the upper dome and later abandoned the web. In the second example an
Argyrodes was in the tangle above a Metepeira retreat. The host was in the retreat feeding
on a coleopteran. The Argyrodes approached the tent and touched the beetle. The host
immediately left the retreat and web. The Argyrodes began to feed on the abandoned
prey. In both of these examples the host made no attempt to reclaim the prey and,
instead, retreated from the web once it was aware of Argyrodes' presence.

(3) Capture of the Host by Argyrodes, Mature Argyrodes were commonly found
feeding on females and males of the host species. We never observed the actual capture of
a Neriene female, but have witnessed the capture of a male Neriene and a juvenile Mete-
peira by a mature A. trigonum.

An Argyrodes was in the dome of a Neriene web, with the displaced female in the
lower portion of the dome. A Neriene male entered the web and was approached and
touched by the Argyrodes. The male did not retreat and the Argy’rodes bit the first right
leg. This leg was then wrapped with silk and the Argyrodes bit the third right leg. All
right legs were wrapped and then the entire spider was wrapped. Approximately 30
minutes later the Argyrodes began to feed on the dead Neriene.

In the other instance an Argyrodes was introduced into a web containing a juvenile
Metepeira that was wrapping a prey in its retreat. The labyrinth spider appeared aware of
the Argyrodes but did not retreat as the Argyrodes approached it by climbing the signal
line that runs from the orb to retreat. The Argyrodes bit the Metepeira, which appeared
to die within a minute.

In both cases of an observed capture of a host species by Argy’rodes, wrapping of the
prey was not observed until after a bite had occurred. This differs from situations in
which Argy^rodes catches small insect prey by using a wrap-bite behavioral sequence. An
observation of a very small Argyrodes attempting to capture a large host indicates that
the bite-wrap sequence miglit regularly be employed when Argyrodes is attempting to
capture large prey. A small A. trigonum was observed to approach an adult Neriene and
bite its fourth leg. The Argy’rodes made no obvious attempt to restrain the host and no
obvious damage to the host resulted from the attack.

DISCUSSION

Argy’rodes trigonum causes web abandonment by Neriene throughout a large portion
of the season that these spiders occur together. The only combination that does not result
in significant host emigration (juvenile Argyrodes and adult Neriene) represents a period
of 3-4 weeks. Significant impact of Argyrodes on Metepeira occurs in June and July
(adult Argy’rodes and juvenile Metepeira) but not in September when Argyrodes are very
small relative to the adult Metepeira, One can tentatively predict that weight ratios of
host to Argyrodes below approximately 10:1 will result in significant host emigration.
This value is only an approximation, since some h.o%i- Argyrodes ratios were not studied.
Unfortunately, M. labyrinthea populations were low during 1982, and no data for combi-
nations common during late July and August were collected. Consequently, it is unknown
if there is a transition in Argyrodes' impact as juvenile Argyrodes enter the webs of late-
stage Metepeira juveniles. A prediction could be made that Argyrodes would have little
effect on larger juvenile Metepeira if the weight ratio of host-to-invader is greater than 10:1 .

LARCHER AND WISE-ARGYRODES AND TWO HOST SPECIES

55

Apparently the critical weight ratio of approximately 10:1 is similar between host
species, which suggests that weight ratios might be applied as a predictor of the outcome
of host-Argyrodes interactions. Previous studies have shown that spiders monitor vibra-
tions in the web to locate the position of other organisms (Witt 1975, Suter 1978, Voll-
rath 1979b). Weight is an indicator of size; and if the host is large enou^, Argy rode s may
not attempt to oust the host because of the threat of injury. Studies have shown that
weight is an important determinant of outcome in intraspecific contests for web occu-
pancy for an agelenid (Riechert 1978) and the labyrinth spider (Wise 1983). Predation
by the conspecific appears to be relatively rare for these species. If, as these studies
indicate, the heavier spider frequently gains control of the web site, then one might
expect Argyrodes to rarely displace the host since Argyrodes is usually the smaller of the
two. However, Argyrodes trigonum clearly has evolved the specialized behavior of preying
upon the original occupant of the web, which no doubt explains the readiness of large
residents to vacate their webs when invaded by Argyrodes.

In situations where a single A. trigonum does not force the host from the web, it was
found that several small Argyrodes also did not cause host emigration. The total weight of
several small Argyrodes does not reach the critical ratio that might lead to host abandon-
ment. Also, these small Argyrodes do not seem capable of injuring the host. It is rare to
see more than three Argyrodes in q\X\vqi Metepeira or Neriene webs.

Fewer than 20% of the adult Neriene were captured by introduced adult Argyrodes.
No juvenile Argyrodes was found with a captured N. radiata, and no juvenile was

found dead with any stage of Argyrodes. Perhaps juvenile Neriene are more likely to
abandon their web than mature females because juveniles are smaller and more vulnerable
to predators such as Argyrodes. Since only 4.9% (6/122) of all Neriene were captured by
Argyrodes in the five experiments investigating the role of size differences, it appears that
Argyrodes is not an important direct source of Neriene mortality. Argyrodes may, how-
ever, be a source of indirect mortality, since displaced Neriene may suffer higher mor-
tality from a variety of sources while off the web.

Metepeira were also not often captured by Argyrodes. Only adult Argyrodes were
observed capturing juvenile Metepeira. The overall rate of successful attacks by Argyrodes
on Metepeira was approximately 8.0% (IjSS). Thus it appears mortality rates from
Argyrodes may be of similar importance for Metepeira and Neriene. These rates for both
host species measure the outcome of introduced interactions, and are most valuable for
comparing interactions between different size classes. These measured mortality rates
provide no direct indication of the role or importance of Argyrodes mortality in the
population dynamics of the fihny dome and laybrinth spiders. These experiments give no
indication of how frequently a host spider is exposed to Argyrodes invasions during its
life, nor do they indicate whether mortality from Argyrodes web invasions, either direct
or indirect, is density-dependent.

Predation by Argyrodes cannot necessarily be assumed when an Argyrodes is found
feeding on a dead host spider, particularly when the host spider is over lOx the size of the
Argyrodes. A demonstration in which 15 dead A/, radiata were introduced into the tangle
of webs containing only Argyrodes resulted in three of these dead spiders being eaten by
A. trigonum within three hours of introduction. This is evidence that Argyrodes will
scavenge dead prey; thus, observations of juveniles feeding on host spiders should not be
unexpected. For instance, towards the end of the season, Metepeira reaching the end of

56

THE JOURNAL OF ARACHNOLOGY

their life are often found dead in their webs. If the web is inhabited by juvenile A. tri-
gonum at the time of the host’s death, it would not be unlikely to find the Argyrodes
feeding on the dead host.

Argyrodes occupying Neriene webs abandoned by the host are often found in the
dome as well as the tangle. From either of these locations Argyrodes can successfully
capture prey that enters the web. Argyrodes do not appear to differentially abandon webs
with or without hosts. This suggests that Argyrodes can benefit from a host web regard-
less of presence or absence of the host, and that the multiple benefits of a potential meal
or a web for catching prey are available to the Argyrodes. Rates of immigration of Argy-
rodes into host-occupied and host-unoccupied webs do not differ for both the filmy
dome and laybrinth spiders. This suggests that the presence of the host is not the princi-
pal attraction to the Argyrodes. If host presence was important in terms of being a
potential food source, or as a source of prey capture for the kleptoparasite, the net
colonization of Argyrodes should have been higlier in host-occupied webs. The primary
factor limiting the use of the web by Argyrodes appears to be the web’s structural integ-
rity. The Neriene web is not particularly sturdy and Argyrodes do not appear to substan-
tially reinforce the structure. Argyrodes has been observed to strengthen the tangle of a
Metepeira web with additional silk. This silk could serve the dual functions of additional
support or creating denser mesh for increased capture efficiency.

1.0

.8

CT>

C

E

a>

Od

«/)

(i>

■g

>■,

cn

1-

<

■a

c

ro

a>

c.

*k_

(U

c:

o

tr

R

o

.6

.2

I I unsupplemented webs

Fig. 4. -Effect of food supplementation on proportion of Neriene and Argyrodes remaining in the
web. Details appear in the text.

LARCHER AND mSE-ARGYRODES AND TWO HOST SPECIES

57

Food supplementation did not improve the probability that a mature Neriene would
remain in a web occupied by a mature Argyrodes, nor did supplemental prey increase the
probability of Argyrodes remaining in the web, whether occupied by the host or not. This
suggests that additional food will not influence Neriene to risk staying in the web with
Argyrodes, nor will additional food influence emigration of Argyrodes. Because of the
limited number of such experiments in this study, interpretations must remain tentative.
Prey may not have been in short supply in 1982 for Neriene or Argyrodes. However, lack
of an effect of added prey on the tendency of Neriene to vacate the web more likely
reflects the fact that Argyrodes* major potential impact upon the adult filmy dome
spiders is that of a predator, not a prey kleptoparasite. Added prey may not have reduced
the emigration rate of Argyrodes because it does not repair the empty filmy dome web,
and thus web-site quality is more a function of web integrity than short-term prey cap-
ture rates. These questions require further study.

Prey kleptoparasitism may play a minor role in Neriene-Argy’rodes interactions, since
the filmy dome spider abandons its web when all but the smallest Argyrodes enter the
web. Kleptoparasitism may play a more important role for Argyrodes that inhabit the
webs of mature labyrinth spiders, since this host and invader coexist in the web for the
latter 2-3 months of the season (Aug.-Oct.). During periods when Argy’rodes and its hosts
coexist in the web, there is no indication that prey stealing by Argyrodes leads to the
increased web abandonment that was found in the Nephila- Argyrodes system studied by
Rypstra (1981). If one defines commensalism in terms of the net fitness of both species,
then possibly the cohabitation of small Argyrodes and either host is a commensal rela-
tionship, since there is no apparent damage to the host spiders. Longer-term experiments
are needed, however, to establish conclusively that the presence of even small Argyrodes
does not lower the fitness of the host. Parameters such as net fecundity of the host with
and without the presence of A. trigonum should be studied before all possibilities of
detrimental effects to the host resulting from Argyrodes presence are ruled out.

The term “kleptoparasitism” is usually used to describe a specialized type of competi-
tive behavior, in which one species steals another’s prey. The theft of the host’s web by
Argyrodes can be viewed as an example of web kleptoparasitism. This behavior of Argy-
rodes fulfills the definition of kleptoparasitism for several reasons. First, the loss of the
web represents an energetic loss to the original owner, since it must build a new web,
using energy that could have been applied to future growth or egg production. Also, the
act of moving to a new web site may increase the probability of being preyed upon.
Either result would reduce fitness. Secondly, the theft of the web represents an energetic
gain by the thief. Argyrodes captures prey in the stolen web, and on occasion produces an
egg sac in the web. Argyrodes will inhabit the dome of the Neriene web as did the host or
will reinforce the tangle of the Metepeira web, thereby possibly increasing the web’s
capture efficiency. Habitation of the web may also improve Argyrodes* chances against
predators during times when it is not feeding. These points suggest that the net fitness of
A. trigonum is improved when inhabiting a web abandoned by its host. Further evidence
that Argyrodes views the web of other species as a resource is the fact that Argyrodes are
equally attracted to webs with and without hosts. Finally, this specialized behavior
towards other species is a one-way competitive interaction— A does not build a
web for other species to capture. Hence the specialized term of “web kleptoparasitism”
appears appropriate.

Argyrodes may experience disadvantages to inhabiting an empty web. If the host were
to remain and continue normal web maintenance in the presence of Argyrodes, the

58

THE JOURNAL OF ARACHNOLOGY

amount of time that Argyrodes could remain at the web site should increase, since
apparently Argyrodes does not have the ability to spin the variety of webs spun by the
species it parasitizes. Also, Argyrodes might conserve energy by not being required to
perform web maintenance. However, these disadvantages may be outweighed by gains
accrued by not behaving as a commensal. The energy provided to the Argyrodes by the
capture of the host may be more than the energy conserved by continued presence of the
host in the web. Also, kleptoparasitism of prey may not be more efficient than capturing
live prey, particularly since possible risk of capture by the host exists.

The lifestyle of Argyrodes is characterized by an ability to generalize its behavior when
invading the webs of different host species. Argyrodes' flexible behavior makes possible
the exploitation of different species with nonsynchronous phenologies. Assuming Argy-
rodes moves from web to web in a largely random manner, the species of potential host
next encountered should primarily be a function of that species’ relative frequency in the
habitat. Argyrodes' quest for a habitable web is made more successful by not having to
search for a single host species, or a particular size class of host.

A. trigonum possibly is a commensal, and certainly behaves as a prey kleptoparasite
and host-predator. Which alternative behavior A. trigonum exhibits appears to depend
primarily upon its stage of development and the size of the host spider. Our research
suggests that, in addition to exhibiting these behaviors, A. trigonum spends considerable
time searching out the web of its host species, independently of whether or not the host
spider is present. We propose that A. trigonum and possibly oX\iQx Argyrodes spp., behave
as web kleptoparasites in addition to being web commensals, prey kleptoparasites, and
predators on their hosts.

ACKNOWLEDGMENTS

We wish to thank the Patuxent Wildlife Research Center, U.S. Fish and Wildlife
Service, for providing undisturbed research sites. The University of Maryland Computer
Science Center furnished support for data analysis. A portion of this research was sup-
ported by National Science Foundation Grant DEB 79-11744.

LITERATURE CITED

Archer, A. F. 1946 (1947). The Theridiidae or comb-footed spiders of Alabama. Alabama Mus. Nat.
Hist. Paper No. 22, 67 pp.

Comstock, J. H. 1948. The spider book. Revised and edited by W. J. Gertsch. Comstock Publ. Co.,
Ithaca, New York. 729 pp.

Exhne, H. 1945. Spiders of the genus Conopistha (Theridiidae, Conopisthinae) from northwestern
Peru and Ecuador. Ann. Entomol. Soc. Amer., 38:505-528.

Exline, H. and H. W. Levi. 1962. American spiders of the gQnus Argyrodes (Araneae, Theridiidae).
Bull. Harvard Mus. Comp. ZooL, 127:75-204.

Gertsch, W. J. 1979. American spiders (Revised ed.). Van Nostrand Reinhold Co., New York. 274

pp.

Kaston, B. J. 1965. Some httle known aspects of spider behavior. Amer. Midi. Nat., 73:336-356.
Kaston, B. J. 1978. How to know the spiders (Third ed.). Van Nostrand Reinhold Co., Dubuque,
Iowa. 272 pp.

KuUman, E. 1959. Beobachtungen und Betrachtungen zum Verhalten der Theridiide Conopistha
argyrodes. Mitt. Zool. Mus. Berlin, 35:275-292.

Lamore, D. H. 1958. The jumping spider, Phidippus audax Hentz, and the spider Conopistha trigona
Hentz as predators of the basilica spider, Allepeira lemniscata Walckenaer in Maryland. Proc. Ent.
Soc. Washington, 60:286.

LARCHER AND WISE-ARGYRODES AND TWO HOST SPECIES

59

Riechert, S. E. 1978. Energy-based territoriality in populations of the desert spider Agelenopsis aperta
(Gertsch). Symp. Zool. Soc. London, 42:711-722.

Robinson, M. H. and J. Olazarri. 1971. Units of behavior and complex sequences in the predatory
behavior oi Argiope aurantia (Fabricus) (Araneae: Araneidae). Smithson. Contrib. Zool., 65:1-36.

Robinson, M. H. and B. Robinson. 1973. The predatory behavior of the giant wood spider , N ephila
maculata (Fabricus) in New Guinea. Smithson. Contrib. Zool., 149:1-76.

Rypstra, A. L. 1981. The effect of kleptoparasitism on prey consumption and web relocation in a
Peruvian population of the spider N ephila clavipes. Oikos, 37:179-182.

Smith-Trail, D. 1980 (1981). Predation by Argyrodes (Theridiidae) on solitary and communal spiders.
Psyche, 87:349-355.

Suter, P. B. 1978. Cyclosa turbinata (Araneae: Araneidae): Prey discrimination via web-borne vibra-
tions. Behav. Ecol. SociobioL, 3:283-296.

Thomas, M. 1953. Vie et moeurs des araignees. Payot, Paris. 339 pp.

Vollrath, F. 1979a. Behavior of the kleptoparasitic spider Argyrodes elevatus (Araneae, Theridiidae).
Anim. Behav., 27:515-521.

Vollrath, F. 1979b. Vibrations: Their signal function for a spider kleptoparasite. Science, 205:1149-
1151.

Wiehle, H. 1928. Beitrage zur Biologic der Araneen, insbesondere zur Kenntnis des Radnetzbaues. Z.
Morph. Oekol. Tiere, 11:115-151.

Wiehle, H. 1931. Neue Beitrage zur Kenntnis des Fanggewebes der Spinnen aus den Familien Argiopi-
dae, Uloboridae und Theridiidae. Z. Morph. Oekol. Tiere, 22:349-400.

Wise, D. H. 1976. Variable rates of maturation of the spider, Neriene radiata (Linyphia marginata).
Amer. Midi. Nat., 96:66-75.

Wise, D. H. 1982. Predation by a commensal spider, Argyrodes trigonum, upon its host: An experi-
mental study. J. Arachnol., 10:111-116.

Wise, D. H. 1983. Competitive mechanisms in a food-hmited species: Relative importance of inter-
ference and exploitative interactions among labyrinth spiders (Araneae: Araneidae). Oecologia,
58:1-9.

Wise, D. H. In press. Phenology and life history of the filmy dome spider (Araneae: Linyphiidae) in
two local Maryland populations. Psyche.

Witt, P. N. 1975. The web as a means of communication. Biosci. Commun., 1:7-23.

Manuscript received November 1983, revised April 1 984.

'

>

•i •!; '!

f I .•-»

.» ■“

■) ;»'? tV< Tnyir'. ,,jv.'; IjU V.-'l*^- 'f;'

: ,. 'tfi

’V- »■ ■ .'»fit|, '; ■^■■. tA

i ■ ‘t--. ■ •,. ■

•lit

'.! U,

‘1- 1.' ■

* X < .

^ < wv :* ^ : .A <

Suter, R. B. 1985. Intersexual competition for food in the bowl and doily spider, Frontinella pyrami-
tela (Araneae, Linyphiidae). J. ArachnoL, 13:61-70.

INTERSEXUAL COMPETITION FOR FOOD IN THE BOWL
AND DOILY SPIDER, FRONTINELLA PYRAMITELA
(ARANEAE, LINYPHIIDAE)

R. B. Suter

Biology Department, Vassar College
Poughkeepsie, NY 12601

ABSTRACT

Among web-building spiders, bowl and doily spiders (Frontinella pyramitela) are unusual because
adult males feed frequently. The males rarely build webs, however, and so depend upon females’
snares for foraging. In field and laboratory experiments I assessed the impact of male competition on
female foraging success and growth rate. During periods when males are abundant, a female is likely to
have a male on her web 22% of the time. These cohabiting males capture about 32% of the prey
that hit the web despite the female’s efforts to capture the same prey. As a result, males decrease
female foraging success by about 7% during periods of male abundance. Adult females grow at a rate
that increases linearly with increased food consumption- thus the presence of cohabiting males causes
a corresponding 7% decrease in female growth rate when males are abundant. The literature on spider
foraging and fecundity permits us to calculate that the resultant impact on female fecundity is be-
tween 6.1% and 7.0% depending upon what proportion of growth in adult females is attributable to
maternal biomass increase and what is attributable to egg production. This detriment to the female is
probably outweighed by the high cost of dislodging the male and by a reduction in the probability
that the female will be killed by spiders that mimic prey.

INTRODUCTION

Several taxa of insects and spiders effectively reduce the availability of food for host
spiders by web parasitism or commensalism (see references in Barth 1982 and Krafft
1982) and some adult female spiders suffer a further effective reduction in food supply
because adult males compete for food while in residence on the females’ webs (Rovner
1968, Robinson and Robinson 1978, and see references in Kraft 1982). This situation,
male use of the female’s web for predation, is rare among spiders because in most species
the males take no food during adulthood (Bristowe 1958, Savory 1977; in contrast, see
Eberhard et al. 1978).

The bowl and doily spider, Frontinella pyramitela (Walckenaer), is one species in
which the influence of male competition on female growth and fecundity may be particu-
larly severe. This is because 1) males live with females on the females’ webs for long
periods of time and 2) males capture and feed on prey on those web despite attempts by
the females to prevent such activities. This paper describes some aspects of the feeding
ecology of bowl and doily spiders and assesses the effect of male competition on female
growth and fecundity.

62

THE JOURNAL OF ARACHNOLOGY

MATERIALS AND METHODS

Spiders.— pyramitela is a common inhabitant of low vegetation througliout
much of temperate North America. Its non-viscid web consists of a bowl-shaped horizon-
tal sheet, an underlying flat sheet, and a barrier or knock-down meshwork of silk that is
above the bowl and “doily.” The spider lives on the underside of the bowl, dorsal surface
downward, and captures prey that are deflected onto the bowl by the barrier silk. The
webs are approximately circular when viewed from above and those of adult females have
a diameter of about 8 cm. The female spiders are small (4 mm long, 6 mg) and the males
are smaller still (3 mm, 3 mg). Mating in this species occurs on the underside of the bowl
of the female’s web and is preceded by a complex vibration- and chemical-mediated
courtship (Suter and Renkes 1982, 1984).

Field Procedures.— Prey capture rates by adult female F. pyramitela were assessed in
the field (Poughkeepsie, NY, USA) by observations at one-hour intervals at marked web
sites. The time of capture, size of prey, and web site were recorded for each captured
prey item. More frequent visits to webs were not possible because of the large number of
webs that were under observation at a given time. I found that little inaccuracy in quanti-
fication of capture rates resulted from hourly observations because spiders fed on most
prey items for more than one hour. Prey capture rates were assessed once in 1981 (July)
and twice in 1982 (May and July) for a total of 1093 web-hours. Observations spanned all
hours of the day and night but periods of inclement weather (during rain, or ambient
temperature less than 10°C) were avoided because the spiders temporarily abandoned
their webs or were unresponsive to prey at those times.

I performed censuses of occupied F. pyramitela webs 16 times during the 1982 season.
All adults and late-instar juveniles (greater than 2.5 mm long) were counted and adults’
sexes were recorded. The presence of a common theridiid inhabitant of bowl and doily
webs, Argyrodes trigonum (Hentz), was also recorded. I confined my censuses to a
single unmanipulated study area and, within the confines of that area, tried to achieve
sample sizes of at least 80 spiders. When the population in the study area was very low, I
could not always achieve that sample size.

I measured male-female cohabitation times (the total consecutive time spent by a male
on a female’s web) both in the field and in the laboratory (see below). In the field, both
marked (fluorescent tempera paint on the dorsal surface of the abdomen) and unmarked
males were used at marked web sites. Each male was swung by its dragline onto the
periphery of a web occupied by a solitary female. There the male would usually begin
courtship immediately (Suter and Renkes 1982), mate, and remain with the female for
some time. Timing began as the male first contacted the web and continued until he
could not be found on the web. Though the initiation of timing was precise, its termina-
tion could have been off by 0.5 hour for marked individuals because they were checked
each 0.5 hour, and by 0.5 minutes for unmarked individuals because they were checked
each 0.5 minutes.

Vestigial-winged (Vg) Drosophila melanogaster were used as experimental prey to
assess competition for prey between male and female spiders. These flies were used
because many similarly sized prey are encountered by F. pyramitela in nature (see results)
and because they cannot fly and so are readily captured by the spiders. Naturally formed
cohabiting pairs of spiders that were neither actively courting nor feeding on prey were
presented with a single fruit fly dropped so that it would land approximately midway
between the spiders on the bowl of the web. The spider that ultimately fed on the fly was
designated the winner despite occasional changes of possession during the frequent

SUTER-INTERSEXUAL COMPETITION FOR FOOD

63

Stereotyped contests that preceded feeding. I presented a second fly about 1.5 hours
later, after the first fly had been entirely consumed and its carcass discarded.

Laboratory Procedures.— Adult female spiders captured on hedges near Poughkeepsie,
New York, were placed on glass or wooden hexapods and enclosed in 3.8 1 plastic aquaria.
Because adult males rarely build webs, males were enclosed in 10 ml test tubes stoppered
with cotton. Both aquaria and test tubes contained wet sand which kept the air near
100% RH. I fed vinegar flies to females on their own webs and to males on webs vacated
by females. The temperature in the laboratory varied between 21®C and 23°C.

I measured cohabitation times in the laboratory via continuous visual monitoring of
unmarked males on females’ webs. Methods were identical to those used in the field
(above) except that the webs were on wooden or glass hexapods and were continuously
lighted. The cohabitation times of 40 pairs were measured in the laboratory and 17 were
measured in the field. Because there were no statistically significant differences between
field and laboratory results, both sets of results were pooled.

I assessed the relationship between food consumption and growth rate in 30 adult
female spiders by feeding the spiders different numbers of Vg vinegar flies each day for
eight days. Spiders were weighted to the nearest 0.01 mg at the beginning and end of the
study and the weights of egg masses produced during the study were added to the weights
of the responsible females. The flies had a group mean weight of 0.96 mg.

The mortahty rates of adult male spiders are usually insensitive to food supply because
adult males don’t eat (see introduction). F. pyramitela males do eat, however, and so may
suffer increased mortality when food is scarce or unavailable due to competition. To
check the relationship between feeding history and mortality rates, I fed variable numbers
of vinegar flies to 33 adult males and then withdrew all food. All of the spiders were
maintained at 100% RH until their deaths.

RESULTS

Prey-capture Rates.— During 1093 web-hours of observation, F. pyramitela females
captured 149 prey that varied in length from 0.5 mm to 11 mm. The mean rate of prey
capture (0.14 prey per hour or 1 prey every 7.3 hours) did not vary systematically with
the hour of the day or with the date during the adults’ foraging season. Because many of
the hours of observation were not contiguous, I could not directly derive a frequency
distribution of times between prey captures from the field data. However, because webs
encounter prey randomly with respect to time of day, the distribution of times between
prey captures can be approximated from the exponential distribution with// = 7.3 hours
(Schaeffer and Mendenhall 1975). The distribution indicates that 50% of the time alone
female spider will have to wait less than six hours between prey captures, that 75% of the
time she will wait less than 12 hours, and that about 10% of the time she will have to go
without food for over 25 hours.

Of course, not all captured prey are equally valuable. The insect prey of F. pyramitela
vary considerably in mass and consequently in nutrient content. Figure 1 shows how prey
length varied among prey captured by spiders in the field. More than 80% of all prey
captures were smaller than 3 mm— that is, about the size of a female D. melanogaster or
smaller, and more than half of all prey captures were smaller than 1 .5 mm. At the other
extreme, the largest prey was an 1 1 mm beetle.

The prey lengths shown in Fig. 1 are readily converted into mass units using the
empirically derived conversion formula given by Suter (1977). I converted the frequency

64

THE JOURNAL OF ARACHNOLOGY

distribution of prey lengths into a frequency distribution of prey masses and then built a
two-dimensional matrix whose elements were the products of those mass frequencies and
the exponential frequencies (above) of times between captures. The resulting frequency
distribution of rates of prey capture (in mg/h) reveals that the median prey capture rate is
0.13 mg/h, that 25% of the time the spiders capture at a rate greater than 0.39 mg/h, and
that 25% of the time they capture prey at a rate less than 0.05 mg/h.

Male impact on the female.— The time spent by males on females’ webs in the labora-
tory and in the field varied between 0.37 and 50 hours and showed no systematic differ-
ences between the two sites. The frequency distribution of cohabitation times of 57 pairs
from both field and laboratory sites is shown in Fig. 2. The distribution has a median of
7.8 hours with more than 20% of the males leaving the webs in less than 2 hours and a-
bout 25% remaining on the webs longer than 12 hours. Two of the males were still on
webs after 50 hours.

An assessment of the impact of cohabitation on female foraging depends on informa-
tion about the prevalence of males in a spider population and on the foraging success of
the males. Web censuses throughout the bowl and doily spider’s active season revealed
only one period when males were present (Fig. 3) though in previous years I had seen a
second but smaller pulse of males during September. During the peak period of male
abundance (May 25 to June 9, 1982), males could be found on 21.9 ± 0.6% (mean ± S.D.,
N = 3 samples days) of all F. pyramitela webs. That apparently skewed sex ratio could
have been underestimated because males wander between females’ webs and might remain
off webs for many hours. To test that possibility I performed a removal experiment. All
males (N = 13) were removed from webs in a small population (48 webs) of spiders, and
the appearance of additional males was monitored over the succeeding three days. Over
those three days, only four males appeared on the 48 female-occupied webs. Thus it is
likely that, in a population of bowl and doily spiders, more than 80% of all males in the
population can be found on female webs at any given time. (I could not eliminate immi-
gration of males into the test population, so any or all of the males that appeared during
the experiment could have been immigrants). I conclude that the sex ratio is only slightly
underestimated by counts of spiders on webs.

Also present on F. pyramitela webs were both sexes of the theridiid Argyrodes tri-
gonum. The mean frequency of web occupancy by A. trigonum on three days in late May
was 0.19 ± 0.04 (SD); of 487 webs checked, 94 harbored at least one A, trigonum.

F. pyramitela males captured 32% (24/75) of all test prey (Drosophila) given them in
field studies of male-female competition for prey. When the first fly was presented to
a cohabiting pair, 37% (20/54) of the time the male captured the fly. Second fly presenta-
tions resulted in only 19% (4/21) capture success by males. The difference between first-

Fig. 1.- Frequency distribution of sizes of
prey captured by bowl and doily spiders during
1093 web-hours. N = 149, median = 1 to 1.9 mm
(vertical dashed line).

Prey Size Category (mm)

SUTER-INTERSEXUAL COMPETITION FOR FOOD

65

and second-presentation capture success is not significant (x^ = 1.50). However, a com-
parison of the success of males capturing both first and second prey (0/11, 0%) with
females capturing both (11/22, 50%) revealed that males are significantly less likely than
females to capture two prey in a row when the second prey comes soon after the first
(binomial test, P < 0.001). During 35 of the presentations of test prey, I noted not only
which spider eventually captured the prey but also which spider contacted the prey
first. Males were the first to contact the prey 46% of the time (16/35).

Growth rates of females and starvation morality of males.— Figure 4 shows the results
of a eight day laboratory study in which I fed vinegar flies to female bowl and doily
spiders and measured the spiders’ growth rates. The two spiders that received no food
during the study lost only 14% of their original mass after eight days. Over the 0 to 0.23
mg/h range of capture rates, spider growth rates increased linearly with capture rates. The
linearity of the data indicates that neither satiation nor decreased nutrient utilization
occurred at high feeding rates.

Figure 5 shows the frequency distribution of deaths of male spiders as a function of
time since the last feeding. Starved males died within 34 days of their last vinegar fly meal
if kept at approximately 22° C. Time of death was neither related to the date (i.e. the age
of the spider) (runs test, P > 0.1) nor to the number of flies consumed prior to withdraw-
al of food (rg = 0.08, P > 0.1) nor to the mass of the spider at the beginning of the study
(r^ = 0.01, P > 0.1). Only the time spent fasting was strongly related to time of death
(runs test, P < 0.025).

DISCUSSION

Impact of male competition for prey.— In spiders, as in many other invertebrate
groups, fecundity is directly proportional to the mass of the female and to her foraging
success (Turnbull 1962, Kessler 1971, Riechert and Tracy 1975, Wise 1975, Van Winger-
den 1978). Any limitation of a female spider’s food supply, then, results in a limitation of

Cohabitation Time (h)

Fig. 2.— Frequency distribution of cohabitation times of pairs of spiders from both field and
laboratory observations. N = 57, median = 7.8 hours (vertical dashed line).

66

THE JOURNAL OF ARACHNOLOGY

her fecundity and a consequent limitation of her fitness. The data presented herein show
that male bowl and doily spiders cause a considerable decrease in female foraging success
by competing with them for prey items. The competition must also cause a decrease in
the females’ fecundity and, if females are not all equally tolerant of male cohabitation, a
decrease in their fitness.

The impact of male competition on F. pyramitela females can now be estimated.
During periods of high male .’female ratio, a female is likely to have a male on her web
about 22% of the time (assuming that cohabitation times are approximately evenly
distributed across females). The cohabiting male captures about 32% of the prey that hit
the web despite the female’s efforts to capture those same prey. Thus during periods of
male abundance, males decrease female foraging success by about 7.0% (0.32 X 0.22 X
100).

Though a 7% decrease in prey capture constitutes a major effect of competition, the
impact must be much greater for many females in a population of bowl and doily spiders.
The number of visits of males to a particular female’s web, the time between prey cap-
tures at the web, and the sizes of prey are all highly variable, apparently random parame-
ters. And the tenacity of the visiting males, though sensitive to female reproductive status
(Austad 1982) and to several identifiable male attributes (Suter, unpublished), is not
entirely under the female’s control. Some females will therefore lose considerably more
than 7% of their food to males during periods of male abundance because they cohabit
with more males who stay longer and are more successful at competing for prey. During
the remainder of the reproductive lives of the female spiders, male impact is low or nil
because males are scarce.

Adult female bowl and doily spiders grow at a rate that depends linearly on the
amount of food they consume (Fig. 4). Turnbull (1962), in studies of the fecundity
of another linyphiid spider, Linyphia triangularis, showed that growth in his mature
females was entirely composed of increase in total egg mass rather than increase in ma-
ternal tissue biomass. Assuming that is also the case for F. pyramitela, the impact of
the male competition is then as great on fecundity (or mean egg mass) as it is on the
“growth” rate in Fig. 4. In contrast, Kessler (1971) showed that in four species of wolf

Date during 1982

Fig. 3.- Relative abundances of male, female, and immature spiders inhabiting webs during the
1982 season. Numbers at the top of the graph indicate the total number of F. pyramitela counted on
each census date.

Growth Rate (X 10” ■mg/h)

SUTER-INTERSEXUAL COMPETITION FOR FOOD

67

spiders (Lycosidae) about 13% of adult female growth was increase in maternal biomass
and the remaining 87% was egg production. If bowl and doily spiders are very similar to
the wolf spiders in their reproductive ecology, then 87% (rather than 100%) of the
growth deficit would be subtracted from the female’s current reproductive effort. In
either case, the fecundity deficit caused by the male competition for prey is large-
bet ween 7.0% (all growth is egg production) and 6.1% (87% of growth is egg production)
during the period when males are most abundant. Figure 3 shows that males are common
for about four weeks early in the summer, time enough for the female to produce at least
two, perhaps three clutches of eggs (Kessler 1971, Eberhard 1979, Austad 1982) and
time enough to encompass a majority of the reproductive life of most females (Austad
1982a). Because the aggregate impact of males varies with their abundance, the 6-7%
estimate of their impact at peak abundance overestimates the impact of males during an
entire season.

Benefit (to males) of competition for prey.— In most spider species, the males appar-
ently do not feed as adults (see introduction) but rather wander from female to female,
from web to web, consuming stored nutrients. If male mortality is high while searching

Fig. 4. -Growth rates of adult female bowl and doily spiders as a function of the rate of prey
capture. The solid line indicates the best fit to the data (r = 0.947) for the growth of the 30 spiders in
the study.

68

THE JOURNAL OF ARACHNOLOGY

for females, then male foraging would confer no nutritional advantage because the
spiders, with their comparatively low metabolic rates (Anderson 1970), would die of
other causes before they died of starvation. Thus high off-web mortality is an adequate
explanation for the absence of feeding by males in some species of spiders. What explana-
tions can be offered for the existence of vigorous foraging by adult male bowl and doily
spiders? 1) Off-web mortality from predators may be low due to crypsis, small size,
unpalatability, location of search, etc.; 2) desiccation may be a primary source of male
mortality in dry habitats and thus feeding may more appropriately be considered drink-
ing; or 3) intrasexual competition for scarce resources may favor heavier spiders and
males can become heavier by eating.

Neither I nor the literature have information on mortality in vagabond male spiders
although Robinson and Robinson (1978) suggest that such mortality may be relatively
low for small male spiders because of their inconspicuousness. Desiccation certainly is a
problem for both sexes of F. pyramitela during periods when there is neither rain nor dew
formation, and both sexes can be observed to drink as dew forms or when it rains. How-
ever, males kept at high relative humidity still feed readily when placed on female webs
and this feeding, in the absence of desiccation, prolongs their lifetimes under laboratory
conditions (Fig. 5). Thus the threat of desiccation is not a sufficient explanation of male
feeding in the bowl and doily spider. Several authors have shown that the results of
intrasexual competition among male spiders are biased by mass: the heavier male has a
higher probability of winning an encounter (Rovner 1968, Dykstra 1969, Christenson and
Goist 1979; in contrast, see Aspey 1977). Austad (1983) and Suter and Keiley (1984)
have shown the same mass bias in F. pyramitela. It may be, then, that increase in mass

Time from Lost Food to Death (days)

Fig. 5. -Male morality due to starvation. Twenty three males kept at high RH and at approximately
22°C died within 34 days after their last meal (median =17 days, dashed line). Time of death was
neither related to the age of the spider nor to the number of flies consumed prior to withdrawal of
food nor to the initial mass of the spider. The time spent fasting was significantly related to time of
death (runs test, P < 0.025).

SUTER-INTERSEXUAL COMPETITION FOR FOOD

69

through opportunistic feeding is adaptive for males both because it prolongs their lives
(above) and because it makes them more formidable opponents during agonistic encoun-
ters. Both prolongation of life and success in agonistic encounters would allow a male to
inseminate more females than if he was short-lived and a frequent loser.

Benefit (to females) of male cohabitation.-F. pyramitela males cohabit with females
far longer than is necessary for insemination of the females (Austad 1982) and yet
far shorter than is necessary to guard the female and thereby ensure paternity. Indeed,
Austad has shown that first male sperm priority is so complete in this species that first
males have no need to guard and subsequent males have no paternity to ensure. Those
conclusions and our data indicate that the males remain in webs to feed.

A female’s tolerance of such prolonged male cohabitation is difficult to understand,
then, because her interests are apparently in direct conflict with his. Austad (1984) argues
that, at least with respect to multiple matings, the costs of compliance with the male
are less than the costs of resistance. His argument was tenable for multiple matings
because aU known costs to the female were small. The nutrient cost of lengthy periods of
cohabitation with males, however, is high.

I propose two explanations for female tolerance of male cohabitation. 1) Males are
probably expensive to dislodge. Observations of males and females both during courtship
(Suter and Renkes 1984) and during competition for prey indicate that the two sexes
are equally agile on the female’s web. Thus, though the female is heavier than her mate,
she is unlikely to be able to throw him off without expending considerably time and
energy in prolonged chase. In this respect, the male’s behavior contrasts sharply with his
behavior when confronted with another male: when confronted by an aggressive female,
the male avoids direct contact by fleeing from the female but remaining on the web;
when confronted by another male, the resident male promptly engages in display and
fighting behavior that ends when one male flees from the web altogether (Austad 1983,
Suter and Keiley 1984). 2) The presence of the male decreases by approximately 50%
the probability of female mortality caused by predation by other spiders. Several spiders
in the Theridiidae and Mimetidae prey upon bowl and doily spiders. Typically the bowl
and doily spider senses the presence of the intruding predator, mistakes it for prey, and
rushes to the attack only to be attacked and consumed itself (pers. obs.). When males
share females’ webs, males and females are equally likely to rush at prey and thus equally
likely to rush at predators that mimic prey. In my study areas, Argyrodes trigonum,
known to be a predator on other spiders (Wise 1982) is very common during the period
when male F. pyramitela are also abundant and may therefore contribute strongly to
mortality during that period. Indeed, I have frequently observed A. trigonum feeding on
both sexes of bowl and doily spiders. I hypothesize, therefore, that a major benefit to the
female of prolonged male cohabitation is the deflection of predation from female to
male. I am currently testing this hypothesis in field populations.

ACKNOWLEDGMENTS

Marilyn Ramenofsky and John Wingfield provided valuable insight into the possible
benefits to the female spiders of prolonged male cohabitation. Their comments are
greatly appreciated. I also thank Alfred M. Dufty and Valerie J. Suter for their helpful
comments on the manuscript. The hours of field observations logged by Michael Keiley
are also appreciated and were supported by an NSF Undergraduate Research Participation
grant (SPI-8025922). Additional support for the study came from the Beadle Fund of
Vassar College.

70

THE JOURNAL OF ARACHNOLOGY

LITERATURE CITED

Anderson, J. F. 1970. Metabolic rates of spiders. Comp. Biochem. Physiol., 33:51-72.

Aspey, W. P. 1977. Wolf spider sociobiology: I. Agonistic display and dominance-subordinance rela-
tionships in adult Schizocosa crassipes. Behaviour, 62:103-141.

Austad. S. N. 1982. First male sperm priority in the bowl and doUy spider, Frontinella pyramitela
(Walckenaer). Evolution, 36:111-185.

Austad, S. N. 1983. A game theoretical interpretation of male combat in the bowl and doily spider
(Frontinella pyramitela.) Amin. Behav., 31:59-73.

Austad, S. N. 1984. The evolution of sperm priority patterns in the spiders. In Sperm Competition
and the Evolution of Animal Mating Strategies, R. L. Smith (ed). Academic Press, New York.

Barth, F. G. 1982. Spiders and vibratory signals: sensory reception and behavioral significance. In
Spider Communication: Mechanisms and Ecological Significance, P. N. Witt and J. S. Rovner (eds).
Princeton University Press, Princeton, N.J.

Bristowe, W. S. 1958. The World of Spiders. Colhns, London.

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

Dykstra, H. 1969. Comparative research of the courtship behaviour in the Pardosa (Arachn.

Araneae): III. Agonistic behaviour mPardosa amentata. Bull. Mus. Nat. Hist., 41:91-97.

Eberhard, W. G. 1979. Rates of egg production by tropical spiders in the field. Biotropica, 11:292-
300.

Eberhard, W. G., M, Barreto and W. Pfizenmaier. 1978. Web robbery by mature male orb-weaving
spiders. Bull. Brit. Arachnol. Soc., 4: 228-230.

Kessler, A. 1971. Relation between egg production and food consumption in species of the genus
Pardosa (Lycosidae, Araneae) under the experimental conditions of food-abundance and food
shortage. Oecologia, 8:93-109.

Kraft, B. 1982. The significance and complexity of communication in spiders. /« Spider Communica-
tion: Mechanisms and Ecological Significance, P. N. Witt and J. S. Rovner (eds.) Princeton Univer-
sity Press, Princeton, N. J.

Riechert, S. E. and C. R. Tracy. 1975. Thermal balance and prey availability: bases for a model relat-
ing web-site characteristics to spider reproductive success. Ecology, 56:265-284.

Robinson, B. and M. H. Robinson. 1978. Developmental studies of Argiope argentata (Fabricius) and
Argiope aemula (Walckenaer). Symp. Zool. Soc. London, 42:31-40.

Rovner, J. S. 1968. Territoriality in the sheet web spider Linyphia triangularis (Clerck) (Araneae,
Linyphiidae). Z. Tierpsychol., 25:232-242.

Savory, T. 1977. Arachnida. Academic Press, New York.

Schaeffer, R. L. and W. Mendenhall. 1975. Introduction to probability: theory and applications.
Duxbury Press, North Scituate, Massachusetts.

Suter, R. B. 1977. Cyclosa turbinata: prey discrimination via web-borne vibrations. Ph.D. Thesis.
Indiana University, Bloomington, Indiana.

Suter, R. B. and M. Keiley. 1984. Agonistic interactions between male Frontinella pyramitela (Ara-
neae, Linyphiidae). Behav. Ecol. Sociobiol.

Suter, R. B. and G. Renkes. 1982. Linyphiid spider courtship: releaser and attractant functions of a
contact sex pheromone. Anim. Behav., 30:714-718.

Suter, R. B. and G. Renkes. 1984. The courtship of Frontinella pyramitela (Araneae, Linyphiidae):
patterns, vibrations, and functions. J. Arachnol., 12:37-54.

TurnbuO, A. L. 1962. Quantitative studies of the food of Linyphia triangularis Clerck (Araneae:
Linyphiidae). Can. Entomol., 94:1233-1249.

Van Wingerden, W. K. R. E. 1978. Population dynamics of Erigone aretica (White) (Araneae, Linyphii-
dae) II. Symp. Zool. Soc. London, 42:195-202.

Wise, D. H. 1975. Food limitation of the spider Linyphia marginata: experimental field studies.
Ecology, 56:637-646.

Wise, D. H. 1982. Predation by a commensal spider, Argyrodes trigonum, upon its host: an experi-
mental study. J. Arachnol., 10: 111-116.

Manuscript received July 1 983, revised April 1984.

Rypstra, A. L. 1985. Aggregations oi Nephila clavipes (L.) (Araneae, Araneidae) in relation to prey
availability. J. Arachnol., 13:71-78.

AGGREGATIONS OF NEPHILA CLA VIPES (L.)
(ARANEAE, ARANEIDAE) IN RELATION TO
PREY AVAILABILITY

Ann Lundie Rypstra

Department of Zoology
Miami University
1601 Peck Blvd.
Hamilton, Ohio 4501 1 , USA

ABSTRACT

Females of the species, clavipes (L.), sometimes build their webs in interconnected clusters.

Aggregations in a Peruvian population of this spider were located in areas of high insect activity. This
locale enabled individuals in aggregations to capture more prey than solitary individuals. Experiments
with prey removal and prey supplementation verified that a high prey capture rate was essential in
maintaining such groups. Agonistic interactions frequently preceded departure of a spider from a
cluster. These data imply that high prey availability is a prerequisite for the evolution of more com-
plex sociality in Nephila and other similar spider species.

INTRODUCTION

The evolution of sociality in various groups of animals has received much attention in
light of the theory of natural selection (Wilson 1975, Maynard-Smith 1983 and references
therein). Spiders are an unlikely recipient of this attention since they are voracious
predators that frequently recognize conspecifics as prey. Nevertheless three levels of
social interaction have been described for spiders based on some degree of tolerance,
interattraction and cooperation (Shear 1970, KuUmann 1972). These social types include
(1) colonial spiders that forage in interconnected webs, (2) spiders that aggregate only for
the protection and rearing of young, and (3) spiders that share a common web in which
they cooperate both in prey capture and in the rearing of young. Aggregate behavior
in arachnids increases foraging efficiency, habitat exploitation, ease of mate location, and
offers protection from predators and parasites (Buskirk 1975, Lubin 1974, Brach 1977,
Rypstra 1979).

There appear to be two evolutionary pathways capable of producing some form
of aggregation in spiders (Shear 1970, KuUmann 1972, Krafft 1979, Buskirk 1981).
Along one route sociality results from an extension of parental care and the family unit.
In the second pathway communal habits are the result of opportunism: selection favoring
those individuals able to reap some ecological advantage from unintentional contact with
conspecifics. In the study presented here, I focus on the evolution of intraspecific toler-
ance via that second pathway.

72

THE JOURNAL OF ARACHNOLOGY

High prey availability is a likely prerequisite for the formation of aggregations in
spiders. In spider species that maintain well-defined territories, field-measured nearest
neighbor distances are smaller in populations that live in areas with high prey densities
(Riechert 1978, 1981, Uetz et al. 1982). In situations where prey abundances are artifi-
cially maintained at exceptionally high levels territories disappear and both inter- and
intraspecific tolerance appear in species that are normally solitary (Rypstra 1983).
These factors make it tempting to hypothesize that occurrences of very high natural prey
abundances make intraspecific competition for resources less necessary and allow the
evolution of more amicable interactions between individuals.

The golden-web spider, Nephila clavipes (L), is the focal species of this study . Nephila
females build large webs in the open forest or edge habitats of tropical and subtropical
America (Gertsch 1949, Peters 1954). The webs are fme-meshed circular orbs with
varying amounts of barrier webbing surrounding them. An insect, detained by the sticky
orb area, is usually subdued, wrapped and transported to the center of the web. Prey can
either be cached or consumed immediately (see Robinson and Merick 1971, Moore 1977
for more details). Smaller spiders, called kleptoparasites, frequently occupy the barrier
webbing and steal captured prey from the host spider (Vollrath 1976, Rypstra 1981).

I selected TV. clavipes for this investigation of the relationship between food availability
and sociality in spiders for two reasons. (1) Although adult females are usually solitary, in
some habitats, individuals build webs adjacent to or intertwined with the webs of others
(Shear 1970, Robinson and Merick 1971, Farr 1977). Understanding the circumstances
when members of this species tolerate conspecifics in close proximity should provide
insight into the factors that make possible the evolution of more advanced sociaUty. (2)
Individuals of this species respond within one or two days to low prey consumption rates
by relocating their webs (Rypstra 1981). Since long-term web site tenacity is apparently
a reflection of an adequate prey supply, one would predict that webs remaining in close
proximity are gathered around a favorable prey source.

METHODS

Aggregations of Nephila clavipes were studied in an area of subtropical moist forest
in the Tambopata Reserve Zone, 29 km SSW of Puerto Maldonado, Department of
Madre de Dios, Peru. Data were collected during June and July 1983. This time is the
beginning of the dry season. Searches were made of all areas associated with major paths
of the reserve. Each TV. clavipes female found was marked with a drop of acrylic paint on
her abdomen. If the barrier silk of two or more webs was contiguous the spiders were
considered aggregated. Solitary spiders had webs with no silk connections to the webs of
others.

I observed natural spider webs for two-hour periods at some time between 0900 and
1500 h. I recorded all activities of spiders and of insects that moved within ten cm of the
webbing. Specific attention was paid to insect activity, prey capture rates, aggressive
interactions in complexes, and kleptoparasite actions. Ten periods (20 hours) were spent
with solitary individuals and 1 1 periods (22 hours) were spent with web complexes of
four or more spiders. On three occasions I attempted to transfer kleptoparasites from the
barrier webbing of one web to another.

An independent measure of prey availability was obtained via adhesive traps. Sheets of
plastic (10x10 cm) covered with Tack Trap^M (Animal Repellents, Inc., Griffin, Geor-
gia) were strung in the forest undergrowth for four-hour intervals. Sixty-four samples

RYPSTRA-AGGREGATIONS 0¥ NEPHILA

73

were taken within a seven-day period. Thirty-two sheets were positioned 50 cm away
from the capture surface of an individual in an aggregate and 32 sheets were placed 50 cm
away from solitary webs. At the end of four hours the traps were collected and the
captured insects were counted and categorized by order and size.

These observations allowed simple comparisons between the webs of solitary and
aggregated individuals. In an attempt to reinforce any conclusions made about the forma-
tion and maintenance of aggregations, I conducted two experiments.

Experiments 1 : Prey Removal.— A natural aggregate of six N. clavipes females located
near a small stagnant stream was selected. Normal prey activity around this complex was
determined by observation for four hours prior to manipulation. For ten consecutive days
I visited this complex at 14:00 h. At that time I removed prey items, both living and
dead, contained in the webbing. The diffuse nature of the barrier webbing made it possi-
ble to remove approximately 80-90% of the insects with little or no damage to the
structure. I also noted changes in position or number of N. clavipes while observing the
group for one hour. Prey removal was terminated after ten days. I returned to the com-
plex on six subsequent days to determine if any other changes in the colony occurred. On
these visits I plucked the webs near each prey item without removing them. This proce-
dure should control for effects of web disturbance on colony integrity.

Experiment 2: Prey Supplementation.— females that have been starved for 48
hours will usually spin a web where they are released. Using this technique I created an
aggregate of three spiders in an area for which both adhesive traps and insect observations
indicated low prey activity. I visited this group daily at 10:30 h. for ten consecutive days.
Ten to 15 live fruit flies {Drosophila spp.) were gently placed onto the capture surface of
each N. clavipes web. I recorded the number and position of the spiders and observed
their actions for one hour after prey addition. Supplementation was terminated after ten
days. I returned to the site for six days to monitor the fate of this artificial aggregate. On
these occasions I carefully touched the webbing in 10 places with a live insect but did not
leave any additional prey in the trap.

RESULTS

Thirty-two N. clavipes females were located in the study area. Four natural aggregates
accounted for 17 of these spiders (one with three members; two with four; one with six).
The remaining 15 spiders were solitary.

During my observations N. clavipes spent about four percent of its time in web main-
tenance activities (clearing debris and repair). No difference is evident between solitary
and aggregated webs (Table 1). Since all observations were of foraging spiders in com-
pleted webs no overall time/energy budget budget was determined.

Seventy-seven percent of the prey captured by N. clavipes in the Tambopata forest
were small dipterans and hymenopterans (1-5 mm in total length). Many larger insects
seemed to be capable of either avoiding the web or escaping before they were attacked.
Overall, the webs captured 79% of the insects that contacted them (Table 1). The activity
of insects in the small size range was significantly higher near aggregated webs than it was
near solitary webs (Table 1). Individuals in aggregations captured more prey than did
those in solitary webs (Table 1).

Two species of kleptoparasites in the genus Argyrodes Simon (Araneae: Theridiidae),
occupy the barrier webbing of N. clavipes web in the Tambopata forest. From 0-12
Argyrodes fed on the prey of each host spider (Table 1). Similar numbers of Argyrodes

74

THE JOURNAL OF ARACHNOLOGY

Table 1. -Comparative data collected for solitary (20 observation hours) and aggregate (22 hours)
webs of TV. clavipes. The presence of a * means there is a significant difference between the two groups
using the Mann-Whitney U-test, p < 0.05.

Solitary Webs

X ± S. D.

Aggregated Webs

X ± S. D.

Prey Activity/h (adhesive traps)*

3.9 ±2.2

10.2 ± 3.4

Prey Activity/h (observed)*

7.6 ± 2.4

11.6 ± 3.8

Prey Captured/spider/h*

5.5 ± 1.4

10.3 ± 3.1

Capture Efficiency (captures/prey in web)

77.1 ± 7.5

82.5 ± 9.1

Web Maintenance (min/h)

2.6 ± 1.7

2.3 ± 2.0

Total Number of Kleptoparasites

3.2 ± 1.4

3.6 ± 1.9

Kleptoparasites/TVep/z//fl*

3.2 ± 1.4

1.2 ± 0.9

Prey Lost to Kleptoparasites/TVep/zz7fl/h

2.2 ± 1.7

1.4 ± 1.2

Prey Captured by Each Kleptoparasite/h*

2.2 ± 1.7

4.2 ± 1.4

occupied solitary and aggregated webs (Table ). Consequently those individuals in col-
onies acquired significantly more prey (Table 1). Nephila females gave no indication that
they were aware of the other spiders in the web. In no instance did they move into the
barrier webbing or recover an item. Prey losses per Nephila individual were not signif-
icantly different between solitary and grouped spiders (Table 1). Kleptoparasite trans-
plants were unsuccessful. Argyrodes, when introduced, retreated to a remote position or
dropped out of the webbing entirely.

Aggressive interactions between TV. clavipes females consisted of a rapid exchange
of web jerks. During interactions individuals would orient toward the other spider with
the anterior legs nearly straight out. In no instance did one female move onto the capture
surface of the other. The spider would, however, move into the barrier webbing within
three to four cm of another before either retreated. Only four interchanges were observed
in natural aggregates. All of these were initiated by the same individual in one four-
member colony. That individual had a capture rate of 5.5 prey per hour, which was the
lowest recorded for any colony member.

Experiment 1.— The prey activity around this six-member colony was 15.4 insects/h
which was the highest measured. Between six and 28 prey items were removed from each
web surface daily during the experimental period (Table 2). On the third day of removal.

Table 2. -Data for individuals involved in prey removal experiment (exp. 1) including: prey capture
prior to manipulation, average number of prey removed during the ten day experimental period, the
day on which that individual departed from the colony, and whether that spider returned in six days
after removal procedures were stopped.

Spider

Initial

Capture Rate/
spider/hour

Number of Prey
Removed

X ± S. D.

Day

Left

Return?

1

7.5

9.0 ± 0.0

3

no

2

8.0

9.3 ± 2.5

4

yes

3

8.5

14.0 ±3.5

5

no

4

12.0

20.4 ± 4.6

8

yes

5

13.0

21.6 ± 7.4

>10

—

6

13.5

22.0 ± 5.4

>10

~

RYPSTRA-AGGREGATIONS OV NEPHI LA

75

Nephila individuals began to relocate away from the group (Figure 1). There is a correla-
tion between the amount of prey captured by a spider in a given position and the time at
which it relocated (Kendall’s Tau = 0.933, p < 0.05) (Table 2). On days nine and ten a
total of eight aggressive interactions took place between the two remaining females in the
complex. After prey removal attempts were terminated, three Nephila females joined the
aggregation (Figure 1). Two of the three were former residents of the colony (Table 2).
The third previously occupied a solitary web about 12 m away. The number of Argyrodes
in the aggregate declined form 12 to six during the experimental period.

Experiment 2.— The natural activity of prey around the artificially created aggregate
was relatively low at 3.9 insects/h. During the period in which prey were added to the
webs no Nephila females left and one previously unidentified female joined the group
(Figure 1). After supplementation ceased, spiders began to disperse out of the area
(Figure 1). Three aggressive interactions were observed on day 11. The initiator of these
encounters was absent from the complex on day 12. Two acts of aggression were ob-
served on day 13 and followed with the departure of the intiator by day 14. No klepto-
parasites colonized these webs during this experiment.

DISCUSSION

Web construction constitutes a large energy investment on the part of a spider (Ford
1977, Prestwich 1977). Therefore it is essential that the web provide the spider with a
suitable return in the form of insect prey. Many spiders, including A! clavipes, appear to
make decisions about relocating their web based on their prey consumption rates (Turn-
bull 1964, Gillespie 1981, Rypstra 1981). Thus, it is not surprising that clusters of N.
clavipes females should be associated with patches of prey. Even with this small sample
size, the prey distributions, capture rates, and aggression levels recorded for the Peruvian
population support this contention. Further evidence is provided by the relatively rapid
dispersal of group members when food consumption was experimentally lowered (Figure
1). Similar results were obtained for the colonial Metapeira spinipes (Araneae;

Araneidae) in Mexico. Uetz et al. (1982) found both the number of individuals remaining
in a colony as well as the nearest neighbor distances within the group related to prey
availability.

Agonistic interactions between N clavipes females appear to be related to their prey
consumption rate and to precede departure form a colony. Aggression between conspe-
cifics at low prey levels is a likely factor operating to break apart aggregates. The metabo-
lic cost of interactions coupled with a marginal prey capture rate could hasten the need
for web relocation of juxtaposed individuals. At high prey levels aggression was not
observed, presumably because competition for prey is reduced. One hypothesized prere-
quisite for a complex social existence to evolve is consistently high prey levels. Interest-
ingly enough most social spiders live in tropical regions which typically have higher
overall insect abundances than do comparable habitats in the temperate zone (Janzen
1973, Janzen and Pond 1975).

It has been suggested that spider’s silk is a preadaptation for the evolution of social
behavior in arachnids (Shear 1970). The silk acts as a communication network that
precludes the need for physical or even visual contact during information transfer. The
reactions of N clavipes females to conspecifics in their web is clearly distinct from their
reaction to potential prey items. During such agonistic encounters genuine communica-
tion occurs via an exchange of signals (Krafft 1982). Discrimination of web signals is key

76

THE JOURNAL OF ARACHNOLOGY

to the development of any more complex social structure (Krafft 1982). However,
the precise structure of an orb web seems to set limits on the amount of communal
behavior that can evolve in a species such as N. clavipes (Burgess and Witt 1976, Burgess
and Uetz 1982). The vibratory information that is transmitted in the circular capture
surface of the web is focused on a single central point where there is only room for one
spider (Burgess and Witt 1976). For this reason gregarious orb-weavers are usually colo-
nial, maintaining individual webs within a matrix of interconnected webs (Buskirk 1981,
Burgess and Uetz 1982, Krafft 1982).

It was not the purpose of this paper to evaluate if N. clavipes reaps any advantages
directly form group participation. There is no difference in web maintenance expenses or
capture efficiency between soUtary and communal webs evident in the data presented
here. No predation attempts were observed. Farr (1977), working with a population of N.
clavipes in Florida, concluded that clumping was a stochastic phenomenon influenced by
population density and the availability of suitable web sites. My data do not negate that
contention in so far as a suitable site is one with a high prey yield for the spider. Farr
(1977) also cited two disadvantages to colony formation in this species; lowered feeding
efficiency and increased direct competition for mates. Prey capture rates did vary among
the positions in Peruvian colonies (Table 2), however, all webs in the complexes were
more productive than solitary webs (Table 1). In addition, experiments suggest that high
prey capture rates are essential for continued group existence. This study generated no
data on the mating hierarchies that might exist within N. clavipes colonies. However in

a

3

o

W

o

c

«

w

a>

2

o.

(O

o

w

0)

n

E

3

z

I

« • ♦

f

T 1 1 1 r 1 1 r— 1 1 1 1 1 T I // T"

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 20

V

Days

(End of Experiment)

Fig. 1. -Number of N. clavipes in aggregations during experiments. In the prey removal experiment
(bottom line) prey were removed from the webbing once a day for the first ten days. In the prey
supplementation experiment (top line) 10-15 fruit flies were provided daily for each spider during the
first ten days.

RYPSTRA-AGGREGATIONS OF NEPHILA

77

Other studies of communal spiders, facilitation of sexual encounters was suggested as an
advantage to group living (Lubin 1974, Valerio and Herrero 1977). The resolution of this
question requires comparable data concerning the fitness of solitary females vs. the fitness
of low-ranking females within an aggregation.

In a previous study on N. clavipes, the activity of kleptoparasites had a substantial
affect on web site tenacity (Rypstra 1981). However in the few webs available for study
here similar numbers of kleptoparasites occupy web complexes as live in single webs
(Table 1). Based on that piece of information one would predict that aggregated individ-
uals should loose fewer prey because the stealing events are spread over all of those in the
group (“selfish herd effecf ’ Hamilton 1971). No difference in prey losses to kleptopara-
sites by individual Nephila females were revealed between the two web situations (Table
1). Alternatively, if these few data reflect actual trends, kleptoparasites are experiencing
significantly higher prey levels in Nephila aggregations (Table 1). The Argyrodes may be
limiting their own density within webs in order to increase their food intake. My inability
to alter Argyrodes densities in host webs made it difficult to test any of this more specif-
ically.

Regardless of the actual position of N. clavipes in the evolutionary progression to
spider sociality, the role of prey consumption in maintenance of aggregations in Peru has
been established. The variability in spacing patterns that this species displays make it a
valuable model system with which the situations that might allow for the appearance of
more complex social interactions can be clarified.

ACKNOWLEDGMENTS

Thanks to M. P. Frisbie and the staff at the Tambopata Wildlife Reserve for help with
field experiments. T. G. Gregg, E. J. Mury Meyer, G. A. Polis, G. E. Stratton and G. W.
Uetz made many helpful suggestions on an earlier draft of this manuscript. Financial
support was derived from the National Geographic Society and the Faculty Research
Committee of Miami University.

LITERATURE CITED

Brach, V. 1977. Anelosimus studiosis (Araneae: Theridiidae) and the evolution of quasisociality in
Theridiid spiders. Evolution, 31:154-161.

Burgess, J. W. and G. W. Uetz. 1982. Social spacing strategies in spiders, Pp. 317-351, In: Spider
Communication: Mechanisms and Ecological Significance (P. N. Witt and J. S. Rovner, eds.).
Princeton Univ. Press, Princeton, New Jersey.

Burgess, J. W. and P. N. Witt. 1976. Spider webs: design and engineering. Interdisc. Sci. Rev., 1:322-
335.

Buskirk, R. E. 1975. Coloniality, activity patterns and feeding in a tropical orb-weaving spider. Ecol-
ogy, 56:1314-1328.

Buskirk, R. E. 1981. Sociality and the arachnida. Pp. 281-367, In: Social Insects. Vol. II. (H. R.
Hermann, ed.). Academic Press, New York.

Farr, J. A. 1977. Social behavior of the golden silk spider, Nephila clavipes. J. Arachnol., 4:137-144.
Ford, M. J. 1977. Energy costs of the predation strategy of the web-spinning spider Lepthyphantes
zimmermanni (Bertkau) (Linyphiidae). Oecologia, 28:341-349.

Gertsch, W. J. 1949. American Spiders. Van Nostrand-Reinhold, Princeton, New Jersey.

Gillespie, R. G. 1981. The quest for prey by the web-budding spider Amaurobius similis (Blackwell).
Anim. Behav., 29:953-954.

Hamdton, W. D. 1971. Geometry for a selfish herd, J. Theor. Biol., 31:295-311.

78

THE JOURNAL OF ARACHNOLOGY

Janzen, D. H. 1973. Sweep samples of tropical foliage insects: effects of seasons, vegetation types,
elevation, time of day, and insularity. Ecology, 54:687-708.

Janzen, D. H. and C. M. Pond, 1975. A comparison, by sweep sampling, of the arthropod fauna of
secondary vegetation in Michigan, England, and Costa Rica. Trans. Royal Entomol. Soc. London,
127:33-50.

Krafft, B. 1979. Organisation et evolution des societes d’araignees. J. Psychol., 1:23-51.

Krafft, B. 1982. The significance and complexity of communication in spiders. Pp. 15-66, /«.' Spider
Communication: Mechanisms and Ecological Significance (P. N. Witt and J. S. Rovner, eds.).
Princeton Univ. Press, Princeton, New Jersey.

Kullmann, E. 1972. The evolution of social behavior in spiders (Araneae, Eresidae and Theridiidae).
Amer. Zool., 12:395-405.

Lubin, Y. D. 1974. Adaptive advantages and the evolution of colony formation in Cyrtophora
(Araneae: Araneidae). Zool. J. Linnean Soc., 54:321-339.

Maynard-Smith, J. 1983. The evolution of social behaviour-a classification of models. Pp. 33-66, In:
Current Problems in Sociobiology (King’s College Sociobiology Groups, eds.). Cambridge Univ.
Press, Cambridge.

Moore, C. W. 1977. The life cycle, habitat and variation in selected web parameters in the spider
Nephila clavipes Koch (Araneidae). Amer. Midi. Nat., 98:95-108.

Peters, H. M. 1954. Estudios adicionales sobre la estructura de la red concentrica de las arahas. Corn-
mum. Inst. Trop., El Salvador, 1:1-18.

Prestwich, K. N. 1977. The energetics of web-building in spiders. Comp. Biochem. Physiol., 54A:321-
326.

Riechert, S. E. 1978. Energy-based territoriality in populations of the desert spider Agelenopsis aperta
(Gertsch). Symp. Zool. Soc. London, 42:21 1-222.

Riechert, S. E. 1981. The consequences of being territorial: spiders, a case study. Amer. Nat., 117:
871-892.

Robinson, M. H. and H. Merick. 1971. The predatory behavior of the goldenweb spider Nephila
clavipes (Araneae: Araneidae). Psyche, 78:123-139.

Rypstra, A. L. 1979. Foraging flocks of spiders: a study of aggregate behavior in Cyrtophora citricola
Forskal (Araneae; Araneidae) in west Africa. Behav. Ecol. Sociobiol., 5:291-300.

Rypstra, A. L. 1981. The effect of kleptoparasitism on prey consumption and web relocation in a
Peruvian population of the spidQi Nephila clavipes. Oikos, 37:179-182.

Rypstra, A. L. 1983. The importance of food and space in limiting web-spider densities; a test using
field enclosures. Oecologia, 59:312-316.

Shear, W. A. 1970. The evolution of social phenomenon in spiders. Bull. Brit. Arachnol. Soc., 1:65-76.

Turnbull, A. L. 1964. The search for prey by a web-building Achaearanea tepidariorum (CL
Koch) (Araneae, Theridiidae). Canadian Entomol. 96:568-579.

Uetz, G. W., T. C. Kane, and G. E. Stratton. 1982. Variation in the social grouping tendency of a
communal web-building spider. Science, 217:547-549.

Valerio, C. E. and M. V. Herrero. 1977. Tendencia social en adultos de la Pam2i Leucauge sp. (Aran-
eae, Araneidae) en Costa Rica. Brenesia, 10/11:69-79.

Vollrath, F. 1976. Konkurrenzvermeidung bei tropischen kleptoparasitism Haubennetzspinnen der
Argyrodes (Arachnida: Araneae: Theridiidae). Entomol. Germ., 3:104-108.

Wilson, E. O. 1975. Sociobiology, the New Synthesis. Belknap Press, Cambridge.

Manuscript received April 1 984, revised June 1984.

Zolnerowich, G. and N. V. Homer. 1985. Gnaphosid spiders of north-central Texas (Araneae, Gnapho-
sidae). J. Arachnol., 13:79-85.

GNAPHOSID SPIDERS OF NORTH-CENTRAL TEXAS
(ARANEAE, GNAPHOSIDAE)

G. Zolnerowich^ and N. V. Homer

Department of Biology
Midwestern State University
Wichita Falls, Texas 76308

ABSTRACT

Thirty-two species representing 1 1 genera of Gnaphosidae are recorded from north-central Texas.
The study has extended the ranges of Callilepis chisos Platnick and Shadab, Gnaphosa altudona
Chamberlin, Haplodrassus chamberlini Platnick and Shadab, Herpyllus hesperolus Chamberlin, Nodo-
cion rufithoracicus Worley, Rachodrassus captiosus (Gertsch and Davis), and Sergiolus angustus
(Banks). Habitat and natural history data for species are presented.

INTRODUCTION

North-central Texas includes, as used here, Wilbarger, Wichita, Baylor, Archer, Clay,
and Montague counties. The eastern portion of the area is included in the Cross Timbers
and Prairies, whereas the western portion is within the Rolling Plains (Gould 1975). The
study area is included in the Texas and Kansan biotic provinces of Blair (1950).

Five araneid studies have confined themselves to the north-central Texas area. Carpen-
ter (1972) conducted a survey of the Salticidae of Wichita County; and Cokendolpher,
Horner, and Jennings (1979) reported on the Philodromidae and Thomisidae. Zaltsberg
(1977) and Salmon and Horner (1977) studied aerial movements of spiders in Wichita
Falls, and Matelski (1982) investigated Peckhamia (Salticidae). Other information on
spiders from this area appears as locality records in revisionary works or checklists. The
lack of information concerning gnaphosid spiders from this area prompted this study.
This paper describes the gnaphosid fauna of north-central Texas and presents natural
history data and range extensions for species.

METHODS AND MATERIALS

The collection of gnaphosid spiders at Midwestern State University was examined,
checked for proper taxonomy and habitat, and locality records recorded. Intensive field
work was conducted in Wichita County from July 1981 to September 1982. The primary
method of collection was overturning ground cover. Large to small stones, logs, card-
board, lumber, sheet metal, and cow manure all harbored gnaphosids. Fallen bark, leaves.

* Present address: Department of Entomology, Texas A & M University, College Station, Texas 77843

80

THE JOURNAL OF ARACHNOLOGY

and other ground litter were placed in a sifter with a 1 cm mesh screen and shaken over a
white cloth. Any spiders present were easily seen and captured in snap-top plastic vials. A
sweep net was employed where the vegetation permitted. The outer bark of trees was
stripped away with a metal probe and examined for gnaphosids. Twenty-one pitfall traps
were employed in Wichita County. Each trap consisted of a 1500 ml can dug flush
into the ground and filled 2/3 full with a 1:1 mixture of water and ethylene glycol. The
trapping periods were July 12-19, August 9-16, September 13-20, and October 11-18,
1981. At the end of each trapping period the traps were removed from the ground and
the contents poured through a fme-meshed strainer and sorted. This method was aban-
doned when only three adult gnaphosids were collected.

Immature specimens were reared to maturity in the laboratory. Rearing chambers were
made of glass or plastic tubes measuring 1 x 9, 2 x 10, or 3.5 x 10 cm. Spiders were
offered food three times a week, principally termites and adult fruit flies {Drosophila
melanogaster Meigen). Cotton stoppers were wetted once a week for five weeks but the
spiders shunned the moist sides of their tubes and were not observed drinking. The
practice was discontinued with no ill effects. Each tube contained a strip of heavy paper
to provide cover and a molting surface. The tubes were cleaned after each molt and as
prey debris accumulated.

Adult spiders collected in the field and those matured in the laboratory were killed
and preserved in 70% ethanol and have been added to the Department of Biology In-
vertebrate Collection at Midwestern State University.

RESULTS AND DISCUSSION

Suitable ground cover seemed to be a major requirement for gnaphosids in this region.
Localities with abundant cover often harbored 10 or more species within a small area.
Notable exceptions were areas with sandy soil; these yielded very few spiders even if
ground cover was present. The majority of the adult and penultimate specimens were
collected from February to May, after which the variety and number of species declined
significantly.

The accounts that follow include natural history data such as behavior, habitat, and
egg sac contents.

(1) Callilepis chisos Platnick.— Two females were collected in May and September;
one was found indoors and the other in bark debris along a small lake. These specimens,
from Wichita County, extend the known range approximately 300 miles eastward from
San Miguel County, New Mexico (Platnick 1975).

(2) Cesonia sincera Gertsch and Mulaik.— Eight of the nine specimens collected were
found under rocks along the margins of two reservoirs and a pond. Similar habitat a few
hundred yards from the collection sites yielded no additional specimens. Seven of the
nine spiders were immature and were reared to adults in the laboratory. Two immature
males collected on 26 March matured in 11 and 14 days. The five immature males col-
lected in late March and April reached maturity between 25 April and 1 July.

(3) Drassodes gosiutus Chamberlin.— Only females were collected, mostly in March.
Of the 16 specimens, 14 were found under rocks in broken country. Twelve of the spiders
were guarding single egg sacs within their hibernacula. The hibernacula consisted of silken
tubes stretched across the undersides of rocks, or of silk-lined burrows extending straight
down into the ground. The female would not leave her hibernaculum until it was broken
open; she would then seize the egg sac in her chelicerae and attempt to drag or push it to

ZOLNEROWICH AND HORNER -GNAPHOSIDAE OF NORTH-CENTRAL TEXAS

81

safety. The 12 egg sacs contained 60, 63, 82, 83, 95, 98, 105, 105, 106, 116, and 147
eggs; one egg sac held 101 deutova. A mature female collected 26 March deposited an egg
sac on 1 1 April. She carried the sac in her chelicerae but dropped it on 12 April to feed.
Three days later she opened the egg sac and scattered the eggs. Although Platnick and
Shadab (1976a) state that mature males may be found from late June to late December,
none were collected by us.

(4) Drassodes saccatus (Emerton).— This species was more common than Drassodes
gosiutus. All 65 specimens were found under some type of ground cover, usually stones.
Gertsch (1979) states that the males of some species of Drassodes enclose immature
females in hibernacula adjacent to their own. Thirteen males were found sharing hiber-
nacula with females. Of these, 12 mature males were found cohabiting with penultimate
females and one mature male was found in a hibernaculum with a mature female. The
hibernacula varied from a sac just large enough to enclose the two spiders to silken tubes
12 cm or longer. The spiders would usually exit from opposite ends of their retreat when
disturbed. Most of the specimens were collected between 18 March and 13 April. AU the
males found during this period were mature. Six penultimate females molted between 1
April and 22 April. A female caught on 19 May had an egg sac which contained 1 1 1 first
instar spiderlings.

(5) Drassyllus aprilinus (Banks).— Platnick and Shadab (1982) list this species from
Montague and Wichita counties and state that mature spiders have been collected year-
round. The species was not encountered in this study.

(6) Drassyllus dromeus Chamberlin.— This spider does not appear to favor a particu-
lar habitat. Sixteen specimens were found from March to June under rocks, boards,
bricks, sheet metal and willow tree bark; indoors, and by sweeping grass. Two penulti-
mate females collected 18 March and 16 April molted 21 March and 24 April.

(7) Drassyllus lepidus (Banks).— This was the most common Drassyllus collected.
Twenty-seven specimens were found from May to October under rocks, concrete, boards,
sheet metal, and indoors. Eight immatures (seven males and one female) collected in April
reached maturity by 30 May. An immature female collected on 30 October molted by 27
November and to an adult on 24 December.

(8) Drassyllus notonus Chamberlin.— Three females and one male were collected
from May to June. One of the females was found under a stone, another under a board
and the third indoors on a garage floor. The male was collected in June by sweeping
vegetation.

(9) Drassyllus orgilus ChamberHn.- Nine female specimens were collected in Wichita
and Clay counties during February and March. Spiders were found under rocks and
boards, in grass and other vegetation, and indoors.

(10) Drassyllus texamans ChamberHn.— Four specimens were taken from collection
sites in Wichita County on the sandy terraces along the Red River. This species seems to
prefer loose, sandy soil. A female from Hardeman County was found on the silty floor of
a cave (Platnick and Shadab 1982). A penultimate male collected on 20 May molted after
five days.

(11) Drassyllus species.— A single unidentified male Drassyllus was found under a
stone in a pasture in Wichita County. Return trips to the same locality failed to yield
additional specimens. This may be a species of which only the female has been described.

(12) Gnaphosa altudona Chamberlin.— Five of the six specimens collected were
males found under stones. All the spiders were found during June and July in rough,
eroded country dominated by stones and bare ground. A female collected on 1 5 July had
an egg sac which contained 27 first instar spiderlings. Two immature males reared in the

82

THE JOURNAL OF ARACHNOLOGY

laboratory reached maturity on 2 and 16 August. The present record extends the range
approximately 400 miles north from Brewster, Presidio, and San Patricio counties of
Texas (Platnick and Shadab 1975a).

(13) Gnaphosa clara (Keyserling).— A single female collected on 10 June was found
with two egg sacs. The first sac had already been vacated and contained only exuviae
where as the second sac held 59 eggs.

(14) Gnaphosa fontinalis Keyserling.— Two specimens were collected in Wichita
County on 6 May. No specific habitat data were recorded.

(15) Gnaphosa sericata (L. Koch).— The four specimens collected were found in
areas with sandy soil and little available ground cover, a niche that other gnaphosids
seemed to avoid. Spiders were taken from under small bits of wood, cow manure, and in a
pitfall trap. Of two immature females caught on 20 May, one molted to maturity on 8
June and the second molted on 13 June (penultimate) and 24 July.

(16) Haplodrassus chamberlini Platnick and Shadab.— Twenty individuals were
found from March to May, eighteen under stones in grassy pastures and rough, eroded
areas. Six penultimate spiders (four females and two males) collected on 16 March all
molted to adults by 27 March. This is a new state record for this species: the closest
previous records are Roosevelt County, New Mexico, and Texas County, Oklahoma
(Platnick and Shadab 1975b). Its presence in north-central Texas extends the range 270
miles to the southeast.

(17) Haplodrassus signifier (C. L. Koch).— Forty-one specimens were found from
March to June, most of them under stones. Three females were collected on 19 May, each
protecting one large and one small egg sac. The large sacs were drab and dirty while the
small sacs were obviously newer because they were whiter. Each female, when exposed,
tried to move the large egg sac to safety by carrying it in her chelicerae. The large sac of
one female contained 232 second instar spiders and the small sac had 50 first instar
spiderlings. The egg sacs of the second female held 94 first instar spiders and 24 eggs. The
sacs of the third female contained 160 second instar spider and 26 first instar spiderlings.
Three penultimate spiders molted on 21 February (female), 21 March (male), and 24
March (female). A late instar spider, probably H. signifer, was parasitized by the acrocerid
fly Ogcodes eugonatus Loew. The larva ruptured the abdomen of the spider, pupated on
18 April, and the adult female emerged on 29 April.

(18) Herpyllus bubulcus ChamberUn.- A single female was collected on 3 March
from a rock pile in Hardeman County by an araneology student.

(19) Herpyllus ecclesiasticus Hentz.— This widespread species is opportunistic as to
habitat. Fourteen specimens collected from February to August were found indoors,
under tree bark, on trees, and in grass. A penultimate female found 11 July molted on 3
August.

(20) Herpyllus hesperolus Chamberlin.— Two females were collected on 20 March
along a rocky, eroded hiUside. This discovery extends the range approximately 400 miles
east of a line from Pecos County, Texas, to the Big Meddy Valley in Saskatchewan
(Platnick and Shadab 1977).

(21) Micaria.-Micaria is present in the area. This genus is currently under revision and
is not dealt with within this paper.

(22) Nodocion floridanus (Banks).— Only four specimens have been recorded from
this area. The single specimen examined was a male collected from a wasp nest in Wichita
County. Records indicate another male from Wichita County was found under tree bark
and two males were found in a tamarisk bower in Baylor County (Platnick and Shadab
1980a).

ZOLNEROWICH AND HORNER-GNAPHOSIDAE OF NORTH-CENTRAL TEXAS

83

(23) Nodocion rufithoracicus Worley. -Eight spiders (4 males and 4 females) were
found from March to August in Wichita County under rocks in broken eroded country. A
penultimate male collected on 18 March molted on 23 March. This material extends the
range 320 miles east of a line extending from Eddy County, New Mexico, to Divide
County, North Dakota (Platnick and Shadab 1980a).

(24) Rachodrassus captiosus (Gertsch and Davis).— A single male was found in June
under a piece of railroad tie at the base of a high bluff. Previous specimens are known
only from San Luis Potosi, Mexico, and Cameron and San Patricio counties of the south
Texas coast. This is a range extension of over 400 miles to the north.

(25) Sergiolus angustus (Banks).— A single female was collected along a rocky hillside
in March. The only other Texas record for the species is Kleberg County on the coast of
south Texas. This is a range extension of approximately 100 miles east of a line from
northern Colorado to Kleberg County, Texas (Platnick and Shadab 1981).

(26) Sergiolus bicolor Banks.— Two males collected in June were found, one under a
stone and the other indoors.

(27) Sergiolus lowelli Chamberlin and Woodbury.— This form is found throughout the
area and is variable in habitat. Nine specimens were collected from March to October in
grass, indoors, from a bird’s nest, a tamarisk bower, and a tarpaulin.

(28) Sergiolus Stella Chamberlin.— A female was collected on 19 May along a stony,
eroded hillside. The spider molted on 2 June and 28 June. The record extends the range
100 miles west northwestward from Denton County, Texas (Platnick and Shadab
1981).

(29) Zelotes aiken Platnick and Shadab.— Four specimens were taken during March
and April from Clay, Montague, and Wichita counties. The spiders were found along a
lake shore, under rocks, and in Bermuda grass. A penultimate male caught 31 March
molted that day.

(30) Zelotes anglo Gertsch and Riechert.— A single male was found in Wichita County
and Platnick and Shadab (1983) record a male from Wilbarger County. The Wichita
County specimen was taken in September from a pitfall trap set in an open pasture.

(31) Zelotes gertschi Platnick and Shadab.— This was the most common Zelotes
detected. Forty specimens were collected from Archer, Clay, and Wichita counties,
mostly between March and July under stones, boards, railroad ties, and cardboard.
Eighteen laboratory reared spiders molted one to four times between 24 March and 4
August. Six were males and all had matured by 13 June, with most maturing in April. The
maturity dates of the females were much later: two matured in April, six in May, two in
June, one in July and one in August.

(32) Zelotes pseustes Chamberlin.— Two males (collected 10 January and 28 Febru-
ary) and two females (collected 24 April and 16 August) were found under a rock and a
board, in dead leaves, and in a pitfall trap set in sand.

(33) Zelotes tuobus Chamberlin.— A single male was taken on 28 April, 1975, from
under a rock in Wichita County.

(34) Addendum.— Dr. Norman Platnick has recently (personal communication March
30, 1984) identified two European species that were collected from north-central Texas.
Two males of Urozelotes rusticus (Koch L.) were collected in homes in Wichita County in
June 1976 and May 1977. A single female of Trachyzelotes lyonneti (Audouin) was
collected in May 1975 from the ground in Baylor Co.

In addition to the confirmed gnaphosid fauna, based on records from surrounding
counties, range maps, and habitat data, an additional two genera and nine species are

84

THE JOURNAL OF ARACHNOLOGY

believed to occur in north-central Texas. Callilepis imbecilla (Keyserling) is present in
Comanche County, Oklahoma, and has been found under leaf Utter and boards (Platnick
1975) and under stones in pastures and dry, sandy areas (Kaston 1981). Cesonia bilineata
(Hentz) ranges from New Mexico to the Atlantic coast and has been collected in open and
tall grass prairies, mesquite woods, and Bermuda grass (Platnick and Shadab 1980b).
Drassodes auriculoides Barrows is known from Comanche County, Oklahoma. Platnick
and Shadab (1976a) list leaf litter and a pasture as habitats and Kaston (1981) reports
this species can be found under stones and logs. Rachodrassus exlineae Platnick and
Shadab (female) is known from Comanche County, Oklahoma. Sergiolus montanus
(Emerton) is found across the contiguous states and has been collected from under rocks,
bark, dry cow dung, and indoors (Platnick and Shadab 1981). Sosticus insularis (Banks)
is known from Comanche County, Oklahoma, and Dallas County, Texas, and has been
found under bark and indoors (Platnick and Shadab 1976b). Synaphosus syntheticus
(Chamberlin) has been recorded from Dallas County, Texas; it has been found indoors
and in salt cedar, cottonwood, and mesquite litter (Platnick and Shadab 1980a). Zelotes
hentzi Barrows ranges across the United States except for the Southwest and has been
collected from under rocks, boards, logs, and in cottonwoods, cotton fields, pecan
groves, prairies, and meadows (Platnick and Shadab 1983). Zelotes lasalanus Chamberlin
is present in Tarrant County, Texas, and has been found under debris, dung, and stones,
and in grass, mesquite, meadows, and prairies (Platnick and Shadab 1983).

Of the 32 species that have been reported from north-central Texas, 16 are represented
by four or less specimens taken in 14 months of intensive collecting. Further collecting
may be expected to produce additional species records and natural history data.

ACKNOWLEDGMENTS

We thank Dr. N. I. Platnick (Amer. Mus. Nat. Hist., NY) for identifying and confirm-
ing problematic specimens, and Dr. E. I. Schlinger (Univ. Calif, Berkeley) for identifying
the acrocerid fly. Appreciation is expressed to the many landowners who permitted
collecting on their property. A special thanks is extended to the many araneology stu-
dents who contributed gnaphosid specimens to the Midwestern State University Collec-
tion. This paper is part of a thesis submitted by the senior author to the faculty Depart-
ment of Biology, Midwestern State University, in partial fulfillment of the requirements
for the degree of Master of Science.

LITERATURE CITED

Blair, W. F. 1950. The biotic provinces of Texas. Texas J. Sci., 2(1):93-117.

Carpenter, R. M. 1972. The jumping spiders (Salticidae) of Wichita County, Texas. Southwest. Nat.,
17(2):161-168.

Cokendolpher, J. C., N. V. Homer and D. T. Jennings. 1979. Crab spiders of North Central Texas
(Araneae: Philodromidae and Thomisidae). J. Kansas Entomol. Soc., 52(4):723-734.

Gertsch, W. J. 1979. American spiders. Van Nostrand Reinhold Co., Dallas, Texas. 274 pp.

Gould, F. W. 1975. The grasses of Texas. Texas A & M University Press. College Station, Texas. 653

pp.

Kaston, B. J. 1981. Spiders of Connecticut. Bull. State Geol, Nat. Hist. Survey, 70:1-1020.

Matelski, E. A. 1982. The bionomics of Peckhamia in North Central Texas. Unpubl. M.S. Thesis,
Midwestern State University. 31 pp.

Platnick, N. I. 1975. A revision of the Holarctic spider genus Callilepis (Araneae, Gnaphosidae). Amer.
Mus. Nov., 2573:1-32.

ZOLNEROWICH AND HORNER-GNAPHOSIDAE OF NORTH-CENTRAL TEXAS

85

Platnick, N, I. and M. U. Shadab, 1975a. A revision of the spider genus Gnaphosa (Araneae, Gnaphosi-
dae) in America. Bull. Amer. Mus. Nat. Hist., 155(l):l-66.

Platnick, N. 1. and M. U. Shadab. 1975b. A revision of the spider genera Haplodrassus and Orodrassus
(Araneae, Gnaphosidae) in North America. Amer. Mus. Nov., 2583:1-40.

Platnick, N. I. and M. U. Shadab. 1976a. A revision of the spider genera Drassodes and Tivodrassus
(Araneae, Gnaphosidae) in North America. Amer. Mus. Nov., 2593:1-29.

Platnick, N. I. and M. U. Shadab. 1976b. A revision of the spider genera Rachodrassus, Sosticus, and
Scopodes (Araneae, Gnaphosidae) in North America. Amer. Mus. Nov., 2594:1-33.

Platnick, N. I. and M. U. Shadab. 1977. A revision of the spider genera Herpyllus and Scotophaeus
(Araneae, Gnaphosidae) in North America. Bull. Amer. Mus. Nat. Hist., 159(1):144.

Platnick, N. I. and M. U. Shadab. 1980a. A revision of the North American spider genem Nodocion,
Litophyllus, dind Synaphosus (Araneae, Gnaphosidae). Amer. Mus. Nov., 2691:1-26.

Platnick, N. I. and M. U. Shadab. 1980b. A revision of the spider genns Cesonia (Araneae, Gnaphosi-
dae). Bull. Amer. Mus. Nat. Hist., 165(4):337-385.

Platnick, N. I. and M. U. Shadab. 1981. A revision of the spider genus Sergiolus (Araneae, Gnaphosi-
dae). Amer. Mus. Nov., 2717:1-41.

Platnick, N. I. and M. U. Shadab. 1982. A revision of the American spiders of the genus Drassyllus
(Araneae, Gnaphosidae). BuU. Amer. Mus. Nat. Hist., 173(l):l-97.

Platnick, N. I. and M. U. Shadab. 1983. A revision of the American spiders of the genus Zelotes
(Araneae, Gnaphosidae). Bull. Amer. Mus. Nat. Hist., 174(2):99-191.

Salmon, J. T. and N. V. Horner. 1977. Aerial dispersion of spiders in North Central Texas. J. Arach-
nol., 5(2):153-157.

Zaltsberg, H. S. 1977. Aerial dispersal of spiders as related to atmospheric parameters in North Central
Texas. Unpubl. M.S. Thesis, Midwestern State University. 20 pp.

Manuscript received March 1 984, revised June 1 984.

■K V

'S

' 'V'U

'■■"■ ■'■' S'!'' ,lS''*S't"

iM >■ . • r, .

. *'u hsit

. pktiJ^-iK r-r^

^ ‘'fT'^^if i/ . ''.V'*’

- V ., > >:■ * ,: Mf‘' y -’■%;»*; ,/WK^ M ' . '■ 1,,^ .-j-

, h •'-•^ ^ 'piiii>'4L<ji iniiiig'K j ■.■>!’. .^’ ''»v'

■■ ‘

' i ^ r •

’ , ? ) ' i •

iirw ■i;t>~

I

‘ •-.

to.’ A

i HnV

‘J' ) r.i^l.'lf':

'l * . ■ t ', '" 1 .- *• •, * ;. I ■

. . :-.*' , '^iU

■* W

' '•’.^cwti'' „

■ .■» ' ■• .*. *

• 1 >

- -li

• ’ ■■ >( A-V

* .^1

V '

I. '

'^■' 1'-- , - -.

.r

' '■

r. -

. *.

“;■ S' «|: --■■■ •

.V ■ " "> ■• ■

ia...

J ' V i

. . r j

•-■3^

I f

' v»r.

r I .* ■ ;* i I r

•T I

Bennett, R. G. 1985. The natural history and taxonomy of Cicurina bryantae Exline (Araneae, Agelen-
idae). J. Arachnol., 13:87-96.

THE NATURAL HISTORY AND TAXONOMY OF
CICURINA BR YANTAE EXLINE (ARANEAE, AGELENIDAE)

R. G. Bennett

Department of Biology
Western Carolina University
Cullowhee, North Carolina 28723

ABSTRACT

Since its description in 1936 Cicurina bryantae Exline has been rarely collected. The recent
discovery of the microhabitat preference of this spider has allowed the observation and collection of
substantial numbers of specimens. The following account is compiled from these data. The natural
history of the species is discussed (including the interesting tubular retreat constructs inhabited by
immatures and adults). Both sexes are described and figured (the male for the first time).

INTRODUCTION

In 1936 Harriet Exline described a new species in the genus Cicurina Menge, 1869 on
the basis of female specimens collected at Newfound Gap on the Tennessee-North Caro-
lina border in 1930 by Nathan Banks. She named the species bryantae in honor of Eliza-
beth Bangs Bryant who was at that time working at the Museum of Comparative Zoology
where the specimens were deposited. Since that date, Cicurina bryantae has rarely been
collected, its behavior and life history have not been recorded, and the male has not been
described.

Chamberlin and Ivie (1940) redescribed the female and mentioned two more females
captured by Ivie in 1933 in East Tennessee. Apparently no other specimens were col-
lected until 1972 when J. O. Howell found an undescribed Cicurina male in a pitfall
trap set in Union County, Georgia. Vincent Roth tentatively identified it as the male of
C bryantae but no effort was made to describe it at that time as no other males were
found (Howell 1972, pers. comm.).

During the late fall of 1982 I began finding populations of C bryantae living within a
fairly specific microhabitat in and around Jackson County, North Carolina. Erom then
until January 1984 substantial numbers of males, females and immatures were observed
and collected. The resultant data are presented in this paper; both sexes are described and
figured (the male for the first time) and a general account is given of the life history.
Specimens collected during this study have been deposited in the American Museum of
Natural History (AMNH), Dr. N. I. Platnick; the Museum of Comparative Zoology (MCZ),
Dr. H. W. Levi; the Canadian National Collection of Insects, Arachnids and Nematodes
(CNC), Dr. C. D. Dondale; and the Florida State Collection of Arthropods (FSCA), Dr. G.
B. Edwards; and in my private collection (RGB).

88

THE JOURNAL OF ARACHNOLOGY

Since my acquaintance with some agelenid taxa is limited, I have here conservatively
retained the classical placement of Cicurina within the Agelenidae. However, I do believe
that the Agelenidae as it is usually envisioned is probably a polyphyletic assemblage and
tentatively concur with Lehtinen’s transfer of Cicurina to the Dictynidae: Cicurininae
(1967 :222-223) as catalogued recently by Brignoli (1983 :5 18).

NATURAL HISTORY

Specimens of C bryantae have been collected at elevations of 300-1350 meters (1000
to 4400 ft.) above sea level in the Blue Ridge and Great Smoky Mountain regions of the
Southern Appalachians (Fig. 6). The true southern and south lateral boundaries of
its range probably correspond closely to the southerly collection locales. The more
northerly Ihuits of its range are not known because of a lack of collection data from
northwestern North Carolina and adjacent areas of Tennessee. It is possible that this
spider ranges well into the Appalachians of Virginia.

Populations are most common in mixed deciduous forests (dominated by oak, hickory
and tulip poplar) and are encountered less frequently in deciduous woods mixed with
conifers such as white pine and hemlock. Within a suitable wooded habitat C. bryantae is
usually found within retreats constructed on the undersurface of rotting wood well
settled in the leaf litter of the forest floor. Retreats are never found on substrates other
than rotting wood, Gryllacridid crickets, cryptocercid roaches and large numbers of
collembolans as well as spiders of the genera Antrodiaetus, Calymmaria, Coras, Wadotes,
Liocranoides and other species of Cicurina commonly occupy the same microhabitat as C.
brymntae. Unidentified pseudoscorpions have been collected from otherwise empty
retreats.

Eugene Simon (1898:265) observed that Cicurina cicurea (Fabricius) “file une toile
tres legere et horizontale, sous les pierres ou au milieu des mousses. . .” (Spins a very
slight and horizontal web, beneath stones or within mosses. . .). Exline (1936) reported
that the delicate webs of Cicurina species can be found in rotting logs or under rocks or
boards and that no retreat is built. Although the observations of Simon and Exline may
apply to most species of Cicurina, C. bryantae is an exception; it builds no fine horizon-
tal sheet web but does construct an interesting retreat (Figs. 7, 8).

The retreats are built in natural cavities on suitable wood surfaces or wherever there is
sufficient space between the wood and the ground. The retreat is a tube with twin, often
turret-like openings at either end. The tube consists of very fine and delicate silk coated
on the outside with organic debris derived from the rotting substrate or the ground
beneath. The wooden substrate forms the ceiling of the retreat and is covered only lightly
with silk. The openings normally point downwards and are never closed by the spider as
are the similar appearing (although normally larger) entrances to the subterranean bur-
rows of Antrodiaetus spiders (see Coyle 1971). The turrets are often connected to the
surrounding substrate with short silk threads which may possibly function as trip wires to
aid in the detection of prey.

Males, females and immatures build retreats similar in all respects except size, although
male retreats occasionally exhibit a slight but pronounced lateral widening between the
twin openings. Adults of both sexes occupy retreats measuring 12 to 15 mm between the
inner edges of the openings (Fig. 8). Immatures build retreats (Fig. 7) ranging from about
2 mm in length (as measured above) to adult size. It is not known whether immatures

BENNETT-C/Ct//?/7V/4 BR YANTAE NATURAL HISTORY AND TAXONOMY

89

expand their retreats as they grow or simply move out and construct new larger retreats.
A number of adult and immature-sized retreats can often be found in close proximity.

In an effort to observe aspects of their behavior I placed five females in an observation
chamber containing a piece of rotting wood. Five others were placed in a similar chamber
with a clean, unrotted piece of wood. The atmosphere was kept humid and there were
ample suitable sites for retreat construction in both chambers. Overnight, spiders in the
first group all built retreats on the undersurface or sides of the rotting wood. These
retreats seemed quite typical although they lacked turrets and were rather sparsely coated
with debris. Over a period of weeks these spiders successfully maintained themselves on a
diet of cricket nymphs. The spiders of the second group built no retreats or obvious
webbing of any kind and did not feed on the cricket nymphs which were introduced
into the chamber. It seems evident from this that the presence of a suitable substrate
and/or debris particles is a necessary prerequisite to normal retreat construction and prey
capture.

Cicurina bryantae retreats are remarkably similar to those built by several species of
Japanese cavernicolous cybaeinine agelenids (Komatsu 1961). Komatsu describes the
cybaeinine retreats as being pipe-like in construction and usually coated with sand grains.
The openings lie at the extreme ends of the tubes rather than at right angles to the tube
axis, as do the entrances to the retreats of C. bryantae. From each opening a pair of long
silk tripwires extends to the surrounding substrate. Some retreats may have as many as
three openings. The retreats are encountered in open locations within caves but insofar as
these areas are dark and of relatively constant humidity, they resemble the dwelling sites
of C bryantae.

Immatures and adult females of C bryantae can be collected year round, but adult
males have been collected only from late August to late January suggesting that mating
occurs in the fall. Additional support for this idea is the observation that females col-
lected during the fall, winter and early spring often have one or both passages leading
from the copulatory opening blocked with what is apparently a copulatory plug of
unknown composition.

Females begin producing egg cases in late spring (May). These flattened lenticular
structures are composed of two silken half shells and are attached to the inner surface of
the ventral-most wall of the retreat. Cases typically are filled with from three to ten eggs,
each roughly spherical and approximately 0.9 mm in diameter. As the summer progresses
additional egg cases are constructed and attached to any already present. Normally each
succeeding egg case will have fewer eggs than its predecessor. One retreat, examined in
early October 1983, contained nine egg cases in a stratified lump coated with substrate
debris. Eggs in five of these had already hatched and the spiderlings had dispersed. Two
contained small numbers of spider nymphs and two were parasitized. No eggs were
present in any of these cases. Out of a total of 43 egg cases examined during 1983, four
were parasitized. In each of these four cases a single ichneumon wasp pupa {Gelis sp.) was
found.

Dispersal of the immatures from the maternal retreat commences during the summer.
Since very small retreats are commonly found in late summer clustered near the retreats
of adults, it is likely that most spiderHngs simply leave the mother’s retreat and build
their own retreats at the first suitable location they come upon. Often these immature
constructs will be built upon the surface of the presumed parental retreat (which may or
may not be occupied).

90

THE JOURNAL OF ARACHNOLOGY

Observations made over 16 months of study of C bryantae populations suggest that
there may be a two year or longer life cycle involved. Maturation may occur when the
spiders are one year old (teneral adults have been collected regularly and only in the late
summer and early fall) but there is a strong possibility that the immatures take more than
one year to reach maturity as a wide range of sizes of immature retreats may be observed
at any one point in time.

That a female may pass more than one winter as an adult is intimated by the collection
in the early spring of 1983 of a female from a retreat containing three empty egg cases.
This capture was made at a high altitude (1350 m) and sometime before any new egg
cases for that year had been observed. The presence of a plug in only one of the two
copulatory tubes of some females collected with egg cases suggests that if females are
capable of surviving to a second breeding season, they may be able to mate again success-
fully.

TAXONOMY

Methods.— All measurements and counts were made with a Leitz stereomicroscope
fitted with lOx eyepieces and a micrometer reticle. Measurements are accurate to 0.025
mm. Abbreviations are as follows; CL, carapace length; CW, carapace width; SL, sternum
length; SW, sternum width; ALE, anterior lateral eyes; AME, anterior median eyes;PLE,
posterior lateral eyes; PME, posterior median eyes; MOQ, median ocular quadrangle.
Statistics are presented as follows: sample range (mean ± standard deviation). All mea-
surements are in millimeters.

Macroseta counts for all legs are presented either literally or through the use of a
standard macroseta formula where for a particular face of a leg segment a series of three
numbers separated by hyphens represents the number of macrosetae on the proximal,
medial, and distal portions respectively.

A 10x10 squared grid reticle and 16x Zeiss wide field eyepieces were used with the
Leitz stereomicroscope to make drawings. Male palps were examined and drawn whole in
80% ethanol. Epigyna were treated likewise and also dissected out of the abdomen and
cleared in clove oil or hot 10% KOH for study and drawing internal sclerotized characters.

Cicurina bryantae Exline
Figs. 1-11

Cicurina bryantae Exline 1936:13, figs. 4, 14 [female holotype from Newfound Gap on the Tennes-
see (Sevier County )-North Carolina (Swain County) border, 9 July 1933 (W. Ivie), in MCZ, exam-
ined] ; Chamberlin and Ivie 1940:25, fig. 12; Bonnet 1956:1087.

Diagnosis.— Specimens of C. bryantae can be separated from the morphologically
similar species Gcurina pallida Keyserling and Cicurina breviaria Bishop and Crosby, and
from all other species of Cicurina, by the following series of characters: abdomen and
cephalothorax unmarked; copulatory tubes smoothly curving to slightly sinuous and
leading to undivided spermathecae (Figs. 3-5); embolus short and thick, terminating
bluntly on distal end of short conductor; conductor with small retrolateral accessory
projection originating from base of hook (Figs. 1, 2); distal retrolateral tibial apophysis
short, shallowly bifurcate distally and slightly cupped dorsally.

BENNETT-C/Cf//?/7V^ BRYANTAE NATURAL HISTORY AND TAXONOMY

91

Description.— Small to medium size spiders. Sexes similar morphologically except for
sexual characteristics; males slightly larger than females. Cephalothorax and legs uniform
light reddish brown (darkening with age), with dark longitudinal streak demarking posi-
tion of thoracic groove. Carapace glabrous except for a few lines of sparse setae radiating
from thoracic groove primarily towards and around eye region. Abdomen light colored,
unmarked and lightly clothed with short setae. Chelicerae robust, each with single strong
macroseta dorsally on prolateral face (Fig. 9). Retromargin of cheliceral fang furrow
with 3 or 4 teeth and 2 to 5 denticles; promargin with 3 or occasionally 2 teeth. Sternum
approximately as wide as long, roughly heart-shaped, with short projection extending
between hind coxae. Colulus indicated by 2 to 4 setae. Spinnerets closely grouped, with

Figs. 1-5. -Genitalia of Gcurina bryantae Exline; 1, left palpal tibia and tarsus of male, ventral
view, specimen from Cane Creek, Cullowhee; 2, same, retrolateral view; 3, epigynum, ventral view,
specimen from Cane Creek, Cullowhee; 4, same cleared to show internal sclerotization; 5, variation of
epigynum, ventral view, specimen from Cataloochee Valley, Abbreviations: C, conductor; E, embolus;
TA, tibial apophysis; CO, copulatory opening; CT, copulatory tube; S, spermatheca; FT, fertilization
tube.

92

THE JOURNAL OF ARACHNOLOGY

median pair smallest; anterior pair close together and relatively stout; posterior pair
longer, thinner and more widely spaced than anterior spinnerets. AME alone dark, others
light (Fig. 9). AME smallest, slightly smaller than PME. PEE largest but only slightly
larger than ALE. MOQ wider behind and slightly higher than wide. Height of MOQ about
1.5 times height of clypeus. Legs with numerous macrosetae. All femora with 3 macro-
setae dorsally along midline, with a pair of lateral macrosetae bracketing the terminal
dorsal macroseta except for femur I which normally bears a retrolateral pair of offset
macrosetae distally. Ventrally on all femora a double row of relatively small thin macro-
setae, ventral retrolateral row heavier than ventral prolateral row, and macrosetae decreas-
ing in size to status of setae from femur I to femur IV. Macrosetation of tibia I and
metatarsus I complicated and somewhat variable (Figs. 10-1 1). Tibia I normally with row
of 5 macrosetae originating on ventral prolateral surface and terminating prolaterally; two
other macrosetae along ventral retrolateral surface and a pair of small macrosetae distally
on ventral surface; one additional macroseta prolaterally above second and third ventral
prolateral macrosetae; dorsally one small macroseta near proximal end. Ventral surface of
metatarsus I with similar macroseta pattern but three macrosetae each on prolateral,
ventral prolateral and ventral retrolateral surfaces. Tibia II 2-2-2 ventrally, with distal pair
of macrosetae and ventral prolaterals reduced; prolaterally two strong macrosetae; two
weak macrosetae dorsally. Metatarsus II 2-2-3 ventrally and 3 macrosetae along prolateral
surface. Tibia III 2-2-2 or 1-2-2 ventrally, and two macrosetae each on prolateral, retro-
lateral and dorsal surfaces. Metatarsus III 2-2-3 ventrally, one macroseta each on pro-
lateral and retrolateral faces and 2-2-2 dorsally with the median pair staggered. Tibia IV
normally 1-2-2 ventrally, otherwise similar to tibia III. Metatarsus IV 2-2-1 ventrally, with
2 macrosetae on retrolateral surface and 2-2-2 dorsally. Metatarsal and tarsal trichobo-
thria patterns typical, arranged in single longitudinal row dorsally and increasing in
length distally along both segments on each leg. Tarsi and metatarsi each with 4 or 5
trichobothria; metatarsi II and III often with 4 and metatarsi I and IV with 5. Legs
arranged in order of decreasing length: IV-I-II-III.

Fig. 6. -Known geographic distribution of C. bryantae.

BENNETT-C/C(//?/7V^ BR YANTAE NATURAL HISTORY AND TAXONOMY

93

Figs. 7-8. -Retreats of C. bryantae, specimens from Cullowhee: 7, retreats of immature specimens,
both approximately 5 mm between inner edges of entrances. Lower entrance of right hand burrow
blurred due to momentary presence of occupant in entrance during time exposure; 8, retreat, 12 mm
between inner edges of entrances, with mature female at one entrance.

94

THE JOURNAL OF ARACHNOLOGY

Fig. 9 -Frontal view of eyes and cheliceral bases of C bryantae.

Figs. 10-11. -Left leg I of male, 10, retrolateral aspect; 11, prolateral aspect.

Male: Figs. 1, 2. 26 specimens measured. CL 1.78-2.45 (2.09 ± 0.16), CW 1.38-1.73
(1.57 ± 0.11), SL 0.88-1.1 (0.99 ± 0.06), SW 0.85-1 .08 (0.96 ± 0.06). Palpus very simple
in comparison to most Cicurina males. Cymbium short and stubby. Terminal conductor
hook short and smoothly curved ventrally. Other characters as in diagnosis.

Female: Figs. 3-5. 30 specimens measured. CL 1.73-2.35 (2.06 ±0.17), CW 1.25-1.63
(1.45 ± 0.12), SL 0.8-1.08 (0.97 ± 0.07), SW 0.83-1.03 (0.92 ± 0.07). Epigynum very
simple. Copulatory opening usually appearing single and dividing into two funnel shaped
bursae just inside opening (Figs. 3, 4). In some specimens, particularly those from the
Cataloochee Valley, Haywood County, North Carolina, the copulatory opening itself
appears to be divided (Fig. 5). This is because the internal division into the pair of bursae
is more ventrally placed within the mouth of the copulatory opening. (No concommitant
variation has been detected in males from the Cataloochee Valley.) A single, short sclero-
tized fertilization tube exits each spermatheca dorsocaudaUy and proceeds for a short
distance anteriorly and dorsally in a tight arch before sclerotization becomes weak and
tubes become invisible. Copulatory tubes thick-walled with a thin canal visible within.
Other characters as in diagnosis.

BENNETT-C/Ct//^r7V/4 BR YANTAE NATURAL HISTORY AND TAXONOMY

95

Specimens examined.- (Collected by author unless otherwise noted.) U.S.A.: GEORGIA;

County, W of Chatooga R. on Highway 708, 11 January 1984, 1 male, 1 female (FSCA), intersec. of
St. Rds. 76 and 197, 12 miles W of Clayton (2160 ft.), 12 January 1984, 3 males, 8 females, 4 imma-
tures (AMNH); Union County, Vogel St. Pk. (2750 ft.), 17 January 1972 (J. O. Howell), 1 male
(JOH), south end of Vogel St. Pk. (2300 ft.), 12 January 1984, 4 immatures (FSCA); White County,
Unicoi St. Pk., Ana Ruby Falls (1800 ft.), 12 January 1984, 2 females (FSCA). NORTH CAROLINA:
Graham County, Stratton Meadows (4400 ft.), 4 April 1983, 5 females, 5 immatures OACZ)', Hay wood
County, Pinnacle Ridge, Blue Ridge Parkway (4400 ft.), 31 May 1983, 1 female (AMNH), Great
Smoky Mtn. Nat. Pk., Cataloochee Valley (2700-3200 ft.), 7-9 October 1983, Caldwell Fork, 1 male
(MCZ), near Caldwell House, 2 males, 1 female (AMNH), above Rough Fork, 2 males, 7 females
(FSCA), above Beech Grove school, 4 males, 6 females (RGB), Hannah Cabin, 2 males, 4 females
(MCZ), near Palmer Cemetery, 1 male, 4 females (AMNH); County, N slope Little Panther
Knob, Long Branch Rd., Cullowhee (2600 ft), 5 November 1982, 1 male, 9 females, 16 immatures
(AMNH), 17 May 1983, 3 females, 3 immatures (AMNH), Cane Creek Valley, Cullowhee (2200-3200
ft), 7 November 1982, 2 males, 11 females, 9 immatures (MCZ), 31 December 1982, 1 female (RGB),
15 March 1983, 3 females (RGB), 16 June 1983, 5 females, 3 immatures (RGB), 25 August 1983,4
females, 1 immature male, 1 immature (AMNH), 5 September 1983, 1 female (MCZ), 13 September
1983, 4 males, 10 females, 2 immatures (MCZ), 30 September 1983, 2 females (FSCA), 19 October
1983, 5 males, 8 females (RGB), Owens Gap, intersec. of Co. Rd. 1763 and Highway 281 (3600 ft.),
22 November 1982, 1 male, 3 females (FSCA), Tuckaseegee River at 40 mile bend, Cullowhee (2050
ft.), 30 September 1983, 6 males, 10 females, 4 immatures (MCZ), Fall Cliff, WCU Preserve, Cul-
lowhee Mtn. Rd., 4 July 1983, 4 females, 1 immature male, 6 immatures (MCZ), Mull Creek, Cullow-
hee (3000 ft.), 27 August 1983, 1 female (MCZ), 5 September 1983, 1 male (MCZ), Webster, Morgan
Farm, 27 April 1983, 3 females, 1 immature (FSCA); Swain County, Newfound Gap near Cherokee, 3
August 1930 (N. Banks), 1 female (MCZ). SOUTH CAROLINA: Oconee County, N side of Highway
76 at Chatooga River (1240 ft.), 11 January 1984,4 immatures (FSCA). TENNESSEE: Po/A: Cowwfy,
Goforth Creek at Highway 64 (1000 ft.), 1 female (AMNH); Sevier County, Laurel Falls Trail, Little
River Rd., Great Smoky Mtn. Nat. Pk., 8 May 1983, 1 female, 1 immature (MCZ).

Additional Records.— TENNESSEE: Unicoi County, Erwin, 8 July 1933, (W. Ivie), 1 female;
County, Little Pigeon Creek, Great Smoky Mtns., 9 July 1933, (W. Ivie), 1 female.

ACKNOWLEDGMENTS

I wish to express my appreciation to Drs. F. A. Coyle, C. D. Dondale and W. A.
Shear for encouraging me to publish this account. I am greatly indebted to Fred Coyle for
his interest, advice and criticism during the preparation of the manuscript and for direct-
ing me to interesting papers. Dr. H. W. Levi kindly loaned me Exline’s holotype and Dr. J.
0. Howell sent me his male for study and comparison with my specimens. Dr. J. Barron is
warmly thanked for the effort he put into identifying ichneumon parasites. The privilege
of collecting in the Great Smoky Mountains National Park was granted through a permit
from the National Park Service. My wife Jennifer helped considerably by providing moral
support and typing the manuscript. Finally, I am forever grateful to my parents for their
support, both moral and financial, the latter being of great significance in these pecuniar-
ily depauperate times.

LITERATURE CITED

Bonnet, P. 1956. Bibliographia Araneorum. Toulouse. Vol. 2(2):919-1925.

BrignoU, P. M. 1983. A catalogue of the Araneae described between 1940 and 1981. Manchester Univ.
Press, 755 pp.

Chamberlin, R. V. and W. Ivie. 1940. Agelenid spiders of the genus Cicurina, Bull. Univ. Utah, 30
(13):108 pp.

96

THE JOURNAL OF ARACHNOLOGY

Coyle, F. A. 1971. Systematics and natural history of the mygalomorph spider genus Antrodiaetus
and related genera (Araneae, Antrodiaetidae). Bull. Mus. Comp. Zool., 141(6):269-402.

Exline, H. 1936. Nearctic spiders of the genus Cicurina Menge. Amer. Mus. Novit., No. 850, 25 pp.
Howell, J. O. 1975. Discovery of the male of Cicurina bryantae (Araneae, Agelenidae). J. Georgia
Entomol. Soc., 10(1):32“33.

Komatsu, T. 1961. Cave spiders of Japan, their taxonomy, chorology, and ecology. Arachnol. Soc.
East Asia, Osaka, 91 pp.

Lehtinen, P. T. 1967. Classification of the cribellate spiders and some allied families, with notes on the
evolution of the suborder Araneomorpha. Ann. Zool. Fenn., 4:199-468.

Simon, E. 1898. Histoire naturelle des Araignees. Paris. Vol. 2(2):193-380.

Manuscript received June 1 984, revised July 1 984.

Lowrie, D. C. 1985. Preliminary survey of wandering spiders of a mixed coniferous forest. J. Arach-
nol., 13:97410.

PRELIMINARY SURVEY OF WANDERING SPIDERS OF A
MIXED CONIFEROUS FOREST

Donald C. Lowrie^

Professor Emeritus, California State University, Los Angeles
5151 State University Drive
Los Angeles, California 90032

ABSTRACT

This study of nomadic riparian ground-surface inhabiting spiders was made at the Los Alamos National
Environmental Park, Mew Mexico. Spiders constituted about 10% of this mixed coniferous forest
community. They were widely distributed in all samples, as the frequency of occurrence was high
(mean for year was 85%). The mean relative densities of spiders, however, was low, ranging from less
than 2% in winter to about 15% in summer. There is a seasonal shift of relative densities indicating
that this population of carnivores may increase proportionately faster than its prey from winter to
summer. Actual numbers of spiders trapped seasonally ranged from 102 individuals at all sites in late
winter to over ten times as many (1140) in early summer. This mean number of species per season per
site ranged from eight in winter to nearly 22 in early summer. The four sites were not significantly
different from each other in total number or mean number of species, number of individuals or
relative densities. Only frequencies show any differences and as indicated they are suspect especially in
this preliminary study where samples are shown to be inadequate in most cases in numbers and length
of time in operation.

INTRODUCTION

This study was part of a more extensive pitfall sampling of the wandering invertebrates
of the ground surface in a streamside coniferous forest community. The sites sampled
were located in Mortandad canyon at the Los Alamos National Laboratory in New
Mexico. The study of spiders was made to determine what the species and populations of
spiders were, what their relative abundances were, their abundances relative to all other
invertebrates at the sites, what the seasonal and habitat preferences were and whether
there was a single or more than one community in the transect studied and what
modifications of techniques might be done to more accurately and definitely assess the
characteristics of the spider community (ies). It was a preliminary study in the National
Environmental Research Park. Lour sites were sampled. The site at the highest altitude
(site I) had constantly running water but no radionuclide contamination from fluid
radioactive wastes whereas the other sites had varying degrees of minor contamination.
An assessment of the affects of the radionuclide contamination was also an object of the
study. Trapping was done at five different times of the year (late winter, midspring, early

* Present address: Country Club Gardens M.H.P. #117, Santa Fe, New Mexico 87501.

98

THE JOURNAL OF ARACHNOLOGY

summer, midsummer and midfall), based on manpower, climatic and other mainly prac-
tical considerations.

This part of the study analyzes the spider portion of the carnivore trophic level. Only
part of the ground inhabitants were sampled; the mobile portion. It did not include all
spiders of the ground surface. Many of the web builders, none of the less mobile spiders,
often only one sex, and often none of the immature portion were collected. Pitfall
samples are selective and need to be supplemented by quadrat or “zeitfang” (equal effort)
samples to census all of the ground dwelling spiders.

MATERIALS AND METHODS

Site Descriptions.— Four sets of ten pitfalls each were placed along the upper elevations
(about 2200 m) of Mortandad Canyon. Each site was located 1000 m down the canyon
from the previous site. Traps were ten meters apart alongside the meanderings of the
stream. Elevation differences between sites were slight, ranging from 5 to 75 m. These
four sites appeared different enough to be different communities.

The canyon extends mainly from west to east and empties into the Rio Grande,
although only severe flash floods actually drain any water into the river. More detailed
descriptions of the soils, contamination, etc. of the canyon are available elsewhere
(Hakonson et al. 1973; Miera et al. 1977). The features of the sites which seem to be
important are summarized in Table 1.

The canyon was a dry canyon which had water in it only at periods of rainfall. Over 15
years ago a cooling tower was installed at the head of the canyon and began releasing
water into the streambed. Later, liquid effluent from a disposal plant began to be released
into the stream about 100 m below the cooling tower. This is released suddenly, creating
a rush of water at Site II which continued for as long as half an hour once or twice a day,
except on weekends. Aquatic forms occur in the stream, and waterside spiders have
become established. Site III commonly is affected by the daily surge of water although
even here its onset is quite gentle. The traps at the sites, however, are set 6-9 cm above
the stream channel and hence are not directly subjected to the effects of the surge. Site
IV seldom receives water from the liquid disposal. The water has usually sunk into the
ground somewhere in Site III. Summer storms still occasionally flood this part of the
streambed.

As may be seen from these descriptions each site differs in several features although
most of them are not sharp and distinct either quantitatively or qualitatively. Absolute
differences occur between Sites I and IV, mainly in some plant occurrences, probably in
soil moisture, soil texture and chemical composition (although not measured), winter and
summer precipitation and streamside slope. However, no close correlation with any of
these features has been made with the spider species occurring at each site. Generally,
shade, soil moisture and density of vegetation is highest at Site I and least at Site IV.
The canyon floor is greatest in area at Site IV and much more dry, level, and sandy than
at Site I, where there is more humus but a much smaller and more rocky and irregular
surface.

Methods.— A pitfall trap was designed so that it could be placed in position and easily
removed with only slightly disturbing the vegetation and soil surface.

Alcohol (75% ethanol) was placed in the bottom to preserve the trapped animals.
Pitfalls were placed in position in the morning and collected one to five days later, again
in the morning. Initial trapping (midsummer and midfall) varied in duration, as indicated.

Table 1. -Characters of invertebrate trapping sites in Mortandad Canyon.

LOWRIE-SURVEY OF WANDERING SPIDERS OF CONIFEROUS FOREST

99

g g

^ o
00

o CO

OJ

s S'

I =

g

O

o

-M

'O O G

5« »H

g

' o

O ro
<N ^

«5

a?

60 _

M O

Sis

■%

-O

>» ^ d

e

0?

a>

•4-*

a>

*+-1

„ O

o ^

a>

4-^

Ti

d

o

d

W5

O

g

o

&<

GJ ^
O ™

^ o

2 a>
tH a

OJ

>

o

Q

o

d

o .g

0 ^

&i 0^

d <u

1 'd

>' g

-t-j > s-'

»d e3
60 OJ ;
s X O

g

s « s

O d S

j-i e >-<

bO ^ a>

o 5*

^ -d (D

S ^ 4-»

e . .

5-^ ^

d ^

S o

^ 'S

^ g

d

s

6«

W5

d

4h

>» o
d

. 3

'd 2

>> w 0^

-d

g .d -S

^ ."d .-H
s« d ^

d ^ Cd

' .g

4->

d

d >>

cd

c/3 d

+-) W3

4-*

Oh

>.

}_4

0 0

03

B g

0

X

0

d d

03

>.

cd

d

d

03 §

g §

03

d

cd

4^

C/5

ȴ^

0

S S)

d

'§

g

CD

d5

o

d

g

'd

d

cd

T3

s

c/3

C/3

;:3

g

d

4>

a

c«

.d

g

03

OJ

&,

'O «

^ P
9 ^ B

g'"- ^

I c -I. “

^ ^ o ^

^ ^ I

Sti« "d 0^

d is

d3 ^ TO c«

X

o

d

g

s->

d

Xi

0)

d-

^ >

C« HH

c« Qi

da

a d

0) d

CD a
.ii d!

o-

o

3

d

id'

C/3

d

d)

d

<

d

<D

CD

'xi

d

g

c3

CD

c/5

(D

(/5

C/5

KS

)-4

60

CD

.g

Sd

o

dl

03

>>

g

in

"o

0^

03

c/5

C/5

cd

M

60

03

d

‘d

o

*03

>.

60 .g

•d

o

03

03 cd
Df 03

>. d 60

d

0

0

4->

a

H-f

©

03

i/3

SO

to

4-»

M

C/5

©

g

C/5

S)

.3

fNJ

0

'0

0

>

ca

>

03

>

o

0 >.

1 ^

d

c/3

d

o

s

>

ed

03

d!

■d

d d
rs 03

o ^

M O

60 03

60 ^

S c

d

O

03 d?

S” _.

d o

coJ d G

c/3 c/3 C

03

d

d

C6

c/3

d
.ii O

ti

^ -g

O 6S

c/3

d ^
.52 d

d §

g d

d:

Q

d

g

03

Cl,

o

g

cS

03

Cu

03

03

c/3

03

d

d

03

d

g

cd

03

a

03

03

d

ro

o

Cl

2 'S

d ^

d Cl g
d g 03
d d

C-^

>:^ c/5

M

03 '60 ^
F F ®
O c/3

ed

d

03

d

'S

X

o

d

Q ^
d

C/5

:3

£

ga

g

o

O C4-1

g s

o ^

ON

n
ro ^

60

.g

d

d

d

c/a

d

o

o

g

o

a $

s s

d, ^

S-s

trt ^
5 03
^ d

43

>

o

o

o

s

ed

03

d

«3

O

e !

cd d
03 03

M N

^ c§

i -

d 03

d

S ^

03

ed

03
d
d 5

g 'S

F o F

60 M

g “ g

^ -TS ^

•- 'g cS

d 3 !-H

F d ^

C« d C«

'F 3

§ I

“S

.03

d

.j i

03 ^

d

C/5

o

g

^ c«

d d

c/3

C/3

cd

a I s
s .c a

C/5

o

g c«
60 ^
.g

g:^

o c^
d o

g

F/

S 03
B 03

^ 03

d d
g F

Cf^

d O 03

ra d +-*

cd </3

d c« ^
03 d ^

o 13 .— I

M 4-5 03

cd -i-i >

9 ’r. 03
»5 Cl *— I

03

d

4l

Cd

d

c/3

O

d

o *1:3
c/5

C/5

C/5

bX)

>.

60

o

g

d
• ^

d

d

o

g

d

cd

O

03

d

g

cd

o

so

r-

ON

w

H

NH

C/3

03

03

d d

$ s

03

d

d

0

-t->

C/3

4-,

d

cd

d

•d

0

g

cd

03

'0

d

Cl

0

©

M

C^

a

HH

0

c/a

C/a

-M

cn

03

d

g d

cd *9

B 2

is o

c/a d

100

THE JOURNAL OF ARACHNOLOGY

mainly to determine the efficacy of one or two days versus longer trapping periods. The
winter, spring and early summer trapping was established at two sets of three days each
(except for one period when pitfalls were accidentally left for four days). Catch numbers
were adjusted (multiplying or reducing numbers actually caught) to make them all as
though six days of trapping had been done at each site each season. Although this adjust-
ment is not statistically valid or as valid as if all data had been collected the same number
of days, it is more comparable than actual numbers since sampling times did vary. I did
not know before making the samples whether the number of specimens would be too
great for identifying and counting in a reasonable period of time. Also, too few samples
might easily be too small to generate any dependable data if they were too variable, for
example. The results will show my conclusions about these methods.

The animals from a bottle at a single pitfall were examined with a dissecting micro-
scope. Each specimen was identified to species, or genus in some immatures, and the
number of individuals for each taxon enumerated. I identified most of them but sent
various groups with which I was not sufficiently familiar to the specialists indicated.
Some have still not been identified as to species and there was one new species (Millidge
1981), and several others not yet described. Micryphantids and small linyphiids were not
all identified to species and exact records as to site and season were not kept because so
many were unidentifiable immatures or females. Voucher specimens of most species
collected have been deposited in the general collection of the American Museum of
Natural History.

The following data were recorded and/or calculated: numbers of species, numbers of
individuals, sex, frequency of occurrence (Ashby 1935, Cox 1967, Curtis and McIntosh
1950), relative density (Cox 1967), site of occurrence and season of occurrence. Most of
these terms are self-explanatory or in common usage and will not be defined. A few terms
unique to this study, or which are not always used uniformly, are the following:

The mean number of individuals, or species, per site per three days of pitfall use have
been recorded (or calculated) from the data. Various factors of the environment, mainly
climatic (rainfall, temperature, wind, etc.) were seldom the same from day to day. There-
fore, as shown by the two sets of three day samples taken successively, the samples were
seldom without great variability.

Frequency (or percent frequency) is a widely-used easily determined ecological statis-
tic which is commonly defined, and so used here, as the percent of the samples in which a
species is found (Cain 1932, Cox 1967, Hoel 1943, McGinnies 1934). However, it is a
poorer statistic than most because it depends upon a number of factors such as size of the
organism, size of the sample, number of samples taken, density of the population, etc.,
instead of only a single factor such as density (number of individuals) or dominance
(size, weight, volume or the like of individuals). Specifically in pitfall trapping there is the
problem of the length of time a pitfall is in operation at any one period of sampling. In
addition there is the problem of whether there are enough samples to get a “true” statis-
tic representing the ubiquity of the population being sampled. The less common species,
more so than the common ubiquitous species, will have larger frequencies if the sample is
in operation for a longer period of time because that gives the individual of the popula-
tion a longer time to fall into a trap. I know of no way to determine the amount of time a
trap should operate to give a “true” value for that population’s “actual” frequency.

Theoretically, the larger the frequency the more widespread and/or common a species
is. To be an adequate sample of the actual population the number of pitfalls in use must
be large enough to sample the less ubiquitous species as well as the wide ranging species.

LOWRIE-SURVEY OF WANDERING SPIDERS OF CONIFEROUS FOREST

101

What skewness occurs in this sampling is on the side of making frequencies lower than
they would actually be if a larger number of samples had been taken. One way to judge
the adequacy of the sample is to determine the distribution of species according to
Preston’s octaves of abundance (Preston 1948). When these data are plotted, it is ap-
parent that there are too few rare species (those represented by one or a few individuals)
as well as not enough individuals of common species, those with large numbers of individ-
uals. This is aside from the environmental affects upon the community. Many more
samples would give a more nearly complete sampling of the total species in the commu-
nity. Nevertheless, a number of the common species have frequencies that are above 20%
which has been determined to be an adequate percentage for most of the calculations
(Ashby 1935, Cox 1967, McGinnies 1934, Morris 1960). However, this statistic is less
reliable than others used here.

Relative densities (RD) are percentages (also called relative abundance. Uetz and
Unzicker 1975) that indicate the proportion of the catch (pitfall, or group of pitfalls)
that consists of a certain species (Cox 1967). The number of individuals of each species in
the catch is divided by the total number of individuals of all species and multiplied by
100. Ideally, and theoretically, this is proportional to the actual numbers of individuals of
the species in the total community but it is affected by the mobility, or nomadism of the
species in the total community, the weather (Lowrie 1971), size, number and placement
of samples, and other factors. However, it is one of the best statistics available and at least
is more valid in comparing species abundance in a single study, like this, where samples
are all the same, taken in the same way.

Catch (or sample) is used here for the number of species, or individuals, collected in a
sample (pitfall or group of pitfalls). The community of spiders is used to encompass all
the individuals and/or species caught in the pitfalls. The population of species inhabit a
basically mixed coniferous forest with the plant species present as indicated in Table I.

Some of the limitations of sampling should be mentioned, as well as specific problems
peculiar to this study. Immatures in most studies of invertebrates are difficult to identify
and with spiders this is also a problem. However, in a specific localized area when many
adults have been captured and there are no, or few closely related species, the immatures
of that genus are the species which has been identified by the adults.

The pitfall technique, as has not been too clearly or generally emphasized in the
literature, samples only moving individuals. It seems better than the quadrat method for
sampling this roving population (Uetz and Unzicker 1975). However, any stage of devel-
opment of a species which is relatively sedentary will only rarely be sampled, so it does
not sample roving and stationary ground species equally. Most ground level species
probably do move at some time during their life cycle, although for each species these
values would be different. They are certainly not trapped always in the same proportion
that they occur in an area. This gives relative densities and numbers per pitfall that must
be markedly different from, usually below, and at least not always in proportion to
their actual numbers. Quadrat sampling can compensate for this to a great extent but the
amount of field work necessary to get an equal amount of information is several times as
great and in addition will not sample the tiny species very well unless it is combined with
the TuUgren Funnel extraction or other methods.

Variability between samples is great. This is at least partly due to the differences in the
weather (rain, snow, temperature and wind mainly) on the sampling days. These data
from this preliminary sampHng show that this variability can be smoothed out or damp-
ened by more days of sampling at any one sampling period. The two sets of three day

102

THE JOURNAL OF ARACHNOLOGY

samples or the three sets with a total of nine days of sampHng in this study give better,
more nearly average representative counts of the wandering spiders of an area than a
short, one or two day, sample.

Finally, there is probably a lack of sampling of some species and “ overcatching” of
others because their vision allows them to avoid the trap or they may be attracted to it as
a possible retreat to avoid rigorous weather, predators, or other features of the environ-
ment. However, there is no way to determine the extent of these effects on numbers
caught— it must simply be recognized that the sampling is biased.

RESULTS AND DISCUSSION

Frequency and Relative Density.— Spiders average nearly 10% of the invertebrate
mobile ground populations of Mortandad Canyon for the entire year.

Relative density figures show that Acarina had an RD of 35%, Collembola 28%,
Formicidae 17%, Diptera 3.6%, Coleoptera 2.4%, Homoptera 1.8%, Heteroptera 1%,
Thysanoptera 0.9%, Hymenoptera (mainly wasps) 0.8%, Orthoptera 0.5% and the remain-
ing insects (Psocoptera, Lepidoptera, Thysanura, Mecoptera, Neuroptera, Siphonaptera)
the remaining 1%.

In considering the spiders as a whole, we find the following frequencies and relative
densities by sites and seasons (Table 2). Although the relative densities are low the
frequencies are quite high most of the year. In other words, spiders are found in most of
the pitfalls most of the time, that is, spiders are quite ubiquitous. Throughout the year
they are found in 85% of the pitfalls. Late winter occurrence was significantly lower than
any other time with a mean frequency of 55%. At other seasons all four sites had similar
frequencies.

Early summer contrasted markedly with winter in that every pitfall had at least one
spider (frequency 100%). This is true even if the great number of active males of Pardosa
yavapa which were in their courtship stage of life, searching for females, were eliminated
form the analysis. Eliminating this species could only change two sets of pitfalls dropping
the F to 90%. Midspring was also high with a mean of 94%. The other seasons were lower
but high also.

The frequency of finding spiders at each site was high and about equal each season
although winter, as might be expected, was lower. The relative density of spiders fluctu-
ated much more than did the frequency (Table 2). Relative densities for spiders ranged
from a low of 2.4% in winter at Site I (they were a small proportion of the total com-
munity) to a high of 25% at Site II in early summer, nearly eleven times as great. How-
ever, this high value is greatly inflated since over 75% of the catch was of the one species,
Pardosa yavapa, most of which were males in search of females with sex, not food, as the
stimulus for movement.

I am at a loss to explain these data as far as density analyses are concerned because of
the large number of P. yavapa. Similar increases in abundances of some other species
probably occur at mating time although not in all species to the same degree. It is possible
also that the absolute density of this species is greater than that of any others in this
community but this one set of early summer samples can only hint at such a possibility.
More extensive sampling seems called for to better analyze this species’ abundance and its
relationship to the other species in the community.

LOWRIE- SURVEY OF WANDERING SPIDERS OF CONIFEROUS FOREST

103

Table 2. -Spider distributions by seasons and sites (for six trap days).

I

II

Sites

III

IV

Mean

Total

Late

No. of species

6

4

6

16

8

25

Winter

No. of individuals

13.7

6.9

22.0

59

25

101.6

% Mean frequency

50

50

50

85

59

% Mean relative density

1.4

3

0.8

2.6

1.9

Mid

No. of species

13

11

16

26

16.5

30

Spring

No. of individuals

70

45

89

53

64

257

% Mean frequency

100

90

100

85

94

% Mean relative density

8.6

5

4.75

3.05

5.4

Early

No. of species

19

19

22

26

21.5

45

Summer

No. of individuals

242

418

319

161

285

1140

% Mean frequency

100

100

100

100

100

% Mean relative density

15.3

25.45

14.45

7.95

15.8

Mid

No. of species

18

15

10

14

14.25

32

Summer

No. of individuals

60

43

25

81

52

209

% Mean frequency

87

83

71

93

84

% Mean relative density

12.4

10.2

4.35

9.3

9.5

Mid

No. of species

9

12

8

11

10

21

Fall

No. of individuals

43

41

61

57

50.5

202

% Mean frequency

83

90

87

93

88

% Mean relative density

11.1

7.5

11.9

2.1

8.1

Yearly

No. of species

31

32

26

43

33

78

Means

Mean no. of species

13.0

12.2

12.4

18.6

14

and

No. of individuals

429

554

516

411

476

1910

Totals

Mean no. of

Individuals per site

85.8

111

103

82

95-97

382

% Mean frequency

84

83

82

91

85

% Mean relative density

9.8

10.2

7.25

5.0

8.1

The mean yearly RDs of spiders at Sites I, II and III were 7% to 10% whereas at IV the
RD was only about half that, 5%. The variability in relative density from site to site and
season to season was similar. Sites I and II were about twice as high in RD as III and IV.
Most seasons showed lowest RDs about half that of the highest, except in midfall.

Statistically significant relative densities between sites and seasons were few as deter-
mined by using the arc-sine conversion of RD figures. However, there is no clear trend or
clue as to what the causes of these differences might be (and for this reason Fm not
presenting these data). There were significantly higher RDs for early summer, but definite
conclusions mustl3e avoided at this time because they are due to the larger number of
Pardosa yavapa as indicated earlier. All other seasons are not statistically significantly
different from one another. More samples at each site might show significant differences.
At present it would seem that the only certain conclusion is that the relative density of
spiders in winter is significantly low while for the rest of the seasons the proportion of
the community that is spiders remains high and about the same.

The RDs of spiders and the actual numbers of all invertebrates (all potential prey) per
pitfall or site are related as follows. Discounting early summer for the reasons already
noted, only Site IV shows RDs that are significantly lower from season to season. When
more than about 50 individuals invertebrates are collected in a pitfall the RDs of spiders
are usually less than 10%. Conversely, when spider RDs are high (more than 20%) then

104

THE JOURNAL OF ARACHNOLOGY

the numbers of individuals per pitfall are less than about 40. This is not applicable to the
early summer figures. The added numbers of male P. yavapa produce higher RDs. When
both values (numbers of individual invertebrates and RD of spiders) are low there are
neither positive nor negative correlations. And, finally, there are no cases where numbers
of individuals and RDs are both high. This seems to mean that when the numbers of
individuals of all species in the environment are high, then the numbers of spiders are not
increasing as rapidly to avail themselves of the added prey. Conversely, when the propor-
tion of spiders is high it might be because the prey have died and left proportionately
more spiders alive. This condition applies (except for the unusual summer condition)
almost exclusively during midfall (about 20 pitfalls out of the 120 censused). Only five of
nearly 200 pitfalls produced over 25% RD (except for early summer).

Finally it must be acknowledged that these are possible conclusions only as many prey
species were probably not sampled in the pitfalls and features such as life cycle durations
and mobility patterns of the insects and spiders were not known or taken into considera-
tion. The collecting may average out differences or even skew them one way or another,
but I present them here as giving some evidence that predator-prey relationships may
show the lag indicated, and should be investigated in any subsequent study of this sort.

Numbers of Individuals.— In this discussion I am eliminating the early summer collec-
tions (Table 3). The number of individuals for that period is significantly higher than for
any of the other collecting periods. But this is due to the large number of Pardosa yavapa.
Of 1140 individuals collected during this period, 819 were P. yavapa, and over 80% of
these were males. Collecting was apparently done at or near the peak of their period of
search for females. This was corroborated by some collecting done at the same time the
following year, although the numbers were only about half as large.

Finally, I am not considering this summer collection in detail also because it was taken
at the beginning of the summer season and comparable collections at the beginning of
each season were not made. Although it was a valid collection in general it seems to me
that because of these factors (including the overwhelming numbers of P. yavapa) it is
reasonable to consider these data as atypical. I will only point to the actual figures and
not compare them with the other collection figures.

In general the number of individual spiders per pitfall, regardless of the species, was
low. Means ranged from 0.4 spiders per trap to 4.45 while actual numbers went as high as
13. Finally, only 34 of 379 traps for the year had six or more spiders in them. The overall
mean for the four seasons was 2.5 spiders per pitfall. The data in Table 2 indicates that
few of the pitfall means were different from one another, except for the winter sampling

Table 3, -Number of individuals of Pardosa yavapa (adjusted to 10 traps totals for six days).

I

Sites

II

III

IV

Season

Total

Late Winter

0

0

0

1

1

Midspring

27

4

8

7

46

Early Summer

177

362

184

99

822

Midsummer

4.5

0

0

11.3

15.8

Midfall

.7

0

0

7.3

8

Totals

209.2

366

192

125.6

892.8

LOWRIE-SURVEY OF WANDERING SPIDERS OF CONIFEROUS FOREST

105

at Sites I, II and III. Some were statistically significant, but no trend is obvious enough to
make the data reportable. They were all significantly lower in spider catches from nearly
all other sites and seasons of collecting (except early summer).

Dispersion of Spiders.— An attempt was made to determine whether the spiders were
dispersed in a random, clumped or uniform fashion (Cole 1945). Using analyses relative
to the Poisson distribution (variance equal to the mean in randomly dispersed populations
and variance greater than the mean in clumped populations) all seasons show clumped
figures. Even when considering one set of pitfalls placed at a site for one, three or five
days only three samples of ten pitfalls showed a dispersion that was not clumped.

All indications in this study are that the spiders are dispersed in a somewhat clumped
fashion. The difference in conclusions between this analysis and Cole’s is presumable due
to the differences in the habitat. The wooded area in which Cole sampled was a rather
homogeneous habitat without marked differences in moisture, litter, temperature or
other ecological factors to which the spiders might react. Although each of the sites in
Mortandad Canyon did not have a great deal of difference between pitfalls, nevertheless,
there were differences in moisture, temperature due to insolation, degree of shade, etc.
These were probably enough to cause spiders to aggregate in certain pitfalls and not so
much in others. This created a tendency to clump because of the environment and
not because of the behavior of the spiders to the presence of other spiders (or other small
carnivores) and/or prey.

Number of Species.— The total number of species of spiders collected at a site seasonal-
ly ranged from four (late winter at Site II) to 26 (early summer at Site IV— Table 2),
about seven times as great. During any season the differences between sites were not as
great, ranging from four to only 1 .4 times as many species at the site with the greatest
number of species as the one with the least number. This means that there were about the
same number of active species at any of the sites at any particular season.

The greatest variation in numbers of species is seasonal. Site II and III show the
greatest variation with early summer showing between four and five times as many species
as the low winter numbers. Site IV is more uniform, with numbers of species varying
from a low of eleven to a high of 26, only 2.4 times as great. (This mean number of
species per season per site varied from eight in late winter to 2.7 times as great (21.5) in
early summer). Consistent with this is the fact that the range in total numbers of species
collected at all sites per season was also great, varying from 21 in midfall to 45 (2,14
times as many) in early summer.

Site IV has the greatest variety of species; more species occurred there in most seasons
than at any other site. From a general assessment of the sites this would seem due to the
site being less extreme, or at least less variable, in temperature, precipitation and mois-
ture. The site would seem to have a greater variety of microhabitats also. Both the total
number of species found here (43) and the mean number per six pitfall days (19) was
over 1.5 times as great as at the lowest site (Site III). At the same time the community of
spiders at Site IV was more stable, showed less variation in numbers, than at the other
sites. This generally coincides with our knowledge of more complex communities such as
rain forests versus less complex communities such as deserts and tundra. It is also an
expression of the fact that the physicochemical parts of the environment (temperature,
winds, humidity, moisture, etc.) are more variable and affect the biotic community more
in a less protected environment than in a more complex community in which the biotic
affects control the organisms mostly (Odum 1971).

106

THE JOURNAL OF ARACHNOLOGY

Seasonally, the late winter is most variable in number of species (from site to site: four
to 16) whereas summer is least variable (14 to 26). Mean seasonal variations are of greater
magnitude (eight to 21.5— seven times greater) than mean site variations (26 to 43—1.65
times greater). Only winter shows a range between sites (four times as great) that is
greater than the magnitude of the range between seasons (eight to 21.5— only 2,7 times as
great). This again illustrates the possibility that physicochemical factors control the
environment in winter.

In summary, the range in numbers of species between sites was never great at any one
season except in late winter. That is, at any season but winter, the number of species at
each site was not significantly different.

Phenology— Seasonal Distribution.— The patterns of seasonal activity shown by the
common species (Appendix) are as follows. Only Hahnia cinerea is active equally, or
nearly so, at all seasons. Most species are active in spring, often in greater numbers at
certain sites. Pardosa yavapa is abundant at all sites in early summer but significantly less
so at IV and with a definite preference for site I. Agroeca pratensis is more common at
Sites I and III whereas Zelotes subterraneus is more common at III and IV. In terms of
the moisture gradient the Cicurim and the Gnaphosa may be tolerant of a wide range
of moisture whereas P. yavapa and the Agroeca prefer moist sites.

Most common species show preference for two sites rather than a single site. The
following distributions can be inferred from this study, but only future replication can
establish them as true consistent generalizations for the species involved. Pardosa sierra, a
typical streamside species (Lowrie 1973) and Hahnia cinerea prefer Sites I and II.
Titanoeca silvicola and Micaria montana are more abundant at Sites II and III and Neo-
antistea gosiuta at Sites III and IV. Schizocosa mccooki and Trochosa gosiuta show a
strange abundance at the extremes, I and IV. What the explanation of this is cannot be
determined for certain but does seem to correlate with the species preference for drier
habitats: Site I does have some dry areas. Only one Haplodrassus, is commonly at
one site only, III. The two species of the genus do occur at other sites but only as one or
two individuals. However, this may be an artifact of collecting and identifying as most
specimens were immature, or it may be just the situation this year.

At each site we can say there are about the same number of common species. It is
suggested that this may be due to the carnivorous habits of spiders and the tendency for
carnivores to be evenly or randomly distributed (Cole 1945), in a uniform habitat.

A cluster analysis of the common spiders at each site was also done, but not presented
here because no significance can be attributed to it since they were not consistent and no
field data or information on the species would give a reason for such correlations. Seven-
teen species of spiders were involved with an unweighted pair-group method used to
produce a dendrogram. This dendrogram showed Sites I and II to be very similar and Sites

III and IV likewise, with a greater distance between II and III. This could fit in with the
evaluation of the situation from year long observation although other evaluations may be
equally plausible, such as that Site I is different from II and III, and III is different from

IV or I. There is little justification for any conclusion of differences between sites. A
much more extensive and intensive sampling would be necessary to determine whether
there were any differences.

LOWRIE- SURVEY OF WANDERING SPIDERS OF CONIFEROUS FOREST

107

CONCLUSIONS AND SUMMARY

1. Spiders constitute about 8% of the wandering streamside ground-inhabiting in-
vertebrates of this mixed-conifer biotic community, in Mortandad Canyon.

2. Spiders’ frequency of occurrence (measure of their ubiquity) is high with a mean
for the year of 85%, winter was significantly low (59%) and summer was high (100%).

3. Spiders occurred about equally at the four sites with frequencies from 82% to 91%
for the year. Site IV showed higher frequencies than other sites, but in light of the
unreliability of this statistic do not seem to warrant conclusion of a difference at Site IV
from this study.

4. The mean relative densities of spiders were low, ranging from less than 2% (late
winter) of the invertebrate population to over 15% (early summer). Their relative densi-
ties were highest (about 10%) at Sites I and II and lowest (5%) at Site IV.

5. At Site IV the RDs were most stable (varied less) throughout the year (2.1% to
9.3%). Site III was most variable ranging from a low of 0.8% to a high of 14.45%. The
seasonal shift in relative densities of spiders indicates that this carnivorous population
increases proportionately more than its prey population from winter into summer. It then
regresses during the rest of the year to a low proportion when prey seems to be corre-
spondingly low.

6. The actual densities of spiders (the numbers of individuals per pitfall or site)
throughout the year were about equal at each of the sites (low mean of 82 individuals per
site at IV to a high mean of 1 1 1 at Site II). Seasonally, their abundance ranged from a
total of 102 individuals from all four sites in late winter to over 200 in the other seasons,
with a high in early summer of 1 140. Thus densities were lowest in late winter (mean of
25 individuals per site) increasing to a high in early summer (mean of 285 individuals per
site) and then back down to a low in winter.

7. The mean numbers of species per season range from eight per site in winter to
nearly 22 per site in early summer and then declined to the near low of ten in midfall.
There was an overall mean of 14 species per site for the year. The mean number of species
per site for the year ranged from lows slightly over 12 at the 3 higher sites to a high of 19
species at IV. The number of species (species diversity) was greatest at Site IV most of the
year, and was more variable at the other sites.

8. The dispersion of spiders was clumped. This may be due to the habitats being
relatively heterogeneous with a variety of micro-environments although no analysis of this
factor was made. If that is the case then the spiders would clump in microhabitats that
they prefer, and avoid those not suitable.

9. This study shows no effect of the contaminating radionuclides introduced into the
stream at Site II. That is, the sites of contamination did not show significantly greater or
fewer numbers of individuals, number of species, relative densities, frequencies or other
measurements than any other site. It is concluded that this study shows no effect of the
radioactive material on the spider population.

108

THE JOURNAL OF ARACHNOLOGY

ACKNOWLEDGMENTS

This study was done at the Los Alamos National Environmental Research Park.
Financial and other aid from the University of California, Los Alamos National Labora-
tory was obtained under contract W-745-ENG-36. The laboratory provided equipment,
such as pitfall trips, alcohol and vials. I would also like to acknowledge the aid of C. D.
Dondale, W. J. Gertsch, H. W. Levi, and N. 1. Platnick in identifying some of the difficult
species and for checking some of my identifications, and Richard T. Carter of the Univer-
sity of Manitoba for identifying the Micryphantidae. Most of the statistical analysis was
provided by Gary Tietjen of LANL, and for this I thank him. G. W. Uetz reviewed the
manuscript and contributed valuable suggestions for which I thank him.

LITERATURE CITED

Ashby, E. 1935. The quantitative analysis of vegetation. Ann. Botany, 49:779-802.

Cain, S. A. 1932. Density and frequency of the woody plants of Donaldson’s Woods, Lawrence Co.
Indiana. Proc. Indiana Acad. Sci. 41:105-122.

Cole, Lamont. 1945. A study of the cryptozoa of an Illinois woodland. Ecol. Monogr., 16:49-86.
Cox, G. W. 1967. Laboratory manual of general ecology. W. C. Brown and Co., Dubuque, Iowa.
Curtis, J. and R. M. McIntosh. 1950. Interrelationship of certain analytic and synthetic phytosocio-
logical characters. Ecology, 31:434456.

Hakonson, T. E. and J. W. Nyhan, L. J. Johnson and K. V. Bostick. 1973. Ecological investigation of
radioactive materials in waste discharge areas at Los Alamos for the period July 1, 1972 through
March 31, 1973. Los Alamos Laboratory Report, LA-5 282-MS.

Hoel, P. G. 1943. The accuracy of sampling methods in ecology. Ann. Math. Stat., 14:289-300.
Lowrie, D. C. 1971. Effects of time of day and weather on spider catches with a sweep net. Ecology,
52:348-351.

Lowrie, D. C, 1973. The microhabitats of western wolf spiders of the Pardosa. Entomol. News.,
84:103-116.

McGinnies, W. G. 1934. The relation between frequency index and abundance as applied to plant
populations in semi-arid regions. Ecology, 15:263-282.

Miera, F. R. Jr., K. V. Bostick, T. E. Hakonson and J. W. Nyhan. 1977. Biotic survey of Los Alamos
radioactive liquid effluent areas. Los Alamos Laboratory Report, LA-6503-MS.

Millidge, A. F. 1981. The Erigonine Spiders of North America. Part IV. The Genus Disembolus.

Chamberlin and Ivie. J. Arachnol., 9:259-284.

Morris, B. F. 1960. Sampling insect populations, Ann, Rev. Entomol., 5:243-264.

Odum, E. P. 1971. Fundamentals of Ecology. W. B. Saunders Co., Philadelphia.

Preston, F. W. 1948, The commonness and rarity of species. Ecology, 29:254-283.

Uetz, G, W. and J, D. Unzicker. 1975. Pitfall traps in ecological studies of wandering spiders. J.
Arachnol., 3:101-111.

Manuscript received March 1984, revised August 1984.

LOWRIE-SURVEY OF WANDERING SPIDERS OF CONIFEROUS FOREST

109

Appendix

Wandering Spiders of Mortandad Canyon collected in pitfalls. Numbers for commonest species
(more than 10 specimens) in parentheses: 1 = Rank (16 most common), 2 = Number of individuals
captured.

Agelenidae

Amaurobiidae

Anyphaenidae

1. Cicurina robusta Simon (4 - 60)

2. Titanoeca silvicola Chamberlin and Ivie (10 - 20)

3. Anyphaena marginalis (Banks)

4. A. paciflca (Banks)

Clubionidae

5. Agroeca pratensis Emerton (2 - 142)

6. Castianeira cingula ta (C. L. Koch)

7. C. de script a (Hentz)

8. Clubiona sp?

9. Phrurotimpus nr. woodburyi Chamberlin and Gertsch

10. Trachelas deceptus Banks

Dictynidae

1 1 . Dictyna apacheca Chamberlin and Ivie

12. D. completa Chamberlin and Gertsch

13. Z). terrestris Emerton

Gnaphosidae

14. Callilepis eremella Chamberlin

15. Drassodes neglectus Keyserling

16. Drassylus nr. argilus Chamberlin

17. Gnaphosa muscorum (L. Koch) (12-14)

18. Haplodrassus chamberlini Platnick and Shadab

19. Haplodrassus bicomis (Emerton) (11 -17)

20. Micaria montana Emerton (3-98)

2\. Nodocion m. florissantus (Chamberlin)

22. Zelotes subterraneus (C. L. Koch) (5-57)

Hahniidae

23. Hahnia cinerea Emerton (7-42)

14. Neoantistea gosiuta GQxXsch. - 11)

Linyphiidae

25. Helophora sp?

26. Lepthyphantes subalpina Emerton

27. Lepthyphantes sp?

28. - 3\. Meioneta sp?

32. Neriene radiata (Walckenaer)

33. Wubana drassoides (Emerton)

Lycosidae

34. Arctosa sp?

35. Pardosa montgomeryi Gertsch

36. P. orophila Gertsch

31. P. sierra Banks (13-12)

38. P. sternalis (Thorell)

39. P. yavapa Chamberlin (1-919)

40. Schizocosa mccooki (Montgomery) (9-21)

41 . Alopecosa kochi (Keyserling)

42. Trochosa gosiuta (Chamberlin) (6 - 47)

Micryphantidae

43. Ceraticelus crassiceps (Chamberlin and Ivie)

44. Ceraticelus sp?

45. Ceratinella sp?

46. Ceratinops n. sp.

47. Collinsia perplexa (Keyserling)

48. C. plumosa (Emerton)

49. Disembolus anguineus Milhdge

110

THE JOURNAL OF ARACHNOLOGY

50. Eperigone taibo Chamberlin and Ivie

5 1 . E’. trilobata (Emerton)

52. Eperigone sp?

53. Grammonota gentilis Banks

54. Islandiana flaveola (Banks)

55. Pocadicnemis pumila (Blackwell) (8 - 22)

56. Spirembolus pallidus Chamberlin and Ivie (14 - 14)

57. S. vallicolens Chamberlin

58. Walckenaera directa (0. P.-Cam bridge)

59. IP. spiralis (Emerton (15 - 12)

60. Walckenaera n. sp.

61. - 63. Three species not placed as to genus

Oonopidae

64. Orchestina saltitans Banks

Philodromidae

65. Thanatus coloradensis Keyserling

66. T. formicinus Clerck

Pholcidae

67. Pholocophora americana Banks

68. Psilochorus imitatus Gertsch and Mulaik

Salticidae

69. Pellenes sp?

Tetragnathidae

70. Tetragnatha laboriosa Hentz

Theridiidae

7 1 . Euryopes sp?

72. Theridion murarium Emerton

Thomisidae

73. Ozyptila sincera canadensis Dondale and Redner

74. Misumenops oi Misumenoides sp?

Dean, D. A. and W. L. Sterling. 1985. Size and phenology of ballooning spiders at two locations in
eastern Texas. J. Arachnol., 13:111-120.

SIZE AND PHENOLOGY OF BALLOONING SPIDERS
AT TWO LOCATIONS IN EASTERN TEXAS^

D. A. Dean and W. L. Sterling

Department of Entomology
Texas A&M University
College Station, Texas 77843

ABSTRACT

The aerial dispersal of spiders at two locations in eastern Texas was studied using a Johnson-Taylor
suction trap. A total of 17,596 spiders were collected at College Station, Texas during a one year
period, and 6248 spiders at the Ellis Unit from April to August. Nineteen families were found includ-
ing Uloboridae and Hahniidae. The greatest numbers of spiders were collected during May and June at
College Station with immatures representing 94% of total spiders collected during the year. The
families Erigonidae, Araneidae, and Oxyopidae comprised ca. 74% of aU spiders collected at College
Station. The size class 1-2 mm represented 64% of aU spiders collected at College Station with those <
1 mm in length next in abundance with 28%.

INTRODUCTION

Suction traps have been used to study the monthly dispersal of spiders in Oklahoma
and Texas (Horner 1975, Salmon and Horner 1977). Ballooning spiders, captured in a
Johnson-Taylor suction trap, peaked during June and September (Salmon and Horner
1977) and spiders in the family Erigonidae were reported to be the most numerous.
Horner (1975) studied the dispersal of salticids in Oklahoma and found immatures
peaking in July. Adult spiders were also observed to balloon.

Other workers used different methods to study dispersal. Click (1939) collected
ballooning spiders at different heights by airplane in Louisiana. The number of spiders
collected was highest near the ground and lowest at 5000 ft. elevation. Freeman (1946),
at a height of 3^m, used nets attached to masts to observe that spiders actively ballooned
at all wind velocities ranging from 6-35 mph. More ballooning male spiders were present
in September and October than any other month in England (Duffey 1956). Ballooning
females were abundant over a longer period of time, from September to March. Bristowe
(1929) observed that immatures of several families balloon in late summer and early
autumn in England but as winter approaches, the proportion of linyphiids increases and
the other families decrease.

Spiders are known to be predaceous and McDaniel and Sterling (1982) and McDaniel
et al. (1981) determined those species of spiders which are predaceous onHeliothis spp.
eggs and larvae in a cotton field in eastern Texas. Dean et al. (1982) listed the species of

^ Approved for publication as TA 19150 by Director, Texas Agricultural Experiment Station.

112

THE JOURNAL OF ARACHNOLOGY

spiders collected from the same cotton field. To eventually predict the phenology and
abundance of spiders colonizing cotton fields, the seasonal profile of potential colonizing
species should be known. The current study was conducted to determine the seasonal
phenological profile of each spider family and the size most likely to balloon in eastern
Texas. The dispersal of adults identified to species will be presented in a separate paper.

METHODS AND MATERIALS

A Johnson-Taylor suction Trap (Johnson and Taylor 1955) was used to study the
aerial dispersal of spiders in eastern Texas. The cylindrical trap is 2.5 m tall and 55.8 cm
in diameter. A cone-shaped fine mesh screen located in the trap funnels the spiders into a
one pint jar of either 70% ethyl alcohol or 50% ethylene glycol. A 0.33 h.p. electric
motor driven fan pulls air and spiders into the trap.

Two locations were used for sampling; at College Station, Texas from 26 March 1979
to 25 March 1980 and at the Ellis Unit of the Texas Department of Corrections from 3
April to 31 August 1980 (through the growing season). The College Station site was
located next to the Entomology Research Laboratory on the Texas A&M University
campus. The area is surrounded by paved streets, buildings, small cotton plots, natural
vegetation, and trees. The Ellis Unit is located 8 km northeast of Huntsville, Texas. This
site is on a flood plain of the Trinity River and surrounded predominantly by pinewoods.
The trap was located in an area surrounded by pastures, silage, sorghum, and corn; the
nearest cotton was ca. two km away. The trap was set at ground level at each location,
and spiders ballooning at ca. 2.5 m in elevation were captured. Jars containing the sam-
ples were changed daily at College Station and daily when possible at the Ellis Unit. It is
possible that large adult salticids may have crawled into the trap since a nest was occa-
sionally seen near the top of the trap. However, adults [e.g. Phidippus audax (Hentz)]
were observed attached to a drag line and being blown about by the wind.

Identification to the family level was recorded along with the life stage (adult/imma-
ture). No attempt was made to determine the instars, but each spider was assigned to an
arbitrary size class (< 1 mm, 1-2 mm, etc.). The size was measured from the anterior of
the carapace to the posterior of the abdomen, excluding the spinnerets.

RESULTS

A total of 17,596 spiders in 19 families were captured in the suction trap at College
Station (hereafter, abbreviated as C.S.) during a one year period (Table 1). Most of the
families have previously been reported to balloon. The Mysmenidae were included in the
Theridiidae but are currently considered to be a separate family (Platnick and Shadab
1978).

Spiders were captured in the trap every month of the year, but the highest catches
were recorded in June and September. Most of the spiders were immature (94.0%). More
adult males (4.4%) were captured than adult females (1.6%). The size class 1-2 mm repre-
sented 64.1% of all spiders collected. Other size classes included < 1 mm (28.1%), 2-3
mm (6.1%), 3-4 mm (1 .0%), 4-5 mm (0.3%), and > 5 mm (0.4%).

Most (84%) of the spiders at C.S. were captured between May and October. Salmon
and Horner (1977) found a similar pattern at Wichita Falls, Texas. Collections made
during the winter months of January through March comprised 2.7% of the total yearly
collection at C.S. compared to 1.7% at Wichita Falls.

Table 1. -Ballooning spiders collected at College Station, Texas (26 March 1979-25 March 1980).

DEAN AND STERLING-SIZE OF BALLOONING SPIDERS IN EASTERN TEXAS

o ^

3

VO

00

C"

CO

OO

ON

ON

ON

ON

VO

o

00

CO

CN

o

CO CO

1—1 VO

o

<N

VO

00

VO

VO

o

(N

(N

(N

H

ON

o

ON

VO

00

OO

VO

<N

(N

CN

T-H

T— 1

VO

H

VO

(N

t''

c

o

S

o

a>

Q

>

o

o

O

Cl,

C/D

feX)

cs

3

c

3

a

a

<

M

OJ

43

<U

[X,

fO ’“t r— <

T™., T— ,

vvvvvvvvvv

r'lcrsooMDOO-^ascys'^'^DCNm

O^Tl-Tl•fScoc^r-lrO'>^^— (

•^(:?^'^cncomi:'~oooio00ooorOLO

CO

coeNi-HOs'OoocNr'-

c^T-(vo(N'^‘or'i’^

LO »— t CO ,-H

LX) CO VO VO CN

^'-lr)^''00(NO^f-'COco<Nco^i^^''loa^
OOCOCNOOV'OVOCOCOCOCOfNfN
I Ov <N CO t-h

! I

I (N I

CO I

I lO

C" VO OO CO
00 <N VO Ov
VO C" <N

O VO O

I 00

VO VO lO
T-l CO CO

VO CO o I <N

COiO-^OiOVOOcOt^VOOOOVO
OCNC-'CNVOvOCOt— I
00 VO c^

CO

i>-,-Hr-'Ovoco<N'<^(N>o

O-lOCOVO-'^-'^VOCOi-H
?N O T-l CO

lo VO CN I I I I

o

VO

I CO
VO

I VO
VO

I OO
00
OO
CO

I VO
VO

o

CN

I VO
VO
ON

VO

o

CO

cor-C'-ooONONcoTt
(N "?}• CO ^ 1—1 <N T— I T— I
t"- ON Tf

I (On VO <N 1—1 I VO I I I

VO VO CO

1—1 ^3 1-H

CN) fS <N

C— t" O ON 1— I VO

CO I I 00 I I I

oaNvoooooovoT^T^cNr^cocN

00 CO CS i-( 1— I (N 1—1

I 1 (N I I I

o

<N

CO

VO

C"

VO

CO

CN

rt- <N VO ON
CS 1— ( t-H t^

Ov C" VO OO

T— I

1— ( 00 On 00 VO 1— ( 1^ 1— ( 1—1 I

r- <N <N CO CO CN VO

I I I OO

CN

I I I

0)

C(3

<D

cd

CD

(U

c^

xi

cd

(D

ec3

Xi

T3

• ^

a

Xi

:3

’S

(U

C

'O

o

Kn

o

cd

M

o

•c

•c

w

X <o >,

O H hJ

0)

<3

TO

•Q

'•C!

3

C/D

(U

CVJ

xi

(D

s

o

-o

o

(U

t3

c

>1

(U (U

Xi xi

:a -3

i.1

>1 43

g

J S .5^ .S 3 q
^ - Q ^ u <

(D

Cj

T3

3

CD

CCJ

43

Cl,

>1

(D

CU

TS

H Ph

c

<D

a

c/i
>1

0)

a

I5l

p'g

E !•§

S s

CD <D
ccj CCJ
Xi Xi

'C

^ s

3 JC

S ffi

3

+-*

o

H

114

THE JOURNAL OF ARACHNOLOGY

At the Ellis Unit 6248 spiders were captured during the growing season (April- August).
The same families found at C.S. were also found at Ellis (Table 2). A lower number of
immatures was found at this location (83.0%) and adult males (9.3%) slightly outnum-
bered adult females (7.7%). The distribution of size classes was as follows: < 1 mm
(21.8%), 1-2 mm (66.4%), 2-3 mm (6.5%), 3-4 mm (2.5%), 4-5 mm (1.4%), and> 5 mm
(1.4%). Peak dispersal occurred during May.

In the < 1 mm size class, the three most abundant families at each location were the
Araneidae, Erigonidae, and Tetragnathidae. The Erigonidae and Araneidae were among
the most abundant families in the 1-2 mm class (Fig. 1). The most abundant families in
the 2-3 mm and > 5 mm classes varied between locations. Although not shown in Fig. 1,
the 3-4 mm and 4-5 mm size classes were dominated by the Salticidae.

Spiders captured in the suction trap may not accurately represent the total ballooning
fauna since it collected only those ballooning at the 2.5 m altitude. However, Click
(1939) reported that spiders were most abundant close to the ground. Other factors such
as amount of silk, wind, habitat, and time of day may affect the ballooning fauna so to
sample the entire spider fauna may require the use of several sampling techniques simul-
taneously.

The Erigonidae and Tetragnathidae are subfamilies of the Linyphiidae (Millidge 1980)
and Araneidae (Levi 1980), respectively, but are raised to the family level here for greater
detail and for comparisons to the study by Salmon and Horner (1977).

Family Dispersal at Erigonidae: One-third (34%) of the spiders collected belong
to this family (Table 1). The size class < 1 mm comprised 17.3% of the erigonids and
peaked in June whereas those > 1 mm (1-2 mm, 80.8%) remained abundant from May to
October. Males peaked in September (20.1% of male erigonids) and females peaked in
August (17.0%). Adults comprised 12.2% of total erigonids.

Araneidae: The most numerous size class of araneids was < 1 mm (57.0%) which
peaked in June. Spiders > 1 mm (1-2 mm, 41.7%) remained abundant from May to
September.

Oxyopidae: This family was most abundant during the fall. Most of the specimens
collected were < 2 mm (86.7%) and were captured in September. A peak occurred in
November for those > 2 mm. The 2-3 mm class contained 1 1 .8% of the oxyopids and < 1
mm with 2.3%.

COLLEGE STATION ELLIS

= ERIGONIDAE
^THERIDIIDAE

Fig. 1.— Percentages of the most abundant families of spiders collected in a suction trap at College
Station and the Ellis Unit in Texas.

DEAN AND STERLING-SIZE OF BALLOONING SPIDERS IN EASTERN TEXAS

115

Tetragnathidae: This family was the fourth most abundant. Those spiders < 1 mm
(49.9%) were the most numerous and peaked in May and June. Those > 1 mm (1-2 mm,
45.1%) also peaked in May.

Lycosidae: Nearly half (48%) of the lycosids captured ballooned in September. The
size class 1-2 mm (66.3%) was most abundant in September. Those > 2 mm (2-3 mm,
27.9%) also peaked in September.

Theridiidae: The peak month was September for the size class < 1 mm (39.2%).
Theridiids > 1 mm (1-2 mm, 56.8%) also peaked in September. Adults were most numer-
ous in the trap in September with more males being captured than females.

Salticidae: The salticids 1-2 mm in length (46.0%) were most numerous in June.
September was the peak month for the size class 2-3 mm (31.2%). The 34 mm class
comprised 14.5%.

Thomisidae: Thomisids peaked during the fall months as did those in the size class 1-2
mm (77.4%) in October and November. The other classes were similar in abundance: < 1
mm (5.4%), 2-3 mm (7.5%), 34 mm and 4-5 mm (4.3% each).

Philodromidae : Like the previous family, the philodromids also peaked during the fall.
The peak month for those < 2 mm (45.4%) was August, whereas December was the peak
month for those > 2 mm (2-3 mm, 45.3%) in length.

Dictynidae: Dictynids were most numerous form August to November. Most of
those 1-2 mm (73.0%) ballooned in August. The size classes < 1 and 2-3 mm comprised
13.8% and 13.2%, respectively.

Linyphiidae: September was the peak month for the 1-2 mm size class (89.0%) and for
total linyphiids. The < 1 mm class represented 9.6%. Males and females were captured in
similar numbers with a peak in September.

Table 2. -Ballooning spiders collected at Ellis (3 April-31 August 1980).

Month

Total

%of

Total

Apr.

May

June

July

Aug.

Erigonidae

380

1173

719

247

50

2569

41

Araneidae

111

483

297

516

79

1486

24

Tetragnathidae

37

173

90

119

38

457

7

Lycosidae

35

92

81

86

11

305

5

Thomisidae

19

21

53

152

19

264

4

Theridiidae

14

35

53

88

48

238

4

Dictynidae

3

8

9

119

98

237

4

Oxyopidae

15

22

38

81

27

183

3

Salticidae

79

32

12

27

19

169

3

Anyphaenidae

14

11

26

32

11

94

1

Linyphiidae

23

34

19

14

3

93

1

Philodromidae

8

10

16

9

7

50

1

Clubionidae

2

6

5

7

16

36

<1

Gnaphosidae

2

12

7

3

2

26

<1

Mysmenidae

-

10

3

2

~

15

<1

Mimetidae

1

1

6

6

14

<1

Pisauridae

-

6

2

1

—

9

<1

Uloboridae

-

1

—

1

—

2

<1

Hahniidae

-

1

—

—

—

1

<1

Total

742

2131

1431

1510

434

6248

116

THE JOURNAL OF ARACHNOLOGY

Clubionidae: The clubionids tended to be more numerous during the fall. The most
individuals were in the size class 1-2 mm (34.0%) which peaked in September. The second
largest grouping were those > 5 mm (32.0%), peaking in November. The 2-3 mm class
comprised 25.0%.

Anyphaenidae: August and September were the peak months for total anyphaenids
and also for the size class 1-2 mm (50.0%). Other size classes included: 2-3 mm (19.2%),
3-4 mm (14.1%), 4-5 mm (5.1%), and > 5 mm (1 1.6%).

Mysmenidae: All mysmenids captured were < 1 mm. Most were collected during
the fall.

Mimetidae: In this study, 22 mimetids were collected from July to December. Salmon
and Horner (1977) reported finding only one mimetid. The size class 1-2 mm contained
63.6% of the total. The other size classes were ca. evenly distributed. No males were
captured but females were collected in September, November and December.

Gnaphosidae: Gnaphosids were captured mainly in April and May with the 2-3 mm
size class containing 80.0% of the total (1-2 mm, 20.0%).

Uloboridae: This may be a new record, as we have not seen a previous report of this
family ballooning. Thirteen specimens were collected from July to November. One male
was captured in September. The most individuals were in the 1-2 mm class (61.5%).

Pisauridae: Three pisaurids were collected.

Hahniidae: This also appears to be a new record for ballooning activity. One specimen
was taken in June. Duffey (1956) reported that Hahnia nava Bl. was common in England
but did not disperse by aerial means.

Size Dispersal.— The percentages by size of six of the most common spiders (where the
immatures could be identified) collected in a suction trap (adults and immatures in-
cluded) are presented in Figure 2. The most individuals of Acanthepeira stellata
(Walckenaer) (As) at C.S. were in the size class 1-2 mm which peaked in September.
Those > 5 mm (all males, x = 6.9 mm) were captured in September. The size class 1-2
mm at Ellis included the most individuals in July. This was also the peak month for those
> 5 mm (all males, x = 8.4 mm). Over 85% of the total at each location were in the 1-2
mm size class.

Gea heptagon (Hentz) (Gh) was not very numerous at C.S. but the size classes < 1 and
1-2 mm included the most individuals which were collected mostly from August to

Fig. 2. -Distribution of the most abundant species by size class at College Station and the Ellis Unit
in Texas. (As = Acanthepeira stellata, Gh = Gea heptagon, Gf = Glenognatha foxi, T1 = Tetragnatha
laboriosa. Os = Oxyopes salticus, and Ms = Misumenops spp.).

DEAN AND STERLING-SIZE OF BALLOONING SPIDERS IN EASTERN TEXAS

117

December. More individuals were collected at Ellis where these two size classes peaked in
July. About 60% of the total at each location were < 1 mm in size. Most of the remaining
spiders were in the 1-2 mm size class.

The most numerous species at C.S. was Glenogmtha foxi (McCook) (Gf) with most
individuals in the smaller size class, < 1 mm, being captured in May and June. August and
September was the peak period for the 1-2 mm size class. June was the peak month at
Ellis for these size classes. Most of the individuals of G. foxi belonged in these two size
classes. Adult males (x = 1 .7 mm) were most abundant in September at C.S. and May at
Ellis. One adult female (x = 2.2 mm) was collected in April at each location.

Tetragnatha laboriosa Hentz (Tl) was the third most abundant species at C.S. The size
classes < 1, 1-2, and 2-3 mm peaked in May and June. May was also the peak month at
Ellis for the size classes < 1 and 1-2 mm. About 50% of the total at each location were <
1 mm. This size class corresponds to the second instar stage that emerges from the egg sac
ready to disperse (LeSar and Unzicker 1978). Few adults were collected (x of males = 5.5
mm at C.S. and 8.0 mm at Ellis; x of females = 9.2 mm).

Oxyopes salticus Hentz (Os) was the second most abundant species at C.S. with the
size classes < 1 and 1-2 mm peaking in September. Those 2-3 mm in size peaked in
November. The numbers at Ellis were much lower with the size classes 1-2 mm to 4-5 mm
peaking in July. Several males were collected with a x of 4.4 mm at C.S. and 4.7 mm at
Ellis.

Misumenops spp. (Ms) were collected in similar numbers at each location with those
spiders 1-2 mm in size the most numerous (ca. 75% of the total). The 1-2 mm size class
peaked in October. The size classes 1-2 mm to 3-4 mm at Ellis peaked in July. Four males
were collected at each location (x = 3.5 mm). Two females at Ellis measured 7.2 mm in
length.

DISCUSSION

It is well known that immature spiders disperse when they emerge from the egg sac,
primarily to avoid the cannibalistic tendencies of their siblings and to increase survival
chances by avoiding overcrowding (Turnbull 1973). “It is usually, but not invariably, very
young spiders that exhibit the aeronautic habit” (Comstock 1948). Horner (1975)
reported that 88% of the salticids collected in a Johnson-Taylor suction trap were imma-
ture (mostly early instar). Salmon and Horner (1977) stated that size is one of the major
restrictions to ballooning. Richter (1970) observed that aerial dispersal occurs generally in
young instars of eight Pardosa species, though different species have different dispersal
capacities.

These observations suggest that smaller species of spiders tend to balloon more fre-
quently than larger species and that early instars balloon more than later instars. Smaller
spiders, such as the erigonids, are caught ballooning more frequently than larger spiders in
other families. However, without knowing the relative abundance of individuals in these
families, the conclusion that small spiders balloon more readily than large spiders is
questionable. Obviously, we see a dramatic reduction in numbers of individuals correlated
with increasing size classes (Fig. 3). But, do spiders in earlier instars within a taxon
balloon more often than later instars?

We suggest that in nature some species exhibit little or no difference in the ballooning
rates of the various sizes and the difference in numbers of individuals in the various size
classes may largely represent normal mortality; i.e. mortality of the young result in fewer

118

THE JOURNAL OF ARACHNOLOGY

Fig. 3.— Numbers of spiders in their respective size ranges captured in a suction trap at College
Station and the Elhs Unit in Texas.

<1 2-3 4-5 6-7 8-9 10-11 12-13 14-15

SIZE (IN MM)

large individuals rather than reduced ballooning behavior in these later instars. In Table 3
we suggest that the number collected in each size class compared to the class with the
most individuals may represent percent survival. For example, at C.S. from the class
containing the greatest number of individuals of a species to the 4-5 mm class the survival
of Tetragnatha laboriosa would be 0.4%, Oxyopes salticus 0.3%, Chiracanthium inclusum
(Hentz) 12.0%, zndi Misumenops spp. 5.9%. In the long term the average number of adults
which must survive from the eggs deposited by a single female must be two (1 male and 1
female) for the population density to remain stable from year to year. Thus, for C.
inclusum, which averages ca. 76 eggs per female (Peck and Whitcomb 1970), survival of
12.0% of the ballooning stages to the 4-5 mm size class leaves ca. 9.1% additional genera-
tion mortality which can take place before 2.9% of the generation remains. This lower
survival rate is equal to 1 male and 1 female or an Ro (increase per female per generation)
of one. However, Peck and Whitcomb (1970) found ca. 10% mortality form egg to first
instar. This results in an Ro of ca. A A . Misumenops spp. was calculated to have an Ro of
14.0 based on data from Muniappan and Chada (1970). Early mortality of O. salticus and
T. laboriosa would leave them with an Ro of ca. 1.0. Since O. salticus and T. laboriosa
appear to have a lower survival rate but are more abundant in the suction trap and in
cotton (Dean et al. 1982) than C. inclusum, numbers in the trap in this study do not
appear to indicate equal ballooning rates among size classes.

Definitive tests of a hypothesis that instars balloon at equal rates should be based on
field collections by species and by classifying individuals into instars. Then with season-
long suction trap samples the yearly survival rates could be estimated and year to year
trends in population dynamics of various species could be more accurately determined.

Other factors that could influence the results of this experiment are: (1) the efficiency
of the suction trap in collecting various sizes of spiders, (2) elevation preference of
ballooning individuals of the various instars within a species, (3) ballooning time of
individuals in different size classes, and (4) actual mortality in the field.

DEAN AND STERLING-SIZE OF BALLOONING SPIDERS IN EASTERN TEXAS

119

Table 3.— Percentage survival from the size class with the most individuals to the other size classes
captured in a suction trap at College Station, Texas.

Size

class in mm

(n) % survival

Tetragnatha laboriosa <1 (909) to 1-2

(823)

90.5

2-3

(66)

7.3

34

(16)

1.8

4-5

(4)

0.4

>5

(5)

0.5

Oxyopes salticus 1-2 (2022) to 2-3

(257)

12.7

3-4

(23)

1.1

4-5

(6)

0.3

>5

(2)

0.1

Chiracanthium inclusum 1-2 (25) to 2-3

(24)

96.0

34

(4)

16.0

4-5

(3)

12.0

>5 imm.

(19)

76.0

>5 ad.

(13)

52.0

Misumenops sp. 1-2 (186) to 2-3

(18)

9.7

34

(7)

3.8

4-5

(11)

5.9

>5

(3)

1.6

However, for certain species of spiders, the catch-frequency may represent expected
abundance due to natural mortality rather than higher ballooning rates of smaller individ-
uals. If this size frequency is due to natural mortality, then continuous suction trapping
throughout the year may be used to obtain data useful in developing life tables and

predicting the dynamics of spider populations.

Of the most abundant species collected during this study, most ballooned in more than
one size class and many ballooned as adults. Thus, Comstock (1948) may be correct in
stating that it is usually young spiders that exhibit the aeronautic habit. However, our

data provide evidence that two other factors should be considered:

(1) differential rates

of ballooning between instars and (2) age dependent survivorship.

ACKNOWLEDGMENTS

We appreciate the cooperation of Billy Moore and Joe Berry of the Texas Department
of Corrections. The authors thanks Mrs. Pebble Lewis for assistance with sorting of some
samples and Norman Horner for use of the suction trap. We thank Lynn Carroll and
Gregg Nuessly for their review of this manuscript.

This research was supported in part by the National Science Foundation and the
Environmental Protection Agency through a grant (NSF GB 34718) to the University of
California and Texas A&M University. Continued support has been provided by EPA
grant no. R-806277-01 and the Texas Agricultural Experiment Station under project
H-2591. This research was conducted in cooperation with SEA, USDA, and the Texas
Department of Corrections.

120

THE JOURNAL OF ARACHNOLOGY

LITERATURE CITED

Bristowe, W. S. 1929, The distribution and dispersal of spiders. Proc. ZooL Soc. London, 43:633-657.

Comstock, J. H. 1948. The Spider Book. Doubleday & Co., New York, rev. by W. J. Gertsch, 729 pp.

Dean, D. A., W. L. Sterling and N. V. Horner. 1982. Spiders of eastern Texas cotton fields. J. Arach-
nol., 10:251-260.

Duffey, E. 1956. Aerial dispersal in a known spider population. J. Anim. EcoL, 25:85-111.

Freeman, J, A. 1946. The distribution of spiders and mites up to 300 feet in the air. J. Anim. EcoL,
15:69-74.

Ghck, P. A, 1939. The distribution of insects, spiders, and mites in the air. United States Dept. Agric.
Tech. Bull. 673, 150 pp.

Horner, N, V. 1975. Annual aerial dispersal of jumping spiders in Oklahoma (Araneae, Salticidae). J.
Arachnol., 2:101-105.

Johnson, C. G. and L. R. Taylor. 1955. The development of large suction traps for airborne insects.
Ann. Appl. Biol., 43:51-61.

LeSar, C. D. and J. D. Unzicker. 1978. Life history, habits, and prey preferences of Tetragnatha
laboriosa (Araneae: Tetragnathidae). Environ. EntomoL, 7:879-884.

Levi, H. W. 1980. The orb-weaver genus Mecynogea, the subfamily Metinae and the genera Pcc/zy-
gnatha, Glenognatha and Azilia of the subfamily Tetragnathinae North of Mexico (Araneae:
Araneidae). Bull. Mus. Comp. Zook, 149:1-74.

McDaniel, S. G. and W. L. Sterling. 1982. Predation of Heliothis virescens (F.) eggs on cotton in east
Texas. Environ. EntomoL, 11:60-66.

McDaniel, S. G., W. L. Sterhng and D. A, Dean. 1981. Predators of tobacco budworm larvae in Texas
cotton. Southwest. EntomoL, 6:102-108.

Millidge, A. F. 1980. The erigonine spiders of North America. Part 1. Introduction and taxonomic
background (Araneae: Linyphiidae). J. Arachnol,, 8:97-107.

Muniappan, R. and H. L. Chada. 1970. Biology of the crab spider, Mswmewops ce/er. Ann. EntomoL
Soc, America, 63:1718-1722.

Peck, W. B. and W. H. Whitcomb. 1970. Studies on the biology of a spider, Chiracanthium inclusum
(Hentz). Arkansas Agric, Exp. Stn, Bull. 753, 76 pp.

Platnick, N, 1. and M. U. Shadab. 1978. A review of the spider genus Mysmenopsis (Araneae, Mysmeni-
dae). American Mus. Novit., 2661:1-22.

Richter, C. J. J. 1970. Aerial dispersal in relation to habitat in eight wolf spider species (Pardosa,
Araneae, Lycosidae). Oecologia, 5:200-214.

Salmon, J. T. and N. V. Homer. 1977. Aerial dispersion of spiders in North Central Texas. J. Arach-
noL, 5:153-157.

Turnbull, A. L. 1973. Ecology of the true spiders (Araneomorphae). Annu. Rev. EntomoL, 18:305-
348.

Manuscript received November 1983, revised August 1984.

Poinar, G. O., Jr. 1985. Mermithid (Nematoda) parasites of spiders and harvestmen. J. Arachnol.,
13:121-128.

MERMITHID (NEMATODA) PARASITES
OF SPIDERS AND HARVESTMEN

George O. Poinar, Jr.

Division of Entomology and Parasitology
University of California
Berkeley, California 94720

ABSTRACT

Nematode parasites of spiders and harvestmen are restricted to members of the family Mermithi-
dae. A literature review shows that nematode parasitism of arachnids is worldwide and at least 51
species of spiders and harvestmen have been recorded as hosts of mermithid nematodes. Infected
spiders have varied habits and it is postulated that two types of parasite life cycles probably exist and
that the indirect life cycle (involving a paratenic host which falls prey to the arachnid) is probably the
common type.

INTRODUCTION

Representatives of the family Mermithidae are the only nematodes known to parasitize
spiders. Their effect on spiders is similar to that on other arthropod hosts, namely host
mortality at the time of parasite emergence.

The difficulty in rearing adult mermithids from postparasitic juveniles that have
emerged from parasitized spiders has prevented a systematic assessment of spider mer-
mithids. However, it is apparent that mermithid parasitism of spiders is widespread and
occurs in various habitats. The present work tabulates previous instances of these associa-
tions, adds some, and discusses the host parasite relationship. Reports of spider parasitism
by horsehair worms are not discussed here. The latter, commonly referred to as Gordius,
are not nematodes and belong to a separate phylum, the Nematomorpha. Early reports of
spiders parasitized by the horsehair worms may actually have involved mermithid nema-
todes and vice versa. The adult forms of both groups are similar superficially and may
have the same type of life cycle.

RESULTS AND DISCUSSION

Parasite identification.— Records of spider parasitism by mermithid nematodes are
summarized in Table 1 . E. Schlinger gave me mermithids that emerged from spiders in
New Guinea and New Zealand but they have not been included in the Table since the
hosts were not identified. Such is the case for a parasitized male clubionid from Papua,
New Guinea, that L. N. Sorkin had in his collection.

122

THE JOURNAL OF ARACHNOLOGY

c

a>

s

c/3

o

>

4:::

C

c3

<D

'Sh

c/3

C/3

C3

Ph

<D

O

C3

E

0)

c

'O

<u

s

(yi

X!

O

O

D

0^

3

cd

H

<D

O

c

a>

M

<u

1/3

03

TS

a>

C

(D

3

o

03

l-H

03

Cu

o

K

VO

00

0)

tyj

c«

D

l-i

a

c

cyj O

W3

(D C

(-1 D

Cl, PQ ro

O

S3

0) <-H

g, 8

(U

T3

c

03

f-4

03

C

S >

c -s

o

CL^

00

o^

D

S

S

o o

o 00

. On

S

E

o

o

C/3

»-l

CD

Cl,

£2

00

a>

Cl,

OJ

O

I I

>

o

CO

,C)

<D

X>

£

■o

o 00 00
• ^ ^

C/3 CN

S 3 s

c

■Jn Sh

Q ^ ^

E

E

o

o

E

E

o

o

CO

M

. O

CO Oh

<U Kn

<U CLh ^

a w S

M 3 C
^ ^ E

ro g O

S ffi

lo

00

■rt" l-H

00 tJ-

t-l ON

0,

.£

3 § QJ

'I E ^
c 5
3 2 3

O 03 "G

K K CQ

o

I ^ r-H

CTn *0 HCt
h-h 00 ON

L-l

o

M)

M ‘S 3
^ ^ X ^

CO

0

ON ‘O

,—1 00

»\

^ r

I-, 73

O) o-

1 ^

o (u

bX)

H-H 00

C -
° O

c

o

00

71-

2 c

" «

^ Gn

3

3

4D

0)

• f«H

00

C

O

>

73

r-1 C

^ o3

0) >-H

Cl o3

&I C

o -g

<D

E

o3

M

-t-»

00 o

0^

,-H ^

<U

00 s

CL>

Dh 3

Cl, N
•C3 e
c/3 D

bJD

«

c

73

o

o

O

00

73

G

E

<N

ctl

B

Os

+-<

0

H

43

0

.£f

«-H

c/3

a>

'c

M

73

(D

*♦-#

c/3

0

w

:c^

0

a

S

3

hC)

(U

•

c/3

C

o

>

60

<3

•»»«.»

-Cl

C3

Co

C

I

C

o
-,— >
G
0)
03

73

§

}-H

C3

.S

o

CIh

-2

s

o

Cl,

to

C

Cl

c

I -

c 5

^ 25

CS 3

c

o

G

G

G

>

O

CO

•H-H

hG

G

13:^

00

K

tU

-2

.C3

!L^

cu

s

O

C

-s;

<3

3

G

o3

E

o

3

3

X

'll

53

Co

O

G

O

-*-<

G

CD

CQ

73

G

cc3

I-,

G

C

'o

(Oh

50

3

'S

X

Co

5S

-C

G

O

G

G

G

e

CO «
Co ^

G O

S G

v2 C
^ G

G G

O O
G G

G

O

G

-Id

G

O

G

,14

G

O

G

GGGGG^t^^^^^^GGG'^

Oh 0

0,

0

Cu

a

CO ^

CO

X.

c«

c/3

CO

to

•S

.to

to

E/3

'1 •§

1

s

1

s

X

^3

!X

Sk

X.

i2

>2

2

2

G

O

G

,14

G

O
G G
,14 44
G G

5S

C)

•«»■>*

.CO

§■

Co

G ‘s:

1 1

(U

•S

0)

+-*

C/!)

2

Sh.

til

o

C

G-

co

Oh

CO

•2 Cd

S 's:

x:

^ S 'I o

S ^ S G
t3 c ^

3 ^ ^ :3

ia 9

CD ^

S- 3 K

^ ^ 2

•»>«4

-c 53 ca

C3

CD

)-,

O

iS

0)

E

o

euD

+->

G

O

S

43

o

o

U

Co

l-H S
0) Xh
^ -2
G :«

Oh Oh Oh
CO C/3 C/3

I

w

« >3 Q «

Co Co Co to

0000

Cj <D Cj o

^ ^ ?S ?S

kQ k,^

X.

tn

<3

to

O

<D

CD

44

o

3

2

o

<3

to

o

o

Co

■x>

c

o.

b

o

43

O

o

u

^ .

2 ^
CO

O

X.

D
G

s

^

G

'o

•n

43

«3

tin

X.

o

0,

^•2
O >*»

s,

^ o

43

.sf

'c

'G

O

O

o

•>->

43

M

G

73

O

O

a

2

2

<D

s

5S

•S-

S

a

c

<3

d s

CO ^

o

^ 2
■g «D

o ^

POINAR- NEMATODE PARASITES OF SPIDERS AND HARVESTMEN

123

73

(U

3

.s

c

o

u

(U

3

cd

H

<u

o

C

(U

i-i

(D

(Si

C3

T3

CD

S

C

«

nd

(U

c3

Ph

t/3

O

K

os

CD

X)

o

so

OS g

X

o

CD

CD

cd

<-i

ed

Ph

a

c«

'C

t-H ^

- £ -

> o >
O ?n O
W2 M

+J ■«-» -(_i

X c X
303

p^: IS Pi:

^ St

O On
£

O

bC

+->

c

o

c

o

-t-<

V5

s 2

>

o

CO

-(->

X

3

.’^

Co

s;

CD

Co

'§

:n.

<3

Ci,

>

O

CO

-*->

X

3

P^

Co

•*«*»

Co

s

CD

CO

Js.

*a

Pi-

co

c

Q c ^

><? X e

' « ^

3

a

CO

"■> •»->

Pd

!S^

I

-

X <1^

CD 3
M O
O 41^

fSb

.CO

.CO E?

'5 s;

.ca CD

2 t:
^ o
bo -p:

C3 C3
Co c<a

^ -2

C3 <3
Cl,

CD

cd

e

CD

X

CD

.Co

S!

3

O

4->

I-I

(D

£

w

X

CD

o

C3

CO

'§

V.

-B* ^

C3

-S

o s:

C3

£f .Pd

K 5:

<3 ds

Co 00

V.

4=* 43

CD

£

o

bO

+-*

e

o

S

S

C d

Co CO

C3 <3
Co Co

^ -2

!k V.

in':* r?
®H ®H

CO

CO

O

£

£

CO

eo

0

0

0

0

a^

ON

,

ir^

r--

H— 1

00

hoi

to

C"

0)

#-N

£

£

o

o

CD

Cu

/-V .£ O

£ c i

m

o

bO

M —

D -

CO Cd

X ,§ 3

3 o a

P^ E X

t-H

ON

fe

0

3

0

G

c

0

ON

T— H

c

0

00

CD

IT)

'd-

0

CD

0

00

0

CO

00

t-H

NO

00

CO

CO

m

to

I.H

00

00

l-H

CO

IH

00

00

ON

hoi

0

ON

M

D

"d

D

D

D

On

ed

£

• ^
X

X

'WH

l-H

d

X

'—H

1— 1

cd

X

.£

3"

Cd

£

0

bO

X

3

d

3"

cd

c£

2

’0

T3

3

D

X

3

0

x-»

Cd

H-H

3

cd

.£

0

X

l-H

0

X

-l-H

fH

D

-l-H

3

0

3

l-H

0

0

V5

Cd

X

H-H

IH

D

X

X

Pi

U

u

X

X/i

X

C/D

X

£

£

o

o

£

£

S -
. ^

CO On

CO CO
CO CO

0) O
-• CIh Qh

I 3 3

o

o

2 3 3

CD <l>

JX PP PP

CD CD

CD a

.£

X

X

l-H

o

PP lx IX PQ

bO
CD

o

-(-»

CO

•c

X Td

. . c 3

gD 3 Cd
X l-H Vh

C 3 cd

-S s s

O O O

Dh Ph Ph

CO CO
CO CO

CD D

l-H l-H

&, CD

.£ .£

3" 3"

o o

-*-> H-H

3 3

OD CD

ffl « X 2

'XS T3 )£ o\ 0\

ON ,-H —

>

o

CO

H-H

X

3

pp;

<3

Co

O

CD

s;

«3

X

U

C3

s

Cj

X S

'P5 5

c-2 CD

1^3

^ S o

s ■§ =

« Q X

bo Jh 3

^ ^ 3

3

Cd

£

D

Td

X

bo

s:

-2

ds

-p;

Pd

.<3

tp;

X

'o

Td

3

p^

<3

CD)

p;

3

o

3

X

3

3

e c

.<3 O

s: 3

-2

Pp 3

tit §

u

xA

u

2

3

<3

CD

CD

P3

Co

^ o

2 3

2 ^
bo C

X =s

3

O

H-H

3

D

X

l-H

a

.£

'o

Ph

.2

3

CD

S-

PD

di

Sh.

CD

V.

3 3
3 3

a

C/D

3

c &

Vj

^ CO

•»««<*

5

0

3

0 -3
3 3

d.

X

2

3

c 2

3 3

3 3 & 3

^ ^ .CO ^

0 0 3 0

3 3 3 3

X X X
3 3

3 3

Cd

T3

cd

S

u

3

o

H-H

M

CD

£

w

c

CD

3

O

d)

. 3

CD C3

CO Pi

dJ dS
Co Co

'§
V. V.

h2 4=*

®H ^

<

IX

X

3

D

X

CO

3

:§

.ds

*»--

■t-H

CD

Cj

3

3

3

X

o

"3

—

S3.

O

<

A

D

CO

3

.£ §

X PP

O -2
•»«»»

32 ds

a- ^
o o

3 -P)

3

So

3

.2

3

3

•l-H

bo

3

3

dJ

3

’»»«*

bo

3

2

3

Co

3

S3

.S3

<

A

D

£

a

X

X

o

D

X

£

bO _£

■s S

CO X
>.

<1^ :3
S

55 3

g s

2 3
^ CD

Co Co

3 3
S3 Pd
.S3 .PD

PP P3 P3

Ip; >3

S3 <3

S

X

D

l-H

H-H

, , Cd

<• ^
00

^ •* •r^

»— t ^

Co Co c2

3 s ^
83 .<D ^

•S-'B 0

P3 3 ‘3
•3 CD
pp; ^ a

CD, 3, Co

3

D

X

.X

H-h

3

+-H
I"— 4

3

Co

Q

Co

o

di

o

N

3

O

" £
< X
A CO

D 3

. '3
S3 ‘s;

CO

2

Co

3
CD

D

l-H

O

X

H

3

3

cd

T3

cd

3

cd

U

X

o

D

U

o

d>

o

N .N Cd

3 ^ *I-H

►C* 10

dj o
CO Co

3 X

C3- CO

3 3

c*^

3
*—

3

CD

. 3

X '*'*
CO 3

;c4

3 O
•p;

*-H ■*'* 22

dJ 3 o

. "PS

2 s:

-B <D>

CD P

3

cd

M

cd

3

O

X

3

cd

l-H

cd

3

‘o

X

X §

O H-H

CD CO
CD CC3

X

D

bJD

>>

3333

0000

H-H H-H H-J H-H

3 3 3 3
CD D D (D

« « X X

I-I l-H l-H l-H

Cd Cd Cd Cd

.£ £ S £

'o 'o ‘o ‘o

X X X X

2 2 -S 2

3 3 3 3

d) dj cj d)

’fxia •'xii •f»«*

Cd Pd a. Cd

Co Co Co Co

*»— » 'H— 'l-H *»—

■*— -l-H H— •»—

3- 3- a. 3-

3 3 3 3

Co Co to Co

•— -I--* •»—

d d d !d
3 3 3 3

3 3 3 3
2 2 2 2

X

X

CO

CO

,Co

.Cd

X

3

3

CO

£

£

.3

3

3

C

X

3

3

2

7—.

D

cd

3

D

X

o

X cd

Co

3

O

P

O

Co

3

D

cd

3

D

X

o

Id

3

♦-4

3

3

2

3

3

D

CO

3

D

l-H

D

o

X

X Q

CO Co

3 3

S eS

N i
H-H S

3 '■
D ^

X J

o PS ^

CO 3 C

3 Q ;

'3 2 c.

-2 ^
a c

2 2 s=

D

bO

bO

3 X
X X

S £

CO ctJ

^ U

D 1

^ X

Co

a o
3 2
^ 2
2 3

^ 3

52 3

g £
1?^

N

124

THE JOURNAL OF ARACHNOLOGY

Tlie earliest reported incidence of mermithid parasitism of spiders was by Hoppe in
1796. No attempt was made to describe the parasite. In 1833, Walckenaer cited a Filaria
from Aranea diadema. At that time, the name Filaria was used as a collective genus name
for representatives of various groups, especially the larger parasitic worms, such as repre-
sentatives of the Mermithidae. It had no taxonomic significance. Kryger (1910) also cited
Filaria from Lycosa sp. and Zora maculata. In 1819, Rudolph! described mermithids he
obtained from Phalangium cornutum and P. opilio as Filaria truncatula. However, his
description was very brief and based on general characters found in the postparasitic
juveniles. Since adult characters are needed for proper taxonomic placement, this must be
cited as a species inquirenda. Also included in this category are Filaria phalangii Halde-
man 1851 and Filaria lycosae Haldeman 1847.

Later, the genus Mermis was used in a broad sense to represent members of the family
Mermithidae. It and the frequently used binomial, Mermis albicans, were assigned to
a range of species collected from arthropods. However, as in the case of Filaria, these
names were used in a collective sense and either lacked a description or the description
was so general that it was useful only to family level. Thus the citations listed in Table 1
for Menge (1866), Holm (1941), Bristowe (1931; 1941), Montgomery (1903), Kastner
(1928) and Bertkau (1888) when Mermis sp. ox Mermis albicans is mentioned must stand
as species inquirendae. Kaston (1945) cited Agamermis decaudata as a parasite of Pardosa
sp. and Phidippus clarus. Those nematodes were identified by G. Thorne, basically a plant
nematologist. Since he was probably examining juveniles, it is doubtful that a specific
designation could have been possible. Also, A. decaudata is a parasite of Orthoptera
and has not otherwise been reported from spiders. It is my contention that this was a
misidentification.

Reports of a Hexameris sp. parasitizing Xysticus deichmanni, X. funestus and Pardosa
glacialis (Kaston, 1945) (Leech, 1966) are also not exact since postparasitic juveniles were
examined and only rarely can a genus be determined from these stages. More recently,
Rubtsov described Amphimermis pardosensis from Pardosa riparia (1977), Arachnomer-
mis araneosa from Pardosa sp. (1978) and Arachnomermis dialaensis from Diaea dorsata
(1980). The descriptions of these species are based on postparasitic juveniles and again,
their true identity remains unknown. From what we now known about mermithid
morphology and systematics, aU of the above mentioned mermithids from spiders have no
systematic position in the classification of the Mermithidae and might well be placed in
the collective genus, Agamomermis, erected to receive mermithids that could not be
placed in existing genera (Poinar and Welch, 1981).

The only completely described mermithid parasite of spiders is Aranimeris aptispicula
Poinar and Benton (in press). The description is based on adult characters comparable
with those of existing genera.

Effects of parasitism.— External symptoms of mermithid parasitism of spiders usually
are associated with the size and shape of the host’s body. A swollen abdomen is a com-
mon symptom and Leech (1966) noted that parasitized P. glacialis had a lopsided or
greatly enlarged opisthosoma, an alterexL epigynum, malformed palpi, legs that were
shorter and thicker than normal and poorly developed or absent male secondary sexual
characteristics. It is possible to see the coils of the parasite through the host’s integument
since the mermithid usually occupies the entire abdomen and occasionally the cephalo-
thorax. Parasitic castration was noted by Bertkau (1888) in a Tarentula inquilina attacked
by a mermithid.

Infection signs generally start with a reduction or absence of the digestive gland. In
extreme examples, other organs may also be reduced. Leech (1966) commented that

POINAR-NEMATODE PARASITES OF SPIDERS AND HARVESTMEN

125

parasitism of P. glacialis resulted in the loss of the main prosomatic muscles, the entire
digestive system and the entire reproductive system.

Behavioral changes in parasitized spiders have also been noted. Leech (1966) (and
personal correspondence) mentioned that some infected individuals of P. glacialis were
sluggish and did not attempt to escape when approached. During the week before the

Fig. 1. -Coils of Aranimeris aptispicula Poinar and Benton filling the abdominal cavity of the
spider, Tmarus [probably angulatus (Hentz)]. (Photo by the author; specimen from C. Benton).
(Mag. X 10).

Fig. 2. -A postparasitic juvenile mermithid that has Just emerged from its phalangid host, a male
Protolophus sp. (Photo by Pat Craig). (Mag. x 5).

126

THE JOURNAL OF ARACHNOLOGY

parasites emerged the spiders ceased feeding but drank a lot of water. This attractiveness
to water was noted in infested A. aptispicula which would come out of neighboring
woods and fields to find a source of water.

Kaston (1945) presented some evidence that mermithids retard the development of
their spider hosts.

Incidence of infection.— Most of the reports of mermithid parasitism of spiders men-
tion only a single incidence of infection. Leech (1966) noted that 1% of the Pardosa
glacialis he collected were parasitized and that most were females. He mentioned that the
rate might have been higher since the infection is very hard to detect in young spiders.

Color of parasites.— Certain species of mermithids can be recognized by their color and
both Hal deman (1851) and Leidy (1856) mentioned that upon emergence, the nematodes
were pale pink to reddish. The former author noted that the color changed to yellowish
after the specimen was heat-killed. This color change was also noted by Poinar and
Benton (in press) in A. aptispicula. Emerging individuals were pinkish, yellowish and
occasionally green, but all became white after some days in water. The initial color may
have been acquired from the host.

Life cycle of mermithids attacking spiders.— Although the life history of no spider
mermithid is completely known, Aranimeris aptispicula is one that probably possesses an
indirect life cycle. Its occurrence in a wide range of spiders suggests this. In this type of
development, the females deposit eggs in an aquatic habitat. The eggs are ingested by
immature insects and the infective stage mermithid hatches, penetrates the gut wall,
invades the parenteral tissues of the host and then enters dormancy. Thus when the host
matures, it carries the parasite. When one of these paratenic hosts falls prey to a spider
and is eaten, the nematode becomes active, enters the spider’s hemocoele and resumes
development. Such a life cycle has been shown to occur in Pheromermis pachysoma, a
parasite of yellowjackets (Poinar et al. 1976).

However, from the descriptions of some postparasitic juvenile mermithids that
emerged from spiders, it is obvious that at least one other mermithid species attacks
spiders in North America. This species could well have a direct cycle, that is, one where
the infective stage emerging from the egg enters a young spider by direct penetration
through the integument and initiates development. A second host is not involved in such
a cycle.

Type of spiders attacked.— Spiders that are attacked by mermithids demonstrate a
wide range of behavior and habitat preference. Thus, it is not just ground-stratum hunters
that show mermithid parasitism but also orb web weavers, aquatic forms, plant climbers,
and even crab spiders that catch insects attracted to flowers. Food preference for parasi-
tized spiders is not restricted to any particular group of insects. It is interesting to note
that all spiders found parasitized would have an opportunity to feed on adult insects
which possess an aquatic larval stage (e.g. Chironomidae, Culicidae, Trichoptera). Such
insects would make ideal paratenic hosts.

Recommended handling of mermithids.— Upon noticing the emergence of a mermithid
from a spider host, the investigator should place the parasite in a small amount of water
in a glass container with a layer of sand in the bottom. It should be left until it has
molted (a single molt composed of the final two shed cuticles) to the adult stage which
normally occurs within a month. During this time, the water should be changed daily to
avoid the accumulation of fungi which can kill the parasite. Adult stages can be recog-
nized by the appearance of the vulva in the female and the spicules (copulatory organs) in
the male [see Poinar (1983) for figures of the appearance and location of these struc-
tures] .

POINAR-NEMATODE PARASITES OF SPIDERS AND HARVESTMEN

127

The adults should be killed by placing them in water heated to 50-60°C. After death,
they can be fixed in 3% formalin or 70% alcohol for taxonomic studies. If the living
worms are placed directly into fixative when they emerge from the spiders, further
taxonomic studies will be prevented.

ACKNOWLEDGMENTS

The author would Hke to express his sincere appreciation and gratitude to Pat Craig for
his assistance in the preparation of this manuscript. Thanks are also extended to the
following individuals for submitting material cited in the text; R. G. Holmberg, L. Vin-
cent, M. Doucet, G. Landau, C. L. B. Benton, Jr., E. L. Schlinger, B. Cutler, G. L. Miller,
J. C. Cokendolpher, G. Lowe and L. N. Sorkin.

LITERATURE CITED

Bertkau, Ph. 1888. UeberAfermw in Tarentula inquilina und die durch den Parasiten bedingte Sterilitat
des Wirthes. Verhandl. naturhistor. Vereines Rheinl., 45:91-92.

Bristowe, W. S. 1931. Notes on the biology of spiders. V. Theridion ovatum Clerck, its habits and
varieties. Ann. Mag. Nat. Hist, 8:466469.

Bristowe, W. S. 1941. The comity of spiders. The Ray Society, London, vol. 2.

Haldeman, S. S. 1847. Lycosa scutulata and his parasite, Filaria. Proc. Amer. Phil. Soc., 4:356.
Haldeman, S. S. 1851. Invertebrates. Pp. 48-196, In: Outlines of General Zoology (S. Baird, ed.) R.
Garrigue, New York.

Holm, A. 1941. Uber Gynandromorphismus und Intersexualitat bei den Spinnen. Zool. Bidrag Fran
Uppsala, 20:397416.

Hoppe, D. H. 1796. Entomologische Taschenbuch fur Anfanger und Liebhaber diesen Wissenschaft auf
das Jahr 1796. Regensburg. 16 pp.

Kaston, B. J. 1945. Notes on nematode parasites of spiders. Trans. Connecticut Acad. Sci., 36:241-
244.

Kastner, A. 1928. Opiliones (Weberknechte, Ranker) 51 pp. In F. Dahl: Die Tierwelt Deutschlands.
Part 8. Gustav Fischer, Jena.

Kryger, J. P. 1910. Snyltere i Edderkoppeaeg. Entom. Medd., 3:257-285.

Leech, R. E. 1966. The spiders (Araneida) of Hazen Camp 81°49*N, 71°18'W. Quaest. Entomol.,
2:153-212.

Leidy, J. 1857. A synopsis of Entozoa and some of their Ecto-conginers, Proc. Acad. Nat. Sci. Phila-
delphia, 8:42-58.

Menge, A. 1866. Preussische Spinnen 1. Schr. d. Naturf. Ges. Danzig, 1:37.

Montgomery, T. H, 1903. Studies on the habits of spiders, particularly those of the mating period.
Proc. Acad. Nat. Sci., Philadelphia, 55:80-90.

Muller, H. G. 1983. Ein Mermithide als Parasitoid von Coelotes inermis (L. Koch 1855) (Arachnida:
Araneae: Agelenidae). Entomol. Zeit., 93:358-360.

Parker, J. R. and M. T. Roberts. 1974. Internal and external parasites of the spidti Pardosa nortensis
(Thorell). (Araneae: Lycosidae). Bull. Brit. Arachnol. Soc., 3:82-84.

Pfeifer, H. 1956. Zur Okologie und Larvalsystematik der Weberknechte. Mitt. Zool. Museum Berlin,
32:59-104.

Poinar, Jr. G. O. 1983. The Natural History of Nematodes. Prentice Hall, Inc., New Jersey. 323 pp.
Poinar, Jr. G. O., R. S. Lane and G. M. Thomas. 1976. Biology and redescription of Pheromermis
pachysoma a parasite of yellowjackets. Nematologia, 22:360-370.

Poinar, Jr., G. O. and H. E. Welch. 1981. Parasites of invertebrates in the terrestrial environment.

Review Adv. Parasitol. Ed. W. Slusarski, Polish Scientific Publishers, Warsaw. 947-954.

Poinar, Jr., G. O. and C. L. B. Benton. Jr. (In Press). Aranimeris aptispicula n. gen., n. sp. (Mermithi-
dae: Nematoda), a parasite of spiders (Arachnida: Araneida), System. Parasitol.,

Rubtsov, I. A. 1977. New species of mermithids from spiders and earwigs. In the book. Fauna of
Siberia. “Nauka”, Novosibirsk 16-22. (In Russian).

128

THE JOURNAL OF ARACHNOLOGY

Rubtsov, I. A. 1978. Mermithids: Classification, importance and use. “Nauka”, Leningrad. 207 pp. (In
Russian).

Rubtsov, I. A. 1980. New species of mermithids from arthropods. In the book: Helminthes of insects.
“Nauka”, Moscow 95-102. (In Russian).

Rudolphi, C. A. 1819. Entozoorum Synopsis cui accedunt mantissa duplex et indices lociepletissimi.
Berlin. 811 pp.

Siebold, C. T. E. von. 1843. Ueber die Fadenwurmer der Insekten. Entomol. Zeit., 4:78-84.

Siebold, C. T. E. von. 1848. Ueber die Fadenwurmer der Insekten. Entomol. Zeit., 9:290-300.

Siebold, C. T. E. von. 1854. Ueber die Fadenwurmer der Insekten. Entomol. Zeit., 15:103-121.
Stipperger, H. 1928. Biologie und Verbreitung der Opilioniden Nordtirols. Arb. a. d. Zool. Institut
Innsbruck, 3:1-63.

Unzicker, J. D. and G. L. Rotramel. 1970. A new parasite record for Opiliones (Arachnida). Trans.
Illinois Acad. Sci., 63:223-224.

Vincent, L. (In press). Pathogens and parasitoids of the fossorial mygalomorph spider Atypoides
riversi O.P.-Cambridge (Antrodiaetidae: Araneae) of various age and size classes. Proc. 9th Int.
Arachnol. Congress, Panama, 1982.

Walckenaer, C. A. 1833. Une Filaire parasite d’une Epeira diadema. Ann. Soc. Entomol. France,
2:74.

Manuscript received May 1 984, revised August 1984.

Lee, R. E,, Jr. and J. G. Baust. 1985. Low temperature acclimation in the desert s,^\6.Qr:,Agelenopsis
aperta (Araneae, Agelenidae). J. ArachnoL, 13:129-136.

LOW TEMPERATURE ACCLIMATION IN THE DESERT SPIDER,
AGELENOPSIS APERTA (ARANEAE, AGELENIDAE)

Richard E. Lee, Jr.^ and John G. Baust

Department of Biology
University of Houston
Houston, Texas 77004

ABSTRACT

Agelenopsis aperta (Gertsch) inhabits desert grasslands and lava beds in the southwestern U.S.A.
The capacity of this species to cold-harden was assessed by exposing second generation laboratory-
reared specimens to an artificial low temperature cycle simulating the “summer-autumn-winter”
transition. Low temperature acclimation had no effect on whole body supercooling points, freeze
tolerance or rates of oxygen consumption. Elevated levels of cryoprotectants were not detected using
high performance liquid chromatographic techniques. Cold tolerance was similar between males,
females and immatures. Exposure to temperatures immediately above the whole body supercooling
point caused no apparent injury. It is hypothesized that movement into protected overwintering
microhabitats may obviate the necessity for the evolution of seasonal mechanisms of cold-hardening in
A. aperta.

INTRODUCTION

Cold-hardening refers to physiological mechanisms by which organisms acquire en-
hanced tolerance to low temperature. Few workers have examined the overwintering
biology of arachnids, despite the fact that for temperate species winter represents a
significant portion of the life span.

Comparable studies with terrestrial insects have revealed two basic patterns of cold-
hardening (Salt 1961). One group tolerates the formation of extracellular ice within the
body, whereas, other species are freeze susceptible or freezing intolerant. The latter a-
void the lethal effects of freezing by lowering the temperature at which the spontaneous
freezing (nucleation) of body water occurs. This value is termed the supercooling point
and is experimentally determined by detecting the released latent heat of fusion as body
water freezes. For intolerant species, the supercooling point represents the temperature at
which body tissues freeze and the lower lethal limit.

Exposure to low environmental temperatures often serves as a proximal cue triggering
the seasonal acquisition of increased cold tolerance (Baust 1981, Baust and Lee 1982).
For both freeze susceptible and freeze tolerant arthropods cold-hardening is often associ-
ated with the accumulation of low molecular weight sugars (glucose and trehalose) and
polyhydric alcohols (glycerol, sorbitol and mannitol). These cryoprotectants are believed

‘Current mailing address: Department of Zoology, Miami University, 1601 Peck Blvd., Hamilton,
Ohio 45011.

130

THE JOURNAL OF ARACHNOLOGY

to afford protection in a number of ways including the colligative depression of super-
cooling and melting points, a reduction in the rate and amount of tissue ice formation
and by the prevention of injurious levels of cellular dehydration (Meryman 1974).

The purpose of this investigation was to assess the capacity of Agelenopsis aperta
(Gertsch) to cold-harden. This species is a funnel-web weaver of the family Agelenidae.
The specimens used in this study were laboratory-reared offspring of spiders collected
from desert grassland and lava bed habitats at an elevation of 1600m in south-central New
Mexico. In this area winters may be severe with heavy snowfall and extended periods of
sub-zero temperatures. Advanced juveniles overwinter from late October until activity
resumes in March (S. E. Riechert, pers. comm.; Riechert 1974). This species population
has been extensively studied with respect to habitat selection and web-site distribution by
Riechert et al (1973), Riechert (1974), Riechert and Tracy (1975), Riechert (1976) and
Riechert and Gillespie (in press).

Second generation laboratory -reared specimens were exposed to an artificial “summer-
autumn-winter cycle” by gradually exposing them to decreasing temperature during a 72
day laboratory experiment. Periodically low temperature tolerance was evaluated by
determining whole body supercooling points, freeze tolerance versus susceptibility,
cryoprotectant titers and rates of metabolism.

MATERIALS AND METHODS

Initial stocks of Agelenopsis aperta were collected from desert grasslands in south-
central New Mexico, U.S.A. (Riechert et al. 1973). All specimens used were juvenile
second-generation laboratory -reared for their entire lives at temperatures between 24 and
28°C at the University of Tennessee. Their age was similar to those of spiders overwinter-
ing in natural habitat. The spiders were mailed to the University of Houston and allowed
to acclimate to 26 ± 1°C for three weeks prior to experimentation. Each week 2-3 live
Musca domestica adults and droplets of water were provided for each spider. However,
spiders were not fed for five days prior to supercooling point, cryoprotectant and respira-
tion rate determinations. Spiders were fasted to remove potential nucleating agents
in the food which might result in artifically high supercooling points.

Spiders were cold-acclimated using 5°C temperature steps decreasing from 20°C to
0°C over a 72-day period. Details of the acclimation schedule are shown in Eig. 1. Groups
of 4-6 males and females (e.g. females and immature individuals) were removed periodic-
ally and tested for supercooling point, cryoprotectant levels and freeze tolerance using
methods described by Lee and Baust (1981). A cooling rate of approximately l°C/min
was used for supercooling point determinations. The cryoprotectant levels were analyzed
using high performance Hquid chromatography with a radially compressed amine modif-
ied silica column (Hendrix et al 1981, Lee et al 1983). Freeze tolerance or susceptibility
was assessed by allowing the body temperature of the spider to return to the temperature
of its supercooling point after the release of the heat associated with the freezing of body
water. Post-freezing survival was evaluated after 24 hr at 5°C.

Oxygen consumption was determined using a microrespirometry system described by
Lee and Baust (1982). Each respirometer consisted of a 5cc disposable plastic syringe
depressed to 2cc with a 20 jul micropipette. Two groups of eight females were acclimated
for 12 days without food at 26 or 10°C prior to oxygen consumption determinations.
Each spider was placed in a separate microrespirometer and equilibrated for 30 min at
15°C followed by oxygen consumption measurements during the next two hours. The

LEE AND BAUST-COLD TOLERANCE mAGELENOPSIS APER TA

131

same individuals were then transferred to 10°C and respiration measurements were
repeated as described above. All determinations were made between 1300-1800 hr.
Oxygen consumption is expressed in nanoliters (nl) per mg live weight per hour.

RESULTS

Survival was high (> 90%) during the 72-day period of low temperature acclimation in
the laboratory and did not differ between cold-acclimated spiders and ones held continu-
ously at 26°C. Fewer males were available for study and as a result were tested only
during the first 30 days of the acclimation schedule. At no time during acclimation did a
spider survive freezing of body fluids. However, cooling to temperatures immediately
above the supercooling point caused no apparent injury in any spider, including individ-
uals maintained at 26° C without cold acclimation.

Supercooling points of males and females were similar and remained unchanged during
the first 30 days of acclimation to low temperature (Fig. 1). The supercooling point
represents the lower temperature limit for survival under natural conditions for organisms
intolerant of freezing. Therefore, these data indicate that males and females have similar
levels of cold tolerance. Although in most samples the supercooling point values were
tightly clustered in the -8 to -12°C range, some individuals extended supercooling to as

9

lu

Qt

<

CE

ly

0.

S

UJ

Acclimation Schedule

• Sampling Time

O

e

0

Z

o

CL

-6

O

O

O

O

flC

lU

CL

CO

-10

-16

A

I

0

♦

I

j,

10

■ Females
□ Males

* * *

20 30 40 70

DAYS

Fig. l.-Top: Laboratory low temperature acclimation schedule foi Agelenopsis aperta. Bottom:
Comparison of supercooling points for males and females of A. aperta determined at intervals during
the acclimation schedule (X ± SE).

132

THE JOURNAL OF ARACHNOLOGY

low as -18.2°C before spontaneous nucleation occurred. Even after 72 days of low
temperature exposure, no change in the lower limit of cold tolerance as judged by whole
body supercooling points was observed (Fig. 1).

Homogenized spider extracts were analyzed using high performance liquid chromat-
ographic techniques to determine whether low temperature acclimation had an effect on
endogenous levels of cryoprotectants. Elevated levels of typical cryoprotectants including
glycerol, sorbitol, trehalose, glucose, mannitol and erythritol were not detected. In fact,
no glycerol, trehalose or erythritol were identified in chromatograms despite detection
limits of greater than 0.05 jUg per mg of live weight.

Two groups of eight female A. aperta were acclimated to 10 or 26° C for 12 days.
Measurement of oxygen consumption at 10 or 15°C revealed no significant differences
between the groups. The combined consumption rates (x ± SEM) at 10 and 15°C were,
respectively, 112 ± 9 and 171 ± 15 nl/mg/hr.

DISCUSSION

Cold-hardening in spiders appears to be restricted to the avoidance of tissue freezing
by depressing the whole body supercooling point. Kirchner (1973) found no evidence for
freeze tolerance in studies conducted during the winter with 50 species belonging to 15
families. The data of other investigators are consistent with his observations (Table 1).
Regardless of the length of cold acclimation, freeze tolerance was never observed in A.
aperta. Cooling to temperatures immediately above the supercooling point, however,
produced no apparent injury. In general it appears that spiders tolerate exposure to low
temperature as long as the limit of supercooling of body fluids is not exceeded. A single
exception is the tropical species, Agelena consociata which dies after a few hours of
exposure to 0°C (Kirchner 1973).

In certain spiders considerable increases in freeze avoidance by supercooling may be
observed on a seasonal basis. Philodromus immatures lower the supercooling point from
-6.2°C in summer to -26.2°C in winter (Duman 1979). Overwintering eggs have the
greatest capacity for freeze avoidance with supercooling points extending below -30°C
(Table 2). In contrast, overwintering stages of some species lack specific mechanisms of
changing cold-hardiness as evidenced by the similarity between summer and winter
supercooling points. In Table 2 five species have summer and winter values within one
degree of each other and none below -6.8° C.

Supercooling points for A. aperta were relatively constant regardless of the tempera-
ture or duration of low temperature exposure. Apparently this species lacks the capacity
for cold-hardening despite the fact that it is exposed to sub-zero temperatures in the field.
A second possibility is that low temperatures alone are not sufficient to induce hardening
in this species. For example, seasonal changes in daylength provide a reliable cue for
cold-hardening (Duman and Horwath 1983). This alternative is perhaps less likely since
the four other agelenids listed in Table 2 apparently, also, lack the capacity for cold-
hardening. For phylogenetic reasons the Agelenidae as a group may not have evolved
biochemical or physiological mechanisms of seasonally enhancing low temperature
tolerance. Seasonal field collections which assess cold tolerance in the natural habitat are
needed to definitively resolve this question for^. aperta.

Physioloigcal mechanisms regulating the supercooling point are poorly understood.
Although glycerol accumulation has been associated with cold-hardening in several species
(Kirchner and Kestler 1969, Duman 1979) other common cryoprotectants, such as
sorbitol, glucose and trehalose have not been identified in overwintering spiders.

LEE AND BAUST-COLD TOLERANCE IN AG ELENOPSIS APERTA

133

Table 1.- Summary of cold-hardiness in overwintering spiders. Blanks in table indicate that specific
information was not provided in the original article or that a given parameter was not investigated.
Mean supercooling points are for spiders collected during the winter unless noted otherwise. Unless
specifically stated the overwintering stage of spiders studied by Kirchner (1973) were adults or im-
matures. All species are intolerant of tissue freezing.

Taxon

Study Site

Overwintering

Stage

Supercooling
Point (°C)

Source

Agelenidae

Agelenopsis aperta

New Mexico,
U.S.A.

(Lab reared)

Adult,

Immature

-8.0

to

-12.0

Present study

Cicurina cicurea

W, Germany

-6.7

Kirchner, 1973

Coelotes terrestris

W. Germany

-6.2 (Winter)
-5,3 (Summer)

Kirchner, 1973
Kirchner, 1973

Histopona torpida

W. Germany

-6.5 (Winter)
-6.0 (Summer)

Kirchner, 1973
Kirchner, 1973

Tegenaria sp.

W. Germany

-8.0

Kirchner, 1973

Amaurobiidae

Amaurobius femes trails

W. Germany

-6.6

Kirchner, 1973

Araneidae

Araneus cornu tus

W. Germany

Adult,

Immature

-23 (Winter)

-8 (Summer)

Kirchner and
Kestler, 1969

Argiope aurantia

Illinois,

U.S.A.

Spiderling

Riddle, 1981

Meta menardi

W. Germany

4.0

Kirchner, 1973

Singa nitidula

W. Germany

-21.4

Kirchner, 1973

Clubionidae

Clubiona sp. *

Indiana,

U.S.A.

Immature

-15.4 (Winter)
-9.2 (Warm-
acclimated)

Duman, 1979

Clubiona phragmitis

W. Germany

-16.1

Kirchner, 1973

Eresidae

Eresus niger

W. Germany

-16.6

Kirchner, 1973

Linyphiidae

Bolyphantes index^

Norway

Adult

-15.2

Husby and
Zachariassen, 1980

Floronia bucculenta

W. Germany

Egg

-30 (Summer
and Winter)

Schaefer, 1976

Linyphia sp.

W, Germany

Egg

-25.2 to
-33.8

Kirchner, 1973

Lycosidae

Pardosa lugubris

W. Germany

-6.8 (Winter)
-5.8 (Summer)

Kirchner, 1973

Nesticidae

Nesticus cellulanus

W. Germany

4.7 (Summer
and Winter)

Kirchner, 1973

Philodromidae

Philodromus sp.‘

Indiana,

U.S.A.

Immature

-26.2 (Winter)
-6.2 (Summer)

Duman, 1979

Philodromus sp.

W. Germany

-21.5

Kirchner, 1973

Tetragnathidae

Pachygnatha clercki

W. Germany

-5.8

Kirchner, 1973

Theridiidae

Ten tana castanea

W. Germany

-9.5

Kirchner, 1973

T. triangulosa

W. Germany

-10.9

Kirchner, 1973

Theridion deuticulatum

W. Germany

-11.4

Kirchner, 1973

T. n Ota turn

W. Germany

-26.1

Kirchner, 1973

T, tepidariorum

W. Germany

-8.2

Kirchner, 1973

‘ Thermal hysteresis factors present.

134

THE JOURNAL OF ARACHNOLOGY

Thermal-hysteresis factors are present in the hemolymph of several overwintering
spiders (Duman 1979, Husby and Zachariassen 1980, Duman and Horwath 1983). These
proteins produce a thermal hysteresis between the freezing and melting points and are
similar in this respect to antifreeze proteins in polar fishes and some insects. Recently,
Zachariassen and Husby (1982) have hyposthesized that these factors may function to
enhance supercooling capacity by absorption to the surface of embryo ice nuclei and,
thereby, prevent additional growth of the ice crystal. Additional studies are needed in
order to determine the general significance of thermal-hysteresis factors in relation to
mechanisms of cold-hardening in spiders.

Terrestrial arthropods from thermostable habitats generally lack the capacity for
compensatory acclimation of respiration rate, whereas ones from variable habitats may
possess considerable potential for adjustment (Hazel and Prosser 1974). Metabolic studies
related to overwintering in spiders have produced varying results. A seasonal decrease in
oxygen consumption has been reported for Pisaura mirabilis (Dondale and Legendre
1971) and Araneus cornutus (Kirchner 1973). This decrease may represent a state of
winter diapause similar to that commonly observed in insects (Dondale and Legendre
1971). On the other hand, no evidence for seasonal acclimation of respiration rate was
observed in spiderlings of Argiope aurantia, a species which overwinters in exposed sites
above the ground (Riddle and Markezich 1981). However, laboratory studies suggest a
potential for metabolic compensation since low temperature acclimation elevated the
respiratory rate of spiderlings with respect to animals held at higher temperatures. Several
species studied by Anderson (1970) respond to acclimation at 30°C by reducing oxygen
consumption, but did not respond to low temperature acclimation at 10°C by increasing
the metabolic rate. The results of our study suggest a lack of compensatory change in
A. aperta.

The capacity for cold-hardening and the limit of cold temperature tolerance are often
closely correlated with the overwintering microhabitat. Spiders which overwinter above
the ground on vegetation or beneath the bark of dead standing trees possess the greatest
cold tolerance (Kirchner 1973). Cave dwelling or species overwintering on the ground or
in the soil are less tolerant with supercooling points between -4 and -8*^0 (Table 2).
Furthermore, the lower lethal temperature for these species often remains essentially the
same throughout the year. Edgar and Loenen (1974) suggest that the more northerly
distribution of Pardosa lugubris relative to congeneric species is possible due to its pro-
tected hibernaculum in leaf litter.

Agelenopsis aperta is a ground dwelling species whose funnel retreat may extend
25-50cm into cracks in the substrate (Riechert etal. 1973). Agelenids, in general, retreat
into their funnels during periods of unfavorable environmental temperature (S. E.
Riechert, pers. comm.). The scorpion, Paruroc tonus aquilonalis, also inhabits desert
grasslands in New Mexico and similar to A. aperta, retreats into burrows in the ground
(Riddle and Pugach 1976). In January with low ambient temperatures of -14°C, burrow
temperatures for P. aquilonalis at 2 and 4 cm depth were respectively -12 and -6°C. These
observations suggest that by moving only a few cm into the substrate A. aperta would be
able to avoid environmental temperatures approaching its supercooling point.

The selection of a protected overwintering microhabitat obviates the need for seasonal
mechanisms enhancing cold tolerance. The combination of a normal (i.e. non-cold-
hardened) supercooling point of -8 to -10°C and movement into a moderate thermally
buffered hibernaculum is likely sufficient protection for a species inhabiting southern
temperature regions. In turn, this provides an explanation for the lack of mechanisms of

LEE AND BAUST-COLD TOLERANCE m AGELENOPSIS APERTA

135

cold-hardening and metabolic compensation in A. aperta and other species overwintering
in protected hibernacula.

ACKNOWLEDGMENTS

We thank Susan E. Riechert for providing the spiders used in this study. John F.
Anderson, Wayne A. Riddle, Susan E. Riechert and Ann L. Rypstra provided useful
suggestions on the manuscript.

LITERATURE CITED

Anderson, J. F. 1970. Metabolic rates of spiders. Comp. Biochem. Physiol., 33:51-72.

Baust, J. G. 1981. Biochemical correlates to cold hardening in insects. Cryobiology, 18:186-198.

Baust, J. G. and R. E. Lee, Jr. 1982. Environmental triggers to cryoprotectant modulation in separate
populations of the gall fly, Eurosta solidagenis. J. Insect Physiol., 28:431-436.

Dondale, C. D. and R. Legendre. 1971. Winter diapause in a Mediterranean population oiPisaura mira-
bilis (Clerck). Bull. Brit. Arachnol. Soc., 2:6-10.

Duman, .1. G. 1979. Subzero temperature tolerance in spiders: The role of thermal hysteresis factors.
J. Comp. Physiol., 131:347-352.

Duman, J. G. and K. Horwath. 1983. The role of hemolymph proteins in the cold tolerance of insects.
Ann. Rev. Physiol., 45:261-270.

Edgar, W. and M. Loenen. 1974. Aspects of the overwintering habitat of the wolf Pardosa

lugubris. J. Zool. Lond., 172:383-388.

Hazel, J. R. and C. L. Prosser. 1974. Molecular mechanisms of temperature compensation in poikilo-
therms. Physiol. Rev., 54:620-677.

Hendrix, D. L., R. E. Lee, Jr., J. G. Baust and H. James. 1981. Separation of carbohydrates and
polyols by a radially compressed high-performance liquid chromatographic silica column modified
with tetraethylenepentamine. J. Chromatography, 210:45-53.

Husby, J. A. and K. E. Zachariassen. 1980. Antifreeze agents in the body fluid of winter active insects
and spiders. Experientia, 36:963-964.

Kirchner, W. 1973. Ecological aspect of cold resistance in spiders (A comparative study). Pp. 271-279,
In: Effects of Temperature on Ectothermic Organisms (W. Weiser, ed.). Springer-Verlag, New York.

Kirchner, W. and P. Kestler. 1969. Untersuchungen zur kalteresistenz der schilfradspinne Amneus
cornutus (Araneidae). J. Insect Physiol., 15:41-53.

Lee, R. E., Jr. and J. G. Baust. 1981. Seasonal patterns of cold-hardiness in Antarctic terresterial
arthropods. Comp. Biochem. Physiol., 70A:5 79-582.

Lee, R. E., Jr. and J. G. Baust. 1982. Respiratory metabolism of the Antarctic tick, Ixodes uriae.
Comp. Biochem, Physiol., 72A: 167-1 71,

Lee, R. E., Jr., D. Friday, R. Rojas, H. James and J. G, Baust. 1983, An evaluation of eluent recycling
and column life for HPLC analysis of carbohydrates. J. Liq. Chromatogr., 6:1139-1151.

Meryman, H. T. 1974. Freezing injury and its prevention in living cells. Ann. Rev. Biophys. Bioengng.,
3:341-363.

Riddle, W. A. 1981. Cold survival of Argiope aurantia spiderlings (Araneae, Araneidae). J. Arachnol.,
9:343-345.

Riddle, W. A. and A. L. Markezich. 1981, Thermal relations of respiration in the garden spider, Ar-
giope aurantia, during early development and overwintering. Comp. Biochem. Physiol., 69A:759-
165.

Riddle, W. A. and S. Pugach. 1976. Cold-hardiness in the scorpion Paruroctonus aquilonalis. Cryobi-
ology, 13:248-253,

Riechert, S. E. 1974. The pattern of local web distribution in a desert spider: Mechanisms and seasonal
variation, J. Anim. EcoL, 43:733-746.

Riechert, S. E. 1976. Web-site selection in a desert spider, Agelenopsis aperta (Gertsch). Oikos, 27:
311-315.

Riechert, S. E. and R. Gillespie. In press. Habitat choice and utilization in web spinners. In: Spider
Webs and Spider Behavior (W. Shear, ed.). Stanford Univ. Press,

136

THE JOURNAL OF ARACHNOLOGY

Riechert, S. E., W. G. Reeder and T. A. AUen. 1973. Patterns of spider distribution [Agelenopsis
aperta (Gertsch)] in desert grassland and recent lava bed habitats, south-central New Mexico. J.
Anim. EcoL, 42:19-35.

Riechert, S. E. and C. R. Tracy. 1975. Thermal balance and prey availability: Bases for a model
relating web-site characteristics to spider reproductive success. Ecology, 56:265-284.

Salt, R. W. 1961. Principles of insect cold-hardiness. Ann. Rev. Entomol., 6:55-74.

Schaefer, M. 1976. An analysis of diapause and resistance in the egg stage of Floronia bucculenta
(Araneida; Linyphiidae). A contribution to winter ecology. Oecologia, 25:155-175.

Zachariassen, K. E. and J. A. Husby. 1982. Antifreeze effect of thermal hysteresis agents protects
highly supercooled insects. Nature, 298:865-867.

Manuscript received April 1 984, revised August 1984.

1985. The Journal of Arachnology 13:137

RESEARCH NOTES

BALLOONING BEHAVIOR OF UMMIDIA SPIDERLINGS
(ARANEAE, CTENIZIDAE)

Ever since Baerg (1928, Entomol. News, 39:1-4) described the pre-ballooning behavior
of Ummidia carabivora spiderlings, it has been assumed that at least some species of
Ummidia disperse by ballooning, an uncommon mode of dispersal for mygalomorph
spiders. Additional observations of pre-ballooning behavior in Ummidia spiderlings (J.
Ragan, pers. comm.), the widely scattered distribution of Ummidia burrows (Coyle, F. A.,
1983, J. ArachnoL, 11:283-286), and the fact that distribution ranges of some Ummidia
species bridge water gaps (Yaginuma, T., 1979, Bull. National Sci. Mus., 13:639-701 ; G.
B. Edwards, pers. comm.) provide further support for this conclusion. The purpose of this
paper is to describe, for the first time, the ballooning behavior of Ummidia, only the
second non-araneomorph taxon that has been observed ballooning (Coyle, ibid.).

The 100-150 Ummidia spiderlings, clustered on top of a 0.9 m tall tombstone, were
discovered by Sandra and Matthew Beachy at 1045 hr on 7 April 1984 in a graveyard on
a flat grassy knoll just below my home five miles south of Cullowhee, North Carolina.
During the period of observation (1100-1345 hr) I was assisted by Lloyd, Phillip, and
Matthew Beachy, and my son, Alec. During this period the air temperature rose from
14-20°C, there were no clouds, and there were frequent gusts of light wind, ranging up to
perhaps 20 knots and variable in direction (but primarily from the north). A single 2-3
mm wide band of multiple draglines, presumably marking the approach route of the
spiderling brood, extended from the base of the tombstone northward through the grass
for only 1.5 m before disappearing. Careful searches over a 10 m^ area on this side
of the tombstone failed to reveal the maternal burrow.

Early in the observation period almost all the spiderlings were clustered at each end of
the central apical ridgeline formed at the junction of the two inclined (45°) surfaces that
formed the top of the tombstone. As time passed, however, more and more of these
spiderlings walked about, primarily along the ridgeline and edges of the top surface of the
tombstone, always trailing draglines. Later observation of two of these spiderlings alive
under a stereomicroscope confirmed that this dragline is a flat band of numerous fibers
issuing from spigots on both sides of the spinning field. During the strongest gusts of wind
the spiderlings would stop walking and “hug” the surface of the stone.

Ballooning activity commenced about 1115 hr and continued until the last spiderling
departed at 1345 hr. Ballooning was accomplished in the following manner: The spider-
ling would move to an edge of the tombstone’s top surface, tilt its cephalothorax upward,
extend its pedipalps and first two pairs of legs off the substrate and out over the edge,
and drop (or be blown off) a few to about 30 cm on its dragline (Fig. lA). The breeze
would push and lift the spiderling and its dragline up towards the horizontal and away
from the attachment point as the dragline lengthened (Fig. IB- 1C). Eventually, after a
few to several seconds, the lengthening dragline would incline slightly above the hori-

1985. The Journal of Arachnology 13:138

Fig. 1.- Diagrammatic representation of ballooning method of Ummidia spiderlings: A, Spiderling
drops on draghne from edge of launching surface; B-D, Wind pushes spiderling away from launch
point, and, as dragline lengthens, spiderling and dragline are lifted to and then above horizontal; E,
Dragline breaks near attachment point and spiderling drifts downwind.

zontal (Fig. ID) and break at or near the attachment substrate so that the spiderling
would be airborne (Fig. 1 E), drifting off in a sUghtly upward trajectory. Close observation
failed to reveal any silk fibers issuing from the spinnerets other than those fibers that
composed the dragline. Successful launchings like this were observed very infrequently.
More commonly, after dropping on a dragline in the manner just described, the spiderling
was either blown against the tombstone’s surface (after which it ascended to repeat the
launching process) or the spiderling drifted to the ground before the lengthening dragline
could be lifted to the horizontal.

Although the method of ballooning is the same, the launching success of these Um-
midia spiderlings was not as great as that of Sphodros spiderlings I have observed balloon-
ing (Coyle, ibid.). Whether this low launching success is typical of Ummidia and is due to
a heavier body (Coyle, F. A., M. H. Greenstone, A. L. Hultsch and C. E. Morgan, in
prep.), a heavier dragline, some other inherent constraint, or whether it was due to the
greater fluctuation in wind velocity and direction, is not clear. Ummidia spiderling
draglines appear to have a higher tensile strength than those of Sphodros since the Um-
midia draglines tended to become much longer before breaking (one was 6 m long) in
spite of the higher wind velocities.

Frederick A. Coyle, Department of Biology, Western Carolina University, CuUowhee,
North Carolina 28723.

Manuscript received June 1984, accepted July 1984.

1985. The Journal of Arachnology 13:139

SPIDERLING SURVIVAL IN A MANTISPA
(NEUROPTERA, MANTISPIDAE) INFESTED EGG SAC

Larvae of the insect subfamily Mantispinae (Neuroptera, Mantispidae) are obligatory
predators on the eggs of spiders (Redborg 1983). A first instar mantispid must gain access
to spider eggs by one of two methods; either the larva boards a female spider and de-
scends into the egg sac during construction, or it locates and penetrates an egg sac that
has already been formed (Redborg and MacLeod 1984).

The survival of the eggs and subsequent spiderlings would appear to be nonexistent or
very limited once a mantispid larva has successfully entered a spider egg sac, as indicated
by previous researchers. No spiders survived from three egg sacs of Scytodes sp. which
were separately preyed upon by larvae of Mantispa fuscicornis Banks (Gilbert and Ray or
1983). An egg sac of Peucetia viridans found by Killebrew (1981) contained d. Mantispa
cocoon which filled the entire space of the egg sac, and it is therefore assumed no eggs or
spiderlings survived. Valerio (1971) studied Mantispa viridis Walker preying upon
Achaearanea tepidariorum and noted that the mantispid almost always devoured all
of the spider eggs and only when the number of eggs was sufficiently high, would a few
remain intact and survive naturally. Limited spiderling survival was also observed by
Capocasale (1971) with 15 juvenile Lycosa poliostoma surviving the larva oi Mantispa
decorata Erichson. Within this egg sac, around 300400 dehydrated eggs were also found.

However, large numbers of spiderlings can survive in an infested egg sac. On September
27, 1983, a female Lycosa rabida Walckenaer with an egg sac was collected from a weedy
field in Wharton, Texas. The Lycosa egg sac was opened on October 5 and contained 367
live spiderlings plus one mantispid cocoon. All spiderlings appeared healthy and scram-
bled from the egg sac when it was opened. An adult Mantispa internipta Say subsequently
emerged from the cocoon on October 8. Measurements were taken of the following
structures to indicate adult size: head capsule width 2.8 mm; pronotal length 5.3 mm;
and forewing length 17.7 mm.

This contrast between numerous spiderling survival versus minimal or no spiderling
survival may be the result of one or a combination of several factors.

It is not known whether M. internipta is an obligate spider boarder, an obligate egg sac
penetrator or a facultative spider boarder/egg sac penetrator although larvae will board
lycosids (Viets 1941). If the species is an egg sac penetrator, and the larva entered the egg
sac an appreciable amount of time after it had been formed, then the failure to consume
all of the eggs may be related to embryonic development. Spiderling eclosion may have
occurred before some of the embryos could have been eaten (Redborg pers. comm.).

Spiderling survival may also relate to the larva’s ability to locate all of the spider eggs.
Under laboratory conditions, Mantispa uhleri Banks larvae will occasionally end up
surrounded by empty, stuck-together chorions from eggs that it has eaten and may be
unable to locate more viable eggs (Redborg pers. comm.).

Another explanation of spiderling survival may be based on a density dependent
factor. The three Scytodes egg sacs studied by Gilbert and Rayor (1983) contained only

1985. The Journal of Arachnology 13:140

31, 37 and 38 eggs, yet that was a sufficient amount of food for M. fuscicornis to com-
plete development to the adult stage. The survival of the 367 L. rabida spiderlings not fed
upon by the M. interrupta larva probably represents a surplus of prey not required by the
predator during development. A Mantispa species may have a prey resource range where-
by a minimum-maximum number of spider eggs would fulfill the necessary nutritional
requirements for biological development. Based upon the limits of this single observation,
it appears plausible that the probability of any spiderlings suiviving a. Mantispa infestation
would be dependent upon the total number of eggs within the sac. A spider species
producing an egg sac with relative large numbers of eggs should be more likely to produce
progeny than a species laying only a few eggs in a sac if both were to become infested.

I would like to thank Dr. William B. Peck for identifying the Lycosa rabida and
graciously providing English translations from the Spanish manuscripts. A thanks also to
Dr. Kurt E. Redborg for confirming the identity oi Mantispa interrupta and commenting
on an early draft of the manuscript.

LITERATURE CITED

Capocasale, R. 1971. Hallazgo de Mantispa decorata Erichson parasitando la ooteca de una Lycosa
poliostoma (Koch) (Neuroptera, Mantispidae; Araneae, Lycosidae). Rev. Brasil. Biol., 31(3):367-
370.

Gilbert, C. and L. S. Ray or. 1983. First record of mantisfly (Neuroptera: Mantispidae) parasitizing a
spitting spider (Scytodidae). J. Kansas Entomol. Soc., 56(4):578-580.

Killebrew, D. W. \9%\ . Mantispa in aPeucetia egg case. J. Arachnol., 10:281-282.

Redborg, K. E. 1983. A mantispid larva can preserve its spider egg prey: evidence for an aggressive
allomone. Oecologia, 58:230-231.

Redborg, K. E. and E. G. MacLeod. 1984. Maintenance feeding of first instar mantispid larvae (Neu-
roptera: Mantispidae) on spider (Arachnida, Araneae) hemolymph. J. Arachnol., 11:337-341.
Valerio, C. E. 1971. Parasitismo en huevos de araha Achaearanea tepidariorum (Koch) (Araneae:

Theridiidae) en Costa Rica. Rev. Biol. Tropical, 18(1, 2):99-106.

Viets, D. 1941. A biological note on the Mantispidae (Neuroptera) J. Kansas Entomol. Soc., 14(2):70-
71.

Marlin E. Rice, Box 151, Wharton, Texas 77488. Present address: Department of
Entomology, Kansas State University, Manhattan, Kansas 66506.

Manuscript received July 1984, accepted August 1984.

, ^ , v'v*V ■• ,.., • M':!f ^

I

' V-

v^_^7 tii’fiTi V*^ '

:iJ y«i

.*'• ’ / K^tflfll A ,j,, ,,

*5^ ■*'^" ^'*-«;

#: J i '■■’*'^^ ' '. A-f

■ • < '■■ (, ! ,.J

f-r.' - ■■ - -lie

|‘'^?;4ii-: . ... ''y

fi’/' "."^ ■ y: ..ifijc .' ^3,

THE AMERICAN ARACHNOLOGICAL SOCIETY

President:

Susan E. Riechert (1983-1985)

Department of Zoology
University of Tennessee
Knoxville, Tennessee 37916

Membership Secretary:

Norman I. Platnick (appointed)

American Museum of Natural History
Central Park West at 79th Street
New York, New York 10024

Secretary:

Brent D. Opell

Department of Biology
V.P.I.S.U.

Blacksburg, Virginia 24061

The American Arachnological Society was founded in August 1972, to promote
the study of the Arachnida, to achieve closer cooperation between amateur and profes-
sional arachnologists, and to publish The Journal of Arachnology.

Membership in the Society is open to all persons interested in the Arachnida. Annual
dues are $25.00 for regular members, $15.00 for student members. Correspondence
concerning membership in the Society must be addressed to the Membership Secretary.
Members of the Society receive a subscription to The Journal of Arachnology. In addi-
j^tion, members receive the bi-annual newsletter of the Society, American Arachnology.

American Arachnology, edited by the Secretary, contains arachnological news and
comments, requests for specimens and hard-to-fmd literature, information about arach-
nology courses and professional meetings, abstracts of the papers presented at the So-
ciety’s meetings, address changes and new listings of subscribers, and many other items
intended to keep arachnologists informed about recent events and developments in
arachnology. Contributions for American Arachnology must be sent directly to the
Secretary of the Society.

Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only from
current members of the Society, and there are no page charges. Detailed instructions for
the preparation of manuscripts appear in the Fall issue of each year, and can also be
obtained from the Editor and the Associate Editor, Manuscripts that do not follow those
instructions will be returned to the author(s) without the benefit of review.

President-Elect:

Jerome S. Rovner (1983-1985)
Department of Zoology
Ohio University
Athens, Ohio 45701

Treasurer:

Norman V. Homer (1983-1985)
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308

Directors:

G. B. Edwards (1984-1986)
William B. Muchmore (1983-1985)
Jonathan Reiskind (1983-1985)

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 13 SPRING 1985 NUMBER 1

Feature Articles

A revision of the Metaphidippus arizonensis group (Araneae, Salticidae),

Bruce Cutler and Daniel T Jennings 1

Revision de genev o Hurius Simon, 1901 (Araneae, Salticidae),

Maria Elena Galiano 9

New groups and species belonging to the nominate subgenus Paruroctonus

(Scoripiones,y2ieioYid2ie), Richard M. Haradon 19

Experimental studies of the interactions between a web-invading spider

and two host species, Scott F. Larcher and David H. Wise 43

Intersexual competition for food in the bowl and doily spider,

Frontinella pyramitela (Araneae, Linyphiidae), R. B. Suter 61

Aggregations of Nephila clavipes (L.) (Araneae, Araneidae) in relation

to prey availability, Ann L. Rypstra 71

Gnaphosid spiders of north-central Texas (Araneae, Gnaphosidae),

G. Zolnerowich and A". V. Horner 79

The natural history and taxonomy of Cicurina byrantae Exline

(Araneae, Agelenidae), R. G. Bennett 87

Preliminary survey of wandering spiders of a mixed coniferous forest,

Donald C. Lowrie 97

Size and phenology of ballooning spiders at two locations in

eastern Texas, D. A. Dean and W. L. Sterling Ill

Mermithid (Nematoda) parasites of spiders and harvestmen,

George O. Poinar, Jr. 121

Low temperature acclimation in the desert spider, Agelenopsis aperta

(Araneae, kgQ\Qw\d2iQ), Richard E. Lee, Jr. mdJohn G. Baust 129

Research Notes

Ballooning behavior of Ummidia spiderlings (Araneae, Ctenizidae),

Frederick A. Coyle 137

Spiderling survival in diMantispa (Neuroptera, Mantispidae) infested

egg sac. Marlin E. Rice 139

Cover photograph, Phrynus marginemaculatus Koch, by Scott A. Stockwell
Printed by The Texas Tech University Press, Lubbock, Texas
Posted at Lubbock, Texas, in May 1985

r,

The Journal of

^ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

NOV 6 " WBS

** ;■ r ‘f • --ipf-'

VOLUME 13

SUMMER 1985

NUMBER 2

THE JOURNAL OF ARACHNOLOGY

EDITOR: Oscar F. Francke, Texas Tech University
ASSOCIATE EDITORS: B. J. Kaston, San Diego State University

William B. Peck, Central Missouri State University
ASSISTANT EDITOR: James C. Cokendolpher, Texas Tech University
EDITORIAL ASSISTANT: Scott A. Stockwell, Texas Tech University
TYPESETTER: Sharon L. Robertson, Texas Tech University
EDITORIAL BOARD: Charles D. Dondale, Agriculture Canada

William G. Eberhard, Universidad de Costa Rica

Maria E. Galiano, Museo Argentine de Ciencias Naturales

Willis J. Gertsch, American Museum of Natural History

Neil F. Hadley, Arizona State University

Vincent F. Lee, California Academy of Sciences

Herbert W. Levi, Harvard University

Emilio A. Maury, Museo Argentine de Ciencias Naturales

William B. Muchmore, University of Rochester

Martin H. Muma, Western New Mexico University

Norman 1. Platnick, American Museum of Natural History

Gary A. Polis, Vanderbilt University

Susan E. Riechert, University of Tennessee

Michael E. Robinson, U.S. National Zoological Park

Vincent D. Roth, Southwestern Research Station

Jerome S. Rovner, Ohio University

William A. Shear, Hampden-Sydney College

William J. Tietjen, Lindenwood College

George B. Uetz, University of Cincinnati

Carlos E. Valerio, Universidad de Costa Rica

Stanley C. Williams, San Francisco State University

THE JOURNAL OE ARACHNOLOGY (ISSN 0161-8202) is published in
Spring, Summer, and Fall by The American Arachnological Society in
cooperation with The Graduate School, Texas Tech University.

Individual subscriptions, which include membership in the Society, are $25,00
for regular members, $15.00 for student members. Institutional subscriptions to
The Journal are $35.00. Correspondence concerning subscriptions and
memberships should be addressed to the Membership Secretary (see back inside
cover). Back issues of The Journal are available from Dr. Susan E. Riechert,
Department of Zoology, University of Tennessee, Knoxville, TN 37916 U.S. A.,
at $10.00 for each number; the Index to Volumes 1-10 is available for $10.00.
Remittances should be made payable to The American Arachnological Society.
Correspondence concerning undelivered issues should be addressed to the
Assistant Editor, Department of Biological Sciences, Texas Tech University,
Lubbock, Texas 79409, U.S. A.

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

Harvey, M. S. 1985. The systematics of the family Sternophoridae (Pseudoscorpionida). J. Arachnol.,
13:141-209.

THE SYSTEMATICS OE THE EAMILY STERNOPHORIDAE

(PSEUDOSCORPIONIDA)

Mark S. Harvey^

Department of Zoology
Monash University
Clayton, 3168, Victoria, Australia

ABSTRACT

A systematic revision of the family Sternophoridae is presented. Three genera are recognized:
Gary ops Banks, Idiogaryops Hoff and Afrosternophorus Beier; the latter is given full generic status.
These genera are distinguished solely on the female genitalia. The following generic synonymies are
proposed (junior synonym first): Sternophorus Chamberlin = Gary ops; Sternophorellus Beier =
Afrosternophorus; Indogaryops Sivaraman = Afrosternophorus. The following species are synony-
mized with A. ceylonicus (Beier): Sternophorus transiens Murthy and Ananthakrishnan, S. indicus
Murthy and Ananthakrishnan, S. montanus Sivaraman, 5. femoratus Sivaraman, S', intermedius Sivara-
man and Indogaryops amrithiensis Sivaraman. Four new species are proposed: Afrosternophorus
anabates (Australia), A. fallax (Vietnam), A. nanus (Australia) and A. xalyx (Australia). Garyops
contains depressus Banks, sini (Chamberlin), centralis Beier and, possibly, ferrisi (Chamberlin).
Idiogaryops consists of paludis (Chamberhn) and pumilus (Hoff). Afrosternophorus contains aethiopi-
cus (Beier), anabates, new species, araucariae (Beier), cavernae (Beier), ceylonicus (Beier), chamberlini
(Redikorzev), cylindrimanus (Beier), dawydoffi (Beier), fallax, new species, grayi (Beier), hirsti,
(Chamberlin), nanus, new species, papuanus (Beier) and xalyx, new species. Two species-groups are
proposed in each of the two genera Idiogaryops and Afrosternophorus to accommodate species with
differing trichobothrial numbers. Male genitalia was found to be a useful adjunct to traditional charac-
ters, and a detailed description of the male genitalia of A. hirsti is presented. Post-embryonic develop-
ment, biogeography and possible evolutionary pathways are discussed.

INTRODUCTION

Members of the family Sternophoridae are small to medium sized pseudoscorpions
that are immediately recognizable by the possession of an extensive pseudosternum, a
feature indicated by their family name. Their pale colour and corticolous habits combine
to render them comparatively rare in collections.

Six sternophorid genera or subgenera have been described to date, but several inconsis-
tencies in the literature indicate that the generic classification is apparently artificial and
in disarray. In particular, authors have relied heavily on the number of chelal trichobo-
thria and the form of the carapace as generic characters, and ignored genitalic features,
even though the latter have proved to be extremely valuable in the delimitation of genera
in other pseudoscorpion families, such as the Chernetidae. For example, females of

* Present address: Biological Survey Department, Museum of Victoria, 71 Victoria Crescent, Abbots-
ford, 3067, Victoria, Australia.

142

THE JOURNAL OF ARACHNOLOGY

the Mexican Sternophorus sini Chamberlin (the type species of the genus) have genitalia
with two spurred, median cribriform plates (Chamberlin 1923, 1931), whereas S. hirsti
Chamberlin, and other Australian sternophorids have only one unspurred plate. Likewise,
females of Garyops depressa Banks (the type species of the genus) have two spurred,
median cribriform plates (Hoff 1963), whereas G. pumila Hoff and Idiogaryops paludis
(Chamberlin) have two unspurred plates (Hoff 1963). Thus, the present study was aimed
at clarifying the taxonomy of this family which appeared to be unreliable, at least as far
as the female genitalic characters were concerned.

MATERIALS AND METHODS

Much of the material used in this study was borrowed from overseas institutions which
are listed in the Acknowledgments section of this paper. My own material has been
lodged in the following depositories (see Acknowledgments for abbreviations): Australian
Museum, Sydney (AM); Australian National Insect Collection, Canberra (ANIC);MHNG;
MV; Museums and Art Galleries of the Northern Territory, Darwin (NTM); Queensland
Museum, Fortitude Valley (QM); and VAM,

My personal accession numbers (e.g. MH237.01) basically follow the system employed
by other workers, and are used to refer to the exact specimen upon which diagrams and
observations are made. The number to the left of the decimal point refers to the lot
number, and the number to the right of the decimal point refers to each individual
specimen.

Specimens were examined in two ways, depending on curatorial preferences. Perma-
nent microscope slides were made as follows: specimens were removed from 75% ethanol,
an incision was made along one pleural membrane, and the specimens were then cleared
overnight at room temperature in 10% potassium hydroxide, dehydrated through a
graded ethanol series, and mounted on microscope slides in Euparal (Chroma-Gesell-
schaft, Schmid GmbH and Co.). One chela of some specimens was dissected off and
mounted on a cavity slide to facilitate the inspection of the trichobothrial pattern.
Temporary mounts were made by clearing whole specimens in clove oil or lactic acid and
mounting on slides in glycerol.

Each specimen was measured with a micrometer eyepiece in a compound microscope,
measurements were made in accordance with those discussed by Chamberlin (1931),
except for coxa I length (Harvey 1981a) [the “accessory length'’ of Chamberlin (1931)] .
When appendages, especially chelae, were not lying in a horizontal plane, Pythagoras'
theorem was employed to determine their true length (Bird et al. 1979). Appendages
were often observed to be slightly larger on one side of the body than the other. Hence,
measurements were taken from both sides (to the nearest 0.005 mm, except for body
length which was taken to the nearest 0.01 mm) to fully record the range of variation.

Benedict and Malcolm (1977) have recently suggested a modified system of reference
Hnes for obtaining accurate measurements of the pedipalpal chela since they found it
“nearly impossible to secure reliable measurements of the chela when it remains attached
to the palp.” They preferred taking measurements from a lateral aspect, rather than a
dorsal or ventral one as suggested by Chamberlin (1931), I had no trouble measuring
undissected specimens (including large, heavily sclerotized garypids) and Chamberlin’s
original reference lines are used here. There seems to be a serious lack of uniformity
between different workers, and consequently, this makes the comparison between the

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

143

different descriptions very difficult. For example, some authors present chelal measure-
ments including the pedicel, whereas others provide them without the pedicel. Mahnert
and Muchmore have obviously tried to rectify this problem by providing measurements of
the pedicel, but I have found that this is slightly misleading because adding the length of
the pedicel to the length of the chela (without pedicel) does not provide the length of the
chela (with pedicel). I have followed Chamberlin’s (1947 and subsequent papers) system
by providing the length of the chela with the pedicel and without the pedicel. Similarly,
the ratios of the chela are given with the pedicel and without the pedicel. For the sake of
uniformity, all workers should present both of these measurements.

The number of setae within the male genital atrium is given in brackets (e.g. [4-6]),
and the number of setae associated with the spiracular plate is shown in parentheses. The
abbreviation for a tergal or sternal tactile seta is T. The dimensions of females are in
parentheses and follow those of males. The numerator refers to the length of a segment,
and the denominator refers to its width.

Drawings were made with the aid of a Leitz camera lucida attached to a Leitz Ortho-
plan microscope or with a Leitz Prado photographic slide projector with a Prado micro-
scope slide attachment.

The spelling of various southeast Asian localities was found to vary from atlas to atlas,
and to differ from the spelling used by Redikorzev (1938) and Beier (1951). Thus, the
spelling advocated by the U.S. Board on Geographic Names was adopted in this study
(Table 1).-

When portions of the locality data were known from published records, but were not
present on locality labels, they are shown in brackets.

Abbreviations for chelal trichobothria and cheliceral setae follow those employed by
Chamberlin (1931), and are commonly used in pseudoscorpionid literature. Genitalic
abbreviations follow Legg (1974b, 1975a), but are listed here for convenience:

aa

anterior apodeme

da

dorsal apodeme

dag

dorsal anterior gland

dmgs

dorsal median genital sac

ejc

ejaculatory canal

ejca

ejaculatory canal atrium

f

foramen

hp

hyaline plate

la

lateral apodeme

Icp

lateral cribriform plate

Igs

lateral genital sac

Ir

lateral rod

mcp

median cribriform plate

mgs

median genital sac

pdg

posterior dorsal gland

pvdv

posterior ventral diverticulum

te

testis

vdv

ventral diverticulum

Two females of Afrostemophorus hirsti were critical point dried, mounted on points
and gold coated for examination in a JEOL JSM-35C Scanning Electron Microscope.

144

THE JOURNAL OF ARACHNOLOGY

Table 1.— Gazetteer of southeast Asian localities. The first column depicts the spelling used by
Redikorzev (1938) and Beier (1951), and the second column shows those used by the U.S. Board on
Geographic Names. The latter are used in this paper.

KAMPUCHEA
Beng Mealea
Prah Khan
Phailin
Rusei Chrum
Sre Umbell
Ream

LAOS

Luang Prabang
Paclay

Plateau von Boloven

VIETNAM

Plateau von Langbian

Dalat

Arbre-Broye
Krongpha
Insel Phu-Quoc

Phumi Boeng Mealea
Prasat Preah Khan
Paihn

Roessei Chrum
Sre Ambel
Phsar Ream

Louangphrabang

Pak-Lay

Plateau des Bolovens

Cao Nguyen Lam Vien
Da Lat

Ap Tram Hahn
Thon Song Pha
Dao Phu Quoc

13°28'N 104°14'E
13°24'N 104°45'E
12°51'N 102°36'E
11°46'N 103°04'E
11°07'N 103°46'E
10°30'N 103°37'E

19°52'N 102°18'E
18°12'N 101°25'E
15°20'N 106°20'E

12°00'N 108°25'E
11°56'N 108°25'E
11°51'N 108°34'E
11°50'N 108°42'E
10°12'N 104°00'E

FAMILY STERNOPHORIDAE CHAMBERLIN
Sternophorinae Chamberlin 1923:370.

Sternophoridae Chamberlin 1931:238, 1932a:140; Beier 1932:15, 1954b:136; Hoff 1956:3, 1963:2;

Murthy and Ananthakrishnan 1977: 1 1 7-1 18 ; Sivaraman 1981:313.

Diagnosis.— This family may be easily distinguished from other pseudoscorpion fami-
lies by the extensive pseudosternum (Eig. 10); other supplementary characters include:
carapace posteriorly angulate and eyeless (e.g., Fig. 12); venom apparatus present in both
chelal fingers (e.g., Fig. 1 5); accessory teeth absent from chelal fingers (e.g., Fig. 15); legs
monotarsate and homofemorate, the junction of the femora perpendicular to the long
axis of the leg (Fig. 11); and pedal tarsi without elevated slit sensilla.

Description.— Pedipalps and anterior portion of carapace red brown, terga with pale
fuscous stripes, remainder of body pale. Pedipalps, carapace and often legs with striations
and ovoid sculpturing. Cheliceral palm with four setae. Is absent, bs short and blunt (Fig.
9); moveable finger with one subdistal seta; flagellum of four blades, anterior blade very
broad and often with several spinules; galea of male always simple, occasionally with one
or two small rami; galea of female always with several distinct rami. Carapace posteriorly
angulate, eyeless, without transverse furrows, and with a small cucullus and cheliceral
condyle. Pedipalpal femur usually with a single, often subbasal, dorsal tactile seta. Fixed
chelal finger with seven trichobothria, moveable chelal finger with two or three trichobo-
thria; eb and esb basal, adjacent, est about halfway between eb and et, et subdistal, ib and
isb basal, adjacent, opposite eb and esb, ist about halfway between level of esb and est, it
absent, b and sb subbasal, adjacent, t submedial, st absent, sb sometimes absent; areole
shape not unusual (Fig. 4); a long seta usually present slightly proximal to t, approxi-
mately three quarters the length of a trichobothrium, not arising from a large areole (Fig.
1). Chelal fingers without accessory teeth; each with a mediolateral row of stout, curved,
spatulate setae (Fig. 6) (usually more on moveable finger than on fixed finger); and with

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

145

Figs. \-S .-Afrosternophorus hirsti (Chamberlin), scanning electron micrographs, females: 1, lateral
aspect of right chela, MH474.18; 2, dorsal aspect of carapace, MH474.17; 3, pleural membrane of
segments VII-VIII, MH474.17; 4, areole of trichobothrium sb, MH474.17; 5, sensory pit slightly
anterior to sb, MH474.18. Scale lines = 0.1 mm (Figs. 1-3), 0.005 mm (Figs. 4-5).

146

THE JOURNAL OF ARACHNOLOGY

Figs. 6-1 .—Afrosternophorus hirsti (Chamberlin), scanning electron micrographs, female,
MH474.18: 6, lateral aspect of right chela; 7, ventral aspect of coxal area. Scale lines = 0.1 mm.

several sensory pits, each with a small, blunt seta (Fig. 5). Venom apparatus present in
both chelal fingers, nodus ramosus midway between et and est in fixed finger, and slightly
proximal to t in moveable finger. Pedal coxae touching in midline (Fig. 7), but with a
large medial section that is unsclerotized, thus appearing as a ‘pseudosternum’ (Fig. 10).
Legs homofemorate, junction of the femora perpendicular to the long axis of the leg;
femur I always shorter than femur II; tarsi unsegmented, much shorter than tibiae; legs III
and IV each with a medial, tibial tactile seta and a proximal, tarsal tactile seta (Fig. 1 1);
tarsi without elevated slit sensilla; arolia shorter than claws. Abdominal terga and sterna
not medially divided. Pleural membrane longitudinally striate (Fig. 3). Spiracles situated
within pleural membrane; anterior pair of tracheae fairly long, ramifying into tracheoles
when above the third or fourth coxae; posterior pair of tracheae very short, branching
almost immediately (Fig. 10). Male genitalia described in detail below. Female genitalia
with one or two median cribriform plates, with or without spurs, and with one pair of
lateral cribriform plates. Spermathecae absent. Tergum and sternum X each with two
pairs of lateral tactile setae. Tergum and sternum XI fused, with several tactile setae
(because of this fusion, these setae are difficult to count with accuracy, and are simply
referred to in the species descriptions with ‘?’). Anus terminal, anal plate oval.

Type gQnus— Gary ops Banks 1909 (= Sternophorus ChamberUn 1923).

Remarks.— The Sternophoridae is a remarkably uniform group that presents few
characters for its subdivision. Previous authors have utilized only external morphological
characters to delimit genera. These characters are insufficient to divide the family into
monophyletic genera. They are discussed in detail below:

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

147

Fig. 8. -Dorsal aspect of Afrosternophorus anabates, s^Qc\Q%,mdk.\io\oXy^Q.

(1) Anterior constriction of the carapace: The degree of constriction of the carapace
has been utilized as a major generic character by previous authors ever since Chamberlin
[1931: based on Banks’ (1909) misleading description of Garyops depressus] separated
Garyops and Sternophorus on the presence or absence, respectively, of this constriction.
This study has revealed that a constriction is indeed present in S. sini (the type species of
the genus) and that the two genera cannot be separated by this criterion. Furthermore,
intraspecific variation has been observed in several species, most notably in Afrosterno-
phoms dawydoffi; some specimens possess no constriction (Fig. 95), others possess a
slight constriction (Fig. 96), and others display a distinct constriction (Fig. 97). General-
ly, it appears that the degree of constriction is related to the overall size of the animal.
The larger species such as A. dawydoffi, Garyops spp. and Idiogaryops pumilus, and to a
lesser extent, /. paludis, A. anabates and A. ceylonicus, often possess a constriction,
whereas the smaller species (the remaining Afrosternophorus species) do not possess this
constriction. Clearly then, this character is too variable to be given serious consideration
as a valid generic character.

(2) CucuUus: The subgenus Afrosternophorus and the gQmxs Indogary ops were charac-
terized by their authors, Beier (1967) and Sivaraman (1981), respectively, as possessing a
distinct cucuUus. This study has revealed that all sternophorids possess a cucullus, and
therefore, it must be rejected as a valid generic or subgeneric character.

(3) Absence of trichobothrium sb: Idiogaryops and Sternophorellus were separated

from other genera by the possession of only two trichobothria on the moveable chelal
finger. The type species of these two genera, /. paludis and S. araucariae, are, in all other
respects, very similar to species which possess three trichobothria on this finger. The
relative position of b and t of all species is the same and the only difference that can be
found is the absence of sb. To retain different genera for species which lack one or more
trichobothria has clearly extended the generic system and obscured obvious affinities.
Thus, the generic limits of the sternophorid genera are herein extended to include species
with either two or three trichobothria on the moveable chelal finger. Nevertheless, in
accordance with the importance that is placed on trichobothria in pseudoscorpion taxon-
omy, two species groups in each of two Idiogaryops 2indi Afrosternophorus,

been erected to accommodate species with different trichobothrial numbers.

148

THE JOURNAL OF ARACHNOLOGY

Many other pseudoscorpion genera are known to include species with varying numbers
of trichobothria. These include Geogarypus Chamberlin (Chamberlin 1930, Harvey,
unpublished observations), Synsphyronus Chamberlin (Chamberlin 1943, Harvey, in
press), Larca Chamberlin (Hoff 1961), Eremogarypus Beier (Beier, 1962, 1973a), Ana-
garypus Chamberlin (Muchmore 1982a) and Thaumastogarypus Beier (Mahnert 1982b),
the vachoniid Paravachonium Beier (Muchmore 1982b), the cheiridiid Neocheiridium
Beier (Mahnert 1982a) and the chernetid Parachernes Chamberlin (Muchmore and Alteri
1974). More detailed analyses of other generic complexes may well yield further genera
whose trichobothrial numbers vary. Indeed, a perusal of the literature reveals that many
African garypid genera are extremely similar, and that the North American garypids Larca
and Archeolarca Hoff and Clawson are probably synonymous. Furthermore, the Solinus-
Aldabrinus complex (Olpiidae) contains several genera that may eventually prove to be
synonymous.

(4) Cheliceral setation; Hoff (1963) stated that the cheliceral seta sbs was absent in
Garyops and Idiogaryops. Chamberlin (1931) indicated that Is was the missing seta in
Sternophoms, and Murthy and Ananthakrishnan (1977) utilized this apparent difference
to separate Sternophorus from the former genera. In fact, the cheliceral setation of all
sternophorids is identical, and it is contended here that Is is the missing seta.

Thus with all of the traditional characters discarded, only the genitalic characters were
found to be of any use at the generic level. Three genera are recognized herein: Garyops,
Idiogaryops and Afrosternophorus. Females of Garyops and Idiogaryops possess two
median cribrifomi plates and females of Afrosternophorus possess only one. Further-
more, females of Garyops possess lateral spurs on these plates. The generic synonymies
are discussed and justified under the relevant genera.

The genus name Garyops has clearly been treated as feminine by Banks (1909) and
subsequent authors, but Article 30a(i)(2) of the International Code of Zoological Nomen-
clature [as amended by the Commission in 1972 (Bull. Zool. Nomen., 29:182)] unequiv-
ocally states that a “genus-group name ending in -ops is to be treated as mascuHne regard-
less of its derivation or of its treatment by its original author.” Only two names are
affected by this rule in the genera Garyops and Idiogaryops: depressa and pumila are
converted to depressus and pumilus, respectively.

AFFINITIES OF THE STERNOPHORIDAE

Opinions of the taxonomic position of the Sternophoridae have varied over the years.
Chamberlin (1931) placed it with the Cheiridiidae and Pseudochiridiidae in the super-
family Cheiridioidea on the grounds that aU three possess homofemorate legs. Beier
(1954b) transferred it to the Cheliferoidea, allying it to the Goniochernetinae, when he
found that Goniochernes gonio thorax (Redikorzev) possesses not only a posteriorly
angulate carapace (Beier 1932), but also a ‘pseudosternum’, features characteristics of the
Sternophoridae. As discussed in the familial description, the ‘pseudosternum’ of all
sternophorids is not a gap between the coxae, but a large unsclerotized region roughly in
the center of the coxal area. However, the ‘pseudosternum’ of the goniochernetines, at
least in the Australian representatives that I have examined [Calymmachernes angulatus
Beier, Conicochernes brevispinosus (L. Koch), C. crassus Beier, C incrassatus (Beier)
and C. spp.j, is an actual space between the coxae [as described by Beier (1954a) for

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

149

Calymmachernes angulatus] , there being no translucent cuticle. Other chernetids exam-
ined by me, including Cherries cimicoides (Fabricius) and C hahni (C. L. Koch), also
have a small gap between coxae II and III, which often extends to coxae IV. Thus, the
sternophorid pseudosternum and the gap of the Chernetidae are quite different structures
and cannot be regarded as homologous.

Where, then, do the relationships of the Sternophoridae lie? The similarity to the
goniochernetines can hardly be gainsaid, at least as regards the posteriorly angulate
carapace, but I consider that the Goniochernetinae truly belongs in the Chernetidae and is
unrelated to the Sternophoridae. Characters supporting this contention include: (1)
elevated slit sensillum present on all pedal tarsi (absent in Sternophoridae); (2) venom
apparatus present in moveable chelal finger only (present in both fingers in Sternophori-
dae); (3) chelal fingers with accessory teeth (accessory teeth absent in Sternophoridae);
and (4) females with spermathecae (without spermathecae in Sternophoridae). The first
three characters are considered by Muchmore (1973) to be diagnostic of the Chernetidae.
Other characters include the grouping of the setae on the genital opercula (usually com-
pact in Chernetidae; not so in Sternophoridae), the suture of the pedal femora (oblique in
Chernetidae; perpendicular in Sternophoridae), the form of the male genitalia, which,
although hard to define, is “chernetid-like” in the Chernetidae, quite unlike the relatively
simple genitalia of the Sternophoridae, and the presence (Chernetidae) and absence
(Sternophoridae) of a medial division of the abdominal terga and sterna (W. B. Much-
more, pers. comm.).

Therefore, I concur with Heurtault (1983) that the Sternophoridae is not closely
related to the Goniochernetinae. This does not solve the problem of which superfamily
they belong to, and I consider that the placement of the Sternophoridae into either the
Cheiridioidea or the Cheliferoidea is a conjectural matter which cannot be resolved at
present. Further research into areas such as male genitalia may lead to a more stable
classification.

GENITALIA

Pseudoscorpion genitalia have not been frequently studied, but the recent papers by
Legg (1973, 1974a, b, c, 1975a, b, c) have allowed for a comprehensive understanding of
the complex morphology of many British species of the order.

As shown in the taxonomic section, the three sternophorid genera are distinguishable
by female genitalic characters alone. Furthermore, the male genitalia often delimit taxa at
the specific level.

Female genitalia.— The most obvious structures are the cribriform plates, which are
porous plates of unknown function (Legg 1974c). All taxa possess one pair of lateral
cribriform plates (Icp) (e.g.. Fig. 16). One genus, Afrosternophorus, also possesses one
median cribriform plate (mcp)(e.g., Fig. 85), whereas Idiogaryops and Garyops possess
two (e.g., Figs. 16, 45). Furthermore, the latter possesses a unique pair of lateral spurs or
projections on these plates (e.g.. Fig. 16). The significance or function of these spurs is
unknown, and the study of them may be of extreme interest.

Spermathecae are present in most “higher” families such as the Cheliferidae, Cherneti-
dae and Atemnidae (Chamberlin 1931), yet are absent or significantly reduced in the
Chthoniidae, Neobisiidae and Cheiridiidae (Legg 1975b, c). Detailed examination of all
sternophorid genera revealed an absence of spermathecae. The “spermathecae” of Gary-
ops sini figured by Chamberlin (1931: Fig. 52o, as Sternophorus sini) are, in fact, the
median glands (cf. Legg 1974c: Fig. 1).

150

THE JOURNAL OF ARACHNOLOGY

Male genitalia.— The male genitalia of pseudoscorpions consist of a complex series of
sclerotized apodemes and rods, which serve as attachment sites for muscles and as support
for the genital atrium (Legg 1975a). As shown in the taxonomic section, the morphology
of the male genitalia was often seen to vary significantly at the species level. Given this
variation, and the fact that male sternophorid genitalia have not been studied in detail
before, a relatively comprehensive account of the armature and glands is presented.
Although the genital armature was examined in every species in which males were known,
the soft portions were examined only in Afrosternophorus hirsti (Chamberlin) (Fig.
51).

Chamberlin (1923) figured the male genitalia of Garyops sini (as Sternophorus sini),
but my observations on material of this species indicate that his diagram is not entirely
accurate. Chamberlin (1932a) noted that he could distinguish between the four sterno-
phorid species known to him, but refrained from quantifying these differences.

All nomenclature follows that of Legg (1975a), but is presented in the Materials and
Methods for convenience.

(1) Genital opercula and aperture: As with all pseudoscorpions (Legg 1975a), the
anterior and posterior genital opercula are formed from the opisthosomal sternites II and
III. An invagination between these plates forms the genital atrium. Within this atrium are
several (2-8) small setae. The genital aperture is relatively small, as in most of the Mono-
sphyronida.

(2) Genital annature: Associated with the genital atrium are a series of apodemes and
rods which constitute the genital armature (Legg 1975a). The lateral apodeme (la) ex-
tends laterally and sometimes may be curved anteriorly; these apodemes meet in the
midline. Arising anteriorly from the dorsal portion of the armature of some species is an
anterior apodeme (aa). This varies considerably in shape and size, and in A. hirsti it is
brush-like (Fig. 56). The paired dorsal apodemes (da) are usually elongate and acute, but
several species (in two genera) show various modifications in size and shape. Between the
dorsal apodeme and the lateral apodeme is a clear area of cuticle, here termed the hyaline
plate (hp). It is often difficult to observe except under high magnification and strong
illumination. The lateral rod (Ir) forms a complete circle, and is often broadest ventrally.
It possesses a ventral midpiece that is often terminally bifurcate. The lateral rod often
may lie anterior to the genital armature, and thus give a totally different appearance to
the genitalia. Such a situation occurs in A. aethiopicus (Beier) (Fig. 52), A. ceylonicus
(Beier) (Fig. 53) and Idiogary’ops sp. (Fig. 33). Although it may represent distortion
arising during the slide making process, it appears that this is not the case, because the
many spirit preserved specimens of A. ceylonicus that I have examined all possess this
condition. Often one, or sometimes two, foramina (/) occur in the area where the lateral
rod and lateral apodeme fuse. Some species of Garyops and Idiogaryops also possess a
foramen where the lateral apodemes join.

(3) Accessory glands, genital sacs and ventral diverticula: Extending posteriorly from
the genital amiature is the posterior dorsal gland (pdg). As noted by Legg (1975a), it
occurs in all pseudoscorpion families that have been examined. Lying anterior to the
genital armature is the dorsal anterior gland (dag); this gland is bilobed in most pseudo-
scorpions, yet is absent in the Cheiridiidae (Legg 1975a). The single lobed structure found
in sternophorids may be an intermediate stage, but much more work needs to be com-
pleted before such a statement can be verified. Ventral anterior glands are not present.

A pair of lateral genital sacs (Igs) originate from the distal ends of the lateral apo-
demes. A bilobate median genital sac (mgs) lies posterior to the genital armature, and is

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

151

connected to the posterior ventral diverticulum (pvdv) via the long, thin duct of the
median genital sac (drugs). The ventral diverticulum (vdv) is semicircular, and the anterior
edge is often gently sinuate.

(4) Testis, ejaculatory canal atrium, and ejaculatory canal: The testis (te) extends
posteriorly into the abdomen and terminates bluntly. The ejaculatory canal atrium (ejca)
is cup-shaped and lies anterior to the genital armature. It is connected to the genital
atrium by the ejaculatory canal (ejc).

POST-EMBRYONIC DEVELOPMENT

Pseudoscorpions characteristically possess four post-embryonic stages, termed proto-
nymph, deutonymph, tritonymph and adult. Apart from the obvious fact that each stage
is slightly larger than the preceding one, the only other apparent differences are the
increase in the number of setae and the development of genitalia at tlie final moult. It is
the first difference that will be addressed now, since the acquisition of genitalia (or, at
least, the sclerotized portions) is apparently confined to the final moult, and does not
exhibit sequential development. In particular, the number of cheliceral setae and chelal
trichobothria will be examined.

Cheliceral setae.— Little may be said concerning the development of the cheliceral
setae, except that protonymphs differ from the adult and remaining nymphal stages by
lacking gs.

Chelal trichobothria.— The chelal trichobothria of pseudoscorpions are added sequen-
tially during their ontogeny, and provide a means for recognizing the three nymphal
stages (Vachon 1934). Vachon (1936) found that for species in which adults possess eight
trichobothria on the fixed chelal finger and four trichobothria on the moveable chelal
fmger (herein abbreviated to 8/4), the nymphal complement was 3/1 for protonymphs,
6/2 for deutonymphs, and 7/3 for tritonymphs. This has subsequently been confirmed
for many genera of most pseudoscorpion families, even though different trichobothria
may be added at each moult (Mahnert 1981). Furthermore, Vachon (1936) and Nelson
(1982) have found that even though adults of Microbisium dumicola (C. L. Koch) andM
confusum Hoff (Neobisiidae), respectively, possessed a reduced trichobothrial comple-
ment of 7/3, the nymphs retained a complement typical of those species whose adults
possessed 8/4. [Beier (1963) and Gabbut (1969) have expressed doubt as to the validity
of M. dumicola and Vachon’s material may belong to a different species] .

Adults of all known sternophorid species possess a reduced complement of either 7/3
{Gary ops spp., Idiogaryops pumilus species group and Afros ternophorus aethiopicus
species group) or Ijl (/. paludis species group and A. araucariae species group). Unfor-
tunately, the nymphal stages of only four sternophorid species are known. The first
observations were made by Murthy and Ananthakrishnan (1977: Fig. 40B), who demon-
strated that tritonymphs of A. ceylonicus (as Sternophorus transiens) possessed Ijl,
whereas the adults possessed 7/3. This situation can now be confirmed in three other
species of the aethiopicus species group, A. hirsti, A. nanus and A. anabates. Further-
more, the deutonymphs of the former species and the protonymphs of hirsti and anabates
are known. Following the format of Vachon (1936) and Gabbut and Vachon (1965, and
subsequent papers), the order in which the trichobothria are added at each moult in A.
hirsti (the only species for which all nymphal stages are known) is summarized in Table 2.
Following the format of Vachon (1973), it may be summarized as follows, where the

152

THE JOURNAL OF ARACHNOLOGY

Table 2.— The order in which trichobothria are added at each moult in Afrosternophorus hirsti
(Chamberhn).

protonymph

deutonymph

tritonymph

adult

moveable finger series

t

b

sb

fixed finger, external series

eb, et

est

esb

—

fixed finger, internal series

ib

isb, ist

—

—

Stage at which a certain trichobothrium appears is shown as a subscript (A = adult, N3 =
tritonymph, N2 = deutonymph, N1 = protonymph):

In 1 St. sb^ bj^ 2 1 2 3 1 2 1 2

Notwithstanding the absence of st and it, this is similar to Mahnert’s (1981) pattern
for the Cheliferinea (Monosphyronida), to which the Sternophoridae currently belongs.

The reduced trichobothrial complement of all sternophorids makes it difficult to
ascertain which trichobothria are absent. My interpretation that it and st are the missing
trichobothria may need modification as our knowledge of pseudoscorpion trichobothrio-
taxies increases.

Althouglr A. hirsti is the only sternophorid species to be studied in detail, there is no
reason to assume that the situation will be any different for the other species whose
adults possess 7/3. The nymphs of those species which possess a reduced adult comple-
ment of 7/2 (I. paludis, A. araucariae, A. cavernae, A. fallax and^. xalyx) will probably
possess a slightly different pattern.

KEY TO GENERA OE STERNOPHORIDAE

1. Females with two median cribriform plates 2

Females with one median cribriform plate Afrosternophorus Beier

2. Female median cribriform plates with a pair of lateral spurs Gary ops Beier

Female median cribriform plates without lateral spurs Idiogaryops Hoff

Genus Garyops Banks

Garyops Banks 1909:305 (in part); Chamberlin 1931:238 (in part); Beier 1932:18 (in part); Hoff
1963:2-3 (in part). Type species by original designation and monotypy Garyops depressus (pro
depressa) Banks 1909.

Sternophorus Chamberlin 1923:371, 1931:238-239; Beier 1932:16 (in part); Murthy and Anantha-
krishnan 1977:18 (in part). Type species by original designation 2Xidi monotypy Sternophorus sini
Chamberlin 1923. NEW SYNONYMY.

Distribution,— Dominican Republic; El Salvador; Mexico; Florida, U.S.A. (Map 1).
Diagnosis.— Females with two spurred, median cribriform plates. Fixed chelal finger
with seven trichobothria, moveable chelal finger with three trichobothria.

Subordinate Garyops depressus Banks, G. sini (Chamberlin), G. centralis Beier,

G. (?)/em5/ (Chamberlin).

Remarks.— Chamberlin (1931) separated Garyops and Sternophorus on the presence or
absence, respectively, of an anterior constriction of the carapace. The degree of narrowing

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

153

Map 1.- North America showing known distribution of Garyops depressus Banks (circles), G.
sini (Chamberlin) (squares), G. centralis Beier (triangle) and G. (?) ferrisi (Chamberlin) (star) (state
record only). Open symbols represent literature records only.

in G. depressus was unknown to Chamberlin, and he overlooked the slight constriction
evident in S. sini, the type species of Sternophorus. This study has revealed that the form
of the carapace of the two genera is not different, and furthermore, that females of both
genera possess unique spurs on the median cribriform plates. On this basis, Garyops and
Sternophorus are here synonymized. Hoff (1949, 1963) and Hoff and Bolsterli (1956)
suggested that the two genera may be identical, yet refrained from formally synonymiz-
ing them.

Garyops depressus Banks
Figs. 9-16, 28, 34, 39; Map 1

Garyops depressa Banks 1909:305-306 (in part); Beier 1932:18; Hoff 1958:19, 1963:4-7, Figs. 14;
Beier 1976:46; Brach 1979:34-38.

nec Garyops depressa Banks: Hounsome 1980:85 (misidentification).

Types.— Lectotype female (designated by Hoff 1963:4), paralectotype male, paralecto-
type female, Punta Gorda, Charlotte County, Florida, U.S.A., date? [A. T. Slosson] ,
MCZ (slides and spirit).

Distribution.- Florida, U.S.A.; Dominican Republic (Map 1).

154

THE JOURNAL OF ARACHNOLOGY

Diagnosis.— Female galea with three distal rami. Male genitalia with long, acute, dorsal
apodemes. Chela (with pedicel) 0.97 to 1.18 (male), 1.015 to 1.19 mm (female) in length,
3.96 to 4.44 (male), 3.78 to 4.37 (female) times longer than broad.

Description.— Supplementary to Hoff (1963). Chela (with pedicel) 3.96 to 4.44 (male),
3.78 to 4.37 (female) times longer than broad. Carapace (Fig. 12) 1.23 to 1.38 (male),
1.34 to 1.35 (female) times longer than broad. Male genitalia (Fig. 28) with long, acute,

Fig. 9-\2. —Garyops depressus Banks: 9, dorsal aspect of left chelicera, female, S-2840.1; 10,
ventral aspect of coxal area, with left tracheae, male, S-2792,4; 11, leg I and leg IV, male, S-2782.2;
12, dorsal aspect of carapace, male, S-2792.4. Scale line = 1.00 mm (Figs. 10-12), 0.25 mm (Fig. 9).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

155

Figs. Gary ops depressus Banks: 13, dorsal aspect of right pedipalp, male, S-2782.2; 14,

same, female, S-2829.2; 15, lateral aspect of left chela, female, S-2829.2; 16, female genitalia and
associated sternites, S-2829.2. Scale line = 1.00 mm (Figs. 13-15), 0.25 mm (Fig. 16).

156

THE JOURNAL OF ARACHNOLOGY

Dimensions (mm): Chela (with pedicel) 0.97-1.18/0.245-0.28 (1.015-1.19/0.26-0.30).

Habitat.— Hoff (1963) and Brach (1979) discussed the habitat preferences of this
species, and all the specimens (with known habitat data) have been taken from under
bark of Pinus elliotti.

Remarks.— Hoff (1963) adequately redescribed this species, and little needs to be
added here except for details of the male genitalia, measurements of the chela including
the pedicel, and carapaceal ratios, which were omitted in Hoffs paper.

Hoff examined three of Banks’ syntypes, and found that two females were referable to
G. depressa, whereas the third female belonged to his new species G. pumila (herein
transferred to the genus Idiogaryops). I have had the opportunity to examine the other
three syntypes and found that one male belongs to G. depressus, whereas the other two
specimens, a male and a female, belong to /. pumilus.

Hounsome (1980) recorded G. depressus from Little Cayman Island based upon
Prof. Beier’s identification. I have been able to examine this material, and it is clearly /.
pumilus.

Garyops depressus is extremely similar to G. sini, and they eventually may be con-
sidered synonymous. They are retained here as separate species on the basis of the slightly
larger size of G. depressus, even though there is considerable overlap (Fig. 39).

Other specimens examined.- DOMINICAN REPUBLIC: Bani (65 m), 24 September 1972 (J. and
S. Klapperich), 5 males, 3 females (MHNG) (spirit). U.S.A.: FLORIDA; Highlands Co., Archbold
Biological Station, under bark of Pinus elliotti, 1 April 1956 (C, C. Hoff), 1 male (AMNH, S-2782.2)
(slide). Same data as above except 10 April 1956, 1 male (AMNH, S-2792.4) (slide). Same data as
above except 15 April 1956, 1 female (AMNH, S-2829.2) (slide). Same data as above except 16 April
1956, 1 female (AMNH, S-2840.1) (slide).

Garyops sini (Chamberlin), new combination
Figs. 17-21, 29, 35, 39; Map 1

Sternophorus sini Chamberlin 1923:371-372, Plate 1, Fig. 6, Plate 2, Fig. 21, Plate 3, Figs. 6, 15,
22-25, 1931-192, 239, Figs. 4d, 10c, llz, 20f, 52o-q, 67, 1932a:142; Beier 1932:17, Figs. 11-12.

Types.— Holotype male, paratype female (designated as allotype by Chamberlin), SE
corner of Tiburon Island, Gulf of California, Mexico, under bark, 4 July 1921 (J. C.
Chamberlin), CAS, Type Nos. 1286, 1287 (slides). Paratype male, paratype female, same
data as above except under bark of mesquite, ICC, JC-1 83.020034 (slides). Paratype
male, paratype female, same data as above except date? (collector?, presumably J. C.
Chamberlin), NHMW, JC-344.01003-4 (spirit). Paratype female. Palm Canyon, Angel de
la Guarda Island, Gulf of CaUfornia, Mexico, 3 May 1921 (J. C. Chamberlin), JCC,
JC- 167.02001 (slides). Paratype male, paratype female, Los Angeles Bay, Baja California,
Mexico, 25-27 June 1921 (J. C. Chamberlin), JCC, JC-1 76.04001-2 (slides). Paratype
male, paratype female. Las Animas Bay, Baja California, Mexico, 8 May 1921 (J. C.
Chamberlin), JCC, JC-714.01001-2 (slides). The type series also included many other
specimens which were not examined.

Distribution.— Baja California, Gulf of California, Sonora, Mexico (Map 1).

Diagnosis.— Female galea with three distal rami. Male genitalia with long, acute dorsal
apodemes. Chela (with pedicel) 0.835 to 0.935 (male), 0.86 to 1.14 mm (female) in
length, 3.80 to 4.00 (male), 3.54 to 3.81 (female) times longer than broad.

Description.— Pedipalpal trochanter large and inflated, 1.95 to 2.05 (male), 1.79 to
2.10 (female), femur 2.77 to 3.05 (male), 2.68 to 3.05 (female), tibia 2.31 to 2.45

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

157

Figs. ll-l\.-Garyops smz (Chamberlin): 17, ventral aspect of right pedipalp, male paratype,
JC-176.04001; 18, same, female paratype, JC-167.02001 ; 19, lateral aspect of right chela, female
paratype, JC-714.01002; 20, dorsal aspect of carapace, female paratype, JC-176.04002; 21, female
genitalia and associated stemites, paratype, JC-167.02001. Scale line = 1.00 mm (Figs. 17-20), 0.25
mm (Fig. 21).

158

THE JOURNAL OF ARACHNOLOGY

(male), 2.11 to 2.35 (female), chela (with pedicel) 3.80 to 4.00 (male), 3.54 to 3.81
(female), chela (without pedicel) 3.59 to 3.84 (male), 3.35 to 3.61 (female) times longer
than broad. Trichobothria as for genus, in usual position (Figs. 17-19). Serrula exterior of
chehcera with 10 to 12 (male, female) lamellae. Two males (JC-1 76.04001, JC-
714.01001) possess an extra gs on the moveable fingers of their chelicerae. Galea of male
simple, of female with three distal rami, one usually smaller than the others (Fig. 35).
Carapace anteriorly constricted (Fig. 20) with 25 (male), 20 to 32 (female) setae; 1.34
to 1.41 (male), 1.35 to 1.48 (female) times longer than broad. Male genitalia with long,
acute dorsal apodemes (Fig. 29). Female genitalia as for genus (Fig. 21); some females
possess extra, smaller projections on the median cribriform plates. Tergal chaetotaxy:
male, 6:5-6 :4-5;5:4-6:5:5-7:5-6:6:TlT4TlT:?:2; female, 6-7:5-6:3-6:4-6:4-6:4-6:4-8:
5-7:5-6:TlT3-4TlT:?:2. Sternal chaetotaxy: male, 0:4-7:(0)4[2] (0):(l)3-6(l):6-7:5-8:
5-6:6:6:TlT4TlT:?:2; female 0:3-9:(0)3-4(0):(l)4-7(l):4-8:5-8:5-8:6-8:5-8:TlT4TlT:
?:2. Coxal chaetotaxy: male, 4-5 :3-5:3-5:3-5; female, 3-6:3-6:2-5:3-4.

Dimensions (mm): Body length 2. 1-2.4 (2.4-3. 8); pedipalps: trochanter 0.36-0.40/
0.18-0.20 (0.34-0.49/0.185-0.245), femur 0.545-0.625/0.195-0.205 (0.54-0.74/0.19-
0.25), tibia 0.44-0.49/0.185-0.205 (0.435-0.60/0.20-0.26), chela (with pedicel) 0.835-
0.935/0.22-0.24 (0.86-1.14/0.23-0.31), chela (without pedicel) 0.79-0.885 (0.82-1.075),
moveable finger length 0.40-0.43 (0.39-0.515); chelicera 0.16-0.17/0.085-0.095 (0.16-
0.20/0.09-0.115), moveable finger length 0.11-0.13 (0.115-0.14); carapace 0.75-0.83/
0.50-0.60 (0.795-1.02/0.56-0.72); leg I: coxa 0.215-0.24/0.23-0.27 (0.23-0.29/0.265-
0.34), trochanter 0.11-0.125/0.09-0.095 (0.13-0.16/0.09-0.115), femur I 0.10-0.14/
0.105-0.12 (0.105-0.15/0.115-0.15), femur II 0.16-0.195/0.105-0.12 (0.165-0.24/0.115-
0.15), tibia 0.19-0.215/0.07-0.075 (0.18-0.265/0.075-0.095), tarsus 0.115-0.14/0.05-
0.055 (0.115-0.17/0.05-0.065); leg IV: coxa width 0.225-0.28 (0.27-0.29), trochanter
0.16-0.185/0.11-0.125 (0.165-0.21/0.125-0.15), femur I 0.195-0.23/0.17-0.19 (0.205-
0.28/0.18-0.22), femur II 0.23-0.28/0.175-0.195 (0.23-0.32/0.18-0.225), tibia 0.34-
0.365/0.10-0.11 (0.33-0.45/0.105-0.155), tarsus 0.17-0.19/0.07-0.075 (0.13-0.235/

0.07-0.105).

Habitat.— Specimens have been taken from under the bark of several species of trees,
including mesquite (Prosopis sp.), palo tinto (the scientific name of this tree could not be
located) and Sideroxylon sp. (Chamberlin 1923).

Remarks.- As discussed above, this species may eventually prove to be identical with
G. depressus. The male genitalia of these species show no constant differences. The only
real differences appear to be the disjunct distributions (Map 1) and the slightly smaller
size of G. sini (Fig. 39). More specimens must be collected and examined before any
definitive statement may be made about the status of G. sini.

Other specimens examined.— MEXICO: BAJA CALIFORNIA: Los Angeles Bay, 5-6 May 1921 (J.
C. Chamberlin), 4 females (JCC, JC-1 19.03001-4) (slides): Gulf of California; San Gabriel Bay, Espiri-
tu Santo Island, 1 June 1921 (J. C. Chamberlin), 1 female (JCC, JC-36 1.03001) (slide): SONORA; San
Carlos Bay, under bark of palo tinto, 8 July 1921 (J. C. Chamberlin), 1 female (JCC, JC-687. 02001)
(slide).

Gary ops centralis Beier
Figs. 22-24, 36, 39; Map 1

Garyops centralis Beier 1953:15-16, Figs. 1-2.

Types.— Paratype female. La Union, Cutuco, El Salvador, 19 April 1951 (A. Zilch),
SM, 7682, (spirit). Paratype female, same data as above, NHMW (spirit).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

159

Distribution.— El Salvador (Map 1).

Diagnosis.— Female galea with five distal rami and one subbasal ramus. Chela (with
pedicel) 1.31 to 1.43 mm (female) in length, 4.09 to 4.15 (female) times longer than
broad.

Description.— Female only. Pedipalpal trochanter large and inflated, 1.97 to 2.03,
femur 2.95 to 3.25, tibia 2.52 to 2.70, chela (with pedicel) 4.09 to 4.15, chela (without
pedicel) 3.86 to 3.95 times longer than broad. Trichobothria as for genus, in usual posi-
tion (Fig. 23). Serrula exterior of chelicera with 13 lamellae. Galea with five distal and
one subbasal rami (Fig. 36). Carapace anteriorly constricted (Fig. 22), with 34 to 35
setae; 1.30 to 1.33 times longer than broad. Genitalia as for genus (Fig. 24). Tergal
chaetotaxy: 6-7:6:4:6-7:6-7:6-8:7-8:7-8:6:TlT4TlT:?:2. Sternal chaetotaxy: 0:9:
(0)4-5(0):(l)5-6(l):7:6-7:7:8:6:TlT4TlT:?:2. Coxal chaetotaxy: 4-5 :4-5 :4-5:5-6.

Dimensions (mm): Body length 3.6; pedipalps: trochanter 0.59-0.635/0.195-0.32,
femur 0.90-0.96/0.28-0.325, tibia 0.72-0.78/0.27-0.31, chela (with pedicel) 1.31-1.43/

Figs. 22-2A.— Gary ops centralis Beier, female paratype: 22, dorsal aspect of carapace; 23, ventral
aspect of left pedipalp; 24, median cribriform plates. Scale line = 1.00 mm (Figs. 22-23), 0.25 mm
(Fig. 24).

160

THE JOURNAL OF ARACHNOLOGY

0.32-0.35, chela (without pedicel) 1 .265-1 .35, moveable finger length 0.61-0.65; chelicera
0.22-0.255/0.12-0.135, moveable finger length 0.05-0.065; carapace 1.16-1.21/0.87-0.93;
leg I: coxa 0.32-0.36/0.40-0.43, trochanter 0.185-0.21/0.21-0.13, femur I 0.19-0.22/
0.145-0.155, femur II 0.28-0.295/0.15-0.16, tibia 0.28-0.31/0.10, tarsus 0.14-0.165/0.07;
leg IV: coxa width 0.39-0.44, trochanter 0.23-0.26/0.155-0.17, femur I 0.33-0.375/
0.22-0.265, femur II 0.445-0.48/0.23-0.275, tibia 0.51-0.56/0.14-0.1 55, tarsus 0.26-0.27/
0.105-0.11.

Habitat.— No habitat data accompanied the specimens.

Remarks.— Beier (1953) erroneously referred to the NHMW specimen as a male,
and labelled it as such. Furthermore, he stated that the holotype male and a paratype
female were deposited in SM; the former specimen is apparently not housed in this
institution (Dr. Grasshoff, pers. comm.), but a lectotype female has not been designated
in the hope that the male might eventually reappear.

Gary ops centralis may be a junior synonym of G. (?) ferrisi, but the paucity of speci-
mens precludes any definite statements. They are virtually identical in size (Fig. 39), but
since males of centralis and females of ferrisi are not yet known, the final decision must
await further collecting.

Females of G. centralis possess spurred median cribriform plates which justifies its
inclusion in this genus. These plates (Fig. 24) appear to be different to those of the other
species of the genus, but this is simply due to the mode of preservation of the specimens.
Females of G. depressus and G. sini were eviscerated and mounted on microscope slides;
this tends to push the cribriform plates so that they lie flat. Females of G. centralis were
not eviscerated (due to curatorial preferences) and the cribriform plates were lying
in a slightly different plane. Several non-eviscerated specimens of G. depressus and G. sini
were examined, and their plates also lay at a different angle, as shown by Chamberlin
(1931 : Fig. 52o) for the latter species.

Gary ops (1) ferrisi (Chamberlin), new combination
Figs. 25-21, 30, 39; Map 1

Sternophorus ferrisi Chamberlin 1932a:143; Beier 1932:18.

Type.— Holotype male, no exact locality, Michoacan, Mexico, under bark of tree, date?
(G. F. Ferris), JCC, JC-275.01001 (slide).

Distribution.— Michoacan, Mexico (Map 1).

Diagnosis.— Male genitalia with long dorsal apodemes. Chela (with pedicel) 1.425 to
1 .45 mm (male) in length, 4.25 (male) times longer than broad.

Description.— Male only. Pedipalpal trochanter large and inflated, 2.05 to 2.08, femur
3.33 to 3.36, tibia 2.76 to 2.86, chela (with pedicel) 4.25, chela (without pedicel) 4.01
times longer than broad. Trichobothria as for genus, in usual position (Figs. 26-27).
Serrula exterior of chelicera with 12 to 13 lamellae. Galea simple. Carapace anteriorly
constricted (Fig. 25), with 23 setae; 1.27 times longer than broad. Male genitalia (Fig. 30)
with long dorsal apodemes, and apparently with a large, median foramen. Tergal chaeto-
taxy: 6:6:4:5:6:6:6:5:?:T1T4T1T:?:2. Sternal chaetotaxy: 0:7:(0)3 [3] (0):(1)6(1):8:
8:6:6:6;T1T4T1T:?:2. Coxal chaetotaxy: 4-5:5-6:4-5:5.

Dimensions (mm): Body length 3.8; pedipalps: trochanter 0.665-0.675/0.325, femur
0.99-1.00/0.295-0.30, tibia 0.80/0.28-0.29, chela (with pedicel) 1.425-1.45/0.335, chela
(without pedicel) 1.34-1.345, moveable finger length 0.59-0.62; chelicera 0.27/0.16-

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

161

0.165, moveable finger length 0.17-0.175; carapace 1.205/0.95; leg I: coxa 0.32-0.33/
0.40-0.41, trochanter 0.18-0.19/0.125-0.145, femur I 0.165-0.17/0.185-0.195, femur II
0.28/0.185-0.195, tibia 0.315-0.32/0.10-0.105, tarsus 0.20/0.07-0.075; leg IV: coxa
width 0.37-0.38, trochanter 0.235-0.24/0.165-0.17, femur I 0.31/0.265-0.27, femur II
0.42-0.43/0.28, tibia 0.54/0.15, tarsus 0.275-0.28/0.105.

Habitat.— No habitat data accompanied the specimen, yet Chamberlin (1932a) noted
that it was “collected under bark of a tree”.

Remarks.— Since females of this species are unknown, the generic position oi ferrisi is
uncertain. It is tentatively placed in Garyops because of its similarity with G. centralis,
with which it may be conspecific (see above).

The apparent lack of setae on one side of the pedipalpal tibia and chela (Fig. 27) is an
artifact which probably occurred during preparation of the slide.

The exact collection site of the holotype is unknown, and Map 1 shows the
state record only.

Figs. 25-11 .-Garyops (?) ferrisi (Chamberlin), male holotype: 25, dorsal aspect of carapace; 26,
lateral aspect of left chela; 27, ventral aspect of right pedipalp. Scale line = 1.00 mm.

162

THE JOURNAL OF ARACHNOLOGY

Figs. 28-33. -Anterior portion of male genitalia, ventral aspect: 28, Garyops depressus Banks,
S-2782.2; 29, G. sini (Chamberlin), paratype, JC-176. 04001 ; 30, G. 0) ferrisi (Chamberlin), holotype;
31, Idiogaryops paludis (Chamberhn), S-2887.?; 32,7, pumilus (Hoff) from Little Cayman Island; 33,
I. sp., S-2886.2. Scale line = 0.25 mm (Figs. 28-29), 0.33 mm (Fig. 30), 0.167 mm (Figs. 31-33).

Genus Idiogaryops Hoff

Garyops Banks 1909:305 (in part); Chamberlin 1931:238 (in part); Beier 1932:18 (in part); Hoff
1963: 2-3 (in part).

Sternophorus Chamberlin: Beier 1932:16 (in part); Murthy and Ananthakrishnan 1977:16 (in part).
Idiogaryops Hoff 1963:10-11. Type species by original designation and monotypy Sternophorus
paludis Chamberlin 1932a.

Distribution.— Little Cayman Island; Arkansas, Florida, Georgia, Illinois, Mississippi,
North Carolina, Texas, U.S.A. (Map 2).

Diagnosis.— Females with two (occasionally three) unspurred, median cribriform
plates. Fixed chelal finger with seven trichobothria, moveable chelal finger with two or
three trichobothria.

Subordinate i20L2i— Idiogaryops paludis (Chamberlin), Idiogaryops pumilus (Hoff).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

163

Figs. 34-3 8. -Female galeae: 34, G. depressus S-2840.1;35, G. sini (Chamberlin), paratype,

JC-714. 01002; 36, G. centralis Beier, paratype; 37, Idiogaryops paludis (Chamberlin), S-2826.2;/.
pumilus (Hoff), paratype, S-3782.3. Not to same scale.

Remarks.— is here redefined to include all sternophorids with two un-
spurred, median cribriform plates, even though it was originally restricted by Hoff (1963)
to include only /. paludis, on the basis of it possessing only two trichobothria on the
moveable chelal finger. As discussed above, to give generic status to species which lack
individual trichobothria has extended the generic system and has obscured obvious
relationships between species. Therefore, this genus, as here interpreted, contains species
with two or three trichobothria on the moveable chelal finger. Those with two trichobo-
thria are placed in the paludis group, and those with three trichobothria are placed in the
pumilus group.

Map. 2. -North America showing known distribution o^ Idiogaryops paludis (Chamberlin) (circles)
and/, pumilus (Hoff) (squares). Open symbols represent literature records only.

164

THE JOURNAL OF ARACHNOLOGY

1.50

CL

1.40

1.30

1.20

1.10

1.00

0.90

0.80

39

8 • •

J L

1.10 -

CL

1.00

0.90

0.80

0.70

0.60 -

J 1 I I 1 I I I 1 I I L

40

I I I I I I I I I 1 I I 1 1 L

0.22 0.24 0.26 0.28 0.30 0.32 0.34 0.36 0.14 0.16 0.18 0.20 0.22 0.24 0.26 0.28

CW CW

Figs. 3 9-40. -Graphs of chela (with pedicel) length (CL) versus width (CW), in mm; open symbols,
males; closed symbols, females: 39, Garyops depressus Banks (circles), G. sini (Chamberlin) (squares),
G. centralis Beier (triangles), G. (?) ferrisi (Chamberlin) (star); 40, Idiogaryops paludis (Chamberlin)
(squares), /. pumilus (Hoff) (circles), I. sp. (triangles).

Paludis group

Diagnosis.— As for genus, except that the moveable chelal finger possesses two tricho-
bothria, b and t.

Subordinate t20i2i.— Idiogaryops paludis (Chamberlin).

Idiogaryops paludis (Chamberlin)

Figs. 31,37, 4045 ; Map 2

Sternophorus paludis Chamberlin 1932a:142-143 ; Beier 1932:17-18; Hoff and Bolsterh 1956:164-
165;Hoff 1958:19.

Idiogaryops paludis (Chamberlin): Hoff 1963:11-13, Figs. 7-9;Weygoldt 1969:27, Fig. 105; Rowland
and Reddell 1976:19; Brach 1979:34-38.

Types.-Holotype male, no exact locality, Alachua County, Florida, U.S.A., 30 March
1925 (T. H. Hubbell), depository unknown, JC-725.01001 (slide?), not examined.
Paratype female, Billy’s Island, Okeflnokee Swamp, Georgia, U.S.A., date? (C. R. Cros-
by), depository unknown, JC-43. 02001 (slide?), not examined.

Distribution.— Arkansas, Florida, Georgia, Illinois, Mississippi, North Carolina, Texas,
U.S.A. (Map 2).

Diagnosis.— Male genitalia with long dorsal apodemes. Small species: chela (without
pedicel) [from Hoff and Bolsterli (1956) and Hoff (1963)] 0.61 to 0.67 (male), 0.61 to
0.715 mm (female) in length.

Description.— Supplementary to Chamberlin (1932a), Hoff and Bolsterli (1956) and
Hoff (1963). Carapace (Fig. 44) only slightly constricted anteriorly. Male genitalia (Fig.
31) with long, tapering dorsal apodemes.

Habitat.— Hoff and Bolsterli (1956), Hoff (1963), Weygoldt (1969) and Brach (1979)
have recorded this species from a variety of cortical habitats (Pinus elliotti. Ilex cassine,
Quercus virginianus, Platanus occidentalis and Carya alba).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

165

Remarks.— Previous authors have adequately described this species and little needs to
be added here except for details of the male genitalia which were omitted in previous
papers. Hoff (1963) recorded the occasional presence of a small, third median cribriform
plate in some females.

Contrary to Chamberlin (1932a), the type specimens are not deposited at Cornell

University (Dr. L. L. Pechuman, pers. comm.).

Specimens examined.— U.S. A.: FLORIDA; Highlands Co., Archbold Biological Station, 14 April
1956 (C. C. Hoff), 1 female (AMNH, S-2826.2) (slide). Same data as above except 22 April 1956, 1
male (AMNH, S-2887.?) (slide). Same data as above except 1 May 1956, 1 female (AMNH, S-2951.3)
(slide).

Pumilus group

Diagnosis.— As for genus, except that the moveable chelal finger possesses three tricho-
bothria, b, sb and t.

Subordinate taxa—Idiogaryops pumilus (Hoff).

Idiogaryops pumilus (Hoff), new combination
Figs. 32, 38, 40, 46-50; Map 2

Gary ops depressa Banks 1909:305-306 (in part); Hounsome 1980:85 (misidentification).

Garyops pumila Hoff 1963:7-10, Figs. 5-6 (in part).

Types.— Holotype male, Parker Islands, near Lake Placid, Highlands County, Florida,
U.S. A., under bark of live oak [Quercus virginianus] , 22 April 1956 (C. C. Hoff),
AMNH, S-2886.8 (slide). Paratype female. Mahogany Hammock, Everglades National
Park, Florida, U.S. A., [under bark of Metopium toxiferum] , 8 February 1958 (F. C.
Craighead), AMNH, S-3 782.3 (slide). Paratype female, Punta Gorda, Charlotte County,
Florida, U.S. A., date? [A. T. Slosson] , MCZ, S-3791.3 (slide) (syntype of G. depressus).
(The type series also consisted of other specimens wliich were not examined.)

Distribution.— Florida, U.S. A.; Little Cayman Island (Map 2).

Diagnosis.— Male genitalia with long, tapering dorsal apodemes. Chela (without pedicel)
[from Hoff (1963)] 0.865 to 0.95 (male), 0.95 to 1.07 mm (female) in length.

Description.— Supplementary to Hoff (1963). Carapace (Fig. 46) 1.31 (male), 1.32 to
1.34 (female) times longer than broad, with 31 (male), 21? to 32 (female) setae. One
female specimen (S-3 782.3) has et missing from one chela (Fig. 49). Male genitalia (Fig.
32) with long, tapering dorsal apodemes.

Habitat.-^Hoff (1963) briefly discussed the habitat preferences of this species and
stated that it had been collected from the bark of oak (Quercus virginianus) and poison-
wood (Metopium toxiferum) and from moss and rotted wood at the base of cabbage
palmetto (Sabal palmetto). This is in contrast to the sympatric species G. depressus,
which in Florida is only known from bark of slash pine (Pinus elliotti). Idiogaryops
pumilus was not recorded in Brach’s (1979) rigorous search for pseudoscorpions under
slash pine bark, and highlights this interesting case of habitat partitioning. The specimens
from Little Cayman were apparently taken from “marl facies with tall scrub” (Hounsome
1980).

Remarks.— Hoff described this Floridian species from four males and five females
which included one of Banks’ syntypes of G. depressus. His description is quite thorough
and all that needs to be added here are details of the male genitalia and carapaceal ratios.

166

THE JOURNAL OF ARACHNOLOGY

Figs. MA5.-Idiogaryops paludis (Chamberlin): 41, dorsal aspect of left pedipalp, male, S-2887.?;
42, lateral aspect of left chela, female, S-2826.2;43, dorsal aspect of right pedipalp, female, S-2826.2;
44, dorsal aspect of carapace, female, S-2826.2; 45, female genitalia and associated sternites, S-2828.2.
Scale line = 1.00 mm (Figs. 41-44), 0.25 mm (Fig. 45).

As discussed below under Idiogary’ops sp., one of the male paratypes of G. pumilus is
not conspecific with the holotype of this species.

Otlier specimens examined.— U.S. A.: FLORIDA; Punta Gorda, date? [A. T. Slosson] , 1 male, 1
female (MCZ) (spirit) (syntypes of G. depressus). LITTLE CAYMAN ISLAND: 3 August 1975 (M. V.
Hounsome), 1 male, 1 female (NHMW) (spirit).

Idiogaryops sp.
Figs. 33, 40

Gary ops pumila Hoff 1963:5-6 (in part).

Diagnosis.— Male genitalia with reduced dorsal apodemes.

Remarks.— Even though this specimen possesses pedipalpal morphometries that are
indistinguishable from /. pumilus (Fig. 40), the form of the male genitalia (Fig. 33) is

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

167

Figs. A6-50.-Idiogaryops pumilus (HofO: 46, dorsal aspect of carapace, male from Little Cayman
Island; 47, ventral aspect of left pedipalp, male from Little Cayman Island; 48, ventral aspect of right
pedipalp, female from Little Cayman Island; 49, lateral aspect of right chela, female paratype, S-
3782.3 (note absence of trichobothrium et) \ 50, female genitalia and associated sternites, from Little
Cayman Island. Scale line = 1.00 mm (Figs. 46-49), 0.25 mm (Fig. 50).

substantially different from that species, and it undoubtedly represents a new species. I
have not formally described it because I believe that more specimens, including females,
should be examined. Indeed, it is possible that one or more of the female paratypes
recorded by Hoff (1963) as G. pumila may be the female of this species.

Specimens examined.— U.S. A.: VhOKYD k. Highlands Co., Parker Islands, near Lake Placid, under
bark of live oak [Quercus virginianus] , 22 April 1956 (C. C. Hoff), 1 male (AMNH, S-2886.2) (slide)
(paratype of G. pumila).

Genus Afrosternophorus Beier, new status

Stemophorus Chamberlin: Beier 1932:16 (in part); Murthy and Ananthakrishnan 1977:118 (in part).
Sternophoms (Afrosternophorus) Beier 1967:81-82. Type species by original designation and mono-
typy Stemophorus (Afrosternophorus) aethiopicus Beier 1967.

Sternophorellus Beier 1971:371-372. Type species by original designation and monotypy Sterno-
phorellus araucariae Beier 1971. NEW SYNONYMY.

168

THE JOURNAL OF ARACHNOLOGY

Jndogaryops Sivaraman 1981:322. Type species by original designation and monotypy Indogaryops
amrithiensis Sivaraman 1981. NEW SYNONYMY.

Distribution.— New South Wales, Northern Territory, Queensland, Victoria, Australia;
Ethiopia; India; Kampuchea; Laos; Papua New Guinea; Sri Lanka; Vietnam (Maps 3 to 6).

Diagnosis.— Females with one unspurred, median cribriform plate. Fixed chelal finger
with seven trichobothria, moveable chelal finger with two or three trichobothria.

Subordinate taxa.—Afrosternophoms aethiopicus (Beier), A. anabates, new species, A.
araucariae (Beier), A. cavernae (Beier), A. ceylonicus (Beier), A, chamberlini (Redi-
korzev), A. cylindrimanus (Beier), A. dawydoffi (Beier), A. fallax, new species, A. grayi
(Beier), A. hirsti (Chamberlin), A. nanus, new species,^, papuanus (Beier), A. xalyx, new
species.

Fig. 51. -Ventral aspect of male genitalia of
Afrosternophonis hirsti (Chamberlin), MH302.41.
Scale line = 0.30 mm.

Remarks.— Beier (1967) originally described Afrosternophonis as a subgenus of the
genus Sternophorus; the latter has been shown above to be a junior synonym of Garyops.
Therefore, it is necessary to reassess the status of Afrosternophorus. Since females are
needed to unequivocally place species in genera, it is unfortunate that S. aethiopicus (the
type species of Afrosternophorus) is currently represented in collections by a single male.
Nevertheless, I have raised Afrosternophorus to full generic status to accommodate those
sternophorids in which females possess genitalia with one unspurred, median cribriform
plate. If in the future it can be shown that A. aethiopicus is not congeneric with the
remaining species I have included in the genus, Sternophorellus is the next available name.

Sternophorellus was erected by Beier (1971) for S.araucariae Beier from Papua New
Guinea. It differed from other genera (except Idiogaryops) by possessing only two

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

169

trichobothria on the moveable chelal finger. This is no longer considered to be a valid
character for separating genera, and we are left with the female genitalia to delimit the
higher taxa. Unfortunately, females of S. araucariae have not been available for study, but
females of the new species A. fallax from Vietnam and A. xalyx from Australia, which
also have only two such trichobothria, possess genitalia with only one median cribriform
plate. IhQXQioxQ.Sternophorellus is synonymized mth Afrosternophorus.

Figs. 5 2-5 7. -Anterior portion of male genitalia, ventral aspect: 52, Afrosternophorus aethiopicus
(Beier), holotype {ejca not shown); 53, A. ceylonicus (Beier), paralectotype from Per Aru, Sri Lanka;
54, A. chamberlini (Redikorzev); 55, A. da wy do ffi (BeiQi), paralectotype, MH430.01; 56, A. hirsti
(Chamberlin), MH302.21 ; 57, A. nanus, new species, holotype. Scale line = 0.10 mm.

170

THE JOURNAL OF ARACHNOLOGY

Figs. 58-64. -Anterior portion of male genitalia, ventral aspect: 58, Afrostemophorus anabates,
new species, paratype, MH044.01 {ejca not shown); 59, A. papuanus (Beier), lectotype; 60, /I. grayi
(Beier), lectotype (slightly distorted); 61, araucariae (Beier), holotype; 62, A. cavernae (Beier),
paratype; 63, A. fallax, new species, holotype; 64, A. xalyx, new species, holotype {ejca not shown).
Scale line = 0.10 mm.

Indogary ops was erected by Sivaraman (1981) to include the Indian species /. amri-
thiensis. It was segregated from other genera by the presence of both a cucuUus and an
anterior constriction of the carapace. As discussed above, all sternophorid genera possess
a cucullus, and the degree of constriction in the carapace may vary intraspecifically (see
A. dawydoffi, Figs. 95-97); therefore, both characters are invalid at the generic level.
More importantly, females of I. amrithiensis possess one median cribriform plate, and
therefore Indogary ops falls into synonymy with Afrostemophorus. Furthermore, /.
amrithiensis is a junior synonym of A. ceylonicus.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

171

Figs. 65-77. -Female galeae: 65-66, Afrosternophorus ceylonicus (Beier), paralectotype from
Chemiyanpattu, Sri Lanka; 67, A. chamberlini (Redikorzev); 68, A. dawydoffi (Beier), paralectotype
from Roessei Chrum, Kampuchea; 69, A. cylindrimanus (Beier), paralectotype; 70, A. hirsti (Cham-
berlin), MH302.48; 71, A. hirsti, MH302.50; 72, A. nanus, new species, paratype, MH230.09; 73, ^4.
anabates, new species, paratype, MH416.10; 74, A. papuanus (Beier), paralectotype; 75, A. grayi
(Beier), paralectotype from Bulolo, Papua New Guinea; 76, A. fallax, new species, paratype; 77, A.
xalyx, new species, paratype, K160. Not to same scale.

KEY TO SPECIES OF AFROSTERNOPHORUS

1 . Moveable chelal finger with three trichobothria, b, sb and t

aethiopicus group. . .2

Moveable chelal finger with two trichobothria, b and t . . . araucariae group. . . 11

2. Male genitalia with reduced dorsal apodemes 3

Male genitalia with long, tapering dorsal apodemes 5

3. Male genitalia with brush-like anterior apodeme; dorsal apodemes parallel sided;

Australia hinti (Chamberlin)

Male genitalia without brush-like anterior apodeme; dorsal apodemes, if visible, not
parallel sided 4

4. Chelal fingers long and strongly curved; male genitalia with short, but prominent,
dorsal apodemes, slightly curved; female galea with two distal and one subdistal to

subbasal rami; India, Sri Lanka ceylonicus (Beier)

Chelal fingers not especially long or strongly curved; male genitalia with much
reduced dorsal apodemes, not visible; female galea unknown; Ethiopia

aethiopicus (Beier)

5. Male genitalia with anterior apodeme distally broad; female galea with three distal,

one subdistal and two (sometimes one) subbasal rami 6

Male genitalia with anterior apodeme not distally broad; female galea not as
above 7

172

THE JOURNAL OF ARACHNOLOGY

6. Chela (with pedicel) 0.83 to 0.895 (male), 0.835 to 1.03 mm (female) in length,

Australia anabates, new species

Chela (with pedicel) 0.61 to 0.69 (male), 0.73 to 0.74 mm (female) in length;
Papua New Guinea papuanus (Beier)

7. Lateral rod of male genitalia with long mid-piece; female galea with two distal, one

subdistal and three subbasal rami; Vietnam, Laos (?)... chamberlini (Redikorzev)
Lateral rod of male genitalia without, or with short, mid-piece; female galea never
with three subbasal rami 8

8. Chela (with pedicel) less than 0.74 mm in length; female galea with three distal to

subdistal rami 9

Chela (with pedicel) greater than 0.90 mm in length; female galea with at least
four distal to subdistal rami 10

9. Chela (with pedicel) 0.645 to 0.70 (male), 0.68 to 0.74 mm (female) in length;

Papua New Guinea grayi (Beier)

Chela (with pedicel) 0.55 to 0.59 (male), 0.56 to 0.61 mm (female) in length; Aus-
tralia new species

10. Chela (with pedicel) 1.07 to 1.35 (male), 1.11 to 1.38 mm (female) in length;

female galea with six (sometimes five) distal to subdistal rami; Kampuchea, Viet-
nam dawydoffi (Beier)

Chela (with pedicel) 0.935 to 0.95 (male), 1 .02 mm (female) in length; female galea
with four distal and one subbasal rami; Laos cylindrimanus (Beier)

11. Male genitalia with reduced dorsal apodemes; female galea with two distal and four

subdistal to subbasal rami; Australia xalyx, new species

Male genitalia with long dorsal apodemes; female galea (when known) not as
above 12

12. Chela (with pedicel) 0.805 to 0.82 mm (male) in length, 4.47 to 4.56 (male)

times longer than broad; Papua New Guinea araucariae (Beier)

Chela (with pedicel) less than 0.70 mm in length, less than 4.00 times longer than
broad 13

13. Lateral rod of male genitalia with short mid-piece; Papua New Guinea

cavemae (Beier)

Lateral rod of male genitalia with long mid-piece; Vietnam . . . fallax, new species

Aethiopicus group

Diagnosis.— As for genus, except that the moveable chelal finger possesses three tricho-
bothria, b, sb and t.

Subordinate tSLX?i.—Afrostemophoms aethiopicus (Beier), A. anabates, new species, A.
ceylonicus (Beier), A. chamberlini (Redikorzev), A. cylindrimanus (Beier), A. dawydoffi
(Beier), A. grayi (Beier), A. hirsti (Chamberlin), A. nanus, new species, A. papuanus
(Beier).

HARVEY-SYSTEMATICS OF STERNOPHORIDAE

173

Map 3. -North Africa showing known distribution of Afrosternophoms aethiopicus (Beier).

Afrosternophorus aethiopicus (Beier), new combination
Figs. 52, 78-79; Map 3

Sternophonis (Afrosternophorus) aethiopicus Beier 1967:81-82, Fig. 6.

Type.— Holotype male, Alomata, Ethiopia (5000 ft) [= 1525 m] , 16 January 1960 (E.
S. Ross), CAS, Type No. 9386 (slide).

Distribution.— Ethiopia (Map 3).

Diagnosis.— Male genitalia with greatly reduced dorsal apodemes; lateral rod lying
anteriorly to the rest of the genital armature. Chela (with pedicel) 0.895 mm (male) in
length.

Description.— Male only. Pedipalpal trochanter large and inflated, 1.76, femur 2.68 to
2.76, tibia 1.90, chela (with pedicel) 3.52, chela (without pedicel) 3.49 times longer than
broad. Trichobothria as for aethiopicus group, in usual position (Fig. 78). Serrula exterior
of chelicera with 12 lamellae. Galea of male simple. Carapace anteriorly constricted (Fig.
79), with at least 12 setae; 1.37 times longer than broad. Male genitalia (Fig. 52) with
greatly reduced dorsal apodemes; lateral rod lying anteriorly to the rest of the genital
armature. Tergal chaetotaxy; 6:7:5;5;5:6:6:6:6:T1T4T1T:?:2. Sternal chaetotaxy:
0:6:(0)6[0?](0):(1)6(1):6:6:6:8?:7:T1T4T1T:?:2. Coxal chaetotaxy: 3:5:2-5:4.

Dimensions (mm): Body length 2.4; pedipalps: trochanter 0.37-0.38/0.21, femur 0.58
-0.59/0.21-0.22, tibia 0.47/0.205, chela (with pedicel) 0.895/0.245, chela (without
pedicel) 0.855, moveable finger length 0.43; chelicera 0.17/0.09, moveable finger length
0.12; carapace 0.82/0.60; leg I: coxa 0.33/0.27, trochanter 0.12-0.13/0.10, femur I
0.12/0.12, femur II 0.17/0.12, tibia 0.21/0.075, tarsus 0.12-0.13/0.05; leg IV: coxa
width 0.24-0.26, trochanter 0.15-0.17/0.13-0.15, femur I 0.22/0.20, femur II 0.26/
0.20-0.21, tibia 0.35-0.36/0.12-0.13, tarsus 0.19/0.08-0.085.

174

THE JOURNAL OF ARACHNOLOGY

Figs. 1 S-1 9 .-Afrosternophorus aethiopicus (Beier), male holotype; 78, ventral aspect of right
pedipalp; 79, dorsal aspect of carapace. Scale line = 1.00 mm.

Habitat— No habitat data accompanied the specimen.

Remarks.— The single available specimen is in poor condition and the sclerotized
portions of the genitaha are ill-defined. The left pedipalpal tibia and chela were missing
from the specimen.

Afrosternophorus ceylonicus (Beier), new combination
Figs. 53, 65-66, 80-85, 90; Map 4

Sternophonis ceylonicus Beier 1973b:47, Fig. 11.

Sternophorus indicus Murthy and Ananthakrishnan 1977:119-121, Fig. 39. NEW SYNONYMY.
Sternophorus (Sternophonis) transiens Murthy and Ananthakrishnan 1977:121-123, Fig. 40. NEW
SYNONYMY.

Sternophorus (Sternophorus) montanus Sivaraman 1981:315-317, Fig. 1. NEW SYNONYMY.
Sternophorus (Afrosternophorus) femoratus Sivaraman 1981:317-319, Fig. 2. NEW SYNONYMY.
Sternophorus (Afrosternophorus) intermedius Sivaraman 1981:319-321, Fig. 3. NEW SYNONYMY.
Indogaryops amrithiensis Sivaraman 1981:322-324, Fig. 4. NEW SYNONYMY.

Types.— Sternophonis ceylonicus: Lectotype male (present designation), paralectotype
female, Cherniy anpattu, 18 mi. [= 29 km] SE of Point Pedro, North Province, Sri Lanka,
under bark of tree-like bush, 13 February 1962 (Brink, Anderson, Cederholm), LU, Type
No. 539 (spirit). Paralectotype male, same data as above, NHMW (spirit). Two paralecto-
type males, paralectotype female, paralectotype tritonymph. Per Aru, 9 mi. [= 14.5 km]
E of Mankulan, North Province, Sri Lanka, under log, 14 February 1962 (Brink, Ander-
son, Cederholm), LU, Type No. 539 (slides and spirit). Two males, same data as above.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

175

Z

Figs. S0-S5.— A frosternophoms ceylonicus (Beier): 80, ventral aspect of left pedipalp, female
paralectotype from Chemiyanpattu, Sri Lanka; 81, same, male lectotype; 82, ventral aspect of right
pedipalp, tritonymph paralectotype from Per Aru, Sri Lanka; 83, dorsal aspect of carapace, male
paralectotype from Per Aru; 84, lateral aspect of left chela, female paralectotype from Per Aru;
85, female genitalia and associated sternites, paralectotype from Per Aru. Scale line = 1.00 mm (Figs.
80-84), 0.25 mm (Fig. 85).

176

THE JOURNAL OF ARACHNOLOGY

NHMW (spirit). Two paralectotype males, two paralectotype females, 5 mi. [= 8 km]
NNE of Puttalam, North-West Province, Sri Lanka, 1 February 1962 (Brink, Anderson,
Cederholm), NHMW (spirit). Five paralectotype males, five paralectotype females, same
data as above, LU, lost (see Remarks).

Sternophoms indicus: Three paratype males, Tirupathi, Andhra Pradesh, India, under
bark, 14 August 1960 (V. A. Murthy), VAM (slides).

Sternophoms transiens: Paratype male, paratype female, Shimoga, Karnataka, India,
under bark, 4 January 1963 (V. A. Murthy), VAM (slides).

Sternophoms montanus: Holotype male, paratype female, Alakarkoil Hill forest,
Madurai, Tamil Nadu, India, under bark, 15 July 1977 (S. Sivaraman), MHNG (slides).

Sternophoms femoratus: Holotype female, paratype male, Amrithi forest. North
Arcot, Tamil Nadu, India, under bark, 2 October 1977 (S. Sivaraman), MHNG (slides).

Sternophoms intermedins: Holotype female, paratype male, Alakarkoil Hill forest,
Madurai, Tamil Nadu, India, under bark, 15 July 1977 (S. Sivaraman), MHNG (slides).

Indogaryops amrithiensis: Holotype female, Amrithi forest. North Arcot, Tamil Nadu,
India, under bark, 2 October 1977 (S. Sivaraman), MHNG (slide).

Distribution.— India, Sri Lanka (Map 4).

Diagnosis.— Male genitalia with short, but prominent, slightly curved dorsal apodemes;
lateral rod lying anteriorly to the rest of the genital armature. Female galea with two
distal and one sub distal to subbasal rami. Chelal fingers long and strongly curved. Chela
(with pedicel) 0.75 to 0.835 (male), 0.80 to 0.94 mm (female) in length.

Description.— ADULTS: Pedipalpal trochanter 1.63 to 1.79 (male), 1.64 to 1.84
(female), femur 2.47 to 2.86 (male), 2.48 to 2.81 (female), tibia 2.00 to 2.29 (male),
1.95 to 2.29 (female), chela (with pedicel) 3.39 to 3.76 (male), 3.35 to 3.72 (female),
chela (without pedicel) 3.20 to 3.56 (male), 3.20 to 3.50 (female) times longer than
broad. Chelal fingers long and strongly curved (Figs. 80-81). Trichobothria as for aethi-
opicus group, in usual position (Figs. 80-81 , 84). Serrula exterior of chelicera with 10 to
12 (male), 1 1 to 13 (female) lamellae. Galea of male simple, of female with two distal and
one subdistal to subbasal rami (Figs. 65-66). Carapace usually unconstricted (Fig. 83), but
sometimes a slight constriction is present, with 20 to 30 (male), 22 to 28 (female) setae;
1.26 to 1.46 (male), 1.21 to 1.33 (female) times longer than broad. Male genitalia (Fig.
53) with short, but prominent, slightly curved dorsal apodemes; lateral rod lying anterior-
ly to the rest of the genital armature. Female genitalia as for genus (Fig. 85). Tergal
chaetotaxy: male, 5-6:4-6:3-4:6-7:5-7:5-7:6-7:5-7:4-8:TlT3-5TlT:?:2; female, 5-7:4-6:
3-5:5-7:4-7:6-8:5-7:6-8:6-8:TlT34TlT;?:2. Sternal chaetotaxy: male, 0:4-8 :(0)3 4 [4-6]
(0):(l)4-5(l):5-8:6-7:6:6-7:6:TlT3-4TlT:?:2; female, 0:7-8:(0)4(0):(l)4-6(l):5-8:5-8:
6-7:6-7:4-8:TlT4TlT:?:2. Coxal chaetotaxy: male, 3-5 :3-6:3-5:3-5; female, 3-5 :3-5:2-5:
3-6.

Dimensions (mm): Body length 1. 7-2.1 (2. 1-2.9); pedipalps: trochanter 0.28-0.32/
0.16-0.195 (0.295-0.36/0.17-0.20), femur 0.465-0.535/0.175-0.205 (0.495-0.605/0.18-
0.225), tibia 0.375-0.44/0.175-0.205 (0.40-0.51/0.185-0.225), chela (with pedicel)
0.75-0.835/0.205-0.24 (0.80-0.94/0.22-0.27), chela (without pedicel) 0.72-0.80 (0.755-
0.92), moveable finger length 0.37-0.43 (0.41-0.48); chelicera 0.155-0.175/0.09-0.095
(0.18-0.185/0.10-0.115), moveable finger length 0.11-0.13 (0.12-0.14); carapace 0.68-
0.76/0.51-0.595 (0.74-0.89/0.56-0.68); leg I: coxa 0.19-0.23/0.235-0.255/0.26-0.30),
trochanter 0.095-0.135/0.09-0.10 (0.12-0.15/0.095-0.11), femur I 0.095-0.115/0.10-0.12
(0.12-0.14/0.11-0.135), femur II 0.15-0.18/0.10-0.12 (0.17-0.20/0.11-0.135), tibia
0.17-0.21/0.065-0.08 (0.20-0.22/0.075-0.085), tarsus 0.105-0.13/0.05-0.055 (0.13-0.14/

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

177

0.05-0.055); leg IV: coxa width 0.21-0.26 (0.275-0.305), trochanter 0.14-0.175/0.105-
0.12 (0.165-0.195/0.12-0.13), femur I 0.17-0.195/0.14-0.185 (0.215-0.24/0.175-0.18),
femur II 0.22-0.26/0.14-0.19 (0.26-0.28/0.18-0.185), tibia 0.27-0.33/0.10-0.115 (0.32/
0.105), tarsus 0.17-0.185/0.06-0.075 (0.18/0.07).

TRITONYMPHS: Pedipalpal trochanter 1.66 to 1.68, femur 2.44 to 2.52, tibia
2.00 to 2.03, chela (with pedicel) 3.47 to 3.63, chela (without pedicel) 3.32 to 3.39 times
longer than broad. Fixed finger with seven trichobothria, moveable finger with two
trichobothria (Fig. 82); it, sb and st absent. Serrula exterior of chelicera with 10 lamellae.
Galea as for female. Carapace unconstricted, with 20 setae; 1.25 times longer than broad.
Tergal chaetotaxy: 4:4:6:6:6?:6:6:6:6:T1T3T1T:?:2. Sternal chaetotaxy: 0:2:(0)4(0):
(1)4(1):6:6:6:6:6:T1T3T1T:?:2. Coxal chaetotaxy; 4:3-4:3:2.

Dimensions (mm): Body length 1.9; pedipalps: trochanter 0.235-0.24/0.14-0.145,
femur 0.39/0.155-0.16, tibia 0.31-0.315/0.155, chela (with pedicel) 0.66-0.69/0.19,
chela (without pedicel) 0.63-0.645, moveable finger length 0.32-0.34; carapace 0.69/
0.55.

Habitat— The Chemiyanpattu specimens were taken from under bark of a “tree-like
bush” and the Per Aru specimens were taken from under logs in jungle. Murthy and
Ananthakrishnan’s and Sivaraman’s material was all taken from under bark.

Remarks.— Unfortunately, one vial containing type material (5 males, 5 females,
Puttalam, Sri Lanka) was lost in transit from Lund to Monash University.

Beier (1973b) did not designate a primary type, and merely published and labelled
some Chemiyanpattu specimens as “Typen”. A lectotype male has been selected from
this vial.

Through the characteristic generosity of Prof. V. A. Murthy, I have been able to
examine some type specimens of S. indicus and S. transiens, as well as many other speci-
mens from southern India. The type material of S. montanus, S. femoratus, S. intermed-
ius and /. amrithiensis was also available for study. All of this material possesses the
characteristic chelal finger shape, female galea and, most importantly, male genitalia of A.
ceylonicus, and therefore these species are hereby synonymized with ceylonicus. Murthy
and Ananthakrishnan (1977) erroneously included the chelal trichobothria it and st in
their description and diagram of S. indicus. My examination of three paratypes reveals

that these trichobothria are absent, as in all sternophorids.

Other specimens examined. -INDIA: TAMIL NADU; Alakarkoil Hill forest, Madurai, under bark,
15 July 1977 (S. Sivaraman), 3 males, 1 female (VAM) (spirit). Amrithi forest. North Arcot, under
bark, 2 October 1977 (S. Sivaraman), 1 male, 1 female (VAM) (spirit). Same data as above except,
date? (V. A. Murthy), 3 males, 1 tritonymph (ANIC, MH472. 01-04) (spirit). No locality data, 2 males,
2 females (VAM) (spirit).

Afrosternophoms chamberlini (Redikorzev), new combination

Figs. 54, 67, 86-90; Map 4

Sternophorus chamberlini Redikorzev 1938:89-91, Figs. 17-18; Beier 1951:71-72, Fig. 16 (in part).

Types.— Lectotype female (present designation), Dalat [= Da Lat, see Table 1], Viet-
nam, January 1931 (C. Dawydoff), MNHN (spirit). Paralectotype female, same data as
lectotype except 3 February 1931, MNHN (spirit). One specimen, Abre-Broye [= Ap
Tram Hahn] , Plateau du Lang-Biang [= Cao Nguyen Lam Vien] , Vietnam, 1,500 m, 20
January 1931 (C. Dawydoff), depository unknown, not examined.

Distribution.— Vietnam, Laos? (Map 4).

178

THE JOURNAL OF ARACHNOLOGY

Figs. ^6-S9.- Afrostemophorus chamberlini (Redikorzev): 86, ventral aspect of right pedipalp,
female; 87, same, male; 88, dorsal aspect of carapace, male; 89, female genitalia and associated ster-
nites. Scale line = 1.00 mm (Figs. 86-88), 0.25 mm (Fig. 89).

Diagnosis.— Male genitalia with long, acute dorsal apodemes; mid-piece of lateral rod
elongate. Female galea with two distal, one subdistal and three subbasal rami. Chela (with
pedicel) 0.74 to 0.81 (male), 0.74 to 0.90 mm (female) in length.

Description.— Pedipalpal trochanter 1.79 to 1.87 (male), 1.75 to 2.00 (female), femur
3.00 to 3.16 (male), 2.73 to 3.17 (female), tibia 2.18 to 2.39 (male), 2.13 to 2.35 (fe-
male), chela (with pedicel) 3.68 to 3.95 (male), 3.47 to 4.20 (female), chela (without
pedicel) 3.55 to 3,74 (male), 3.30 to 3.98 (female) times longer than broad, Trichobo-
thria as for aethiopicus group, in usual position (Figs. 86-87). Serrula exterior of chelicera
with 10 to 11 (male), 11 to 12 (female) lamellae. Galea of male simple, of female with
two distal, one subdistal and three subbasal rami (Fig. 67). Carapace (Fig. 88) uncon-
stricted, with 23 (male), 22 to 28 (female) setae; 1.50 to 1.51 (male), 1.44 to 1.56 (male)
times longer than broad. Male genitalia with long, acute dorsal apodemes; mid-piece of
lateral rod elongate (Fig. 54). Female genitalia as for genus (Fig. 89). Tergal chaetotaxy;
male, 6:5 :2:6:6:6:7:6:7:T1T4T1T:?:2; female, 6:4-6:24 :6:6-8?:6-7:5-8:5-8:6-8:TlT4
T1T:?:2. Sternal chaetotaxy: male, 0:5 :(0)5 [8] (0):(1)6(1):8:8:6:7:6:T1T4T1T:?:2;
female, 0:6-9:(0)5-6(0):(l)6(l):5?-9:4-9:7-8:6:6-7:TlT4TlT:?:2. Coxal chaetotaxy:
male, 4-5:3-5:5:4; female, 3-6:4-6:3-6:3-5.

Dimensions (mm): Body length 1.6-1. 8 (2.0-2. 7); pedipalps: trochanter 0.285-0.305/
0.155-0.17 (0.28-0.36/0.16-0.19), femur 0.48-0.53/0.155-0.175 (0.45-0.59/0.165-0.205),
tibia 0.37-0.44/0.165-0.185 (0.37-0.49/0.165-0.215), chela (with pedicel) 0.74-0.81/
0.195-0.22 (0.74-0.90/0.20-0.25), chela (without pedicel) 0.705-0.78 (0.705-0.865),
moveable finger length 0.345-0.38 (0.36-0.40); chelicera 0.13-0.16/0.08-0.09 (0.15-0.17/
0.08-0.10), moveable finger length 0.105-0.115 (0.105-0.12); carapace 0.665-0.69/

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

179

0.44-0.46 (0.64-0.86/0.49-0.555); leg I: coxa 0.185-0.19/0.20-0.21 (0.195-0.22/
0.21-0.25), trochanter 0.105-0.11/0.07 (0.11-0.12/0.075-0.08), femur I 0.10/0.105-0.11
(0.105-0.12/0.095-0.11), femur II 0.14/0.10-0.11 (0.15-0.16/0.095-0.105), tibia 0.16-
0.165/0.065 (0.18/0.07), tarsus 0.10-0.11/0.045 (0.11-0.12/0.045); leg IV: coxa width
0.18-0.21 (0.21-0.24), trochanter 0.12-0.155/0.09-0.10 (0.15-0.165/0.10-0.11), femur I
0.175-0.20/0.15-0.18 (0.205/0.165), femur II 0.205-0.255/0.14-0.18 (0.22/0.165), tibia
0.275-0.32/0.09-0.10 (0.275-0.32/0.09-0.10), tarsus 0.165-0.17/0.065 (0.16-0.17/0.06-
0.07).

Habitat.— No habitat data accompanied the specimens.

Remarks.— Redikorzev did not designate a holotype and thus a lectotype has been
selected. Beier’s (1951) material was composed of two species: chamberlini fallax,
new species. Afrosternophorus fallax is substantially different from A. chamberlini and
can be separated by several characters, particularly the presence of only two trichobothria
on the moveable chelal finger. Beier’s (1951) description of chamberlini was a composite
of these two species. His male pedipalp measurements were of fallax, and his female
measurements were of chamberlini.

Beier (1951) erroneously included the chelal trichobothrium it in his diagram of this
species. He also recorded one female from Plateau des Bolovens, Laos. This specimen
needs to be re-examined to determine its true status. A deutonymph of Stenatenmus
annamensis Beier (?) was present in one vial.

The NHMW material is in poor condition, and few specimens could be fully scored for
setation or leg measurements.

Specimens examined.— VIETNAM: Plateau von Langbian [= Cao Nguyen Lam Vien], 1938-1939
(C. Dawydoff), 3 males, 9 females (NHMW) (spirit).

Afrosternophorus dawydoffi (Beier), new combination
Figs. 55, 68, 90-97 ; Map 4

Sternophorus dawydoffi Beier 1951 : 70-7 1 , Fig. 1 5 .

Sternophorus cylindrimanus Beier 1951:73-74, Fig. 17 (in part).

Sternophorus dawydoffi: Lectotype male (present designation), four paralec-
totype males, two paralectotype females, Rusei Chrum [= Roessei Chrum, see Table 1 ] ,
Kampuchea, March 1939 (C. Dawydoff), NHMW (spirit). Paralectotypes: one male, one
female, same data as above, ANIC, MH430.01-02 (slides). One male, two females, same
locality as above, April 1939 (C. Dawydoff), NHMW (spirit). One male, three females,
Beng Mealea [= Phumi Boeng Mealea] , Kampuchea, April 1939 (C. Dawydoff), NHMW
(spirit). Two females, Phailin [= Pailin] , Kampuchea, March 1939 (C. Dawydoff), NHMW
(spirit). Two males, three females, Prah Khan [= Prasat Preah Khan] , Kampuchea, April
1939 (C. Dawydoff), NHMW (spirit). Five males, two females. Ream [= Phsar Ream],
Kampuchea, April 1939 (C. Dawydoff), NHMW (spirit). Two males, one female, Sre
Umbell [= Sre Ambel] , Kampuchea, March 1939 (C. Dawydoff), NHMW (spirit). One
male, three females, Insel Phu-Quoc [=Dao Phu Quoc] , Vietnam, March 1939 (C. Dawy-
doff), NHMW (spirit).

Sternophorus cylindrimanus: Paralectotype female, Krongpha [= Thon Song Pha] ,
Vietnam, 30 April 1939 (C. Dawydoff), NHMW (spirit).

Distribution.— Kampuchea, Vietnam (Map 4).

Diagnosis.— Male genitalia with long, acute dorsal apodemes. Female galea with six
(sometimes five) distal to subdistal rami. Large species: chela (with pedicel) 1.07 to 1.35
(male), 1.11 to 1 .3 8 mm (female) in length.

180

THE JOURNAL OF ARACHNOLOGY

Figs. 90. -Graph of chela (with pedicel) length (CL) versus width (CW), in mm; open symbols,
males; closed symbols, females: Afrosternophoms ceylonicus (Beier) (inverted triangles),/!, chamber-
lini (Redikorzev) (squares), A. dawydoffi (Beier) (upright triangles), A. dawydoffi, paralectotype
female of A. cylindrimanus (Beier) (crosses), A. cylindrimanus (circles).

Description.— Pedipalpal trochanter relatively inflated, 1.88 to 2.27 (male), 1.81 to
2.02 (female), femur elongate, 3.18 to 3.72 (male), 2.96 to 3.30 (female), tibia 2.60 to
2.88 (male), 2.35 to 2.63 (female), chela (with pedicel) 4.00 to 4.46 (male), 3.67 to 4.13
(female), chela (without pedicel) 3.84 to 4.40 (male), 3.44 to 3.97 (female) times longer
than broad. Trichobothria as for aethiopicus group, in usual position (Figs. 91-93).
Serrula exterior of chelicera with 12 to 15 (male), 11 to 15 (female) lamellae. Galea of
male simple, of female with six (occasionally five) distal to subdistal rami (Fig. 68).
Carapace usually unconstricted (Fig. 95), but sometimes a slight (Fig. 96) or a conspicu-
ous constriction is present (Fig. 97), with 24 to 32 (male), 24 to 31 (female) setae; 1.24
to 1.52 (male), 1.30 to 1.55 (female) times longer than broad. Male genitalia with long,
acute dorsal apodemes (Fig. 55). Female genitalia as for genus (Fig. 94). Tergal chaeto-
taxy; male, 4-6:4-6:34:4-6:5-6:5-7;6;6-7;6:TlT4-5TlT:?:2; female, 4-6:5-6;2-4;4-7:6:
5.6:4-7:5-7:5-6:T1T4T1T:?:2. Sternal chaetotaxy: male, 0;2-6:(0)4-6[7-8] (0):(1)4-6(1):
4-9:6-8:5-6:5-8;5-7:TlT4TlT:?:2; female, 0;5-9:(0)4-5(0):(l)4-6(l):6:5-6:6:6-7:5-7;
T1T4T1T:?;2. Coxal chaetotaxy: male, 3 -6:3 -6:3 -7:3 -5; female, 3-6:3-6:3-6:3-5.

Dimensions (mm): Body length 2.0-3.3 (2.4-3.3); pedipalps: trochanter 0.43-0.545/
0.215-0.265 (0.44-0.555/0.23-0.29), femur 0.72-0.93/0.21-0.27 (0.74-0.95/0.23-0.295),
tibia 0.60-0.77/0.22-0.275 (0.595-0.76/0.24-0.305), chela (with pedicel) 1.07-1.35/
0.24-0.305 (1.11-1.38/0.275-0.35), chela (without pedicel) 1.02-1.275 (1.05-1.32),
moveable finger length 0.47-0.61 (0.53-0.66); chelicera 0.20-0.23/0.11-0.13 (0.21-0.28/

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

181

0.12-0.14), moveable finger length 0.14-0.16 (0.15-0.175); carapace 0.90-1.13/0.645-
0.795 (0.96-1.215/0.71-0.83); leg 1: coxa 0.25-0.31/0.29-0.36 (0.275-0.35/0.32-0.41),
trochanter 0.145-0.185/0.105-0.125 (0.15-0.20/0.11-0.14), femur 1 0.13-0.165/0.14-
0.175 (0.15-0.19/0.15-0.19), femur 11 0.24-0,30/0.14-0.18 (0.25-0.315/0.15-0.19),
tibia 0.25-0.29/0.085-0.115 (0.26-0.32/0.095-0.115), tarsus 0.12-0.175/0.06-0.07 (0.15-
0.195/0.06-0.075); leg IV: coxa width 0.27-0.33 (0,29-0.37), trochanter 0.20-0.25/0.14-
0.16 (0.21-0.275/0.14-0.175), femur I 0.255-0.305/0.22-0.285 (0.27-0.345/0.25-0.325),
femur II 0.39-0,51/0.225-0.285 (0.40-0.50/0.25-0.33), tibia 0.405-0.51/0.13-0.165
(0.45-0.56/0.14-0.18), tarsus 0.205-0.255/0.08-0.10 (0.23-0.27/0,075-0.105).

Habitat.— No habitat data accompanied the specimens.

\ / /

Figs. 9\-9A.-Afrostemophoms dawydoffi (Beier): 91, dorsal aspect of left pedipalp, male lecto-
type; 92, ventral aspect of right pedipalp, female paralectotype from Roessei Chrum, Kampuchea; 93,
lateral aspect of left chela, male paralectotype, MH430.01; 94, female genitalia and associated ster-
nites, paralectotype from Roessei Chrum. Scale line = 1.00 mm (Figs. 91-93), 0.25 mm (Fig. 94).

182

THE JOURNAL OF ARACHNOLOGY

Figs. 95-91 .-Afrosternophoms dawydoffi (Beier), dorsal aspect of carapace, specimens from
Roessei Chrum, Kampuchea: 95, female paralectotype; 96, male lectotype; 97, female paralectotype.
Scale line = 1.00 mm.

Remarks.— The anterior constriction of the carapace of this species is quite variable
(Figs. 95-97; all specimens from Roessei Chrum) and clearly shows that this character
cannot be used at the generic level.

The female paralectotype of S. cylindrimanus referred to above is discussed under that
species.

Beier (1951) erroneously included the chelal trichobothrium it in his diagram of this
species, as well as mis-sexing many specimens.

Beier did not designate a primary type in the original description, and merely pub-
lished and labelled one vial as “Typen”. A lectotype male has been selected from this vial.

The alternative spelling of the localities given above is discussed in the Materials and
Methods and Table 1.

Afrosternophoms cylindrimanus (Beier), new combination
Figs. 69, 90, 98-100; Map 4

Sternophorus cylindrimanus Beier 1951:73-74, Fig. 17 (in part).

Types.— Lectotype male (present designation), paralectotype male, paralectotype
female, Paclay [= Pak-Lay, see Table 1], Laos, January 1938 (C. Dawydoff), NHMW
(spirit).

Distribution.— Laos (Map 4).

Diagnosis.— Male genitalia with long, acute dorsal apodemes. Female galea with four
distal and one subbasal rami. Chela (with pedicel) 0.935 to 0.95 (male), 1.02 mm (fe-
male) in length.

Description.— Pedipalpal trochanter 1.95 to 2.00 (male), 2.00 to 2.05 (female), femur
3.31 to 3.37 (male), 3.42 to 3.50 (female), tibia 2.48 to 2.60 (male), 2.48 (female), chela
(with pedicel) 4.37 to 4.42 (male), 4.34 to 4.53 (female), chela (without pedicel) 4.19 to
4.26 (male), 4.15 to 4.31 (female) times longer than broad. Trichobothria as for aethiopi-
cus group, in usual position (Figs. 98-99). Serrula exterior of chelicera with 9 to 10
(male), 12 (female), lamellae. Galea of male simple, of female with four distal and one
subbasal rami (Fig. 69). Carapace (Fig. 100), with 23 to 24 (male), ? (female) setae; 1.34

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

183

Map 4. -South-east Asia showing known distribution of Afrosternophoms ceyloniciis (Beier)
(circles), A. chamberlini (Redikorzev) (squares), A. dawydoffi (Beier) (upright triangles), A. cylindri-
manus (Beier) (inverted triangles) and A. fallax, new species (star). Open symbols represent hterature
records only.

to 1.47 (male), 1.41 (female) times longer than broad. Male genitalia very obscure, dorsal
apodemes long and acute. Female genitalia as for genus. Tergal chaetotaxy: male, 5-6: 5-6:
3-4:6:5-6:5-6:5-6:6:4-5:TlT?TlT:?:2; female, ?:?:?:?:?:6:6:6:6:T1T4T1T:?:2. Sternal
chaetotaxy: male, ?; female, 0:?:(0)?(0):(?)4(1):6:5 :?:?:6:T1T4T1T:?:2. Coxal chaeto-
taxy: male, 4:3 :3-4:2; female, ?.

Dimensions (mm): Body length 2. 1-2.2 (2.5); pedipalps: trochanter 0.37-0.39/0.19-
0.195 (0.41-0.42/0.205), femur 0.59-0.605/0.175-0.18 (0.615-0.63/0.18), tibia 0.515-
0.52/0.20-0.21 (0.52/0.21), chela (with pedicel) 0.935-0.95/0.215 (1.02/0.225-0.235),
chela (without pedicel) 0.88-0.915 (0.97-0.975); chelicera 0.16/0.08-0.095 (?/?), move-
able finger length 0.10-0.12 (?); carapace 0.75-0.765/0.52-0.56 (0.86/0.61); legs: ?.

Habitat.— No habitat data accompanied the specimens.

Remarks.— This species is very similar to A. dawydoffi, from which it can be separated
only by its smaller size (Fig. 90) and the form of the female galea (Fig. 69). The latter
character appears to be slightly variable, and may prove to be of little significance for
these two species. Only further collecting will determine whether A. cylindrimanus is
merely a smaller form of A. dawydoffi, and hence synonymous with it.

Beier (1951) described this species from two other specimens: a female from Louang-
phrabang, Laos, and a female from Thon Song Pha, Vietnam. The former specimen was
not available for study, and should be re-examined to determine its true status. It is
geographically intermediate between the known distributions of A. cylindrimanus and A.
dawydoffi, and may resolve the probable synonymy of these two species.

184

THE JOURNAL OF ARACHNOLOGY

Figs. 9'&-\Q{i.-Afrostemophoms cylindrimamis (Beier): 98, ventral aspect of left pedipalp, male
lectotype; 99, dorsal aspect of right pedipalp, female paralectotype; 100, dorsal aspect of carapace,
male lectotype. Scale line = 1.00 mm.

Beier identified the Thon Song Pha specimen as cylindrimanus because it possessed a
broad pseudosternum, which was his criterion for separating dawydoffi and cylindri-
manus. My observations indicate that the relative sizes of the pseudosternum are variable
and must be viewed with caution. The following values were obtained for the width of
the pseudosternum over the width of the second coxa [the ratio used by Beier (1951)] :
3.00 to 6.00 for 1 1 specimens of A. dawydoffi; 1.65 to 2.12 for the two Pak-Lay males
of A. cylindrimanus (the female was damaged and unmeasurable); and 2.08 for the Thon
Song Pha female. Nevertheless, on the basis of pedipalpal morphometries (Fig. 90) and
distribution (Map 4), it is believed that the latter specimen is better placed in A. dawydof-

fi-

The Pak-Lay specimens are in poor condition, and many characters could not be
properly scored or drawn (e.g. genitalia).

Beier (1951) erroneously included the chelal trichobothrium it in his diagram of this
species.

Beier (1951) did not designate a primary type in his original description, and merely
published and labelled one vial as “Typen”. A lectotype male has been selected from this
vial.

The alternative spelling of the type locality is discussed in the Materials and Methods
and Table 1 .

Afrostemophorus hirsti (Chamberlin), new combination
Figs. 1-7, 51, 56, 70-71, 101-108, 129;Map 5

Sternophoms hirsti Chamberlin 1932a:143; Beier 1932:18; Harvey 1981b:244.

“Sternophorus” hirsti Chamberlin: Harvey 1982:192, Fig. 1.

Type.— Holotype male, Barringun, New South Wales, Australia, 1927 (F. S. Hirst),
JCC, JC-480.01001 (slide).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

185

Distribution— New South Wales, Queensland, Australia (Map 5).

Diagnosis.— Male genitalia with brush-like anterior apodeme; dorsal apodemes parallel
sided, much reduced. Female galea with three distal, one subdistal and one subbasal rami,
the subdistal ramus occasionally absent. Chela (with pedicel) 0.715 to 0.815 (male), 0.66
to 0.91 mm (female) in length.

Description.— ADULTS: Pedipalpal trochanter 1.64 to 1.90 (male), 1.60 to 1.94
(female), femur 2.42 to 2.89 (male), 2.33 to 2.89 (female), tibia 2.05 to 2.44 (male),
1.96 to 2.35 (female), chela (with pedicel) 3.25 to 3.77 (male) 3.08 to 3.71 (female).

Figs. \{)\-\0^.—Afrosternophorus hirsti (Chamberlin): 101, female genitalia and associated ster-
nites, MH302.24; 102, ventral aspect of right pedipalp, female, MH237.02; 103, ventral aspect of left
pedipalp, male holotype; 104, ventral aspect of right pedipalp, tritonymph, MH237.05; 105, same,
deutonymph, MH302.57; 106, same protonymph, MH237.07; 107, lateral aspect of right chela,
male holotype; 108, dorsal aspect of carapace, female, MH210.03. Scale line = 1.00 mm (Figs. 102-
108), 0.25 (Fig. 101).

186

THE JOURNAL OF ARACHNOLOGY

chela (without pedicel) 3.11 to 3.58 (male), 2.94 to 3.55 (female) times longer than
broad. Trichobothria as for aethiopicus group, in usual position (Figs. 1, 102-103, 107).
One male (MH302.15) has sb missing from one chela. Serrula exterior of chelicera with
11 to 12 (male), 10 to 12 (female) lamellae. Galea of male simple, of female with three
distal, one subdistal and one subbasal rami, the subdistal ramus occasionally absent (Figs.
70-71). Carapace (Figs. 2, 108) usually unconstricted, but sometimes a slight constriction
is apparent, with 20 to 24 (male), 20 to 27 (female) setae; 1.43 to 1.67 (male), 1.32 to
1.55 (female) times longer than broad. Male genitalia (Figs. 51, 56) with distinctive
brush-like anterior apodeme, dorsal apodemes parallel sided, much reduced; foramen
relatively large. Female genitalia as for genus (Fig. 101). Tergal chaetotaxy: male, 4-7;
3.6:4-6:4-7:5-7:5-7:6-7:5-7:5-6:T1T22-5T1T:?:2; female, 4-7:4-6:3-6:5-8:5-8:6-7;5-7:
6-7:6-7:TlT2-4TlT:?:2. Sternal chaetotaxy: male, 0:3-8:(0)4-5 [5-8] (0):(l)3-6(l);5-8:
5-7;5.7:4.7;5.6;T1T3-4T1T:?:2; female, 0:7-1 1 :(0)3-5(0):(l)4-6(l):5-9:5-8:5-7:3-8:5-8:
T1T3-4T1T:?:2. Coxal chaetotaxy: male, 4-5:5:2-4:4-5; female, 4-6:4-6:3-5:3-5.

Dimensions (mm): Body length 2.0-2.3 (1.8-2. 9); pedipalps: trochanter 0.275-0.33/
0.155-0.18 (0.255-0.36/0.15-0.225), femur 0.46-0.545/0.165-0.21 (0.41-0.60/0.16-0.24),
tibia 0.39-0.475/0.175-0.22 (0.36-0.51/0.16-0.26), chela (with pedicel) 0.715-0.815/
0.20-0.24 (0.66-0.91/0.19-0.28), chela (without pedicel) 0.68-0.775 (0.63-0.87), move-
able finger length 0.33-0.38 (0.315-0.42); chelicera 0.15-0.16/0.08-0.095 (0.135-0.19/
0.08-0.11), moveable finger length 0.10-0.11 (0.095-0.13); carapace 0.67-0.80/0.45-0.52
(0.65-0.87/0.47-0.65); leg I: coxa 0.19-0.20/0.22-0.23 (0.185-0.25/0.20-0.29), trochanter
0.09-0.11/0.08-0.09 (0.08-0.13/0.07-0.11), femur I 0.10-0.11/0.095-0.105 (0.085-0.13/
0.09-0.13), femur II 0.145-0.16/0.10-0.105 (0.13-0.19/0.09-0.13), tibia 0.17-0.19/
0.065-0.07 (0.15-0.22/0.06-0.08), tarsus 0.07-0.095/0.045-0.05 (0.095-0.13/0.045-
0.055); leg IV: coxa width 0.18-0.24 (0.19-0.27), trochanter 0.125-0.14/0.10-0.11
(0.125-0.17/0.09-0.125), femur I 0.15-0.19/0.165-0.175 (0.15-0.22/0.145-0.225), femur
II 0.22-0.26/0.16-0.175 (0.21-0.32/0.145-0.225), tibia 0.275-0.31/0.09-0.105 (0.255
-0.37/0.085-0.12), tarsus 0.16/0.06-0.07 (0.14-0.19/0.055-0.075).

TRITONYMPHS: Pedipalpal trochanter 1.33 to 1.83, femur 2.32 to 2.72, tibia 1.94 to
2.30, chela (with pedicel) 3.10 to 3.84, chela (without pedicel) 3.00 to 3.65 times longer
than broad. Fixed finger with seven trichobothria, moveable finger with two trichobo-
thria (Fig. 104); it, sb and st absent. Serrula exterior of chelicera with 10 to 11 lamellae.
Galea as for female. Carapace with 21 to 23 setae; 1.41 to 1.55 times longer than broad.
Tergal chaetotaxy: 4-6:4-5:4:5-6:5-6:6:6:6:6:TlT2TlT:?:2. Sternal chaetotaxy: 0:2:
(0)4(0):(l)4(l):6:6:5-6:5-6:6:TlT2TlT:?:2. Coxal chaetotaxy: 3-5:4-5:3:3.

Dimensions (mm): Body length 1.5-2. 2; pedipalps: trochanter 0.22-0.265/0.125-
0.165, femur 0.34-0.36/0.135-0.165, tibia 0.305-0.36, chela (with pedicel) 0.59-0.72/
0.155-0.195, chela (without pedicel) 0.56-0.685, moveable finger length 0.275-0.33;
carapace 0.61-0.71/0.39-0.48.

DEUTONYMPH: Pedipalpal trochanter 1.40 to 1.71, femur 2.08 to 2.25, tibia 1.85 to
1.92, chela (with pedicel) 3.13, chela (without pedicel) 3.10 times longer than broad.
Fixed finger with six trichobothria, moveable finger with two trichobothria (Fig. 105); it,
esb, sb and st absent. Serrula exterior of cheHcera with 9 lamellae. Galea as for female.
Carapace with 16 setae; 1.43 times longer than broad. Tergal chaetotaxy: 4:4:4:5:4:4:4:
4:4:T1T2T1T:?:2. Sternal chaetotaxy: 0:0:(0)4(0):(1)4(1):4:4:4:4:4:T1T4T1T:?:2.
Coxal chaetotaxy: 3:3;2:2.

Dimensions (mm): Body length 1.5; pedipalps: trochanter 0.175-0.18/0.105-0.125,
femur 0.27/0.12-0.13, tibia 0.24/0.125-0.13, chela (with pedicel) 0.485/0.155, chela
(without pedicel) 0.48, moveable finger length 0.24; carapace 0.50/0.35.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

187

PROTONYMPH: Although the single available protonymph is in poor condition,
the following observations could be made. Pedipalpal trochanter 1.63, femur 2.67, tibia
1.94, chela (with pedicel) 3.50, chela (without pedicel) 3.45 times longer than broad.
Fixed finger with three trichobothria, moveable finger with one trichobothrium (Fig.
106); eb, et, ib and t present. Serrula exterior of chelicera with 9 lamellae ; seta ^5 absent.
Carapace with 10 setae; 1.23 times longer than broad.

Dimensions (mm); Pedipalps: trochanter 0.13/0.08, femur 0.20/0.075, tibia 0.165/
0.085, chela (with pedicel) 0.385/0.11, chela (without pedicel) 0.38, moveable finger
length 0.20; carapace 0.38/0.31.

Habitat.— This species has been taken from under bark of Eucalyptus spp., Melaleuca
sp. and Gyrocarpus americanus.

Remarks.—Afrosternophorus hirsti is a widely distributed species which is easily
recognized by the form of the male genitalia.

I (Harvey 1982) recorded the presence of a nematode (Mermithidae) from a female
(MH302.51) which was collected near Tenterfield, N.S.W.

Other specimens examined.— AUSTRALIA: NEW SOUTH WALES; 32 km SW of Forbes, under
bark of Eucalyptus sp., 6 May 1981 (M. S. Harvey and M. Kotzman), 1 male (AM, KS 8365, MH
295.01) (spirit). 3 km E of Tabulam 28°53*S 152°34'E, under bark of F. sp., 23 November 1983 (M.
S. Harvey and D. C. F. Rentz), 3 males, 1 tritonymph (ANIC, MH5 36.0 1-04) (spirit). 53 km W of
Tenterfield, under bark of E. sp., 13 May 1981 (M. S. Harvey and M. Kotzman), 13 males, 9 females,
2 tritonymphs, 1 deutonymph (MV, MH302. 29-32, 35-45, 48-57) (slides and spirit). Same data as
above, 5. males, 5 females (AM, KS 8366, MH302. 01-10) (slides and spirit). Same data as above, 5
males, 5 females (QM, S994-1003, MH302. 11-20) (slides and spirit). Same data as above, 2 males, 2
females (ANIC, MH302. 21-24) (slides). Same data as above, 2 males, 2 females (VAM, MH302. 33-34,
46-47) (spirit). Same data as above, 2 males, 2 females (MHNG, MH302. 25-28) (spirit). 13 km SW of
Texas, under bark of E. sp., 1 June 1980 (C. Silveira), 1 female (AM, KS 8364, MH218.01) (slide). 16
km S of Texas, 29°00'S 151°09'E, under bark ofF. sp., 24 November 1983 (D. C. F. Rentz and M. S.
Harvey), 19 males, 17 females, 1 tritonymph (ANIC, MH538. 05-41) (spirit). 19 km SSW of Texas,
29°00'S 151°05'E, under bark of F. sp., 24 November 1983 (D. C. F. Rentz and M. S. Harvey), 2
females, 1 tritonymph (ANIC, MH539. 01-03) (spirit). 20 km SW of Texas, under bark of F. sp., 18
May 1980 (C. Silveira), 2 females (MV, MH210. 03-04) (slides): QUEENSLAND; 6 km E of Chillagoe,
under bark of Melaleuca sp., 10 December 1980 (M. Kotzman), 1 female, 1 tritonymph (MV, MH
268.01-02) (slides). Chillagoe, under bark of F. sp., 13 December 1980 (M. Kotzman), 3 females (QM,
S988-990, MH270. 01-03) (slides). 3 km W of Chillagoe, under bark of F. sp., 14 December 1980 (M.
Kotzman), 1 female (QM, S991, MH271.01) (slide). 3 km S of Chillagoe, under bark of F sp., 17
December 1980 (M. Kotzman), 1 male (QM, S992, MH272.01) (slide). 10 km NNE of Condamine,
under bark of F. sp., 28 August 1980 (M. S. Harvey), 1 male, 2 females, 3 tritonymphs, 1 protonymph
(QM, S980-986, MH273. 01-07) (slides). Davies Creek National Park, 39 km E of Mareeba, under bark
of F. sp., 23 August 1981 (N. and S. Wentworth), 1 female (QM, S1004, MH345.01) (spirit). Dome
Rock, near Chillagoe, under bark of Gyrocarpus americanus, 31 December 1982 (M. Kotzman), 5
males, 8 females (ANIC, MH474. 06-18) (spirit and points). 14 km N ofMt. Isa, under bark of F. sp.,
22 August 1980 (M. S. Harvey), 1 female (QM, S979, MH234.06) (slide). 2 km S of Rookwood
Homestead, near Chillagoe, under bark of sp., 12 December 1980 (M. Kotzman), 1 female (QM,
S987, MH269.02) (slide). 68 km WSW of Warwick, under bark of F. sp., 7 May 1981 (M. S. Harvey
and M. Kotzman), 1 female (QM, S993, MH296.01) (spirit).

Afrosternophorus nanus, new species
Figs. 57, 72, 109-113, 129; Map 5

Types.— Holotype male, three paratypes males, paratype female, paratype female?, two
paratype tritonymphs. Rum Jungle, Northern Territory, Australia, under bark of Euca-
lyptus sp., 13 August 1980 (M. S. Harvey), NTM, A1-A8, MH230.01-04, 09-12 (slides).

188

THE JOURNAL OF ARACHNOLOGY

Two paratypes males, same data as above, MV, K054-K055, MH230.07*08 (slides). Two
paratype males, same data as above, ANIC, MH230.05-06 (slides).

Etymology.— The specific epithet refers to the small size of this species {nanus L. a
dwarf).

Distribution.— Northern Territory, Australia (Map 5).

Diagnosis.— Male genitalia with long, tapering dorsal apodemes. Female galea with
three distal and one subbasal rami. Small species: chela (with pedicel) 0.55 to 0.59
(male), 0.56 to 0.61 mm (female) in length.

Description.— ADULTS: Pedipalpal trochanter 1.57 to 1.91 (male), 1.75 to 1.84
(female), femur stout 2.35 to 2.57 (male), 2.40 to 2.62 (female), tibia 1.97 to 2.21
(male), 1.94 to 2.07 (female), chela (with pedicel) 2.95 to 3.47 (male), 3.03 to 3.22
(female), chela (without pedicel) 2.82 to 3.32 (male), 2.89 to 3.05 (female) times longer
than broad. Trichobothria as for aethiopicus group, in usual position (Figs. 109-110).
Serrula exterior of chelicera with 1 1 (male, female) lamellae. Galea of male simple, of
female with three distal and one subbasal rami (Fig. 72). Carapace (Fig. 112) uncon-
stricted, with 20 to 24 (male), 21 (female) setae; 1.33 to 1.56 (male), 1.36 to 1.81
(female) times longer than broad. Male genitalia (Fig. 57) with long, tapering dorsal
apodemes which are often difficult to observe. Female genitalia as for genus (Fig. 113).

Map. 5. -Australia showing known distribution of Afrosternophonis hirsti (Chamberlin) (circles),
A. nanus, new species (squares), A. anabates, new species (triangles) and A. xalyx, new species (star).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

189

Tergal chaetotaxy: male, 4-6:3-6:3-6:4:4-5:3-5:4-6:4-7:4-7:TlT24TlT:?:2; female,
5:5:4:4:?:?:?:?:?:?:?:2. Sternal chaetotaxy: male, 0:6-7:(0)4[6-8] (0):(l)4-5(l):4-5:4-6;
4-6:4-6:4-7:T1T2-4T1T:?:2; female, 0:7:(0)4(0):(1)4(1):?:?:?:?:?:?:?:2. Coxal chaeto-
taxy: male, 3-6:4-6:3-5:4; female, 4-5:4-5 :4-5:4.

Dimensions (mm): Body length 1.6-1. 7 (?); pedipalps: trochanter 0.21-0.23/0.11-0.14
(0.21-0.23/0.12-0.125), femur 0.345-0.385/0.135-0.155 (0.36-0.38/0.145-0.155), tibia
0.295-0.33/0.14-0.16 (0.31-0.34/0.15-0.165), chela (with pedicel) 0.55-0.59/0.165-0.19
(0.56-0.61/0.18-0.19), chela (without pedicel) 0.51 5-0.565 (0.535-0.58), moveable finger
length 0.26-0.285 (0.27-0.29); chelicera 0.12-0.13/0.065-0.085 (0.13-0.135/0.08),
moveable finger length 0.07-0.09 (0.09); carapace 0.53-0.57/0.36-0.40 (0.56-0.61/0.31-
0.45); leg I: coxa 0.15-0.18/0.17-0.18 (0.16/0.19), trochanter 0.075-0.08/0.065-0.07
(0.08-0.11/0.065-0.075), femur I 0.07-0.08/0.08-0.09 (0.085-0.10/0.075-0.10), femur II
0.11-0.12/0.08-0.09 (0.105-0.13/0.075-0.10), tibia 0.135-0.15/0.055-0.065 (0.14-0.15/
0.06-0.105), tarsus 0.07-0.09/0.035-0.045; leg IV: coxa width 0.17-0.19 (?), trochanter
0.105-0.11^.075-0.085 (0.11-0.13/0.085-0.095), femur I 0.13-0.14/0.125-0.145 (0.145-
0.15/0.14-0.16), femur II 0.16-0.19/0.13-0.145 (0.18-0.20/0.145-0.16), tibia 0.215-0.23/
0.075-0.085 (0.22-0.245/0.085-0.10), tarsus 0.12-0.125/0.05 (0.115-0.15/0.055-0.075).

TRITONYMPHS: Pedipalpal trochanter 1.60 to 1.70, femur 2.29 to 2.64, tibia 2.00 to
2.19, chela (with pedicel) 3.15 to 4.07, chela (without pedicel) 3.05 to 3.93 times longer
than broad. Fixed finger with seven trichobothria, moveable finger with two trichobo-
thria (Fig, 111); it, sb and st absent. Serrula exterior of cheUcera with 10 to 11 lamellae.
Galea as for female. Carapace unconstricted, with 18 to 22 setae; 1.38 times longer than
broad. Tergal chaetotaxy: 4-5:4:34:3-6:5:6:5:5:6:TlT2TlT:?:2. Sternal chaetotaxy:
0:2:(0)4(0):(1)3(1):6:6:6:6:6:T1T2T1T:?:2. Coxal chaetotaxy: 3-4:4:2-4:3-4.

Dimensions (mm): Body length 1.5; pedipalps: trochanter 0.16-0.24/0.10-0.1 5, femur
0.28-0.39/0.11-0.17, tibia 0.23-0.36/0.115-0.175, chela (with pedicel) 0.59-0.65/0. 145-
0.205, chela (without pedicel) 0.57-0.625, moveable finger length 0.235-0,31; carapace
0.62/0.45.

Figs. 109-\\3. -Afrostemophoms nanus, new species: 109, ventral aspect of right pedipalp, male
holotype: 110, same, female paratype, MH230,09; 111, same, tritonymph paratype, MH230.12; 112,
dorsal aspect of carapace, male holotype; 113, female genitalia and associated sternites, paratype,
MH230.09. Scale line = 1.00 (Figs. 109-112), 0.25 mm (Fig. 113).

190

THE JOURNAL OF ARACHNOLOGY

Habitat.— The types were collected together under the bark of a single tree {Eucalyptus
sp.). The Roper Bar tritonymph was collected from under the bark of a tree.

Remarks.— This species is easily distinguished from all other species of the genus by its
small size and the form of the male genitalia. It appears to be most similar to A. grayi, but
the two are separable on size (Fig. 129). The “paratype female?” referred to above is in
poor condition and its gender is difficult to ascertain. It possesses an adult trichobothrial
pattern, but due to the contorted and shrivelled state of the abdomen, the presence of
cribriform plates cannot be confirmed. The tritonymph from Roper Bar has not been
designated as a type specimen because it is slightly larger than tritonymphs of the type
series. Nevertheless, it is believed to be a member of this species, but adult material is
needed to confirm this record.

Other specimens examined.— AUSTRALIA: NORTHERN TERRITORY; 13.5 km SE of Roper
Bar, under bark of tree, 17 July 1980 (C. Silveira), 1 tritonymph (NTM, MH222.04) (slide).

Afrosternophonis anabates, new species
Figs. 8, 58, 73, 114-120, 129; Map 5

Types.— Holotype male, one paratype male, three paratype females, one paratype
tritonymph, 15 km WNW of Yaapeet, Lake Albacutya Park, Victoria, Australia, under
bark of Eucalyptus camaldulensis, 2 July 1982 (M. S. Harvey and B. E. Roberts), MV,
K125-K129, K145, MH416. 05-06, 11-14 (slides and spirit). Two paratype males, two
paratype females, same data as above, ANIC, MH4 16.07-10 (slides). Paratype male,
paratype female, paratype protonymph, same data as above except 3 July 1982, MV,
K130-132, MH419.03-05 (slides). Paratype male, paratype tritonymph, Mt. Killawarra,
17 km NW of Wangaratta, Victoria, Australia, ex Delena cancerides Walckenaer (Sparas-
sidae; Araneae), 7 November 1978 (M. S. Harvey), MV K053-053a, MH044. 01-02
(slides). Two paratype females, Reedy Lake, near Nagambie, Victoria, Australia, ex D.
cancerides or Isopoda sp. (Sparassidae), 23 November 1979 (H. E. Parnaby), MV, K050-
051, MH151.01-02 (slides). Paratype female, 6.5 km SSW of Stuart Mill, Victoria, Austra-
lia, ex D. cancerides, 3 December 1977 (M. S. Harvey), MV, K052, MH004.01 (slide).

Etymology.— The specific epithet refers to the phoretic habit exhibited by some of the
specimens {anabates Gr. rider, passenger).

Distribution.— Victoria, Australia (Map 5).

Diagnosis.— Male genitalia with anterior apodeme distally broad; lateral apodemes
elongate, tapering. Female galea with three distal, one subdistal and two (sometimes one)
basal to subbasal rami. Chela (with pedicel) 0,83 to 0.895 (male), 0,835 to 1.03 mm
(female) in length. Sometimes phoretic on huntsman spiders.

Description.— ADULTS: Pedipalpal trochanter 1.71 to 1.89 (male), 1.51 to 1.90
(female), femur elongate, 2.77 to 3.08 (male), 2.43 to 3.13 (female), tibia 2.31 to 2.59
(male), 2.04 to 2.52 (female), chela (with pedicel) 3.56 to 4.15 (male), 3.38 to 4.00
(female), chela (without pedicel) 3.38 to 3.90 (male), 3.27 to 3.79 (female) times longer
than broad. Trichobothria as for aethiopicus group, in usual position (Figs. 114-116).
Serrula exterior of chelicera with 11 to 13 (male), 11 to 12 (female) lamellae. Galea of
male simple, of female with three distal, one subdistal and two (sometimes one) basal to
subbasal rami (Fig. 73). Carapace usually unconstricted (Fig. 119), but occasionally a
slight constriction is apparent; with 16 to 23 (male), 18 to 27 (female) setae; 1.46 to
1.59 (male), 1.38 to 1.55 (female) times longer than broad. Male genitalia (Fig. 58) with
anterior apodeme distally broad, lateral apodemes long and tapering. Female genitalia as

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

I9I

for genus (Fig. 120). Tergal chaetotaxy: male, 6:5-7:4:5-8:6-7:6:6-7:6:6:TlT3-4TlT:?:
2; female, 5-7:5-6:3-4:4-6:4-7:4-7:6-7:5-7:6-7:TlT4-5TlT:?:2. Sternal chaetotaxy: male,
0:4-8:(0)4»5[6-8](0):(l)5-6(l):6-8:5-7:6-7:6-8:6:TlT3-4TlT:?:2; female, 0:6-8:(0)4-5
(0):(l)2-6(l):6-8:6-9:6-8:6-7:6-7:TlT4TlT:?:2. Coxal chaetotaxy: male, 3-5:4-6:4-5:
3-5; female, 3-6 :3-7:3-5:3-5.

Dimensions (mm): Body length 2. 2-2. 5 (2.4-3.4); pedipalps: trochanter 0.325-0.375/
0.175-0.205 (0.335-0.40/0.175-0.245), femur 0.55-0.595/0.175-0.215 (0.52-0.695/0.19-
0.235), tibia 0.475-0.525/0.185-0.225 (0.465-0.60/0.195-0.27), chela (with pedicel)

Figs. .-Afrostemophorus anabates, new species: 114, lateral aspect of left chela, male

paratype, MH044.01; 115, ventral aspect of left pedipalp, male holotype; 116, same, female paratype,
MH4 16.09; 117, same, tritonymph paratype, MH416.14; 118, ventral aspect of right pedipalp, proto-
nymph paratype, MH4 19.05; 119, dorsal aspect of carapace, male holotype; 120, female genitalia
and associated sternites, paratype, MH416.09. Scale line = 1.00 mm (Figs. 114-119), 0.25 mm (Fig.
120).

192

THE JOURNAL OF ARACHNOLOGY

0.83-0.895/0.205-0.25 (0.835-1.03/0.21-0.28), chela (without pedicel) 0.795-0.855
(0.795-0.98), moveable finger length 0.39-0.425 (0.40-0.475); chelicera 0.16-0.185/
0.095-0.10 (0.165-0.20/0.10-0.11), moveable finger length 0.11-0.125 (0.12-0.14);
carapace 0.82-0.89/0.52-0.59 (0.83-1.005/0.555-0.695); leg I: coxa 0.21-0.24/0.24-0.265
(0.23-0.29/0.26-0.30), trochanter 0.12-0.13/0.08-0.10 (0.12-0.15/0.095-0.12), femur I
0.11-0.13/0.11-0.13- (0.12-0.145/0.115-0.15), femur II 0.175-0.195/0.115-0.135 (0.18-
0.22/0.12-0.155), tibia 0.20-0.225/0.07-0.085 (0.21-0.255/0.075-0.09), tarsus 0.11-0.13
/0.05-0.055 (0.12-0.13/0.05-0.06); leg IV: coxa width 0.22-0.24 (0.26-0.29), trochanter
0.155-0.16/0.105-0.12 (0.16-0.19/0.115-0.135), femur I 0.205-0.22/0.175-0.225 (0.215-
0.27/0.20-0.26), femur II 0.285-0.33/0.185-0.235 (0.30-0.37/0.205-0.26), tibia 0.34-
0.375/0.11-0.135 (0.35-0.43/0.115-0.14), tarsus 0.16-0.19/0.07-0.085 (0.18-0.21/0.07-
0.09).

TRITON YMPHS: Pedipalpal trochanter 1.51 to 1.81, femur 2.38 to 2.73, tibia 2.08 to
2.29, chela (with pedicel) 3.44 to 3.85, chela (without pedicel) 3.30 to 3.69 times longer
than broad. Fixed finger with seven trichobothria, moveable finger with two trichobo-
thria (Fig. 117); it, sb and st absent. Serrula exterior of chelicera with 1 1 lamellae. Galea
with three distal to subdistal and two subbasal rami. Carapace with 17 setae; 1.46 to 1.53
times longer than broad. Tergal chaetotaxy: 6-7:4-5:4:4-5:4-5:4-6:5-6:5-7:5-6:TlT2TlT:
?:2. Sternal chaetotaxy: 0:2:(0>t(0):(l)4-5(l):6:5-6:6-7:6:6:TlT2TlT:?:2. Coxal
chaetotaxy: 3-4:34:3:3.

Dimensions (mm): Body length 2.3-2. 5; pedipalps: trochanter 0.265-0.28/0.155-
0.175, femur 0.44-0.46/0.165-0.185, tibia 0.375-0.39/0.17-0.18, chela (with pedicel)
0.735-0.75/0.195-0.215, chela (without pedicel) 0.705-0.725, moveable finger length
0.345-0.36 ; carapace 0.70-0.75/0.48-0.49.

PROTONYMPHS: Pedipalpal trochanter 1.68 to 1.78, femur 2.53 to 2.61, tibia
1 .95 to 2.05, chela (with pedicel) 3.60 to 3.71, chela (without pedicel) 3.52 to 3.54 times
longer than broad. Fixed finger with three trichobothria, moveable finger with one
trichobothrium (Fig. 118); eb, et, ib and t present. Serrula exterior of chelicera with 9 to
10 lamallae. Seta gs absent. Galea with three distal to subdistal rami. Carapace with 14
setae; 1.26 times longer than broad. Tergal chaetotaxy: 4:2:0:4:4:4:4:4:4:TTTT:TTTT:
2. Sternal chaetotaxy: 0:0:(0)2(0):(1)2(1):4:4:4:4:4:TTTT:TT:2. Coxal chaetotaxy:
1:1:1:1.

Dimensions (mm): Body length 1.5; pedipalps: trochanter 0.16/0.09-0.095, femur
0.235-0.24/0.09-0.095, tibia 0.205/0.10-0.105, chela (with pedicel) 0.445-0.45/0.12-
0.125, chela (without pedicel) 0.425-0.44, moveable finger length 0.22-0.23; carapace
0.48/0.38.

Habitat.— As discussed in detail below, some specimens were taken from spiders of the
family Sparassidae (Delena cancerides and Isopoda sp.), whereas others were taken from
under bark of Eucalyptus camaldulensis.

Remarks.— This species is easily distinguished from other members of the genus except
A. papuanus by the shape of the male genitalia (distally broad anterior apodeme) and the
form of the female galea. It differs from A. papuanus only in size; papuanus is smaller
than anabates. It is thought that these two species are sister-species because they possess a
synapomorphy in the form of the anterior apodeme.

Three of the five known collections of A. anabates have been taken from sparassid
spiders. The Mt. Killawarra specimens were found clinging to leg setae of a male of D.
cancerides which was found under bark of Eucalyptus melliodora. The Reedy Lake
material was removed from a vial containing specimens of D, cancerides Isopoda sp.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

193

which were collected under bark of Eucalyptus sp. Since the sternophorids were not
collected from the bark of the tree (H. E. Parnaby, pers. comm.), they were obviously
phoretic on one or both of the spider species. The Stuart Mill specimen was taken from a
leg seta of a male of D. cancerides which was found under bark of a log. Extensive collect-
ing at Stuart Mill and Mt. Killawarra failed to disclose any further sternophorids, even
though many corticolous species of the families Atemnidae, Chernetidae and Cheliferidae
were found. Conversely, at Lake Albacutya, A. anabates was only found under bark
of E. camaldulensis and was not found on any sparassids that also occurred in the area (D.
cancerides, Isopoda spp.). It is of interest to note that the phoretic nature of this species
may be seasonal. It has been collected from spiders in November and December, and it
has been found under bark in July. Naturally, much more collecting is needed before any
definite statements may be made. It would be very interesting to examine this aspect of
the pseudoscorpion’s biology in relation to the time of year that mating and egg produc-
tion takes place. Unfortunately, no such data is available for A. anabates, or, for that
matter, any other sternophorid species.

The only other pseudoscorpion species that is known to be phoretic on a spider is the
chernetid Lustrochernes grossus (Banks) from southern U.S.A. (Hoff and Jennings 1974).
The spider from which it was taken was also a sparassid (= Heteropodidae; Platnick and
Levi 1973), Olios fasciculatus Simon. In contrast to A, anabates, L. grossus has been
found in a variety of habitats, including the bark of trees and under the elytra of ceram-
bycid beetles (Hoff and Jennings 1974, Benedict and Malcolm 1982). Dr. Jennings (pers.
comm.) kindly informed me that no further records of L. grossus on O. fasciculatus have
been detected.

Hoff and Jennings suggested that the aggressive nature of spiders accounted for the
lack of phoretic pseudoscorpions. The two specimens of L. grossus that Hoff and Jen-
nings examined were found clinging to “dorsal abdominal setae”, where, they suggest, the
spider could not reach, and hence, dislodge and eat the pseudoscorpions. Afrosterno-
phorus anabates has been taken from leg setae, a position which would be easily acces-
sible to cleaning. Mites also are not uncommon on sparassids, which suggests that they do
not clean themselves as rigorously as Hoff and Jennings imply. Heavily infested specimens
are often in poor physical condition, yet it cannot be ascertained whether the mites are
responsible for this state, or whether they are attracted to, or reproduce vigorously on
weak specimens.

Afros ternophorus papuanus (Beier), new combination
Figs. 59, 74, 121-124, 129; Map 6

Sternophorus papuanus Beier 1974:211-212, Fig. 5.

Types.— Lectotype male (present designation), paralectotype male, two paralectotype
females, Americ, Madang District, Papua New Guinea, 1972 (B. Gray), NHMW (spirit).
Distribution.— Papua New Guinea (Map 6).

Diagnosis.— Male genitalia with anterior apodeme distally broad; lateral apodemes long
and tapering. Female galea with three distal, one subdistal and two subbasal rami. Chela
(with pedicel) 0.61 to 0.69 (male), 0.73 to 0.74 mm (female) in length.

Description.— Pedipalpal trochanter 1.82 to 1.89 (male), 1.83 to 1.87 (female), femur
3.00 to 3.19 (male), 2.87 to 2.93 (female), tibia 2.38 to 2.57 (male), 2.12 to 2.32 (fe-
male), chela (with pedicel) 3.97 to 4.18 (male), 3.65 to 3.79 (female), chela (without

194

THE JOURNAL OF ARACHNOLOGY

pedicel) 3.73 to 4.20 (male), 3.45 to 3.56 (female) times longer than broad. Trichobo-
thria as for aethiopicus group, in usual position (Figs. 121-122). Serrula exterior of
chelicera with 1 1 (male), 1 2 (female) lamellae. Galea of male simple, of female with three
distal, one subdistal and two subbasal rami (Fig. 74). Carapace (Fig. 123) unconstricted,
with 24 (male, female) setae; 1.45 (male), 1.51 to 1.52 (female) times longer than broad.
Male genitalia (Fig. 59) with distally broad anterior apodeme; lateral apodemes long and
tapering. Female genitalia as for genus (Fig. 124). Tergal chaetotaxy: male, 6:5:2:4:5:6:
6;6:6:T1T4T1T:?:2; female, 6:6:2:4;5-6:5-6:5-6:6:6:TlT3-4TlT:?:2. Sternal chaeto-
taxy: male, 0:6:(0)4[6] (0):(1)5(1):?:?:?:6:6:T1T4T1T:?:2; female, 0:8-9: (0)4(0):
(l)4(l):6-7:6:6:6:6-7:TlT4TlT:?:2. Coxal chaetotaxy: male, 3 :2-4:3:3 ; female, 2-4:2-4:
3:3-4.

Dimensions (mm): Body length 1. 7-2.0 (2.3-2. 5); pedipalps: trochanter 0.255-0.265/
0.14 (0.275-0.285/0.15-0.155), femur 0.405-0.42/0.135 (0.425-0.445/0.145-0.155), tibia
0.345-0.36/0.14-0.145 (0.35-0.36/0.155-0.165), chela (with pedicel) 0.655-0.69/0.165-
0.17 (0.73-0.74/0.195-0.20), chela (without pedicel) 0.61-0.645 (0.69-0.695), moveable
finger length 0.30-0.335 (0.375-0.385); chelicera 0.125-0.13/0.08-0.085 (0.13-0.135/
0.08-0.09), moveable finger length 0.085-0.095 (0.10-0.105); carapace 0.64/0.44 (0.68-
0.70/0.45-0.46); leg I: coxa ?/? (0.195-0.20/0.225-0.23), trochanter 0.095/0.07 (0.10/
0.075), femur I 0.10/0.09 (0.095/0.085-0.095), femur II 0.10/0.09 (0.15-0.155/0.085-
0.095), tibia 0.16/0.055 (0.16/0.06), tarsus 0.10/0.035 (0.1 1/0.045); leg IV: coxa width
0.17 (0.23), trochanter 0.135/0.085 (0.14/0.095), femur I 0.18/0.125 (0.18/0.145),
femur II 0.21/0.13 (0.23/0.15), tibia 0.23/0.085 (0.255/0.09), tarsus 0.12/0.06 (0.12/
0.055).

Figs. \2\-\2A.-Afrostemophoms papuanus (Beier): 121, ventral aspect of left pedipalp, male
lectotype; 122, same, female paralectotype; 123, dorsal aspect of carapace, male lectotype; 124,
female genitalia and associated sternites, paralectotype. Scale line = 1.00 mm (Figs. 121-123), 0.25
mm (Fig. 1 24).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

195

Map 6. -Papua New Guinea and Irian Jaya showing known distribution of Afrosternophorus
papuanus (Beier) (circle) (district record only), A. grayi (Beier) (squares), A. araucariae (Beier) (tri-
angle) and A. cavemae (Beier) (star). Open symbol represents literature record only.

Habitat.— No habitat data accompanied the specimens.

Rem2p[\!is.— Afrosternophorus papuanus is easily separated from its congeners, except
A. anabates, by the form of the male genitalia (distally broad anterior apodeme) and the
female galea. It can be distinguished from A. anabates only by its smaller size (Fig. 129).
These two species are thought to be sister-species because both possess the distally broad
anterior apodeme, which is a synapomorphy.

The locality of “Americ” cannot be traced, and the map shows the district (Madang)
record only.

Beier (1974) recorded the type material as “Type und Paratypen; 2 d, 2 9”, yet failed
to distinguish a holotype in the collection. Therefore, a male lectotype has been herein
selected.

Afrosternophorus grayi (Beier), new status, new combination
Figs. 60, 75, 125-129;Map6

Sternophorus hirsti grayi Beier 1971:370-371, Fig. 2.

Types.— Lectotype male (present designation), six paralectotype males, five paralecto-
type females, Bulolo, Morobe District, Papua New Guinea, 18 August 1970 (B. Gray),
NHMW (spirit). Two paralectotype females, Bunu, Lake Kutubu, Southern Highlands
District, Papua New Guinea, 22 November 1969 (B. Gray), NHMW (spirit).

Distribution.— Papua New Guinea (Map 6).

Diagnosis.— Male genitalia with long, acute dorsal apodemes. Female galea with two
distal, one subdistal and one subbasal rami. Chela (with pedicel) 0.645 to 0.70 (male),
0.68 to 0.74 mm (female) in length.

Description.— Pedipalpal trochanter 1.68 to 1.86 (male), 1.66 to 1.86 (female), femur
stout, 2.30 to 2.65 (male), 2.21 to 2.58 (female), tibia 2.03 to 2.29 (male), 1.92 to 2.23
(female), chela (with pedicel) 3.42 to 3.94 (male), 2.96 to 3.94 (female), chela (without
pedicel) 3.21 to 3.51 (male), 2.70 to 3.75 (female) times longer than broad. Trichobo-

196

THE JOURNAL OF ARACHNOLOGY

thria as for aethiopicus group, in usual position (Figs. 125-126). Serrula exterior of
chelicera with 12 (male), 11 to 13 (female) lamellae. Galea of male simple, of female with
two distal, one subdistal and one subbasal rami (Fig. 75). Carapace (Fig. 127) uncon-
stricted, with 21 to 23 (male), 19 to 24 (female) setae; 1.40 to 1.51 (male), 1.09 to 1.39
(female) times longer than broad. Male genitalia (Fig. 60) obscure, slightly distorted;
dorsal apodemes long and tapering. Female genitalia as for genus (Fig. 128). Tergal
chaetotaxy: male, 6:5-6:2-3:4-5:4-5:5:5-6:5-6:6:TlT2?-4TlT:?:2; female, 5-6:5-7:14:
4.6:3?-6:4-6:5-6:5-6:5-6:T1T2?-3T1T:?:2. Sternal chaetotaxy: male, 0:3-6: (0)4-6 [5-6]
(0):(l)4(l):4-6:5-6:3?-6:5-6:5-6:TlT24TlT:?:2; female, 0:6-10:(0>f-6(0):(l)4-5(l):6:
5-6:4-6:5-6:5-6:TlT24TlT:?:2. Coxal chaetotaxy: male, 3-6:3-5:3-5:3-5; female,
3-5:3-5:3-4:4-5.

Dimensions (mm): Body length 1. 4-2.1 (1 .8-2.7); pedipalps: trochanter 0.235-0.27/
0.135-0.155 (0.24-0.295/0.14-0.165), femur 0.385-0.445/0.155-0.185 (0.39-0.465/
0.165-0.195), tibia 0.32-0.37/0.14-0.175 (0.33-0.39/0.155-0.195), chela (with pedicel)
0.645-0.70/0.17-0.195 (0.68-0.74/0.18-0.23), chela (without pedicel) 0.61-0.66 (0.62-
0.70), moveable finger length 0.315-0.36 (0.325-0.38); chelicera 0.145-0.16/0.08-0.085
(0.14-0.165/0.08-0.10), moveable finger length 0.10-0.11 (0.09-0.1 15); carapace 0.65-
0.68/0.43-0.47 (0.645-0.72/0.49-0.59); leg I: coxa 0.18-0.19/0.20-0.21 (0.185-0.21/
0.205-0.25), trochanter 0.095-0.11/0.07-0.085 (0.10-0.115/0.075-0.08), femur I 0.095-

Figs. 125-12S. -A frosternophorus grayi (Beier): 125, ventral aspect of right pedipalp, male lecto-
type; 126, ventral aspect of left pedipalp, female paralectotype from Bulolo, Papua New Guinea; 127,
dorsal aspect of carapace, male lectotype; 128, female genitalia and associated sternites, paralectotype
from Bulolo. Scale line = 1.00 mm (Figs. 125-127), 0.25 mm (Fig. 128).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

197

0.11/0.085-0.11 (0.10-0.115/0.095-0.105), femur II 0.14-0.155/0.085-0.11 (0.14-0.16/
0.095-0.105), tibia 0.16-0.165/0.07-0.075 (0.145-0.185/0.065-0.075), tarsus 0.09-0.10/
0.045 (0.09-0.125/0.045-0.055); leg IV: coxa width 0.18-0.23 (0.215-0.22), trochanter
0.125-0.15/0.09-0.10 (0.14-0.155/0.09-0.115), femur I 0.165-0.175/0.155-0.18 (0.16-
0.185/0.14-0.175), femur II 0.215-0.23/0.165-0.18 (0.215-0.25/0.14-0.18), tibia 0.255-
0.27/0.095-0.105 (0.265-0.29/0.09-0.115), tarsus 0.135-0.14/0.065-0.07 (0.11-0.155/
0.055-0.07).

Habitat— This species has been taken from under bark of Araucaria cunninghamii and
A. hunsteinii (Beier 1971), a fact which is not stated on the locality labels.

Remarks.— This species was originally described from many specimens, most of which
were apparently returned to Dr. B. Gray, Entomology Section, Department of Forests,
Bulolo, Papua New Guinea; despite repeated requests to this institution concerning the
remaining material, it has not been available for study.

Beier labelled the Bunu specimens “Typen” and failed to designate a holotype. Fur-
thermore, he stated in the original description that one male and one female were present
in the vial. Both specimens are in fact females, and since it appears advantageous that the
primary type be a male, a male lectotype has been selected from the Bulolo vial.

Beier also labelled the Bunu specimens as ''Sternophorus grayi n. sp.”, and the Bulolo
specimens as ''Sternophorus hirsti grayi n. sp.”, even though he published the description
under the latter combination.

As can be seen from the description, A. grayi is substantially different from A. hirsti,
and Beier’s decision to relegate it to subspecific rank cannot be justified. Afrosterno-
phorus grayi appears to be most closely related to the Australian species A. nanus, from
which it can be distinguished only by its larger size.

Araucariae group

Diagnosis.— As for genus, except that the moveable chelal finger possesses two tricho-
bothria, b and t.

Subordinate tax^.—Afrosternophorus araucariae (Beier), cavernae (Beier), A. fallax,
new species, A. xalyx, new species.

Afrostemophorus araucariae (Beier), new combination
Figs. 61, 130-132;Map6

Sternophorellus araucariae Beier 1971:372-373, Fig. 3.

Types.— Holotype male, two chelae of paratype adult, Mt. Dayman, Milne Bay District,
Papua New Guinea, under bark of Araucaria cunninghamii, 20 July 1969 (B. Gray),
NHMW (spirit). The type series also contained two male and a female paratype which
were not available for study.

Distribution.— Papua New Guinea (Map 6).

Diagnosis.— Male genitalia with long, acute, slightly curved dorsal apodemes. Chela
(with pedicel) 4.47 to 4.56 (male) times longer than broad.

Description.— Male only. Pedipalpal trochanter 1.88, femur elongate, 3.34 to 3.55,
tibia 2.58 to 2.84, chela (with pedicel) slender, 4.47 to 4.56, chela (without pedicel) 4.19
to 4.28 times longer than broad. Trichobothria as for araucariae group, in usual position
(Fig. 131). Serrula exterior of chelicera with 10 lamellae. Galea of male simple. Carapace

198

THE JOURNAL OF ARACHNOLOGY

CW CW

Figs. 129-1 30. -Graphs of chela (with pedicel) length (CL) versus width (CW), in mm; open sym-
bols, males; closed symbols, females: 129, A frostemophonis hirsti (Chamberlin) (circles),^, nanus,
new species (squares), A. anabates, new species (upright triangles), A. papuanus (Beier) (stars), A. grayi
(Beier) (inverted triangles); 130, A. araucariae (Beier) (squares), A. cavernae (Beier) (triangle), A.
fallax, new species (circles), A. xalyx, new species (inverted triangles).

slightly constricted (Fig. 132), with 23 setae; 1.41 times longer than broad. Male genitalia
with long, acute, slightly curved dorsal apodemes; mid-piece of lateral rod small (Fig.
61). Tergal chaetotaxy: 6:5 ;4:5 :6:6:6:6:6:T1T3T1T:?:2. Sternal chaetotaxy: 0:5:
(0)4[6](0):(1)4(1):7:6:7:7:7:T1T3T1T:?:2. Coxal chaetotaxy: 4-5:3 :3-5:4-5.

Dimensions (mm): Body length 2.2; pedipalps: trochanter 0.30/0.16, femur 0.535-
0.55/0.155-0.16, tibia 0.425-0.44/0.155-0.165, chela (with pedicel) 0.805-0.82/0.18,
chela (without pedicel) 0.755-0.77, moveable finger length 0.375-0.39; chelicera 0.16/
0.095, moveable finger length 0.11-0.115; carapace 0.705/0.50; leg I: coxa 0.20/0.18,
trochanter 0.12/0.085, femur I 0.105/0.105, femur II 0.175/0.105, tibia 0.185/0.07,
tarsus 7/0.05; leg IV: coxa width 0.19, trochanter 0.16/0.095, femur I 0.175/0.145,
femur II 0.26/0.145, tibia 0.30/0.09, tarsus 0.14/0.06.

Habitat.— The type specimens were taken from under bark of Araucaria cunninghamii
(Beier 1971).

Remarks.— Beier originally described this species from three males and one female. The
holotype male is deposited in NHMW, but the remaining specimens were probably re-
turned to Dr. B. Gray of the Entomology Section, Department of Forests, Bulolo, Papua
New Guinea. Repeated requests to this institution concerning the material have failed to
elicit a response, and as a result, the female of araucariae remains unstudied. This is
unfortunate since the form of the female genitalia is crucial in correctly assigning species
to genus. Nevertheless, females of the closely related species A. fallax and A. xalyx
possess the typical Afrosternophorus genitalia.

Two badly crushed, isolated chelae were present in the vial containing the holotype.
They most probably belong to one of the aforementioned paratypes.

This species is easily distinguished from other species of the species group on the basis
of its elongate pedipalps (Fig. 130).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

199

Figs. l?)\-\2)2-Afrosternophorus araucariae (Beier), male holotype: 131, ventral aspect of right
pedipalp; 132, dorsal aspect of carapace. Figs. \2>'h-\l>A .-Afrosternophonis cavernae (Beier), male
paratype: 133, dorsal aspect of carapace; 134, ventral aspect of left pedipalp. Scale line = 1.00 mm.

Afrosternophorus cavernae (Beier), new combination
Figs. 62, 130, 133-134; Map 6

Sternophorellus cavernae Beier 1982:44-45, Fig. 1.

Types.— Holotype male, Selminum tern Cave [near Tifalmin] , West Sepic District,
Papua New Guinea, 1 November 1975 (P. Chapman), National Natural History Museum,
Sofia (spirit), not examined. Paratype male, same data as above except 1 October 1975
(P. Beron), NHMW (spirit).

Distribution.— Papua New Guinea (Map 6).

Diagnosis.— Male genitalia with long, acute dorsal apodemes; mid-piece of lateral rod
not elongate. Chela (with pedicel) 0.66 mm (male) in length.

Description.— Male only. Pedipalpal trochanter 1.92 to 2.00, femur 2.93 to 2.97, tibia
2.20 to 2.27, chela (with pedicel) 3.88, chela (without pedicel) 3.62 to 3.65 times longer
than broad. Trichobothria as for araucariae group, in usual position (Fig. 134). Serrula
exterior of chelicera with 12 lamellae. Galea of male simple. Carapace unconstricted (Fig.
133), with 20 setae; 1.39 times longer than broad. Male genitalia with long, acute dorsal
apodemes; mid-piece of lateral rod small (Fig. 62). Tergal chaetotaxy: 7:6:6:6:6;7:6:7:8:
T1T4T1T:?:2. Sternal chaetotaxy: ? Coxal chaetotaxy:?

Dimensions (mm): Body length 1.7; pedipalps: trochanter 0.25-0.26/0.13, femur
0.425-0.43/0.145, tibia 0.33-0.34/0.15, chela (with pedicel) 0.66/0.17, chela (without
pedicel) 0.615-0.62, moveable finger length 0.32-0.325; chelicera 0.135-0.14/0.08,
moveable finger length 0.10; carapace 0.61/0.44; legs: ?.

Habitat.— The type specimens were taken from a cave, but it is not known whether or
not this is simply fortuitous.

200

THE JOURNAL OF ARACHNOLOGY

RQm2ixks>.—Afrosternophorus cavernae is very similar to A. fallax and the two species
can be distinguished only on details of the male genitalia. Additional material of both
species may yield subtle differences in pedipalpal morphometries, but the paucity of
specimens precludes such a study at the present time. Both of these species differ from A.
araucariae by their stout pedipalps (Fig. 130), and from A. xalyx in possessing long dorsal
apodemes.

Unfortunately, females of A. cavernae have not been available for study, but it is
presumed that they will possess the genitalia typical of the remaining species of the genus.
Females of the closely related species A. fallax and A. xalyx possess such genitalia.

The occurrence of this species in a cave is very surprising because all other sterno-
phorid species (with known locality data) are corticolous. It does not display any troglo-
bitic adaptations such as attenuate appendages or unusually pale colouration and this
cavernicolous record may be simply fortuitous. Selminum tern Cave is 20 km long (Moore
1978) and the locality data given by Beier (1982) or the data on the label of the paratype
that I examined did not state from where in the cave system the specimens were taken.
Further collections and more detailed observations on its biology are needed to support
the alleged cavernicolous nature of the species.

Beier (1982) stated that Selminum tern Cave was situated in the Western District of
Papua New Guinea. Although I cannot locate this cave on any maps or gazetteers at my
disposal, Moore (1978) indicated that it is situated near the township of Tifalmin, which,
in fact, is in the West Sepic District.

Afrosternophorus fallax, new species
Figs. 63,76, 130, 135-138;Map4

Sternophonis chamberlini Redikorzev: Beier 1951:71-72 (in part).

Types.— Holotype male, paratype male, paratype female, Plateau von Langbian [= Cao
Nguyen Lam Vien, see Table 1], Vietnam, 1938-1939 (C. Dawydoff), NHMW (spirit).
Paratype male, same data as above, ANIC, MH431.01 (slide).

Etymology.— The specific epithet refers to this species being confused with chamher-
lini by Beier {fallax L. deceitful, false).

Distribution.— Vietnam (Map 4).

Diagnosis.— Male genitalia with long, acute dorsal apodemes; mid-piece of lateral rod
elongate. Female galea with three distal to subdistal rami. Chela (with pedicel) 3.70 to
3.91 (male), 3.45 to 3.59 (female) times longer than broad.

Description,— Pedipalpal trochanter 1.68 to 1.88 (male), 1.86 to 1.89 (female), femur
2.88 to 3.08 (male), 2.83 to 2.90 (female), tibia 2.14 to 2.38 (male), 2.16 to 2.25 (fe-
male), chela (with pedicel) 3.70 to 3.91 (male), 3.45 to 3.59 (female), chela (without
pedicel) 3.45 to 3.73 (male), 3.33 to 3.38 (female) times longer than broad. Trichobo-
thria as for araucariae group, in usual position (Figs. 135-136). Serrula exterior of chel-
icera with 12 to 13 (male), 12 (female) lamellae. Galea of male simple, of female with
three distal to subdistal rami (Fig. 76). Carapace unconstricted (Fig. 137), with 22 (male),
20 (female) setae; 1.33 to 1.41 (male), 1.45 (female) times longer than broad. Male
with long, acute dorsal apodemes; mid-piece of lateral rod elongate (Fig. 63). Female
genitalia as for genus (Fig. 138), Tergal chaetotaxy: male, 6:4-6;2:6:5-6:6:6:5-6:6:TlT3
T1T:?:2; female, 6:5 :2:6:6:6:6:6:6:T1T4T1T:?:2. Sternal chaetotaxy: male, 0:4:(0)4[5-
6](0):(1)4(1):6:6:6:6:6:T1T2-3T1T:?:2; female, 0:5:(0)4(0):(1)4(1):6:6:6:6:6:T1T3
T1T:?:2. Coxal chaetotaxy: male, 4-5 :3-5:24:3-5 ; female, 5:4-5:3:3.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

201

Dimensions (mm): Body length 1.5-1. 7 (1.9); pedipalps: trochanter 0.21-0.235/0.12-
0.125 (0.26-0.265/0.14), femur 0.375-0.385/0.125-0.13 (0.425/0.15), tibia 0.30-0.31/
0.13-0.14 (0.345-0.36/0.16), chela (with pedicel) 0.585-0.625/0.15-0.165 (0.69-0.70/
0.195-0.20), chela (without pedicel) 0.55-0.595 (0.66-0.665), moveable finger length
0.28-0.30 (0.32); chelicera 0.125-0.13/0.07-0.075 (0.145/0.08), moveable finger length
0.095-0.10 (0.105); carapace 0.56-0.58/0.41-0.42 (0.665/0.46); leg I: coxa 0.155/0.165
(0.18/0.195-0.20), trochanter 0.08-0.085/0.065 (0.095-0.10/0.07), femur I 0.09/0.075
(0.085/0.085), femur II 0.12/0.08 (0.135-0.14/0.085), tibia 0.13/0.055 (0.14/0.06),
tarsus ?/? (?/?); leg IV: coxa width 0.16-0.19 (0.195-0.21), trochanter 0.11-0.12/0.075-
0.085 (0.14/0.085-0.09), femur I 0.14-0.165/0.11-0.115 (0.175-0.18/0.125-0.13), femur
II 0.17-0.18/0.115-0.12 (0.21-0.215/0.13-0.135), tibia 0.19-0.20/0.07-0.075 (0.22-0.24/
0.08-0.085), tarsus 0.105-0.13/0.045-0.05 (0.13/0.06).

Habitat.— No habitat data accompanied the specimens.

Remarks.— As discussed under Afrosternophorus chamberlini, Beier’s (1951) descrip-
tion of that species was a composite of chamberlini and fallax. His male pedipalp mea-
surements were of the latter, and his female measurements were of the former.

This species is very similar to the Papua New Guinean species A. cavernae. It may be
distinguished from this species by the presence of a long mid-piece on the lateral rod of
the male genitalia.

Afrosternophorus xalyx, new species
Figs. 64, 77, 130, 139-143; Map 5

Types.— Holotype male, two paratype males, four paratype females, Townsville,
Queensland, Australia, under bark of eucalypt tree, date? (collector?), MV, K154-K160
(slides and spirit).

Etymology.— The specific epithet is an arbitrary combination of letters.

Figs. Afrosternophorus fallax, new species: 135, ventral aspect of right pedipalp, female

paratype; 136, ventral aspect of left pedipalp, male holotype; 137, dorsal aspect of carapace, male
holotype; 138, female genitalia and associated sternites, paratype. Scale line = 1.00 mm (Figs. 135-
137), 0.25 mm (Fig. 138).

202

THE JOURNAL OF ARACHNOLOGY

Diagnosis.— Male genitalia with reduced dorsal apodemes; mid-piece of lateral rod with
distinct posterior notch. Female galea with two distal and four subdistal to subbasal rami.
Chela (with pedicel) 0.685 to 0.72 (male), 0.74 to 0.79 mm (female) in length, 3.49 to
3.81 (male), 3.43 to 3.52 (female) times longer than broad.

Description.— Pedipalpal trochanter 1.61 to 1.89 (male), 1.67 to 1.81 (female), femur
2.64 to 2.93 (male), 2.71 to 2.94 (female), tibia 2.03 to 2.25 (male), 2.08 to 2.24 (fe-
male), chela (with pedicel) 3.49 to 3.81 (male), 3.43 to 3.52 (female), chela (without
pedicel) 3.32 to 3.58 (male), 3.26 to 3.38 (female) times longer than broad. Trichobo-
thria as for araucariae group, in usual position (Figs. 139-141). Serrula exterior of chel-
icera with 12 to 13 (male), 12 (female) lamellae. Galea of male simple, of female with
two distal and four subdistal to subbasal rami (Fig. 77). Carapace not constricted (Fig.
142) with 16 to 19 (male), 21 to 23 (female) setae; 1.30 to 1.37 (male), 1.20 to 1.34
(female) times longer than broad. Male genitalia with reduced dorsal apodemes; midpiece
of lateral rod with a distinct posterior notch (Fig. 64). Female genitalia as for genus (Fig.

Figs. \3>9-\A2>.-Afrosternophoms xalyx, new species: 139, lateral aspect of left chela, male para-
type, K155; 140, dorsal aspect of right pedipalp, male holotype; 141, same, female paratype, K157;
142, dorsal aspect of carapace, female paratype, K157; 143, female genitalia and associated sternites,
paratype, K157. Scale line = 1.00 mm (Figs. 139-142), 0.25 mm (Fig. 143).

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

203

143), Tergal chaetotaxy: male, 5-6:6-7:5-6:6-8:6-8:5-8:6-7:6-8;6:TlT4TlT:?:2; female,
5.7:6.8:4:6;6-7:6-7:6-7:6;6-8:T1T4-5T1T:?:2. Sternal chaetotaxy: male, 0:4-7:(0)4[10-
12] (0):(l)4-5(l):6:6:6-7:6-8:6:TlT3-4TlT:?:2; female, 0:8:(0)4(0):(l)4-6(l):6-8:6-8:5-
6:5-6:5-7:TlT3-4TlT:?:2. Coxal chaetotaxy: male, 34:4-5:3-5:4; female, 3-4:4-6:4-6:3-
4.

Dimensions (mm): Body length 1.6-1. 8 (1.7-2. 5); pedipalps: trochanter 0.25-0.265/
0.135-0.155 (0.265-0.295/0.155-0.17), femur 0.42-0.455/0.145-0.165 (0.45-0.50/0.16-
0.18), tibia 0.345-0.375/0.155-0.17 (0.375-0.405/0.17-0.19), chela (with pedicel) 0.685-
0.72/0.18-0.205 (0.74-0.79/0.21-0.23), chela (without pedicel) 0.645-0.69 (0.70-0.76),
moveable finger length 0.34-0.365 (0.375-0.405); chelicera 0.13-0.145/0.07-0.08 (0.155-
0.16/0.085-0.10), moveable finger length 0.10 (0.11-0.12); carapace 0.635-0.67/0.48-
0.49 (0.66-0.73/0.52-0.61); leg I: coxa 0.18-0.21/0.19-0.205 (0.20-0.21/0.23-0.25),
trochanter 0.095-0.105/0.07 (0.125/0.09), femur I 0.09-0.095/0.115-0.12 (0.105/0.13-
0.135), femur II 0.155-0.16/0.11-0.12 (0.18/0.08), tibia 0.175-0.18/0.05-0.075 (0.19/
0.08), tarsus 0.13/0.05 (0.13-0.14/0.05-0.06); leg IV: coxa width 0.21-0.22 (0.235-0.24),
trochanter 0.135-0.145/0.10 (0.155/0.11), femur I 0.155-0.165/0.175-0.18 (0.175-
0.185/0.175-0.19), femur II 0.22-0.23/0.175-0.18 (0.23-0.255/0.18-0.205), tibia 0.29/
0.09-0.10 (0.30-0.31/0.105-0.12), tarsus 0.165/0.06 (0.17-0.18/0.06-0.07).

Habitat.— The specimens were taken from “under bark of eucalypt tree”.

Remarks.— Even though the type specimens are in poor condition, there can be no
doubt that they represent a new and distinct species of the araucariae group.

Males of A. xalyx differ from the other three species of the araucariae group in posses-
sing reduced dorsal apodemes, and females differ from A. fallax (the only other species of
the group in which females are known) by possessing more galeal rami. In addition, A.
xalyx is slightly larger (Fig. 130).

A second label with the specimens reads “presented by G. F. Hill 24/4/23”.

EVOLUTION

A speculative phytogeny of the three sternophorid genera, and their species or groups,
is presented in Fig. 144. Table 3 summarizes the characters utilized in the cladogram. The
techniques advocated by Hennig (1965, 1966) for cladogram construction were used.

1 . Female median cribriform plate(s). Even though data for many genera are lacking,
most genera of all of the monosphyronid families (with the exception of the Pseudo-
garypidae, whose female genitalia are unknown) possess only one median cribriform
plate. Exceptions that possess two plates include most Cheliferini (Chamberlin 1932b),
the chernetids Metagoniochernes milloti Vachon (Vachon 1951) and Lamprochernes sp.
(pers. obs.) and the two sternophorid genera Garyops and Idiogaryops. Neocheiridium
africanum Mahnert and N. pusillum Mahnert (Cheiridiidae) possess three such plates
(Mahnert 1982a), as do some females of Idiogaryops paludis (Hoff 1963). Chamberlin
(1952) stated that two plates represent the primitive condition, and that several indepen-
dent fusions have occurred to form one plate. If this is so, then fusion has occurred in all
of the monosphyronid families; the alternative hypothesis is preferred here because
this involves a minimum of only four changes within the suborder (one each in the
Sternophoridae and Cheiridiidae, and one or more in the Cheliferidae and Chernetidae).

Thus, the character state displayed by the genus Afrosternophorus is thought to be
plesiomorphic, and the character state displayed by Garyops and Idiogaryops is assumed
to be apomorphic.

204

THE JOURNAL OF ARACHNOLOGY

Table 3. -Characters used in the construction of the cladogram of the Sternophoridae (Fig. 144).

Character

Plesiomorphic state

Apomorphic state

1. Female median cribriform plate(s)

2. Female median cribriform plate(s)

3. Trichobothria of moveable chelal finger

4. Trichobothria of moveable chelal finger

1

without spurs

3

3

2

with spurs

2

2

Afrosternophorus Afrosternophorus idiogaryops Idiogaryops

aethiopicus group araucariae group pumilus paludis Garyops spp.

Fig. 144. -Cladogram of the Sternophoridae: open squares, plesiomorphic character states; closed
squares, apomorphic character states.

2. Female median cribriform plates(s). Lateral spurs on the median cribriform plates
are unique within the Pseudoscorpionida, and undoubtedly represent an autapomorphy
for Garyops.

3. and 4. Trichobothria of moveable chelal finger. Because most pseudoscorpions
possess 8/4 (i.e., eight trichobothria on the fixed chelal finger and four trichobothria on
the moveable chelal finger), it is assumed that this is the most plesiomorphic condition
within the order. If a lower number is considered to be the primitive state, then one must
postulate a series of independent additions to the trichobothriotaxies of all of the pseudo-
scorpion families. If 8/4 is the primitive condition, then one need only assumed that
reductions have occurred independently in a few families, such as the Garypidae, Olpi-
idae, Neobisiidae, Sternophoridae, Cheiridiidae and Chernetidae. The highest number in
the Sternophoridae is 7/3 and this is regarded as the plesiomorphic condition for the
family. Deviations from this pattern are known in five species of two genera (A. arau-
cariae, A. cavernae, A. fallax, A. xalyx and/, paludis) where each species lacks the tricho-
bothrium sb of the moveable chelal finger. This is considered to be apomorphic. Since it
has occurred in two separate genera, it obviously represents an interesting case of parallel-

ism.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

205

This loss of a trichobothrium appears to be neotenic, since the tritonymphs of the 7/3
species possess 7/2 trichobothria and once again sb is absent. Thus, it is not so much a
case of “losing” a trichobothrium, but of having a trichobothrium fail to appear in the
post-embryonic development of a species. Developmental data on these species with
reduced trichobothriotaxies are completely lacking.

The analysis of phylogenetic trends within the genera Garyops and Afrostemophorus,
especially the large A. aethiopicus group, is an extremely difficult task which has not
been attempted. This is mainly due to a series of apomorphic character states whose
sequence in time are virtually impossible to interpret. They are only briefly summarized
here: G. centralis (increased number of female galeal rami), A. aethiopicus (reduction of
dorsal apodeme), A. ceylonicus (reduction of dorsal apodeme; decreased number of galeal
rami), A. chamberlini (midpiece of lateral rod elongate; increased number of female galeal
rami), A. hirsti (reduction of dorsal apodeme; anterior apodeme brush-like),^, anabates
(anterior apodeme distally broad; often phoretic on sparassid spiders), A. papuanus
(anterior apodeme distally broad), A. fallax (midpiece of lateral rod elongate) and A.
xalyx (dorsal apodemes reduced). Afrostemophorus anabates and A, papuanus are
synapomorphic sister-species that are united by the presence of a distally broad anterior
apodeme.

Since Afrosternophoms and Idiogaryops are not based upon apomorphic character
states, and are defined by plesiomorphies, the possibility exists that they are paraphyletic.
Nevertheless, at the expense of a strict cladistic classification where such groups would
not be recognized, I have retained the three generic names because they are useful labels
denoting distinct differences in the female genitalia.

BIOGEOGRAPHY

The Sternophoridae is currently distributed in most parts of the world (Maps 1-6), but
it appears to conform quite readily to a “trans-antarctic” pattern, and may have had its
origins in Gondwanaland. Nevertheless, anomalies occur which need clarification before
definite statements may be made. These include;

(1) The absence of Neotropical records: If it is assumed that sternophorids originated
in Gondwanaland, then there are two hypotheses that may be presented to explain their
absence from South America: (a) they are extinct, or (b) they have yet to be collected. I
consider that the latter is the most plausible hypothesis since sternophorids are small,
pale, corticolous pseudoscorpions that, unless specifically searched for, are difficult to
find.

(2) Presence in Laurasia (North America): Many “Gondwanaland” organisms invaded
North America from South America when those two continents joined during the Creta-
ceous (Dietz and Holden 1970). Thus, it is not particularly surprising to find that at least
one pseudoscorpion group has undertaken a similar journey. Others apparently include
the Pseudogarypidae, Garypidae, Tridenchthoniidae and Atemnidae.

(3) Presence in Indo-China: This apparent anomaly may be explained by two models.
The first assumes that migration occurred from India after that subcontinent became
attached to Asia during the Cretaceous (Dietz and Holden 1970), and the second has its
foundations in recent studies which have shown that various portions of south-east Asia
may have been an integral part of Gondwanaland (Burton 1970, Ridd 1971, Crawford
1974, Stauffer 1974, Cooper 1980 and others), possibly lying between India and Austra-
lia (Ridd 1971).

206

THE JOURNAL OF ARACHNOLOGY

The presence of Idiogaryops pumilus on Little Cayman Island in the Caribbean is also
quite interesting. Perfit and Heezen (1978) hypothesized that during the Miocene, local-
ized uplift elevated the Cayman Islands (among others) above sea level. This suggests that
/. pumilus arrived on Little Cayman Island by phoresy or rafting subsequent to the is-
land’s appearance.

It is of interest to note that two sister-species united by a single synapomorphy,
Afrosternophorus anabates and A. papuanus, are geographically isolated, the former in
south-eastern Australia (Map 5) and the latter in Papua New Guinea (Map 6). New Guinea
(which includes Papua New Guinea and Irian Jaya) is now firmly considered to be an
integral part of the Australian continent (Nix 1982), and the two countries are thought to
have separated during the Eocene or Pliocene (Doutch 1972).

It is of interest to note the congruence between the generic classification proposed in
this study and the biogeography of each of the genera. Garyops and Idiogaryops are
found only in the New World and are more closely related to each other than to Afro-
sternophorus.

AC KNO WLEDGM ENTS

I wish to thank Dr. E. M. Benedict and Dr. D. R. Malcolm, Pacific University, Forest
Grove (JCC), Dr. L. Cederholm, Lund University, Lund (LU), Dr. M. Grasshoff, For-
schungsinstitut Senckenberg, Frankfurt (SM), Dr. J. Gruber, Naturhistorisches Museum
Wien (NHMW), Dr. B. Hauser, Museum d’Histoire Naturelle, Geneve (MHNG), Dr. J.
Heurtault, Museum National d’Histoire Naturelle, Paris (MNHN), Mr. J. M. Hunter,
Museum of Comparative Zoology, Cambridge (MCZ), Dr. D. H. Kavanaugh and Dr. W. J.
Pulawski, California Academy of Sciences, San Francisco (CAS), Prof V. A. Murthy,
Loyola College, Madras (VAM), Dr. A. Neboiss and Mr. K. Walker, Museum of Victoria,
Abbotsford (MV), Dr. N. I. Platnick, American Museum of Natural History, New York
(AMNH) for the loan of specimens, M. Kotzman, H. E. Parnaby, C. Silveira and N. and S.
Wentworth for collecting and forwarding to me Australian sternophorids. Dr. G. Etter-
shank. Dr. V. Mahnert, Dr. W. B. Muchmore, Dr. P. J. Gullan, D. Cukier and A. Boulton
for critically reviewing the manuscript (or part thereof), the National Parks and Wildlife
Service of Victoria for permission to collect pseudoscorpions in areas under their control,
C. D. Beaton and K. A. Pickerd for the Scanning Electron microscopy, and CSIRO,
Division of Entomology, for laboratory facilities. This work was funded in part by an
Austrahan Biological Resources Study grant.

LITERATURE CITED

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

Beier, M. 1932. Pseudoscorpionidea 11. Suborder Cheliferinea. Tierreich, 58:1-294.

Beier, M. 1951. Die Pseudoscorpione Indochinas. Mem. Mus. Nat. Hist. Nat., Paris (N. S.), 1:47-123.
Beier, M. 1953. Pseudoscorpione aus El Salvador und Guatemala. Senckenbergiana Biol,, 34:15-28.
Beier, M. 1954a. Report from Professor T. Gislen’s expedition to Australia in 1951-1952. 7. Pseudo-
scorpionidea. Lunds Univ. Arsskr., Avd 2, 50(3): 1-26.

Beier, M. 1954b. Pseudoscorpioniden aus den Belgischen Congo. Ann. Mus. R. Congo Beige N. S.
Zool., 1:132-139.

Beier, M. 1962. Pseudoscorpioniden aus der Namib-Wiiste. Ann. Transvaal Mus., 24:223-230.

Beier, M. 1963. Ordnung Pseudoscorpionidea (Afterskorpione). Bestimmungsbiicher Bodenfauna
Europas, 1:1-313.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

207

Beier, M. 1967. Pseudoskorpione aus dem tropischen Ostafrika (Kenya, Tansania, Uganda etc.). Ann.
Naturhist. Mus. Wien, 70:73-93.

Beier, M. 1971. Pseudoskorpione unter Araucarien Rinde in Neu-Guinea. Ann. Naturhist. Must. Wien,
75:367-373.

Beier, M. 1973a. Weiteres zur Kenntnis der Pseudoscorpioniden Sudwestafrikas. Cimbebasia, Ser. A,
2:97-101.

Beier, M. 1973b. Pseudoscorpionidea von Ceylon. Entomol. Scand. Suppl., 4:39-55.

Beier, M. 1974. Neue Pseudoskorpione aus Australien und Neu-Guinea. Ann. Naturhist. Mus. Wien,
78:203-213.

Beier, M. 1976. Pseudoscorpione von der dominicanischen Republik (Insel Haiti). Rev. Suisse Zook,
83:45-48.

Beier, M. 1982. Zoological Results of the British Speleological Expedition to Papua New Guinea 1975.
9. Pseudoscorpionidea. Acta Zook Bulgarica, 19:43-45.

Benedict, E. M. and D, R. Malcolm. 1977. Some garypoid false scorpions from western North America
(Pseudoscorpionida: Garypidae and Olpiidae). J. Arachnok, 5:113-132.

Benedict, E. M. and D. R. Malcolm. 1982. Pseudoscorpions of the family Chernetidae newly identified
from Oregon (Pseudoscorpionida, Cheliferoidea). J. Arachnok, 10:97-109.

Bird, J., D. P. Faith, L. Rhomberg, B. Riska and R. R. Sokak 1979. The morphs oi Pemphigus populi-
transversus: allocation methods, morphometries and distribution patterns. Ann. Entomol. Soc.
Amer., 72:767-774.

Brach, V. 1979. Species diversity and distributional relationships of pseudoscorpions from slash pine
{Pinus elliotti Eng.) in Florida (Arachnida, Pseudoscorpionida). Bull. Southern California Acad.
Sek, 78:32-39.

Burton, C, K. 1970. The palaeotectonic status of the Malay Peninsula. Palaeogeogr., Palaeoclimatok,
Palaeoecok, 7 :51-60.

Chamberlin, J. C. 1923. New and little known pseudoscorpions principally from the islands and
adjacent shores of the Gulf of California. Proc. California Acad. Sci., ser. 4, 12:353-387.

Chamberlin, J. C. 1930. A synoptic classification of the false scorpions or chela-spinners, with a report
on a cosmopolitan collection of the same. Part II. The Diplosphyronida (Arachnida-Chelonethida).
Ann. Mag. Nat. Hist., ser. 10, 5:1-48, 585-620.

Chamberlin, J. C. 1931. The arachnid order Chelonethida. Stanford Univ. Pubk, Univ. Ser., Biol. Sci.,
7(1): 1-284.

Chamberlin, J. C. 1932a. On some false scorpions of the super-family Cheiridioidea. Pan-Pacific
Entomol., 8:137-144.

Chamberlin, J. C. 1932b. A synoptic revision of the generic classification of the chelonethid family
Cheliferidae Simon. (Arachnida). Canadian Entomol., 64:17-21.

Chamberlin, J. C. 1943. The taxonomy of the false scorpion genus Synsphyronus, with remarks on the
sporadic loss of stability in generally constant morphological characters. Ann. Entomol. Soc.
Amer., 36:486-500.

Chamberlin, J. C. 1947. The Vachoniidae - a new family of false scorpions represented by two new
species from caves in Yucatan. Bulk Univ. Utah, Biol. Ser., 38:1-15.

Chamberlin, J. C. 1952. New and little-known false scorpions (Arachnida, Chelonethida) from Mon-
terey County, California. Bulk Amer. Mus. Nat. Hist., 99:262-312.

Cooper, M. R. 1980. Palaeobiogeography and the reassembly of Gondwanaland. South African J. Sci.,
76:35-37.

Crawford, A. R. 1974. A greater Gondwanaland. Science, 184:1179-1181.

Dietz, R. S.and J. C. Holden. 1970. The breakup of Pangaea. Sci. Amer., 223:30-41.

Doutch, H. F. 1972. The Paleogeography of northern Australia and New Guinea and its relevance to
the Torres Strait area. Pp. 1-10, In: Bridge and barrier: the natural and cultural history of Torres
Strait (D. Walker, ed.). Australian National University Press, Canberra.

Gabbutt, P. D, 1969. A key to all stages of the British species of the family Neobisiidae (Pseudo-
scorpiones: Diplosphyronida). J. Nat. Hist., 3:183-195.

Gabbutt, P. D. and M. Vachon. 1965. The external morphology and life history of the pseudoscorpion
Neobisium muscomm. Proc. Zook Soc. London, 145: 335-358.

Harvey, M. S. 1981a. Geogarypus rhantus sp. nov. (Pseudoscorpionida: Garypidae: Geogarypinae); a
generic addition to the Australian fauna. Mem. Queensland Mus., 20:279-283.

Harvey, M. S. 1981b. A checklist of the Australian Pseudoscorpionida. Bulk British Arachnok Soc.,
5:237-252.

208

THE JOURNAL OF ARACHNOLOGY

Harvey, M. S. 1982. A parasitic nematode (Mermithidae) from the pseudoscorpion '‘Sternophorus”
hirsti Chamberlin (Sternophoridae). J. Arachnol., 10:192.

Hennig, W. 1965. Phylogenetic systematics. Ann. Rev. Entomol., 10:97-116.

Hennig, W. 1966. Phylogenetic systematics. University of Illinois Press, Urbana, xiii + 263 pp.
Heurtault, J. 1983. Pseudoscorpions de Cote d’Ivoire. Rev. Arachnol., 5:1-27.

Hoff, C. C, 1949. The pseudoscorpions of Illinois. Illinois Nat. Hist. Surv., Bull., 24:413-498.

Hoff, C. C. 1956. Pseudoscorpions of the family Cheliferidae from New Mexico. Amer. Mus. Novit.,
No. 1804:1-36.

Hoff, C. C. 1958. List of the pseudoscorpions of North America north of Mexico. Amer. Mus. Novit.,
No. 1875:1-50.

Hoff, C. C. 1961. Pseudoscorpions from Colorado. Bull. Amer. Mus. Nat. Hist., 122:409464.

Hoff, C. C. 1963. Sternophorid pseudoscorpions chiefly from Florida. Amer. Mus. Novit., No. 2150:
1-14.

Hoff, C. C. and J. E. Bolsterli. 1956. Pseudoscorpions of the Mississippi River drainage basin area.
Trans. Amer. Micros. Soc., 75:155-179.

Hoff, C. C. and D. T. Jennings. 1974. Pseudoscorpions phoretic on a spider. Entomol. News, 85:21-22.
Hounsome, M. V. 1980. The terrestrial fauna (excluding birds and insects) of Little Cayman. Pp.
81-90, in: Geography and ecology of Little Cayman (D. R. Stoddart and M. E. C. Giglioli, eds.).
Atoll Res. Bull., No. 241.

Legg, G. 1973. Spermatophore formation in the pseudoscorpion Chthonius ischnocheles (Chthon-
iidae). J. Zool., 170:367-394.

Legg, G. 1974a. An account of the genital musculature of pseudoscorpions (Arachnida). Bull. British
Arachnol. Soc., 3:38-41.

Legg, G. 1974b. A generalised account of the female genitalia and associated glands of pseudoscor-
pions (Arachnida). Bull. British Arachnol. Soc., 3:4248.

Legg, G. 1974c. The genitalia and accessory glands of the pseudoscorpion Cheiridium museomm
(Cheiridiidae). J. Zool., 173:323-339.

Legg, G. 1975a. A generalised account of the male genitalia and associated glands of pseudoscorpions
(Arachnida). Bull. British Arachnol. Soc., 3:66-74.

Legg, G. 1975b. The genitaha and associated glands of five British species belonging to the family
Chthoniidae (Pseudoscorpiones: Arachnida). J. Zool., 177:99-121.

Legg, G. 1975c. The genitalia and associated glands of five British species belonging to the family
Neobisiidae (Pseudoscorpiones: Arachnida). J. Zool., 177:123-151.

Mahnert, V. 1981. Sigles trichobothriaux chez les pseudoscorpions (Arachnida). Atti Soc. Toscana Sci.
Nat. Pisa, Mem. B, 88 suppl.:185-192.

Mahnert, V. 1982a. Die Pseudoskorpione (Arachnida) Kenyas 11. Feaellidae, Cheiridiidae. Rev. Suisse
Zool., 89:115-134.

Mahnert, V. 1982b. Die Pseudoskorpione (Arachnida) Kenyas, IV. Garypidae. Ann. Hist.-Nat. Mus.
Nat. Hung., 74:307-329.

Moore, B. P. 1978. A remarkable new genus and three new species of cavernicolous Carabidae (Cole-
optera) (Zoological results of the British Speleological Expedition to Papua-New Guinea 1975, 2).
Int. J. SpeleoL, 9:357-363.

Muchmore, W. B. 1973. A new genus of pseudoscorpions based upon Atemnus hirsutus (Pseudoscor-
pionida: Chernetidae). Pan-Pacific Entomol., 49:43-48.

Muchmore, W. B. 1982a. The genus Anagarypus (Pseudoscorpionida: Garypidae). Pacific Insects,
24:159-163.

Muchmore, W. B. 1982b. Some new species of pseudoscorpions from caves in Mexico (Arachnida,
Pseudoscorpionida). Ass. Mexican Cave Stud. Bull., 8:63-78.

Muchmore, W. B. and C. Alteri. 1974. The genus Parachernes (Pseudoscorpionida, Chernetidae) in the
United States, with descriptions of new species. Trans. Amer, Entomol. Soc., 99:477-506.

Murthy, V, A. and T. N. Ananthakrishnan. 1977. Indian Chelonethi, Oriental Insects Monogr., 4:1-

210.

Nelson, S., Jr. 1982. The external morphology and hfe history of the pseudoscoipion Microbisium
confusum Hoff. J, Arachnol., 10:261-274.

Nix, H. 1982. Environmental determinants of biogeography and evolution in Terra Australis. Pp.
47-66, In: Evolution of the flora and fauna of arid Australia (W. R, Barker and P, J. M. Greenslade,
eds). Peacock Publications, Frewville.

HARVEY— SYSTEMATICS OF STERNOPHORIDAE

209

Perfit, M. R. and B. C. Heezen. 1978. The geology and evolution of the Cayman Trench. Geol. Soc.
Amer. Bull., 89:1155-1174.

Platnick, N. I. and H. W. Levi. 1973. On family names of spiders. Bull. British Arachnol. Soc., 2:166-
167.

Redikorzev, V, 1938. Les pseudoscorpions des I’lndochine fran^aise receullis par M. C. Dawydoff.

Mem. Mus. Nat. Hist. Nat., Paris (N. S.), 10:69-116.

Ridd, M. F. 1971. South-east Asia as a part of Gondwanaland. Nature (London), 234:531-533.
Rowland, J. M. and J. R. Reddell. 1976. Annotated checkhst of the arachnid fauna of Texas (exclud-
ing Acarida and Araneida). Occ. Pap. Mus. Texas Tech Univ., No. 38:1-25.

Sivaraman, S. 1981. Systematics of some South Indian sternophorid pseudoscorpions (Pseudoscor-
pionida, Monosphyronida). Rev. Suisse Zook, 88:313-325.

Stauffer, P. H. 1974. Malaya and southeast Asia in the pattern of continental drift. Geol. Soc. Malay-
sia, BuU., 7:89-138.

Vachon, M. 1934. Sur le developpement postembryonnaire des pseudoscorpionides. Premiere note.
Bull. Soc. Zook, France, 59:154-160.

Vachon, M. 1936. Sur le developpement postembryonnaire des pseudoscorpion (quatrieme note).
Bulk Mus. Nat. Hist. Nat. Zook, (2) 8:77-83.

Vachon, M. 1951. Les pseudoscorpions de Madagascar. 1. Remarques sur le famille des Chernetidae J.
C. Chamberhn, 1931, a propos de la description d’une nouvelle Metagoniochernes milloti.

Mem, Inst. Sci. Madagascar, Ser. A, 5:159-172.

Vachon, M. 1973. Etude des caracteres utilises pour classer les families et les genres de Scorpions
(Arachnides). 1. La trichobothriotaxie en Arachnologie. Sigles trichobothriaux et types de tricho-
bothriotaxie chez les Scorpions. Bull. Mus. Nat. Hist. Nat. Zook, (3) 140:857-958.

Weygoldt, P. 1969. The biology of pseudoscorpions. Harvard University Press, Cambridge, xiv + 145

pp.

Manuscript received March 1984, revised June 1984.

* • * . r V . . ,

^•. i iGBL

'''••• '•

■ ;.j) . ,

^ ■ .tj

iA‘ ,

^ .Arr-fe5^ ,;f^n J»mA

<xi ,v/ , j; ,,

iiM

...vt r><:r . ^f:^' ,/ .-. .;;'rbW4>^ ‘ S_

?f,; c ^ ^ .1 ■-.■ .tKo •

■■..•I. .;;, . . V> , ' ■ ' !»)*»«» ■.^•V ‘f ■■ • „ ;/:'lt'? ,.to‘ • ' ■

- . -, '^ vrtfclf^.vOT : ■ -f ■ ]

• ' ' iflBaH '^ '' -'■'’ ' * ...;rf.<??.M ;ii'iw|ff™. ^ '»

■ ~~ ® Jv >A

■' ' ’’V', r-:4 n »' ''■ w

’ '*♦ -* ! !gf ■‘Ci>v:ir ■' >A^'':Y''- /'‘^,a\'4 .‘'I'JwV '

'< . .., ,. V: v.^:,»' \A,7.vv,ir.|,,; M • ‘ ■, . > J i

. : V: ' ■ ^

«• Vi-„1 . ;■, ... ^ IlV0i»

■ *• ■ -■ ?ii '- ' > , ' • ‘i'. . 1 ■ T i 'h-f!^^ '.i4Mfi' ni . , * . -i^' ' ■ • * iiii

•• ■-', ' -il'r^. 7'x\ .;x»-^ r ,H .i|iK .f4>M n ;(i

i»

■*.;’

,-r

. i

Ct:

M

7, .V-

it *4

a

M.25^ t'.

i: I

*« • •« if 1 4 (". V ■ f

'. I'-.'

y'’

- r^- ■ fJ . .1 ■ J 4 V * ''^:',;ijiSj^«

if flMf"^ A , ifecrt^vf 4lt- 1 *£?:

Tr>>sm%

■ -v^' -i
■® ” "

' ■ - V .=^

^V:K 'r

1**4 .■* ;

'It' '* ' '

* . y':

. . ^ i '

‘ .* •<”■'

*,*

^ V'

i t •J.J'U.g

t'M

‘(itftfjli' '

^v.'M

• ^ :*ii ..: ■ ■ ^«S«a|||(

■^ ,, .t •' 1

" ... . ■ ‘ . ■ It ._ . "• -'•’ « »- -’ijl- - A'tf'"'

O

Coyle, F. A. 1985. Two-year life cycle and low palpal character variance in a Great Smoky Mountain
population of the lamp-shade spider (Araneae, Hypochilidae, Hypochilus). J. Arachnol., 13:211-
218.

TWO-YEAR LIFE CYCLE AND LOW PALPAL CHARACTER
VARIANCE IN A GREAT SMOKY MOUNTAIN POPULATION

OF THE LAMP-SHADE SPIDER
(ARANEAE, HYPOCHILIDAE, 7/FPDC///LU5^)

Frederick A. Coyle

Department of Biology
Western Carolina University
Cullowhee, North Carolina 28723

ABSTRACT

Size-frequency histograms and other data generated from four samples (totaling 926 specimens)
collected during a complete year show that a Hypochilus population in the Great Smoky Mountains
has a two-year life cycle with the following schedule: spiderlings emerge from egg sacs and construct
their first webs in late May; 15 to 18 months later, during their second autumn, these spiders mature,
mate, and lay eggs. The growth rate and adult body size variances of this population are very large.
The coefficients of variation of three palpal dimensions in a sample of 38 males are significantly
smaller than those of tibia I length or carapace length. Such relative constancy of palpal characters
within a population may be common in spiders and may result from stabilizing selection in one or
both of the following forms: selection for the mechanical compatibility necessary for effective sperm
placement during copulation, and sexual selection by female choice.

INTRODUCTION

In his study of several populations of the lamp-shade spider, Hypochilus,
Fergusson (1972) concluded, chiefly from the frequency distribution of tibia I
lengths of 164 individuals collected in late October, that this species has a two-
year life cycle. The current study was undertaken primarily to rigorously test this
life-cycle hypothesis and, secondarily, to determine whether the coefficients of
variation (standard deviation x 100/ mean) of palpal dimensions are less than the
large coefficients of variation of male body dimensions in this population. Even
though the relative constancy of genital characters is often noted in taxonomic
revisions (for example Reiskind 1969, Levi 1981) and may be one of the reasons
taxonomists rely so heavily on genital characters to diagnose species of spiders
and many other animals (Mayr 1969), I am not aware of any rigorous statistical
attempt to confirm its occurrence in any species of spider. It is important to note
that Hoffman (1982, pers. comm.) recently concluded that the southern Blue
Ridge Province populations of Hypochilus which Fergusson (1972) and I have
studied represent a new species (as yet undescribed) that is distinct from the
Appalachian Plateau populations of Hypochilus thorelli.

212

THE JOURNAL OF ARACHNOLOGY

METHODS

Life History. — Four collections of individuals from a single, dense Hypochilus
population living on rock outcrops along Little River Road on the Tennessee side
of the Great Smoky Mountain National Park were made between 1.5 miles
upriver and 1.5 miles downriver from The Sinks on the following days (sample
size in parentheses): 1 October 1982 (207), 8 May 1983 (134), 1 June 1983 (287),
and 30 September 1983 (298). For each collection an attempt was made to
carefully examine entire areas of rock outcrop surface (different for each
collection) and collect every Hypochilus individual within reach. However, during
the October 1 collection a number of adult males and females that were spotted
were not collected, and during the June 1 collection it was possible that not all
reachable spiderlings were collected since these are very small and difficult to see.

All specimens were preserved in 70% ethanol. The length of the first tibia of
each specimen was measured by me with a Wild M5 stereomicroscope fitted with
20x eyepiece lenses with a 100-unit eyepiece reticle scale. Tibia I length was
selected as an indicator of body size because measuring carapace or sternum
dimensions would have required the removal of all legs. Tibia I length is defined
as the straight line distance between the proximal and distal ends of the article
in lateral view with the article on the horizontal plane. In order to avoid using
a value for a regenerating leg, both the left and right tibia I were measured and
the value for the left recorded unless it was more than three scale units shorter
than the right tibia. Measurements were recorded to the nearest half unit and
repeated measurements of two specimens indicated that measurements were
precise to within 0.04 mm for the smallest specimens and 0.23 mm for the largest
ones. If a specimen belonged in any of the following categories, such was noted:
adult female (any female with developing or fully developed eggs and a markedly
hirsute area just anterior to the genital groove); adult male; penultimate male
(identified by swollen pedipalp tarsi); about to molt (any individual with dark
setae of new exoskeleton visible under old exoskeleton).

Palpal Character Variance. — Measurements were recorded for two non-genital
characters that are frequently used as indicators of body size and three palpal
characters on each of the 38 adult Hypochilus males collected on September 30.
Extreme care was taken to minimize measurement imprecision, an especially
important goal when generating coefficients of variation (Rohlf, Gilmartin, and
Hart 1983). Structures were selected which can be consistently positioned, and
whose linear dimension has well-defined end points. The two non-genital
characters are defined as follows: ITL = tibia I length as defined above; CL =
distance along median longitudinal line connecting anterior edge of carapace to
median indented posterior edge of carapace with all legs removed and carapace
horizontal. The three palpal characters were measured on the left pedipalp after
it was removed from the body and are defined as follows: PTC = width of palpal
tarsus “clasper” in retrolateral view (Fig. 1); PL = length of palpus in retrolateral
view (Fig. 1); CdL = length of palpal conductor in prolateral view (Fig. 2). The
measurements were performed by me with the optical equipment described above.
ITL was measured to the nearest end point of a scale-unit 0.077 mm long, CL
to the nearest 0.037 mm scale-unit end point, and the palpal characters to the
nearest 0.009 mm scale-unit end point. In order to determine the imprecision
involved in making these measurements, one specimen was remeasured ten times

COYLE-PALPAL VARIANCE AND LIFE CYCLE OF HYPOCHILUS

213

2

Fig. 1-2.— Left palpal tarsus and palpus of Hypochilus male showing three palpal measurements
as defined in text: 1, retrolateral view; 2, prolateral view.

over the period of a week (after each set of five measurements was recorded the
examining dish was swirled so that each part would have to be repositioned
during the next set of measurements), and the coefficient of variation was
calculated for each of the five measurement characters for that sample of ten
remeasurements.

RESULTS

Life History. — Frequency distribution histograms of tibia I length for each of
the four collections are presented in Fig. 3, and statistical values for the age, sex,
and molting classes represented in these histograms are presented in Table 1. The
May 8 collection histogram indicates that there is only a single age class — all
juveniles — present at that time. Although the size range is large, the distribution
is monomodal and is not skewed markedly to the right, indicating that it does
not comprise two or more age classes. The relatively small size of individuals
about to molt also supports this conclusion because, in a homogeneous and
relatively young age class with non-synchronous molting, small individuals should
be further in time than large individuals from their last molt, and thus more
likely to molt than large individuals. The June 1 collection reveals two age
classes, a class of spiderlings recently emerged from egg sacs and a class of older
juveniles. The latter class, despite its large size range, exhibits those characteristics
(non-skewed shape and restriction of molting to smaller individuals) which
indicate that it is one age class.

The age class patterns of the histograms for both fall collections are similar
to each other. In each there is a single juvenile class with large size variance but
with the non-skewed size distribution and molting activity pattern which indicate
age homogeneity. In addition, each collection contains an adult age class with an
extremely large size variance but with only a slight size overlap with the juvenile
class. That this adult class is made up of same aged individuals in their second
autumn is indicated by the absence of adults during the spring and early summer

214

THE JOURNAL OF ARACHNOLOGY

and by the simultaneous presence of a single juvenile age class. In both fall
collections some penultimate males, half of which are beginning to molt, are
present. The distinct difference between the frequency distributions of adult
females and adult males is due to the allometric growth of legs in the final male
molt.

The large body size variance of the juvenile classes as compared to the
spiderling class (Table 1) shows that there is a large variance in growth rate, with
some individuals growing five to six times (linear dimensions) faster than their
slowest growing contemporaries during the first full year of growth following
spiderling emergence. This early growth rate variance helps create the large
variance in body size of adult female and adult male subsamples of the
population. This large growth rate variance and the slight overlap between the

20. 10CT .

1982 jH

I JUVENILES

T 1 1 — 1 r

0-

ADULT MALES ▼

^

, , ^

10H

0

0

20 1

10

10

CO

_l

<

3

^ 0
< 0

8 MAY
1983

T

I , JUVENILES

10

q40

>

Q 30

SPIDERLINGS

1 JUNE
1983

ADULT FEMALES

n 1 1 1 r

Ll 20“ I JUlNtl ^

^ Ijj JUVENILES

1 ,11,, iiliiH^ii, ,

^ 0 5 10

20.

30 SEPT

10-

TiBIA I LENGTH (mm)

^ ■, F ^

— I 1 1 1 r 1 1 1 1 1 1 1 1 1

15 20 25

ADULT MALES

Fig. 3. — Frequency distributions of tibia I lengths for four collections from a Hypochilus
population. Adults of each sex grouped on separate scale lines. Each square represents one specimen.
White squares represent penultimate males. White circles represent individuals ready to molt.
Triangles mark means of each age/ sex class sample.

COYLE— PALPAL VARIANCE AND LIFE CYCLE OF HYPOCHILUS

215

Table 1. — Statistical
Hypochilus population

values for tibia I
samples represented

length (in
in Fig. 1 .

mm) for the

age/ sex/

' molting

classes of the

Class

Date

N

Range

Mean

Variance

Coefficient

of

Variation

Spiderlings

1 June ’83

44

0.92-1.30

1.13

0.004

5.60

Juveniles (Year 1)

1 Oct. ’82

156

2.72-7.85

5.09

1.07

20.32

About to Molt

1 Oct. ’82

7

3.85-5.39

4.48

0.31

12.43

Juveniles (Year 1)

30 Sept. ’83

155

1.72-9.09

4.68

1.48

26.00

About to Molt

30 Sept. ’83

7

2.59-4.70

3.61

0.44

18.37

Juveniles (Year 2)

8 May ’83

134

3.70-10.32

6.80

1.49

17.95

About to Molt

8 May ’83

4

3.81-5.01

4.57

0.20

9.79

Juveniles (Year 2)

1 June ’83

243

3.18-13.09

7.66

3.23

23.46

About to Molt

1 June ’83

23

4.16-7.70

6.53

1.03

15.54

Adult Females

1 Oct. ’82

34

7.16-17.79

12.20

9.74

25.58

Adult Females

30 Sept. ’83

103

7.85-19.02

14.63

4.92

15.16

Adult Males

1 Oct. ’82

9

15.40-26.80

23.09

14.66

16.58

Adult Males

30 Sept. ’83

38

15.02-27.87

21.56

12.49

16.39

juvenile and adult classes in tibia I length in the fall raise the possibility, albeit
very remote, that a very fast growing individual might occasionally mature in its
first fall and that a very slow growing individual might, on occasion, fail to
mature until its third fall.

Each of the ten egg sacs collected on October 1 and September 30 contained
eggs, many of which contained embryos with appendage buds readily visible. Of
the eight egg sacs collected on May 8, five were preserved and contained faintly
pigmented spiderlings, whereas the remaining three were kept in the laboratory
until spiderlings emerged on 13, 22, and 24 May. These spiderlings had longer
legs, more hair, and more pigment than those preserved on May 8, and had the
same tibia I lengths as those collected in their webs on 1 June. They spun silk
in the glass vial within which they emerged and, when disturbed, bounced on this
webbing. Each of the 12 egg sacs collected on June 1 had been vacated.
Apparently the great majority of spiderlings in this population had emerged from
their egg sacs between May 8 and June 1. In spite of diligent searching on June
1, fewer of these spiderlings were found than expected. These newly emerged
spiderlings were found (in their webs) only on the more sheltered outcrop surfaces
where the light intensity was lowest (and probably the humidity was highest). On
the more exposed outcrop surfaces these spiders and their webs were absent or
only the bases of their abandoned webs were present, even though older spiders
were common. Perhaps these spiderlings tend to hide in crevices during dry
periods and build capture webs only at night and/ or during humid days.

In summary, the size/ age frequency distribution, egg sac, and spiderling data
show that this Hypochilus population has a two-year life cycle with the following
schedule; eggs are deposited in the fall, spiderlings emerged from egg sacs and
construct their first webs during the second half of May, these spiderlings then
eventually mature, mate, and lay eggs 15 to 18 months later in their second fall,
and adults seldom, if ever, survive to reproduce for more than one breeding
season.

216

THE JOURNAL OF ARACHNOLOGY

Palpal Character Variance. — For the sample of 38 males the coefficient of
variation of each palpal character (PTC, PL, CdL; Table 2) is significantly less
(using Lewontin’s (1966) method, P < 0.01) than the coefficient of variation of
either non-genital character (ITL, CL). Since the measurement imprecision indices
(Table 2) indicate that the three palpal characters are more difficult to measure
accurately than are the non-genital characters, factoring out the effect of
measurement imprecision on the coefficient of variation of the sample should not
reduce (indeed, it would increase) the significance of these differences. It is
important to point out that the difference between the coefficients of variation
of ITL and CL is largely due to the allometric increase of leg length during the
final molt.

DISCUSSION

Life History. — These results strengthen Fergusson’s (1972) hypothesis that
Hypochilus populations in the southern Blue Ridge Province have a two-year life
cycle, with adult females rarely, if ever, surviving to a second year of
reproduction. Although spiders generally exhibit large intrapopulation size
variation as adults (Jocque 1981), the very large variation in growth rate and,
consequently, adult body size observed in this Hypochilus population and in
some araneid populations (Crome and Crome 1961, Vollrath 1980, Levi 1981)
appears to surpass that of most other spiders investigated (for example Dondale
1961, Reiskind 1969, Vogel 1970, Brady 1979, Jocque 1981). It seems reasonable to
postulate that the most probable primary cause of this especially large growth
rate variance in Hypochilus and some other spiders is a large variance in prey
capture success. Whether this prey capture variance results mainly from a patchy
prey distribution and an inability to move webs quickly to more profitable sites,
or from other factors, remains to be investigated. Riechert and Cady (1983)
present evidence suggesting that the spiders Achaearanea tepidariorum and
Coelotes montanus, both common inhabitants of the rock outcrops where
Hypochilus lives, may have an important impact on Hypochilus foraging success
by serving as a source of food (A. tepidariorum) or by competing for space (C

Table 2. Statistics for five measurement (mm) characters for a population sample of 38 adult male
Hypochilus. Measurement imprecision index is the coefficient of variation of a sample of ten
remeasurements of a single specimen. Characters are defined in text. The coefficient of variation for
each a palpal character is significantly less (P<0.01) than that of either non-genital character.

Character

Range

Mean

Variance

Coefficient

of

Variation

Measurement

Imprecision

Index

Non-Genital

ITL

15.02-27.87

21.56

12.49

16.40

0.348

CL

2.73-4.54

3. '69

0.23

12.85

0.293

Palpal

PTC

0.52-0.70

0.61

0.002

8.15

1.149

PL

1.07-1.37

1.25

0.006

6.39

0.362

CdL

0.65-0.82

0.75

0.002

6.37

1.005

COYLE— PALPAL VARIANCE AND LIFE CYCLE OF HYPOCHILUS

217

montanus). It is therefore possible that a patchy distribution of these species over
the outcrop surfaces inhabited by Hypochilus might help create a large variance
in Hypochilus foraging success.

Palpal Character Variance. — The widespread reliance of spider systematists
upon genital characters to diagnose species is presumably due primarily to the
tendency of these characters to evolve more rapidly and divergently than other
characters. However, the analysis of variance in this sample of Hypochilus males
suggests that the usefulness of genital characters for diagnosing species may also
be due to a tendency for genitalia to vary less than other characters within
populations and groups of freely interbreeding populations. Such genital
character constancy is frequently implied or demonstrated indirectly in taxonomic
papers (for example McCrone 1963, Reiskind 1969). Vollrath’s (1980) data for a
sample of 38 Nephila clavipes males indicate relatively low variance in palpal
conductor length. With a sample of only two Tetragnatha elongata males, Levi
(1981) has illustrated the same phenomenon of relative palpal character
constancy. Whether this phenomenon is widespread in spiders awaits analyses of
variance in other taxa, but it seems heuristic to suggest two possible causes for
such a phenomenon.

If, as seems likely, there is some optimal range of palpus size and shape
required for quick and accurate insertion into the copulatory bursae and
spermathecae of the females of this Hypochilus population in order to achieve
effective semen transfer, there should then be selection for a developmental
mechanism that insures that a male’s palpus will approach that optimum form
independent of his body size. In other words, although there is no evidence to
suggest that the species-specific shape of Hypochilus genitalia (Gertsch 1958,
1964, Hoffman 1963) is a mechanical (lock and key) reproductive isolating
mechanism (indeed the unsclerotized nature of the female genital region, bursa,
and spermathecae [Coyle et al. 1983] argue against such a function), selection
could be acting to maintain some degree of mechanical compatibility in these
copulatory structures.

Sexual selection by female choice acting late in courtship or during copulation
may be another cause for relatively low intrapopulation variance in palpal
characters. As suggested by Eberhard (in press), the tactile information produced
by the male palpus during copulation may affect the behavior and/or physiology
of the female in a way that affects the male’s success in fertilizing her eggs. That
is, the female may be selecting mates partly on the basis of her tactile perception
of palpus form. Once a preference is established by a proportion of the
population’s females, there may be strong stabilizing selection for that palpus size
and shape and, consequently, a developmental mechanism that regulates palpus
size independent of body size. Although West-Eberhard (1983) has justifiably
emphasized the probable importance of sexual selection in causing rapid
evolution that may accelerate speciation, it is also important to recognize that
under some circumstances sexual selection may instead have a stabilizing effect.

218

THE JOURNAL OF ARACHNOLOGY

ACKNOWLEDGMENTS

I thank J. M. Palmer and R. C. Bruce for helping me collect the Hypochilus
samples. R. C. Bruce, W. G. Eberhard, and A. M. Moore provided helpful advice
concerning data analysis and criticized drafts of this paper.

LITERATURE CITED

Brady, A. R. 1979. Nearctic species of the wolf spider genus Trochosa (Araneae; Lycosidae). Psyche,
86:167-212.

Coyle, F. A., F. W. Harrison, W. C. McGimsey and J. M. Palmer. 1983. Observations on the structure
and function of spermathecae in haplogyne spiders. Trans. Amer. Microsc. Soc., 102:272-280.

Crome, W. and J. Crome. 1961. “Wachstum ohne Hautung” und Entwicklungsvorgange bei den
Weibchen von Argiope hruennichi (Scopoli). Deutsch. Entomol. Z., 8:443-464.

Dondale, C. D. 1961. Life histories of some common spiders from trees and shrubs in Nova Scotia.
Canadian J. Zool., 39:777-787.

Eberhard, W. In press. Why are genitalia good species characters? Proc. 9th Internat. Congr.
Arachnol., Panama 1983.

Fergusson, I. C. 1972. Natural history of the spider, Hypochilus thorelli Marx (Hypochilidae). Psyche,
79:179-199.

Gertsch, W. J. 1958. The spider family Hypochilidae. Amer. Mus. Novitates No. 1912, 28 pp.

Gertsch, W. J. 1964. A review of the genus Hypochilus and a description of a new species from
Colorado (Araneae, Hypochilidae). Amer. Mus. Novitates No. 2203, 14 pp.

Hoffman, R. L. 1963. A second species of the spider genus Hypochilus from eastern North America.
Amer. Mus. Novitates No. 2148, 8 pp.

Hoffman, R. L. 1982. The Appalachian species of Hypochilus. Amer. Arachnol. No. 26, p. 1 1.

Jocque, R. 1981. Size and weight variations in spiders and their ecological significance. Biol. Jb.
Dodonaea, 49:155-165.

Levi, H. W. 1981. The American orb-weaver genera Dolichognatha and Tetragnatha north of Mexico
(Araneae: Araneidae, Tetragnathinae). Bull. Mus. Comp. Zool., 149:271-318.

Lewontin, R. C. 1966. On the measurements of relative variability. Syst. Zool., 15:141-142.

Mayr, E. 1969. Principles of Systematic Zoology. McGraw-Hill, New York, 428 pp.

McCrone, J. D. 1963. Taxonomic status and evolutionary history of the Geolycosa pikei complex in
the southeastern United States (Araneae, Lycosidae). Amer. Midi. Nat., 70:47-73.

Reiskind, J. 1969. The spider subfamily Castianeirinae of North and Central America (Araneae,
Clubionidae). Bull. Mus. Comp. Zool., 138:163-325.

Riechert, S. E. and A. B. Cady. 1983. Patterns of resource use and tests for competitive release in
a spider community. Ecology, 64:899-913.

Rohlf, F. J., A. J. Gilmartin and G. Hart. 1983. The Kluge-Kerfoot phenomenon — a statistical
artifact. Evolution, 37:180-202.

Vogel, B. R. 1970. Taxonomy and morphology of the sternalis and falcifera species groups of Pardosa
(Araneida: Lycosidae). The Armadillo Papers No. 3, 31 pp.

Vollrath, F. 1980. Male body size and fitness in the web-building spider Nephila clavipes. Z.
Tierpsychol., 53:61-78.

West-Eberhard, M. J. 1983. Sexual selection, social competition, and speciation. Quart. Rev. Biol.,
58:155-183.

Manuscript received June 1984, revised August 1984.

Ibarra N., G. 1985. Egg feeding by Tegenaria spiderlings (Araneae, Agelenidae). J. Arachnol., 13:219-
223.

EGG FEEDING BY TEGENARIA SPIDERLINGS
(ARANEAE, AGELENIDAE)i

Guillermo Ibarra N.

Centro de Investigaciones Ecologicas del Sureste^

Area Agroecologica, 2a. Poniente 36
Ap. Postal 36, 30700 Tapachula, Chiapas, Mexico

and

Laboratoire d’Ethologie et Sociobiologie, Universite Paris XIII
Av. Jean-Baptiste-Clement, 93430 Villetaneuse, France

ABSTRACT

Egg feeding activity has been observed within the egg sac in spiderlings of two species of Tegenaria
Latreille (Agelenidae), in which this phenomenon had not been previously reported. The two species
differ in their proportion of egg consumption; their ecological characteristics and life cycle are
probable factors that account for the proportional difference in egg feeding.

INTRODUCTION

In spiders, eclosion leads to a stage in which spiders are still incompletely
developed. They have little motility and cannot weave or capture prey. In order
to complete their development they must remain within the egg sac during a
certain length of time which varies according to the species and the conditions
of development. This stage, named incomplete (Holm 1940), larval (Vachon
1957), deutovum (Gertsch 1949) or quiescent (Valerio 1974), consists of one or
more phases, separated by the shedding of the vitelline membranes. At the end
of this stage the first true molt takes place and the larva transforms into a
nymph, which is defined as being complete, active and with functioning venom
and silk glands (Vachon 1957, Foelix 1979). This nymph remains inside the egg
sac for some time before emerging to begin its solitary life. Some authors (Burch
1979, Turnbull 1973) assume that when the nymphs emerge from the egg sac they
still have vitelline reserves from which they feed until dispersion occurs.
Nevertheless it is known that in several species, the spiderlings (larvae or nymphs)
feed on inviable eggs before emergence, as in the families Loxoscelidae (Galiano

'This work was supported by grant QCECBFR-005 104 from the Consejo Nacional de Ciencia y
Tecnologia, Mexico.

^Present address.

220

THE JOURNAL OF ARACHNOLOGY

1967), Clubionidae (Lecaillon 1904, Mansour et al. 1980, Peck and Whitcomb
1970), Gnaphosidae (Holm 1940), Thomisidae (Schick 1972) and Theridiidae
(Juberthie 1964, Kaston 1970, Valerio 1974).

I have detected egg feeding by first nymphs in two species of Tegenaria
Latreille ( Agelenidae), with a remarkable difference in the proportion in which
the phenomenon' occurs in each species. I believe the difference might probably
be determined as much by ecological dissimilarities in the habitat of each of these
species as by their life cycles.

MATERIALS AND METHODS

The first observations were made on Tegenaria sp. during a study on the
predatory behavior of this species (Ibarra 1983, in press). Adults were collected
in a tropical forest (Chiapas, Mexico) and were kept and mated in the laboratory.
The females wove their egg sacs, from which the spiderlings later emerged, within
the cages. A record was kept of the number of spiderlings that emerged from
each egg sac and later they were separated individually to start the predatory
behavior study.

Later, adult T saeva Blackwall were collected in buildings in the suburbs of
Paris (Prance). These spiders were mated in the laboratory and four egg sacs were
removed from the cages a short time after they were woven. Each of these was
cut approximately one centimeter in length along its external silk layer, in order
to open it enough to display the eggs in its interior. Each egg sac was placed
in a plastic box with the observation window facing upwards. Two pieces of
sponge at either side supported a slide which covered the observation window in
order to protect the eggs from air currents and possible intruders, without
obstructing their observation. The egg sacs were observed once a day, until
eclosion took place. From that moment on, three observations were made daily,
each consisting of five minutes. One of the egg sacs was exposed to programmed
photographic equipment and photographs were taken every two hours.

RESULTS

In 350 spiderlings of Tegenaria sp. that emerged from six egg sacs, no evident
difference in size was observed, except for 15 individuals which had a
considerably larger opisthosoma than the rest, all these 15 emerged from the same
egg sac (this being the last produced by one of the reproductive females). These
spiderlings had an accelerated rate of development and underwent the second
molt between four and seven days after they emerged, whereas the rest took from
21 to 45 days to do so.

In the four egg sacs of T saeva 162 spiderlings hatched from the eggs, but four
died before the emergence from the egg sac. I have observed egg feeding by 14
individuals, and a progressive enlarging of the opisthosoma in many others
(Figures 1 and 2). At emergence all of the 158 survivors had an expanded
opisthosoma, and there was no evident difference in size between them. The
observations on the behavior of these spiderlings inside the egg sac are

IBARRA— EGG FEEDING BY TEGENARIA SPIDERLINGS

221

Figs. 1-2. — First nymphs ot Tegenaria saeva inside the egg sac; 1, two days after molting, a nymph
is feeding on an egg (arrow); 2, four days after molting, note the more bulky opisthosomae because
of egg feeding. The circle is 5 mm in diameter.

summarized as follows: A) There is a close synchronization between individuals
for the eclosion, and later for the first molt. B) The motility of the larvae is
highly reduced in comparison to that of the nymphs. C) The nymphs can feed
repeatedly on eggs from the day after they undergo the first molt. D) An egg
is not consumed all at once, but progressively. In the majority of these cases, the
egg is consumed completely; only the chorion, which looks like a deflated ball,
is left. The increase in the volume of the opisthosoma is then clearly noticeable
(Figures 1 and 2). E) No aggressive act of behavior was observed, either among
the nymphs or the larvae. F) In the majority of these cases (153) the nymphs
molted the second time without any additional food.

DISCUSSION

The enlarged opisthosomae of 15 individuals of Tegenaria sp. made me suspect
a case of egg feeding. Forster (1977) observed that in salticid spiderlings, any
reinforcement in feeding results in the reduction in length of the first nymphal
instar. So the accelerated rate of development of the same 15 individuals of
Tegenaria sp. is a further evidence to assume that they fed on eggs. Therefore
in this species the proportion of egg feeding detected is 4.2% of the total number
of eclosed spiderlings. The egg feeding directly observed in several individuals of
T saeva, the similarity in size in all the survivors at emergence, and the fact that
a great number of survivors molted the second time without having any
additional food, permit me to state that all of them fed on eggs. So in this
species, the proportion of egg feeding is 97.6% for all the eclosed spiderlings; this
value differs greatly from that found in Tegenaria sp.

I can tell that the cases reported here are of spiderlings that fed on inviable
eggs and not on other spiderlings because: A) the nymphs tolerated one another
while they were feeding on the eggs, I saw no instance of aggression between
them, and the small number of spiderlings that died before the emergence is proof
that this should be rare; B) the Tegenaria spiderlings become capable of feeding
at the first nymphal stage, the larvae being too immobile to do it; C) The reduced

222

THE JOURNAL OF ARACHNOLOGY

motility of the larvae make them probably subjects of cannibalism by the
nymphs, but the synchronized development prevents this, and allows only the
inviable eggs (when these are present in the egg sac) as food for the nymphs.

Thus the difference in the percentages of egg feeding between the two species
is due in all probability to the proportion of inviable eggs which the female
deposits in the egg sac. So in Tegenaria sp. practically all of the eggs were viable,
except those in one sac, whereas in each sac of T. saeva there was a proportion
of inviable eggs sufficiently large enough for the nymphs to feed on before they
emerged from the sac. In various species of spiders, it has been observed that the
proportion of viable eggs drops in the last egg sacs of each female (Christenson,
Wenzl and Legum 1979, Horner and Starks 1972, Jackson 1978). This might be
the explanation for the single case of egg feeding detected in Tegenaria sp. It
might further indicate that the presence of eggs (whether they be viable or not)
is capable of evoking the egg feeding behavior in these nymphs.

Valerio (1974) demonstrated that egg feeding helps to prolong the survival of
the first nymphs. Thus, egg sacs with a higher proportion of inviable eggs have
a higher probability of egg feeding; though their rate of eclosion is diminished;
the possibilities of survival for the emerged spiderlings are increased.

There seem to be two ways for optimizing the reproduction as a function of
the number of viable and inviable eggs in each sac. In one case, the vitellum
produced by the female is used to produce the largest number possible of
spiderlings (as in Tegenaria sp.). On the other hand the vitellum is used to
produce stronger individuals, but in smaller numbers (as in T saeva). What
determines the predominating way in each species? I believe the answer lies on
the particular life cycle and the characteristics of the habitat of each species.

In Tegenaria sp. the females make their egg sacs from November to March,
whereas T saeva does so from September to December. In the habitat of the first
species potential prey are abundant during the whole year, but for the second
species potential prey become scarce in the autumn, precisely when the spiderlings
emerge. The females of T saeva must then spend part of their reproductive
potential in inviable eggs so that a small number of nymphs develop enough to
survive the winter season, when the potential prey have practically disappeared
and the weather is unsuitable for active life. Because the Tegenaria sp. nymphs
do not have to face the scarcity of prey (neither at the moment of emergence nor
later), it is more profitable for the females of this species to produce the highest
number of nymphs possible. Thus each species optimizes the use of its own
resources (vitellum produced) in function of its possibilities to exploit the
resources of its habitat (potential prey).

ACKNOWLEDGMENTS

I would like to thank Dr. P. Jaisson and his coworkers (Laboratoire
d’Ethologie, Universite de Paris XIII, France) for their support and
encouragement while I was preparing my doctoral thesis, on a part of which this
paper is based. Gratitude is extended to Dr. A. Hoffmann and M. C. M. L.
Jimenez (Laboratorio de Acarologia, Facultad de Ciencias, Universidad Nacional

IBARRA— EGG FEEDING BY TEGENARIA SPIDERLINGS

223

Autonoma de Mexico) and to Drs. C. E. Valerio (Escuela de Biologia,
Universidad de Costa Rica) and W. A. Shear (Hampden-Sydney College, U.S.A.)
for their helpful comments and critical review of the manuscript. Dr. P. S. Baker
made the review of the English language.

LITERATURE CITED

Burch, T L. 1979. The importance of communal experience to survival for the spiderlings of Araneus
diadematus {P^V2LnQ2Lt: Araneidae). J. Arachnol., 7:1-18.

Christenson, T. E., P. A. Wenzl and P. Legum. 1979. Seasonal variation in egg hatching and certain
egg parameters of the golden silk spider, Nephila clavipes (Araneidae). Psyche, 86:137-147.

Foelix, R. F. 1979. Biologie der Spinnen. Thieme, Stuttgart, Germany.

Forster, L. M. 1977. Some factors affecting feeding behaviour in young Trite auricoma spiderlings
(Araneae, Salticidae). New Zealand J. Zool., 4:435-443.

Galiano, M. E. 1967. Ciclo biologico y desarrollo de Loxosceles laeta (Araneae: Scytodidae). Acta
Zool. Lilloana, 23:431-464.

Gertsch, W. J. 1949. American spiders. Van Nostrand, New Jersey.

Holm, A. 1940. Studien iiber die Entwicklung und Entwicklungsbiologie der Spinnen. Zool. Bidr.
Uppsala, 19:1-214.

Horner, N. V. and K. J. Starks. 1972. Bionomics of the jumping spider Metaphidippus galathea. Ann.
Entomol. Soc. Amer., 65:602-607.

Ibarra Nunez, G. 1983. L’ethogenese de la predation chez les araignees du genre Tegenaria (Araneae,
Agelenidae). Doctorate Thesis. University of Paris XIII.

Ibarra Nunez, G. (In press). La etogenesis de la predacion en las arafias del genero Tegenaria
(Agelenidae): 1. La discriminacion de las presas en las ninfas sin experiencia. Folia Entomologica
Mexicana.

Jackson, R. R. 1978. The life history of Phidippus johnsoni (Araneae: Salticidae). J. Arachnol., 6:1-
29.

Juberthie, C. 1964. Sur les cycles biologiques des araignees. Bull. Soc. Hist. Nat. Toulouse, 39:299-
318.

Kaston, B. J. 1970. Comparative biology of american black widow spiders. Trans. San Diego Soc.
Nat. Hist., 16:33-82.

Lecaillon, A. 1904. Sur la biologie et la psychologie d’une araignee {Chiracanthium carnifex Fab.).
Annee Psychol., 10:63-83.

Mansour, F, D. Rosen and A. Shulov. 1980. Biology of the spider Chiracanthium mildei (Arachnida:
Clubionidae). Entomophaga, 25:237-248.

Peck, W. B. and W. H. Whitcomb. 1970. Studies on the biology of a spider, Chiracanthium inclusum
(Hentz). Bull. Arkansas Agr. Expt. St., 753:1-76.

Schick, R. X. 1972. The early instars, larval feeding and the significance of larval feeding in the spider
genus ( Araneida: Thomisidae). Notes Arachnol. Soc. Southwest, 3:12-19.

Turnbull, A. L. 1973. Ecology of the true spiders ( Araneomorphae). Ann. Rev. Entomol., 18:305-348.
Vachon, M. 1957. Contribution a I’etude du developpement postembryonaire des araignees. lere note:

generalites et nomenclature des stades. Bull. Soc. Zool. France, 82(5-6):337-354.

Valerio, C. E. 1974. Feeding on eggs by spiderlings of Achaearanea tepidariorum (Araneae:
Theridiidae), and the significance of the quiescent instar in spiders. J. Arachnol., 2:57-63.

Manuscript received March 1984, revised August 1984.

■^... J •

•V;l^
W ■•

IT;

^7^ . ■€ ■*■ '1

■* /

Mifs

?• ♦;>?>■ ♦*>♦13^^''.; 'p 1

• ' ' '• ' ^ ■'Jp^w'’^ - " SUwjBBBk • ','’ , • ■ *-<i • ■' •• *’'• .1

Aib'-sisC

\j:‘^

. I •- ■■

‘I- ^. } *)f^i

^L:<p.->.-.'fH'C^^'4:i

Jt>-1

,-, 1, .'V *..-P i^>-\ . :'f'^f'3«f;y^ ■

■■ I r-rr ■ '■i' • i.Vk tV

)• L\:i' r^lF.tji

'tii 'y* ^ ,' > i I y i ' 1 . <;,P .i*.’i.

*'.('■- /..f . V'W'

i

:\iU

i •' ' #.

3W^4r*iIST< '• .* ' -3

, 4I' ■^''^ it* i^' pB^UvtA,;^ ■

r :V« V ; VC ' -^4 .1.

./ ••’;,'i--,vr.,;. •„ - -rj^wp^^ .V, ,;'0. -il^fi/ 73iUM Ai^'W 'a

l„ 1 • ..^-^,||C . ' ’• ■^■'‘•'■■'■i.(i.i-...i‘ ,S

..t.

^■tif

.. ■. J>'.r «,...» ;%5m iu9/<r M ,

**' -*^.'''' '*! ’*•’ 'I**’’’'* ’*(Q »'*i>.4c<|B ,

.* : '.r ,.v. ■ . . i,ik* Ma. i 1 Lil /■ i

' ■ ' ! ’*•’ 'I**’’’'* ''Q ‘.Aiel^Vi

i--' ' 1

mi j - • ' .V ‘ 1 .w»f

>t*t'

< ” t/.: '»

saif.

= liV -.'■•I

r*>

;.ti ^ji»l‘’ Ui. v»’ >(^ h^} «PW<I i-T^j vbi i

!■ .'-.ip;. iV*. J .. rP.,^ JAU^M,k->'A.t>‘V-^ W

'; -.■■■'.a.* ^ ■■ •’raijL . ■ . -y;' ■5ir*i ‘‘^V..*

■3.

*> '.' ♦'‘4„’/:

-.Nvf - ■ i’

i ‘‘'"'a •..« , ;■ ' • '^<

. A:^’3|(iisari~.3llfe -. '■■ ■ ,b r .

’ ' *.

’^' fcu

iM *i»,) ,- ’*

ti^^-

•VpVt*'jAi,.I.',y, i'' ..;•,

Greenstone, M. H., C. E. Morgan and A.-L. Hultsch. 1985. Ballooning methodology; Equations for
estimating masses of sticky-trapped spiders. J. Arachnol., 13:225-230.

BALLOONING METHODOLOGY: EQUATIONS FOR
ESTIMATING MASSES OF STICKY-TRAPPED SPIDERS

Matthew H. Greenstone, Clyde E. Morgan, and Anne-Lise Hultsch

U. S. Department of Agriculture
Biological Control of Insects Research Laboratory
P. O. Box 7629, Research Park
Columbia, Missouri 65205

ABSTRACT

Most empirical studies of spider ballooning use sticky traps to sample the aeronaut fauna. Once
the animals are removed from the adhesive and passed through solvents into preservative, biologically
meaningful masses cannot be determined directly. We use simple linear regressions to describe
relationships between live masses of wild-caught animals and a volume estimate which treats the
spider as a cylindrical solid with diameter equal to the mean of greatest carapace and abdomen width
and height equal to total length. Of regressions for six families studied in detail, the slope for
tetragnathids differs significantly from those for all but one of the other five. Pair-wise comparisons
of slopes for the five non-tetragnathid families show no statistically significant differences. We
therefore present two linear regression equations, one for tetragnathids and other similar-shaped
spiders, and the other for “typical” (all other) spiders. Limited data from one species each of pisaurid
and linyphiid are statistically indistinguishable from the “typical” regression but highly significantly
different from the tetragnathid regression, lending added support to a dichotomy in shape between
tetragnathids and other spiders. Simple linear regressions of mass on the volume estimate tend to
explain more of the variation in mass than traditional power functions of mass on measurements of
single body dimensions.

INTRODUCTION

Ballooning (aerial dispersal) in spiders is a dramatic and widespread
phenomenon. Although a fair amount is known about the taxonomic
composition, seasonal occurrence, and reproductive maturity of aeronauts,
nothing has been published on their mass frequency distributions. The masses of
ballooning spiders are of fundamental interest since one wonders how massive a
flightless animal can be and still be passively carried by the wind. The masses
of ballooners also have profound biogeographic implications, since, within a
species, larger (more massive) animals will have higher reproductive value and
hence higher colonizing potential (MacArthur and Wilson 1967). By the same
token certain species may have very low colonizing potential because they are
larger as adults and can only balloon at very early stages when reproductive value
is low.

226

THE JOURNAL OF ARACHNOLOGY

The most desirable way to sample the aeronaut fauna would be with
continuously operating suction traps (Taylor 1974), which can be calibrated to
give absolute estimates of aerial density. However they are expensive and require
the availability of electricity. This has led most workers to use sticky traps of one
form or another (Duffey 1956, van Wingerden and Vugts 1974, Yeargan 1975).
Of course after a spider has been removed from the trap and placed in one or
more solvents to remove the adhesive and preserve it, direct measurements of
mass are meaningless. The present study was undertaken to develop methods to
estimate the live masses of spiders collected from sticky traps.

Live spiders were collected from late may to late July in a variety of habitats
in North Central Missouri, viz., the Tucker Preserve (native tall grass prairie).
The University of Missouri Ashland Wildlife Area (mixed hardwood forest
understory and woodland clearings), The University of Missouri, Columbia
campus (ornamental shrubs), the U.S. Department of Agriculture, Biological
Control of Insects Research Laboratory study plot (alfalfa), and the University
of Missouri South Farms (soybeans). With the exception of Tucker Prairie, which
is in Callaway County, all localities are in Boone County. Spiders were separated
from vegetation by sweeping, beating or aspiration, dumped in a sleeve cage or
on a drop cloth for sorting and placed in individual shell vials. The live spiders
were transported in a styrofoam chest to the laboratory where they were weighted
to the nearest 0.1 mg in a tared gelatin capsule on a Mettler AE 160 electronic
balance. After weighing they were killed and preserved in 70% ethanol.

A few days later the following three measurements were made on each animal
to the nearest 0.05 mm, using a wild M5 stereo microscope with ocular
micrometer at 120 x magnification: total length, greatest width of cephalothorax,
and greatest width of abdomen. An estimate of each spider’s volume was
obtained by treating it as a cylindrical solid having radius equal to half the mean
of the two width measurements and height equal to total length, i.e.

Volume = (Total length) tt [(Width Abd + Width Ceph)/4]2

Assuming further that the density of material is homogeneous throughout a
spider’s body and fixed regardless of absolute mass, we expected to find a linear
relationship between the measured mass and estimated volume.

Because spiders are not perfectly cylindrical we expected families to show
different mass-volume relationships. In particular we expected to find significant
differences among the tetragnathids, which are most nearly cylindrical, the
thomisids (in the broad sense, including philodromids) which tend to be flattened
dorsoventrally, and most other families, whose members can be idealized as two
more or less overlapping subspherical pieces. Only animals less than 20 mg in
mass were studied because more massive animal were not expected to be common
ballooners.

RESULTS AND DISCUSSION

The raw data for eight families are presented in Fig. 1. Sample sizes and least
squares regression parameters are presented in Table 1. All are highly significant
(p < 0.01). The data were subjected to Bartlett’s Test for Homogeneity of

8

4

0

20

16

12

8

4

0

ig. I.

com

, MORGAN AND HULTSCH— SPIDER MASS ESTIMATION

227

flaw data and regressions for mass-volume relationships: A, Tetragnathidae (dashed line)
ed data for the remaining five families (solid line); B, Raw data for the Lyniphiidae and
lines in Fig. 1 A).

228

THE JOURNAL OF ARACHNOLOGY

Table I. — Individual regression parameters. V = linear regression of mass on volume, C = power
curve of mass versus cephalothroax width, TL = power curve of mass versus total length, N = sample
size, a = regression intercept, b = regression slope, R^ = coefficient of determination.

Family

N

av

bv

R'v

R'c

R^tl

Lycosidae

6

1.69

1.14

0.99

0.30

0.99

Salticide

37

-0.01

1.14

0.97

0.91

0.76

Araneidae

36

0.55

1.02

0.97

0.84

0.89

Oxyopidae

7

1.14

1.17

0.97

0.16

0.58

Tetragnathide

8

0.31

1.61

0.98

0.74

0.90

Thomisidae

15

-0.01

1.05

0.92

0.92

0.97

Pisauride

8

-0.33

1.57

0.97

0.75

0.82

Linyphiide

28

0.26

1.20

0.94

0.52

0.92

Variance (Sokal and Rohlf 1969) and found to be significantly heterogeneous (X?
= 62.159, p < 0.001) prohibiting overall parametric analysis. Inspection of Fig.
1 suggests that this result was due to the smaller range of variation exhibited by
the pisaurids (comprising eight Pisaurina spiderling) and the linyphiids
(comprising only individuals of the single species Frontinella pyramitela). When
only the remaining six families data are subjected to Bartlett’s Test there is no
evidence of significant heterogeneity of variances (X5 = 3.848, p > 0.5),
permitting an analysis of Covariance (Snedecor and Cochran 1967) to test the
hypothesis of no differences among mass-volume regressions. This revealed highly
significant differences among slopes (F5, 97 = 3.339, p < 0.01). In order to
determine where these differences lie, all possible pair-wise t-tests on slopes were
performed. The method of Bonferroni (Morrison 1983) was used to derive the
critical value for an overall error rate error less than 0.05. Results of these tests
are presented in Table 2. As expected the tetragnathids differ significantly from
most of the other families (the sole exception was the Oxyopidae, with 0.05 <
p < 0.10). However the thomisids do not differ significantly from the other
families, owing perhaps to their higher variance (See Fig. 1 and the R ’s in Table
1), i.e. to their failure to conform as well to the assumptions of the mass-volume
model. Given these data, we suggest that workers wishing to estimate the masses
of preserved spiders may do so by using the cylindrical approximation to volume
as the independent variable for one of two regressions. If the animal is a
tetragnathid, use:

Mass(in mg) = 1.61 (volume in mm^) + 0.31

(see Table 1 and Fig lA)

(We predict that other patently cylindrical spiders, such as Tibellus and Larinia
spp. which were not collected in this study, will also fit this model). The following

Table 2. — t values and levels of significance for pair-wise t-tests on slopes. Levels of significance are
coded as n.s. (>0.1) and * (<0.05). See text for explanation.

Salticidae

Araneidae

Oxyopidae

Tetragnathide

Thomisidae

Lycosidae

—0.034, n.s.

1.444, n.s.

0.292, n.s.

-3.760, *

0.544, n.s.

Salticidae

2.615, n.s.

—0.244, n.s.

4.363, *

1.213, n.s.

Araneidae

—

—

-1.542, n.s.

6.710, *

—0.428, n.s.

Oxyopidae

—

—

—

—3.014, n.s.

0.6541 , n.s.

Tetragnathidae

Thomisidae

—

—

—

—

3.450, *

GREENSTONE, MORGAN AND HULTSCH— SPIDER MASS ESTIMATION

229

overall regression for all other families, which could be considered as having
“typical” shape, was derived by combining the data from the non-tetragnathid
families in Fig. lA:

Mass (in mg) =1.12 (volume in mm^) + 0.23

The utility of these equations, and particularly the assumption that all spiders
besides tetragnathids and their look-alikes are essentially the same shape and
homogeneous with respect to density, will be determined with time. As a partial
test we offer the “goodness of fit” of the pisaurid and linyphiid data, which have
lower variance than the other six families but show an obvious linear relationship
between mass and volume (Fig. IB), to the two regressions. Lines of least squares
are fit such that half the points lie above and half lie below the line. If pisaurids
and linyphiids are shaped like “typical” spiders, then the proportion of points
lying above (and below) the “typical” regression line should not differ
significantly from 0.5, whereas the proportions of points and below the
tetragnathid line should differ significantly from 0.5. The numbers of points on
each side of both lines for both regressions in each family are given in Table 3.
The proportions of points above or below the “typical” regression line do not
differ significantly form 0.5 using the Binomial Test (Siegel 1956) (p > 0.29 and
p > 0.34, pisaurids and linyphiids, respectively) but are highly significantly
different from 0.5 for the tetragnathid regression (p < 0.008 and p < 0.0001,
respectively), supporting our contention that the linyphiids and pisaurids have a
“typical” spider shape different from that of tetragnathids.

The most commonly used index of size for spiders has been cephalothorax
width. It is a logical index of instar, since as a single sclerotized plate the
carapace is not apt to change its dimensions very much during the instar (see
Hagstrum 1971 and references cited therein). On the other hand it cannot be
expected to be a good indicator of mass, which can vary widely within an instar
due to nutritional and reproductive state. Breymeyer (1967) used simple linear
regression and Sage (1982) used step-wise linear regression to derive a
relationship between mass and total length. Rogers et al. (1977) found that a
power function gave the best fit for the total length vs. mass relationship for
spiders. In Table 1 we present the coefficients of determination for the simple
linear regressions of mass on volume (V) and for the power curves of mass on
cephalothorax width (C) and mass on total length (TL), for each family. The
simple linear regression of mass on volume is clearly a better description of the
relationship than either power curve in most cases. An alternative approach is to
take the oven-dried weight of the spider after removal from the trap and apply
a conversion factor (S. E. Riechert, pers. commun.). This is less time-consuming
but not useful if one wishes to save the specimen for identification.

Table 3. — Numbers of pisaurid and linyphiid points lying above (+) and below (— ) calculated overall
and tetragnathid regression lines.

Overall Regression

Tetragnathid Regression

+ -

+

Pisauridae

2 6

0 8

Lmyphiidae

16 12

I 27

230

THE JOURNAL OF ARACHNOLOGY

As Table I shows, an estimate of volume will tend to be a more accurate
predictor of mass than a single linear measurement. The method presented here
is tedious but accurate. A logical extension of our technique would be to
determine the frontal sectional area of the image of the spider by projecting the
surface of a digitizing tablet into the microscope field with a drawing tube. This
area estimate could then be rotated by an appropriate algorithm in a
microcomputer to produce a spider-shaped, rather than a cylindrical, estimate of
the volume.

ACKNOWLEDGMENTS

We thank G. F. Krause and N. L. Marston for assistance with Analysis of
Covariance, and A. Kronk, N. L. Marston, S. E. Riechert, G. W. Uetz, and J.
C. Weaver for valuable comments on the manuscript. A.-L. H. was supported by
the U.S.D.A.-A.R.S. Research Apprenticeship Program.

LITERATURE CITED

Breymeyer, A. 1967. Preliminary data for estimating the biological production of wandering spiders.
Pp. 821-834, In Secondary Productivity of Terrestrial Ecosystems (K.. Petrusewicz, ed.) Institude of
Ecology, Polish Academy of Sciences, Warsaw.

Duffey, E. 1956. Aerial dispersal in a known spider population. J. Anim. Ecol., 25: 85-1 1 1.

Hagstrum, D. W. 1971. Carapace width as a tool for evaluating the rate of development of spiders
in the laboratory and field. Ann. Entomol. Soc. Amer., 64: 757-760.

MacArthur, R. H. and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University
Press, Princeton. 203 pp.

Morrison, D. F. 1983. Applied Linear Statistical Methods. Prentice Hall, Inc., Englewood Cliffs. 562

pp.

Rogers, L. E., R. L. Buchsbomb, and C. R. Watson. 1977. Length-weight relationships of shrub-
steppe invertebrates. Ann. Entomol. Soc. Amer., 70(1): 51-53.

Sage, R. D. 1982. Wet and dry-weight estimate of insects and spiders based on length. Amer. Midi.
Nat., 108: 407-411.

Siegel, S. 1956. Nonparametric Statistics. McGraw Hill, New York. 312 pp.

Snedecor, G. W. and W. G. Cochran. 1967. Statistical Methods. Iowa State University Press, Ames.
593 pp.

Sokal, R. R. and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Fransisco. 776 pp.

Taylor, L. R. 1974. Insect migration, flight periodicity and the boundary layer. J. Anim. Ecol., 43:
225-238.

van Wingerden, W. K. R. E. and H. F. Vugts. 1974. Factors influencing aeronautic behavior of
spiders. Bull. British Arachnol. Soc., 3: 6-10.

Yeargan., K. V. 1975. Factors influencing the aerial dispersal of spiders (Arachnida: Araneida). J.
Kansas Entomol. Soc., 48: 403-408.

Manuscript received May 1984, revised September 1984.

Gilbert, C. and L. S. Ray or, 1985. Predatory behavior of spitting spiders (Araneae, Scytodidae) and
the evolution of prey wrapping. J. Arachnol., 13:231-241.

PREDATORY BEHAVIOR OF SPITTING SPIDERS (ARANEAE:
SCYTODIDAE) AND THE EVOLUTION OF PREY WRAPPING^

Cole GilberU and Linda S. Rayor^

Department of Entomology^ and of Systematics and Ecology^
University of Kansas, Lawrence, Kansas 66045

ABSTRACT

The predatory behavior of the spitting spider Scytodes sp. was studied in the laboratory, and an
ethogram of the predatory behavior was developed. The principal components usually occur in the
order: tapping, spitting, biting, wrapping, feeding. Spitting results in a pair of sticky, zig-zag,
transverse bands which pin the prey to the substrate. At the capture site scytodids wrap the prey using
the typical form seen in the “higher” spiders: the spider holds the prey in both third legs and
alternates the use of right and left fourth legs in applying silk. Prey are eaten at the capture site.

A comparison of prey wrapping by spiders in primitive aerial-web building species with that used
by typically “vagrant” species which forage on elevated substrates shows two very different forms of
prey wrapping. We argue that prey wrapping at the capture site is an early adaptation of spider
radiation into the aerial niche based on the presence of one form or the other in most taxa foraging
above ground. Further, the extreme similarity of form of prey wrapping in “higher” spiders which
build aerial webs is indicative of a stronger selective pressure for efficient prey handling than for
actual prey capture behavior or web geometry.

INTRODUCTION

Spiders of the genus Scytodes (Latreille) have the curious behavior of ejecting
a mucilagenous glue from the chelicerae at their prey during attack (Monterosso
1927, 1928, Kovoor and Zylberberg 1972) or at predators in self-defense
(McAlister 1960, Gilbert and Rayor 1983). Although these spiders are considered
primitive on the basis of web structure and several morphological characters
(Lehtinen 1967), several aspects of their use of silk during predatory behavior are
analogous with those of aerial-web building spider taxa with more advanced
characteristics. We examined these behaviors in an undescribed species of Scytodes
in order to make inferences about the evolutionary stages in the transition from
ground-dwelling to aerial-web weaving.

The evolution of web-building spiders from primitive vagrant ancestors to
species using silken aerial webs has been the focus of many studies (Kaston 1964,

'Contribution no. 1895 from the Department of Entomology, University of Kansas, Lawrence,
Kansas.

232

THE JOURNAL OF ARACHNOLOGY

1966, Robinson 1975). It is generally agreed that early aerial webs were derived
from an accumulation of draglines around the spider’s resting place or retreat.
Selective pressures for more efficient prey capture favored the construction of
more elaborate, structured, and sticky webs.

As spiders radiated into the aerial niche they faced different selective pressures
on the mechanics of prey capture. Prey immobilized on an aerial web can fall
to the ground and be lost if the spider loses contact with it. Most aerial spiders
wrap their prey during predation. Wrapping serves, among other functions, to
sequester the prey and frees a spider from the necessity of eating its prey
immediately. A further evolution in the use of prey wrapping has moved this
behavior from late to earlier in an attack sequence. In situations when struggling
prey could injure a spider as it approaches to bite the prey or when its struggles
could allow it to escape quickly, many spiders first wrap the prey in silk to
immobilize it, then approach and bite (Robinson and Robinson 1976). This
method of prey immobilization is obligatory in cribellate orb-weavers of the
family Uloboridae in which the poison glands are absent (Marples 1962).

Because Scytodes possesses several primitive morphological characteristics, yet
builds an aerial web, we consider that our observations on its predatory
contribute insight into possible stages in the transition from living on the ground
to aerial habits.

MATERIALS AND METHODS

Spiders of a currently undescribed species in the genus Scytodes were collected
from under picnic shelter eaves and around stones at two Texas localities: Lake
Corpus Christi, San Patricio Co., and Tyler State Park, Smith Co. Voucher
specimens are deposited in the American Museum of Natural History and Museo
de Zoologia, Universidad de Costa Rica. The spiders were transported to
Lawrence, Kansas, and housed individually in clear plastic boxes, 11 x 11 x 3
cm high, each supplied with a cotton-stoppered vial of water. The temperature
was 25° C and the light cycle was irregular, but approximately L:D, 14:10. Prey
were principally vestigial-winged fruit flies. Drosophila melanogaster (Meig.).
However, cockroach nymphs (Blattaria), lacewings (Neuroptera), and small moths
(Lepidoptera) were occasionally presented as well, A second generation box
design employed a cork-stoppered hole through which prey were delivered singly
or in groups of three to ten individuals. Subsequent predatory behavior was
directly observed with (usually) two observers reporting their observations into a
tape recorder. The tapes were later transcribed and the data analyzed as detailed
below. After the conclusion of the sequence, the pattern of spit was observed
through a binocular microscope equipped with an ocular micrometer, then
sketched.

From observations of more than 70 predation attempts on a variety of prey
by 15 spiders (males, females, and late juvenile instars) we developed an ethogram
of the components of predatory behavior. Complete predation sequences (N =
31, 8 individuals) were then described in terms of the defined behavioral
components. Durations were not recorded. The principal behaviors were serially
ordered for analysis and a particular component could not follow itself in

GILBERT AND RAYOR^SCYTODES PREDATORY BEHAVIOR

233

successive acts in a sequence. The components were then put into a first-order
transition matrix and transition frequencies calculated. This procedure was
performed for each capture sequence of each spider. Individual results were
compared and no major differences were observed between the sexes or instars.
Data from all sequences were summed into one transition matrix and displayed
graphically as a flow diagram.

RESULTS

Twelve behaviors comprise the ethogram of predatory behavior for Scytodes
sp.

Alert posture. — Structures used: Entire body. Action: Space-filling posture with
spider “up” on its legs. Much extension at all joints. Context: When walking
about or just after prey contacts web lines.

Retracted posture. — Structures used: Entire body. Action: Spider appears
dorsoventrally compressed in one plane. All the legs are held at the sides of the
body in typical latigrade position, i.e. the femora are directed posteriorly and the
more distal segments directed anteriorly. Context: Diurnal resting or defensive
posture.

Tap. — Structures used: Legs I (or II). Action: Leg is extended, metatarsus and
tarsus then flex and re-extend. Contralateral legs alternate in tapping. Context:
Initial localization of prey after it has touched the web or spider.

Spit . — Structures used: Chelicerae. Action: Spider is in slightly elevated posture
by leg extension. Spitting is accompanied by a convulsive shudder and slight
posterior movement of the body. Multiple spits occasionally occur. Context:
After prey has been contacted and is roughly positioned between extended legs
I and the spider’s body.

Reach and Roll (RR). — Structures used: Legs I (or II). Action: Leg is
extended, then tibia, metatarsus, and tarsus are flexed and slightly retracted, then
elevated and re-extended. The movement describes a circular motion with
contralateral legs moving in unison. Context: Immediately after spitting, RR is
performed distal or lateral to prey and serves to entangle it in the drying spit.
It may also help to localize the prey.

Bite. — Structures used: Chelicerae and pedipalpi. Action: Pedipalpi are
extended as in PALP EX. As they palpate the prey and a suitable surface is
found (e.g. an appendage) the spider leans forward and bites the prey with the
chelicerae. Context: Occurs after spitting and as the prey is cut from the spit or
during wrapping.

Nibbling . — Structures used: Chelicerae, pedipalpi, and legs I (or II). Action:
Distal leg segments (metatarsi or tarsi) are brought to the mouth and held there
by pedipalpi. Leg is pulled out dorsally as the chelicerae nibble proximo-distally.
Context: After RR the same legs are nibbled as were used in RR. This behavior
appears homologous with nibbling seen during grooming.

Pedipalp Reciprocal Scraping (RECIP). — Structures used: Pedipalpi and
chelicerae. Action: Pedipalpi scrape ipsi- or contra- lateral chelicera then scrape
against one another. Context: Often follows nibbling and cleans dried spit from
the chelicerae. It also occurs during grooming.

234

THE JOURNAL OF ARACHNOLOGY

Pedipalp extension (PALP EX). — Structures used: Pedipalpi, legs I (or II), and
chelicerae. Action: Both pedipalpi are extended with slight lateral oscillation
toward a thread (dried spit or silk) held by a leg. A single pedipalp pulls the
thread to the chelicerae which cut it. Context: Freeing the prey from the substrate
by cutting the dried spit around it.

Hind leg wrap (HLW) . — Structures used: Legs IV. Action: Legs alternate
wiping spinnerets. Each wipe is accompanied by a lateral movement of the
abdomen toward the leg which pulls silk from the spinnerets and places it around
the prey which is held by legs III. Context: After partially or completely freeing
the prey from the substrate, the prey is bound into a silk package.

Dragline attachment (DGL). — Struetures used: Spinnerets and legs IV. Action:
Abdomen flexed toward the surface to which attachment disc is applied. When
the surface is a web line it is pulled to the spinnerets with either leg IV. Context:
1. Disc is applied to substrate or web line upon initial contact of prey to spider
or web. 2. During wrapping, especially near completion, discs are applied to the
prey package, the substrate, web lines, or several of these structures.

Feed. — Structures used: Chelicerae and pedipalpi. Action: Prey is bitten with
the chelicerae and held against the mouth with the pedipalpi. Context: At the
conclusion of the predation sequence.

Typical prey capture is similar to that reported by Monterosso (1927, 1928) for
Scytodes thoracica (Latr.). When prey first contacts the web lines or spider’s legs,
the spider assumes an alert posture and usually fastens its dragline to the
substrate with its abdomen or to a web thread using its abdomen occasionally
aided by a leg IV. The spider then orients toward the prey and approaches it
slowly, tapping with legs I, occasionally touching the prey. When the prey is
approximately centered between the forelegs and the spider, it spits a net of glue
at the prey (Fig. 1). The spider steps quickly to the prey; we seldom observed
the leisurely saunter reported by Gertsch (1979:222). As it approaches the prey,
the spider uses legs I and sometimes II in the RR motion which further entangles
the prey in the rapidly drying glue. The spider infrequently spits a second time.
As the spit dries on the immobilized prey, the spider palpates and bites the prey,
usually on an appendage. At this point the spider alternately nibbles legs I and
II with the chelicerae. This behavior is similar to nibbling of the legs observed
in grooming. After nibbling, the spider begins to free the immobilized prey from
the net of spit. If the prey is not securely fastened, the spider simply bites it with
the chelicerae and pushes down on the substrate with all eight legs, thus pulling
the prey free from the net of spit. With more securely fastened prey the spider
cuts through the securing threads around the prey by drawing them to its
chelicerae using the pedipalps (PALP EX) and legs I and 11. Once prey is freed
or has only one side attached to the substrate the spider begins to wrap the prey.

The form of scytodid hind-leg wrapping is analogous to that of araneids,
theridiids, and other spiders (e.g., Robinson 1975, Eberhard 1982, for further
references see DISCUSSION). The prey is held with the short legs III, while legs
IV alternate in distributing loops of silk, stripped from the spinnerets, over the
prey. Occasionally one leg wraps several (3-5) loops over the prey before the
opposite leg is used. During wrapping the abdominal apex repeatedly waggles
from side-to-side toward the leg which will apply the next loop of silk. In
Scytodes the strands of silk are fine and we could not determine whether strands

GILBERT AND RAYOR—SCYTODES PREDATORY BEHAVIOR

235

Fig. I. — Dorsal view of Scytodes sp. to show the orientation of the spider and its spit immediately
after spitting. Prey has been omitted for clarity.

were composed of multiple fibers. Nor could we determine from which spinnerets
the silk was pulled. The strands are definitely not the swathing bands seen in
Argiope Audouin and some other araneid genera (Robinson, Mirick and Turner
1969). The spider punctuates its wrapping by placing dragline attachment discs
on the prey, the substrate, or the threads which the spider is contacting. Finally
the spider holds the trussed prey in its chelicerae and begins to feed. Occasionally
the spider carries the prey a short distance, but feeding generally occurs at the
capture site. It is possible that the observed feeding at the capture site was an
artifact of the small cages. One of us (LSR) has observed predation by Scytodes
longipes Lucas in Costa Rica where it is found in association with human
dwellings. Scytodes longipes typically bites its prey (in one case, a wolf spider
twice its body size) and wraps it at the capture site, then carries it to its retreat
before feeding.

Localization of the prey by tapping occurred prior to spitting (Fig. 2). Vision
does not appear to play a role in prey localization. Prey were almost always
bitten at least once before being wrapped. If prey continued to struggle, biting
occurred intermittently with wrapping. A long period of feeding terminated a
sequence. The sequence is not stereotyped even for the same individual capturing
the same type of prey. There is variability in the number of acts per capture and
in the components used. The mean number of acts per capture sequence is 19.5;
the range in this study is 3 - 97. The lower bound is close to the minimum given
the resolution of our component descriptions. A shorter sequence could be BITE-
FEED, but this was never observed even for prey much smaller than the spider.

236

THE JOURNAL OF ARACHNOLOGY

Fig. 2. Predatory sequence of Scytodes sp. Main components are represented by circles; numbers
within the circles are the number of occurrences of the behavior in the data set analyzed. The numbers
beside the arrows and the width of the arrows represent the percentage of transitions in which a
behavior was followed by the next. Transition percentages smaller than 4% have been omitted for
clarity.

The upper limit may depend entirely upon the prey item. The 97 acts were used
to subdue a pyralid moth 30% longer than the spider. In nature it is possible that
even relatively larger prey are caught and require more acts to subdue.

Another source of variability in the predatory behavior is in the component
composition of the sequence. Obviously, all the behavioral components listed in
the ethogram cannot be present in a sequence of fewer acts. However, even
several of the longer capture sequences did not display all behavioral components
listed. We could not discern any systematic relationship between prey size and the
extent of or position of wrapping in the sequence, although few large prey items
were offered.

DISCUSSION

Prey capture by Scytodes. — Though our report of predation in this species of
Scytodes agrees with previous general accounts, it differs significantly in one
respect from the thorough study of Scytodes thoracica by Dabelow (1958) (see
also Schaller 1956 and Kaestner 1963). The net of spit observed in 5’. thoracica
reportedly consists of a single block oriented so that the parallel bands are
parallel to the spiders’s longitudinal axis. In the Scytodes sp. used in the present
study, although the net itself is similar to that of S. thoracica, the parallel bands
are perpendicular to the spider’s longitudinal axis (Fig. 2). Further, the net seems
always to consist of two discrete, yet often overlapping blocks of spit each
composed of 5 - 17 parallel bands (mean = 10, N = 8). Typically the left and
right blocks are composed of unequal numbers of bands. Similar orientations
have been observed in N. intricata Banks (McAlister 1960), S. venusta (Thorell),
S. longipes (= marmorata L.K.), and S. fusca Walkn. (= domestica Dol.)
(Bristowe 1931).

GILBERT AND RAYOR^SCYTODES PREDATORY BEHAVIOR

237

Two other genera have recently been included in the family Scytodidae:
Drymusa Simon and Loxosceles Heinecken and Lowe (Gertsch 1967, but see
Gertsch and Ennik 1983). Drymusa dinora Valerio has not been observed to use
spit during prey capture (Valerio 1974). Loxosceles has perhaps been observed to
spit, for Kaestner (1963:579) says, "^Loxosceles Lowe, greift aus viel geringerem
Abstand als Scytodes durch Speien von Leim aus den Cheliceren an und erzeugt
dabei viel weniger regelmassig angeordnete sehr zarte Faden.” However numerous
other authors (Hite et al. 1966, Kaston 1972, Gertsch 1979, Foelix 1982) did not
report spitting in Loxosceles, and this agrees with our own observations of L.
reclusa Gertsch and Muliak and Loxosceles sp. made with the help of a binocular
microscope.

Although the actual components of predation differ among taxa, the form of
these behaviors in Scytodes sp. are analogous to those reported in predation of
many other labidognath spiders with one exception, the component termed reach-
and-roll. This maneuver, performed with both legs I and occasionally legs H
serves to entangle the prey in the drying spit. Further, the spider may be applying
additional fine strands of spit with its legs as it alternates between reach-and-roll
and nibbling during the pre-bite portion of the sequence.

Comparative prey wrapping. — Given the present understanding of the use of
silk in the predatory behavior of Scytodes, the remainder of the discussion will
focus on the evolution of prey wrapping as an adaptation accompanying the use
of aerial webs. Primitive ground-dwelling spiders, such as liphistiomorphs, do not
build trapping snares and do not wrap their prey (Bristowe 1976), whereas spiders
in more derived groups with aerial snares or elevated cursorial habits do exhibit
prey wrapping. This difference leads to the inference that prey wrapping is an
adaptive response to the increased chance of losing contact with prey in an aerial
habitat.

Descriptions of the predatory behavior of “vagrant” spiders which capture prey
in an elevated setting support this interpretation (Table 1). In a study of spiders
of four species in the family Lycosidae, Greenquist and Rovner (1976) reported
that individuals of the ground-dwelling genus Schizocosa Chamberlin never
wrapped their prey. On the other hand, individuals of Lycosa punctulata Hentz
and L. rabida Walckenaer which spend significantly more time foraging on
elevated foliage exhibit post-immobilization prey wrapping at the capture site.
The spider immobilizes the prey with a bite then circles it and applies silk directly
from the spinnerets. We designate this as the ‘primitive’ prey wrapping form. This
method is equally efficient on the ground or on elevated substrates. In our view,
it represents a form of prey wrapping that is adaptive for above ground prey
retention, not just specifically adapted to use with aerial webs. Similar forms of
circular prey wrapping, occasionally involving the application of silk by legs IV,
have been reported for at least one species in the following, primarily vagrant,
families: Theraphosidae (M. Teeter pers. comm.), Lycosidae (Rovner and Knost
1974, Greenquist and Rovner 1976) Gnaphosidae and Hersiliidae (Bristowe 1930),
Uroctiidae (Crome 1937), Oecobiidae (Glatz 1967), Psechridae (Robinson and
Lubin 1979), Theridiidae (Carico 1978), and Ctenidae (Melchers 1967).

Spiders which employ aerial capture webs tend to use a second form of prey
wrapping which we designate as ‘derived.’ The spider hangs from its web and
using legs IV applies silk to the prey which may be in contact with the spider,
the web, or both. This method has been reported for species in the families:

238

THE JOURNAL OF ARACHNOLOGY

Table 1. — Summary of capture and wrapping elements of predatory behavior of spider taxa at
different stages in the evolution of prey wrapping. See text for details of elements.

Taxon

Web

Structure

Method of

Immobilization

Wrapping

Location Form

Feeding

Location

Source

Hypochilus
ge rise hi
( Hypochilidae)

Aerial

inverted

funnel

Bite

No Wrapping

Capture

site

Shear

1969

Lycosa
rah id a

L. punctulata
(Lycosidae)

Elevated
capture
(no web)

Bite

Capture

site

Primitive

Capture

site

Greenquist

and

Rovner

1976

Fecenia

a ng us tat a
(Psechridae)

Aerial

inverted

funnel/

planar

Bite

Capture

site

Primitive

Retreat

Robinson

and Lubin

1979

Diguetia

alholineata

(Diguetidae)

Aerial

inverted

funnel

Bite

Retreat

Unique

Retreat

Eberhard

1974

Drymusa

dinora

(Scytodidae)

Aerial

space-

filling

Bite

Capture

site

Primitive

Removed

from

capture

site

Valerio

1974

Scytodes

sp.

(Scytodidae)

Aerial

space-

filling

Bite

Capture

site

Derived

Capture

site

Present

study

Modisimus

spp.

(Pholcidae)

Aerial

space-

filling

Wrap

Capture

site

Derived

Retreat

Eberhard

and

Briceno

1976

Diguetidae and Linyphiidae (Eberhard 1967), Theridiidae (Kaston 1965),
Araneidae (Robinson 1975), Theridiosomatidae and Uloboridae (Eberhard 1982),
Pholcidae (Eberhard and Briceno 1983), and Scyotdidae (present study).

There is relatively little variation in the derived form of prey wrapping
employed by morphologically and phylogenetically diverse taxa. This is in
contrast to the enormous variation in predatory technique and to the great
variation in web structure, from primitively unstructured to highly derived and
secondarily reduced, in these same taxa. This similarity of form of prey wrapping,
though convergent, leads us to speculate that once a taxon moves into the aerial
niche there is a greater evolutionary premium associated with efficient prey
handling than is associated with the actual method of prey capture or details of
web structure. Thus selection has not only favored some form of prey wrapping
by spiders when they capture prey above ground, but selection has tended to
channelize the form of this behavior in those species well adapted to prey capture
in an aerial habitat.

Examination of primitive aerial-web weavers may reflect the early steps in the
evolution of prey wrapping. Predatory behavior has been described for several
taxa (Table 1) in the group Filistatides (sensu Lehtinen 1967) which includes the

GILBERT AND RAYOR-SCYTODES PREDATORY BEHAVIOR

239

Hypochilomorpha and Haplogynae of other authors (Simon 1892, Petrunkevitch
1933, Platnick 1977, Brignoli 1978). The hypochilids are the most primitive taxon
in this group, sharing many characters with orthognaths (Gertsch 1958, Marples
1968). Hypochilus gertschi Hoffmann does not wrap its prey, but merely bites the
prey, pulls it through the web, and feeds at the capture site. This represents a
very early stage in the evolution of aerial webs and prey wrapping.

A slightly more advanced stage may be represented by species in the cribellate
family Psechridae. “The [phylogenetic] position of Psechridae is enigmatic.”
According to Lehtinen (1967:383) who considers them closer to his Amaurobiides
than Filistatides. The family, as delimited by Forster and Wilton (1973), includes
both terrestrial vagrant genera and aerial-web building genera such as Fecenia
and Psechrus. Fecenia augustata (Thorell) immobilizes its prey by biting, then
binds it to the web by applying silk directly from the spinnerets. After this
primitive wrapping of the prey, it is cut from the web, carried in the chelicerae
to the retreat, and eaten immediately or re-attached to the substrate.

The next stages may be represented by primitive aerial-web building haplogyne
spiders. Diguetia albolineata (O. P.-Cambridge) (Diguetidae), also bites its prey
and pulls it through the web, then transports it to the retreat where it is wrapped
and eaten. The wrapping however, has a form unique to this species (Eberhard
1967:179) and may represent an independent response to the selective pressures
favoring prey wrapping.

Drymusa dinora (Scytodidae) does not wrap small prey, but after immobilizing
larger prey with a bite, the prey is wrapped at the capture site. The form of prey
wrapping is similar to that used by Fecenia and spiders in “vagrant” taxa.
Drymusa applies silk directly from the spinnerets while moving around the prey.
The spider then carries it a short distance before feeding. The situation reported
for Scytodes is different. The form of prey wrapping it uses is the typical derived
form of hind-leg wrapping: alternate use of opposite legs IV in casting loops of
silk over the prey.

The Pholcidae is the only other family of haplogyne aerial-web building spiders
for which we are aware of detailed accounts of predatory behavior. Prey
wrapping in this group is advanced both in form and in position in the predation
sequence. Modisimus spp. wrap their prey at the capture site using the derived
from. Wrapping is the first means of prey immobilization. Prey are then bitten,
carried into the retreat, and eaten.

We agree with previous authors (e.g. Eberhard 1967, Robinson 1975, Lubin
1980) that the use of wrapping as the primary method of prey immobilization is
the most derived use of the behavior. We suggest that because of the selective
pressures on aerial prey capture, post-immobilization wrapping at the capture site
was one of the earliest stages in the evolution of prey capture in aerial webs.
Spider phylogeny is poorly understood and it is reasonable to assume that just
as aerial webs have evolved several times (Lehtinen 1967, Kullman 1972), prey
wrapping also may have evolved independently in many taxa as their members
adapted to an aerial niche.

240

THE JOURNAL OF ARACHNOLOGY

ACKNOWLEDGMENTS

We are grateful to members of R. E. Beer’s invertebrate zoology class for
collecting the spiders and to V. Roth and W. Gertsch for identifying them. We
thank Byron Leonard for enthusiastically translating Monterosso’s papers for us
and members of-C. D. Michener’s scientific writing class for comments on a
rough version of the manuscript. We especially thank W. G. Eberhard and C.
Valerio for their interest, helpful discussions, and criticisms of a later version of
the manuscript. R. Buskirk’s thoughtful comments helped clarify our Discussion.
Publication costs were generously supported by a grant from the Joseph H.
Camin Memorial Fund.

LITERATURE CITED

Brignoli, R M. 1978. Some remarks on the relationships between the Haplogynae, the
Semientelegynae, and the Cribellatae. Pp. 285-292. In Arachnology (P. Merrett, ed.). Symp. Zool.
Soc. London, 42.

Bristowe, W. S. 1930. Notes on the biology of spiders. III. Ann. Mag. Nat. Hist., ser. 10, 6:347-353.
Bristowe, W. S., 1931. Notes on the biology of spiders, IV. Ann. Mag. Nat. Hist., ser. 10, 8:469-471.
Bristowe, W. S. 1976. A contribution to the knowledge of liphistiid spiders. J. Zool. (London), 178:1-
6.

Carico, J. E. 1978. Predatory behavior in Euryopis funebris (Hentz) (Araneae: Theridiidae) and the
evolutionary significance of web reduction. Pp. 51-58. /« Arachnology (P. Merrett, ed.), Symp. Zool.
Soc. London, 42.

Crome, W. 1937. Bau und Function des Spinnapparates und Analhugels, Ernahrungsbiologie und
allgemeine Bemerkungen zur Lebenweise von Uroctea durandi (Latreille). Zool. Jahb. (Syst.),
85:571-606.

Dabelow, S. 1958. Zur Biologic der Leimschleuderspinne Scytodes thoracica (Latreille). Zool. Jahb.
(Syst.), 86:85-126.

Eberhard, W. G. 1967. Attack behavior of diguetid spiders and the origin of prey wrapping in spiders.
Psyche, 74:173-181.

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

Eberhard, W. G. and R. D. Briceno L. 1983. Chivalry in pholcid spiders. Behav. Ecol. Sociobiol.,
13:189-196.

Foelix, R. F. 1982. Biology of Spiders. Harvard University Press: Cambridge, Massachusetts. 306 pp.
Gertsch, W. J. 1958. The spider family Hypochilidae. Amer. Mus. Nov., 1912:1-28.

Gertsch, W. J. 1967. The spider genus Loxosceles in South America (Araneae: Scytodidae). Bull.
Amer. Mus. Nat. Hist., 136:117-174.

Gertsch, W. J. 1979. American Spiders (2nd ed.). Van Nostrand Reinhold: New York. 274 pp.

Gertsch, W. J. and F. Ennik. 1983. The spider genus Loxosceles in North America, Central America,
and the West Indies (Araneae: Loxoscelidae). Bull. Amer. Mus. Nat. Hist., 175:264-360.

Gilbert, C. and L. S. Rayor. 1983. First record of Mantisfly (Neuroptera: Mantispidae) parasitizing
a spitting spider (Scytodidae). J. Kansas Entomol. Soc., 56:578-580.

Glatz, L. 1967. Zur Biologic und Morphologic von Oecobius annulipes Lucas (Araneae: Oecobiidae).
Zeit. Morph. Okol. Tiere, 61:158-214.

Greenquist, E. A. and J. S. Rovner. 1976. Lycosid spiders on artificial foliage: Stratum choice,
orientation preferences, and prey-wrapping. Psyche, 83:196-209.

Forster, R. R. and C. L. Wilton. 1973. The spiders of New Zealand. IV. Otago Mus. Bull., 4:1-309.

Hite, J. M., W. J. Gladney, J. L. Lancaster, and W. H. Whitcomb. 1966. Biology of the brown recluse
spider. Univ. Arkansas Exp. Sta. Bull., 711:1-27.

Kaestner, A. 1963. Lehrbuch der Speziellen Zoologie. Teil I, Wirbellose. 2 Halbband. G. Fischer
Verlag: Stuttgart.

GILBERT AND RAYOR— ^’CTTOZ) £5 PREDATORY BEHAVIOR

241

Kaston, B. J. 1964. The evolution of spider webs. Amer. Zool., 4:191-207.

Kaston, B. J. 1965. Some little known aspects of spider behavior. Amer. Midi. Nat., 73:336-356.

Kaston, B. J. 1966. Evolution of the web. Nat. Hist., 75:26-33.

Kaston, B. J. 1972. How to know the spiders. W. C. Brown: Dubuque, Iowa. 272 pp.

Kovoor, J. and L. Zylberberg. 1972. Histologic et infrastructure de la glande chelicerienne de Scytodes
delicatula Sim. (Araneidae: Scytodidae). Ann. Sci. Nat., Zool., Paris, ser. 12, 14:333-388.

Kullman, E. J. 1972. The convergent development of orb-webs in cribellate and ecribellate spiders.
Amer. Zool., 12:395-405.

Lehtinen, P. 1967. Classification of the cribellate spiders and some allied families, with notes on the
evolution of the suborder Araneomorpha. Ann. Zool. Fenn., 4:199-468.

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

Marples, B. J. 1962. Notes on spiders of the family Uloboridae. Ann. Zool. (Agra), 4:1-1 1.

Marples, B. J. 1968. The hypochilomorph spiders. Proc. Linn. Soc. London, 1979:1 1-31.

McAlister, W. H. 1960. The spitting habit of the spider Scytodes intricata Banks (Scytodidae). Texas
J. Sci., 12:17-20.

Melchers, M. 1967. Der Beutefang von Cupiennius salei (Keyserling) (Ctenidae). Zeit. Morph. Okol.
Tiere, 58:321-346.

Monterosso, B. 1927. Osservazioni preliminari sulla del genere Scytodes (Walk.) (Araneae verae:
Sicariidae). Rend. Reale Accad. Naz. Lincei, ser. 6, 6:171-174.

Monterosso, B. 1928. Note araneologiche — Su la biologia degli scitodidi e la ghiandola glutinifera di
essi. Arch. Zool. Italiano, 12:63-122.

Petrunkevitch, A. 1933. An inquiry into the natural classification of spiders. Trans. Connecticut Acad.
Sci., 31:299-389.

Platnick, N. 1977. The hypochiloid spiders: A cladistic analysis with notes on the Atypoidea
(Arachnida: Araneae). Amer. Mus. Nov., 2627:1-31.

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

Robinson, M. H. and Y. D. Lubin, 1979. Specialists and generalists: The ecology and behavior of
some web-building spiders from Papua New Guinea. 11. Psechrus argentatus and Fecenia sp.
(Araneae: Psechridae). Pacific Insects, 21:133-164.

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

Robinson, M. H. and J. Olazarri, 1971. Units of behavior and complex sequences in the predatory
behavior of Argiope argentata (Fab.) (Araneae: Araneidae). Smithsonian Contrib. Zool., 65:1-36.

Robinson, M. H. and B. Robinson 1976. Discrimination between prey types: An innate component
of the predatory behavior of araneid spiders. Zeit. Tierpsychol., 41:266-276.

Rovner, J. S. and S. J. Knost. 1974. Post-immobilization wrapping of prey by lycosid spiders of the
herbaceous stratum. Psyche, 81:398-415.

Schaller, F. 1956. Deskriptive und analytische Instinktforschung bei Spinnentieren Instecten. Forsch.
Fort., 8:225-231.

Shear, W. A. 1969. Observations on the predatory behavior of the spider Hypochilus gertschi
Hoffman (Hypochilidae). Psyche, 6:407-417.

Simon, E. 1982. Histoire naturelle des Araignees. I (1): 1-256. Paris.

Valerio, C. 1974. Prey capture by Drymusa dinora (Araneae: Scytodidae). Psyche, 81:284-287.

Manuscript received June 1984, revised November 1984.

'-.V"

^ 't-rf. .rt

.’ .1 ■■ ,\. ' ! ' »i '‘.i ,'• :

V ' •' t •'■ \

‘ ‘'^' !w • ‘ vjj t , J% • ■ •

♦ ’■. ,i

■ 0; I.'Vt' •,■.•■,■

,0 - ’ . .

.r-i' iw

■J

J /j

''rv\

I '

I

' ■ Vor’1^.tV.V l.; •;: ^ • . ' •* ‘ j^i

»v/warj 'i

j ij'

••' (. J

Vh

■■

.',!*^l « . '‘.ifiitMY 'UJ- i '■

,■ :•; • \*T;i:0rf •• T, .t>,vt»|

Iv.

H

['M

■<t.

V'

i\f

ihM

• 'OA' O' ., !»' f',..‘...:7 V

, • -■ ' ■■ . * v7, ■' . - -■ I f, ,

i <1 » ■ t

■ :.'t

•"V

r*l

t=^'.

'■' ,v t'vf®

'/iAf.f --K'

^

.•*.i/>.''. 1 1.;

’ >J ■ ■ 4j

,,■?);

I . 'C* ;»(? , *..t ,'■]/

■. >- ',; ■' v> , s . , ;.», i'

'■■‘ 1-

,V '. ; '

In •

v: • .. ', ‘tiifii

r ^

*ti J,

<’>>?

f

,'. - W

C . .; ■'• i^ri' »»»*:■

. ;■ '■'•• v, U ;i'^ ?1n\ V

t„»

•i 7 :-._ ,‘i .\', ,,■<;. a.

, . i . ' ■ i 1 ', ' '”>

’*»! ' ■.:•■* 7-. >.!f.r' . '; ' ’Ji

n I

ft .»..A . i"4.a

ft"

I*.

*^L/ -.4^: ’v'V fV’J

v»’f jl-1

■-:,.yV;’^V4a::

■: • w

* ■ 4''^' '.' i ' ' !,’ ■'■ 1,“' I '■ .

1

, •r ' i:^,

V'; /,

./•' •* <U/4.. ' . 7 'I ■• *! 'Wills' 1.

. A .■•'.^s' '-I- >■'

<♦ r-r-'-.v' ' nA'-.iV*- '..* 5 , ■' ■ .•' ■

>■ ■ /f - V, . , .vv . ^■

V.' ' ' . 'i •,. ' n W''

. ,1

»■ ’ . . ' . ' ')^ I'v^

1(1

• t

! m' ' ■

*•'»

*» I

A

V

Sissom, W. D. and 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.

REDESCRIPTIONS OF SOME POORLY KNOWN SPECIES
OF NITIDULUS GROUP OF THE GENUS VAEJOVIS
(SCORPIONES, VAEJOVIDAE)

W. David Sissom

Department of General Biology, Vanderbilt University
Nashville, Tennessee 37235

and

Oscar F. Francke

Department of Biological Sciences, Texas Tech University

Lubbock, Texas 79409

ABSTRACT

The nitidulus group of Vaejovis is defined on the basis of pedipalp chela morphology and dentition,
trichobothrial patterns, and carinal structure of the pedipalp chela and the metasoma. Five species
are treated herein; V. mexicanus decipiens Hoffmann is elevated to specific status and transferred to
the nitidulus group; V. intermedius Borelli, V. nigrescens Pocock, and V. nitidulus C. L. Koch are
all considered valid species and redescribed; and V. minckleyi Williams is transferred to the group
and a revised diagnosis given.

INTRODUCTION

Vaejovis C. L. Koch, 1836, with over 90 described species and subspecies, is
the largest and most widespread genus of scorpions in North America. Through
the years, ^_number of distinct species groups have been established, largely
through the efforts of Williams (1970a, 1970b, 1971, 1980) and Soleglad (1972,
1973). Several of these groups are now recognized as valid genera (i.e.,
Paruroctonus Werner and Paravaejovis Williams; Williams 1970c, 1972).

Currently, four species groups remain in the genus: the eusthenura, mexicanus
(with the minimus subgroup), punctipalpi, and wupatkiensis groups. The purpose
here is to provide a preliminary definition of a fifth group, the nitidulus group,
and to clarify the status of five poorly understood taxa within it.

The most important features of the nitidulus group are: (1) the pedipalps are
long and slender with chela length/ palm width ratios greater than 3.3 (in most
species, greater than 4.0); (2) the pedipalp chela fingers are long and tenuous, and

244

THE JOURNAL OF ARACHNOLOGY

each terminates in an enlarged clawlike denticle bearing distally a distinct white
patch; (3) chela trichobothria ib and it are located at the base of the fixed finger;
(4) the pectinal teeth of the female are subequal in size, never with the proximal
teeth enlarged; (5) the primary row of denticles on the chela fixed finger is
divided into five, six or seven distinct subrows: (6) the dorsointernal carina of
the pedipalp chela is the strongest and usually bears enlarged, sharp granules; and
(7) the ventral submedian carinae of the metasoma are usually obsolete (but in
some species are weak to moderate). The following described species are referable
to the nitidulus group: V. carolinianus (Beauvois), V. decipiens Hoffmann (n.
comb.), V. gracilis Gertsch and Soleglad, V. intermedins Borelli, V. janssi
Williams, V. jonesi Stahnke, V. minckleyi Williams, V. nigrescens Pocock, V.
nitidulus C. L. Koch, V. peninsularis Williams, and V. spicatus Haradon. A
complete revision of this species group, with descriptions of new species, is in
preparation by the senior author.

The status of three of these species, V. nitidulus, V. nigrescens, and V.
intermedins, has been particularly confusing. Until now, each has been considered
a subspecies of V. nitidulus; with the nominal subspecies in Oaxaca, V. nitidulus
nigrescens in central and eastern Mexico, and V. nitidulus intermedins in
northwestern Mexico (Hoffmann 1931). Examination of the primary types of
these species and considerable material from pertinent areas indicates that all
three taxa are distinct species, and that V. nitidulus has been continuously
misidentified since its original description in 1843. A second source of confusion
in the identification of these taxa is the occurrence in central and southern
Mexico of the aforementioned new taxa. Some of these have almost certainly
been called V. nitidulus or V. nigrescens in the past (e.g., Diaz Najera 1964, 1975).
Unfortunately, our attempts to obtain specimens from Mexican collections have
failed, and we can neither confirm nor deny many of the old records for these
species. To clarify the identity and taxonomic status of the three taxa, complete
redescriptions are provided below.

Two other species are dealt with here. Vaejovis mexicanus decipiens Hoffmann
is elevated to specific status and redescribed as a valid species of the nitidulus
group based on examination of the types. Vaejovis minckleyi Williams, which has
lately been considered a member of the genus Paruroctonus (Stahnke 1974), is
returned to Vaejovis and placed in the nitidulus group. Because Williams’ original
description (1968) is largely adequate and the species is very distinctive, only a
revised diagnosis and some comments are given below, emphasizing new
characters of taxonomic importance.

Vaejovis nitidulus C. L. Koch
(Figs. 1-13)

Vaejovis nitidulus C. L. Koch 1843: 4, fig. 758; Peters 1861: 510; Karsch 1879: 135; nec Pocock 1902:

12; Borelli 1915: 7; Moritz & Fischer 1980: 320.

Vejovis nitidus (sic), Thorell 1876: 186.

Vejovis spinigerus, Kraepelin 1894: 202 (part).

Vejovis nitidulus, nec Kraepelin 1899: 186; Biicherl 1959: 271 (?), 1971: 329; Stahnke 1974: 135.

Vejovis nitidulus nitidulus, nec Hoffmann 1931: 371, nec 1937: 204, nec 1939: 318; nec Gertsch 1958:5;
Diaz Najera 1964: 20 (part?), 1975: 7 (part?)

SISSOM AND FRANCKE—A/r/DC/Lt/N GROUP REDESCRIPTIONS

245

Type data. — Of four syntypes of Vaejovis nitidulus used by Koch (1843) in his
original description, only two remain (Moritz and Fischer 1980). Both are adult
females, and represent two different taxa. Koch’s description is too general to
determine whether he based it on either specimen or all four; likewise, his
measurements cannot be assigned with certainty to either specimen (Although his
measurements, which are few and general, are similar to those of the smaller
female, it is not certain that they indeed belong to that female; it is equally
possible that the measurements refer to one of the missing syntypes).

We hereby select the larger female from Mexico (Deppe, coll.), bearing the
label ZMB 10a, as the lectotype for V. nitidulus. We have identified conspecific
material from Hidalgo and eastern Queretaro, Mexico, and the coloration of
these specimens match very well the description by Koch and his color plate
(Koch 1843: fig. 758).

The lectotype, originally a dry, pinned specimen, is in very poor condition. The
pedipalps, metasoma, and most of the legs are detached from the body; some of
the legs themselves are broken into separate segments, as is one pedipalp. Some
damage by dermestid beetles is evident. The carapace is also detached from the
body, and the specimen is strongly discolored. However, in spite of its poor
condition, no body parts are missing and taxonomic characters are readily
observable.

The identity of specimen ZMB 10 is unknown. It is definitely not referable to
V. nitidulus, and we have not yet seen any material which is conspecific with it.
In the vial containing this specimen is a label written by H. L. Stahnke, selecting
this specimen as lectotype. Dr. Stahnke has never published this designation, and
therefore, it is not valid according to the International Code of Zoological
Nomenclature, Art. 74(a)(i). The reference to Stahnke’s invalid lectotype by
Moritz & Fischer (1980) also does not constitute proper designation, as it was
not intended as such [ICZN 74(c)]. Stahnke’s paralectotype label in the vial
containing ZMB 10a should likewise be disregarded.

Both specimens are deposited in the Zoologisches Museum of Humboldt
Universitat in Berlin.

Distribution. — Known from Hidalgo, and eastern Queretaro, Mexico.

Diagnosis. — Adults 45-65 mm in length. Base color yellow brown, without
conspicuous underlying markings. Sternite VH with lateral keels weak, granular.
Metasoma with ventral submedian carinae obsolete on I-IV; ventrolateral carinae
moderate to weak, smooth; distalmost granules of dorsolateral and lateral
supramedian carinae enlarged, spinoid. Pedipalp: tibia with 15 trichobothria (3
et, 1 est, 2 em, 3 esb, 5 eb, 1 v) on external face; fixed finger with primary row
divided into seven subrows by six larger granules; keels of chela developed.
Pectinal tooth count 24-28 in males, 21-27 in females.

Vaejovis nitidulus, in possessing an extra esb trichobothrium on the tibia and
high pectinal tooth counts, is most similar to V. minckleyi. It is easily
distinguished from that species by possessing seven subrows of denticles on the
pedipalp chela fixed finger (instead of six subrows), by having smooth or obsolete
keels on the dorsal and external surfaces of the pedipalp chela (instead of
granulose keels), by having obsolete ventral submedian keels on the metasoma
(rather than weak, granular keels), and by having metasomal segments I-H wider
than long (instead of longer than wide).

246

THE JOURNAL OF ARACHNOLOGY

Redescription. — The following redescription is based on adults. Parenthetical
statements refer to females. Measurements of the female lectotype and an adult
male appear in Table 1.

Coloration (in alcohol). Carapace and tergites yellow brown. Metasomal
segments I-III yellow brown to light orange brown, IV-V orange brown to
reddish brown. Telson yellow brown to light orange brown; aculeus dark brown.

Figs. 1-10 — Vaejovis nitidulus Koch, male from Hidalgo, Mexico: 1, dorsal aspect of pedipalp
femur; 2, dorsal aspect of pedipalp tibia; 3, external aspect of pedipalp tibia; 4, ventral aspect of
pedipalp tibia; 5, lateral aspect of metasomal segments IV and V, and telson; 6, dorsal aspect of
pedipalp chela; 7, external aspect of pedipalp chela; 8, ventral aspect of pedipalp chela; 9, dentition
pattern on fixed finger of pedipalp chela; 10, dentition pattern on movable finger of pedipalp chela.

SISSOM AND FRAl^CKE—NITIDULUS GROUP REDESCRIPTIONS

247

Pedipalps: femur and tibia yellow, lighter than body; chela orange brown, darker
at base of fingers with yellowish fingertips; denticles of dentate margin brown.
Legs yellow with some faint dusky markings; tarsi yellowish. Venter: coxosternal
region yellow brown; pectines yellowish white; sternites light yellow brown; third
sternite in male with whitish patch along posteromedial margin (absent).

Prosoma. Carapace approximately as long as wide. Median ocular prominence
moderately raised above carapacial surface. Three pairs of lateral eyes. Anterior
carapacial margin obtusely emarginate; median notch shallow, narrow. Median
longitudinal furrow deep, wide at anterior margin; deep, narrow posteriorly.
Posterior lateral furrow curved, deep, wide. Entire carapacial surface densely
granular.

Mesosoma. Tergites I-VI: median carina on I obsolete, on II-VI weak,
granular; submedian carinae vestigial. Tergite VII: median carina weak, finely
granular, present on anterior one-third to one-half; submedian and lateral carinae
strong, serratocrenulate. Genital operculi distinctly lobed posteriorly; without
median longitudinal membranous connection (with short basal connection, one-
half length of genital operculum). Genital papillae well developed (absent).
Pectinal teeth numbering 24-28 (21-27). Sternites III-VI smooth, stigmata about
three times longer than wide. Sternite VII with median pair of carinae obsolete;
lateral carinae weak, granular.

Metasoma. Segments I-IV: Segments I-II, occasionally III, wider than long,
others longer than wide. Dorsolateral carinae on I-III strong, serrate; on IV
strong, irregularly crenulate to serrate; distalmost denticle distinctly enlarged,
spinoid (Fig. 5). Lateral supramedian carinae on I strong, serrate; on II-lII
strong, finely crenulate to finely serrate; on IV moderate, smooth to finely serrate;
distalmost denticle distinctly enlarged, spinoid on I-III; flared and winglike on IV
(Fig. 5). Lateral inframedian carinae on 1 complete, strong, crenulate to serrate;
on II present only on posterior one-third, crenulate to serrate; on III present on
posterior one-fourth, granular to crenulate; on IV absent. Ventrolateral carinae
on I moderate, smooth, sometimes with few posterior serrations; on II-IV weak,
smooth, posterior serrations sometimes present. Ventral submedian carinae on I-
IV obsolete. Intercarinal spaces irregularly granular, lustrous. Segment V (Fig. 5):
slightly more than twice as long as wide. Dorsolateral carinae weak, granular.
Lateral median carinae present on anterior three-fourths, weak, granular.
Ventrolateral and ventromedian carinae moderate, finely serrate. Intercarinal
spaces irregularly granular, lustrous.

Telson (Fig. 5.). Dorsal surface flattened, smooth; ventral surface with
irregularly spaced granules and punctations.

Chelicera. Movable finger internally with well developed serrula.

Pedipalp. Femur (Fig. 1) tetracarinate. Dorsointernal, dorsoexternal, and
ventrointernal carinae strong, crenulate. Ventroexternal carina strong, with
irregularly spaced, large rounded granules. Internal face with large, conical
granules; dorsal face coarsely granular; ventral and external faces with
granulation on basal portion. Orthobothriotaxia C (Vachon 1974).

Tibia (Fig.s 2-4) tetracarinate. Dorsointernal and ventrointernal carinae strong,
crenulate. Dorsoexternal carina moderate, granular to crenulate. Ventroexternal
carina moderate, crenulate. Internal face with moderate basal tubercle and large,
sharp granules. Dorsal face smooth; ventral face with few scattered, coarse

248

THE JOURNAL OF ARACHNOLOGY

granules; external face with weakly granular, longitudinal series of granules.
Neobothriotaxia C (Vachon 1974), with three esb trichobothria; esbi petite (Fig.
3). Positions of em\ and errii variable (see Figs. 1 1-13).

Chela (Figs. 6-10): Dorsal marginal carina weak, granular. Dorsal secondary
and digital carinae weak, smooth. External secondary carina obsolete.
Ventroexternal and ventromedian carinae weak, smooth. Ventrointernal carina
weak, granular. Dorsointernal carina moderate, with large sharp granules.
Dentate margin of fixed finger with primary row broken up into seven subrows
by six larger granules; six internal accessory granules of which all but distal one
paired with larger granule in primary row (Fig. 9). Dentate margin of movable
finger with primary row broken up into seven subrows by six enlarged granules;
apical row consisting of one or two small granules; seven internal accessory
granules of which distal granule not paired with larger granule of primary row,
basalmost granule distinctly basal to corresponding granule in primary row (Fig.
10). Both fingers terminating in large, sharp, claw-like tooth bearing distally an
oblong elongate whitish cap, Orthobothriotaxia C (Vachon 1974).

Legs. Tarsomere I on legs I-II with one retrolateral and two ventrolateral rows
of spinules; ventral rows complete, interrupted at irregular intervals by large
spines. Retrolateral row present on distal one-half, interrupted by one or two
spines. Tarsomere I spinule rows rudimentary on legs III-IV, with spines well
developed. Tarsomere II on all legs with single ventromedian row of spinules,
procurved basally, terminating distally between single pair of medium-sized
spines. Spinule rows flanked by pairs of setae as in Soleglad (1975: Fig. 17); seta
formula given in Table 3.

Variation. — Although only a small number of specimens have been located and
examined, pectinal tooth counts varied as follows: among males, 5 combs with
24 teeth, 3 combs with 25 teeth, 3 combs with 26 teeth, 2 combs with 27 teeth,
and 3 combs with 28 teeth; among females, 2 combs with 21 teeth, 3 combs with
22 teeth, 4 combs with 23 teeth, 4 combs with 24 teeth, 4 combs with 25 teeth,
3 combs with 26 teeth, and 2 combs with 27 teeth. As is typical of the group,
adult males have relatively shorter, wider chelae than females and slightly longer
and narrower metasomal segments. No other significant variation was observed.

Figs. 1 \ -U.— Vaejovis nitidulus Koch, males from Hidalgo, Mexico: external aspect of pedipalp tibia
showing variability in position of em and esh trichobothria.

SISSOM AND FRANCKE-^yV/r/Dt/L^/6'GROUP REDESCRIPTIONS

249

Remarks. — V. nitidulus and its relationship to other species in the group have
been poorly understood since the original description. Hoffmann (1931), in his
treatment of the scorpions of Mexico, considered V. nitidulus to be polytypic,
containing three subspecies: V. n. nitidulus, K n. intermedius, and V. n.

nigrescens. The new characters established by examination of the types of the
three nominal taxa (i.e., number of subrows in the dentate margin of the chela
fingers, structure of pedipalp chela keels, trichobothrial pattern of the tibia,
pectinal tooth counts, and morphometries) show conclusively that each is a good
species. Problems in identifying these taxa were compounded by the fact that
Hoffmann (1931) misidentified K nitidulus. We have examined some of

Hoffmann’s specimens from Cuicatlan, Oaxaca, and they are not referable to V.
nitidulus (actually they represent a new species being described by the senior
author). All records of V. nitidulus from Guanajuato and western Queretaro (Diaz
Najera 1975) need to be confirmed.

Specimens examined. — MEXICO: no specific locality or date (Deppe), 1 lectotype female (ZMB)
Hidalgo: Jacala (east side), W99.ll: N21.01, 27 July 1966 (J. & W. Ivie), 1 female (AMNH), 1 juv
(AMNH); 18 July 1963 (R. E. Woodruff) (under cow dung), I female (FSCA); Jacala, W99.12:
N21.01, 20 April 1963 (W. J. Gertsch & W. Ivie), 1 female (AMNH); Taxquillo (Tzindejeh), W99.I9:
N20.33, 29 July 1966 (J. & W. Ivie), 1 male (AMNH), 1 juv (AMNH): 1 mi SE Danghu (3 mi S
Taxquillo), 22 Aug 1984 (W. D. Sissom, C. Myers, L. Born) (N slope of rocky hillside, UV light),
5 males, 2 females (WDS), 1 male, 1 female (OFF); Queretaro: Mountains near San Joaquin, July
1976 (S. Minton), 1 female (SAM); Mission Bucareli, 31 Dec 1981 (S. Minton), I female (SAM),
1 Juv female (MEB).

Vaejovis nigrescens Pocock
(Figs. 14-23)

Vaejovis nigrescens Pocock 1898:396.

Vejovis nitidulus, Kraepelin 1899: 186 (part); 1901: 274 (part).

Vaejovis nitidulus, Pocock 1902: 12.

Vejovis nitidulus nigrescens, Hoffmann 1931: 365, 1937: 204, 1939: 318; Gertsch 1958: 5.

Vaejovis nitidulus nigrescens, Diaz Najera 1964:20, 24, 25, 29; 1975: 7, 22, 25, 26, 30, 34.

Type data. — Adult female holotype from Mexico (no date or collector);
deposited in the British Museum (Natural History); examined. The holotype,
formerly a pinned specimen, is in poor condition: metasomal segments H-V and
the telson are detached from the body and held together with a pin; the right
pedipalp, its movable finger, and most of the legs are also detached; the pectines
are shriveled and broken; and the specimen is strongly discolored.

Distribution.. — The specimens we have examined are from the central Mexican
states of Aguascalientes, Distrito Federal, and Guanajuato. Hoffmann (1931,
1937) and Diaz Najera (1964, 1975) record V. nigrescens additionally from
Hidalgo, Queretaro, and adjacent parts of Jalisco, Michoacan, San Luis Potosi,
and Zacatecas. As in the case of V. intermedius, these records need to be verified.
We have also examined a series of specimens from Dinamita, Durango collected
by Duges; this record is far outside the range of nigrescens (including the records
of Hoffmann and Diaz Najera) and may possibly be the result of a labeling error.

Diagnosis. — Adults 42-68 mm in length. Base color orange brown to reddish
brown with variable faint underlying dusky markings. Sternite VII with
submedian keels obsolete; lateral keels weak to moderate, granular. Metasomal

250

THE JOURNAL OF ARACHNOLOGY

segments I-II wider than long, III as long as wide, V less than twice as long as
wide; ventrolateral carinae of I-IV weak, smooth to finely granular; ventral
submedian carinae obsolete. Pedipalp: tibia with 14 trichobothria (3 et, 1 est, 2
em, 2 esb, 5 eb, 1 v) on external face; fixed finger of chela with primary row
of denticles broken into six subrows by five enlarged denticles; femur as long as
carapace; dorsointernal carina of chela with weak to moderate, rounded granules.
Pectinal tooth count 19-21 in males; 17-21 in females.

Vaejovis nigrescens is most similar to V. intermedins. For comparisons between
these two species, see the diagnosis of the latter species.

Redescription. — The following redescription is based on adult males and
females. Parenthetical statements refer to females. Measurements of the female
holotype and an adult male appear in Table 1.

Coloration. Base color of carapace and tergites orange brown to reddish
brown, with faint underlying dusky markings. Venter yellow brown to orange
brown; third sternite in males with whitish patch along posteriomedial margin
(absent). Metasomal segments I-III orange brown to reddish brown, IV-V dark
reddish brown. Telson reddish brown, with dark brown aculeus. Chelicerae yellow
brown at base, mottled with brown distally; teeth brown. Pedipalps: Femur and
tibia yellow brown to orange brown with carinae somewhat darkened; chela
manus reddish brown to dark orange brown; fingers dark brown basally, reddish
brown to dark orange brown distally. Legs yellow to orange brown, usually
lighter than body. Pectines whitish.

Prosoma. Carapace slightly longer than wide. Median ocular prominence only
slightly raised above carapacial surface. Three pairs of lateral eyes, diameter of

Table 1. Measurements in mm and meristic characters of Vaejovis nigrescens Pocock and Vaejovis
nitidulus C. L. Koch.

Character

V. nitidulus

lectotype $ adult $

(AMNH, Taxquillo)

V

holotype $

nigrescens

adult $

(MNHN, RS-0671)

Total length

64.8

50.4

61.8

49.5

Carapace length

7.7

5.8

7.4

5.9

Mesosoma length

20.8

14.8

20.6

14.4

Metasoma length

27.5

23.1

25.9

22.0

I length/ width

3.6/4.8

3.0/3.6

3.4/4.4

2.8/3. 8

11 length/ width

4.0/4.8

3.6/3.5

3.9/4.5

3.4/3.8

III length/ width

4.5/4.8

3.9/3.5

4.3/4.4

3. 8/3. 8

IV length/ width

6.2/4.6

5.1/3.4

6.0/4.4

5.0/3.9

V length/ width

9.2/4.6

7.5/3.4

8.3/4.4

7.0/3.S

Telson length

8.8

6.7

7.9

12

Vesicle length/ width/depth

5.8/3.4/2.8

43/2.4/ 1.9

5.1/3.4/2.6

4.7/2.8/2.2

Aculeus length

3.0

2.4

2.8

2.5

Pedipalp length

26.7

22.2

27.3

21.3

Femur length/ width

l.ljl.l

6.0/ 1.6

7.2/2.0

5.6/ 1.6

Tibia length/ width

7.5/2.3

6.2/ 1.5

7.5/2.2

6.0/ 1.7

Chela length/ width/depth

12.0/2.6/3.0

10.0/2.2/2.8

12.6/2.8/3.2

9.7/2.9/2.9

Movable finger length

8.0

6.6

8.6

6.0

Fixed finger length

6.5

5.5

7.1

4.7

No. primary rows (FF/MF)

7/7

7/7

6/6

6/6

Int. accessory granules (FF/MF)

6/7

6/7

6/7

6/7

Pectinal teeth ( 1 / r)

27-26

26-24

18-19*

24-23

*as reported by Pocock (1898).

SISSOM AND FRANCKE— yV/r/Z)f/Lt/5’GROUP REDESCRIPTIONS

251

posteriormost pair approximately one-half the diameter of the preceding pairs.
Anterior carapacial margin obtusely emarginate, median notch distinct, shallow,
wide. Median longitudinal furrow deep, wide at anterior margin; deep, narrow
posteriorly. Posterior lateral furrows curved, deep, wide. Entire carapacial surface
coarsely granular.

Figs. 14-23. — Vaejovis nigrescens Pocock, male from Guanajuato, Mexico: 14, dorsal aspect of
pedipalp femur; 15 dorsal aspect of pedipalp tibia; 16, external aspect of pedipalp tibia; 17, ventral
aspect of pedipalp tibia; 18, lateral aspect of metasomal segments IV and V, and telson; 19, dorsal
aspect of pedipalp chela; 20, external aspect of pedipalp chela; 21, ventral aspect of pedipalp chela;
22, dentition pattern on fixed finger of pedipalp chela; 23, dentition pattern on movable finger of
pedipalp chela.

252

THE JOURNAL OF ARACHNOLOGY

Mesosoma. Tergites I-VI: Median carina on I obsolete, on II-VI weak,
granular; submedian carinae vestigial. Tergite VII: median carina weak, granular,
present on anterior one-half to two-thirds; submedian and lateral carinae strong,
crenulate. Genital operculi distinctly lobed posteriorly, without median
longitudinal membranous connection (with basal connection two-thirds length of
genital operculum); genital papillae well developed (absent). Pectinal teeth
numbering 19-21, mode = 21 (17-21, mode = 18). Sternites III-VI smooth, with
slit-like stigmata. Sternite VII with one pair of lateral carinae, these weak to
moderate, granular.

Metasoma. Segments I-IV: Segments I-II shorter than wide. III as long as or
longer than wide, IV longer than wide. Dorsolateral carinae on I-III strong,
crenulate; on IV strong, irregularly crenulate; distalmost denticle on I-IV
distinctly enlarged, spinoid (Fig. 18). Lateral supramedian carinae on I-III strong,
crenulate to finely crenulate; on IV strong, granular to finely crenulate; distalmost
denticle on I-III distinctly enlarged, spinoid, on IV widely flared (Fig. 18). Lateral
inframedian carinae on I complete, strong, crenulate; on II present on posterior
one-half, strong, crenulate; on III present on posterior one-third, moderate,
crenulate; on IV absent. Ventrolateral carinae on I weak, finely granular; on II-

IV weak, smooth to finely granular. Ventral submedian carinae obsolete.
Intercarinal spaces dorsally and laterally with scattered, coarse granules. Segment

V (Fig. 18): Averaging 1.95 (1.84) times as long as wide. Dorsolateral carinae
weak to moderate, granular. Lateromedian carinae weak, finely granular.
Ventrolateral and ventromedian carinae weak, finely serrate. Intercarinal spaces
with scattered coarse granules.

Telson (Fig. 18). Dorsal surface smooth, flattened; ventral surface with
irregularly spaced granules and punctations; subtle subaculear tubercle sometimes
present.

Chelicera. Dentition typical of genus, with well developed serrula on ventral
margin of movable finger.

Pedipalp. Femur (Fig. 14) tetracarinate: Dorsointernal, ventrointernal, and
dorsoexternal carinae strong, granulose. Ventroexternal carina strong, composed
of large irregularly spaced, sharp granules. Internal face with about 15 large,
subconical granules; dorsal face with scattered coarse granules; ventral face with
coarse granulation proximally. Femur as long as or slightly longer than carapace.
Orthobothriotaxia C (Vachon 1974).

Tibia (Figs. 15-77): Dorsointernal and ventrointernal carinae strong, granulose.
Dorsoexternal carina moderate, granular. Ventroexternal carina moderate,
somewhat crenulate. Internal face with about 10 large, sharp, subconical granules;
dorsal and ventral faces with scattered coarse granules; external face with
vestigial, granular external keel. Orthobothriotaxia C (Vachon 1974);
trichobothrium enii consistently basal to em\ (Fig. 16).

Chela (Figs. 19-23): Dorsal marginal carina moderate, granular. Dorsal
secondary keel weak, smooth. Digital carina weak basally, moderate distally,
smooth. External secondary and ventromedian carinae weak, smooth.
Ventrointernal carina weak, finely granular. Dorsointernal carina weak to
moderate, with medium-sized granules more rounded than sharp. Dentate margin
of fixed finger with primary row divided into six subrows by five larger denticles;
six internal accessory granules, of which distalmost not paired with larger granule

SISSOM AND FRANCKE^A/r/Z)f/Lf/5’ GROUP REDESCRIPTIONS

253

in primary row (Fig. 22). Dentate margin of movable finger with primary row
divided into six subrows by five larger granules; apical subrow with only one or
two granules; seven internal accessory granules, of which distalmost and
basalmost not paired with larger granule in primary row (Fig. 23). Both chela
fingers terminating in large, sharp, clawlike tooth bearing distally an oblong
whitish cap. Fingers, of male with weak scalloping. Orthobothriotaxia C (Vachon
1974).

Legs. Arrangement of setae, spines, and spinules as in V. nitidulus; seta formula
given in Table 3.

Variation. — Male pectinal tooth counts varied as follows: 6 combs with 19
teeth, 11 combs with 20 teeth, and 12 combs with 21 teeth. Female pectinal tooth
counts varied as follows: 12 combs with 17 teeth, 22 combs with 18 teeth, 7
combs with 19 teeth, 3 combs with 20 teeth, and 1 comb with 21 teeth. Pectinal
tooth counts of males and females are statistically different (Fi,75 71.0, p <

0.001).

Sexual dimorphism occurs in body size and morphometries of the pedipalp
chela and metasoma (see Table 4). Adult males range from 42-52 mm in length;
adult females from 46-68 mm.

Specimens examined. — MEXICO: no specific locality, date, or collector, 1 holotype female (BM);
no specific locality, date or collector, 1 female (MCZ), 3 males, 1 female (MCZ); no date, 1 female
(AMNH); Aguascalientes: Rio de Pirules (no date or collector), 1 female (AMNH); Mexico, D.F:
Mexico City, no date (Bononsea), 4 males, 2 females (Sc516, ex. 617) (TOR); Durango: Dinamite
(no date or collector), 3 males, 1 female (Sc 517, ex. 656) (TOR); Guanajuato: Guanajuato, no date
(Duges), 4 females (RS-0668) (MNHN), 2 females (RS-0678) (MNHN), 1 female (RS-0684) (MNHN),
1 male, 5 females, 7 first instars (RS-0671) (MNHN), 1 female (BM); Guanajuato, no date (Duges)
(in houses) 3 males, 4 females (Sc 515, ex. 616) (TOR), 2 females (BM), 1 female (BM), Guanajuato,
June 1963 (S. A. Minton), 1 female (SAM).

Vaejovis intermedius Borelli
(Figs. 24-33)

Vaejovis intermedius Borelli 1915: 6.

Vejovis nitidulus intermedius, Hoffmann 1931: 368, 1937: 204, 1939:318; Gertsch 1958: 5.

Vaejovis nitidulus intermedius, Diaz Najera 1964: 20, 24, 25; 1975: 7, 22, 25, 26.

Type data. — The type series consists of three adult males and three adult
females taken from Dinamita, Durango, Mexico (collector and date unknown).
From this type series we designate one adult male, clearly labeled as lectotype
and the remaining specimens as paralectotypes. One of the female syntypes has
been labeled^ as a lectotype by H. L. Stahnke; however, this designation is not
valid as it has never been published [International Code of Zoological
Nomenclature, Art. 74 (a) (i)]. We have chosen to select the male as lectotype
because it is intact; the female is broken into three parts.

The lectotype and five paralectotypes are permanently deposited in the Museo
ed Istituto di Zoologia Sistematica della Universita di Torino; Torino, Italy (cat.
no. Sc 508, ex. 799).

Distribution. — Vaejovis intermedius is known from southwestern Texas
(Brewster, Crockett, Presidio, Terrell, and Val Verde Counties), U.S.A. and the
state of Durango in Mexico. Hoffmann (1931) reports the species also occurs in
the Sierra de Guadalupe, Distrito Federal, Mexico. Diaz Najera (1964, 1975)

254

THE JOURNAL OF ARACHNOLOGY

reports additional records from Hidalgo and Jalisco. We have been unable to
obtain these specimens, and, in light of the confusion previously surrounding the
species of the nitidulus group, we must consider these records as tentative.

Diagnosis. — Adults 40-57 mm in length. Base color yellowish to reddish brown;
pedipalps and metasoma dark reddish brown distally. Sternite VII with
submedian keels obsolete; lateral keels moderate, smooth to crenulate. Metasoma:
segments I-II wider than long. III as wide as long; IV-V longer than wide;
ventrolateral carinae moderate, smooth to finely granular; ventral submedian

Figs. 24-33. — Vaejovis intermedius Borelli, lectotype male from Durango, Mexico: 24, dorsal aspect
of pedipalp femur; 25, dorsal aspect of pedipalp tibia; 26, external aspect of pedipalp tibia; 27, ventral
aspect of pedipalp tibia; 28, lateral aspect of metasomal segments IV and V, and telson; 29, dorsal
aspect of pedipalp chela; 30, external aspect of pedipalp chela; 31, ventral aspect of pedipalp chela;
32, dentition pattern on fixed finger of pedipalp chela; 33, dentition pattern on movable finger of
pedipalp chela.

SISSOM AND FRANCKE— A/r/Dt/LC/NGROUP REDESCRIPTIONS

255

carinae on I-IV obsolete; distalmost denticle of dorsolateral and lateral
supramedian carinae of I-IV distinctly enlarged, spinoid. Pedipalp: tibia with 14
trichobothria (3 et, 1 est, 2 em, 2 esb, 3 eb, 1 v) on external face; fixed finger
of chela with primary row of denticles broken into six subrows by five enlarged
denticles; pedipalp femur distinctly shorter than carapace; keels of dorsal and
external faces of chela faint to weak, smooth. Pectinal tooth count 21-26 in
males, 19-23 in females.

Vaejovis intermedins is most similar to V. nigrescens Pocock from central
Mexico. The most significant differences between the two are as follows: (1) male
pectinal tooth counts range from 21-26 (mode = 23, 24) in K intermedins, 19-
24 (mode = 21) in V. nigrescens; (2) female pectinal tooth counts range from
19-23 (mode = 21) in V. intermedins, 17-22 (mode = 18) in K nigrescens: (3) the
pedipalp chela is proportionately wider and deeper in V. intermedins, with shorter
fingers (see Table 4); (4) the dorsointernal carina of the pedipalp chela possesses
strong, sharp granules in V. intermedins, but weak, more rounded granules in V.
nigrescens; and (5) metasomal segment V length/ width is distinctly greater in
both males and females of V. intermedins than in V. nigrescens (Table 4).

Redescription. — The following redescription is based on adult males and
females. Parenthetical statements refer to females and indicate sexual differences.
Measurements of the lectotype male and a paralectotype female appear in Table
2.

Coloration. Base color of carapace and tergites yellowish to orange brown,
with variable underlying dusky markings; in some specimens, dusky markings
appear as faint submedian stripes on tergites. Venter yellow to yellow brown;
third sternite in males with whitish patch along posteriomedial margin (absent).

Table 2. — Measurements in mm and meristic characters of Vaejovis decipiens Hoffmann and
Vaejovis intermedius Borelli.

Character

V. intermedius

lectotype $ paralectotype 9

V. decipiens
holotype

Total length

46.3

51.6

53.7

Carapace length

5.5

6.5

6.4

Mesosoma length

12.7

16.0

14.9

Metasoma length

21.7

22.3

25.1

I length/ width

2.9/3.4

3. 1/3.8

3.5/3.6

11 length/width

3.4/ 3.4

3.4/3.7

4.0/3.5

III length/width

3.6/3. 3

3. 8/3.6

4.3/3.6

IV length/ width

4.8/3.2

5.0/3.5

5.6/3.4

V length/ width

7.0/3. 1

7.0/3. 5

7.7/3.4

Telson length

6.4

6.8

7.3

Vesicle length/ width/depth

4.1/2.4/ 1.8

4.4/2.8/2.1

4.8/2.6/2.0

Aculeus length

2.3

2.4

2.5

Pedipalp length

18.9

21.1

24.9

Femur length/ width

5.0/ 1.6

5.7/ 1.9

6.7/ 1.7

Tibia length/ width

5.2/ 1.7

5. 8/2.1

7.0/ 1.8

Chela length/ width/depth

8.7/2.5/2.7

9.6/2.4/2.6

11.2/2.6/3.2

Movable finger length

5.5

6.1

7.2

Fixed finger length

4.5

5.1

6.1

No. primary rows (FF/MF)

6/6

6/6

6/6

Int. accessory granules (FF/MF)

6/7

6/7

6/7

Pectinal teeth

25-25

21-21

24-24

256

THE JOURNAL OF ARACHNOLOGY

Pectines whitish. Metasomal segments I-III (sometimes IV) yellowish brown; IV-

V dark reddish brown; ventral submedian and ventrolateral carinae usually with
faint underlying dusky pigment. Telson reddish brown; aculeus dark reddish
brown. Chelicerae yellowish brown; teeth dark reddish brown. Pedipalps: femur
yellowish brown; tibia yellowish brown, often reddish brown distally; chela
reddish brown, with palm somewhat lighter than fingers. Legs yellow brown.

Prosoma. Carapace slightly longer than wide. Median ocular prominence only
slightly raised above carapacial surface. Three pairs of lateral eyes, diameter of
posteriormost pair approximately one-half the diameter of preceding pairs.
Anterior carapacial margin obtusely emarginate; median notch distinct, shallow,
narrow, median longitudinal furrow deep, wide at anterior margin; deep, narrow
posteriorly. Posterior lateral furrow curved, deep, wide. Entire surface of
carapace moderately granular, shagreened (less granular, lustrous).

Mesosoma. Tergites I-VI: Median carina on I obsolete, on II-VI weak,
granular; lateral carinae vestigial. Tergite VII pentacarinate: median carina weak,
granular, present on anterior one-third to one-half; submedian and lateral carinae
strong, serrate. Genital operculi distinctly lobed posteriorly; without median
longitudinal membranous connection (with short basal connection, one-half
length of genital operculum); genital papillae well developed (absent). Pectinal
teeth numbering 21-26, mode = 23, 24 (19-23, mode = 21). Sternites III-VI
smooth, agranular; stigmata approximately three times longer than wide. Sternite

V with submedian carinae obsolete; lateral carinae moderate, smooth to crenulate
(smooth).

Metasoma. Segments I-II (occasionally III) wider than long, III-IV longer than
wide. Dorsolateral carinae strong, serrate; distalmost denticle on I-IV distinctly
enlarged, spinoid (Fig. 28). Lateral supramedian carina on I-II strong, serrate; on
III strong, finely serrate (finely crenulate); on IV moderate, granular (granular to
finely crenulate); distalmost denticle distinctly enlarged, spinoid on I-III, flared
and winglike on IV (Fig. 28). Lateral inframedian carinae on I complete,
moderate, irregularly serrate (crenulate); on II-III present on posterior one-half,
moderate, crenulate; on IV absent. Ventrolateral carinae moderate; on I-II
smooth anteriorly, finely granular posteriorly; on III-IV smooth to finely serrate.
Ventral submedian carinae obsolete. Intercarinal spaces dorsally and ventrally
with faint luster, all faces with scattered small granules. Segment V (Fig. 28):
Dorsolateral carinae moderate, anteriorly serrate, posteriorly granular. Lateral
median carinae present on anterior three-fifths, weak, smooth to granular.
Ventrolateral carinae moderate, finely serrate (crenulate). Ventromedian carina
moderate, finely crenulate to finely serrate. Intercarinal spaces as on I-IV.

Telson (Fig. 28). Dorsal surface smooth, flattened; ventral surface with
irregularly spaced punctations. Subtle subaculear tubercle present.

Chelicera. Dentition typical of genus, with well developed serrula on ventral
margin of movable finger.

Pedipalp. Femur (Fig. 24) tetracarinate: Dorsointernal and ventrointernal
carinae strong, serratocrenulate. Dorsoexternal carina strong, serrate.
Ventroexternal carina strong, with irregularly spaced, large serrate granules.
Internal face with numerous, large conical granules; dorsal and ventral faces
granular; external face with few proximal granules. Orthobothriotaxia C (Vachon
1974).

SISSOM AND FRANCKE—A/r//)^/Lt/5 GROUP REDESCRIPTIONS

257

Tibia (Figs. 25-27) tetracarinate: Dorsointernal carina strong, serratocrenulate.
Ventrointernal carina serratocrenulate to serrate. Dorsoexternal carina moderate,
granular (smooth). Ventroexternal carina moderate, granular to crenulate.
Internal face with moderate basal tubercle and short row of enlarged conical
granules; external face granular; dorsal and ventral faces smooth.
Orthobothriotaxia C (Vachon 1974). Trichobothrium erni consistently basal to
em\.

Chela (Figs. 29-33): Dorsal marginal carina moderate, with medium sized sharp
granules. Dorsointernal carina strong, composed of large, serrate granules.
Ventrointernal carina weak to moderate, granular. Ventroexternal carina weak,
granular (smooth). Ventromedian carina weak, smooth. Other carinae faint,
smooth. Dentate margin of movable finger with primary row divided into six
subrows by five larger granules; six internal accessory granules, of which
distalmost not paired with larger granule in primary row (Fig. 32). Dentate
margin of movable finger with primary row divided into six subrows by five
larger granules; distal subrow with only one or two granules; seven internal
accessory granules, of which distalmost and basal most not paired with larger
granules in primary row (Fig. 33). Chela fingers moderately (weakly) scalloped.
Both chela fingers terminating distally in large, clawlike tooth bearing distally an
oblong whitish cap. Orthobothriotaxia C (Vachon 1974).

Legs. Arrangement of setae, spines, and spinules as in V. nitidulus; seta formula
given in Table 3.

Variation. — Male pectinal tooth counts varied as follows: 1 comb with 21 teeth,
8 combs with 22 teeth, 25 combs with 23 teeth, 25 combs with 24 teeth, 9 combs
with 25 teeth, and 4 combs with 26 teeth. Female pectinal tooth counts varied
as follows: 2 combs with 19 teeth, 12 combs with 20 teeth, 68 combs with 21
teeth, 37 combs with 22 teeth, and 4 combs with 23 teeth. Pectinal tooth counts
of males and females are statistically different (Fi,i93 = 340.3, p < 0.001).

Coloration varies with age. Juveniles (including subadults) tend to be light
yellow with more prominent dusky markings on the body. The distal segments

Table 3. — Seta formulae and variation in setal counts of the ventral surface of the tarsi of legs I-
IV in Vaejovis spp. The numerator of the fractions indicates the number of setae in the prolateral
row; the denominator indicates the number in the retrolateral row. The number outside the
parentheses is the mode (=most common observation); the numbers within parentheses indicate the
range.

SPECIES

1

left

1

right

left

II

right

left

III

right

left

IV

right

nitidulus

2(1-2)

2(1-2)

4

4(3-4)

4(4-5)

4(4-5)

5(4-5)

5(4-5)

1(0-1)

1(0-1)

2(2-3)

2(2-3)

3

3(3-4)

4(3-4)

4(3-4)

nigrescens

3(2-4)

3(2-4)

4

4

4(4-5)

4(4-5)

5(4-5)

5(4-5)

I

1(1-2)

3

3(3-4)

3-4

3(3-4)

4(3-5)

4(3-5)

intermedius

2(1-2)

2

4(3-4)

4(3-4)

4(4-5)

4(4-5)

5(4-5)

5(4-5)

1(1-2)

1(1-2)

2(2-3)

2(2-3)

3(3-4)

3

4(3-5)

4(3-4)

decipiens*

2_

2

3-4

3-4

4

4-5

4-5

4-5

1

1-2

2-4

2

3-4

3-4

4

3-4

minckleyi*

2_

J_

_3_

3-4

1

I

2

2

3

3

3

3

*Only the range of observations is given due to the small sample sizes (i.e., n<10).

258

THE JOURNAL OF ARACHNOLOGY

Table 4. Morphometric and meristic characteristics (mean and standard deviation) of Vaejovis inter-
medius Borelli and V. nigrescens Pocock. For V. intermedins, morphometries are based on 20 $$ and
20 9$; lor nigreseens, 8 SS and 13 9$- Pectinal tooth counts are taken from 36 SS and 62 9$ F in-
termedins', and 4 and 23 9$ of F nigrescens.

Character

V. intermedins

males females

V. nigrescens
males females

Slenderness of pedipalp chela

1. chela length width

T

3.78

4.35

3.94

4.54

s

0.20

0.25

0.31

0.29

2. chela length/depth

X

3.10

3.70

3.70

4.04

s

0.21

0.16

0.32

0.31

Relative length of pedipalp

3. fixed finger length /carapace length

X

0.79

0.81

0.84

0.88

s

0.02

0.03

0.01

0.04

4. femur length/ carapace length

X

0.91

0.88

1.02

0.99

s

0.02

0.03

0.03

0.03

Slenderness of metasoma

5. 1 length width

X

0.81

0.76

0.81

0.75

s

0.03

0.02

0.04

0.03

6. HI length/ width

X

1.05

1.00

1.03

0.98

s

0.03

0.04

0.05

0.04

7. V length/ width

X

2.05

2.00

1.95

1.89

s

0.09

0.07

0.08

0.05

Relative length of metasoma

8. carapace length/ met V length

X

0.88

0.90

0.84

0.90

s

0.03

0.03

0.04

0.03

Meristics

9. Pectinal tooth counts

mode

23,24

21

21

18

range

21-26

19-23

19-21

17-21

of the pedipalps and metasoma are light orange brown, rather than dark reddish
brown as in adults.

Sexual dimorphism also occurs in body size and morphometries of the pedipalp
chela and metasoma (see Table 4). Adult males range from 40-48 mm in length;
adult females from 50-57 mm.

Specimens examined. — MEXICO: Dnrango: Dinamita (no date or collector), 1 lectotype male, 2
paralectotype males, 3 paralectotype females (Sc. 508, Ex. 799) (TOR). USA: Texas: Brewster Co.:
NE rim of Chisos Basin, Chisos Mts., Big Bend National Park (56000, 6 Sept 1969 (Cazier, Bigelow),
I male, 1 female (OFF); Chisos Mts., Juniper Canyon, Big Bend National Park, July 1921 (no
collector), 1 female (AMNH); Chisos Mts., The Basin, Big Bend National Park, 1-10 Aug 1937 (K.
P. Schmidt), I female (FMNH); Chisos Mts., The Basin, Big Bend National Park, 6 Aug 1937 (K.
P. Schmidt), 1 female (FMNH); Crockett Co.: 14.3 mi E Sheffield, 11 October 1972 (T. R.
VanDevender), 1 female (FSCA); Presidio Co.: no specific locality, 2 Dec 1978 (under rock) (W. W.
Dalquest), 1 female (OFF); 9 mi E Bandera Mesa, 11 Mar 1975 (W. W. Dalquest), 1 female (OFF);
Terrel Co.: 5 mi N Sanderson, 15 June 1974 (L. Draper, M. A. Cazier, O. F. Francke), 2 females
(OFF); Val Verde Co.: 1 mi SSE Langtry, 7 June 1974 (L. Draper, M. A. Cazier, O. F. Francke),
I female (OFF); 0.5 mi S Langtry, 14 June 1974 (L. Draper, M. A. Cazier, O. F. Francke), 18 males,
19 females (OFF); Langtry, 21 Apr 1973 (no collector), I female (OFF); 10 mi W Comstock, E side
of Pecos River Bridge on US Highway 90 (among large rocks), 2 Sept 1983 (W. D. and J. C. Sissom),
1 female (WDS); Amistad Reservoir, railroad cuts off US Highway 90 (1000"), 6 Sept 1969 (Cazier,
Bigelow), 12 males, 26 females (OFF); Amistad Reservoir, 11 June 1974 (L. Draper, M. A. Cazier,
O. F. Francke), 3 males, 5 females (OFF).

SISSOM AND FRANCKE— yV/r/Z)f/Lt/5'GROUP REDESCRIPTIONS

259

Vaejovis decipiens Hoffmann
(Figs. 34-43)

Vejovis mexicanus decipiens Hoffmann 1931: 349, 399-401; 1939: 318.

Vaejovis mexicanus decipiens, Diaz Najera 1975: 7, 19; nec Vasquez 1960: 219-

221; Williams 1980: 105-107.

Type data. — Hoffmann’s (1931) original description was based on nine
specimens: three males, four females, and two juveniles from Batopilas,
Chihuahua, Mexico (no date or collector). We have been able to locate and
examine only two males permanently deposited in the American Museum of
Natural History. One of these, carrying Hoffmann’s label, “#1, $ type”, is
considered to be the holotype, and the other a paratype.

Both the holotype and paratype are in good condition. However, the holotype
has the left pedipalp loosely articulated at the femur-tibia joint, and leg H loose
at the coxa-trochanter joint. The paratype has the right pedipalp loosely
articulated at the tibia-chela joint, right leg I loose at the trochanter-femur joint,
right leg II loose at the patella-tibia joint, left leg HI loose at the coxa-trochanter
joint, and left leg IV completely detached.

Distribution. — Known only from the type locality.

Diagnosis. — Adults 50-60 mm in length. Base color dark brown to dark reddish
brown with variable underlying dusky markings. Sternite VII with submedian
keels obsolete; lateral keels strong, granular to crenulate. Metasoma with ventral
submedian carinae on I obsolete or weak, smooth; on H-IV weak, smooth with
fine posterior granulation. Pedipalp: tibia with 14 trichobothria (3 et, 1 est, 2 em,
2 esb, 5 eh, 1 v) on external face; fixed finger of chela with primary row of
denticles broken into six subrows by five enlarged denticles. Pectinal tooth count
22-25 in males, 21-22 in females.

V decipiens is most similar to Vaejovis janssi Williams from Isla Socorro
(Revillagigedo Islands). It may be distinguished from V janssi by pectinal tooth
counts of 22-25 in males and 21-22 in females, rather than 21-22 in males and
18-21 in females; and by having the ventral submedian carinae on metasomal
segments H-IV smooth with fine posterior granulation, rather than crenulate.
Morphometric differences cited by Williams (1980) apparently based on
Hoffmann’s (1931) measurements of V. decipiens, are not consistent or cannot be
evaluated due to lack of adult females of V decipiens.

Redescription. — The following redescription is based on adult males.
Measurements of the holotype appear in Table 2.

Coloration. Carapace, tergites, and metasoma brown to dark reddish brown
with variable underlying dusky markings. Pedipalp femur and tibia same color
as body; chela dark reddish brown, usually lighter than body, with yellow brown
fingertips. Chelicerae yellow brown with teeth dark brown. Legs yellow brown to
brown with variable underlying dusky markings. Venter light brown to medium
brown with dusky markings; third sternite with conspicuous yellowish white patch
posteromedially. Pectines yellowish white.

Prosoma. Carapace slightly longer than wide. Median ocular prominence
slightly raised above surface of carapace. Three pairs of lateral eyes, diameter of
posteriormost pair approximately one-half the diameter of preceding pairs.

260

THE JOURNAL OF ARACHNOLOGY

Anterior carapacial margin obtusely emarginate; median notch distinct, shallow,
narrow. Median longitudinal furrow deep, wide at anterior margin; deep, narrow
posteriorly. Posterior lateral furrow curved, deep, wide. Entire carapacial surface
coarsely granular.

Mesosoma. Tergites I-VI with median carina weak, granular; lateral carinae
vestigial. Tergite VII pentacarinate: median carina weak, granular, present on
anterior one-third; submedian and lateral carinae strong, serrate. Genital operculi
distinctly lobed posteriorly; without median longitudinal membranous connection.

Figs. 34-43. — Vaejovis decipiens Hoffmann, holotype male from Chihuahua, Mexico: 34, dorsal
aspect of pedipalp femur; 35, dorsal aspect of pedipalp tibia; 36, external aspect of pedipalp tibia;
37, ventral aspect of pedipalp tibia; 38, lateral aspect of metasomal segments IV and V, and telson;
39, dorsal aspect of pedipalp chela; 40, external aspect of pedipalp chela; 41, ventral aspect of
pedipalp chela; 42, dentition pattern on fixed finger of pedipalp chela; 43, dentition pattern on
movable finger of pedipalp chela.

SISSOM AND FRANCKE— A/r/Z)t/Lt/NGROUP REDESCRIPTIONS

261

Genital papillae well developed. Pectinal teeth numbering 22-25. Sternites III-VI
smooth, agranular; stigmata approximately 2.5 to 3 times longer than wide.
Sternite VII with submedian carinae obsolete; lateral carinae strong, granular to
crenulate.

Metasoma. Segments I-IV longer than wide (except sometimes I). Dorsolateral
carinae strong, serrate; distalmost denticle distinctly enlarged and spinoid (Fig.
38). Lateral supramedian carinae on I strong, serrate; on II-III strong, serrate to
finely serrate; on IV strong, granular to finely serrate; distalmost denticle
distinctly enlarged and spinoid on I-III, flared and winglike on IV (Fig. 38).
Lateral inframedian carinae on I complete, strong, serrate; on II present on
posterior one-third, strong, serrate; on III present on posterior one-fourth, strong,
crenulate to serrate; on IV absent. Ventrolateral carinae on I-III strong, finely
serrate; on IV strong, finely crenulate to finely serrate. Ventral submedian carinae
on I obsolete or weak, smooth; on II-IV weak, smooth, usually with fine
serrations on posterior portion. Intercarinal spaces with scattered coarse granules,
primarily on dorsal surfaces. Segment V (Fig. 38): Dorsolateral carinae strong,
granular. Lateromedian carinae present on anterior three-fourths of segment,
strong, granular to finely serrate. Ventrolateral carinae strong, finely serrate
(sometimes finely crenulate posteriorly). Ventromedian carina strong, finely
crenulate to finely serrate. Intercarinal spaces essentially agranular.

Telson (Fig. 38). Dorsal surface smooth, flattened; ventral surface with
irregularly spaced fine granules and punctations.

Chelicera. Dentition typical of genus, with well developed serrula on ventral
margin of movable finger.

Pedipalp. Femur (Fig. 34) tetracarinate: Dorsointernal and ventrointernal
carinae strong, serratocrenulate. Dorsoexternal carina strong, serrate.
Ventroexternal carina strong, composed of large, irregularly spaced sharp
granules. Internal face with large, scattered, conical granules; ventral face with
moderate granulation basally; dorsal face smooth. Orthobothriotaxia C (Vachon
1974).

Tibia (Figs. 35-37) tetracarinate: All keels strong, crenulate to serratocrenulate.
Internal face with moderate basal tubercle plus series of large, conical granules;
external face granular; dorsal and ventral faces smooth. Orthobothriotaxia C
(Vachon 1974). Trichobothrium errii basal to or in juxtaposition with em\.

Chela (Figs. 39-43). Dorsal marginal carina strong, granular. Dorsal secondary
carina weak, smooth to finely granular. Digital carina moderate, smooth to finely
granular. External secondary carina weak, smooth to finely granular.
Ventroexternal carina moderate, granular. Ventromedian carina weak, smooth.
Ventrointernal carina weak to moderate, finely granular. Dorsointernal carina
strong, composed of large, sharp granules. Dentate margin of fixed finger with
primary row divided into six subrows by five larger granules; six internal
accessory granules, of which distalmost not paired with larger granule in primary
row (Fig. 42). Dentate margin of movable finger with primary row divided into
six subrows by five larger granules; distal subrow with only one or two granules;
seven internal accessory granules, of which distalmost and basalmost not paired
with larger granule in primary row (Fig. 43). Both chela fingers terminating in
large, sharp, clawlike tooth bearing distally an oblong whitish cap.
Orthobothriotaxia C (Vachon 1984).

262

THE JOURNAL OF ARACHNOLOGY

Legs. Arrangement of setae, spines, and spinules as in V. nitidulus; seta formula
given in Table 3.

Variation. — Male pectinal tooth counts varied as follows: 1 comb with 22 teeth;
3 combs with 23 teeth; 3 combs with 24 teeth; 1 comb with 25 teeth. Only a single
female (a juvenile) was available for study; its pectinal tooth count is 21-21.
Hoffmann (1931) reported pectinal tooth counts of 22 for the females he studied.

Juveniles differ from adults in general coloration. In juveniles the body is
yellow brown with dusky underlying markings: the tibia and chela of the pedipalp
are orange brown; the metasoma is yellow brown basally (segments I-III) and
gradually becomes orange brown distally (segments IV-V); the telson is orange
brown.

Remarks. — Hoffmann (1931), in his original description, noticed the similarity
between V. decipiens and the V. nitidulus group, but was reluctant to group it
with those taxa because its ventral submedian keels were distinctly developed (not
obsolete). As a result, he considered it to be a subspecies of V. mexicanus. In
possessing the enlarged terminal denticles on the pedipalp chela fingers (each with
a distinct whitish cap), it is clear that V. decipiens is a member of the nitidulus
group and not a subspecies of V. mexicanus.

Specimens examined. MEXICO; Chihuahua: Batopilas (no date or collector), 1 male holotype
(labeled “#l, male type” by Hoffmann), 1 male paratype (AMNH); Barranca de Rio Batopilas, 120
km S Creel (approx. 1000 m), 26 Feb. 1966 (W. Bell, J. Reddell), 2 males, 3 juvs. (AMNH).

Vaejovis minckleyi Williams
(Figs. 44-53)

Uc/ovA mmcA'/cW Williams 1968; 21-24, figs. 11-12; Soleglad 1972; 180, 1973;357.
Paruroctonus minckleyi, Stahnke 1974; 138.

Type data. — Holotype male from 5.3 km NW Cuatro Cienegas, Coahuila,
Mexico, 3 January 1965 (W. S. Parker). Paratype male from second canyon from
the eastern tip of San Marcos Mountain, 12 km SW Cuatro Cienegas, 3 January
1965 (W. L. Minckley, W. S. Parker, and W. K. Taylor). Deposited in the
California Academy of Sciences; not examined.

Distribution. — Vaejovis minckleyi is apparently endemic to the Cuatro Cienegas
Basin, Coahuila, Mexico. This remarkable intermontane valley in the Sierra
Madre Oriental contains a highly diversified biota with a conspicuous percentage
of endemic taxa, a considerable portion of which are Pleistocene relicts or older
(Minckley 1969).

Diagnosis. — Adults 60-70 mm in length. Base color yellow; pedipalp chelae
yellow orange with reddish brown fingers. Sternite VII tetracarinate: submedian
keels faint, smooth; lateral keels strong, crenulate. Metasoma: all segments
distinctly longer than wide; ventrolateral carinae on I-IV strong, serrate; ventral
submedian carinae on I-III weak to vestigial, smooth, on IV weak, granular.
Distalmost denticle of dorsolateral carinae of I-IV not distinctly larger than
others or spinoid (Fig. 48). Pedipalp (Figs. 44-47; 49-53): tibia with 15
trichobothria (3 et, 1 est, 2 cm, 3 esb, 5 eh, 1 v) on external face (Fig. 46): fixed
finger of chela with primary row of denticles broken into six subrows by five
enlarged denticles (Fig. 52); fixed finger distinctly longer than carapace; keels of

SISSOM AND FRANCKE— A/r/Df/Lf/^GROUP REDESCRIPTIONS

263

dorsal and external surfaces of chela strong in males (moderate in females),
granulose (Figs. 49-51). Pectinal tooth count 31-32 in males, 28-29 in females.

Vaejovis minckleyi is a very distinctive member of the nitidulus group. In
pectinal tooth counts and tibial trichobothrial pattern, it most closely resembles
V. nitidulus. However, it may be easily distinguished from that species by having
six subrows of granules on the chela fingers rather than seven subrows; by having
distinct ventral submedian carinae on metasomal segments I-IV, rather than

Figs. 44-53. — Vaejovis minckleyi Williams, male from Coahuila, Mexico: 44, dorsal aspect of
pedipalp femur; 45, dorsal aspect of pedipalp tibia; 46, external aspect of pedipalp tibia; 47, ventral
aspect of pedipalp tibia; 48, lateral aspect of metasomal segments IV and V, and telson; 49, dorsal
aspect of pedipalp chela; 50, external aspect of pedipalp chela; 51, ventral aspect of pedipalp chela;
52, dentition pattern on fixed finger of pedipalp chela; 53, dentition pattern on movable finger of
pedipalp chela.

264

THE JOURNAL OF ARACHNOLOGY

obsolete carinae; by having all metasomal segments longer than wide; and by
having granulose carinae on the dorsal and external surfaces of the chela, rather
than weak, smooth carinae.

Variation. — Only one male and one female (both topotypes) were studied.

Remarks. — Stahnke (1974) considered V. minckleyi to be a member of the
genus Paruroctonus, apparently because the distalmost denticles of the
dorsolateral carinae of metasomal segments I-IV are not enlarged and spinoid
(Fig. 48). We interpret this condition to result from the elongation of the
metasomal segments (as seen in other nitidulus group species and also in
Syntropis), and it does not indicate phylogenetic affinity with Paruroctonus. In
addition, V. minckleyi lacks denticles or scallops on the inferior margin of the
movable cheliceral finger and has the anterior margin of the carapace distinctly
notched. Both of these characteristics exclude it from Paruroctonus. The structure
of the pedipalp chelae and the trichobothrial pattern of the tibia and chela,
however, indicate that V. minckleyi is a member of the nitidulus group of
Vaejovis.

Specimens examined. — MEXICO: Coahuila; Cuatro Cienegas Basin, large canyon '/^mi E of W tip
of Sierra San Marcos, 12 May 1968 (S. C. Williams, M. Bentzien), 1 male, 1 female (OFF).

ACKNOWLEDGMENTS

Numerous curators and institutions provided assistance during the course of
this study, either through loans of specimens or by searching through their
collections for pertinent material.

for the loan of type specimens, we are deeply grateful to the following: the
types of V. intermedius, O. Elter, Museo ed Istituto di Zoologia Sistematica della
Universita di Torino, Torino, Italy (TOR); the types of V. nitidulus, M. Moritz,
Zoologisches Museum, Humboldt Universitat, Berlin (ZMB); the types of V.
mexicanus decipiens, Normal I. Platnick, the American Museum of Natural
History, New York (AMNH); and the types of V. nigrescens, F. R. Wanless and
P. D. Hillyard, The British Museum (Natural History), London (BM). These
curators and institutions also gave needed extensions of the loans, and for this,
we are also grateful.

Drs. Elter (TOR), Platnick (AMNH), and Wanless and Hillyard (BM) allowed
us to examine other nitidulus group material used in this study. Others providing
such assistance were Larry Watrous, J. B. Kethley, and D. A. Summers, the Field
Museum of Natural History, Chicago (FMNH); G. B. Edwards, Florida State
Collection of Arthropods, Gainesville (FSCA); H. W. Levi, Museum of
Comparative Zoology, Harvard University, Cambridge, Massachusetts (MCZ);
Max Vachon and Wilson R. Louren^o, Museum National d’Histoire Naturelle,
Paris (MNHN); Sherman A. Minton, Indiana University School of Medicine
(SAM); and Matt E. Braunwald (MEB). We thank them for their courtesy and
assistance.

We also wish to thank several individuals who responded to our requests for
specimens, although their respective collections did not have any pertinent
material: V. F. Lee, W. J. Pulawski, and S. C. Williams, the California Academy
of Sciences, San Francisco; C. E. Griswold, the Division of Entomology and

SISSOM AND FRANCKE— A/r/Z){/Lfy^GROUP REDESCRIPTIONS

265

Parasitology, University of California, Berkeley; W. H. Hutchinson, University of
Mississippi Medical Center, Jackson, for information about the collection of the
late H. L. Keegan; James R. Reddell, Texas Memorial Museum, the University
of Texas, Austin; C. Carlton, University of Arkansas Agricultural Experiment
Station; and M. Alvarez del Toro, Instituto de Historia Natural, Tuxtla
Gutierrez, Chiapas, Mexico.

For correspondence and cooperation during the course of this study, we wish
to acknowledge Max Vachon and M. E. Soleglad. James C. Cokendolpher
provided stimulating taxonomic discussions for the senior author, and his
comments are sincerely appreciated. For reviewing the manuscript we are grateful
to M. E. Soleglad and S. C. Williams. To the friends and colleagues who assisted
in the field, we express our warmest appreciation: Mont A. Cazier, Linda Draper
(neeWelch), Joe Bigelow, Neal McReynolds, Chris Colwell, Chris Myers, Lynne
Born, and John Sissom.

REFERENCES CITED

Borelli, A. 1915. Scorpion! nuovi o poco noti del Messico. Boll. Mus. Zool. Anat. Comp. Univ.
Torino, 30(703): 1-7.

Biicherl, W. 1959. Escorpioese escorpionismo no Brasil. X. Catalogo da cole^ao escorpionica do
Instituto Butantan. Mem. Inst. Butantan, 29: 255-275.

Biicherl, W. 1971. Classification, biology, and venom extraction of scorpions. Pp. 317-347 In
Venomous animals and their venom. III. (W. Biicherl and E. E. Buckley, eds.). Academic Press,
New York.

Diaz Najera, A. 1964. Alacranes de la Republica Mexicana: Identificacion de ejemplares capturados
en 235 localidades. Rev. Inst. Salubr. Enferm. Trop., Mexico, 24: 12-30.

Diaz Najera, A. 1975. Listas y datos de distribucion geografica de los alacranes de Mexico
(Scorpionida). Rev. Inst. Salubr. Enferm. Trop., Mexico, 35: 1-36.

Gertsch, W. J. 1958. Results of the Puritan-American Museum expedition to western Mexico. 4. The
scorpions. Amer. Mus. Notitates, No. 1903, pp. 1-20.

Hoffmann, C. C. 1931. Los Scorpiones de Mexico. Primera parte: Diplocentridae, Chactidae,
Vejovidae. An. Inst. Biol., Mexico, 2:291-408.

Hoffmann, C. C. 1937. Nota acerca de los alacranes del Valle del Mezquital, Hgo. An. Inst. Biol.,
Mexico, 8: 201-206.

Hoffmann, C. C. 1939. Nuevas consideraciones acerca de los alacranes de Mexico. An. Inst. Biol.,
Mexico, 9: 317-337.

Koch, C. L. 1843. Die Arachniden. X. Niirnberg.

Kraepelin, K. 1894. Revision der Scorpione. II. Scorpionidae und Bothriuridae. Jahrb.
Hamburgischen Wiss. Anst., 1 1: 1-248.

Kraepelin, K. 1899. Scorpiones und Pedipalpi. Das Tierreich, 8:1-265.

Kraepelin, K. 1901. Catalogue des scorpiones des collections du Museum d’Histoire Naturelle de
Paris. Bull. Mus. Paris, 7: 265-274.

Minckley, W. L. 1969. Environments of the Bolson of Cuatro Cienegas, Coahuila, Mexico, with
special reference to the aquatic biota. Univ. Texas El Paso Sci. Ser., 2:1-65.

Moritz, M. and S. Fischer. 1980. Die Typen der Arachniden-Sammlung des Zoologischen Museum
Berlin. III. Scorpiones. Mitt. zool. Mus. Berlin, 56:309-326.

Peters, W. 1861. Ubereine neue Eintheilung der Skorpione. Monber. Ak. Berlin, 1861: 507-516.

Pocock, R. I. 1898. The scorpions of the genus Vaejovis contained in the collection of the British
Museum. Ann. Mag. Nat. Hist., ser. 7, 1: 394-400.

Pocock, R. I. 1902. Arachnida: Scorpiones, Pedipalpi, and Solifugae. Biologia Centrali-Americana.
Francis and Taylor, London. 72 pp.

266

THE JOURNAL OF ARACHNOLOGY

Soleglad, M. E. 1972. Two new scorpions of the wupatkiensis group of the genus Vejovis
(Scorpionida: Vejovidae). Wasmann J. Biol., 30: 179-195.

Soleglad, M. E. 1973. Scorpions of the mexicanus group of the genus Vejovis (Scorpionida,
Vejovidae). Wasmann J. Biol., 31:351-372.

Soleglad, M. E. 1975. A redescription of Vaejovis gracilis Gertsch & Soleglad based on the adult
(Scorpionida, Vejovidae). Wasmann J. Biol., 33:107-120.

Stahnke, H. L. 1974. ' Revision and keys to the higher categories of Vejovidae (Scorpionida). J.
Arachnol., 1: 107-141.

Thorell, T. 1876. Etudes scorpiologiques. Atti Soc. Italiana Sci. Nat., 19: 75-272.

Vachon, M. 1974. Etude des caracteres utilises pour classer les families et les genres de Scorpions
( Arachnides). 1. La trichobothriotaxie en Arachnologie. Sigle trichobothriaux et types de
trichobothriotaxie chez les Scorpions. Bull. Mus. Nat. Hist. Nat., Paris, 3e Ser., No. 140, Zook,
104:857-958.

Vasquez, G. L. 1960. La Isla Socorro, Archipielago de Las Revillagigedo. 10. Observaciones sobre
los Artrdpodos. Monogr. Inst. Geofisica, Univ. Nac. Autonoma Mexico, 2: 219-221.

Williams, S. C. 1968. Scorpions from northern Mexico: Five new species of Vejovis from Coahuila,
Mexico. Occas. Pap. California Acad. Sci., 68:1-24.

Williams, S. C. 1970a. Scorpion fauna of Baja California, Mexico: Eleven new species of Vejovis
(Scorpionida: Vejovidae). Proc. California Acad. Sci., 37:275-331.

Williams, S. C. 1970b. New scorpions belonging to the Eusthenura Group of Vejovis from Baja
California, Mexico (Scorpionida: Vejovidae). Proc. California Acad. Sci., 37:395-418.

Williams, S. C. 1971. New and little known scorpions belonging to the Punctipalpi Group of the
genus Vaejovis from Baja California, Mexico, and adjacent areas (Scorpionida: Vaejovidae).
Wasmann J. Biol., 29:37-63.

Williams, S. C. 1972. Four new scorpion species belonging to the genus Paruroctonus (Scorpionida:

Vaejovidae). Occas. Pap. California Acad. Sci., 94: 1-16.

Williams, S. C. 1980. Scorpions of Baja California, Mexico, and adjacent islands. Occas. Pap.
California Acad. Sci., 135: 1-127.

Manuscript received June 1984, revised November 1984.

1985. The Journal of Arachnology 13:267

RESEARCH NOTES

STICKY BALLS IN WEBS OF THE SPIDER
MODISIMUS SP. (ARANEAE, PHOLCIDAE)

Sticky or viscid balls have been found on thread of the webs of araneid, theridioso-
matid, anapid, symphytognatliid, theridiid, nesticid and linypliiid spiders; these families
are all in the superfamily Araneoidea, and it has recently been proposed that the balls
represent a synapomorphy for this group [Coddington, J. in press, In: Orb Webs (W.
Shear, ed.). Stanford University Press; see this article also for a review of evidence].

Tliis note documents the presence of balls in the webs oi Modisimus sp. of the family
Pholcidae, which is not related to the araneoids. It also shows that they are liquid and
water-soluble and that they are produced during a particular stage in web construction.

Web construction was elicited by damaging webs. Some samples (“controls”) were
taken from undisturbed webs (presumably built the night before), whereas others were
taken after interrupting spiders which were at various stages of web construction. The
edges of a glass microscope slide were covered with vaseline, and the slide was held against
one sector of the web which was cut free from the rest with scissors. Only one sample was

Fig. 1. -Photograph of a hne with globules, from a mature female of Modisimus sp.

1985. The Journal of Arachnology 13:268

taken per web. The threads were observed under a compound microscope, in some cases
several days after the web was built.

Voucher specimens of the spiders are deposited in the collection of the Escuela de
Biologia of the Universidad de Costa Rica, and in the Museum of Comparative Zoology in
Cambridge, Mass.

In order to detennine the relative numbers of different kinds of lines present on each,
five surveys were made from left to right across the entire width of the slide at a magnifi-
cation of 40X. Each time a new line was encountered it was placed at the extreme left of
the field and all lines present in the field were classified into four categories: (a) thick
without baUs; (b) thin without balls; (c) thick with balls; and (d) thin with balls (see Fig.
1). The slide was then moved on until the thread at extreme right was at the left of the
field. Then the slide was moved onward until the next thread was encountered, and the
process was repeated. A total of 14 sample fields was examined in each pass across
the slide, giving a total of 70 samples (5 x 14) per slide.

At least four types of line were seen in finished webs. In some cases an apparently
“thick” line split into thin lines, so “thick” means, at least sometimes, several thin lines
together.

The spiders’ construction behavior had tow distinct stages: (1) “ Frame lines.” The
spider laid most lines at the edge of the web, expanding the area covered. Every attach-
ment of the line being laid to other lines was made just posterior to the point where one
leg III held it. (2) “Fill In.” The spider walked back and forth across the area already
covered, using its legs IV to push the line it was producing upward against the network of
lines already in place. Only when it reached the edge of the web did it attach as above,
using one leg III and touching the other line with its spinnerets. Once stage 2 had begun
the spider did not return to stage 1 behavior. The lines laid in two stages were different,
thick lines without balls being more common in stage 1 (Table 1).

The liquid nature of the balls was demonstrated by pressing lines with balls into
contact with the slide, and noting that they spread out to fomi puddles.

The balls were shown to be water soluble just like those of at least some araneoid
spiders (Kavavaugh, E. S. and E. K. Tillinghast. 1979. J. Morph., 160:17-31) by placing
drops of water on slides; all of the balls were gone from the threads when the droplet
evaporated.

The wrapping thread of some theridiid spiders is covered with viscous balls, but
samples of wrapping thread of Modisimus sp. showed no balls.

The liquid nature of the balls and the fact that they had not evaporated even days
after being produced suggest that they are viscous and function by causing prey to adhere

Table 1. -Percentage of types of threads present during different stages of web construction in
webs of Modisimus sp.

Thin Thread

Thick Thread

with

without

with

without

balls

balls

balls

balls

Stage 1 . (<25 lines)

0

15.0

0

85.0

Stage 2. (>25 lines)

40.7

48.7

6.9

3.7

Web repair completed

3.8

73.4

3.5

29.4

Completed webs (control)

41.8

48.4

4.9

4.9

1985. The Journal of Arachnology 13:269

to spider’s web. The stickiness is only slight however, and attempts to localize ball-bearing
threads in completed webs by dusting the webs with talcum powder and then jarring
the web to knock the powder from non-sticky lines resulted in the powder being dis-
lodged from nearly all the threads. The relative small sizes of the balls and their often
relatively dispersed nature may account for this. The attack behavior oi Modisimus sp. is
extremely rapid, and perhaps only relatively brief retention is necessary to insure prey
capture.

The prevalence of thick lines in the first stage of web buildings suggests that this stage
serves to establish a scaffolding or frame for the rest of the web. The high frequency of
thin threads both with and without balls in finished webs suggest that the webs may trap
prey by entanglement as well as by adhesion.

The reason for the lower frequency of sticky balls in repaired webs as compared to
controls is not clear. Possibly it is related to a lack of material from which sticky balls are
made, or the control webs may contain threads that have accumulated over a period of
days.

The question of whether the pholcid balls are homologus with those in araneoid webs
cannot be answered at the moment. The pholcid Pholcus phalangioides is known to
possess two pairs of ampullate (non-sticky silk) glands (Kovoor, J. 1977. Ann. Biol.,
16:7-171) but their other silk glands are difficult to homologize with the silk glands of
other spiders (Apstein, C. 1889. Arch. Naturgesch., 55:29-74) so the glandular source of
the balls cannot be detennined.

I thank Yamilet Brenes and Edgar Goitia for help with photography. I want to give my
special thanks to William Eberhard and P. M. Brignoli for comments and suggestions in
the elaboration of this paper.

R. Daniel Briceno., Escuela de Biologia, Universidad de Costa Rica, Cuidad Universi-
taria Rodrigo Facio, Costa Rica.

Manuscript received September 1982, revised July 1984.

ARBORINE AND METHAQUALONE ARE NOT SEDATIVE IN THE
WOLF SPIDER LYCOS A CERATIOLA GERTSCH AND WALLACE

Glomerin and homoglomerin, two quinazolinone alkaloids in the defensive secretion of
the pill milliped, Glomeris marginata, produce delayed sedation of prolonged duration in
wolf spiders (Lycosa spp.). The compounds are sedative at small doses (1-7 pg per spider),
representing but a fraction of the total secretory output of a medium sized milliped
(Carrel and Eisner 1984). Glomerin and homoglomerin are structurally related to ar-
borine, a plant natural product, and to methaqualone, a synthetic drug. Both arborine
and methaqualone are sedative to vertebrates (Dey and Chatterjee 1967, Inaba et al.
1973), which suggests that they might also be sedative to spiders. We here present evi-
dence indicating that this hypothesis is incorrect, since neither arborine nor methaqua-
lone given in large doses produced sedation (= hypnosis) in the wolf spider, Lycosa
ceratiola Gertsch and Wallace.

1985. The Journal of Arachnology 13:270

Arborine, also known as glycosine, was synthesized using the procedure described by
Kametani et al. (1977). Chromatography of the reactant residue on a silica gel column
eluted with ethylacetate-ethanol (4:1, v/v) yielded pure arborine, whose melting point
and UV, IR, NMR, and mass spectral data were identical with published values (Chak-
ravarti et al. 1961, Kametani et al. 1977). Methaqualone hydrochloride (Parest-200®,
Parke-Davis and Co.-, Detroit, Michigan) was locally purchased. Dosages of methaqualone
were calculated as the HCl-free base (0.875 times the weight of the methaqualone-HCl).

Lycosa ceratiola, stemming from the same population at the Archbold Biological
Station near Lake Placid, Florida, as those used to test glomerin and homoglomerin, were
maintained as described by Carrel and Eisner (1984). As before, spiders were of relatively
uniform body size (x ± S.E.M.;body mass = 344 ±13 mg).

Tlie sedative potencies of arborine and methaqualone in spiders were measured by
injection. In these tests (N = 140, including 20 controls), essentially the same as in those
for glomerin and homoglomerin (Carrel and Eisner 1984), injection (5 jul) was accom-
plished with a micrometer-activated syringe into the abdomen of the spider. Arborine and
methaqualone as sonified suspensions were injected at six dosages (N = 10 spiders per
dosage) in the range of 1-50 {ig per spider. Spider saline (Rathmayer 1965) containing 1%

(w/v) methylcellulose was used as sample carrier and was itself tested as the control.
Spiders were checked for sedation at 4, 12, 24, and 48 hours after injection. The criterion
for sedation, as in earlier tests, was the spider’s inability to right itself promptly when
flipped on its back with a curved glass rod.

None of the L. ceratiola became sedated and none died within 48 hours after injection
of 1-50 jUg of arborine or methaqualone. Control spiders also showed no behavioral
abnormalities. Maximum doses of either compound used in our study 150 mg/kg
spider body weiglit) were large compared with doses of these compounds that are seda-
tive/ hypnotic in humans, mice, and rats (Gujral et al. 1955, Swift et al. 1960, Wheeler
1963, Dey and Chatterjee 1967, Muklierjee and Dey 1970, Hardtmann et al. 1971, Ochiai
et al. 1972, American Medical Association 1980). Elence, the absence of sedation in wolf
spiders treated with arborine or methaqualone definitely did not result from using dosages of the
compounds below their established phannacological ranges.

Our findings cannot be explained by a general insensitivity of spiders to sedative drugs.
Phenobarbital (Luminal) and diazepam (Valium), two sedatives commonly used by
humans, each in low doses (10-100 mg/kg) cause the cross-spider, Araneus diadematus, to
curtail construction of its orb web (Reed and Witt 1968). Both phenobarbital and dia-
zepam bear little structural resemblance to the quinazolinones we tested and preliminary
evidence indicates that, at least in mammals, they act via different neurobiochemical
mechanisms to depress the central nervous system (Mukherjee and Dey 1970, Smith
1977, Martin 1982). We think the ineffectiveness of some, but not all sedative drugs in
spiders likely is explained by basic and as yet undescribed neurophysiological differences
between spiders and mammals more than it is by the vagaries of various drug bioassays.

Our study confirms the long standing view of pharmacologists (Witt 1968, 1971) that
spiders are imperfect substitutes for humans for the characterization of psychoactive
drugs. Our findings also illustrate how little is known about the short and long term
responses of spiders to pure substances, especially natural products contained in their
prey. We suspect that many dietary chemicals may alter a spider’s physiological state,
causing changes— ranging from profound to subtle ones— in feeding, reproduction, or
maintenance activities. The chemical ecology of spiders, an emerging field of study, is
bound to be diverse and complex, perhaps rivaling that of herbivorous insects, about
which so much has recently been written (Rosenthal and Janzen 1979, Harborne 1982).

1985. The Journal of Arachnology 13:271

This work was supported in part by grants from the University of Missouri Research
Council and the National Institutes of Health. We thank the Archbold Biological Station
for the use of their facilities and David Richman for identifying the spiders.

LITERATURE CITED

American Medical Association. 1980. Methaqualone hydrochloride (Parest). P. 159, In: AMA Drug
Evaluations (4th Ed.). American Medical Association, Chicago.

Carrel, J. E. and T. Eisner. 1984. Spider sedation induced by defensive chemicals of milliped prey.
Proc. Natl. Acad. Sci., U.S.A., 81:806-810.

Chakravarti, D., R. N. Chakravarti, L. A. Cohen, B. Dasgupta, S. Datta, and H. K. Miller. 1961. Alka-
loids of Glycosrnis arborea - II Structure of arborine. Tetrahedron, 16:224-250.

Dey, P. K. and B. K. Chatterjee. 1967. Effect of glycosine on higher nervous activity. Naturwissen-
schaften, 54:21-22.

Gujral. M. E., P. N. Saxena, and R. S. Tiwari. 1955. Comparative evaluation of quinazolinones: a new
class of hypnotics. Indian J. Med. Res., 43:637-641.

Harborne, J. B. 1982. Introduction to Ecological Biochemistry (2nd Ed.). Academic Press, London.
278 pp.

Hardtmann, G. E., B. S. Huegi, J. H. Gogerty, E. C. lorio, and H. W. Barnes. 1971. Tetracyclic quina-
zolinone derivatives. J. Med. Cheni., 14:878-882.

Inaba, D. S., G. R. Gay, J. A. Newmeyer, and C. Whitehead. 1973. Methaqualone abuse, “luding out”.
J. Amer. Med. Assoc., 224:1505-1509.

Kametani, T., C. V. Loc, T. Higa, M. Koizumi, M. Ihara, and K. Fukumoto. 1977. Iminoketene cyclo-
addition. 2. Total syntheses of arborine, glycosminine, and rutecarpine by condensation of im-
inoketene with amides. J. Amer. Chem. Soc., 99:2306-2309.

Martin, W. R. 1982. Hypnotics. Pp. 305-329, In: Psychotropic Agents. Part III. Alcohol and Psycho-
mimetics. Psychotropic Effects of Central Acting Drugs (F. Hoffmeister and G. Stille, eds.).
Springer-Verlag, Berlin.

Mukherjee, A. and P. K. Dey. Changes in amino acid metabolism in rat brain following glycosine
administration. Indian J. Exper. Biol., 8:263-265.

Ochiai, T., R. Ishida, S. Nurimoto, 1. Inoue, and Y. Kowa. 1972. Pharmacology of a new hypnotic
drug: HQ-355. Japanese J. Pharmacol., 22:431434.

Rathmayer, W. 1965. Polyneuronale Innervation bei Spinnen. Naturwissenschaften, 52:114.

Reed, C. F. and P. N. Witt. 1968. Progressive disturbance of spider web geometry caused by two
sedative drugs. Physiol. Behav., 3:119-124.

Rosenthal, G. A. and D. H. Janzen, eds. 1979. Herbivores: Their Interaction with Secondary Plant
Metabolites. Academic Press, New York. 718 pp.

Smith, C. M. 1977. The pharmacology of sedative/hypnotics, alcohol, and anesthetics: Sites and
mechanisms of action. Pp. 423-527, In: Drug Addiction. 1. Morphine, Sedative/Hypnotic, and
Alcohol Dependence (W. R. Martin, ed.). Springer-Verlag, Berlin.

Swift, J. G., E. A. Dickens, and B. A. Becker. 1960. Anticonvulsant and other pharmacological activi-
ties of Tuazolone (2-methyl-3-o-tolyl4(3H) quinazolinone). Arch. Int. Pharmacodyn., 128:112-
125.

Wheeler, K. W. 1963. Non-barbiturate hypnotics. Pp. 1-245, In: Medicinal Chemistry, Vol. 6. (E. E.
Campaigne and W. H. Hartund, eds.). J. Wiley and Sons, New York.

Witt, P. N. 1968. The story of the drug web. Pp. 14, In: A Spider’s Web, Problems in Regulatory
Biology (P. N. Witt, C. F. Reed, and D. B. Peakall, eds.). Springer-Verlag, New York.

Witt, P. N. 1971. Drugs alter web-building of spiders. Behav. Sci., 16:98-113.

James E. Carrel, Division of Biological Sciences, James P. Doom, Department of
Chemistry, and John P. McCormick, Department of Chemistry, University of Missouri,
Columbia, Missouri 65211.

Manuscript received May 1 984, revised August 1 984.

1985. The Journal of Arachnology 13:272

PARASITIC FUNGI AS A MORTALITY FACTOR OF SPIDERS

Several times during arachnological studies in Panama I found dead spiders covered
with fungi. As the notes which record fungi attacking spiders are scattered and few in
number, my observations are reported here in detail.

The field studies were carried out in Panama (Province of Panama) at three locations:
On Pipeline Road near Gamboa; on Cerro Galera, a hill at the Pacific coast near Arraijan;
on the road from El Llano to Carti, km 13. The first two areas represent typical tropical
moist lowland forest, the third premontane moist forest. In Panama, there is a distinct
seasonality with a dry season (January to April), and a wet season (May to December).
Annual rainfall is more or less restricted to the wet season and amounts to 2000-3000
mm per year. Fungus-covered spiders were found only from August to October, i.e. in the
second and wetter part of the wet season, though field-work was carried out during
the whole year.

Immature specimens of the deuteromycete Nomuraea, probably N. atypicola (Ya-
suda), were found on the dirdnQ\<\s Argiope argentata (F.) (three observations), .4. savignyi
Levi (one observation) and Nephila clavipes (L.) (two observations). A compact white
mycelial stroma was observed on the opisthosoma of the spiders but fruiting structures
(conidi ophores) were lacking (Fig. 1). Fungal colonies are initially white but typically
become purple as a powdery spore bloom develops. The heavy overgrowth of a mucorace-
ous fungus (“bread mould”) is characteristic of recently invaded or badly dried immature
(= non-mummified) specimens (Evans, pers. comm.). On the araneid Eriophora fuliginea
(C. L. Koch) and an unidentified web building spider (one observation each) the fungus
overgrowth was also immature but could tentatively be identified as the Granulomanus
state of a Gibellula sp. (Deuteromycetes).

All araneids were found hanging with their legs in a relatively normal position on the
hub of the orb web. The web, however, was always reduced to a silken platform with a
network of irregular threads, similar to the moulting webs of orbweavers (Fig. 2). Because
old spiders build orb webs of increasing irregularity but never without sticky spiral
(Nentwig, unpubl.) this web type indicates that the fungi probably did not attack a dead
spider but rather killed a living one. The period between infection and death of the spider
must have been at least two days to enable the spiders to build this specific type of web.
It is not possible that these webs represented real moulting webs since no exuviae were
found and all the spiders were full-sized adult females.

An overview of insect-parasitism among fungi has been presented by Madelin [1968,
pp. 227-238, In: The Fungi (G. Ainsworth and A. Sussmann, eds.). Academic Press, New
York, Vol. 3] . Most ectoparasites are Laboulbeniales (Ascomycetes), only a few dozen
are Deuteromycetes, most records originate from tropical and subtropical countries.
Tliese include the Nomuraea (= Spicularia) and Gibellula species which are mentioned
here and are well-known to parasitise spiders [Samson and Evans 1977, Proc. Konink.
Nederland. Akad. Wet., Amsterdam, ser. C, 80(2):120-133] . Further records of fungi
which infect spiders include Engyodontium species (Hyphomycetes) (Gams, De Hoog and

1985. The Journal of Arachnology 13:273

Fig. 1,-A dead Argiope argentata (Araneidae) covered with the white mycelial stroma of the
deuteromycete Nomuraea cf. atypicola.

Fig, 2.— The web of a dead A. argentata, infected by N. cf. atypicola, is reduced to a silken plat-
form and irregular suspension threads.

1985. The Journal of Arachnology 13:274

Samson 1984, Persoonia, Leiden, 12:135-147) and the ascomycete Cordyceps sp. which
had been found on dead linyphiids on the arctic island Jan Mayen (Bristowe 1941,
The comity of spiders, London, The Ray Society, vol. 2:332-333). In India, the endo-
parasitic deuteromycete Beauveria alba has been found in a theridiid (Chandrashekhar,
Suryanarayanan and Narasimham 1981, Curr. Sci., Bangalore, 50:248). Apart from the
last two cases, none of the several dozen records in the mycological literature mentioned
here gives an identification of the spiders, with the exception of one reference to an
ant-mimicking Salticidae by Samson and Evans (1977) and one reference to an “Opilionid
spiders” by Gams et al. (1984). It is possible that the exposed posture of the araneids
mentioned here facilitates infection by airborn fungus spores. This indicates perhaps that
entomophagous fungi are an important mortality factor among spiders, especially web
building species in the tropics.

I thank Dr. H. C. Evans, Commonwealth Mycological Institute, Kew, England, for the
identification of the fungi and for helpful comments.

Wolfgang Nentwig, FB Biologie-Zoologie, Universitat, D-3550 Marburg, Federal
Republic of Germany.

Manuscript received July 1984, revised August 1 984.

NOMENCLATURAL NOTES

On 2 April 1985 the Commission gave six months notice of the possible use
of its plenary powers in the following cases:

Z. N. (S.) \AU—Argyrodes Simon, 1864 and Rohertus O. Pickard-Cambridge,
1879 (Arachnida, Araneae): proposed conservation by the suppression of
Argyrodes Guenee, 1845 and Ctenium Menge, 1871.

Z. N. (S.) lA^A—Olpium L. Koch, 1873 (Arachnida, Pseudoscorpionida,
Olpiidae): proposed designation of type species and related problems.

Z. N. (S.)2480— EngoAie Audouin, 1826 (Arthropoda, Araneae): proposed
designation of type species.

The Commission welcomes comments and advice from interested Zoologists
(Bull. Zool. Nomencl., vol. 42 pt. 1).

1985. The Journal of Arachnology 13:275

BOOK REVIEW

Roberts, M. J. 1985. The Spiders of Great Britain and Ireland. Harley Books,
Martens, Great Horkesley, Colchester, Essex, C06 4AH England. Vol. I, £45.00,
229 pp. 100 textfigs., 1985. Vol. Ill, £55.00, 256 pp., 237 plates of colored
drawings, 1985.

Besides a similarity in titles between Blackwall’s great work and that of
Roberts’ there is also the similarity in page size, both being much larger than the
usual. The 212 by 290 mm page of Roberts’ is just about twice that of the usual
manual size. The recently published “Spiders of the Britain and Northern
Europe” by Jones, and the “Spiders of the World” by the Preston-Mafhans are
each about half the page size of the Roberts’ work.

In volume I the first 30 pages are devoted to spider morphology, followed by
general information about spiders. The rest of the volume is devoted to
descriptions of those spiders belonging to the 27 families: Atypidae to
Theridiosomatidae. The description of the remaining family, the Linyphiidae
(sensu lato) is reserved for volume II, due to appear in 1986.

Most of the keys are of the standard dichotomous type. However, the key to
the families is not a completely dichotomous one, e.g., there are 12 alternatives
in “couplet” 7, and four each in “couplets” 6 and 10. There is merely a statement
of the chief characters for these taxa. Moreover, there is no key to the genera
of Lycosidae, but rather a listing of the nine genera and the characters of each
to be noted.

Interspersed throughout the text the author has included some “Taxonomic
notes” from which the reader can see that the author is well versed in current
and recent literature. For example, in these notes he explains why he does not
use either of the names Trachyzelotes and Urozelotes both accepted recently by
Platnick and Shadab. For the most part he does not follow Lehtinen (1967) but
does accept the latter’s generic name Nigma in the Dictynidae. The three orb-
weavers, originally in Araneus (or Epeira) referred to by Emerton as the three
house Epeira’s, but more recently shown to belong to Nuctenea Simon are by
Roberts placed, following Grasshoff, in Larinoides Caporiacco.

For each genus a section is given over to “distinguishing the species,” followed
by a section indicating the distribution in the British Isles. For those spiders in
which there are intraspecific variations in body size, carapace markings, and in
genitalia, such as in Pardosa, the author supplies lengthy comments in his
“taxonomic notes.” There may also be an assortment of illustrations. All this,
together with the excellent drawings should facilitate identification. Although
most of these drawings appear as black and white textfigures of genitalia, three
of the latter are in color, and some are of bodies, legs, and abdominal patterns.

1985. The Journal of Arachnology 13:276

In volume III all of the plates are in color. Most show a dorsal view including
the legs of both sides, of one spider. But a few plates show at a smaller scale
the bodies only of four different spiders. Accompanying each of these figures is
an inked outline showing the spider’s actual size. The plates 1-157 belong with
the descriptions given in volume I; the remaining 80 plates are of the Linyphiidae.
The author’s stated intent is to facilitate identification of specimens and he
succeeds admirably. Although the author is not an araneologist of long standing
any Britisher using this fine work will find himself deeply in Roberts’ debt.

B. J. Kaston, San Diego State University, San Diego, California, 92182.

GRANTS-IN-AID FOR RESEARCH

Grants-in-Aid for research on Arachnida (excluding Acarina) and Myriapoda are made
available to students and researchers through the ' ‘Exline-Frizz ell Fund for Arachnological
Research'’ of the California Academy of Sciences. Applications, which will be evaluated by
the American Arachnological Society and the Department of Entomology, California
Academy of Sciences (Golden Gate Park, San Francisco, California 94118-9961, phone
[415] 221-5100), may be submitted to the latter at any time. Application forms may be
obtained upon request. Awards will be made upon the approval of the Academy’s Director
shortly after March 1 and September 1 yearly. Grants will normally not exceed $750. The
Exline-Frizz ell Fund may be used for fieldwork, museum research (including travel),
expendable supplies, and costs of publications (including artwork).

THE AMERICAN ARACHNOLOGICAL SOCIETY

President:

Jerome S. Rovner (1985-1987)

Department of Zoology
Ohio University
Athens, Ohio 45701

Membership Secretary:

Norman I. Platnick (appointed)

American Museum of Natural History
Central Park West at 79th Street
New York, New York 10024

Secretary:

Brent D. Opell

Department of Biology
V.P.I.S.U.

Blacksburg, Virginia 24061

The American Arachnological Society was founded in August, 1972, to
promote the study of Arachnida, to achieve closer cooperation between amateur
and professional arachnologists, and to publish The Journal of Arachnology.

Membership in the Society is open to all persons interested in the Arachnida.
Annual dues are $25.00 for regular members, $15.00 for student members.
Correspondence concerning membership in the Society must be addressed to the
Membership Secretary. Members of the Society receive a subscription to The
Journal of Arachnology. In addition, members receive the bi-annual newsletter
of the Society, American Arachnology.

American Arachnology, edited by the Secretary, contains arachnological news
and comments, requests for specimens and hard-to-find literature, information
about arachnology courses and professional meetings, abstracts of papers
presented at the Society’s meetings, address changes and new listings of
subscribers, and many other items intended to keep arachnologists informed
about recent events and developments in arachnology. Contributions for
American Arachnology must be sent directly to the Secretary of the Society.

Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only
from current members of the Society, and there are no page charges. Detailed
instructions for the preparation of manuscripts appear in the Fall issue of each
year, and can also be obtained from the Editor and the Associate Editors.
Manuscripts that do not follow those instructions will be returned to the
author(s) without benefit of review. Manuscripts and all related correspondence
must be sent to Dr. William B. Peck, Associate Editor, Department of Biology,
Central Missouri State University, Warrensburg, Missouri 64093, U.S.A.

Presiden t- Elect:

William A. Shear (1985-1987)
Department of Biology
Hampden-Sydney College
Hampden-Sydney, Virginia 23943

Treasurer:

Norman V. Horner (1985-1987)
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308

Director:

Frederick A. Coyle (1985-1987)

G. B. Edwards (1984-1986)

Susan E. Riechert (1985-1987)

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 13 SUMMER 1985 NUMBER 2

Feature Articles

The systematics of the family Sternophoridae (Pseudoscorpionida),

Mark S. Harvey 141

Two-year life cycle and low palpal character variance in a Great

Smoky Mountain population of the lamp-shade spider (Araneae,

Hypochilidae, Hypochilus), Frederick A. Coyle 211

Egg feeding by Tegenaria spiderlings (Araneae, Agelenidae),

Guillermo Ibarra 219

Ballooning methodology: Equations for estimating masses of

sticky trapped spiders, Matthew H. Greenstone, Clyde E. Morgan

and Anne-Lise Hultsch 225

Predatory behavior of spitting spiders (Araneae, Scytodidae) and the

evolution of prey wrapping. Cole Gilbert and Linda S. Ray or 231

Redescriptions of some poorly known species of the nitidulus group
of the genus Vaejovis (Scorpiones, Vaejovidae), W. David Sissom
and Oscar E Francke 243

Research Notes

Sticky balls in webs of the spider Modisimus sp. (Araneae, Pholcidae),

R. Daniel Briceho 267

Arborine and methaqualone are not sedative in the wolf spider Lycosa
ceratiola Gertsch and Wallace, James E. Carrel, James P Doom

and John P. McCormick 269

Parasitic fungi as a mortality factor of spiders,

Wolfgang Nentwig 272

Others

Nomenclatural Notes 274

Book Review: The Spiders of Great Britain and Ireland by M. J. Roberts,

B. J. Kaston 275

Grants-in-Aid for Research 276

Cover photograph, Phrynus marginemaculatus Koch, by Scott A. Stockwell
Printed by the Texas Tech University Press, Lubbock, Texas
Posted at Lubbock, Texas, in September 1985

The Journal of

ARACHNOLOGY

OFFICIAL ORGAN OF THE AMERICAN ARACHNOLOGICAL SOCIETY

VOLUME 13

FALL 1985

NUMBER 3

THE JOURNAL OF ARACHNOLOGY

EDITOR: Oscar F. Francke, Texas Tech University

ASSOCIATE EDITOR: William B. Peck, Central Missouri State University
ASSISTANT EDITOR: James C. Cokendolpher, Texas Tech University
EDITORIAL ASSISTANT: Scott A. Stockwell, Texas Tech University
TYPESETTER: Sharon L. Robertson, Texas Tech University
EDITORIAL BOARD: C. D. Dondale, Agriculture Canada; W. G. Eberhard,
Universidad de Costa Rica; M. E. Galiano, Museo Argentino de Ciencias
Naturales; W. J. Gertsch, American Museum of Natural History; N. F.

Hadley, Arizona State University; V. F. Lee, California Academy of Sciences;
H. W. Levi, Harvard University; E. A. Maury, Museo Argentino de Ciencias
Naturales; W. B. Muchmore, University of Rochester; M. H. Muma, Western
New Mexico University; N. 1. Platnick, American Museum of Natural
History; G. A. Polis, Vanderbilt University; S. E. Riechert, University of
Tennessee; M. E. Robinson, U.S. National Zoological Park; V. D. Roth,
Southwestern Research Station; J. S. Rovner, Ohio University; W. A. Shear,
Hampden-Sydney College; W. J. Tietjen, Lindenwood College; G. B. Uetz,
University of Cincinnati; C. E. Valerio, Universidad de Costa Rica; S. C.
Williams, San Francisco State University.

HONORARY MEMBERS: P. Bonnet, W. J. Gertsch, H. Homann, fB. J. Kaston,
R. F. Lawrence, H. W. Levi, G. H. Locket, A. F. Millidge, M. Vachon, T.
Yaginuma.

THE JOURNAL OF ARACHNOLOGY (ISSN 0161-8202) is published in
Spring, Summer, and Fall by The American Arachnological Society in
cooperation with The Graduate School, Texas Tech University.

Individual subscriptions, which include membership in the Society, are $25.00
for regular members, $15.00 for student members. Institutional subscriptions to
The Journal are $35.00. Correspondence concerning subscriptions and
memberships should be addressed to the Membership Secretary (see back inside
cover). Back issues of The Journal are available from Dr. Susan E. Riechert,
Department of Zoology, University of Tennessee, Knoxville, TN 37916 U.S. A.,
at $10.00 for each number; the Index to Volumes 1-10 is available for $10.00.
Remittances should be made payable to The American Arachnological Society.
Correspondence concerning undelivered issues should be addressed to the
Assistant Editor, Department of Biological Sciences, Texas Tech University,
Lubbock, Texas 79409, U.S. A.

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

Casper, G. S. 1985. Prey capture and stinging behavior in the Emperor scorpion, Pandinus imperator
(Koch) (Scorpiones, Scorpionidae). J. Arachnol., 13:277-283.

PREY CAPTURE AND STINGING BEHAVIOR IN THE EMPEROR

PANDINUS IMPERATOR (KOCH)
(SCORPIONES, SCORPIONIDAE)

Gary S. Casper

Section of Vertebrate Zoology
Milwaukee Public Museum
Milwaukee, Wisconsin 53233

ABSTRACT

I

Prey capture behavior in the emperor scorpion, Pandinus imperator, is described and an
ontogenetic change in prey capture behavior reported. Young scorpions up to 6 cm in length stung
nearly all prey items. At 10 cm in length these scorpions stung only large, violently struggling prey
items. Adult scorpions never used the sting, dispatching prey with the pedipalps and apparently
refusing prey too large to subdue with the pedipalps alone. Prey capture behavior in P. imperator
is compared with other species and the possible survival values of sting use is discussed.

INTRODUCTION

The stinging behavior of scorpions is well known as a means of prey capture
and defense. However, the willingness to use the sting appears to vary greatly
with the species. The emperor scorpion, Pandinus imperator (Koch), is one of the
largest living scorpions, reaching a total length of over 17 cm. Inhabiting tropical
west Africa, it is a forest dwelling species (Cloudsley-Thompson 1958).

During feeding captive adult emperor scorpions were never observed to sting
their prey (usually common house crickets, Acheta domesticus Linnaeus). In the
fall of 1980 one of the scorpions gave birth to six young. Interesting was the fact
that the mother scorpion had been individually caged, without contact with other
scorpions, for at least two years prior to giving birth. This suggests that sperm
retention or some type of developmental interruption can occur in this species.
The mother died of unknown causes soon after giving birth and subsequently
only two young survived. The young scorpions, in direct contrast to the adults
of their species, stung their prey at every opportunity, often stinging crickets 2
or 3 times. As they grew larger this behavior waned, until at some 6-8 cm in
length they no longer used their sting on crickets.

Having observed this declining use of the sting with growth, it was decided to
explore under what, if any, conditions the sting would be employed in prey
subdual. Various sizes of prey items were offered to the scorpions, to test the idea
that prey over a certain size threshold would stimulate sting use.

278

THE JOURNAL OF ARACHNOLOGY

METHODS

Seven individuals of R imperator were maintained in 32 cm X 7 cm terraria
with 1-2 cm sanitary processed ground clay (Hartz® cat litter) as substrate. A
bottle cap containing water was the only cage furnishing. Five individuals were
adults at least four years of age and ranging form 14 cm to 17 cm in length
(measured from tip of chelicerae to tip of telson). Adults were housed
individually. Two individuals were 2.5 years of age and 10 cm in length. These
were sibling littermates and were housed together. Temperatures during the three
month study period ranged from 16-26° C, varying with outside temperature. No
artificial sources of heat or light were provided. Various sizes of the common
house cricket, A. domesticus, and the common house mouse, Mus musculus
Linnaeus, were offered to the scorpions as prey. Time between feedings was
generally one week.

RESULTS

Table 1 shows the results of 35 feeding trials with mice. In four trials the sting
was used, in one trial there was an unsuccessful attempt to sting, in 24 trials the
prey was subdued without using the sting, and in six trials the scorpion refused
to feed. All four instances of sting use were by the smaller, younger scorpions.
In no instance did the large adult scorpions attempt to sting. Moreover, none of
these scorpions stung crickets of any size, generally devouring them alive. Mice
which were not stung were killed by the crushing action of the pedipalps, except
in two instances where very small mice were devoured alive.

A typical encounter was as follows (terminology after Bub and Bowerman
1979). Upon opening the terraria the scorpion would either back into a corner
with the legs and pedipalps retracted, making itself as small as possible, or would
assume an alert stance in which the scorpion is supported above the substrate by
the legs, the pedipalps are extended anteriorly, and the metasoma is curled over
the back. Initially the pedipalps would often be raised 1-2 cm above the substrate.
Upon introducing the mouse and closing the terraria the alert stance would
usually be modified by placing the movable fingers of the pedipalpal chelae and
the pectines in contact with the substrate. If the scorpion had cowered in a corner
an alert stance would be assumed a minute or two after the terraria was closed
with the mouse inside.

Next the scorpion orients, directing its anterior aspect towards the prey.
Orientation occurred only when the prey was active, the scorpion seemingly being
unable to orient if the mouse remained motionless. The scorpion would next
approach to within 5-10 cm of the prey. During orientation and approach the
pedipalpal chelae and the pectines are raised off the substrate, only to be lowered
again when the scorpion halts its progress.

Finally there occurs the attack and grasp attempt, in which the scorpion rushes
at the prey with the pedipalps extended and held widely apart, so as to form a
sort of corral. Contact is often made seemingly by accident, the scorpion
apparently bearing down on the general vicinity of the prey with the extended
pedipalps sweeping a wide enough area to make contact likely. The attack
culminates in the grasp attempt, in which the scorpion attempts to obtain a firm

CASPER— SCORPION PREY CAPTURE BEHAVIOR

279

Table 1. — Feeding response of P. imperator to Mus musculus. Scorpion lengths measured from tip
of chelicera to tip of telson. Mice: Group A-3.0 cm (snout to vent length), hairless, eyes closed; B-
3.8 cm, furred, eyes closed; C-4.0 cm, furred, eyes closed; D-5.0 cm, furred, eyes open; E-5.1 cm,
furred, eyes open.

Scorpion

Mouse group

Sting use

Remarks

1-10 cm

A

no

devoured alive

B

no

killed by pedipalps

C

yes

stung once midbody

D

yes

stung once midbody

E

—

refused

2-10 cm

A

no

devoured alive

B

no

killed by pedipalps

C

unsuccessful

attempt

attempted sting in
head but did not
penetrate, subsequently
killed by pedipalps

D

yes

stung once midbody

E

yes

stung once at shoulders

3-14 cm

A

no

killed by pedipalps

B

no

killed by pedipalps

C

no

killed by pedipalps

D

—

refused

E

—

refused

4-14 cm

A

no

killed by pedipalps

B

no

killed by pedipalps

C

no

killed by pedipalps

D

no

killed by pedipalps

E

no

killed by pedipalps

5-15 cm

A

no

killed by pedipalps

B

no

killed by pedipalps

C

no

killed by pedipalps

D

—

refused

E

—

refused

6-15 cm

A

no

killed by pedipalps

B

no

killed by pedipalps

C

no

killed by pedipalps

D

no

killed by pedipalps

E

no

killed by pedipalps

7-17 cm

A

no

killed by pedipalps

B

no

killed by pedipalps

C

no

killed by pedipalps

D

no

killed by pedipalps

E

—

refused

hold on the prey with at least one pedipalp. If the prey at any time runs away
or otherwise eludes the scorpion or manages to free itself from a successful grasp,
the scorpion performs the sequence of orientation, attack, and grasp attempt all
over again.

Once the prey is successfully grasped by one pedipalp the scorpion immediately
obtains a hold with the other pedipalp as well. At this point the prey is held well
away from the mouthparts and often slightly elevated, so as to prevent purchase
on the substrate which might facilitate its struggling. Biting and clawing by the
prey is also thus restricted to attacks on the pedipalps, the cuticle of which is
sufficiently durable to withstand any damage. Should the prey struggle violently
it may be subdued by one or both of two methods.

280

THE JOURNAL OF ARACHNOLOGY

In the first method, the scorpion may use the pedipalps as killing or maiming
weapons. Often this entails obtaining new and more effective holds on the prey
than those found with the initial grasp. Consequently one pedipalp may release
its grip and regrasp elsewhere. Usually the scorpion will grasp the prey at either
end, thereby obtaining a head grip with one pedipalp that is frequently lethal
when force is applied. Violently struggling prey is often repeatedly passed from
pedipalp to pedipalp in an attempt to find an effective grip. Scorpions will
occasionally intentionally release and retreat from violently struggling prey, only
to attack again. Larger prey items often escaped during regrasping attempts,
necessitating reorientation and a new attack by the scorpion.

In the second method, the sting may be used to subdue prey after a successful
grasp attempt. This only occurred if the prey struggled violently. In the four
instances of stinging observed, struggling ceased almost immediately upon aculeus
penetration, with cessation of breathing and apparent death from 90 to 180
seconds later. In the four instances observed the prey was stung only once.
Rather than a quick jab, the metasoma was leisurely arched over the back and
the telson used to probe the prey for a soft spot if one was not immediately
encountered. The aculeus was inserted for 5 to 15 seconds, presumably injecting
venom. The sting was always preceded by attempts to subdue the prey with the
pedipalps. A firm grip was maintained by the pedipalps during stinging.

Once the prey is subdued ingestion begins. The pedipalps bring the prey in
contact with the chelicera, which tear off pieces of flesh and convey them to the
oral cavity. E imperator may feed for two days on a carcass before leaving it,
apparently disdaining stale food. The prey is often alive when ingestion
commences.

DISCUSSION

Notes on sting employment in prey subdual are very scarce in the literature.
McDaniel (1968) divided California scorpions into two groups on the basis of
their habits and morphology. Errant types are characterized by long legs, a
slender body, a large thick cauda with a large telson, and chelae with a long
slender tarsus and tibia. These are described as actively pursuing prey and having
rapid stinging reflexes. Paruroctonus sylvestrii (Borelli) is of this type. The second
type is the obligate borrower, with a stouter body, shorter and more slender
cauda, and broad chelae with short, sturdy tarsus and tibia. These are described
as waiting for prey to come to them rather than pursuing it and relying on the
pincers rather than the sting. Here Anuroctonus phaiodactylus (Wood) is an
example. Williams (1966) also discusses burrowing activities in the scorpion A.
phaiodactylus and again notes that the pedipalps are distinctively thick, heavy,
and powerful; they are the primary means of catching and immobilizing prey.

Stahnke (1966) notes that the sting is used as an offensive weapon ''when the
prey is obstreperous and will not quietly submit to being devoured alive” and that
“scorpions with powerful chelae depend largely upon their pinching and crushing
ability for both offensive and defensive action.” Fabre (1923), in working with
Buthus occitanus (Amoreux), noted that the sting was frequently employed to
subdue struggling insects. Southcott (1955) found that Urodacus manicatus
(Thorell) invariably stings its prey as soon as it is captured. By contrast, Schultze

CASPER— SCORPION PREY CAPTURE BEHAVIOR

281

(1927) recorded that he had never seen the large Philippine forest scorpion,
Heterometrus longimanus (Herbst), sting its prey. Schultze believed that “the
poisonous stinger us used only as a defensive weapon against its enemies.” In
Schultze’s experiments the prey (cockroaches) was held clear of the ground and
eaten while still struggling. Vachon (1953) writes that in capturing its prey the
scorpion “moves slowly forward, supported on its hind legs, with claws open and
extended and tail raised and pointing forwards. Often the scorpion will then
hesitate, and the final act of capture seems almost accidental, the scorpion may
even withdraw for a time, but it waits patiently and finally achieves its aim. Then,
especially if the victim struggles, it inserts its sting where best it can, often
without any delay.”

Hadley and Williams (1968) made observations on Vaejovis confusus Stahnke,
Pamroctonus mesaensis Stahnke, Paruroctonus baergi (Williams and Hadley),
Hadrurus arizonensis pallidus Williams, and Centruroides exilicauda (Wood).
They found that “scorpions generally used their venom apparatus at the time of
prey capture.” They note that several species of mice and lizards preyed upon by
H. a. pallidus appeared immune to the venom, however. Although the scorpions
observed by Hadley and Willliams usually grasped the prey in the pedipalps
before employing the sting, this was not always so. If the prey fought back
aggressively, the scorpion sometimes stilted on its walking legs with the
mesosoma and metasoma arched in almost a vertical position, from which
posture the scorpion could strut slowly or twirl around in small circles, stinging
blindly at its target.

Bub and Bowerman (1979) studied prey capture in Hadrurus arizonensis Ewing
and found that the prey was stung at least once in all sequences observed. Baerg
(1961) points out that scorpions with large pedipalps and reduced metasomas
probably do not use the sting to immobilize their prey. Burton (1975) writes that
scorpions will only use the sting if the prey offers resistance. Cloudsley-Thompson
(1951) notes that Euscorpius italicus (Herbst) seldom, if ever, uses the sting to
subdue prey. On the other hand, Scorpio maurus Linnaeus, Buthus occitanus
(Amoreaux), and Androctonus australis (Linnaeus) will lash out with their sting
at the slightest provocation (Cloudsley-Thompson 1958).

In sum, accounts in the literature tend to state generally that the sting is only
used in prey subdual if the prey struggles excessively. However, when species are
individually examined some tend never to use their sting and others always use
their sting. There appears to be an inverse relationship between the size of the
pedipalps and the frequency of stinging behavior. Those species with large
powerful pedipalps appear to rarely use their sting (for example P imperator, H.
longimanus, and A. phaiodactylus), while those species with small slender
pedipalps use the sting frequently (//. arizonensis, U. manicatus, B. occitanus,
Vaejovis spp.).

STING USE IN P IMPERATOR

Pandinus imperator is a species with large powerful pedipalps. The metasoma
is also well developed. Although the sample size was too small for statistical
analysis, the present observations indicate that P. imperator seldom, if ever, uses
the sting in prey subdual as an adult. In this respect P. imperator is similar to

282

THE JOURNAL OF ARACHNOLOGY

E. italicus and H. longimanus (Cloudsley-Thompson 1951, Schultze 1927).
Heterometrus longimanus is also a large species with well developed pedipalps.
(Information on the morphology of E. italicus was not available.)

An ontogenetic change in prey capture behavior was evident in the present
study. The young P. imperator used their stings on crickets frequently as they
grew. At 6-8 cm in length these scorpions changed their prey capture behavior
and dispatched crickets with the pedipalps or devoured them alive. These
scorpions were then entered into the feeding trials with mice, where they reverted
to their earlier stinging behavior to subdue mice larger than 4 cm in length but
continued to dispatch smaller mice via pedipalpal action alone or by devouring
them alive.

Why use of the sting in prey capture should be phased out as R imperator
matures is an interesting question. Possibly the use of the sting increases the prey
capture success of the young scorpions, thereby imparting a survival advantage.
No information on the natural prey items of P. imperator was available, but it
is likely that by using the sting the young scorpions make available a greater
variety of prey, obtain prey more efficiently, and consequently grow rapidly
during a period of life when mortality is undoubtedly high. But if use of the sting
is advantageous in capturing prey, why is its use lost in the adults? Possibly, once
the scorpion has attained a certain size its pedipalps alone are large enough and
powerful enough to dispatch any normal prey item, and prey items large enough
to necessitate stinging are rejected in favor of smaller items which pose less of
a risk of injury to the scorpion. Also, venom production may be costly from an
energetics standpoint. Whether or not this is so, it was clear in the present
experiments that the use of the sting would have saved much energy in struggling
with the prey item. In fact, if not for the confines of the terraria it is unlikely
that the scorpion could have caught the mouse in the first place, much less be
able to make a second attempt after the mouse escaped an initial grasp. Clearly,
further investigations will be necessary to explain the biological significance of
this ontogenetic behavioral change.

CONCLUSION

An ontogenetic change in prey capture behavior, involving loss of stinging
behavior in prey subdual, was observed in two captive born P. imperator. At 6-
8 cm in length these scorpions abandoned sting use in prey subdual, but at 10
cm in length were shown to revert to the juvenile stinging behavior if confronted
with prey too formidable to be subdued via pedipalpal action alone. Wild caught
adult scorpions at least 14 cm in length refused to utilize the sting in prey
subdual, apparently rejecting prey too large to subdue with the pedipalps alone.
The ontogenetic change in prey capture behavior demonstrated in P imperator
may occur in other scorpions as well, and needs to be investigated especially in
those species noted for infrequent or absent stinging behavior {H. longimanus, E.
italicus, A. phaiodactylus). The biological significance of this ontogenetic change
is at present unknown.

CASPER— SCORPION PREY CAPTURE BEHAVIOR

283

ACKNOWLEDGMENTS

The author would like to thank Joan P. Jass for assistance with taxonomic
searches, Terry J. Cullen for use of the scorpions, and the reviewers who offered
many useful comments.

LITERATURE CITED

Baerg, W. J. 1961. Scorpions: Biology and effects of their venom. Univ. Arkansas Agr. Exp. Stn.
Bull., 649:1-34.

Bub, K. and R. F. Bowerman. 1979. Prey capture by the scorpion Hadrurus arizonensis Ewing
(Scorpiones, Vejovidae). J. Arachnoi., 7:243-253.

Burton, R. 1975. Encyclopedia of insects and arachnids. Octopus, London.

Cloudsley-Thompson, J. L. 1951. Notes on Arachnida, 16. — The behavior of a scorpion. Entomol.
Mon. Mag., 86(105).

Cloudsley-Thompson, J. L. 1958. Spiders, scorpions, centipedes and mites. Pergamon Press, New
York. 228 pp.

Fabre, J. H. 1923. The life of the scorpion. Dodd, Mead & Co., New York.

Hadley, N. F. and S. C. Williams. 1968. Surface activities of some North American scorpions in
relation to feeding. Ecology, 49(4):726-734.

McDaniel, M. M. 1968. Notes on the biology of California scorpions. Entomol. News, 79:278-284.

Schultze, W. 1927. Biology of the large Philippine forest scorpion. Philippine J. Sci., 32(3):375-389,
plates 1-4.

Southcott, R. V. 1955. Some observations on the biology, including mating and other behavior of
the Australian scorpion Urodacus abruptus Pocock. Trans. Royal Soc. Australia, 78:145-154.

Stahnke, H. L. 1966. Some aspects of scorpion behavior. Bull. Southern California Acad. Sci.,
65(2):65-80.

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

Williams, S. C. 1966. Burrowing activities of the scorpion Anuroctonus phaiodactylus (Wood)
(Scorpionida: Vejovidae). Proc. California Acad. Sci., 34(8):4 19-428, 2 figs., 2 tables.

Manuscript received July 1984, revised December 1984.

'■Vi

V ’.,1

it.m 1^' -

s||iar H»r

M 1

f ifn wilt — -^y

i •,...’

w

.:$!

' -'k* '■’

■r. -.‘tVi*! ,

(■ ir. »¥

^nh’ ■ I

’i

. !

»> i

‘ - 'W:‘V /.'.V* ' i ’ ,'-x-.j'. ^

•1 .^.t<t-,-{^, -I w ,.''‘<",.1 .' '.

■ •'^'

, '■■ ■ , r 1

■ ■ '* •A' i3

> '* * ■* ' '. ’ • ■ /*>< ' '•■t’UTfT'll

</*';• ' r-',- ■•’:>> f ^ >iqf^

•»■• ' ^

•» . _v' ,■ .=•>, -

I ^ “Ir- • ■ 'r, -.■ /.'■ j-,i' M J/4#|

’■

.'^ '1 ,1. 1

' 4,.

Jfliii*'' .'i*i^ j

' •?

%..-,» r,y If, V^f,

'• ' :■ -*d

**4'^ H ••

PU

■

fi-

4‘:\,.r

•,| -fv.’

' E

v^.- J f » |V ^‘4

, A fjilrf'' - s? •

■ i/- •."

- . '■'‘■'i '-iL ^ i^'‘*''^!''JiT-.-iiV

te'SSI

-'JL- >-''i .

.'■’fe.

' "* f •'■

>E

1^^ ^ f*

.1

. l■.•^^.4ll^ )v-\ >. ii tU' *' vv'jir|fy

i: in , *’ f tiwsii i • L.I

I Tf, J I

* I'*:' .■ '.‘"C ,-, . .», , -,,,♦, .,*.

’ a-/ ‘ *f>

rvn

,. , \tc-

;>fr

^ ' if*

Raven, R. J. 1985. Two new species of Ixamatus Simon from eastern Australia (Nemesiidae,
Mygalomorphae, Araneae). J. Arachnol., 13:285-290.

TWO NEW SPECIES OF IXAMATUS SIMON
FROM EASTERN AUSTRALIA
(NEMESIIDAE, MYGALOMORPHAE, ARANEAE)

Robert J. Raven

Queensland Museum

Gregory Terrace, Fortitude Valley, Queensland, 4006, Australia

ABSTRACT

Two new species of Ixamatus — /. lornensis has a low, broad tarsal organ; /. rozefeldsi has a spinose
cymbium — present characters previously unknown and probably plesiomorphic for the genus.

INTRODUCTION

Ixamatus was revised in a two-part study (Raven 1980, 1982) and includes eight
species. Some initial confusion between species from eastern Australia and
apparently similar species in the south and west resulted in a much wider
distribution being ascribed to the genus than is actually the case (see Main 1983).
After the second revision, new material of Ixamatus was found. Because the
changes required in the diagnosis of Ixamatus are cladistically noteworthy I have
chosen to describe both the species and the changes prior to making a general
biogeographical history (in prep.) of Ixamatus and other Australian
mygalomorphs that have been revised.

The terminology, methods, and abbreviations are consistent with my previous
studies and any of the larger studies (e.g.. Raven 1982) will provide a full list.

Ixamatus Simon

Ixalus L. Koch 1873:469. Type species by monotypy: Ixalus varius L. Koch 1873.

Ixamatus Simon 1887:195 {nomen novum for Ixalus L. Koch 1873); Raven 1982:1036.

Diagnosis. — Ixamatus differs from Xamiatus in the absence of plumose hairs
on the palpal trochanters of adults, and from the remaining nemesiid genera by
the elevated tarsal organ.

Remarks. — A full synonymy and description are given in Raven (1982, 1985).
Males described here require two modifications of that description. First, in most
species of Ixamatus, the tarsal organ is high and raised and the cymbium is not
spinose. In contrast, the tarsal organ of /. lornensis is short and broad, and the

286

THE JOURNAL OF ARACHNOLOGY

cymbium of 1. rozefeldsi is spinose. The cladistic implications will be discussed
elsewhere. Suffice it to say here that both of these newly described conditions are
plesiomorphic in the Nemesiidae.

Ixamatus lornensis, new species
Figs. 1-7, Table 1

Type. — Holotype male, Lome State Forest, N. S. W., 31°35' S — 152°38' E
(1 l.v-19.vi.l978, D. Milledge), Australian Museum No. KS 1562.

Diagnosis. — Differs from /. caldera Raven by the low tarsal organ, spinose
cymbium, and the absence of megaspines on tibia I. Medium-sized spiders,
carapace ca. 5-6 long. Dorsal abdomen anteriorly mottled. Maxillary serrula
group of about 15 low teeth. Tibia I of male unmodified; metatarsus I with slight
retrolateral excavation proximally; palpal bulb spherical with short embolus,
tarsus with several distinct spines apically. Tarsal organ low, broad. Female
unknown.

Etymology. — The specific epithet refers to the type locality.

Description. — Holotype male. Carapace 5.69 long, 4.63 wide. Abdomen 5.00
long, 2.88 wide. Total length 11.88.

Color in alcohol: carapace, chelicerae, and legs orange brown; abdomen
dorsally brown with white mottling anteriorly forming two irregular lines,
ventrally almost entirely white with brown areas medially.

Carapace: fovea broad, slightly procurved; lateral margins with silver hairs on
dorsal coxae, caput, and interstrial ridges; sparsely clothed; 3 pairs of foveal
bristles. Eyes: tubercle low but distinct; group 0.4 of head-width, 1.82 times wider
than long; back row recurved; ratio MOQ back width: front width: length,
35:26:23; ratio AME:ALE:PME:PLE, 11:10:7:8; eye interspaces: AME-AME, 4;
AME-ALE, 1; PME-PLE, 1; ALE-PLE, 1. Chelicerae: with brown bristles and
silver hairs on prodorsal surface; 2 depressions in anterolateral surfaces;
promargin of furrow with 12 teeth; basally with 8 fine teeth.

Maxillae: front length, 1.56; back length, 2.16; width, 1.04, with about 40 blunt
cuspules on inner mound; serrula consisting of about 15 low teeth. Labium: 1.08
wide, 0.48 long. Sternum: 2.92 long, 2.44 wide; shape, length, and distance from
margin of sigilla: posterior, oval, 0.32, 0.28; middle, oval, 0.20, 0.08; anterior,
circular, 0.12, 0.04.

Palp: bulb spherical; embolus short; with eight spines and three stout bristles
on apical tarsi.

Legs: (Table 1). 1423; tibia I unmodified; metatarsus I with slight retrolateral
excavation and mid-distal retrolateral cuticular point; retroventral metatarsus IV

Table 1. — Leg measurements of Ixamatus lornensis. Values are for holotype male.

I

II

III

IV

Palp

Femur

4.75

4.06

3.69

4.63

3.13

Patella

2.81

2.19

1.94

2.06

1.61

Tibia

3.31

2.69

2.06

3.25

2.56

Metatarsus

3.19

2.94

3.19

3.81

-

Tarsus

1.63

1.50

1.56

1.88

1.00

Total

15.69

13.38

12.44

15.63

8.30

raven— NEW IXAMATUS SPECIES

287

Figs. 1-7. — Ixamatus lornensis, holotype male: 1, carapace, chelicerae, and abdomen, dorsal view;
2, abdomen and spinnerets, ventral view; 3, sternum, maxilla, and labium; 4, tibia and metatarsus
I, prolateral view; 5, bulb and cymbium, retrolateral view; 6, palpal bulb, cymbium, and tibia,
retrolateral view; 7, metatarsus I, dorsal view. All scale lines = 1 mm.

288

THE JOURNAL OF ARACHNOLOGY

with two close setae-like preening combs; silver hairs on femora I-IV; scopulae
entire but thin on metatarsi and tarsi 1. Spines: no spines on leg tarsi. First leg:
femur, pi d4; patella, p2; tibia, p2 dl v9; metatarsus, p2 v4. Second leg: femur,
p2 d3; patella, p2; tibia, p2 v8; metatarsus, p3 v7. Third leg: femur, p3 d3 r3;

patella, p2 rl; tibia, p2 dl r2 v6; metatarsus, p5 r3 v7. Fourth leg: femur, p3 r3;

patella, rl; tibia, p2 dl r2 v7; metatarsus, p3 dl r3 v9. Palp: femur, pi d3;

patella, pi; tibia, pi v2; tarsus, 9 apical. Claws: STC of legs I, II with 10 teeth

in each of two rows; STC of legs III, IV with 8-9 teeth per row; ITC without
teeth. Trichobothria: two rows, each of 9 for full length of tibiae; 12 on metatarsi;
18 on tarsi; tarsal rod low, broad.

Spinnerets: PMS 0.23 long, 0.18 wide, 0.40 apart; lengths of basal, middle,
apical, and total segments of PLS, 1.05, 0.93, 1.18, 3.16, respectively.

Distribution and Habitat. — Ixamatus lornensis is known only from the
rainforest of Lome State Forest, New South Wales.

Material Examined. — Only the type.

Ixamatus rozefeldsi, new species
Figs. 8-14, Table 2

Type. — Holotype male. Byfield near Rockhampton, Q., 22°5FS — 150°39'E
(27. vi. 1982, A, Rozefelds). Queensland Museum No. S1314.

Diagnosis. — Differs from all other Ixamatus species by the distinct process on
retrolateral metatarsus I of males. Large spiders, carapace ca. 8 long. Dorsal
abdomen with large white area anteriorly. Maxillary serrula absent. Tibia I of
male with three large megaspines on raised bases; metatarsus I with excavation
proximal to retrolateral cuticular process; cymbium not spinose; bulb pyriform
with short embolus. Female unknown.

Etymology. — The specific epithet in a patronym in honor of Mr. Andrew
Rozefelds, an enthusiastic and fearless collector of mygalomorphs.

Description. — Holotype male. Carapace 8.25 long, 8.13 wide. Abdomen 9.70
long, 6.00 wide.

Color in alcohol: carapace, chelicerae, and legs red brown; abdomen dorsally
brown with large white area anteriorly and three pairs of irregular areas forming
broken chevrons, ventrally cream with brown areas between PMS and laterally.

Carapace: fovea deep, recurved; several paired bristles in front of fovea;
uniformly covered with black bristles and pile of silvery hairs on interstrial ridges
and caput. Eyes: tubercle low but distinct; group 0.31 of head-width, 2.07 times
wider than long; back row slightly recurved; ratio MOQ back width: front width:
length, 44:31:26; ratio AME:ALE:PME:PLE, 13:14:9:12; eye interspaces: AME-
AME, 8; AME-ALE, 2; PME-PLE, 3; ALE-PLE, 3 (PME and PLE of one side

Table 2. — Leg measurements of Ixamatus rozefeldsi. Measurements are for holotype male.

I

II

III

IV

Palp

Femur

8.13

7.19

6.69

8.25

5.00

Patella

4.44

4.06

3.38

3.88

2.69

Tibia

5.38

4.75

4.19

6.31

3.94

Metatarsus

5.56

5.00

5.31

7.06

-

Tarsus

2.94

2.75

2.63

3.13

1.75

Total

26.45

23.75

22.02

28.63

13.38

RAVEN— NEW IXAMATUS SPECIES

289

fused). Chelicerae: with silver hairs and long black bristles; promargin of furrow
with 1 1 teeth; basally with 8 fine teeth.

Maxillae: front length, 2.48; back length, 3.52; width, 1.60, with about 50 stout,
pointed cuspules on inner edge; serrula absent. Labium: 1.52 wide, 1.04 long.

Figs. 8-14. — Ixamatus rozefeldsi, holotype male; 8, carapace and chelicerae, dorsal view; 9, palpal
bulb, cymbium, and tibia (right), ventral view; 10, metatarsus I (right), ventral view; 11, tibia and
metatarsus I, prolateral view; 12, sternum, maxillae, and labium; 13, 14, abdomen and spinnerets,
dorsal view (13), ventral view (14). All scale lines = 1 mm.

290

THE JOURNAL OF ARACHNOLOGY

Sternum: 4.88 long, 3.56 wide; all sigilla oval. Length and distance from margin
of sigilla: posterior, 0.63, 0.40; middle, 0.40, 0.28; anterior, 0.15, 0.20.

Palp: bulb pyriform with short embolus.

Legs: (Table 2). 4123; tibia I with 3 large megaspines on raised bases, most
distal thickest; metatarsus I with excavation proximal to retrolateral cuticular
process; preening combs absent; scopulae on tarsi I, II; fine brown hairs on
femora; no modified hairs anywhere. Spines: no spines on leg tarsi. First leg:
femur, p2 d2; patella, pi; tibia, p3 v7; metatarsus, v3. Second leg: femur, p3 d2;
patella, p2; tibia, p2 v8; metatarsus, pi v7. Third leg: femur, pi d3 r2; patella,
pi rl; tibia, p2 r2 v7; metatarsus, p2 rl v8. Fourth leg: femur, p3 r2; patella,
0; tibia p2 r3 v7; metatarsus, p2 r2 v8. Palp: femur, p2; patella, 0; tibia, vl;
tarsus, 0. Claws: STC with 10 teeth in each of two rows; ITC without teeth.
Trichobothria: two rows, each of 11 extending to % length of tibiae; 16 in
straight line on metatarsi; 23 in slightly irregular line on tarsi; tarsal rod large,
elevated, distal.

Spinnerets: PMS 0.96 long, 0.32 wide, 0.88 apart; lengths of basal, middle,
apical, and total segments of PLS, 2.40, 1.88, 2.80, 7.08, respectively.

Distribution and Habitat. — Ixamatus rozefeldsi is known only from Byfield,
near Rockhampton, Queensland. The holotype was found in a small temporary
web under a log in a gully that is part of a small area of low “poor” rainforest.

Material examined. — Only the type.

ACKNOWLEDGMENTS

I am grateful to Dr. Mark Harvey (Division of Entomology, CSIRO,
Canberra) for reviewing the manuscript, and to Dr. V. E. Davies (Queensland
Museum, Brisbane) and Mr. M. R. Gray (Australian Museum, Sydney) for
loaning the types. This paper was produced while I was in receipt of a C.S.I.R.O.
Post-doctoral Award at the Division of Entomology, CSIRO Australia,
Canberra; I am most grateful for the award and the resources provided by the
Division.

LITERATURE CITED

Koch, L. 1873. Die Arachniden Australiens, nach der natur beschrieben und abgebildet. pp. 369-472
(Niirnberg).

Main, B. Y. 1983. Further studies on the systematics of the Australian Diplurinae (Chelicerata:
Mygalomorphae: Dipluridae): Two new genera from southwestern Australia. J. Nat. Hist., 17:923-
949.

Raven, R. J. 1980. The Australian mygalomorph spider genus Ixamatus Simon (Dipluridae:

Diplurinae) and its affinities. Bull. British Arachnol. Soc., 5:43-49.

Raven, R. J. 1982. Systematics of the Australian mygalmorph spider genus Ixamatus Simon
(Dipluridae: Diplurinae: Chelicerata). Australian J. ZooL, 30:1035-1067.

Raven, R. J. 1985. The spider infraorder Mygalomorphae (Araneae): Cladistics and systematics. Bull.
Amer. Mus. Nat. Hist., 182:1-180.

Simon, E. 1887. Quelques observations sur les arachnides. Ann. Soc. Entomol. France, 6(7): 193-195.

Manuscript received November 1984, revised January 1985.

Coyle, F. A., M. H. Greenstone, A.-L. Hultsch and C. E. Morgan. 1985. Ballooning mygalomorphs:
Estimates of the masses of Sphodros and Ummidia ballooners (Araneae: Atypidae, Ctenizidae).
J. Arachnol., 13:291-296.

BALLOONING MYGALOMORPHS: ESTIMATES OF THE MASSES
OF SPHODROS AND UMMIDIA BALLOONERS
(ARANEAE: ATYPIDAE, CTENIZIDAE)

Frederick A. Coyle

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

Matthew H. Greenstone, Anne-Lise Hultsch and Clyde E. Morgan

U. S. Department of Agriculture
Biological Control of Insects Research Laboratory
P. O. Box 7629, Research Park
Columbia, Missouri 65205

ABSTRACT

The masses of Sphodros atlanticus and Ummidia sp. spiderlings capable of ballooning, and of
Antrodiaetus unicolor spiderlings believed not to balloon, were estimated by means of a volume-mass
regression developed with data from five araneomorph families, and compared with the mass
frequency distribution of a sample of 216 aerially dispersing araneomorphs trapped over a one week
period in a soybean field. Mean estimated masses of the Sphodros and Ummidia spiderlings were 1.25
mg and 3.45 mg, respectively. Although these were in at least the 95th percentile of masses of the
trapped araneomorphs, larger araneomorphs were trapped. These results and the finding that A.
unicolor spiderlings were intermediate in estimated mass (mean = 2.02 mg) between S. atlanticus and
Ummidia, indicate that mass is not the only constraint on ballooning behavior in mygalomorphs.
Habitats of the three species support the habitat predictability hypothesis of ballooning, A. unicolor
tending to be found in more predictable habitats than the other two species. Live mass measurements
of ballooning Ummidia spiderlings indicate that the volume-mass regression estimates of
mygalomorph spiderling masses are consistently slightly high. This may be due to tissue density
differences between araneomorphs and mygalomorphs.

INTRODUCTION

Although the great majority of mygalomorph spider species do not appear to
disperse aerially, observations of ballooning Sphodros atlanticus (Gertsch and
Platnick) and Ummidia spiderlings (Coyle 1983, 1985) and of pre-ballooning
behavior in spiderlings of Sphodros rufipes (Latreille) (Muma and Muma 1945),
Atypus affinis Eichwald (Knock 1885, Bristowe 1939), Ummidia carabivora
(Atkinson) (Baerg 1928), and Conothele malayana (Doleschall) (Main 1957) show
that some species in two mygalomorph families (Atypidae and Ctenizidae) are
aerial dispersers. Although Sphodros and Ummidia ballooning (which involves
dropping and hanging from a dragline that is lifted and lengthened by a breeze.

292

THE JOURNAL OF ARACHNOLOGY

breaks near the attachment substrate, and serves as the ballooning thread) is a
more primitive (and probably less effective) form of ballooning than that
practiced by most araneomorphs, it appears to increase significantly the vagility
of these animals (Coyle 1983). The water gaps bridged by the distribution ranges
of A. affinis (Kraus and Baur 1974; Locket, Millidge, and Merrett 1974), C.
malayana (Main 1957), and by other Atypus and Ummidia species in Japan
(Yaginuma 1970) and the Florida Keys (G. B. Edwards, pers. comm.) also suggest
that mygalomorph ballooning can significantly increase vagility.

Because spiders are not capable of active flight, there must be some upper limit
to the mass at which they can disperse aerially. Ballooning in mygalomorphs has
therefore been an arachnological curiosity, since average mygalomorph adults and
spiderlings (data presented in this paper) are far more massive than the
corresponding stages of araneomorphs. This study was designed to estimate the
masses of ballooning mygalomorph spiders and to compare them with the masses
of typical ballooning araneomorphs, so that the following questions can begin to
be addressed: is ballooning rare in mygalomorphs primarily because of large
spiderling size? In those mygalomorph species which do balloon, to what degree
is large mass a handicap which has been overcome by special adaptations?

METHODS AND STUDY SITES

Sphodros atlanticus spiderlings were collected on 9 October 1980 with their
mother in her burrow on a grassy roadbank 6.5 km south of Cullowhee in
Jackson County, North Carolina. These were third instar spiderlings which
overwinter in the maternal burrow and disperse in the spring (see Coyle and
Shear 1981, for a description). Spiderlings of an undetermined species of
Ummidia were collected while performing pre-ballooning behavior (climbing
around in the top of a boxwood shrub about 1 m above ground) on 19 April
1982 near Apex in Wake County, North Carolina, by J. M. Ragan. Twenty
dispersal stage (second instar) spiderlings of Antrodiaetus unicolor, collected on
16 January 1973, eight km south of Cullowhee in Jackson County on the Wolf
Creek Biological Preserve, were included in the study to provide mass estimates
for a mygalomorph which is believed not to balloon (Coyle 1971, Reagan and
McGimsey pers. comm.). All spiderlings were preserved in 70% ethanol.

Ballooning araneomorphs were collected on vertical sticky traps set out
between 30 and 200 cm high in an eleven ha. soybean planting at the University
of Missouri South Farms, eight km SE of Columbia in Boone County, Missouri,
during the seven-day period ending 9 August 1983. Details of the trapping
procedure have been presented elsewhere (Greenstone 1984). Briefly, the trap
supports were banded with adhesive to prevent spiders from walking onto them,
and all vegetation was cleared from within 3 m of the traps to reduce the
probability that spiders would be inadvertently trapped while “rappelling” (J. E.
Carico, personal communication) or “bridging”, i.e., floating silk “lines on the
wind to establish paths to distant objects” without releasing hold of the substrate
(W. G. Eberhard, personal communication). The traps were returned to the
laboratory and examined with a Wild M5 stereomicroscope to locate all trapped
spiders. The spiders were removed from the adhesive (Tacktrap®, Animal
Repellents, Inc., Tifton, Ga) and placed for three days each in paint thinner and
toluene before final preservation in 70% ethanol.

COYLE £7 /IL.— MASSES OF BALLOONING MYGALOMORPHS

293

Mass estimates for both the mygalomorphs and sticky-trapped araneomorphs
were made by use of a volume-mass regression developed with 101 araneomorph
specimens ranging in mass from 0.7 to 19.5 mg. Briefly, previously massed
animals were preserved in 70% ethanol and their volume estimated by treating
them as cylindrical solids having diameter equal to the mean of greatest carapace
and abdominal widths and height equal to total length (anterior edge of carapace
to posterior end of abdomen, exclusive of spinnerets). Measurements were made
with the Wild M5 stereomicroscope with ocular micrometer at 12X magnification.
Separate regressions for five araneomorph families do not differ significantly in
slope and intercept, and the overall regression combining the data for those five
families provides a good fit for limited data from two other araneomorph families
(Greenstone et al., in press, a). This regression is valid for sticky-trapped as well
as directly ethanol preserved specimens (Greenstone et al., in press, b). As the
seven families include spiders of varying proportions and shapes, we assumed that
mygalomorphs would also fit the overall regression (they certainly do not
resemble tetragnathids, the one family which did have a significantly different
volume-mass regression).

On April 7, 1984, we were presented with an opportunity to check this
assumption partially, when F.A.C. located a group of ballooning Ummidia
spiderlings near Cullowhee in Jackson County, North Carolina (Coyle 1985).
Twenty of these were weighed within 48 h of collection, preserved in 70% ethanol,
and mailed to A.-L.H. for measurement. Because these mass determinations and
those for the araneomorph volume-mass regressions were made at different times
and places, special care was taken in the calibration of both microbalances (a
Mettler M5 and Mettler 160, respectively). The twenty Ummidia were measured
four months after preservation, to ensure that the preserved volumes had come
to equilibrium [Araneids require six weeks (Greenstone et al., in press, b)].

RESULTS AND DISCUSSION

Comparison of Estimated Masses of Araneomorphs and Mygalomorphs. — Two

hundred sixteen araneomorphs were collected off the soybean field sticky traps
and measured. The frequency distribution of their estimated masses is shown in
Fig. 1. Means (and standard errors) for mass estimates of the Wake County
Ummidia and the S. atlanticus spiderlings are 3.45 mg (0.13 mg), and 1.25 mg
(0.03 mg), with N =9, and 15, respectively, and are indicated by arrows in Fig.
1. Although the araneomorph mass frequency distribution appears representative
of other weeks in the summer and fall of 1983 (Greenstone et al., unpublished
data), it would be premature to characterize the shape of this distribution in
order to derive a parametric assessment of the deviation of mygalomorph
ballooner masses from those of araneomorphs. However it is clear that the
ballooning mygalomorphs studied here are more massive than most other
ballooners, and will probably turn out to be in the ninetieth percentile or higher
(they are in the ninety-fifth and ninety-eighth in Fig. 1). On the other hand they
are not the most massive spiders ballooning. Although we cannot entirely rule
out the possibility that some of the larger trapped animals were “rappelling”,
spiders ranging in mass from 4.8 to 19.2 mg have been trapped by nets suspended
more than 100 m above the ground (animals collected by R. A. Farrow in New

294

THE JOURNAL OF ARACHNOLOGY

South Wales, Greenstone et al., MS in preparation). This suggests that in many
mygalomorph taxa ballooning may not be precluded solely or even primarily by
large mass.

Although the Sphodros and Ummidia spiderlings used in this study were not
caught in the act of ballooning, we have strong evidence that they were capable
of doing so. The age (third instar), time of collection, and behavior of the S.
atlanticus spiderlings indicate that the animals were overwintering prior to
dispersal (see Coyle and Shear 1981 for observations on the phenology of other
Sphodros spp). Twenty-five of their siblings were kept alive in a glass-topped
terrarium containing humid soil. They wandered freely over the soil for a few
days and performed pre-ballooning behavior (climbing up the corners of the
terrarium and depositing large amounts of silk at the upper ends of the corners
just under the glass lid) before finally excavating burrows and constructing their
pursewebs. Furthermore the Sphodros spiderlings seen ballooning by F.A.C.
(Coyle 1983) were almost certainly S. atlanticus. The behavior and location of
the Ummidia spiderlings at the time of collection were also indicative of pre-
ballooning behavior.

The twenty spiderlings of A. unicolor, the non-ballooning species, had an
estimated mean mass of 2.02 mg (standard error =0.19 mg), which is intermediate
between those of the two ballooning mygalomorph species. This demonstrates
that mass is not the only constraint on ballooning behavior in mygalomorphs and
suggests that for at least some species, other factors may be more important than
mass in determining whether ballooning occurs. One of us (Greenstone 1982) has
suggested that the predictability of the habitat will be the major selective factor
in the evolution and maintenance of ballooning behavior, with less predictable
habitats selecting for higher rates of ballooning. The habitats of the three species
studied here support that hypothesis. A. unicolor is most often found in mesic
forests (Coyle 1971), whereas the other two species, while sometimes found in
forests, are as apt to be found in forest edge habitats, lawns and old fields. Mesic
forests are inherently longer lived and hence more predictable than successional
habitats like lawns and old fields (Southwood 1962), and they also provide better
buffered and hence more predictable microclimates for their inhabitants. It is also
possible that the kinds of air currents conducive to mygalomorph ballooning are
prohibitively rare in forest habitats.

Comparison of Actual and Estimated Masses of Mygalomorphs. —

Measurement of the twenty Ummidia spiderlings collected on April 7, 1984, while
their siblings were ballooning (Coyle 1985) indicated that the araneomorph
volume-mass regression may not be entirely accurate for mygalomorphs. The
mean (and standard error) measured mass of this sample was 2.48 mg (0.02),
whereas the mean estimated mass was 2.93 mg (0.06). All measured masses were
between two and 35% less than those estimated from the volume-mass regression,
with a mean deficit of 17.4% (arc-sine transformation of original percent data).
If this is representative of the equation’s overestimation for all mygalomorphs of
this volume range, the arrows in Fig. 1 should be moved to the left one mass
class for Sphodros and four for Ummidia: this does not change their percentile
ranks.

The consistent overestimation of Ummidia mass by the volume-mass regression
is counterintuitive, since the volume estimate does not include the chelicerae and
legs, which look more massive in mygalomorphs than in araneomorphs. We can

COYLE £r/4L.— MASSES OF BALLOONING MYGALOMORPHS

295

think of three possible explanations for the overestimation: 1) the regression
consistently overestimates the masses of all spiders in this volume range; 2)
mygalomorph body shape differs sufficiently from araneomorph body shape that
the araneomorph mass-volume regression is not valid for them; 3) mygalomorphs,
at least in this volume range, do in fact have lower density than similar sized
araneomorphs.

The first possibility was ruled out by comparison of the actual and estimated
masses of eight animals (six araneids and two thomisids) from among the 101
used to construct the regression which happened to fall in the same volume range
(2.2 to 2.9 mm^) as the twenty Ummidia: exactly half of the estimates were
overestimates and half underestimates.

The second possibility will have to be determined by further research. The third
is most interesting. If in fact mygalomorph spiderlings are less dense than
araneomorphs of the same volume, it may simply reflect some unknown
physioanatomical difference unrelated to ballooning. On the other hand such a
capacity to be less massive at a given volume could be a preadaptation for the
evolution of still lower density to permit ballooning at larger sizes, an ability that
might permit the transport of larger energy stores, reduce the rate of body water
loss, and enhance colonizing potential by allowing older animals to balloon
(Mac Arthur and Wilson 1967). Given sufficiently strong winds and very long or

Fig. 1. — Frequency distribution of estimated masses of ballooning araneomorphs collected in a
soybean field near Columbia, Missouri, with approximate mean masses of S. atlanticus (S), Wake
County Ummidia (U) and A. unicolor (A) spiderlings indicated (arrows). Mass classes are 0.2 mg
wide. Class labels designate upper bound of each class.

296

THE JOURNAL OF ARACHNOLOGY

multi-stranded balloons, spiders much larger than those studied here should be
capable of ballooning (R. Buskirk and R. B. Suter, pers. comm.). However, since
large spiders seldom balloon, perhaps there is selection against ballooning at large
sizes due to such disadvantages as decreased lift resulting from a decreased
surface to volume (or mass) ratio, dangerously high terminal velocities, or
enhanced visibility to aerial predators. At this point we really do not understand
why the distribution of aeronaut masses is so heavily skewed to the light end.

ACKNOWLEDGMENTS

We thank W. G. Eberhard, N. L. Marston, J. S. Rovner, W. A. Shear and R.
B. Suter for valuable comments on the manuscript. A.-L. H. was supported by
the U.S.D.A.-A.R.S. Research Apprenticeship Program.

LITERATURE CITED

Baerg, W. J. 1928. Some studies of a trapdoor spider (Araneae; Aviculariidae). Entomol. News,
39(1); 1-4.

Bristowe, W. S. 1939. The Comity of Spiders. Vol. 1. Ray Society, London, 228 pp.

Coyle, F. A. 1971. Systematics and natural history of the mygalomorph spider genus Antrodiaetus
and related genera (Araneae: Antrodiaetidae). Bull. Mus. Comp. Zool., 141(6):269-402.

Coyle, F. A. 1983. Aerial dispersal by mygalomorph spiderlings (Araneae, Mygalomorphae). J.
Arachnol., 11:283-286.

Coyle, F. A. 1985. Ballooning behavior of Ummidia spiderlings (Araneae, Ctenizidae). J. Arachnol.,
13:137-138.

Coyle, F. A. and W. A. Shear. 1981. Observations on the natural history of Sphodros abboti and
Sphodros rufipes (Araneae, Atypidae), with evidence for a contact pheromone. J. Arachnol., 9:317-
326.

Enock, F. 1885. The life history of Atypus piceus Sulz. Trans. Entomol. Soc. London, pp. 389-420.
Greenstone, M. H. 1982. Ballooning frequency and habitat predictability in two wolf spider species
(Lycosidae: Pardosa). Fla. Entomol., 65:83-89.

Greenstone, M. H. 1984. Spider ballooning: development and evaluation of trapping protocols. Amer.
Arachnol., 30:11 (Abstract).

Greenstone, M. H., C. E. Morgan and A.-L. Hultsch. Ballooning methodology: equations for
estimating masses of sticky-trapped spiders. J. Arachnol., In Press, a.

Greenstone, M. H., A.-L. Hultsch, and C. E. Morgan. Effects of method and time of preservation
on volumetric mass estimates of spiders (Araneae). J. Arachnol., In Press, b.

Kraus, O. and H. Baur. 1974. Die Atypidae der West-Palaarktis: Systematik, Verbreitung, und
Biologie (Arach.: Araneae). Abh. Verk. Naturwiss. Var. Hamburg, 17:85-116.

Lockett, G. H., A. F. Millidge and P. Merrett. 1974. British Spiders. Vol. III. Ray Society, London,
314 pp.

MacArthur, R. H. and E. O. Wilson. 1967. The theory of island biogeography. Princeton University
Press, Princeton, 203 pp.

Main, B. Y. 1957. Occurrence of the trap-door spider Conothele malayana (Doleschall) in Australia
(Mygalomorphae: Ctenizidae). West. Australian Nat., 5(7):209-216.

Muma, M. H. and K. E. Muma. 1945. Biological notes on Atypus bicolor Lucas (Arachnida).
Entomol. News, 56(5); 122-126.

Southwood, T. R. E. 1962. Migration of terrestrial arthropods in relation to habitat. Biol. Rev.,
37:171-214.

Yaginuma, T. 1970. The spider fauna of Japan. Bull. Nat. Sci. Mus., 13(4):639-701.

Manuscript received November 1984, revised February 1985.

Poinar, G. O., Jr., and G. M. Thomas. 1985. Laboratory infection of spiders and harvestmen
(Arachnida: Araneae and Opiliones) with Neoaplectana and Heterorhabditis nematodes
(Rhabditoidea). J. Arachnol., 13:297-302.

LABORATORY INFECTION OF SPIDERS AND HARVESTMEN
(ARACHNIDA: ARANEAE AND OPILIONES) WITH
NEOAPLECTANA AND HETERORHABDITIS NEMATODES

(RHABDITOIDEA)

George O. Poinar, Jr. and G. M. Thomas

Department of Entomological Science
University of California
Berkeley, California 94720

ABSTRACT

Specimens of the aerial spiders, Pholcus phalangiodes and Latrodectus mactans, a ground spider,
Pirata sp., and a harvestman, Phalangium sp., were placed on damp filter paper containing the
entomogenous nematodes, Neoaplectana carpocapsae and Heterorhabditis heliothidis. Representatives
of all four hosts were killed by the above nematodes. In Pholcus, Pirata and Latrodectus, the
nematodes developed to the adult stage but did not multiply. In the case of Phalangium, the
nematodes reproduced and formed infective juveniles. The present report establishes that under ideal
conditions, neoaplectanid and heterorhabditid nematodes are capable of infecting, killing and with one
host, reproducing in arachnids.

INTRODUCTION

With the commercialization of nematodes belonging to the genera
Neoaplectana and Heterorhabditis for insect control, studies are being undertaken
to determine what effect these nematodes may have on non-insect arthropods.
Neoaplectana carpocapsae and Heterorhabditis heliothidis are known to infect a
range of insects under laboratory conditions but have never been tested against
members of the class Arachnida. The host range and biology of these nematodes
are summarized by Poinar (1979).

The present paper reports infectivity tests made with N. carpocapsae and H.
heliothidis in the laboratory against three species of spiders and a harvestman.

MATERIALS AND METHODS

For the present tests, the aerial spiders, Pholcus phalangiodes (Pholcidae) and
Latrodectus mactans, a ground spider, Pirata sp. (Lycosidae), and a harvestman,
Phalangium sp. (Phalangidae) were used.

The nematodes employed were the 42 strain of Neoaplectana carpocapsae
Weiser and the NC strain of Heterorhabditis heliothidis (Khan, Brooks and
Hirschmann).

298

THE JOURNAL OF ARACHNOLOGY

The infection chambers for Pholcus, Pirata and Latrodectus were plastic vials
(65 mm long by 25 mm in diameter) which were lined with filter paper. The inner
area of the exposed filter paper was 65 mm x 80 mm or 52 cm.^ A single spider
was placed in each vial. The inoculum consisted of 0.5 cc of infective stage
nematodes applied in an aqueous mixture to the filter paper in each vial. The
spiders were exposed to nematodes over most of the surface (except for the
bottom and top of the vial). The nematode concentrations consisted of 10.7 x
lO'^/cc for H. heliothidis and 12 x W/cc for N. carpocapsae, making the dosage
rate approximately 1028 nematodes/ cm^ for H. heliothidis and approximately
1150 nematodes/ cm2 for a. carpocapsae. Fifteen adult specimens of each spider
were used in these experiments. Six were challenged with N. carpocapsae, six
with H. heliothidis and three served as controls. In the controls, only 0.5 cc of
water was added to the filter paper.

Because of their larger size the harvestmen were placed together in containers
measuring 140 mm x 190 mm x 90 mm containing filter paper in the bottom.
The area of the filter paper was 266 cm^. Approximately 10 cc of the nematode
mixtures were added to the filter paper making the nematode concentration
approximately 4022 nematodes/ cm^ for H. heliothidis and approximately 4511
nematodes/ cm2 for N. carpocapsae. Ten harvestmen were placed in a container
with H. heliothidis, ten with N. carpocapsae and nine served as controls (with
water only).

The experiments lasted for 20 days. Water was periodically added to the filter
paper to keep the nematodes viable. At the time of death, the arachnid was
removed and a sample of blood drawn and plated out on Tergitol 7 plus TTC
(triphenyltetrazolium chloride) agar. The symbiotic bacteria that are carried by
the nematodes and released when they enter the host’s hemocoel (Xenorhabdus
spp.) (Thomas and Poinar 1979) turn a characteristic blue color on Tergitol 7
plus TTC agar. A positive color reaction from blood samples indicates a
successful infection and the probable cause of death. This test is especially useful
to determine infections when the nematode is not able to reproduce or perishes
after entering the host.

RESULTS

The results of challenging three species of spiders and a harvestman with N.
carpocapsae and H. heliothidis nematodes are summarized in Table 1. In every
category, mortality as a result of nematode activity was obtained. The controls
of Pholcus, Pirata and Latrodectus were all alive at the end of the experimental
period, yet 67% of the control harvestmen perished. None of the controls showed
evidence of nematode infection and death of the harvestmen was attributed to
possible cannibalism.

With all hosts, those that died showed the presence of Xenorhabdus bacteria
in their hemocoel shortly after death and later also exhibited mature nematodes
in their body cavities (Figs. 1-4). However, although both N. carpocapsae and H.
heliothidis were able to penetrate and develop to the adult stage in the hemocoel
of the test spiders, reproduction and the production of infective stages occurred
only in the phalangid host. With the latter, six harvestmen infected with N.
carpocapsae produced a total of 43,000 infective juveniles (ca. 7,200 per host) and
five harvestmen infected with H. heliothidis produced a total of 130,000
nematodes (26,000 per host).

POINAR AND THOMAS— NEMATODE INFECTION OF SPIDERS AND HARVESTMEN 299

Fig. I. — Pholcus phalangiodes killed by Neoaplectana carpocapsae. Adult nematodes removed from
the spider’s body are on the right side of the figure.

Fig. 2. — Detail mature females and males (arrows) of Neoaplectana carpocapsae removed from the
body cavity of an infected Pholcus phalangiodes.

300

THE JOURNAL OF ARACHNOLOGY

Fig. 3.— A Phalangium sp. killed by Neoaplectana carpocapsae. Note developing nematodes
adjacent to the opened body.

Fig. 4.— Detail of developing female Neoaplectana carpocapsae removed from the body cavity of
a Phalangium sp. Note juvenile nematodes developing inside a mature female (arrow).

POINAR AND THOMAS— NEMATODE INFECTION OF SPIDERS AND HARVESTMEN 301

Table 1. — Results of challenging three spiders and a harvestman with N. carpocapsae and H.
heliothidis.

Number of Individuals Dead After 20 Days'

Host

with N.

carpocapsae

with H.

heliothidis

control

Pholcus phalangiodes

6(100%)

3(50%)

0

Pirata sp.

3(50%)

4(66%)

0

Latrodectus mactans

6(100%)

6(100%)

0

Phalangium sp.

6(60%)

5(50%)

6(67%)

' The number in parenthesis refers to the percentage of those exposed that died.

DISCUSSION

The present tests are interesting because they show that some members of the
class Arachnida are subject to infection by neoaplectanid and heterorhabditid
nematodes.

These nematodes are able to invade and develop to the adult stage in the aerial
spiders Pholcus phalangiodes, and Latrodectus mactans, a ground spidQr{Pirata
sp.), and a harvestman {Phalangium sp.). However, only in the harvestmen was
nematode reproduction complete and infective juveniles formed.

One reason for the lack of reproduction in the spider hosts could be due to
the presence of foreign bacteria. In many hosts, a strain of Pseudomonus
aeruginosa appeared soon after the spider died. This bacterium was noted to
reproduce rapidly and fill the cadaver, thus competing with the symbiotic bacteria
{Xenorhabdus sp.) which are necessary for nematode reproduction (Poinar 197.9).

In conclusion, although spiders and harvestmen can be infected by
neoaplectanid and heterorhabditid nematodes, they are only slightly susceptible
in comparison with most insects. There is no record of a spider parasitized by
these nematodes in nature, though there are reports of spider parasitism by
mermithid nematodes (Poinar 1985).

Most spiders would normally not become infected during the mass release of
nematodes on the soil surface for insect control. The aerial spiders would be
protected by the nature of their habitat. The rapid movement of many ground
spiders would be detrimental for nematode attachment and during periods of
activity the phalangids would normally carry their bodies too high for the
nematodes to reach. Such forms could be infected only during periods of
quiescence.

ACKNOWLEDGMENTS

The authors would like to thank Pat Craig for assistance in collecting and
identifying the arachnid hosts.

302

THE JOURNAL OF ARACHNOLOGY

LITERATURE CITED

Poinar, G. O. Jr. 1979. Nematodes for biological control of insects. CRC Press, 277 pp.

Poinar, G. O. Jr. 1985. Mermithid (Nematoda) parasites of spiders and harvestmen. J. Arachnol.,
13:121-128.

Thomas, G. M. and G. O. Poinar, Jr. 1979. Xenorhabdus gen. nov. a genus of entomophathogenic,
nematophilic bacteria of the family Enterobacteriaceae. Int. J. System. Bacteriol. 29:352-360.

Manuscript received December 1984, revised February 1985.

Toolson, E. C. 1985. Uptake of leucine and water by Centruroides sculpturatus (Ewing) embryos
(Scorpiones, Buthidae). J. Arachnol., 13:303-310.

UPTAKE OF LEUCINE AND WATER
BY CENTRUROIDES SCULPTURATUS (EWING)
EMBRYOS (SCORPIONES, BUTHIDAE)

Eric C. Toolson

Department of Biology, University of New Mexico,
Albuquerque, New Mexico 87131

ABSTRACT

In vitro radiotracer studies using tritiated leucine revealed that C. sculpturatus embryos
accumulated leucine by a process that is characterized by a rate constant of approximately 0.02 min
Cyanide inhibits leucine uptake, indicating that active transport is probably involved. Leucine is
incorporated into embryonic proteins, and uptake is more rapid during early developmental stages
when growth is most rapid. Although leucine uptake continues throughout development, dry mass
of embryos does not increase during the later stages. These stages are characterized by differentiation,
maturation, and by accumulation of water. Total body water increases from 53% of body mass in
embryos with unpigmented median eyes to more than 80% at birth.

INTRODUCTION

The data on embryology and parturition in scorpions was recently reviewed by
Francke (1982), who concluded that the term ovoviviparous could not justifiably
be applied to scorpions (see also Williams 1969). Instead, he argued that the
terms katoikogenic and apoikogenic, first suggested by Laurie (1896), were more
appropriate. Katoikogenic scorpions (Diplocentridae and Scorpionidae) are
characterized by relatively small eggs, lack of persistent extraembryonic
membranes, long parturition times, and nutrition of the embryos via a placenta-
like interaction between the uterine wall and the oral cavity of the embryos.
Apoikogenic scorpions (Buthidae, Chactidae, luridae, and Vaejovidae) are
characterized by well-developed extraembryonic membranes (the amnion and
serosa) that fuse to enclose completely the embryo, short parturition times, and
relatively large eggs. It has generally been assumed that apoikogenic scorpion
embryos receive no nourishment from the mother, and thus the term
ovoviviparous has been applied, in spite of the fact that embryonic growth in
apoikogenic scorpions results in considerable increase in size (Laurie 1896).

Preliminary work in my laboratory had revealed that tritium-labelled leucine
(^H-leucine) injected into the hemolymph of gravid female Centruroides
sculpturatus could be detected in the embryos within 72 hours. The experiments
described in this paper were undertaken to analyze the kinetics of leucine entry
into the embryos and to determine whether the leucine is actually incorporated

304

THE JOURNAL OF ARACHNOLOGY

into the embryonic proteins. Data on mass changes during development were also
gathered.

MATERIALS AND METHODS

Gravid female C. sculpturatus were collected from a population occupying
cobblestone beaches along the Salt River approximately 50 km east of Phoenix,
Arizona and returned immediately to the laboratory. All scorpions were
maintained at 25° C and given ad libitum access to food (crickets) and drinking
water.

Uptake of Radiolabeled Leucine (^H-leucine). — Embryos were obtained from
females that had been sacrificed by severing their anterior mesosoma. Embryos
were dissected from the ovariuterus and placed in scorpion ringers (pH 7.4, 25° C)
solution (Ahearn and Hadley 1976). Only embryos with intact extraembryonic
membranes and, with one exception (see “^H-leucine Uptake by Early Embryos,”
below), pigmented median eyes were used.

Kinetics of ^H-leucine Uptake. — The time-course of ^H-leucine penetration into
the embryos was determined by placing three embryos into each of seven tubes
containing 2.0 ml aliquants of scorpion ringers to which had been added ^H-
leucine and *^‘I-albumin (ICN Radiochemicals). At intervals (see Table 1), the
embryos in a tube were removed and their extraembryonic membranes were
dissected free and discarded. The embryos were briefly washed with a gentle
stream of distilled water and then homogenized in a known volume (1.88-2.20 ml)
of distilled water. The ^H and '^*1 activity in a 1.0 ml aliquant of each
homogenate was determined by liquid-scintillation counting. The total ^H activity
for each set of 3 embryos was then calculated by correcting the observed counting
rate for the total volume of the respective homogenates.

Two replicate experiments were performed, the first one day, and the second
two days after collection.

Effect of Metabolic Inhibition of Embryos. — To determine if the uptake of ^H-
leucine was energy-dependent, during each replicate experiment an additional
tube containing 2.0 ml of radiolabeled scorpion ringers and sufficient NaCN
(aqueous solution) to yield a CN~concentration of 1 mM was set up. Three
embryos were placed in the tube and after 120 min incubation (at 25° C) the
embryos were homogenized’ and ^H activity was determined as described above.

Incorporation of ^H-Ieucine into Embryonic Proteins. — As part of the second
* replicate experiment, a tube containing 2.0 ml of ^H-leucine ringers and three
embryos was incubated (at 25° C) for 48 h. The embryos were then homogenized
in 0.5 ml distilled water and the proteins in a 40 pi aliquant of the homogenate
were precipitated with 10% trichloroacetic acid (TCA). The precipitated proteins
were washed with 5% TCA and ethanol/ether (50/50, v/v). The washed proteins
were solubilized and ^H* activity was determined. Total ^H activity in the proteins
of the three embryos was then estimated as described above.

^H-Ieucine Uptake by Early Embryos. — One of the females sacrificed for the
first replicate contained embryos that were at a very early developmental stage
as evidenced by their small size and the lack of visible differentiation of a
metasoma or appendages. Three of these embryos (with intact extraembryonic

TOOLSON— SCORPION EMBRYO NUTRITION

305

membranes) were placed into 2.0 ml of ^H-leucine ringers and uptake was
determined after 60 min incubation.

Mass Changes During Embryogenesis. — Seven of the gravid females were
sacrificed by exposure to CN~, and their embryos were removed and dissected
free of their extraembryonic membranes. Each female’s clutch of embryos was
assigned to one of two developmental classes, based on whether or not the
median eyes were pigmented (metasomal and appendage development was evident
in all embryos). The total mass of each clutch was immediately determined to the
nearest 0. 1 mg, after which the embryos were freeze-dried to a constant mass.

Five females were allowed to deliver broods. After all of the young had
climbed to their mother’s dorsum, the total wet and dry masses of the newborn
scorpions and their mothers were determined as described above.

Statistical comparisons involving the data were accomplished with appropriate
non-parametric and parametric tests.

RESULTS

counting rates did not significantly exceed background counting rates in
any set of embryos, indicating that the extraembryonic membranes were intact
and that large molecules were unable to penetrate the membranes at significant
rates.

The time-course of entry of ^H-leucine into C. sculpturatus embryos is
indicated in Table 1. In both replicates, embryo counting rates (CPM) increased
rapidly during the first 5 minutes, with a declining rate of increase thereafter. The
presence of the metabolic inhibitor cyanide in the incubation medium significantly
reduced the rate of leucine entry; even after 120 min incubation, embryo CPM
were less than 1/3 those of the normal embryo 45 min CPM (t = 5.20, p «
0.001). Leucine entry into the early embryos was more rapid, as evidenced by the
more than two-fold higher CPM in the earlier embryos.

After 48 h incubation in medium containing ^H-leucine, the embryonic proteins
are definitely labelled with ^H, indicating that the ^H-leucine is incorporated into
the embryonic proteins.

Estimates of rate constants for entry of leucine into the embryos may be
obtained from the data in Table 1. The time-dependent increase in concentration
of a solute entering a reservoir is described by the equation:

— ' In Coo — Ct — Equation 1

Coo

where Coo and Ct are, respectively, the solute concentrations in the reservoir at time
= (i.e., at equilibrium) and at time t after initiation of the experiment, and k

is the rate constant (Kotyk and Janacek 1975). Equation 1 also describes the
time-dependent increase in counting rate (CPM) when radiotracers are used to
monitor transport processes. In the present context, k is a measure of how
rapidly the amino acid concentration in the embryos approaches equilibrium with
the incubation medium. Numerically, it equals the proportion by which the
difference between Ct and Coo is reduced each minute. Because Ct approached Coo

306

THE JOURNAL OF ARACHNOLOGY

Table 1. — Uptake of ^H-leucine by C. sculpturatus embryoes.
embryos.

CPM =

counts per minute for 3

First Replicate

Second Replicate

Incubation

%of

%of

Time (min)

CPM

Medium CPM'

CPM

Medium CPM^

2.5

273

0.32

380

0.53

5.0

408

0.48

702

0.97

10.0

701

0.83

618

0.86

20.0

942

1.12

842

1.17

30.0

848

1.01

997

1.38

45.0

1182

1.40

1182

1.64

60.0

1348

1.60

3

—

CN

(120 min)

Protein

369

0.44

375

0.52

Incorporation
(48 h)

Early

13,238

18.34

Embryos
(60 min)

2972

3.53

'Medium CPM = 42,125 CPM ml '
^Medium CPM = 36,084 CPM ml”'
^Sample destroyed during processing

asymptotically, I used an iterative approach to estimate the value of Coo that
maximized the r^-value when Equation 1 was fitted to the data in Table 1 by
least-squares regression. The CPM values for t = 2.5 min were not used in the
regression analysis, because the data suggested the presence of a rapidly
exchanging compartment that resulted in anomalously high CPM at t = 2.5 min.
The results of this analysis are presented in Table 2.

Similarly, estimates of k for CN'-poisoned embryos may be obtained from
Equation 1. Here, however, the estimates are biased somewhat by the fact that
the effects of the rapidly-exchanging compartment cannot be factored out, and
the values of k in Table 2 are probably somewhat larger than they should be.

In a strict sense, the fact that the leucine is incorporated into proteins requires
use of a more complicated model (Kotyk and Janacek 1975). The consequence
of not including the incorporation process in the model is that the estimated
values of k for ‘uninhibited’ embryos (Table 2) are somewhat too high. The re-
values suggest, however, that the errors are not large enough to preclude use of
the k-values for purposes of discussion.

Wet and dry masses of different ontogenetic stages are presented in Table 3.
Dry mass did not change significantly during the developmental stages listed, but

Table 2. — Kinetic analysis of ^
derived by iterative fitting of the
regression.

H-leucine uptake
data in Table 1

by C. sculpturatus embryos,
using Equation 1 as a model

Parameters were
for least-squares

Coo (CPM)

k(min"'

2

r

k(CN )

First Replicate 1660

0.023

0.94

.0021

Second Replicate 1700

0.018

0.95

.0021

TOOLSON— SCORPION EMBRYO NUTRITION

307

Table 3. — Ontogenetic changes in wet and dry mass of individual C sculpturatus embryos. Values
are presented as x ± S.D.. Total number of offspring in the litters is given in parentheses below the
number of litters. Litter sizes ranged from 19 to 34 (x = 26.6, S.D. = 5.30).

Development

stage

Number of
litters

Average

Dry Mass (mg)

Average

Wet Mass (mg)

% H2O

Median eyes unpig-
mented

3

2.70 ± 0.46

5.77 ± 1.04

53.1 ± 0.6

Pigmented median
eyes

(78)

4

2.65 ± 0.64

9.99 + 2.29

73.3 ± 1.5

Newborn

(121)

6

2.50 + 0.15

12.72 ± 1.45

80.8 ± 0.7

Adults

(147)

—

—

71.2 ± 1.2

wet masses increased significantly (Kruskal-Wallis single factor ANOVA; H =9.45,
p <0.01). This reflected a greater than 3-fold increase in the water content of
embryos, from an average of 3.07 mg H2O in embryos with unpigmented median
eyes to 10.22 mg H2O at birth.

DISCUSSION

Leucine not only penetrates the extraembryonic membranes of C. sculpturatus
embryos, but is also incorporated into the proteins of the embryos at a fairly
rapid rate (Table 1). Presumably, other amino acids and nutrients such as lipids
and carbohydrates are also accumulated and utilized by the embryos. In the case
of leucine, at least, the accumulation is energy-dependent, as evidenced by the
order of magnitude decrease in k, the rate constant for leucine uptake, brought
about by the addition of cyanide to the incubation medium (Table 2). These
findings suggest that C. sculpturatus embryos are not metabolically isolated from
their mother, but rather, are adapted to utilize nutrients provided by her
throughout embryonic development. From that perspective, C. sculpturatus (and
probably other apoikogenic scorpions) is not qualitatively different from the
katoikogenic scorpions, in spite of the lack of obvious morphological
specializations for embryo nutrition.

Ontogenetic changes in amino acid uptake are also suggested by the data in
Table 1. It is not possible to calculate a rate constant for leucine transport in
the “Early Embryo” stage, but the 60 min CPM is significantly greater than the
corresponding CPM for the embryos whose median eyes have become pigmented
(i.e., the stage used for all other uptake measurements listed in Table 1). This
indicates that leucine uptake during the early stages of embryonic development
is more rapid than during later stages. As is discussed below, the early
developmental stages of C. sculpturatus are characterized by considerable increase
in dry mass, while the later stages grow very little, if at all. The higher leucine
uptake rate in the “Early Embryos” undoubtedly reflects this.

The nature of the energy-dependent process responsible for leucine uptake is
unclear. The energy-dependence could result from active transport across the
extraembryonic membranes, active transport into the embryonic tissues.

308

THE JOURNAL OF ARACHNOLOGY

incorporation of leucine into the embryonic proteins, or a combination of these.
The data presented here do not allow discrimination among the possibilities. In
mammalian small intestine, transmembrane flux of alanine (like leucine, a
neutral, lipophilic amino acid) results from a combination of passive diffusion
along concentration gradients and active transport. At low luminal alanine
concentrations, active transport is essential for amino acid uptake, whereas at
higher concentrations, passive diffusion becomes the predominant uptake mode
(Stevens, Kaunitz and Wright 1984).

The presence of ^H-leucine in the embryos that had been incubated with
cyanide indicates that leucine can diffuse through the extraembryonic membranes
and into the embryonic tissues (Tables 1 and 2). However, the large effect of
cyanide on the value of k (Table 2) makes it seem unlikely that the energy
dependence of leucine uptake stems solely from the incorporation of leucine into
proteins. Active transport through the extraembryonic membranes is probably
involved, although further work is necessary to test this hypothesis.

Accumulation of dry mass is essentially completed by the time most of the
characteristic features of scorpion external morphology (metasoma, pedipalps,
etc.) have developed (Table 3). Growth during stages is considerable. If we
assume that C. sculpturatus oocytes are spherical and similar in dimension to
those reported for C. vittatus (approximately 0.5 mm; Francke 1982), and have
a specific gravity of 1.0, then an oocyte would have a total (wet) mass of
approximately 65 pg. By the earliest developmental stage listed in Table 3, even
dry mass has increased by at least 40-fold. Dry mass changes very little during
the later stages, indicating that differentiation and maturation, rather than
growth, predominate.

There is, however, a marked increase in water content of embryonic C.
sculpturatus during the final phase of development. Water content of the average
embryo increases by more than 7 mg prior to birth. This results in an increase
in the % H2O from 53.1% to more than 80% at birth. Comparable data for earlier
developmental stages of other arthropods do not appear to be available, so the
generality of the increase in water content cannot be assessed. Water absorption,
apparently energy-dependent, does occur in insect eggs, however (Edney 1977).
The mechanism of H2O accumulation in embryonic C. sculpturatus is unknown,
but the constancy of dry mass of the embryos suggests an extraembryonic source.

The water content of newborn C. sculpturatus is higher than that of most other
arthropods, being comparable to levels reported for lepidopteran larvae (Edney
1977). The increase in water content prior to birth may be adaptive. Newborn
scorpions are, in many respects, extrauterine embryos: the sting is enclosed in a
‘sheath’, the cuticle is not sclerotized or tanned, and the young apparently do not
feed while on the mother’s back. The cuticle of newborn scorpions is probably
relatively ineffective at retarding water loss. Prior to leaving the mother and
beginning independent existence, the young molt. Molting scorpions seem to be
very susceptible to relative humidity (O. F. Francke, personal communication),
and successful completion of molting may be impossible if the scorpion desiccates
too much prior to molting. The relatively high water ‘stores’ of newborn
scorpions may thus increase the probability of completing the first molt.

The water stores could also be critical to the survival of the first instar larvae
after leaving the mother. Organisms as small as first instar C. sculpturatus have

TOOLSON— SCORPION EMBRYO NUTRITION

309

high surface:volume ratios, which makes the danger of fatal desiccation
particularly acute. Because of their flattened shape, scorpions have relatively large
surface areas for a given mass (Toolson and Hadley 1977), which would tend to
exacerbate the susceptibility of the first instar young to transcuticular water
losses. I suspect that desiccation is a leading cause of mortality for immature
scorpions, particularly during the first instar. Although I do not have evidence
that the high water content of the newborn scorpions is maintained through the
molting to first instar, if it were this would certainly enhance the probability of
surviving.

The data presented in this paper are consistent with Francke’s contention that
the term ovoviviparous should not be applied to the Buthidae or other
apoikogenic families of scorpions. Although the details of the uptake processes
remain to be worked out, it is apparent that the embryos of C. sculpturatus, at
least, are not nutritionally isolated from their mother. Embryos accumulate
leucine at the expense of energy, particularly during the earlier developmental
stages (when most of the increase in embryonic mass occurs), and incorporate the
leucine into proteins. Even during the later developmental stages, leucine uptake
continues but at reduced rates. Moreover, during the final developmental stages,
the embryos accumulate large amounts of water, which must necessarily be
derived, at least in part, from the water stores of the mother.

The demonstration that ‘large-egged’ apoikogenic scorpions provide nutrients to
the embryos raises some interesting questions about the evolution of viviparity
in the order. Little work has been done on the metabolism of gravid females or
of embryos. In the katoikogenic scorpion, Heterometrus fulvipes (C. L. Koch),
a member of the family Scorpionidae, gravid females exhibit enhanced glycolytic
activity in the hepatopancreas and “reproductive tissues” (Jayaram et al. 1978).
In the same tissues, total lipid content and total protein content decrease
significantly during embryogenesis apparently to meet the metabolic demands of
the embryos. Comparable work on apoikogenic scorpions does not appear to
have been done. Comparative studies on several species should be undertaken.
The resulting data should provide insights into the reason(s) why the two
markedly different ‘approaches’ to embryogenesis have been maintained during
the evolution of the Scorpiones.

ACKNOWLEDGMENTS

Dr. Neil Hadley provided laboratory space and equipment for this work. Drs.
Gregory Ahearn and Elliot Goldstein provided invaluable advice and technical
assistance. Dr. Clifford Crawford read and criticized an early draft of this
manuscript. I thank them all.

LITERATURE CITED

Ahearn, G. A. and N. F. Hadley. 1976. Functional roles of luminal sodium and potassium in water
transport across desert scorpion ileum. Nature, 261:66-68.

Edney, E. B. 1977. Water balance in land arthropods. Springer-Verlag, Berlin. 282 pp.

Francke, O. F. 1982. Parturition in scorpions (Arachnida, Scorpiones): a review of the ideas. Revue
Arachnologique, 4:27-37.

310

THE JOURNAL OF ARACHNOLOGY

Jayaram, V., V. Venkata Reddy, S. Krupanidhi and B. Padmanabha Naidu. 1978. Changes in organic
reserves in tissues of the scorpion Heterometrus fulvipes (C. L. Koch) with special reference to
reproduction. Indian J. Exp. Biol., 16:607-608.

Kotyk, A. and K. Janacek. 1975. Cell membrane transport. Principles and techniques, 2 Ed. Plenum,
New York. 583 pp.

Laurie, M. 1896. Further notes on the anatomy and development of scorpions, and their bearing on
the classification of the order. Ann. Mag. Nat. Hist., Series 6, 18:121-133.

Stevens, B. R. J. D. Kaunitz and E. M. Wright. 1984. Intestinal transport of amino acids and sugars:
advances using membrane vesicles. Ann. Rev. Physiol., 46:417-434.

Toolson, E. C. and N. F. Hadley. 1977. Cuticular permeability and epicuticular lipid composition in
two Arizona vejovid scorpions. Physiol. Zool., 50:323-330.

Williams, S. C. 1969. Birth activities of some North American scorpions. Proc. California Acad. Sci.,
37:1-24.

Manuscript received December 1984, revised February 1985.

Cokendolpher, J. C. and D. Lanfranco L. 1985. Opiliones from the Cape Horn Archipelago: New
southern records for harvestmen. J. Arachnol., 13:31 1-319.

OPILIONES FROM THE CAPE HORN ARCHIPELAGO:
NEW SOUTHERN RECORDS FOR HARVESTMEN

Janies C. Cokendolpher

The Museum
Texas Tech University
Lubbock, Texas 79409, U.S.A.

and

Dolly Lanfranco L.

Seccion Entomologia

Departamento de Recursos Naturales Terrestres
Institute de la Patagonia
Casilla 102-D

Punta Arenas, Magallanes, Chile

ABSTRACT

Previous subantarctic and extreme southern cold temperate island localities for harvestmen are
reviewed. The record of Spinicrus tasmanicum (Hogg) from South Georgia is questioned.
Thrasychirus dentichelis Simon and Thrasychirus modestus Simon are now recorded from Isla Deceit
(55°49'S latitude), and T. dentichelis alone from Isla Bayly (55°37'S) and Isla Wollaston (55°44'S),
Cape Horn Archipelago.

RESUMEN

Se revisan las localidades ya conocidas de los Opiliones recolectados el las islas del extremo sur
templado-frio e islas subantarticas. Se pone en duda la validez del registro de Spinicrus tasmanicum
(Hogg) en Georgia del Sur. Thrasychirus dentichelis Simon y Thrasychirus modestus Simon se citan
aqui’' por primera vez de la isla Deceit (55°49'S) y T dentichelis se cita ademas de la isla de Bayly
(55°37'S) e isla Wollaston (55°44'S), en el archipielago del Cabo de Hornos.

INTRODUCTION

Although Opiliones are reported (Marx 1892) as far north as 81°44' (Fort
Conger, Ellesmere Island, Canada), there are few reports from extreme southern
latitudes. There are no reports of harvestmen from Antarctica proper and there

312

THE JOURNAL OF ARACHNOLOGY

c/5

■o

c

03

03

<u

D,

B

(D

o

o

-C

3

O

c/5

(U

6

(U

X

<u

T3

C

03

3

w

C

c3

3

c/5

6

o

T3

(U

■4— »
l-l

O

CL

(U

(U

3

_o

'S.

o

3

03

H

u

4U

4U

4U

CJ

(U

4U

OJ

(U

<u

■4—4

•4—4

■4—4

■4—4

■4-4

■4=4

■4—4

•4-4

■4-4

■4—4

■4-4

3

3

3

3

3

3

3

3

3

3

3

3

3

3

3

1-4

1-.

1-

O

u

Lh

u.

1-4

1-.

Ul

u

1-

L-

1-4

1-4

(U

<U

<u

w

o

P)

<u

<u

a>

OJ

OJ

<u

lU

(U

<U

CP

CP

CP

0

1 ,

CP

CP

CP

CP

CP

CP

CP

CP

CP

CP

CP

CP

(U

6

B

B

S

S

B

6

s

B

B

B

S

£

B

S

w

a>

<u

<D

4^

(U

(D

(U

<u

<u

(U

<D

(U

<u

■4-4

■4—4

C

■^

■-^

■4—4

■4—4

■4—4

■4-4

■4—4

■4-4

■4-4

■4-4

■4-4

■4—4

6

X)

2

2

2

2

2

2

2

2

2

’o

'o

"o

D

o

"o

"o

o

2

o

2

O

2

2

2

'o

u

40

o

o

c/3

O

o

o

o

o

o

o

O

o

o

o

o

CJ

(U

Dh

c/1

6

3

UU

Os

os

c

3

B

o

\0 sD lO

Os OS OS OS

-3-

so

OS

3

•4—4

(U

^ '3‘ '3’
m >^5

o\

<D ^ (D

O

o

u.

o

Ul,

CJ

o

>5

hJ

-D

S

3

H

'3'

OS

OS

>^5

•3'

OS

Os

OS

C/l

c/5

C/l

OJ

(L>

OJ

L4

l-

1-4

3

3

3

O

O

O

CO

CO

C/2

0^

c/5

c/5

c/5

OJ

W

(U

u

L-

1-4

3

3

3

O

O

O

CO

C/2

C/2

CO

u

X

o

OS

OS

Os

«s

X

k-

X

'3'

■3'

-'Ct

OS

o

OS

d

SO

X

X

so

X

D,

—

OS

OS

OS

Os

OS

■4—4

3

■4—4

_o

«

"

2

CP

2

k-

k-

k-

u

k-

2

•4=4

dJ

o

<u

(U

<u

OJ

3

c

3

w

■4-4

C«

■4—4

C/5

•4-4

C/5

-*-*

CO

■4—4

1/5

aa

<u

ao

k-

k-

k-

u

k-

c

c/5

c

o

O

O

O

o

L-

<U

pp

PP

PP

Uh

pp

5

D-

5

u

3

O

o

O

«~n

CO

o

o

0

X

0

X

<u

X

X

X

T3

r-

44

i

^ 4~

3

f/S

fSJ

fN

o

O

rsj

(N

rsi

X

o

o

rn

rn

rn

—

X

X

X

X

X

X

o

o

0

o

0

o

0

o

o

0

o

o

o

o

o

CSJ

O

o

X

X

rs)

<N

o

O

o

rsi

oo

•3‘

■3-

2

X

X

X

■3’

X

X

X

X

X

X

X

•3'

X

X

C/l

•o

C

3

(U

XI

CL

c/5

c/5

3

c/5

c/l

c/5

~o

T3

3

"O

■3

■o

T3

c

C

3

3

3

3

3

_3

_3

u

05

-TD

C/5

cd

c/5

cd

2

2

2

2

T3

"c^

c/5

C

c

c

c/l

c/5

on

c/l

C

<u

o

c

T3

“0

•o

c/5

(U

■4—4

"O

TS

T3

2

CO

<3^

2

C

3

C/5

cd

3

3

3

2

U

X

2

2

W

O

2

2

2

X

CO

D

D,

2

(U

N

K

2

2

2

CP

O

o

G

3

O

3

o

3

O

u

2

CO

CO

o

3

o

3

o

3

G

3

c

o

u

<

<

u

c/5

CO

<

<

<

u

C/)

P-1

<

Q

s

u

>

z

o

z

w

<

5

H

(U

c/5

L-

o

Ph

I

a

•2

3

C

O

4^ 45e

_c

'_S

00

° c

I (S

2

a

§ bu
a w

2 H

•^ c >■

^ ^
< < O

O

•2

c

3

C

o

w

<

Q

H

3

o

S' o
•2 W
s ^
-2 2
^ c

3

L)

oc

ao

o

X PP

r <

■S Q

2

o

P-

>

3

-2

>

-2

-2

'2

ac

"33

s •-

<
oo
CP

o

J

<

U 4^

o

pp ^
2
3

iio

- 2
^ -3

o

O

PP

o

4)

■4-4

C/5

Lh

o

pp

o

I

O

>/^

<u

o

DC

3

(U

■4-4

C/1

o

X

o
aa

c« 3

- "aj
CP

c/5

O

T3

3

■4-4

C/1

PP

C/1

s

(U

T3

O

X

4—4 3

„ >
C/5 3

o 2

■o

3 _3
c/5 c/i

PP ^

0/5 <U

o

fU

CP

3

P2 U

3

'i-' >T«

o pp

3 3

5: PP

3 3

c^ oo

o o

3 3

3 3

0, 4^

Cp

a

■4-4 ■^^

3 3

/2 2

4)

■4-4

C/5

L-

O

PP

C/1

3

400

^ w

3
3

Co

Z

2 o

o' d

<

Q

X

O

PP

z

c

o

_S

-3

•2

3

■2

c<5

3

.1^

P)

3^

c

= c
o B

B

3

Co -4.-

3 ^
O

"2 2
ac 3

40

3

-3 >3
P) 3

40 O

3 3

L. !>.

Isla Deceit 55°49' cold temperate present paper

COKENDOLPHER AND LANFRANCO L.— OPILIONES FROM CAPE HORN

313

are only two species reported from true subantarctic habitats (one record of
which is probably in error). The known records of Opiliones from subantarctic
and southern cold temperate localities are listed in Table 1.

Roewer (1956) reported Spinicrus tasmanicum (Hogg), family Megalopsalidi-
dae, from “Siid-Georgien — 3(5, 3?.” Although South Georgia lies at 54°00'-50',
it is within the Antarctic Convergence and is a true subantarctic island. As the
only other records of 5*. tasmanicum are from Tasmania (Roewer 1956, Hickman
1957), South Georgia is an unlikely locality for this species. We suspect the South
Georgia collections are mislabeled. Further support for our suspicion comes from
the knowledge that numerous specimens from the Roewer collection are believed
to be mislabeled (see Cokendolpher and Cokendolpher 1984, Levi 1983, and
citations contained therein). Furthermore, no other collections of Opiliones are
known from South Georgia (Tambs-Lyche 1954, Gressitt 1970), making Roewer’s
collection of three pairs appear unrealistic.

Until now no harvestmen have been reported from the Cape Horn Archipelago.
The most southern records from South America were previously reported from
Isla Navarino and Isla Hoste by Simon (1884, 1902). We have two species of
neopilionid harvestmen which were collected from the Cape Horn Archipelago:
Thrasychirus dentichelis Simon and Thrasychirus modestus Simon. Our records
not only extend the known ranges for these species, but also set a new southern
latitude for the Order.

THE STUDY AREA

The eight main islands making up the Cape Horn Archipelago are situated
between 55°S-67°W and 56°S-68°W. Despite their extreme southern latitude they
lie north of the Antarctic Convergence and have a relatively mild climate
compared to other islands of similar latitudes (Figs. 1, 2). The Cape Horn
Archipelago has been included in the Magellanic Tundra Complex because of its
climatic, floristic and vegetational characteristics (Pisano V. 1977, 1983).

There are no meteorologic stations on this archipelago. The nearest stations are
located at Isla Navarino (Puerto Williams) and at Islas Diego Ramirez (Isla
Gonzalo). The climate of the Cape Horn Archipelago is similar to that of Diego
Ramirez. However, the latter supports an Isothermic Tundra clime with a high
oceanic influence, which is more moderate at the area studied (Zamora and
Santana 1979, Pisano V. 1980). Nevertheless the most reliable data registered
from nearby the Cape Horn Archipelago are those obtained from October 1882
to August 1883 by the French Scientific Mission to Cape Horn at Bahia Orange,
Isla Hoste (Lephay 1887). From interpolation of these data, Pisano V. (1983)
reported 1,357 mm of annual rain and 5.2° C of mean annual temperature for the
area. The relative humidity fluctuates from 87 to 93%. Another important
climatic factor is the wind, especially that from the SW. This element has a high
constancy in the archipelago and a mean monthly speed of approximately 30km/
hr, with maximal squalls of over 100km/ hr during any season.

Because of these characteristics the area of Cape Horn can be included in the
Isothermic Tundra type of climate, with the notation ETi(w')k'c, from the
Koppen classification (Llorente 1966). It is snowy during winter (E); supports a
Tundra vegetation type (T); is isothermic with a difference between the warmest

314

THE JOURNAL OF ARACHNOLOGY

Fig. 1. — Map of southern tip of South America, showing Cape Horn Archipelago and other
localities where opilionids have been reported.

and the coldest month medium temperatures of 5°C (i): its rainiest season,
although not so marked, is autumn (w'); it is very cold, with a mean monthly
temperature lower than 18°C, but those of the coldest months are higher than
-18°C (k') and during the four warmest months the mean is lower than 10° C (c).

The distribution and floristic composition of the vegetational communities
depends primarily on the edaphic characteristics, such as the low availability of
organic nutrients, the differential capacities of the soils to retain water, and high
organic acidity. In the study area, the dominant vegetational formations are
moorlands and dwarf forests communities. Bogs can be found at low or high
elevations, in flat or slightly undulating sites with poor drainages. There are three
basic types of bogs based on floristic composition: cyperoideus, cushion and
mossy. Intermediate forms are frequently encountered and were sampled on
Bayly, Wollaston and Deceit islands. The cyperoideus bog develops mostly at low
elevations, with dominant species such as Schoenus andinus (Phil.) Pfeiffer being
associated with Carpha alpina R. Br. var. shoenoides (Banks & Sol. ex Hooker
f.) Kiiken, Schoenus antarcticus (Hooker f.) Dusen and the Juncaceae
Marsippospermum grandiflorum (L. f.) Hooker f. In the lower stratum cushion
growing species predominate. They form large convex cushions of low height with
caespitose species such as Astelia pumila (Forster f.) Gaudich. (which dominates),
Donatia fascicularis Forster & Forster f.. Call ha dionaefolia Hooker f., Drapetes
muscosus Banks ex Lam., Gunnera lobata Hooker f., Phyllachne uliginosa

COKENDOLPHER AND LANFRANCO L.— OPILIONES FROM CAPE HORN

315

Fig. 2. — Map of Cape Horn Archipelago, showing collection sites of Opiliones.

Forster & Forster f., Tetroncium magellanicum Willd, Tapeinia pumila (Forster
f.) Baillon, Gaimardia australis Gaudich and Oreobolus obtusangulus Gaudich.
Mosses, ferns and liverworts form a very dense cover in moist areas with high
humidities. At drier sites, herbaceous species (mostly cyperoideus) dominate.

The forest communities grow in wind protected places like gorges, hill slopes
and the fringes of the valleys. The floristic composition and physiognomy changes
depend upon the wind exposure (Pisano V. 1980). The arboreal level is formed
almost exclusively by Nothofagus betuloides (Mirbel) Oersted, associated in some
sectors with a few specimens of Drimys winteri Forster & Forster f. These species
form a semidense stratum with tortuous trees of low height and flat treetops (due
to effects of winds). The forest shrub layer is composed by Berberis ilicifolia L.
f., Pernettya mucronata (L. f.) Gaudich. ex G. Don. and a few specimens of
Berberis buxifolia Lam., Chiliotrichium diffusum (Forster f.) Kuntze, Empetrum
rubrum Vahl ex Willd. and Escallonia serrata J. E. Sm. Several species of ferns,
lichens, mosses and liverworts constitute the basal level. The ground cover of
dead leaves does not exceed 7 cm in depth.

THE OPILIONID FAUNA

Pitfall Barber traps with 7% formaldehyde, without attractive fats, were used.
The traps, in each of the studied islands, were located in two sectors of forest
and bog communities. All of the material captured was stored in 70% isopropyl
alcohol. The localities and dates samples, and number of traps used were: (1) Isla
Bayly, Surgidero Romanche (55°37'S-67°33'W) 28 Feb. -4 March 1980, 20 traps;
(2) Isla Wollaston, Caleta Lientur (55°44'S-67° 19'W) 15-25 Feb. 1980, 20 traps;

316

THE JOURNAL OF ARACHNOLOGY

(3) Isla Deceit, Caleta Toledo (55°49'S-67°06'W) 18 Nov.-3 Dec. 1982, 80 traps
(Lanfranco L. 1980, 1981, in press).

Specimens were identified by the use of taxonomical keys by Ringuelet (1959)
and Cekalovic K. (1976). Species identifications were verified by comparisons to
type specimens (as part of a generic revision). Identified specimens are deposited
at the Instituto de la Patagonia, American Museum of Natural History and in
the senior authors personal collection.

Fig. 3-6. — Thrasychirus spp.: 3, T. dentichelis female; 4, T. dentichelis male; 5, T. modestus female;
6, T. modestus male. Scale lines = 1 mm.

COKENDOLPHER AND LANFRANCO L.— OPILIONES FROM CAPE HORN

317

Thrasychirus dentichelis Simon
Figs. 3, 4

This species was described by Simon (1884) from specimens collected from Isla
Hoste and it has been reported from several more northern localities in Argentina
and Chile (Ringuelet 1959, Cekalovic K. 1976). Recent studies of museum
specimens reveal that the specimens reported as T. dentichelis from Valdivia and
Santiago provinces of Chile and the Nahuel Huapi region of Argentina are
representatives of two different undescribed species of Thrasychirus Simon.
Thrasychirus dentichelis is restricted to the cold forest of the Magallanes province
(Chile), the Argentinean Territory of Tierra del Fuego and Islands of Cape Horn
Archipelago. The northern limit for this species is from a specimen collected at
about 50°50'S.

Thrasychirus dentichelis can be separated from T modestus (the only other
species of harvestmen from Cape Horn Archipelago) by the differences in size
and dorsal body color patterns. The dorsum of T dentichelis is covered with
numerous white spots which are absent on T modestus. The wedge shaped
pattern on the dorsum is generally sharply defined posteriorly on T modestus.
The pattern is weakly defined posteriorly or absent on T. dentichelis (compare
figs. 3 and 6). Adults of T. dentichelis are noticeably larger than those of T
modestus. The males of T. dentichelis also differ from those of T. modestus by
having enlarged chelicerae and a large apophysis on the mesal margin of each
palpal tibia distally. The genitalia of these species differ greatly. Illustrations
showing these differences will be published elsewhere in the context of a generic
revision.

New records. — CHILE: Magallanes, Isla Wollaston, Caleta Lientur, from both bog and forest
habitats (15-25 Feb. 1980, D. Lanfranco L.), 4 females, 25 juveniles. Isla Bayly, Surgidero Romanche,
coastal mixed forest, transitional scrub and moorland communities (28 Feb. -4 March 1980, D.
Lanfranco L.), 8 females, 19 Juveniles. Isla Deceit, Caleta Toledo, from bog and forest communities
(majority from forests) (18 Nov. -3 Dec. 1982, D. Lanfranco L.), 7 males, 22 females.

Thrasychirus modestus Simon
Figs. 5, 6

This species was described by Simon (1902) from specimens collected in
southern Tierra del Fuego and Isla Navarino. Cekalovic K. (1976) reports other
localities in southern Magallanes province. The northern record appears to be
Punta Arenas at about 53° lO'S (Cekalovic K. 1976).

New record. — CHILE: Magallanes, Isla Deceit, Caleta Toledo, forest communities (single specimen
from moorlands) (18 Nov. -3 Dec. 1982, D. Lanfranco L.), 4 males, 17 females.

Thrasychirus sp.

Among the collections already mentioned were numerous examples of early
instar juveniles that could not be identified. The majority of these, if not all,
probably represent examples of T. dentichelis as they lack the characteristic dark,
sharply-defined pattern on the abdomen. As the lack of a pattern on the
abdomen could also be due to the animal being recently molted we prefer not
to attach specific names.

318

THE JOURNAL OF ARACHNOLOGY

New Records. — CHILE; Magallanes, Isla Wollaston, Caleta Lientur, bog and forest habitats (15-
25 Feb. 1980, D. Lanfranco L.), 2 juveniles. Isla Deceit, Caleta Toledo, forest and moorlands (18
Nov. -3 Dec. 1982, D. Lanfranco L.), 74 juveniles.

ACKNOWLEDGMENTS

We thank Drs. Oscar F. Francke (Texas Tech University), Edmundo Pisano V.
and Claudio Venegas C. (Instituto de la Patagonia) for their comments on the
manuscript. Collections from the Cape Horn Archipelago were financed by
SERPLAC XII Region Magallanes. Partial support for the senior author was
provided by the Department of Biological Sciences, Texas Tech University.

LITERATURE CITED

Cekalovic K., T. 1976. CatMogo de los Arachnida; Scorpiones, Pseudoscorpiones, Opiliones, Acari,
Araneae y Solifugae de la XH Region de Chile, Magallanes incluyendo la Antartica chilena (Chile).
Gayana, Zool., No. 37, 108 pp.

Cokendolpher, J. C. and J. E. Cokendolpher. 1984. A new genus of harvestmen from Costa Rica
with comments on the status of the Neotropical Phalangiinae (Opiliones, Phalangiidae). Bull.
British Arachnol. Soc., 6:167-172.

Forster, R. R. 1954. The New Zealand harvestmen (Sub-order Laniatores). Canterbury Mus. Bull.,
No. 2, 329 pp.

Forster, R. R. 1964. The Araneae and Opiliones of the subantarctic islands of New Zealand. Pacific
Insects Monograph, 7:58-1 15.

Gressitt, J. L. 1970. Subantarctic Entomology and Biogeography. Pacific Insects Monograph, 23:295-
374.

Gressitt, J. L., K. P. Rennell and K. A. J. Wise. 1964. Insects of Campbell Island, Ecology. Pacific
Insects Monograph, 7:515-530.

Hickman, V. V. 1939. Opiliones and Araneae. B. A. N. Z. Antarctic Research Expedition 1929-1931.
Reports Series B, 4(5): 159-187.

Hickman, V. V. 1957. Some Tasmanian harvestmen of the sub-order Palpatores. Pap. Proc. Royal
Soc. Tasmania, 91:65-79.

Lanfranco, L., D. 1980. Estudios entomofaunisticos en el archipielago del Cabo de Hornos. 1.
Prospeccion preliminar de suelo-superficie en Caleta Lientur (isla Wollaston). Ans. Inst. Patagonia,
Punta Arenas (Chile), 1 1:281-291.

Lanfranco, L., D. 1981. Estudios entomofaunisticos en el archipielago del Cabo de Hornos. 2.
Prospeccion preliminar de suelo-superficie en Surgidero Romanche, (isla Bayly). Ans. Inst.
Patagonia, Punta Arenas (Chile), 12:229-238.

Lanfranco, L., D. in press. Estudios entomofaunisticos en el archipielago del Cabo de Hornos. 3.
Composicion y estructura de la entomofauna de suelo-superficie asociada a bosques y turbales en
Caleta Toledo (isla Deceit: 55°49'S-67°06'0). Ans. Inst. Patagonia, Punta Arenas (Chile), 14.

Lephay, J. 1887. El clima de la Tierra del Fuego. Miss. Scient. du Cap Horn. Ans. Hidr. Marina
de Chile, Ano XXH, pp. 245-277.

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.

Llorente, J. M 1966. Meteorologia. Ed. Labor. Barcelona, 286 pp.

Marx, G. 1892. A contribution to the study of the spider fauna of the Arctic regions. Proc. Entomol.
Soc. Washington, 2:186-200.

Pisano, V., E. 1977. Fitogeografia de Fuego-Patagonia Chilena. I: Comunidades vegetales entre las
latitudes 52° y 56° S. Ans. Inst. Patagonia, Punta Arenas (Chile), 8:121-250.

Pisano, V., E. 1980. Distribucibn y caracteristicas de la vegetacion del archipielago del Cabo de
Hornos. Ans. Inst. Patagonia, Punta Arenas (Chile), 11:191-224.

Pisano, V., E. 1983. Distribucion y caracteristicas de la vegetacibn en el archipielago del Cabo de
Hornos. En: Investigacibn de Recursos Naturales en el Archipielago del Cabo de Hornos y
Territorios al Sur del Canal Beagle. Informe Final. Primera Parte. Inf. Inst. Patagonia, 28:96-176.

COKENDOLPHER AND LANFRANCO L.— OPILIONES FROM CAPE HORN

319

Ringuelet, R. A. 1959. Los Aracnidos Argentines del orden Opiliones. Rev. Mus. Argentine Cien.
Nat. “B. Rivadavia”, Zool., 5(2): 127-439, pi. I-XX.

Roewer, C. F. 1956. Uber Phalangiinae (Phalangiidae, Opiliones Palpatores). (Weitere Weberknechte
XIX). Senckenberg. Biol., 37:247-318.

Simon, E. 1884. Arachnides recuellis par la Mission du Cap Horn en 1882-1883. Bull. Soc. Zool.
France, 9:117-144, pi. III.

Simon, E. 1902. Arachnoideen, excl. Acariden und Gonyleptiden. Ergeb. Hamb. Magalh. Sammelr.,
2:1-47.

Soares, B. A. M. and H. E. M. Soares. 1949. Monografia dos generos de Opilioes Neotropicos II.
Arquiv. Zool., Sao Paulo, 7:149-240.

Soares, B. A. M. and H. E. M. Soares. 1954. Monografia dos generos de Opilioes Neotrdpicos.
Arquiv. Zool., Sao Paulo, 8:225-302.

Tambs-Lyche, H. 1954. Arachnoidea from South Georgia and The Crozet Islands with remarks on
the subfamily Masoninae. Scientific Results of teh Norwegian Antarctic Expeditions 1927-1928 et
sqq., instituted and financed by Consul Lars Christensen. Norske Videnskaps-Akad., No. 35, pp.
5-19.

Zamora M., E. and A. Santana A. 1979. Caracteristicas climaticas de la costa occidental de la
Patagonia entre las latitudes 46° 40' y 56°30'S. Ans. Inst. Patagonia, Punta Arenas (Chile), 10:109-
144.

Manuscript received December 1984, revised February 1985.

.♦i \JZ‘J4

I , *,.1 ,>r,

■ 1 'I'.

l!!l!

'■ k.j,

#9: 4-1* 'Vi’ ..*1 tW vV“

,I#P 1 :^, ■,V,r ' , ; ’ '■ j,

rtfeAiif..* ‘'’V^'^' ^<f:i >h4' ^ '■

JL' .V/> ’'f^ illsS^.Vs'f ' r * ^,.- iwiiijUirnv f

iiifr,.v li;;!rvC>i: :;. : :'s^X' ',’;..fi&'’4«wS<‘tmisff <^:M' m*;>X . • ‘‘O.

;[£,■>_ ^ A .' .' ^ ' .rM

911'

h li f» ^'*' '’Vn/V

■K ■

•Tjr

"i?:'

' ,'. , n( . I *

*v«' '^^■,'.'>■■^^,#11 • ■'

,< •,^,,tif,,*frf*/' • i ^■, I ''j'.r.-'ifW .r !-•■'-«*► '' '

', '■ "'^ '’ 'li^ r"* 'j[, Tf»ij

•f;

% *,» W ■I.T^ Ifli

— . „'/i,f'''' I ■ -,>

M

'■ ' ■'■ ^ :■ ^ ■'

!•■' lt;^(i|'«»' f4i *4*,' ’ '■' t V. « 44 ( i I. *\

•'y/^ rs,-.' ' ' ;■! .Jf*';. . t I ; >W'i*

I **

, ■‘T. , ;' ^.'^m ■■’^■,, ../^

•

'r

IM' ' .i<-,^i ij4i|y|BHHMkr ':pifp6kf.uK 'V'^

■ ■ ..-I- .' ■^"' »». ■ '"<4i

. . •

|,vt a f t > f i ‘ ■ .,, .jg^i'^;' •■ - *’“*|'1'‘'' ',

!;.!> -,

tf *

^ V • 'T > ,Tl\

&

}-tt 1 1

I . 'f

“ i ■’L

Martyniuk, J. and D. H. Wise. 1985. Stage-biased overwintering survival of the filmy dome spider
(Araneae, Linyphiidae). J. Arachnol., 13:321-329.

STAGE-BIASED OVERWINTERING SURVIVAL
OF THE FILMY DOME SPIDER
(ARANEAE, LINYPHIIDAE)

John Martyniuk*

Department of Biology
Tufts University
Medford, Massachusetts 02155

and

David H. Wise

Department of Biological Sciences
University of Maryland Baltimore County
Catonsville, Maryland 21228

ABSTRACT

Overwintering success of the filmy dome spider Prolinyphia marginata (Koch) [= Neriene radiata
(Walckenaer) and Linyphia marginata Koch] was observed with respect to stage of development and
differences in body weight within a stage. Studies conducted in three different areas measured winter
survivorship rates of 60 spiders housed in outdoor cages (Michigan), 243 spiders in outdoor jars
(Maryland), and 245 spiders in a combination of cages and jars (New York). The overall proportion
of spiders successfully overwintering was 0.31 in New York, 0.69 in Maryland, and 0.75 in Michigan.
All three studies showed that older stages (instars) had substantially higher survival rates than younger
stages. Comparison between the autumn weights of overwintering survivors and non-survivors
indicated that differences in spider weight within a stage had no significant influence on winter
survival.

These experimental studies suggest that overwintering mortality may be significant in natural
populations of P. marginata, and that different overwintering survival rates among stages can alter
the composition of the population.

INTRODUCTION

Not much information is available on overwintering mortality in spiders.
Schaefer (1977) studied the winter ecology of several species and found that
winter mortality was not correlated with overall generation mortality. Buche

*Current mailing address: 1 160 N W North River Dr., #14, Miami, FL 33136.

322

THE JOURNAL OF ARACHNOLOGY

(1966, cited by Toft 1976) discovered that the smaller stages (instars) of linyphiids
do not survive cold temperatures as well as later stages. On the other hand,
Schaefer found that only adults of the linyphiid Centromerita bicolor suffered
substantial winter mortality. Other information is available (e.g. Kirchner 1973,
Edgar and Loenen 1974), though we lack much basic data, especially considering
the potential importance of winter mortality as a key factor in the dynamics of
spider populations. Not much is known about the relative susceptibility of the
younger stages to winter conditions, despite the fact that populations of over 70%
of temperate spider species overwinter, at least partially, in juvenile stages
(Schaefer 1977).

The filmy dome spider, Prolinyphia marginata (C. L. Koch) [- Neriene radiata
(Walckenaer) = Linyphia marginata, C. L. Koch], is a sheet-web building
linyphiid common in North America. The species overwinters in forest leaf litter
(J. Martyniuk and D. H. Wise, pers. obs.), as do many temperate linyphiids
(Schaefer 1977, Aitchison 1978). Census data for a Michigan population of P.
marginata indicate that winter disappearance is significant in this population
(Wise 1976). The complex phenology of the filmy dome spider produces an
overwintering population that is a mixture of subadult stages, ranging from
recently emerged spiderlings to large, late-stage juveniles (Wise 1976, 1984). This
mixture of stages reflects within-season differences in the timing of reproduction
and development patterns of young spiders, and also reflects the effects of
variable prey abundance upon rates of growth and development (Wise 1975,
submitted, Martyniuk 1983).

This paper reports the results of three separate studies, each with a different
population of P. marginata. Considered together, the studies examined the
relative overwintering survival of different stages and the influence of weight
differences within a stage on winter mortality. Study I was conducted by J.
Martyniuk, and Studies II and III by D. H. Wise.

STUDY I

Methods. — The population that was studied occurred on the Nature Preserve
of the State University of New York at Binghamton, Broome Co., New York.
During February 1981, leaf-litter samples were collected and examined in order
to document the presence of P. marginata in the leaf litter during the winter. Five
30 cm X 30 cm forest floor plots were excavated and removed (10 cm depth) from
known P marginata habitat and transported to the laboratory in plastic bags.
Three viable, immature P marginata were found in the samples. It was not
possible to determine the exact location of the spiders in the litter due to the
disruption of excavation and transportation.

To determine overwintering mortality in P. marginata, 145 spiders in 1981-82
and 55 spiders in 1982-83 were individually housed in 400 ml glass jars containing
ca 5 cm of leaf-litter and soil. The jars were securely covered with fabric netting
and placed next to a wall on an open terrace. A sheet of plywood was suspended
10 cm above the top of the jars, minimizing snow accumulation and possible
drainage problems. As a control for this artificial situation, in 1981-82 three 30
cm X 30 cm x 30 cm net cages, each containing soil and leaf-litter with 15 spiders,
were embedded 10 cm into the ground in an area known to harbor P marginata.

MARTYNIUK AND WISE— OVERWINTERING SURVIVAL OF A LINYPHIID

323

The mesh size of the cages was large enough to allow snow to accumulate inside.
Leaf litter and soil used in the jars and cages were collected from the field and
examined for resident R marginata prior to use in experimental containers. Jars
were placed outside and spiders were introduced into the cages with the final
disappearance of R marginata in the field, 1 November 1981 and 5 November
1982. Likewise, jars and cages were examined for survivors with the reappearance
of spiders in the field, 18-28 April 1982 and 15-25 April 1983. Following these
monitoring periods, the soil and leaf-litter were inspected in the laboratory for
further survivors.

To investigate further the degree of similarity of the jar environment to that
of field conditions, jar leaf-litter temperatures were compared to actual leaf-litter
temperatures measured in the field. The leaf-litter temperature and snow covered
in 13 previously identified R. marginata web-sites were monitored from 23
January 1981 to 24 February 1981. Glass-encased mercury thermometers, which
had been calibrated with a total emersion, mercury/ nitrogen thermometer over
a range of temperatures including those encountered in the field, were placed ca
5 cm down in the litter. Litter temperatures and snow covered at these sites were
recorded daily (1200 hours), and the total emersion, mercury/ nitrogen
thermometer was used to record the daily ambient air temperature. Similarly,
from 13 January 1983 to 24 February 1983 the litter temperature of a sample jar
was recorded every six days (1200 hours) using the total emersion mercury/
nitrogen thermometer.

Spiders used in the study were collected one week prior to the start of the
experiments. For each collection period (1981 and 1982), a 20 m x 20 m plot was
marked in R. marginata habitat and all observed individuals were captured. The
cephalothorax width of the collected spiders was measured so as to identify the
particular stage of development for each individual (Martyniuk 1983). Thus
samples for both years reflect the composition of stages in the population
entering the winter. No stage I, II or III spiders were found in 1982, whereas 15
stage III spiders were collected in 1981, but escaped shortly after the start of the
experiment. Fifteen stage I and 15 stage II individuals, captured in 1981, were
combined into one group to increase the sample size. Such variation in
population structure is characteristic of this species (Wise 1984). None of the
stage VI spiders used in these studies were mature.

To determine if there was a difference in average weight between entering and
surviving spiders within a stage, 20 spiders in 1981, and 10 spiders in 1982, all
stage V, were individually weighed to provide a direct comparison of the autumn
weight between survivors and non-survivors.

Results. — Overwintering survival rates for the various developmental stages of
R marginata are presented in Table 1. The combined data for all three
experiments indicate an average overwintering survival rate of 0.31 (76/245).
However, mortality was not evenly distributed among stages. Older stages, V and
VI, had a much higher success rate than younger stages (x^- 37.03, p < 0.001,
2X4 contingency table).

Differences in spider weight within stage V seem to have had little influence
on overwintering success. Analysis of Variance shows no significant difference (F
= 1.25, p > 0.05) between the average autumn weight of 19 stage V survivors,
2.6 ± 0.6 mg.

324

THE JOURNAL OF ARACHNOLOGY

Table I. — Overwintering survivorship of P. marginata in jars and cages, Study I.

1981-82 (jars)

1981-82 (cages)

1982-83 Gars)

Stage

n

Proportion

Surviving

n

Proportion

Surviving

n

Proportion

Surviving

1,11

30

0.03

-

-

-

_

IV

35

0.06

15

0.13

20

0.25

V

55

0.42

15

0.47

20

0.65

VI

25

0.44

15

0.40

15

0.33

TOTAL

145

0.26

45

0.33

0.42

The overwintering field parameters of leaf-litter temperature, snow cover, and
ambient air temperature from 23 January 1981 to 24 February 1981 are depicted
in Figure 1. Comparison of survival rates in jars and cages, in conjunction with
the recording of leaf-litter temperatures, suggests that conditions in the jars
adequately mimicked overwintering field conditions. Analysis of variance showed
no significant difference (F = 4.05, p > 0.05) between the average temperature
of litter in the field, -2.6 ± 2.3 °C, and the average temperature of litter in the
jars -0.1 ± 5.7°C. Elimination of stage I and II data from the 1981-82 jar study,
since they were not included in the cage study, produces a survival rate of 0.31
compared to the 0.33 rate found for the field cages and 0.42 in the 1982-83 jar
study. Z-score comparisons show no significant differences (p > 0.05) between the
1981-82 survival frequencies in cages and jars for stages IV, V and VI.

STUDY II

Methods. — A total of 242 immature spiders was collected from two Maryland
populations at the end of October 1982. One population inhabited an oak forest
on the Liberty Watershed, 40 km northwest of Baltimore. The second site was
a predominately oak forest located 35 km south of Liberty on the Patuxent
Wildlife Research Center, near Laurel. Details on the phenology and size
structure of both populations appear elsewhere (Wise 1984).

The spiders were assigned to stage based upon the relationship between length
of the fourth tibia and developmental stage (unpubl. data). The two youngest
stages have been combined for the analysis since only two Stage I spiderlings
were collected. Stages V and VI were pooled because many spiders cannot be
assigned unambiguously to one or the other of these stages when tibia length is
the criterion. The spiders were kept in the laboratory until 10 November, when
they were placed individually in 240 ml glass jars containing a layer of
vermiculite, crumpled tissue paper to mimic leaf litter and a wire framework for
web attachment. The top of each jar was covered tightly with fine-meshed nylon
cloth. The jars were placed on shelves on a screened porch in the forest at
Patuxent. On 11 November each spider was given three (stages I-II) or six (stages
III-VI) fruit flies. On this date all spiders were alive and many were in small
webs.

The jars were left on the porch through the winter, and were examined for
living spiders on 20 March 1983. By this date P. marginata had appeared on webs
in the forest at Patuxent.

MARTYNIUK AND WISE— OVERWINTERING SURVIVAL OF A LINYPHIID

325

E

I

h-

O.

LU

Q

o

z

Fig. 1. — The overwintering field environment of Prolinyphia marginata from 23 January 1981 to
24 February 1981. Average daily leaf litter temperature, snow cover, and ambient air temperature for
13 sites. Study I.

Results. — The overall survival rate (stages I-VI combined) was 0.69 (168/242).
Probability of surviving the winter was not the same for all stages (x^- 42.8,
p < 0.001, 2x4 contingency table). The youngest stages (primarily stage II)
suffered a markedly higher winter mortality than the larger stages (Fig. 2).

STUDY III

Methods. — The third study was conducted 1972-73 in an oak forest on the E.S.
George Reserve, Livingston County, Michigan, in P. marginata's natural habitat.
Detailed information on this population of the filmy dome spider appears
elsewhere (Wise 1975, 1976).

THE JOURNAL OF ARACHNOLOGY

326

1.0

I - H IK F ¥-YI

STAGE

Fig. 2. — Overwinter survival of different stages, Study II.

The overwintering study was part of an experiment to determine the influence
of feeding rate on rate of development. Sixty immature P. marginata were
collected from various sites in the forest, weighed and placed individually in
cylindrical cages ca 15 cm high and 10 cm in diameter, made of hardware cloth
with openings small enough to retain fruit flies. The tops were covered with nylon
cloth. The cages, which contained a layer of soil and leaf litter, were placed in
leaf litter on the forest floor. Levels of soil and leaf litter were the same inside
and outside the cages.

Each spider was randomly assigned to one of three feeding treatments: High
(60 fruit flies/ 3 days). Medium (6 fruit flies/ 3 days), and Low (0 fruit flies/ 3
days); the only prey available to these spiders were very small insects that
penetrated the screening, or insects that emerged from the soil). The spiders were
fed from 3 August through 27 September. On 6 October they were removed,
weighed and returned. Filmy dome spiders are found in webs through the end
of October in this part of Michigan (Wise 1976); however, supplemental feeding
was not continued through October, primarily because by then the weather is
cold enough that feeding and growth in natural populations are probably
minimal. The cages were searched for living spiders on 30 March, when P.
marginata began to appear in webs on natural vegetation.

Results. — The overall survival rate over the winter was 0.75 (41/55). The
smaller spiders suffered a significantly higher winter mortality rate (Table 2; x^-
24.4, p < 0.001, 2x3 contingency table). Apparently the size within a stage did
not affect survival, since within the Low feeding treatment the October weight

MARTYNIUK AND WISE— OVERWINTERING SURVIVAL OF A LINYPHIID

327

Table 2. — Results of study III.

Feeding treatment

Low

Medium

High

August 3

No. alive

20

20

20

Weight (mg)

0.90 ± 0.05

0.97 + 0.06

0.91 + 0.06

October 6

No. alive

18

18

19

Weight (mg)

1.59 + 0.16

6.04 + 0.26

5.86 + 0.33

No. Molts 3 Aug.-6 Oct.

0.7 + 0.3

2.0 + 0.3

1.8 + 0.4

March 30

No. alive

6

16

19

Weight (mg)

1.32 + 0.14

4.71 + 0.21

4.50 + 0.26

Winter survival (%)

33

89

100

of those that survived the winter did not differ significantly from that of the
spiders that died before spring (1.50 ± 1.8 mg and 1.63 ± 0.60 mg, respectively;
t = 0.42 p > 0.5).

The rate of feeding clearly affected growth rate and rate of development, as
measured by weight gain and by the number of molts from August through
September (Table 2). In addition to being heavier, spiders in the High and
Medium feeding treatments were 1-2 stages more advanced than those in the Low
feeding group. Since linear dimensions were not recorded, it is not possible to
assign the spiders to a particular stage. However, it is clear that the larger spiders
in October were more advanced developmentally i.e., they had completed more
molts and were closer to adulthood than the smaller spiders.

DISCUSSION

Depth of the snow cover and condition of the litter affect the microclimate
experienced by overwintering spiders (Aitchison 1978). We used outdoor cages in
Studies I and HI in order to mimic natural conditions as much as possible.
Spiders overwintering in jars in studies I and H did not benefit from the
insulative effects of snow, and thus their mortality rates may have been higher
than in natural populations. However, comparison of survival rates in outdoor
cages with those in jars (Study I) suggests that mortality rates in the jars did
adequately reflect those under entirely natural conditions. No such comparison
is possible for Study H, but there is usually no significant snow cover for most
of the winter for the part of Maryland in which this study was conducted. Thus
it is unlikely that the physical conditions in the jars in this study caused higher
mortality rates than would have been observed in the natural population
inhabiting the surrounding forest.

Overall winter survival rates varied between the three studies, from a low of
0.26 in Study I (jars, 1981-82) to a high of 0.75 in Study HI. This variation likely
reflects both differences in experimental technique and differences in severity of
winters between years and sites. Similar variation appears in rates estimated from
census data of natural P. marginata populations. For example, two estimates of
juvenile survival from the first week in October to the middle of April in the
Liberty population in Maryland range from 0.24 (115/473; 1980-81) to 0.62 (190/
306; 1981-82) (calculated from census data presented in Wise 1981). Direct

328

THE JOURNAL OF ARACHNOLOGY

comparisons of survival in these three studies with census data of natural
populations are of course not justified, because estimates in natural populations
1) encompass a longer time span than the period of winter stress; 2) include
disappearances due to predation and dispersal; and 3) are subject to sampling
error, particularly with the smaller stages. Taken as a whole, though, these
experimental results and censuses of natural populations suggest that winter
mcprtality may be a significant factor in the dynamics of P. marginata
populations.

The markedly lower winter survival of the younger stages, a result consistent
with the finding of Buche (1966), is a striking pattern that was found in all three
studies. This differential survival increases the homogeneity of the size structure
of the population after the winter. For example, in the New York population
(Study I) representation in the experimental population decreased from 12%
(autumn) to 1% (spring) for stages I and II, and from 29% to 12% for stage IV.
Contrastingly, the proportion of stage V spiders increased from 37% to 57%, and
stage VI from 23% to 29%. In this situation overwintering mortality not only
reduces the overall population density, but also produces a more homogeneous
population in developmental stage.

Higher winter mortality rates among the smaller stages may have led to
selection against females that mature and reproduce late in the season. P.
marginata populations display two peaks in adult abundance, one in spring and
a second in August. The latter results from progeny of spring adults that develop
rapidly (Wise 1976, 1984). These rapidly developing spiders usually mature no
later than August. They deposit egg sacs during September and October and die
by winter. Maturation and reproduction too close to the onset of winter has
likely been selected against, because progeny of late-maturing spiders would face
the winter as smaller stages [this aspect of P. marginata"^ phenology is discussed
in more detail in Wise (1984)].

Results of Study I and III suggest that within a stage, differences in weight
do not influence overwintering survival. Perhaps physiological changes occurring
during development affect the physiology of cold resistance. Alternatively,
differences in the relative amount of food reserves, or differences in the surface
area-to-volume ratio, may better explain the lower overwintering survival of the
younger stages. More experimentation is required to answer these questions.
Particularly fruitful would be a detailed examination of the effects of size
differences within the smaller stages. The lower survival rate of the smaller
spiders in Study III may have resulted at least partly from their feeding history.
Perhaps they had fewer energy reserves at the start of the winter than the better
fed spiders.

In summary, these three experimental studies suggest that overwintering
mortality in natural populations of the filmy dome spider may be particularly
important for the smaller stages. Such stage-biased mortality rates have
consequences for population structure, and for individual fitnesses of spiders with
particular feeding histories and reproductive patterns.

MARTYNIUK AND WISE— OVERWINTERING SURVIVAL OF A LINYPHIID

329

ACKNOWLEDGMENTS

R. S. Wilcox made helpful comments on an earlier draft of a portion of the
manuscript. We would like to acknowledge the support of the State University
of New York at Binghamton; the University of Michigan; and the Patuxent
Wildlife Research Center, U.S. Department of the Interior, for providing research
sites. Study II was supported by a NSF Graduate Fellowship, NSF Grant BG-
25986 to the University of Michigan for research in systematic and Evolutionary
Biology, and a University of Michigan Graduate Student Dissertation Research
Grant. Study III was supported by NSF Grant DEB 81-19309.

LITERATURE CITED

Aitchison, C. W. 1978. Spiders active under snow in southern Canada. Symp. Zool. Soc. Lond.
42:139-148.

Buche, W. 1966. Beitrage zur Okologie und Biologie winterreifer Kleinspinnen mit besonderer
Berucksichtigung der Linyphiiden Macrargus rufus rufus (Wider), Macrargus rufus carpenteri
(Cmb.) und Centromerus sylvaticus (Bl.) Z. Morph. Okol. Tiere, 57:329-448 (cited in Toft, 1976).
Edgar, W. and M. Loenen. 1974. Aspects of the overwintering habitat of the wolf spider Pardosa
lugubris. J. Zool., 172:383-388.

Kirchner, W. 1973. Ecological aspects of cold resistance in spiders. (A comparative study). In Effects
of temperature on ectothermic organisms. W. Wieser (ed.), pp. 271-279. Springer Verlag, Berlin.
Martyniuk, J. 1983. Aspects of habitat choice and fitness in Prolinyphia marginata (Araneaef
Linyphiidae): Web-site selection, foraging dynamics, sperm competition and overwintering survival'.
Unpubl. Ph.D. Thesis, State University of New York at Binghamton.

Schaefer, M. 1977. Winter ecology of spiders. Z. Angew. Entomol., 83:1 13-134.

Toft, S. 1976. Life-histories of spiders in a Danish beech wood. Nat. Jutl., 19:5-40.

Wise, D. H. 1975. Food limitation of the spider Linyphia marginata: Experimental field studies.
Ecology, 56:637-646.

Wise, D. H. 1976. Variable rates of maturation of the spider Neriene radiata (Linyphia marginata).
Am. Midi. Nat., 96:66-75.

Wise, D. H. 1984. Phenology and life history of the filmy dome spider (Araneae: Linyphiidae) in two
local Maryland populations. Psyche. In press.

Wise, D. H. Submitted. Rearing studies with a spider exhibiting a variable phenology: no evidence
of substantial genetic variation.

Manuscript received October, 1984, revised February 1985.

.Jf;

■ •• ■ ..*■-< 1«
w. ''^J

,^.i*v.», .u, I

l%ry'- ^ iJr' ,-s^“>>i!, iv.-;, </fl^

'‘•irU- '.‘'M 5 ■'?■

’t?'

f - ■ VhI r< -'10^ »ai#^f*gJC 'J ;vv);ttMpji

■JC' '-...T^'jt^:* *iiT V. ■ ^ V3V' V'-*'

..r ^ : '„'V> ■4l;■.tM.,v^ "f‘'Vvr.’ 'I.

1. ■•JiV r.'M*uiiii.\ri 'kvc’/»t<Stf>*jlW- -:t. .1'.' *'- ,

»(''^,fe.'.‘ '^W;->i''V-\/'V'«*'#<*.%ir:\^ ; - X: :’v1“!;’f‘r-r‘.'V^^ ■'•■'’

. ■-*.■ ' •'T-i • . 'i*^, -V' ‘ ' ' -'•

pji *.PtW^ ff

CJ I -: .M % -»-.

■'■■..>^r; f r-

.’*-■■ f ‘ >,..C' j> '■

irT^-u4®rT^»Sii» ■ »’ I'- ■'■*

■r =i i^Ji^ . nHulH ' ■>. ,'V ‘ .'l*S‘tei* '’ "

-s^i" , ■ ■ 'mvtrn', \.rT\Tm

t *- » '■■ '■*■'"■ ^*2i

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

EVIDENCE OF INSEMINATION OF MULTIPLE
FEMALES BY THE MALE BLACK WIDOW SPIDER,
LATRODECTUS MACTANS (ARANEAE, THERIDIIDAE)

Robert G. Breene and Merrill H. Sweet

Department of Entomology and Department of Biology
Texas A&M University
College Station, Texas 77843

ABSTRACT

In a series of laboratory matings, adult iMtrodectus mactans males were allowed to mate with up
to three virgin females. The results demonstrate that L. mactans males are capable of successfully
inseminating at least three females and that the tip of the embolus (apical sclerite) is not required
for sperm transfer.

An escape mechanism used by the male involving possible spatial disorientation of the female in
her web during courtship and mating is described.

INTRODUCTION

Courtship and mating in Latrodectus has been studied by a number of workers
(Ross and Smith 1979, Kaston 1970, Abalos and Baez 1963, Bhatnagar and
Rempel 1962, Baerg 1959, Smithers 1944, D’Amour et al. 1936, Burt 1935, Herms
et al. 1935, Blair 1934). Despite myths to the contrary, it is well documented that
the male need not be killed by the female after mating. The incidence of escape
by the male has been correlated with the degree of hunger in the female (Ross
and Smith 1979, Kaston 1970). In addition, the ability of the male black widow
to successfully inseminate a number of females has been discredited (Foelix 1982,
Abalos and Baez 1963, Bhatnagar and Rempel 1962). As Foelix (1982) stated of
Latrodectus “the lengthy emboli often break off during copulation and remain
stuck in the epigynum of the female preventing either sex from mating again.”
Abalos and Baez (1963) noted of Latrodectus “the breaking of the apical element
(sclerite) of the male palpus is a mutilation that renders the male unable to carry
out further matings.” They claimed that if the male is not killed by the female
after mating, it perishes after a few days. They explained away as mistaken
observation reports by Herms et al. (1935) and Montgomery (1903) of males
mating more than once. Bhatnagar and Rempel (1962) conceded that in
Latrodectus curacaviensis (Muller) (= Latrodectus hesperus Chamberlin and Ivie,
Kaston 1970) the female could perform multiple matings unlike the males. They
further noted “males that have copulated can be identified on the basis of a
broken embolus, a male individual can copulate only once, for two reasons.

332

THE JOURNAL OF ARACHNOLOGY

Firstly, an embolus with a broken tip cannot penetrate a bursa, secondly, the
whole structure of the palpal organ is such that, after the first copulation, it will
not return to its original position and the organ may not be stimulated to enact
copulation again.”

To test the validity of these statements, a series of matings were staged where
males were given the opportunity to copulate with two or three virgin females.

METHODS

Spiderlings from three eggsacs spun by a laboratory reared Latrodectus
mactans (Fabricius) female that had been mated with a wild male (College
Station, Texas) were used. The spiderlings were reared to adulthood individually
from the second instar (third true instar) to ensure virginity. The group consisted
of thirty females and eighteen males. Eight males were allowed to mate once,
eight twice and two males were allowed to mate three times using the virgin
females. Once introduced into the female’s snare, the male was permitted to
remain until all courtship and mating behavior ceased, usually between three and
ten hours. Between matings, the males were placed into fresh containers where
(as with every male in which this was done) they built a normal web, and
behaved similarly to immature or unmated adult males in captivity. The males
were killed and checked for the presence of the apical sclerite following their last
mating. All eggsacs spun by the mated females were collected as they were made
and attached to the foam top of 100 ml glass vials until emergence of the
spiderlings or until the eggsac was found to be infertile. Insemination of a female
was considered successful only when living, active spiderlings emerged or were
dissected out of an eggsac spun by the female, and not by inspection of the
epigynum for traces of semen. As a result, more females were probably
inseminated than the data would account for, but this insures that only a
parthenogenetic event would provide false positive data. Parthenogenesis is
unknown in Latrodectus, although Kaston (1968) reports numerous examples of
infertile eggsacs being spun by virgin females.

RESULTS

No apical sclerite (the tip of the embolus) remained on either palpus of the first
group of males allowed to mate once. Five of the eight females of this first group
later spun eggsacs from which active spiderlings emerged. The second group of
eight males were allowed to mate with virgin females, then removed after
copulation and rested for a few days to two weeks before being allowed to mate
with another group of virgin females. One male of this second group retained
both apical sclerites, another had the left sclerite still present and the six others
were missing both apical sclerites. The male that had retained both sclerites was
also the only male unsuccessful in inseminating a female in both attempts based
on data from spiderling emergence from ensuing eggsacs. Two of the eight males
in the second group were successful in inseminating females in both matings,
three males only in the first, and one male only in the second. The third and
final group consisted of two males that were allowed to mate with three separate

BREENE AND SWEET— MULTIPLE INSEMINATION BY MALE LATRODECTUS

333

virgin females each. The apical sclerites were missing on both males and all six
females involved later spun eggsacs from which active spiderlings emerged.

DISCUSSION

The males allowed to mate once had both of their apical sclerites missing and
this loss after the first copulation is probably normal in the majority of males.
The male’s success in inseminating a second virgin female might have been due
to retention of at least one of the apical sclerites which they then used in the

Figs. 1-2. — Latrodectus mactans, male genitalia: 1, Left palpus of unmated male, apical sclerite
intact; 2, Condition of embolus after removal of apical sclerite, right palpus of mated male. Arrow
indicates the point of attachment of the apical sclerite to the balance of the embolus. Scale line =
0.05 mm.

334

THE JOURNAL OF ARACHNOLOGY

second copulation. However, the two males which inseminated three virgin
females each demonstrates that not all L. mactans males require the presence of
an apical sclerite to successfully transfer sperm. The gross morphology of the
apical sclerite appears insufficient to somehow discourage or obstruct a second
male from copulating with a previously mated female, and indeed, multiple apical
sclerites have been found lodged in the seminal receptacles of Latrodectus
(Kaston 1970, Abalos and Baez 1963), nor does it serve as a cap or plug as in
many species of Araneus (Levi 1974). Perhaps the apical sclerite functions in the
initial charging of the palpal organs with sperm.

Courtship and mating can be divided into three sequences in L. mactans. The
first sequence is one of initiation. Upon introduction to the female’s web, the
male apparently contacts female pheromone that is incorporated in her silk (Ross
and Smith 1979), which serves as a set of instructions or prime releasers,
triggering courtship behavior. Secondly is the cutting sequence. The period
between initiation and the original approach to the female is the first cutting
sequence. After initiation, the male starts cutting away and loosely wrapping
sections of the female’s web. Concentrated sheets and bands (Kaston 1970) as well
as balls are the shapes commonly observed formed by the male from the female’s
snare. The cutting sequence ends when the male slows, then cautiously
approaches the female. The mean duration of the first cutting sequence (n = 30)
was 58.5 minutes. In all our observations, the male was repulsed on the first
approach, after which he resumed a cutting sequence. If rejected a second time,
he began a third cutting sequence, and so on. Throughout the cutting periods the
male vibrates his abdomen, preparing the female for mating by putting her in a
trance (see Kaston 1970). The third sequence is the copulatory one. The female
lets the male approach, spin the “bridal veil” and inseminate her, however,
copulation can be interrupted by the female at any point, sending the male into
another cutting sequence. We suggest that the cutting behavior of the male black
widow prior to mating may serve to spatially disorient the female in her own
web, especially if the male’s silk, as is probable, contains pheromone that placates
the female (Ross and Smith 1979). This behavior could effectively “blind” the
female since her primary sensory contact with the environment is tactile and
through her web. This “blinding” may serve to enhance the male’s success in
evading an attack by the female. We also propose that the “bridal veil” spun by
the male prior to copulation, even though it may detain the female for a second
or less, is sufficient to allow the male to escape by immediately dropping out of
range, with little or no chance of successful pursuit by the female due to her
disorientation in what is now the unfamiliar territory of her cut web. Ross and
Smith (1979) working with L. hesperus suggested if the male could not
successfully mate again, its reproductive success would be best served by
providing the female with the energy of its body instead of wandering off to die
as was generally believed. Only two of the eighteen males used were consumed
by females. Both had inseminated a female in their first mating and both had
apparently failed on their second and final mating to inseminate a female. These
data, coupled with the surprising longevity and agility of the adult male black
widow, lead us to suggest that the failure of an L. mactans male to escape the
female after (or before) copulation may be due more to the physiological
condition of the male at the time of mating than to the relative hunger of the

BREENE AND SWEET— MULTIPLE INSEMINATION BY MALE LATRODECTUS

335

female. We further suggest that a larger study using the easily reared L. mactans
might reveal more realistic frequencies of many different behavioral events
observed in black widow spiders, including the number of copulations a male
may participate in over his life span. A variety of useful methods for such a
project can be found in Buskirk et al. (1984).

ACKNOWLEDGMENTS

We thank Stephen Skinner profusely for all the photographs. Susan E.
Riechert graciously reviewed the manuscript meticulously, more than once. S. B.
D. Flom and L. J. F. Hribar provided much needed support.

LITERATURE CITED

Abalos, J. W. and E. C. Baez. 1963. On spermatic transmission in spiders. Psyche, 70:197-207.

Baerg, W. S. 1959. The black widow and five other venomous spiders in the United States. Univ.
Arkansas Agr. Exp. Sta. Bull. 608, 43 pp.

Bhatnagar, R. D. S. and J. E. Rempel. 1962. The structure, function and postembryonic development
of the male and female copulatory organs of the black widow spider Latrodectus curacaviensis
(Muller). Canadian J. Zool., 40:465-510.

Blair, A. W. 1934. The life history of Latrodectus mactans. Arch. Int. Med., 54:844-850.

Burt, C. E. 1935. A review of the biology and distribution of the hourglass spider. J. Kansas Entomol.
Soc., 8(4):1 17-130.

Buskirk, R. E., C. Frohlich and K. G. Ross. 1984. The natural selection of sexual cannibalism. Amer.
Nat., 123(5):612-625.

D’Amour, F. E., F. E. Becker and W. Van Riper. 1936. The black widow spider. Quart. Rev. Biol.,
11 (2): 123- 160.

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

Herms, W. B., S. F. Bailey and B. Mclvor. 1935. The black widow spider. Calif. Agric. Exp. Stat.
Bull. 591, pp. 1-30.

Kaston, B. J. 1968. Remarks on black widow spiders, with an account of some anomalies. Entomol.
News, 79(2): 113-124.

Kaston, B. J. 1970. Comparative biology of American black widow spiders. San Diego Soc. Nat.
Hist., Trans., 16(3):33-82.

Levi, H. W. 1974. Mating behavior and presence of embolus cap in male Araneidae. Proc. 6th Int.
Arachnol. Congr., 1974:49-50.

Montgomery, T. H. 1903. Studies on the habits of spiders, particularly those of the mating period.
Proc. Acad. Nat. Sci., Philadelphia, 1903:59-149.

Ross, K. and R. L. Smith. 1979. Aspects of the courtship behavior of the black widow spider
Latrodectus hesperus (Araneae: Theridiidae), with evidence for the existence of a contact sex
pheromone. J. Arachnol., 7:69-77.

Smithers, R. H. N. 1944. Contributions to our knowledge of the genus Latrodectus (Araneae) in
South Africa. Ann. South African Mus., 36:263-312.

Manuscript received October 1984, revised March 1985.

* *w-‘-iMSk. ' • ii -tii ■ •

^■S:>. ^1/^ ihilMtd '.^iat^rii:'’ .^v bUilt^

.., ,' V .' • ^ t JS^/i 1.^-^

' ’.1’ .’'-'^J^f

Si" .

L^m

. .'U:? : ’:%fi « ? ■ . Xp' '. ,

!2*S "y'^ ^ ’ "

■,-^^v T^^ H ,„.,y ,,*\^ -H <. ' .\

: , ^ .K'l

v,r

I

]n.-‘'»'^'‘;i> ’■/ i^gli

. 'X^ .•> i

Vl V .t‘*j!

' • . , ■ . .

-

^ KTe ;

-i

' ■

}i, . ;,.

;'■“ 'V ‘

,rU:^f *.(

'•t ■

../• V'*'*.'.

•

*.-\

' ■- -;; ':

* /

■ • •

< ,ij»
/(»■

I ^

^ S.

‘ »•

■' , i-

■V.. '/

f !... ■g ifr;i‘'

'-■■W.'i'-'

• 'C

:ir'i.

, ‘5»V

' t

'. * ■ \ ')

1

\

.ii :

f s

lil', ■' '■

rt v:v.. ‘'^tb • ■ .Hf.;^;jis,w ::■:* l^’r' =r:]ii^S' i;^

!•■■ :■,'■*«; : -I. :.gi ':;'--,r^ mr^i>

^ Am! JAp I • . V ■ ***-i

■Jj4..<;'W''l:.1 ■»,w!l6j|,)iaiWiM«Jfl 5

1 ^

f ■‘S’. iV/A'r ’■'

iu'

'■ ■'I®'"' ' ^iSii ■'■^t

' -« V ' V/.'j.i^>y<t*‘ifid*'/ J V ? 1 ; j ?, , /;

V

ar, -.mu \;,., i-,c:.:^

♦S i '•‘‘ -K

. "■■ 1i", - <V,.). '/- vJ. ■ i .

n

• K

■v'|. ■

'■! r /*• I 'J ; ;

4ct

; > •' ■ *>".*1# i.( ■%>! " ■•.. ■- ^ ' ' , I'li -' ‘u..,^'ie^' -■ . * ,

c. , „':.

vf '®V'

?'\»;‘Vi '‘VrtTy

bi'Z? *y*i *,•■ .' .-I'.' ..r'qafiAl^

‘ ^ V •■ • P - ■ , • . J . ..

??•. .1!^''!*-v/4' ^ ij. ' I '■“

;: '■ ' ‘1^ • jIs

' € **

C!

■r. :

r/i,

'• (,-v ■ '■ ;'

Greenstone, M, H., C. E. Morgan and A.-L. Hultsch. 1985. Spider ballooning: Development and
evaluation of field trapping methods (Araneae). J. Arachnol., 13:337-345.

SPIDER BALLOONING: DEVELOPMENT AND EVALUATION
OF FIELD TRAPPING METHODS (ARANEAE)

Matthew H. Greenstone
Clyde E. Morgan
Anne-Lise Hultsch

USD A, ARS Biological Control of Insects Research Laboratory
R O. Box 7629, Research Park
Columbia, Missouri 65201

ABSTRACT

Sets of two types of sticky traps, horizontal wires and vertical panel traps, the latter including clear
polyester, W hardware cloth, and '/i" hardware cloth substrates, were run concurrently for eighteen
weeks in a Missouri soybean field to see which gave the best taxonomic- and mass-frequency
representations of the aeronaut fauna. The wire traps significantly underrepresented the largest family
(Linyphiidae) and mass class (< 0.6 mg) in the fauna. Of the three panel trap substrates, there were
no differences between the two hardware cloth meshes but the polyester trap catches declined
significantly during cold periods in late fall. An estimate is given of the number of hardware cloth
traps needed for an effective sampling program.

INTRODUCTION

Spiders are among the most abundant and consistently present arthropod
predators in crop fields and may contribute significantly to biological control of
pests (Riechert and Lockley 1984). As with other natural enemies there may be
a time lag between their population buildup and that of their prey, due in part
to the need to recolonize fields following harvest or diapause. Spiders may
recolonize either by walking or by ballooning [passive aerial dispersal, or
aeronautic behavior (Richter 1967, Greenstone 1982)].

The composition of the aeronaut fauna has been assessed by mechanical
suction traps (Taylor 1974), by kite-borne nets (Farrow and Dowse 1984), and
by sticky traps of various designs. These have been mounted on airplanes (Click
1939) or on ground level supports (Duffey 1956, Yeargan 1975, Van Wingerden
and Vugts 1976). The purpose of the present investigation was to compare the
efficacy and convenience of two simple and inexpensive sticky trap types,
horizontal wires (Van Wingerden and Vugts (1976) and vertical panels (Yeargan
1975), and to see whether there were detectable differences among various
substrates for the panel traps. We also wished to know whether height above the
ground or compass direction affected the catches.

338

THE JOURNAL OF ARACHNOLOGY

Fig. 1. — A, Wire trap setup in the field; B, Panel trap setup in the field. Note random arrangement
of Mylar, V2' hardware cloth and '4" hardware cloth at each of three heights (Photograph was taken
in mid-summer); C, Panel trap setup photographed in late fall. Compare near opacity of Mylar traps
with their translucence in mid-summer (Fig. IB).

GREENSTONE SPIDER BALLOONING TRAPS

339

MATERIALS AND METHODS

The studies were performed in a 2.0 ha field at the University of Missouri
South Farms, 9.7 km SE of Columbia in Boone County, Missouri. The field was
planted in 18.3 m x 22,9 m plots of soybeans with admixtures of sunflowers
ranging from 60 to 240 plants per plot. The placement of sunflower treatments
among plots was determined by a randomized complete block design, as part of
a study on the influence of cropping scheme on natural enemy diversity (N. L.
Marston, personal communication). The treatments were set out in a three
column by four row array, separated by 18.3 m wide alleyways, which produced
six intersections in which the traps could be set up. Two of these were selected
by use of a table of random digits (Rohlf and IB 1969) and the wire traps set
in one and the panels in the others. The two set-ups were 50-m apart.

Each set of traps was supported by a structure comprising four vertical posts
set into the ground and braced with 2 x 4's at the top, forming a cube 2.5 m
on a side with its vertical faces perpendicular to the cardinal compass directions.
For the wire traps, six 2 mm diameter horizontal wires were nailed to the posts
at heights of 35, 75, 115, 155, 195 and 235 cm (Fig. lA). The central 2.0 m of
each stretch of wire was coated with adhesive (Tack Trap,® Animal Repellents,
Inc., Griffin, GA) to form the trapping surface. For the panel traps, pulleys were
placed on the inside corners of the four posts at heights of 20 and 84, 95 and
160, and 170 and 235 cm, and a plastic coated three-wire clothesline run through
each set of four pulleys and secured with a turnbuckle. Wooden frames holding
the traps were attached to adjacent pairs of clotheslines using standard one-inch
binder clips. The frames were 38 cm x 64 cm rectangles made up by gluing and
nailing together lap-jointed 1 x 2's. They were dipped twice in varnish before use.
Stapled to each frame was one of the following three panel substrates: 0.076-mm
(.003 in) clear polyester (Mylar®) (Air Plastics, Inc., Mt. Vernon, NY), 12.7-mm
(Vi") galvanized hardware cloth, or 6.35-mm (14 ") galvanized hardware cloth. In
order to facilitate coating the panels without fouling the wooden frames, an
aluminum template having a 25-cm x 50 cm cutout in the center was placed over
the panel, and prewarmed adhesive was painted on the unmasked area with a
brush. The frames were hung from the clotheslines in three rows of three traps
on each face (Fig. IB), one of each substrate type, with position (left, right, or
middle) determined by consulting the table of random digits. For transport to
and from the field, the panel traps were stacked in groups of 18 on a wooden
pallet with 1.8-cm diameter vertical pipes cemented into the corners to prevent
shifting.

The two sets of traps were run concurrently for eighteen consecutive 1-wk
periods beginning on June 15, 1983, and the panel traps for an additional five.
(The wire traps were discontinued after it became clear that the panel traps were
easier to handle and sampled the most important family more effectively [see
below]). After each collection the panel traps were replaced with a fresh set and
returned to the laboratory for microscopic examination at 6X with a Wild®M-
5 stereo microscope. The wire traps were examined in the field with the aid of
a 2!4 X magnifying glass. Mylar traps were renewed by disposing of and
replacing the Mylar. Hardware cloth traps were renewed by removing and
soaking the hardware cloth panels overnight in Stoddard’s solvent, cleaning them

340

THE JOURNAL OF ARACHNOLOGY

with a wire brush, and drying, restapling and recoating them with prewarmed
adhesive. The wire traps were cleaned in the field with paint thinner and paper
towels and recoated each week.

To ensure that animals caught by the traps were ballooning and not crawling
or “rappelling” (i.e., travelling via bridge lines, J. Carico, personal
communication) onto the traps, all vegetation was cleared from within 3.0-m of
all trap faces (see Figs. lA and IB) by three applications of Roundup®
(Monsanto Chemical Co., St. Louis, MO), and 10 cm barrier bands of adhesive
were placed encircling the tops and bottoms of all posts and the ends of all wire
(trapping and clothesline) segments. These were checked for stickiness and
renewed as needed.

The spiders collected were placed for three days each in paint thinner and
toluene before final preservation in 70% ethanol. They were identified to family
and the individual masses estimated using volume-mass regressions from
previously live-massed and measured preserved animals (Greenstone et al., 1985).
Because of the tedium of these measurements, a subset of seven samples, chosen
to span the season and include a range of catches from very low to very high,
was selected.

The numbers of spiders caught were subjected to multi-way analysis of variance
of the factors date, height, compass direction, and, for the panel traps, substrate,
using the SAS general linear models procedure at the University of Missouri,
Columbia, Computing Center. The date x height x direction mean square was
used to provide an error term for the wire trap ANOVA’s, and the sums of
squares for the four- and all three-way interactions were pooled to produce an
error mean square for the panel trap ANOVA’s.

RESULTS

Results of the ANOVA’s are given in Table 1. All factors except compass
direction show significant main effects, but all are also involved in significant
interactions. The results are most easily understood if the two sets of comparisons
are taken separately.

Comparison of Hardware Cloth and Mylar Panel Trap Substrates. — There
were no obvious trends in any of the interactions except for that of date x type.

Table. 1 — Results of F Tests.

PANEL TRAPS

WIRE TRAPS

Source

F

d.f

P

F

ds

P

Date

73.6635

22,628

***

150.1029

17,255

Height

6.0902

2,628

5.6602

5,255

*

Date X Height

1.2563

44,628

1.3965

85,255

*

Type

20.1498

2,628

*

—

—

Date X Type

3.8025

44,628

—

—

Height X Type

3.9545

4,628

—

—

Direction

6.8794

3,628

1.8818

3,255

Date X Direction

1.6339

66,628

♦ sjc

1.4478

51,255

Height X Direction

0.3480

6,628

2.7032

15,255

♦

Type X Direction

1.4564

6,628

—

—

***p<.001; **p<.01; *p<.05

GREENSTONE SPIDER BALLOONING TRAPS

341

800

Fig. 2. — Relationship between total number of
spiders caught on the hardware cloth traps and
wire traps on the eighteen concurrent dates.

2

0

0

100

200

300

400

500

600

Panel Traps

All six significant F’s for panel trap type occurred toward the end of the season
(the last week of September, the last two weeks of October, and the first and
third week of November). We noticed a tendency for the Mylar traps to be more
visible from a distance when we picked up the traps during this period (compare
Figs. IB and 1C). Close inspection showed that the adhesive, which is usually
translucent, had become a nearly opaque white. Directions on the Tack Trap can
indicate that it will not work consistently at temperatures below 35° F. There were
eight weeks in which the temperature at the site fell to 35° F or below at least
once. Five of these were included among the six weeks for which the F-test on
trap type was significant, while the other three fell among the seventeen non-
significant F-tests. This difference is highly significant by the log-likelihood ratio
test (Sokal and Rohlf 1969, G = 8.47, p. < 0.01).

In order to determine where the differences among the panel substrates lay,
Student-Newman-Keuls tests (Sokal and Rohlf 1969) were run on the data of the
significant dates. In all six cases the Vi' and 14" hardware cloth means were not
significantly different, while the Mylar traps were always significantly (p < 0.05)
less than the Vi hardware cloth and in all but one case significantly less than
the V4" hardware cloth as well. The mean numbers on the hardware cloth panels
on these dates ranged between 1.8 and 2.4 times those on the Mylar panels.

Comparison of Wire Traps and Panel Traps. — The relationship between the
total numbers of spiders trapped on the wire traps and the Vi" plus hardware
cloth panel traps for the eighteen dates on which they were operated concurrently
is shown in Fig. 2. These data were fitted to simple linear and polynomial
regressions forced through zero. Both regressions were significant (p < 0.0001)
but the polynomial gives a significantly better fit to the data (F2, 12 = 5.27, p.
< 0.025). This is the curve which has been fitted to the data in Fig. 1. It describes
93.1% of the variance in wire trap catches.

The wire and panel traps were compared further in their representations of the
mass-frequency and taxonomic-frequency distributions of trapped spiders. Five

342

THE JOURNAL OF ARACHNOLOGY

Fig. 3. — Proportions of the totals (heights of
bars) and sample sizes (above the bars) for all
spiders (hatched) and major families which were
caught on the wire traps. See text for further
explanation.

families each contributed at least five percent of all trapped individuals from the
seven representative identified samples. Fig. 3 shows the proportions (heights of
bars) of all spiders and of each family caught on the wire traps and the total
numbers (above the bars) of all spiders and of each family caught on all traps.
Of the 2165 animals in the sample, 1236, or 0.57 of the total, were caught on
the wire traps. This is the expected proportion of animals on the wire traps for
each family. The departures from expected are highly significant (G = 94.398, p
< 0.0001). It can be seen by inspection that the family Linyphiidae, which makes
up almost half of the total catch, is underrepresented on the wire traps.

Fig. 4 shows the proportions and totals for the mass classes of the same sample
(the total sample size is slightly higher in Fig. 4 than in Fig. 3 because not all
animals which could be measured could also be placed taxonomically). Again the
departures from expected are highly significant (G = 86.296, p < 0.001). The
largest mass class, animals 0.6 mg or less which make up slightly more than half
the total, is underrepresented on the wire traps.

Fig. 4. — Proportions of the totals (heights of
bars) and sample sizes (above the bars) for all
spiders (stippled) and principal mass classes
which were caught on the wire traps. See text for
further explanation.

GREENSTONE ET/IL.— SPIDER BALLOONING TRAPS

343

DISCUSSION

The curvilinear relationship depicted in Fig. 2 is partially explained by the data
in Figs. 3 and 4. The smallest spiders tend to be linyphiids, so that Figs. 3 and
4 both reflect the tendency for the wire traps to undersample this family. There
is a persistent, moderate to large number of linyphiids ballooning throughout the
season. The other families are rare or absent early in the season, increase in
numbers through early fall and then wane (Greenstone et al., unpublished data).
This makes the wire traps appear to become more efficient as total numbers
increase and the proportion of linyphiids, which they undersample, decreases
(Fig. 2).

Our experience prior to data analysis led us to favor the panel traps since they
can be brought into the lab and examined microscopically under uniform lighting
conditions. Although all three authors seemed to get comparable counts on
adjacent wires on the wire traps on any given day and compass direction, we did
not feel confident about seeing the smallest spiders, particularly under changing
lighting conditions in the field. The data in Fig. 4 bear this out. Whether the
difference is due to actual differences in trapping efficiency or the difference in
magnification and lighting used in scanning them is immaterial, because the wire
traps must be checked in the field and therefore do not lend themselves to
microscopic examination. At our site, where the linyphiids make up such a large
proportion of the aeronaut fauna, wire traps can be expected to give a distorted
picture of the taxonomic- and mass-frequency distributions of ballooners. In fact,
the use of wire traps is probably not wise throughout mid to high latitudes in
the Northern hemisphere, where linyphiids tend to be the largest family of
ballooners.

Of the panel traps. Mylar is clearly at a disadvantage in cold weather.
Furthermore, high winds destroyed two Mylar traps, so they are also less reliable.
Because there were no significant differences between the Vi' and 14" hardware
cloth, we recommend the because, with less surface area they are easier to
scan and require less adhesive.

We can think of two possible explanations for the halving of the catch on the
Mylar traps in six of the late season samples. First, higher winds in the fall may
tend to blow spiders around these solid traps more so than around the
perforated, hardware cloth traps, a difference which may disappear at lower wind
speeds (R. Suter, personal communication). Unfortunately we lack an a priori
criterion for determining what windspeed is “high.” The lowest mean windspeed
recorded during the six significant weeks was 8.4 kph. If we take this as the
threshold for a wind effect, then six of six significant weeks exceeded this
threshold and 13 of the 17 non-significant samples also exceeded it. This
difference is not significant (G = 2.70, p > 0.1). If we take as our threshold for
a wind effect 12.2 kph, which was the next highest mean wind speed among the
six significant dates, then five of six significant dates and five of twelve non-
significant dates meet or exceed it. This difference is significant (G = 5.49, p <
0.02). We can look at the windspeed hypothesis in another way. If the drop in
catches on the Mylar traps is indeed due to this wind effect, then the smallest
animals should be most strongly affected (R. Suter, pers. commun.). The spiders
caught on one of the significant dates, October 18, have been measured. Fig. 5

344

THE JOURNAL OF ARACHNOLOGY

22

Fig. 5.— Proportions of the totals (heights of
bars) and sample sizes (above the bars) for all
spiders (stippled) and principal mass classes
which were caught on the Mylar traps on
October 18, 1983. See text for further

explanation.

shows the proportions of the totals and sample sizes for all spiders (stippled) and
principal mass classes which were caught on the Mylar traps. The departures
from expected are highly significant (G = 11.89, p < 0.01) with the smallest
spiders underrepresented and the largest overrepresented on the Mylar traps.
These results are consistent with the wind-speed hypothesis.

One alternative explanation is that the spiders are better able to see the Mylar
traps due to the large area of nearly opaque adhesive visible on cold days, and
actively avoid them, e.g., by changing the length of the ballooning threads to rise
above or drop below the traps. If correct, this would indicate that spiders may
be at least partially able to guide their flight and thereby effect some degree of
habitat selection before alighting (cf. Meijer 1977). This working hypothesis could
be tested by setting up Mylar traps which are either clear or opaque and
comparing the catches.

Our extensive trapping data can be used to estimate the number of traps
required for a long-term ballooning study, using the formula presented by Sokal
and Rohlf (1969, p. 247). In order to use this formula, the investigator must
specify the size of difference between samples he or she wishes to detect, the
probability that such a difference will be found if it exists, and the level of
significance. There must also be an estimate of the variability of the data. The
modal coefficient of variation for the hardware cloth trap data for this study was
44.9%. Taking 50% as a conservative expected coefficient of variation, a = 0.05
for rejection of the null hypothesis, P = 0.8 as our probability of finding a certain
difference between two means, and d = 0.3 as that difference, we get n ~ 16 traps.
Given the absence of main effects of height and compass direction (Table 1), a
four-sided trap with two traps at each of two heights on each side would support
this number conveniently.

GREENSTONE SPIDER BALLOONING TRAPS

345

ACKNOWLEDGMENTS

We thank Well Ju, Gary Krause and Marti Rhodes for assistance with
statistical analysis, the Department of Atmospheric Sciences, University of
Missouri, Columbia, for providing meteorological data from the trapping site,
and Norm Marston, Susan Riechert, Robert Suter and Ken Yeargan for
reviewing the manuscript. A.-L. H. was supported by a U.S.D.A. Research
Apprenticeship. We are especially grateful to the Department of Entomology,
University of Missouri, Columbia, for use of the field site.

LITERATURE

Duffey, E. 1956. Aerial dispersal in a known spider population. J. Anim. Ecol., 25:85-1 11.

Farrow, R. A. and J. E. Dowse. 1984. Method of using kites to carry two nets in the upper air for
sampling migrating insects and its application to radar entomology. Bull. Entomol. Res., 74:87-
96.

Glick, P. A. 1939. The distribution of insects, spiders, and mites in the air. U.S.D.A. Tech. Bull.,
673:1-150.

Greenstone, M. H. 1982. Ballooning frequency and habitat predictability in two wolf spider species
(Lycosidae: Pardosa). Florida Entomol., 65:83-89.

Greenstone, M. H., C. E. Morgan and A.-L. Hultsch. 1985. Ballooning Methodology: equations for
estimating masses of sticky-trapped spiders. J. Arachnol., 13:225-230.

Meijer, J. 1977. The immigration of spiders (Araneida) into a new polder. Ecol. Entomol., 2:81-90.
Richter, C. J. J. 1967. Aeronautic behavior in the genus Pardosa (Araneae: Lycosidae). Entomol.
Monthly Mag., 103:73-74.

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

Rohlf, F. J. and R. R. Sokal. 1969. Statistical tables. Freeman, San Francisco. 253 pp.

Sokal, R. R. and F.J. Rohlf. 1969. Biometry. Freeman, San Francisco. 776 pp.

Taylor, L. R. 1974. Insect migration, flight periodicity, and the boundary layer. J. Anim. Ecol.,
43:225-238.

Van Wingerden, W. K. R. E. and H. F. Vugts. 1976. Meteorological aspects of aeronautic behavior
in spiders. Oikos, 27:433-444.

Yeargan, K. V. 1975. Factors influencing the aerial dispersal of spiders (Arachnida: Araneida). J.
Kansas Entomol. Soc., 48:403-408.

Manuscript received January 1985, revised April 1985.

:s;

-,.V

4*

1

I ^ \.-

“ ■* -. ^ ■■ ■ • rpV,

';,<.JV/ . ■'■(.> .>M

•*:?»?■»•.■■■<:.>,. I’i-VT” ;..i. v.i-

-‘V / * :-4 ?> ♦ I . ' •- .vltl- .) ■■^‘.■•^'il',.- %.';

' ' .^£' ' -1

y|- , L

' » ^f.4‘

r; ■. -i

i\i ..>■*'

. ,4S

^ ;;rf; V ;■»?>. 7>

- 4 •

. ■/

,^i*/ : f^i-v' -

•* 4?’' t j * 1

if*«-

s' ^^'■'

.it ■'-’-^'•■■iw*^ i*n# ‘T.

■* - K ^r ^ <*► .*.-■•4^'- ^ . •■ . '■'TjfiSIfil

..• • .)*' 1» . J- iTJ . #A jl».-^ '.: •■» ,* ■i^ ■' .!r‘(.Ti ,u*'. '.1»4M

/ , ■^•. . 'US ■” •■• Ti^

' '■' ■ Ar-.i

2

4k^.

'. . . r . 71.’ ; ■..'^i:

• ;. -.4.' -, ; ;M ii_ ; •!) ■ '■ '1 •? . ( ■ j_ ' ^3 .ji

i.’ ^ -'j ^.’ -'l-'-T' i'l'. 'V fJ :'■>•?>*. t ;L .ylfil ^

ft i,

•’>'1! ■-*.; ' '•’5 .-' -■ '.' J>V. {

‘Bt^-'Vxi -4 i. '.> -.r. •'".••

ii ’ r<v.

‘ \ ... .- ■ ‘ •.•, Vi,' f.. . ‘

i ■ ?iikU4

.!,,„ -yi,.."’' :‘ J

'^i

< ■■ ■■' '' , ., :5,^ r--- ,; V' '■- . V, -i L''%>i|i|i . .,

.•;^^.'’v ■*''* '4 r.'^ •■ • ■“•* .r , Vt-5?v

i 'f iG.

. . '’■■ Av '■ ’i • ^ *

•■'. '''''virf . ■■■'

• ■^' •■ r., 1/4. ,. ’'is'-V '‘*'- y,'-'. i^r V'

,. ..... .A E:.f ,

-^s

' ‘-Vila -S.-. -V'

r. •i'i'-’t .

. u.i \ . j ^.-4 ■- •! •

■..)i ■ li X ■ ^ |W

- , • , • • r ‘ ■ -'- " •

, . •»

I.; ■4'..<

i _ 4 I*

■^‘...* W^'ltl". '(> . .J u.',*'

<if\. >

■■■Vi ' i -

1“ * ■ . V'

•' ■ ■

» ''■>‘''’tm^sfS^ -■ .. - t , . 'V :

'--,v

1

.... -- ,.. .. ., ■, ■■

•ij^ I ^f<, fi . > ;, ' - i

VJ

V, . »'

■.-.'v ' "■ '>i;^ I»1

■i^ tyr Iti’

, ■•(#'’■'■'» V* 5^; .' '

• 'Jf^ - •*.

■sy,

fF «di . .-'j

“ ■ # - ,/>3' ,

>SM' ^r:,y'^:

Cokendolpher, J. C. and W. F. Rapp. 1985. Leiobunum lineatum: A synonym of Leiobunum cretatum
(Opiliones, Gagrellidae). J. Arachnol., 13:347-354.

LEIOBUNUM LINEATUM: A SYNONYM 0¥ LEIOBUNUM
CRETATUM GAGRELLIDAE)

James C. Cokendolpher

The Museum
Texas Tech University
Lubbock, Texas 79409

and

William F. Rapp

430 Ivy Avenue
Crete, Nebraska 68333

ABSTRACT

The harvestman Leiobunum lineatum Edgar is synonymized with Leiobunum cretatum Crosby and
Bishop. Leiobunum cretatum is rediagnosed, briefly compared to congeners, and numerous new
records from central and eastern U.S.A. are provided. The labra and seminal receptacles are
illustrated for the first time and the penis is redrawn.

INTRODUCTION

The original description of Leiobunum lineatum by Edgar (1962) mentions that
this species may be confused with Leiobunum aurugineum Crosby and Bishop.
Because L. aurugineum differs remarkably (leg and body relative lengths, body
shape and microsculpturing, and genitalia) from L. lineatum no further
comparisons need be made. Judging from the specific name, Edgar (1962)
considered the silver lines on the abdomen to be diagnostic for the species. This
character ~^^s used in his key (Edgar 1966) to separate L. lineatum from all
congeners of the Great Lakes region. In his revision of Leiobunum Koch, Davis
(1934) reports that Leiobunum cretatum Crosby and Bishop has the “dorsum
golden-yellow, central marking obsolete.” Presumably, Edgar (1966) was
following Davis’ description when he too reported no dorsal abdominal pattern
on L. cretatum. This is indeed unfortunate, as in the original description of L.
cretatum Crosby and Bishop (1924) described the species as having narrow
longitudinal light lines on the abdomen. Furthermore, they used that character
to separate their new species in a key to males of Leiobunum. Our examinations
of the type materials and numerous other specimens of L. lineatum and L.
cretatum reveal they are conspecific.

348

THE JOURNAL OF ARACHNOLOGY

MATERIALS

The American Museum of Natural History, New York, is abbreviated
“AMNH” in the text and specimens referred to as “JCC” and “WFR” are in the
personal collections of the authors.

SYSTEMATICS

Leiobunum cretatum Crosby and Bishop
Figs. 1-12

Leiobunum cretatum Crosby and Bishop 1924:14, 19, 20, fig. 15; Davis 1934: 664, 673, figs. 14, 36;

Goodnight and Goodnight 1942:12; Edgar 1966: 354, 361, 365.

Leiobunum lineatum Edgar 1960:377 (nom. nud.), 1962:146-149, figs. 1-3, 1966:6, 61, 62; Levi and
Levi 1968:237, fig. (13-11); Cokendolpher 1982:89. NEW SYNONYMY.

Figs. 1-4. — Leiobunum cretatum: 1, female holotype from Georgia; 2, female from Tennessee; 3,
male from Michigan; 4, juvenile paratype from Georgia.

COKENDOLPHER AND RAPP— LEIOBUNUM CRETATUM

349

Type data. — Leiobunum cretatum holotype female (reported “probably male”
by Crosby and Bishop 1924) from Oglethorpe, Macon Co., Georgia (1 July 1910,
J. C. Bradley), Cornell Collection in AMNH, examined, and two immature
paratypes from Unadilla, Dolly Co., Georgia (28 June 1910, J. C. Bradley), one
Cornell Collection in AMNH, examined — other juvenile not located. Holotype
male and “allotype” female of L. lineatum from Locus Key TIN R5W SI, Eaton
Co., Michigan (coll. 24 July 1959, reared to maturity /fixed 17 Aug. 1959, A. L.
Edgar), in AMNH, examined.

Diagnosis. — Small species (body lengths: 2. 3-3. 5 mm for males, 2.6-5.25 mm
for females) with femora I longer than body length. Dorsum with silvery-white
longitudinal lines (Figs. 1-4). Ocular tubercle with several pointed spines over
each eye. Male with alate penis (Fig. 5) and with only few weak teeth on palpal
tarsi. Female seminal receptacles as in Figs. 6, 7. The labra of males are pointed
and simple, similar to those of females (Figs. 8-10). All legs uniform in color,
lacking white bands.

Comparisons. — The use of the shape of the labra is a relatively new taxonomic
character in the study of Leiobunum (Suzuki 1976). The first North American
species to be illustrated were by Tsurusaki (1985). Simple pointed labra, similar
to those of L. cretatum, occur in European Leiobunum spp. and in U.S.A.
Leiobunum townsendi Weed, Leiobunum vittatum (Say), and Leiobunum
relictum Davis (Tsurusaki 1985: pers, obs.). The labra of the U.S.A. Leiobunum
aldrichi Weed and Leiobunum gordoni Goodnight and Goodnight are unlike
those of L. cretatum (see Tsurusaki 1985: figs. 31 and 3M).

Based on morphologies of penes (Davis 1934; Cokendolpher 1982; pers. obs.),
L. townsendi, L. vittatum, and L. relictum can be eliminated as near kin to L.
cretatum. The genitalia of L. cretatum appear similar to those of several
European Leiobunum spp. (Martens 1978) and those of L. aldrichi and L.
gordoni (Davis 1934; Cokendolpher pers. obs.).

Tsurusaki (1985) noted that most European Leiobunum spp. lack denticles on
the palpal tarsi and that males of U.S.A. species have such denticles. Males of
L. cretatum have only a few small denticles on the proximal end of the tarsi
ventrally. Unlike European Leiobunum spp.; specimens of L. cretatum, L.
aldrichi, and L. gordoni have several denticles over the eyes.

Variation. — Throughout the range of L. cretatum (Fig. 11), little variation in
genital morphology, granulation of dorsa, spination of ocular tubercles, or
coloration was noted. Obvious differences in leg lengths were noticed when
comparing specimens from Michigan and Georgia. Femora H lengths for all adult
specimens examined were plotted against latitude and longitude. The plots of
lengths vs. longitude for both males and females showed no observed correlation,
but the plots of lengths vs. latitude revealed variation in femora H lengths.
Specimens from more northern localities have shorter legs than those from more
southern localities (Fig. 12). North/ south dines in leg lengths have been
previously recorded in numerous Northern Hemisphere gagrellid opilionids (Weed
1892, 1893; Suzuki 1971, 1973; McGhee 1977) and the reverse dine has been
reported in two gagrellid species from the Southern Hemisphere (Ringuelet 1960).

Unlike most studies on dines in harvestmen, we can not demonstrate a uniform
decrease in body size with increase in latitude. Leg length dines were reported
for the eastern U.S.A. Leiobunum politum Weed and Leiobunum bracchiolum

350

THE JOURNAL OF ARACHNOLOGY

Figs. 5-10. — Leiobunum cretatum: 5, dorsal aspect of penis; 6, seminal receptacle of holotype from
Georgia; 7, seminal receptacle of Tennessee female; 8, lateral aspect of male labrum; 9, ventral aspect
of male labrum; 10, lateral aspect of female labrum.

COKENDOLPHER AND K\¥V—LEIOBUNUM CRETATUM

351

McGhee by McGhee (1977). Plots of body lengths vs. latitude of adult males and
females of L. cretatum revealed that: (1) males are smaller than females, (2) males
do not vary with latitude, and (3) females from northern latitudes are generally
larger than females from southern localities. These differences between females
though are possibly due to the dates of capture. All females studied from
southern localities were collected in summer and those from more northern
localities were collected during late summer and fall. Unless females in the
southern regions are gravid earlier than those from the north, we suggest that the
noted differences in body sizes are due to the extended abdomen of egg-laying
females.

Comments. — Leiobunum cretatum is one of the smaller members of the genus
from North America. Due to this small size, it is possible specimens have been
disposed of by collectors, thinking they were juveniles. This could account for the
scarcity of specimens in museum collections.

Distribution and natural history. — Leiobunum cretatum is widely distributed in
the eastern half of the U.S.A. (Fig. 11). It appears to be most abundant in

Fig. 11. — Distribution of Leiobunum cretatum in the eastern U.S.A., localities mapped by open
circles are from Davis (1934) and Edgar (1962, 1971).

352

THE JOURNAL OF ARACHNOLOGY

deciduous forest and does not appear to exist in the eastern coastal woodlands.
It extends west into the Grassland Biome (Stipa-antilocapra biotic formation),
but is not found in the grasslands, rather in the subclimax woodlands along
streams. It is limited to floodplain woodlands where there is a heavy shrub and
herb layer. From our observations it appears that there are no true grassland
species of Leiobunum. However, in habitats which have high moisture content at
least Leiobunum vittatum and L. cretatum can exist.

Specimens examined. — U.S.A.: MICHIGAN; Eaton Co., 4 miles S. Charlotte, 4 Sept. 1979 (J. C.
& J. E. Cokendolpher), 3 males, 1 female (1 male WFR, others JCC); Washtenaw Co., 20 Sept. 1930
(J. S. Rogers), 1 male, 1 female (AMNH): MISSOURI; Marion Co., Hannibal, 29 Sept. 1979 (W.
F. Rapp), 1 female (JCC): NEBRASKA; Brown Co., Long Pine, 10 Sept. 1975 (W. F Rapp), 1 female

Femora II Lengths

Fig. 12. — Leiobunum cretatum femora II lengths (in mm): male (solid circles), female (open circles).

COKENDOLPHER AND RAYV—LEIOBUNUM CRETATUM

353

(JCC), 11 Sept. 1979 (W. F. Rapp), 1 male (WFR); Holt Co., O’Neill, 9 Sept. 1975 (W. F. Rapp),
2 males, 2 females (1 pair JCC, 1 pair WFR), Chambers, 10 Oct. 1979 (W. F. Rapp), 1 female (WFR),
Swan Lake, 5 Sept. 1978 (W. F. Rapp), 1 female (JCC); Merrick Co., Central City, 31 Aug. 1976
(W. F. Rapp), 1 female (WFR): TENNESSEE; Williamson Co., 17 miles SW of Nashville, Mary
Mount Campground, 20 Sept. 1981 (W. D. Sissom), 1 female (JCC): GEORGIA; Lowndes Co.,
Valdosta, 1 Sept. 1940 (C. J. Goodnight), 4 males, 3 females (AMNH); Turner Co., Ashburn, 31 Aug.
1940 (C. J. Goodnight), 3 males, 5 females (AMNH): FLORIDA; Gadsden Co., Ochlochonee River
and Marianna Road, 24 July 1930 (N. W. Davis), 3 males, 2 females, 3 juveniles (AMNH); Jackson
Co., Marianna, 24-29 July 1930 (N. W. Davis), 8 males, 7 females, 2 juveniles (AMNH); Jefferson
Co., Monticello (date & collector unknown), 1 male (AMNH).

ACKNOWLEDGMENTS

Dr. Norman 1. Platnick (American Museum of Natural History, New York)
loaned many of the specimens examined during this study. Dr. Robert Krai and
Mr. W. David Sissom (Vanderbilt University, Nashville) kindly provided
information on the single collection site in Tennessee. The manuscript was
reviewed by Dr. Nobuo Tsurusaki (Hokkaido University, Sapporo) and Mr.
Stephen W. Taber (Texas Tech University) calculated the regression equation for
Fig. 12. We thank all of the above for their help. This study was supported in
part by the Department of Biological Sciences, Texas Tech University.

LITERATURE CITED

Cokendolpher, J. C. 1982. Comments on some Leiobunum species of the U.S.A. (Opiliones:
Palpatores, Leiobunidae). J. Arachnol., 10:89-90.

Crosby, C. R. and S. C. Bishop. 1924. Notes on the Opiliones of the southeastern United States with
descriptions of new species. J. Elisha Mitchell Sci. Soc., 40:8-26 + pi. 1-3.

Davis, N. W. 1934. A revision of the genus Leiobunum (Opiliones) of the United States. Amer.
Midland Nat., 15:662-705.

Edgar, A. L. I960. The biology of the order Phalangida in Michigan (Ph.D. Thesis, Univ. Michigan.

251 pp. Univ. Microfilms, Ann Arbor, Michigan). Dissertations Abstr., 21(2):377.

Edgar, A. L. 1962. A new phalangid (Opiliones) from Michigan, Leiobunum lineatum sp. nov. Trans.
Amer. Microscop. Soc., 81:146-149.

Edgar, A. L. 1966. Phalangida of the Great Lakes region. Amer. Midland Nat., 75:347-366.

Edgar, A. L. 1971. Studies on the biology and ecology of Michigan Phalangida (Opiliones). Misc.
Publ., Mus. Zool., Univ. Michigan, No. 144, 64 pp.

Goodnight, C. J. and M. L. Goodnight. 1942. New and little known Phalangida from Mexico. Amer.
Mus. Novitates, No. 1163, 16 pp.

Levi, H. W. and L. R. Levi. 1968. Invertebrate Zoology. Arthropod relatives, Chelicerata, Myriapoda.
Vol. H. Trans, and adapted from A. Kaestner: Lehrbuch der Speziellen Zoologie, Band I, John
Wiley and Sons, New York, 472 pp.

Martens, J. 1978. Spinnentiere, Arachnida. Weberknechte, Opiliones. Die Tierwelt Deutschlands, Teil
64, Gustav Fischer Verlag, Jena, 449 pp.

McGhee, C. R. 1977. Observations on the use of measurements in the systematic study of Leiobunum
(Arachnida: Phalangida). J. Arachnol., 5:169-178.

Ringuelet, R. A. 1960. Clines en Opiliones. Un estudio analitico y biometrico de dines ecologicos
en dos especies de la fauna Argentina. Acta Zool. Lilloana, 17(1959):225-247.

Suzuki, S. 1972. Geographical variation in Melanopa grandis Roewer of East Asia (Arach.,
Opiliones). Proc. 5th Intern. Congr. Arachnol., Brno, 1971:65-70.

Suzuki, S. 1973. [Clines in Opiliones]. Japanese Soc. Syst. Zool., Circular, No. 46, pp. 6-10. (in
Japanese).

Suzuki, S. 1976. The genus Leiobunum C. L. Koch of Japan and adjacent countries (Leiobunidae,
Opiliones, Arachnida). J. Sci. Hiroshima Univ., Ser. B., Div. 1, 26:187-260.

354

THE JOURNAL OF ARACHNOLOGY

Tsurusaki, N. 1985. Taxonomic revision of the Leiobunum curvipalpe-gvon^ (Arachnida, Opiliones,
Phalangiidae). I. hikocola-, hiasai-, kohyai-, and platypenis-snhgron^s. J. Fac. Sci. Hokkaido
Univ., Ser. VI, Zool., 24:1-42.

Weed, C. M. 1892. The striped harvest-spider: A study in variation. Amer. Nat., 26:999-1008 + pi.
27.

Weed, C. M. 1893. The cinnamon harvest-spider and its variations. Amer. Nat., 27:534-541 + pi. 13.

Manuscript received March 1985, revised May 1985.

Stockwell, S. A. 1985. A new species of Heteronebo from Jamaica (Scorpiones, Diplocentridae). J.
Arachnol., 13:355-361.

A NEW SPECIES OF HETERONEBO FROM JAMAICA
(SCORPIONES, DIPLOCENTRIDAE)

Scott A. Stockwell

Department of Biological Sciences
Texas Tech University
Lubbock, Texas 79409

ABSTRACT

Heteronebo franckei, new species, is described from Jamaica and new Jamaican locality records are
reported for Heteronebo jamaicae portlandensis Francke and Heteronebo jamaicae occidentalis
Francke.

INTRODUCTION

The diplocentrid scorpion fauna of Jamaica was described by Francke (1978)
and reviewed by Armas (1982). Two species (one with three subspecies) are
known to occur on Jamaica: Heteronebo elegans Francke, Heteronebo jamaicae
jamaicae Francke, H. jamaicae occidentalis Francke, and H. jamaicae
portlandensis Francke. In the present paper, a newly discovered species of
Heteronebo from Portland and St. Thomas Parishes is described and new locality
records are reported for H. jamaicae portlandensis and H. jamaicae occidentalis.

METHODS

The description format and terminology follow that of Francke (1978) and
Francke and Sissom (1980) except for the measurements of the chelicerae. In the
present paper, the cheliceral chela length is measured along the dorsal surface of
the chela from the posterior margin to the distal tip of the fixed finger. One
hemispermatophore of the holotype was removed and examined in 100%
lactophenol as described by Cokendolpher (1984) for opilionid genitalia. A Wild
M7A dissecting microscope equipped with an ocular micrometer and a drawing
tube was used to make all measurements (in millimeters) and drawings of the
holotype. Acronyms for collections from which specimens were examined are
given in the acknowledgments.

356

THE JOURNAL OF ARACHNOLOGY

Heteronebo franckeL new species
Figs. 1-8

Type data. — Holotype male from near the mouth of Christmas River, Portland
Parish, Jamaica, West Indies, 15 December 1969 (P. A. Drummond), deposited
in FSCA. Seven paratypes are listed under specimens examined.

Etymology. — Named in honor of Dr. Oscar F. Francke in recognition of his
contributions to scorpion biology and systematics.

Distribution. — Known from near the mouths of Christmas and Priestman’s
Rivers, Portland Parish, and the Horse Savanna and Crookshank Rivers, St.
Thomas Parish, Jamaica, W. I. (Fig. 9).

Diagnosis. — Heteronebo franckei, new species, is placed in the bermudezi
species group (Francke 1978) with Heteronebo bermudezi (Moreno), Heteronebo
caymanensis Francke, and H. jamaicae. This group is characterized by dense,
small and minute granulation on the lateral and ventral carinae of metasomal
segment V. Heteronebo franckei is distinguished from other species of Heteronebo
by the following combination of characteristics. Pedipalp chela length in adult
males less than twice chela width, in females and immature males greater than
twice chela width; pedipalp chelae weakly carinate; metasomal segment II wider
than long; metasomal segment III in adult males slightly longer than wide, in
females and immature males as wide or wider than long; setation on dorsal lateral
carinae of metasomal segments I-IV, 1:1: 1:1; metasomal segment V ventral
median and ventral lateral carinae moderately strong with small to medium sized,
low tubercles; pectinal tooth count in males 8, in females 7; prolateral pedal spurs
large, well developed on all legs; tarsomere II spine formula 4-5/4:5/5:6/7:7/7.

Description. — With the characters of the genus (Francke 1978). Measurements
of the holotype and a female paratype are given in Table 1. The following
description is based on the adult male; parenthetical statements refer to females
and immature males.

Prosoma: Carapace brown in color, median ocular tubercle and lateral ocular
tubercles dark brown; carapace densely granular, weakly emarginate anteriorly;
median anterior notch shallow (moderately deep), rounded (v-shaped). Venter
brownish-yellow, moderately setate, densely punctate.

Mesosoma: Tergites brown in color, densely granular interspersed with small
rounded tubercles. Tergite VII strongly bilobed posterolaterally; submedian
carinae weak on posterior one-fourth, with single row of low, close tubercles;
lateral carinae moderately strong on posterior half, with irregular row of low
tubercles. Genital operculi yellow, subelliptical; pectinal tooth count in males 8,
in females 7; sternites yellow, lustrous, densely punctate. Sternite VII submedian
carinae weak on posterior one-half, crenate; lateral carinae moderately strong to
weak on posterior half, crenate.

Metasoma: Reddish-brown in color, punctate, minutely coarsely granular;
segments I and II wider than long, segment III slightly longer than wide (as wide
as or wider than long). Ventral submedian carinae on segments I-III weak,
crenate (with irregular rows of medium sized tubercles); on segment IV very weak
with small, scattered tubercles. Ventral lateral carinae on segments I-IV weak,
crenate (with irregular rows of medium sized tubercles) on I-III, with small
scattered tubercles on IV. Lateral inframedian carinae weak on segment I, very
weak on segments II and III, vestigal on segment IV; with a few small, scattered

STOCKWELL— NEW SPECIES OF HETERONEBO

357

Table 1. — Measurements (in millimeters) of Heteronebo franckei.

Character

Holotype

male

Paratype

female

Total length

31.20

37.10

Carapace length

4.95

5.85

Anterior width/ posterior width

3.00/5.25

3.90/6.60

Width at median eyes

4.30

5.20

Mesosoma length

10.20

13.75

Metasoma length

16.05

17.50

I length/ width

2.40/3.20

2.60/3.70

II length/ width

2.70/2.90

2.95/3.30

III length/ width

2.90/2.80

3.20/3.20

IV length/ width

3.50/2.65

3.85/3.05

V lenght/ width

4.55/2.50

4.90/2.90

Telson length

4.35

5.10

Vesicle length/ width/depth

3.50/2.20/1.70

4.15/2.95/2.30

Aculeus length

0.90

0.90

Pedipalp length

15.65

19.15

Femur length/ width/ depth

3.50/1.65/1.80

4.05/1.95/2.20

Tibia length/ width/ depth

3.60/1.80/2.05

4.30/2.10/2.50

Chela length/ width/depth

8.55/4.65/2.80

10.80/5.30/3.40

Movable finger length

5.40

7.00

Fixed finger length

3.60

4.80

Chelicera chela length/ width

2.40/1.15

2.80/1.35

Movable finger length

1.50

1.80

Fixed finger length

0.90

1.10

tubercles. Lateral supramedian carinae on segments I-IV weak with small
irregularly spaced tubercles. Dorsal lateral carinae weak on segments I-IV,
irregularly granulose. Setation of dorsal lateral carinae on segments I-IV, 1:1; 1:1:.
Segment V ventral median carina moderately strong with wide row of small to
medium sized tubercles and strong subterminal bifurcation with large tubercles
(Fig. 5). Ventral lateral carinae moderately strong, irregularly tuberculate
anteriorly; with a single row of large tubercles distally. Lateral median carinae
very weak on anterior three-fourths with small scattered tubercles. Dorsal lateral
carinae weak to vestigal, irregularly granulose. Anal arc rounded; subterminal
keel weak to moderately strong with a row of close, medium sized tubercles;
terminal keel obsolete, smooth. Telson slightly elongate (globular), moderately
granulose, sparsely setate, densely punctate.

^edipalps: Reddish-brown in color, orthobothriotaxia C, densely punctate.
Femur trapezoidal in cross section, deeper than wide; dorsal internal keel
obsolete; dorsal external keel weak with medium sized tubercles; ventral external
keel obsolete; ventral internal keel moderately strong, coarsely granular; dorsal
and internal faces densely and coarsely granular; ventral face minutely granular.
Tibia much shorter than chela width (Fig. 4); dorsal internal keel obsolete; basal
tubercle weak, represented by a few small tubercles; dorsal median keel
moderately strong, weakly tuberculate; dorsal external keel moderately strong,
weakly granular; external keel weak with obscure, scattered granules; ventral
internal keel moderately strong, smooth to vestigally granulose; ventral external
keel moderately strong, smooth; internal faces shagreened, dorsal external face
moderately granular; other external and ventral faces very minutely shagreened.

358

THE JOURNAL OF ARACHNOLOGY

Chela robust (Figs. 1-3) (inflated), moderately granulose, reticulate, densely
punctate; digital keel weak (very weak), vestigally granulose (smooth); dorsal
secondary keel vestigal, granular; external secondary keel obsolete; ventral
external keel weak distally (vestigal distally), vestigally granular (smooth); ventral
median keel weak, crenate (smooth); ventral internal keel weak (vestigal), smooth.
Dentate margins of fingers densely tuberculate.

Legs: Yellow in color. All segments except tarsomeres shagreened. Prolateral
pedal spurs large, well developed on all legs. Variation in tarsomere II spine
counts shown in Table 2.

Hemispermatophore: See Figs. 6-8. Lamina length, 3.9 mm.

Habitat. — All specimens of H. franckei were collected from near the mouths
of rivers in organic litter amid limestone pebbles.

Figs. 1-5. — Heteronebo franckei, new species, male holotype: 1, right pedipalp chela, external aspect
showing trichobothrial pattern; 2, right pedipalp chela, ventral aspect; 3, right pedipalp chela, internal
aspect; 4, right pedipalp tibia, external aspect; 5, metasomal segment V, ventral aspect.

STOCKWELL— NEW SPECIES OF HETERONEBO

359

Table 2. — Variability in tarsomere II spine counts for Heteronebo franckei.

Leg I

Leg II

Leg III

Leg IV

No. of spines in row

4

5

5

6

3

6

7

6

7

8

Males

Prolateral row

4

4

7

1

-

8

-

-

8

-

Retrolateral row

6

2

7

1

-

3

5

-

7

I

Eemales

Prolateral row

2

6

7

1

1

6

1

1

7

-

Retrolateral row

6

2

8

-

-

2

6

1

5

2

Specimens examined. — JAMAICA: Portland Parish, mouth of Priestman’s River, 15 December
1969 (P. A. Drummond), 1 adult female (FSCA), 1 immature female, 2 immature males (FSCA); St.
Thomas Parish, along Crookshank River near sea, 16 March 1969 (P. A. Drummond), 1 subadult
male (FSCA), just above mouth of Horse Savanna River, 16 December 1969 (P. A. Drummond), 1
immature female (FSCA), near N bank of mouth of Horse Savanna River, 15 June 1970 (P. A.
Drummond), 1 adult female (OFF).

DISTRIBUTION OF HETERONEBO ON JAMAICA

This genus is represented on Jamaica by three species (one with three
subspecies) (Fig. 9). Heteronebo franckei is known from the coastal regions of

Figs. 6-8. — Heteronebo franckei, new species, male holotype, right hemispermatophore: 6, external
(lateral) aspect; 7, detail of capsular region, dorsolateral aspect; 8, dorsal aspect.

360

THE JOURNAL OF ARACHNOLOGY

Fig. 9. — Map showing distribution of Heteronebo species on Jamaica, West Indies. Triangle = H,
elegans Francke; star = H. jamaicae jamaicae Francke; solid circle = H. jamaicae occidentalis Francke;
square = H. jamaicae portlandensis Francke; circled star = H. franckei, new species.

Portland and St. Thomas Parishes. Heteronebo elegans is known only from St.
Thomas Parish and Heteronebo jamaicae portlandensis is known only form
Portland Parish. Heteronebo jamaicae jamaicae is known from St. Andrews and
St. Thomas Parishes and H. jamaicae occidentalis is known from the largely
mountainous regions of Manchester, St. Ann, St. Catherine, St. James, St.
Thomas, and Trelawny Parishes.

New locality records. — Heteronebo jamaicae portlandensis was previously known by a single
specimen from the John Crow Mountains, Portland Parish. It is now known from the following
locality. JAMAICA: Portland Parish, mouth of Priestman’s River, 15 December 1969 (P. A.
Drummond), 1 immature female (FSCA).

Heteronebo jamaicae occidentalis, formerly known only from St, James and Trelawny Parishes, is
recorded from the following localities. JAMAICA: Manchester Parish, 2.4 mi. S. along road to Harry
Watch of Craig Head, 27 March 1969 (P. A. Drummond), 1 female (FSCA); St. Ann Parish, 2.0 mi.
along Hollymount Road from junct. with Al, 28 March 1969 (P. A. Drummond), 2 females (FSCA),
3.9 mi. along Hollymount Road from junct. with Al, 28 March 1969 (P. A. Drummond), 1 female
(FSCA), 4.2 mi. along Hollymount road from junct. with Al, 28 March 1969 (P. A. Drummond),
1 male, 1 female (OFF), Rose Hill above Runaway Bay, 5 December 1969 (P. A. Drummond), 1 male,
1 female (FSCA); St. Catherine Parish, 0.2 mi. W of Sligoville, 18 March 1969 (P, A. Drummond),
1 female (FSCA), 2.2 mi. N of Worthy Park Crossroads, 19 June 1970 (P. A. Drummond), 1 female
(FSCA); St. Thomas Parish, along path to Corn Puss Gap from S, 20 May 1969 (P. A. Drummond),
I male (FSCA); Trelawny Parish, Tyre (2 mi. N of Troy), 16 May 1969 (P, A. Drummond). 1 male,
1 female (FSCA).

ACKNOWLEDGMENTS

I thank Mr. James C. Cokendolpher for his helpful comments and critical
review of this manuscript, and Dr. Oscar F. Francke (OFF) for obtaining a loan
of the specimens reported on herein. I am also grateful to Dr. G. B. Edwards
of the Florida State Collection of Arthropods (FSCA) for providing the loan.

STOCKWELL— NEW SPECIES OF HETERONEBO

361

LITERATURE CITED

Armas, L. F. de. 1982. Algunos aspectos zoogeograficos de la escorpiofauna antiliana. Poeyana, No.
238, 17 pp.

Cokendolpher, J. C. 1984. A new genus of North American harvestmen (Arachnida: Opiliones:
Palpatores). Pp. 27-43, In Festschrift for Walter W. Dalquest in honor of his sixty-sixth birthday
(N. V. Horner, ed.). Midwestern State Univ., Wichita Falls, xx + 168 pp.

Francke, O. F 1978. Systematic revision of diplocentrid scorpions (Diplocentridae) from circum-
Caribbean lands. Spec. Publ., Mus., Texas Tech Univ., No. 14, 92 pp.

Francke, O. F. and W. D. Sissom. 1980. Scorpions from the Virgin Islands (Arachnida, Scorpiones).
Occas. Pap., Mus., Texas Tech Univ., No. 65, 19 pp.

Manuscript received February 1985, revised May 1985.

i

r /:

ti

' 'V'- . ■ i

■v- - iOe* T'-'

V

» •’

Smith, D. R. R. 1985. Habitat use by colonies of Philoponella republicana (Araneae, Uloboridae).
J. Arachnol., 13:363-373.

HABITAT USE BY COLONIES OF
PHILOPONELLA REPUBLICANA (ARANEAE, ULOBORIDAE)

Deborah R. R. Smith

Department of Entomology
Cornell University
Ithaca, New York 14853*

ABSTRACT

Philoponella republicana (Araneae, Uloboridae) is a communal orb-weaving spider. Colonies of this
spider were found more frequently in interface forest than in high forest or mountain savannah forest.
This does not appear to be due to differences in insect abundance among forest types, but is
correlated with greater complexity of the understory in the interface forest. This may be due to the
need for supports for colony attachment lines. Within the interface forest, the location of colonies
is correlated with local insect abundance. When flying insects are excluded from colonies, individual
spiders can respond by increasing the distance between orbs in the colony, and colonies can respond
by abandoning the site and moving to a new location.

INTRODUCTION

Philoponella' republicana (Simon) is a communal orb-weaving uloborid spider,
found in Panama, Trinidad, and northern South America (Opell 1979). It occurs
in the rainforest understory, frequently in small tree-fall gaps and other openings
in the forest. It is a seasonal species, with as many as three discrete generations
per year in Panama (Lubin 1980).

The colonies consist of attachment lines, individual prey capture orbs, and a
central retreat area (Figure 1). The retreat is an irregular tangle of non-sticky
threads; individuals leave their orbs and move to the retreat in the evenings and
when disturbed. Females with egg-cases and adult males may also spend much
of their time in the retreat (see also illustration in Simon 1891). Prey capture
generally takes place in the orbs. The orbs are placed above and around the
retreat, sometimes several layers deep (rarely directly below the retreat); orbs are
occupied by one individual at a time. The body of the colony is suspended a
short distance above the ground by the attachment lines. These are large
conspicuous bundles of non-sticky threads running from the colony to objects in
the environment used as supports (e.g., shrubs, herbs).

‘Current address: Museum of Zoology, Insect Division, University of Michigan, Ann Arbor, Michigan
48109.

364

THE JOURNAL OF ARACHNOLOGY

The communal societies of P. republicana are simple compared with those of
cooperative spider species such as Agelena consociata (Agelenidae; Kraft 1970),
Anelosimus eximius (Theridiidae; Brach 1975, Christenson 1984, Vollrath 1982),
or Stegodyphus sarasinorum (Eresidae, Jambunathan 1905). There is no maternal
care of the young other than guarding the egg-case, and no cooperation in orb
construction. Nor do females cooperate in prey capture: although several females
may be attracted to a large struggling insect and help to wrap it, a short
aggressive interaction ensues and one female claims the prey packet. There may,
however, be more integration of colony members than this description implies,
since colony mates share the support lines and the retreat, and there is some
evidence (presented below) that the colony may respond as a group to
unfavorable conditions.

Fig. 1. — Sketch of a Philoponella republicana
colony; a = support lines; b = individual prey
capture orbs; c = central retreat area; d = objects
used as supports (herbs, lianas, palms).

Study of the facultatively communal species Philoponella oweni in Arizona,
U.S.A. (Smith 1982, 1983) showed that this species forms communal groups in
response to several environmental factors. Philoponella oweni builds its long-
lasting webs in protected sites, such as hollow trees or clefts among rocks. These
sites may be scarce in some habitats, and the same sites are often used year after
year by succeeding generations. Females are solitary if such sites are abundant,
or if food is scarce. Communal groups form in areas where suitable sites for web
construction are in short supply and insects are locally abundant, allowing several
females to share a protected site and still obtain enough prey.

Lubin (1980) reports that new colonies of P. republicana are often founded by
groups of immatures dispersing en masse. It is possible that P republicana, with
its larger and more complex groups, has evolved from an ancestor in which
groups of immatures responded to patchily distributed resources in a way similar

SMITH— HABITAT USE BY COLONIES OF PHILOPONELLA

365

to that of R oweni. For instance, if food were abundant groups of siblings might
remain together, whereas if food were scarce they would disperse. Later they
might evolve the habit of remaining in groups even when local food supplies were
low, moving as a group to a better location.

Here I examine the location of R republicana colonies with respect to those
environmental factors already known to be important to R oweni — insect
abundance and substrates for web attachment — and with respect to forest type.
I also present natural history information on colony size and development.

METHODS

Forest type. — I carried out observations of R republicana in the Voltzberg-
Raleighvallen reserve, Saramacca Province, Suriname (04° 32’ N, 56° 32’ W)
during February- April 1980 and February 1982. The Voltzberg reserve is located
in primary lowland rainforest. The vegetation of Suriname is relatively well
known and several forest types have been described from the Voltzberg region.
The names used here for forest types, and the brief descriptions below follow
Schultz (1960).

High forest is characterized by having two or three stories, the lower stories
appearing very open. The main canopy is ca. 30 m tall, with emergent trees
reaching about 40 m. Palms, particularly ''boegroe makka'' {Astrocaryon
sciophilum), are abundant in the understory and form a fairly continuous layer
at ca. 8 m. The understory is sparse.

Mountain savannah forest is a semi-deciduous forest which occurs on shallow
stony soils, as on the edges of granite plates and bergs. It resembles true
savannah somewhat in appearance (hence the name) but differs floristically. Trees
are thin-stemmed and there is little stratification. There may be a dense herb
layer.

I also included a third type: interface forest. Interface forest occurs where two
or more forest types meet. This forest is characterized by a very dense understory
of palms, lianas, shrubs, woody plants and herbs.

The forest in the Voltzberg reserve was essentially undisturbed except for trails,
which passed through tracts of each of the three forest types mentioned here. I
located colonies by searching along trails; because the trails did not pass through
equal distances of each forest type, the amount of each forest type sampled was
not equal. The understory in mountain savannah forest was much less dense than
in either high or interface forest; although colonies a few meters off the trails in
the latter two forest types might not be visible, one could easily see objects which
were reasonable distances from the trail in mountain savannah forest.

Insect Abundance. — I measured insect abundance using sticky traps; my traps
were fresh-cut leaves of Heliconia sp. (Because all equipment and food for two
weeks at a time had to be backpacked into the study area, it was necessary to
rely on natural materials as much as possible. I selected Heliconia leaves because
they were large, abundant, and relatively uniform in size, and provided a smooth
tough surface to spread the trap substance on.) I traced a 15 X 30 or 10 X 20
cm rectangle on the underside of the leaf, and coated an area larger than the
rectangle with Stick’em Special. Insects which crawled onto the leaves would

366

THE JOURNAL OF ARACHNOLOGY

presumably be caught before they reached the rectangle; insects captured inside
the rectangle were assumed to be flying insects. By coating the underside of the
leaf I ensured that the trapping surface would not be obscured if the leaf began
to wilt. The leaf traps were suspended from trees and saplings.

I measured insect abundance at colony sites and at non-colony sites using a
paired sampling scheme. I placed a Heliconia leaf trap next to each of seven
colonies at the same height as the colony’s prey capture surface, and a second
trap at an arbitrarily chosen site 5 m due north, at the same height. Two trap
sizes were used in different trials — 10 X 20 and 15 X 30 cm. The traps in any
paired comparison were the same size. In most cases the traps were examined
after 24 hrs, but in some cases pairs were examined after 48 or 72 hours. I
analyzed these data with the Wilcoxon signed rank test for paired comparisons
(Seigel 1956) to allow for the variation in size and time among pairs. When traps
were examined I recorded the number of insects captured, their size (length to
the nearest mm; insects less than 1 mm were placed in one of two size classes:
those less than 0.25 mm, and those greater than 0.25 mm and less than 0.5 mm),
and taxonomic order.

I also compared insect abundance in the three forest types. I placed five
Heliconia leaf traps with a 10 X 20 cm capture area in each forest type. Points
for trap placement were randomly selected by laying a 50 m forester’s tape along
a trail passing through the appropriate forest type, and selecting two numbers
from a random number table. The first number dictated how many meters I
moved along the meter tape, the second how many meters I moved into the forest
perpendicular to the tape, alternating left and right of the tape. I collected data
on the number of insects captured as above, every 24 hours for five days.

Not all insects captured in sticky traps are potential prey for P. republicana.
Philoponella republicana typically takes insects 5 mm or less in total body length,
and usually does not take Orthoptera or Hemiptera (personal observation). I
called the subset of insects captured that were 5 mm or less in length, exclusive
of Orthoptera and Hemiptera the “small insects.” The potential prey truly
available to P republicana probably consists of some of the “small insects” and
also some insects not captured by the sticky traps at all. During data analysis
I used four measures of insect abundance — total number of insects captured per
trap per day, total number of ’’small insects” captured per trap per day, sum of
the lengths of all insects captured per trap per day, and the sum of lengths of
“small insects” per trap per day. Although watching actual prey capture is the
best way to assess what a particular spider species is taking (Castillo L. and
Eberhard 1983), this method cannot be used to compare insect abundances in
habitats where spiders occur and where they do not. The sticky trap data can
be useful to compare abundance of certain classes of insects among different
locations, but cannot be used to calculate total insect prey available.

Forest understory. — I measured the structure of the forest understory at the
sites of five P republicana colonies and in each of the three forest types to find
the relative numbers of potential supports for colony attachment lines. I
randomly selected 10 points in each forest type using a meter type and random
number table as described above. At 1 m north, south, east and west of each
randomly chosen point I suspended a 160 cm plumb line and recorded the
number of plant stems and leaves that intersected the line. At colony sites I took

SMITH— HABITAT USE BY COLONIES OF PHILOPONELLA

367

measures 1 m north, south, east, and west of the center of the colonies. The total
number of plant parts intersecting the four plumb lines for each point were
summed for each point, since the four were not independent measurements. If a
plumb line fell on a point occupied by a tree or boulder, that point was
discarded.

Food Deprivation. — Orb- weaving spiders can respond in a number of ways to
decreasing food supplies, for example by spinning larger orbs or by relocating the
web. In communal groups spiders can also change the distance between orbs (a
change in the diameter of the orb can also cause a change in orb spacing). I
measured the response of colony members to food deprivation in terms of the
distance between an orb and its nearest neighbor. I first gathered six days of
baseline data on the nearest neighbor distances (NND) in three unmanipulated
colonies (No. 1, 4, and 5). Each day I measured the distance (to the nearest cm,
hub to hub) to the nearest neighboring orb for 10 to 22 orbs in each colony. If
the orbs were readily accessible I measured the distance with a ruler. If direct
measurement would have disturbed the spiders or the webbing I estimated the
distance. To test the accuracy of my estimates I first estimated the NND’s for
a set of readily accessible orbs, and then measured the distance. My estimates
were not significantly different from the direct measurements.

Next I built a large tent around colony 1. The tent consisted of a framework
of saplings and rope covered with cheese cloth. The tent was left in place for five
days, during which time it excluded most flying insects from the colony. After
five days I measured NND for orbs in the experimental and two control colonies.

I repeated the experiment using colonies 3, 4, and 5. I gathered baseline data
for one day and then built a cage around colony 3. After three days of insect
exclusion I measured NND in the experimental and control colonies.

To test for the effect on NND of general disturbance during tent building I
gathered baseline data for one day on colony 8 and then built a tent framework
around it, consisting of poles and ropes without the cheese cloth. I recorded
NND in this colony for three more days.

Colony growth and size. — I censused seven P. republicana colonies (No. 1 and
3-8) from 11 February to 10 April 1980. I classed the spiders in the colonies as
adult males, adult females (7 mm or more in total body length) or one of four
size classes of immatures: less than 1 mm, 1-2 mm, 3-4 mm, and 5-6 mm. When
counting numerous tiny hatchlings much less than 1 mm in length I took three
counts and used the average.

I measured the size of four colonies: height and horizontal diameter of the
main body of the colony (retreat plus orbs) and number and length of attachment
lines, all to the nearest 10 cm. I also noted the objects used as supports for the
attachment lines.

RESULTS

Forest type. — In 1980 I located seven (or five — see the section on Food
Deprivation below) large colonies of P. republicana. These colonies were in
interface forest (five colonies) or in gaps in forest created by boulder fields (two
colonies). The trails passed through large tracts of high forest and mountain

368

THE JOURNAL OF ARACHNOLOGY

Table 1. — Mean number of insects of all types captured per sticky trap per day in three Suriname
forest types. Trap sites are of three types: random sites were randomly selected points in each forest
type; arbitrary sites were 5 m due north of Philoponella republicana colonies in interface forest;
colony sites were next to colonies of P. republicana. Data from all trap sites were compared using
the Mann Whitney U-test. Means with the same group letter do not differ significantly at the 0.05
level.

Forest type:

Trap site

Mean

SD

N

Group

High

random

3.6

2.5

25

A

Mt. Savannah

random

5.2

3.0

25

B

Interface

random

5.8

3.8

25

B

arbitrary

5.3

3.5

28

B

colony

7.9

4.2

28

C

savannah forest and only a small belt of interface forest. Because most of the
colonies were found in interface forest, even though less of this forest type was
sampled, this implies that R republicana occurs more frequently in interface than
in high or mountain savannah forest.

Insect abundance. — Insects were more abundant at colony sites than at
arbitrarily selected sites 5 m north of colonies for all four measures of insect
abundance: total number of insects (p « 0.01), total number of “small insects”
(P «o .01), sum of lengths of all insects (p < 0.01), and sum of lengths of “small
insects” (p « 0.02) captured per trap per day (Wilcoxon signed rank test for
paired samples, Seigel 1956).

A comparison of insect abundance in the three forest types is given in Table
1; these data were analyzed using three-way Analysis of Variance and Duncan’s
multiple range test (Barr et al. 1976). There is no difference in the mean number
of insects captured per trap day in interface and mountain savannah forest, and
significantly fewer captured per trap day in high forest than in the other forest
types. It is also possible to compare insect abundance at these randomly selected
points in the three forest types with insect abundance at colony sites and the
arbitrarily chosen points 5 m north of colonies (Table 1). Data from 28 pairs of
traps of the same size and duration of exposure as the forest samples (10 X 20
cm, 24 hrs) showed that there were significantly more insects captured per trap/
day at colony sites (7.9 ± 4.2) than in any other site (Mann Whitney U-test, p
< 0.04 or better). There was a mean of 5.3 + 3.1 insects per trap/ day at sites
5 m north of colonies, which is not significantly different from the values for
randomly selected points in interface or mountain savannah forest (p > 0.75,
Mann Whitney, U-test).

Understory structure. — Table 2 shows the complexity of the understory in three
forest types and at sites occupied by colonies. There is no significant difference
in the mean number of plant parts intersecting plumb lines in interface forest and
at colony sites, and there is also no significant difference between high and
mountain savannah forest in this respect. There are significantly more plant
parts — potential web supports — in colony sites and in interface forest than in high
and mountain savannah forest.

Food deprivation. — In replicates 1 and 2, in which functional insect exclusion
tents were used, the NND increased when insects were excluded. In replicate 1

SMITH— HABITAT USE BY COLONIES OF PHILOPONELLA

369

Table 2. — Structure of the forest understory: mean number of plant parts intersecting four 160 cm
plumb lines in three forest types and at sites of Philoponella republicana colonies. Means with the
same group letter do not differ significantly at the 0.05 level. Interface differs from High and
Mountain Savannah forest at p<0.002; Colony sites differ from High and Mountain Savannah forest
at p<0.05 (Mann Whitney U-tests).

Forest type:

Mean

SD

N

Group

High

2.3

1.6

10

B

Mt. Savannah

1.6

1.8

10

B

Interface

9.9

2.8

9

A

Colony

6.8

2.9

4

A

the NND in the experimental colony increased from 8.2 ± 1.5 cm (n = 19 orbs)
on the day before insect exclusion to 11.1 ± 3.7 cm after exclusion (n = 17; p
< 0.02, two-tailed Mann Whitney U-test; Seigel 1956). There was no significant
change in the NND in the control groups. In control colony 4 the NND was 6.8
± 1.7 cm (n = 20) before the experiment and 6.8 ± 1.5 after (n = 16; p > 0.10);
in control colony 5 the NND was 6.1 ± 0.8 cm before (n = 20) and 6.4 + 1.7
cm after (n = 20; p > 0.10).

In replicate 2 the NND in the experimental colony increased from 7.6 ± 2.1
cm (n = 16) on the day before insect exclusion to 11.2 ± 3.8 cm after (n = 18;
p < 0.002, two-tailed Mann Whitney U-test). In control colony 4 the NND was
7.0 ± 1.7 cm before (n = 20), and 7.1 ± 1.0 after (n = 20; p > 0.10); control
colony 5 the NND did increase significantly after the course of the experiment
(p < 0.05) but the magnitude of the change was small (6.7 ± 1.6 cm before, n
= 20, to 7.6 ± 0.9 cm, n = 20 after).

In colony 8 the NND did not increase after the sham tent was built. Before
the tent was built NND was 12.1 ± 2.1 cm (n = 10 orbs). On the first day after
the sham cage was built NND decreased significantly to 9.0 ± 1.5 cm (n = 7;
p > 0.02). On the next two days NND increased back to values not significantly
different from the original values: 11.8 ± 1.7 (n = 7) and 11.0 ± 2.2 (n = 5; p
>0.10).

In addition, in replicates 1 and 2 the experimental colony abandoned its old
site after the insect excluding tent was removed; apparently the colony moved as
a group to a new location. Within a few days after removal of the tent the old
site was abandoned and two new colonies, 6 and 7, appeared two to three meters
away from the sites of the old colonies 1 and 3. The experimental colony 1
contained three marked individuals, one of whom was later seen in colony 6. In
addition a female was seen walking on the ground from near the old site of
colony 1 towards colony 6. This circumstantial evidence indicates that each
colony moved as a group.

Colony Growth and Size. — Figure 2 presents the development of Suriname
colonies over the period from 11 February to 10 April 1980. Hatchlings, i.e.,
spiderlings in their first post-emergence instar, are easily identified by their non-
sticky “sheet” orbs (Eberhard 1971, Szlep 1961). All the hatchlings were found
in the attachment lines of the colony, not in the body of the colony. Females
with egg-cases often leave the colonies (Lubin 1980) perhaps in an attempt to
avoid egg-case parasites. The hatchlings in the attachment lines may be the young

370

THE JOURNAL OF ARACHNOLOGY

Fig. 2. — Census data from six colonies of Philoponella republicana (colonies no. 1, 3, 4, 5,
6, and 7). Horizontal axis, time in days; vertical axis, number of individuals in units of 10; striped
horizontal bars, duration of insect exclusion experiments.

SMITH— HABITAT USE BY COLONIES OF PHILOPONELLA

371

of females who left the colony along the attachment lines. The seven colonies
were roughly synchronous in development, and most of the adults of a colony
die or disappear before the young produced by their generation mature. Thus
there is no overlap of adults between generations in these populations.

The four colonies measured to the nearest 10 cm were 85 ± 5 cm tall and 60
+ 0 cm in diameter. The retreats were 0 to 30 cm above the ground. There were
an average of 8 ± 2.2 attachment lines per colony, each an average of 78.6 ±
41 cm long (range 40 to 200 cm). The attachment lines were fastened to lianas,
leaves of hardwood trees, and most commonly, palms.

DISCUSSION

Colonies of R republicana in Suriname were found more frequently in interface
forest than high or mountain savannah forest. This does not appear to be a result
of differences in insect abundance among the three forest types, since insects were
no more abundant in interface than in mountain savannah or high forest. The
size of R republicana colonies, their height above the ground, and the length of
their attachment lines means that they are using objects for support from ground
level to 1.5 to 2.0 m above the ground. The understory in interface forest is
denser than in high or mountain savannah forest and thus provides more
potential supports for colony attachment lines.

Within interface forest the location of R republicana colonies does appear to
be influenced by insect abundance, inasmuch as colonies were found at sites
where insects were locally abundant. (Considering insects caught in sticky traps
as a measure of insect abundance.)

Individuals within colonies also respond to changes in insect abundance. If
food supplies are reduced by building a cage around the colony, the individual
spiders spin their orbs farther apart. When the cage is removed and the colony
is given the opportunity to move the group abandons the old, “poor” site and
relocates in a new site. Thus this species responds to environmental conditions
at three levels: at the level of the forest type occupied, at the level of colony
location within the chosen forest type, and at the level of individual spacing
within colonies.

The responses of R oweni and R republicana to food supply and web building
sites can be compared. Rhiloponella oweni, the facultatively communal species
found in the southwestern United States, builds its webs in protected sites which
may be in short supply in some habitats. Because the locations of these protected
web sites do not change much from year to year, the location of R oweni webs
are also rather stable from year to year. Some immatures of R oweni overwinter
at their mother’s web site and emerge the following spring to begin a new colony.
Some sites were occupied by R. oweni colonies for at least six consecutive years.
The number of adults which ultimately remain at a site appears to be largely
governed by insect abundance at the site (Smith 1983). These results agree with
those of Uetz et al. (1982), who found that the number of individuals in
communal groups of Metepeira spinipes (Araneidae) and interindividual spacing
within the colonies, varied in response to the abundance of prey.

372

THE JOURNAL OF ARACHNOLOGY

Compared to P. oweni, colonies of R republicana are more mobile. In habitat
used by P. republicana (interface and second growth forest) suitable attachments
for web building appear to be relatively abundant, and the results of the insect
exclusion study indicate that a colony may move as a group in response to
changes in food supply. Thus, whereas P. oweni remains at a web site and adjusts
group size to food supply, P republicana maintains its communal group and
moves the colony in response to food shortages. This can only be done in
habitats where potential web attachment sites are plentiful. Philoponella
republicana could be derived from a species in which immatures dispersed in
groups and responded to changes in food supply by moving the entire group to
a better site rather than by breaking up the group into individuals.

ACKNOWLEDGMENTS

The field portion of this study was supported by funds from the Explorers
Club, Sigma Xi, and the Grace Griswold Fund of the Cornell University
Department of Entomology. I wish to thank C. Craig, G. Eickwort, and R.
Hagen for reading and criticizing the manuscript; M. Robinson and an
anonymous reviewer commented on an earlier draft of this manuscript, and the
final version was improved by comments from B. Opell and G. Stratton. I also
benefitted from conversations with W. G. Eberhard, who has given me ideas for
better ways to measure insect abundance in future studies. STINASU, the
Suriname Nature Conservancy, and LBB, the Suriname Forestry Service, allowed
me to work in the excellent Suriname parks and forest reserves.

LITERATURE CITED

Barr, A. J., J. H. Goodnight, J. P. Sail and J. T. Helwig. 1976. A user’s guide to SAS 76. SAS
Institute, Inc., Raleigh, N. C.

Brach, V. 1975. The biology of the social spider Anelosimus eximius (Araneae: Theridiidae). Bull. So.
Cal. Acad. Sci., 74:37-41.

Castillo, L., J. A. and W. G. Eberhard. 1983. Use of artificial webs to determine prey available to
orb-weaving spiders. Ecology, 64:1655-1658.

Christenson, T. E. 1984. Behavior of colonial and solitary spiders of the Theridiid species Anelosimus
eximius. Anim. Behav., 32:725-734.

Eberhard, W. G. 1977. The webs of newly emerged Uloborus diversus and of a male Uloborus sp.
(Araneae: Uloboridae). J. Arachnol., 4:201-206.

Jambunathan, N. S. 1905. The habits and life history of a social spider (Stegodyphus sarasinorum
Karsch). Smithsonian Miscellaneous Collections No. 47, Vol. 2, no. 1573, pp. 365-372.

Kraft, B. 1970. Contribution a la biologic et a I’ethologie d'Agelena consociata (Araignee sociale du
Gabon). Biologia Gabonica, 6:199-369.

Lubin, Y. D. 1980. Population studies of two colonial orb-weaving spiders. Zool. J. Linn. Soc.,
70:265-287.

Opell, B. D. 1979. Revision of the genera and tropical American species of the spider family
Uloboridae. Bull. Mus. Comp. Zool., 148:443-547.

Schultz, J. P. 1960. Ecological studies on rainforest in northern Suriname In: de Hulster, 1. A. and
Lanjouw, J., eds.. The vegetation of Suriname — a series of papers on the plant communities of
Suriname and their origin, distribution, and relation to climate and habitat. Vol. 2. pp. 1-127.
Amsterdam.

SMITH— HABITAT USE BY COLONIES OF PHILOPONELLA

373

Seigel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill Book Co., N. Y.
Simon, E. 1891. Observations biologiques sur les arachnides. Ann. Soc. Entomol. France, 60:1-14,
4 plates.

Smith, D. R. R. 1982. Reproductive success of solitary and communal Philoponella oweni (Araneae:
Uloboridae). Behav. Ecol. Sociobiol., 11:149-154.

Smith, D. R. R. 1983. The ecological costs and benefits of communal behavior in a presocial spider.
Behav. Ecol. Sociobiol., 13:107-114.

Szlep, R. 1961. Developmental changes in the web spinning instinct of Uloboridae: construction of
the primary type web. Behavior, 17:60-70.

Uetz, G. W., T. C. Kane and G. E. Stratton. 1982. Variation in the social grouping tendency of a
communal web-building spider. Science, 217:547-549.

Manuscript received August 1984, revised December 1984.

3

I V ^

1. >*

It} ‘ fii-. ■ ‘v j

-■

Xm

•<• ai4TO

•l'- ‘*«k ‘♦•r" •' I ■ V ■ ■ » •^^'.. .. «'■ '’jfllj

LIT ^ ■■ -« .^; :■ ■■' ■'fr ^■■*:< ;■ .i

;*f i^ , ■ ■?! - ■•?“l. ■•

‘ ■-■-• ■■*•* M ... . ^ •' .^ .-^ir - r^l Jbw'.v ^

,,. .’.>y . J J

b“J‘i :i: 4p : 3* • > a*.* --, ■>>

r^i|i^„„ . . ' V-/''' Ti *' " ''ij \/' 1^ ’vj"' > *• ^ .

>iin •' .»■ *•-- . 1 *■ •• ■ ■ '■ * ’

ryi**. ^ X.'

‘’* ,, ■- -fev. ;x.( ■ • ■

' • * • .

^S» -iTi

r>i. -•■■•'

■‘^„« feSf

j*’ .i? i,':^-'—"

' \'SrtT r sa'4“

yEMi^

xVA- ]

K-'IUN

l.»»

•r. 1:-/' •l>*i . ufi; ■' . i?3<|

<.

y«

»^J ^4' ? ■'jr.‘r55J^.._ 'J,fi

_K

.^i- *.if f'. i, ^ . . ...v'^V;l(igJjJj|H

ii M * ■* ■♦ ■* * J*Tm ■

'4^ .

-■"-•ii

■'Ti

« ;

• ^ f\-'

■■ -i a,
‘i , k '..u
vf

J ^ A';,

M*

4„

iicr^iv

= '-. " ,V‘ k'l

•■5 ' ' '', ^ iui^_ ,

I KV'!

‘■^ %»“ -■ — '■ *•*' ’^it#

';u

• 4'' *•■:..» ■ . . ,i* y\ - tl^J-

• •■^ ix:'-’? 4 r't'A

' ’^ L*

■-••TJfc’'

3

n

k

•!.; r*Tr»-^i

I

9 4:'* " ..y?.\.a

” ^ .4 * ^ •

*• * -

Csi-

'1/ *. '

.V* V

/t

u. -yvT

4

f i

jwi ,, ,

Wi - ir-j^

HU .., . .' M'r* f4«UM-iBiiip

•'4, ’^i'^ ■ >,4t'4 jIk-

a,* •' ■““ - ' ' •

4- .

i:^-..fw. -^ . ■ I-

i ' I Aa=^4-4-'. . 'v*

ri»

’ t| •■-- «S»ijJ i« i >(•* -•V,

■■iAkLi'Vf.*' '■?..' i:- •■ ;#

Hr ■ '.;va».jL_Br. ’-i

*■■ . '.il »

■i*

■:e:

i-

Si.

J? 4»r

■ - ' ■* J'

.1 '^j'*-^>'‘-’S

i«* lj,l

■-1 r-i

■ ■'.* ^ -in

, ■ ■ a- •:' ,■* " .'r f'‘'._V:’

- 1 iki

- *

-» - '- ■ sT «t._ ^ ^ ^

Killebrew, D. W. and N. B. Ford. 1985. Reproductive tactics and female body size in the green lynx
spider, Peucetia v/nt/am (Araneae, Oxyopidae). J. Arachnol., 13:375-382.

REPRODUCTIVE TACTICS AND FEMALE BODY
SIZE IN THE GREEN LYNX SPIDER,
PEUCETIA VIRIDANS (ARANEAE, OXYOPIDAE)

Don W. Killebrew and Neil B. Ford

Department of Biology
The University of Texas at Tyler
Tyler, Texas 75701

ABSTRACT

Data collected on life history traits of Peucetia viridans, representing three generations, were
analyzed. A total of 61 females and their clutches were examined to determine the following: female
mass, female tibia length, clutch mass, clutch size, mean mass per young, egg sac mass, egg sac mass
per young, and relative clutch mass (an estimate of reproductive effort). Clutch mass, clutch size and
egg sac mass are strongly correlated with female mass. Peucetia viridans has the reproductive tactic
of producing young of optimum size, with the number of young varying in response to environmental
pressures that influence female size, thus producing optimal reproductive effort. Relative clutch mass
is also optimized and apparently not influenced by the variation in female mass, clutch mass or size.
Egg sac mass per young is similar in small clutches and large clutches. These findings are discussed
in terms of the species’ life history characteristics.

INTRODUCTION

Most information on reproductive tactics in spiders comes from literature
dealing with spider taxonomy, where some observations of spider size (usually
body length) and clutch size (number of eggs or young in an egg sac) have been
reported (Bristowe 1939, Kaston 1948, Gertsch 1949). For Peucetia viridans, the
green lynx spider, Brady (1964) gives the number of eggs from egg sacs taken
from several geographical regions. There are very few studies that deal with the
precise relationships between the size of female spiders, number and size of
offspring, and estimates of reproductive effort.

Any reproductive event can be described in terms of a triumvirate: reproductive
effort, expenditure of energy per progeny, and clutch size (Pianka, 1983). Though
the concept of reproductive effort (the proportion of total energy allocated to a
reproductive event) is very useful, it is often quite difficult to quantify. An
operational estimate of reproductive effort that has been used is a ratio of clutch
mass to female mass (Vitt and Congdon, 1978). This ratio, termed relative clutch
mass, is used in this paper.

Though many authors have noticed the general trend that larger females
produce larger clutches, few (Peterson 1950, Kessler 1973, Riechert and Tracy

376

THE JOURNAL OF ARACHNOLOGY

1975, Eberhard 1979) have studied this relationship closely in spiders. Enders
(1976) has shown, using published data from a variety of studies, that clutch size
among species of spiders is positively correlated with length of female and that
other life history traits (hunting manner, habitat selection) also appear to be
related to clutch size. The energy content of spider eggs was studied by Anderson
(1978). He included data on mass-specific energy content of P. viridans eggs from
five clutches along with data from eleven other species. He discussed these data
in relation to female mass and clutch mass and suggested that caution by used
when selecting an estimator of reproductive effort.

Our three-year study of certain reproductive parameters of Peucetia viridans
(Hentz) enables us to examine closely its reproductive tactics as they relate to the
energetics associated with the reproductive biology of the species. We analyze the
relationships between such parameters as clutch size, clutch mass, egg sac mass,
mean egg mass, relative clutch mass, and female mass. We also examine how
some parameters change significantly and others remain constant from one
generation to the next.

STUDY AREA AND NATURAL HISTORY

Two study areas in east Texas were selected. One study area was an old field
located 1.6 km northwest of the city of Whitehouse (Smith County) near the
western edge of the Australoriparian Biotic Province. This 1.22 hectare field is
surrounded by an Oak-Hickory-Pine Forest. In late summer and early fall the
tallest herbaceous vegetation consists of composites (Veronia sp., Lactuca sp..
Ambrosia sp.) and an euphorb {Croton capitatus), these being the most
commonly encountered plants on which there are female P. viridans and their egg
sacs. Samples were taken from this site in 1980 and 1981. The second study site
was an old field very similar to the first, located approximately 11.2 km north-
northeast of the first site, 0.8 km south of The University of Texas at Tyler.
Samples were taken from this site in 1983.

The life history and phenology of P viridans have been examined by
Whitcomb, Hite and Eason (1966) and Turner (1979). In east Texas, there
appears to be one reproductive period with a single clutch produced by a female
between mid-September and late October. Though multiple clutches have been
reported from other areas, none was observed during this study.

METHODS

Samples were taken from September 29 to October 12, 1980, and from
September 17 to October 25, 1981, and on October 11, 1983. Twenty-one females
with their corresponding egg sacs were taken in 1980, 14 were taken in 1981, and
26 were taken in 1983. The following information was determined for each
specimen: female mass (mg) after egg sac construction, length of first tibia of
female (mm), mass of egg sac (mg), mass of clutch (mg) and clutch size (number
of eggs or juveniles). Relative clutch mass (RCM, clutch mass/ female mass),
mean mass of young (MMY, clutch mass/ clutch size), and mean egg sac mass
per young (MEY, egg sac mass/ clutch size) were calculated.

KILLEBREW AND FORD— REPRODUCTIVE TACTICS OF PEUCETIA VIRIDANS

377

These data were statistically examined using the computer-packaged Statistical
Analysis System (Freund and Littell, 1981). The significance level chosen was a
- 0.05. In addition to basic descriptive statistics the following were used: Duncan
multiple range test to compare sample statistics by year; correlation; general
linear regression, with coefficient of determination and test for homogeneity of
slopes used to compare certain parameters to female size.

RESULTS

Tables 1 and 2 summarize sample statistics for each of the three years by
parameter and for parameter totals.

Female mass and female tibia length were significantly different by year; the
largest females were taken in 1981. Because the coefficients of variation (C.V.) did
not vary greatly from year to year, the female size differences by year appear to
be real.

Average clutch mass in 1980 was not significantly different form the other two
years, but clutch mass in 1981 was significantly different from that of 1983. Note
that in 1980 the C.V. is large compared to the other two years. Clutch mass
appears to vary more than female size. There is a significant positive correlation
of clutch mass and female mass by year and for the data combined. When clutch
mass is regressed against female mass (Figure 1), 55.14% of the total variation
in clutch mass is explained by the linear relationship to female mass in 1980,
58.12% in 1981, 63.38% in 1983, and 66.53% for the data combined. Regression
coefficients are not significantly different for these data.

Clutch sizes in years 1980 and 1983 were not significantly different from each
other, but clutch size in 1981 was significantly different from that of the other

Table 1. — Means (standard errors) / ranges of reproductive parameters of Peucetia viridans.
Statistics underlined are not significantly different by year (P>0.05).

Year

1980

1981

1983

Total

Sample

21

14

26

61

Female mass (mg)

106.12(4.81)

226.07(14.30)

155.51(7.38)

154.70(7.45)

Range

77.10-164.10

168.10-335.40

85.90-214.10

77.10-335.40

Female tibia length (mm)

6.89(0.16)

9.07(0.17)

7.63(0.15)

7.70(0.14)

Range

6.66-8.21

7.92-10.18

6.22-8.86

5.66-10.18

Clutch mass (mg)

147.94(14.46)

266.78(21.82)

193.65(13.75)

194.70(10.66)

Range

42.10-360.10

131.60-403.60

66.70-375.00

42.10-403.60

Clutch size

98.38(9.64)

169.86(12.40)

1 14.08(7.66)

121.48(6.42)

Range

29-242

90-240

42-183

29-242

Mean mass per young (mg)

1.5.(0.03)

1.56(0.03)

1.70(0.04)

1.60(0.02)

Range

1.29-1.72

1.44-1.76

1.31-2.25

1.29-2.25

Egg sac mass (mg)

5.71(0.43)

13.16(0.75)

7.32(0.46)

8.11(0.47)

Range

3.60-11.90

9.60-18.40

3.40-12.20

2.40-18.40

Relative clutch mass

1.31(0.11)

1.18(0.07)

1.23(0.06)

1.27(0.04)

Range

0.49-2.23

0.69-1.50

0.55-1.91

0.49-2.23

Egg sac mass per young (mg)

0.06(0.01)

0.08(0.01)

0.07(0.00)

0.07(0.00)

Range

0.03-0.12

0.05-0.13

0.04-0.09

0.03-0.13

378

THE JOURNAL OF ARACHNOLOGY

Table 2. —Coefficients of variation (coefficients of determination) of reproductive parameters of
Peucetia viridans. Statistics underlined are not significantly correlated to female mass (P>.05);
*= Regression coefficients for these parameters against female mass are not significantly different by
year when compared to regression line of total.

Year

1980

1981

1983

Total

Female mass

20.75

23.67

24.19

37.62

Female tibia length

10.75

6.96

9.94

14.09

Clutch mass

44.80(0.5514)

30.60(0.5812)

36.22(0.6338)

42.77(0.6653)*

Clutch size

44.91(0.5735)

27.32(0.5419)

34.26(0.6802)

41.25(0.6435)*

Mean mass per young

8.10(0.0001)

6.17(0.1868)

13.16(0.0022)

11.80(0.0129)*

Egg sac mass

34.33(0.1601)

21.31(0.0160)

32.43(0.6486)

45.44(0.5883)

Relative clutch mass

30.88(0.1071)

21.84(0.0003)

24.62(0.0249)

27.09(0.0403)*

Egg sac mass per young

36.77(0.1766)

29.68(0.3476)

20.03(0.0462)

30.35(0.0006)*

two years. Clutch size also appears to vary more than female size. Significant
positive correlation occurs between female mass and clutch size by year and for
the data combined. Coefficients of determination indicate that 57.35% of the
variation in clutch size is explained by a linear relationship to female mass
(Figure 2) in 1980, 54.19% in 1981, 67.02% in 1983 and 64.53% for the data
combined. Regression coefficients are significantly different for these data.

Mean mass of young was significantly different only in 1983. The C.V.’s are
very small when compared to the C.V.’s of other parameters. There is no
significant correlation of mean mass of young and female mass for any year nor
for the data totals; neither is there a difference in regression coefficients by year.

Egg sac mass was significantly different by year. It was not significantly
correlated to female mass in 1980 and 1981, though data from 1983 and the
combined data show a significant correlation between these two parameters.
Figure 3 shows the best-fit linear relationship between egg sac mass and female
mass, with very little of the variation in egg sac mass being explained by female
mass in 1980 and 1981, 16.01% and 1.6% respectively. The coefficients of
determination are 64.86% in 1983 and 58.85% for the data combined. Slopes by
year are significantly different from the slope of the composite regression line.

Relative clutch mass is not significantly different when compared by year.
There is no significant correlation with female mass, and regression coefficients
are not significantly different by year.

Egg sac mass per young is significantly different from other years only in 1981.
There is no significant correlation between egg sac mass per young and female
mass, nor are there significant differences in regression coefficients by year.

DISCUSSION

The results suggest that an analysis from two perspectives would be useful.
First we examine the life history parameters as they relate to individual
phenotypes, i.e., the expression of the parameters and their interactions. Next we
consider the broader implications of these individual reproductive tactics as they
relate to the species’ life history characteristics.

KILLEBREW AND FORD— REPRODUCTIVE TACTICS OF PEUCETIA VI RI DANS

379

Fig. 1. — Regression of clutch mass against female mass of Peucetia viridans for three years, is
the coefficient of determination.

It is clear that a reproductively successful female can vary greatly in size, the
largest being some 4.4 times more massive than the smallest (Table 1). This allows
for the exceptional range in clutch mass (42.10 - 403.60 mg) and clutch size (29 -
242) because these two parameters show high coefficients of determination when
regressed against female mass (Figures 1 and 2). Although natural selection
appears to favor individual variation in these three parameters, that variation
does not affect reproductive effort. These data show that small females may have
the same relative clutch mass as large females and that RCM does not change
from one year to another (Table 1 and Table 2), implying an optimum RCM.
This is striking, given that reproductive effort in some females is 4.5 times greater
than in others. Since essentially none of this variation is explained by a linear
relationship with female mass (or any other correlate of female mass) and since
the two components of RCM covary so strongly, which would normally yield a
constant RCM, we interpret the variation in RCM to be due to changes in female
mass after oviposition. This effect is random in that, even though some females
may have fed after oviposition, feeding frequency in the short time between
oviposition and being collected is due to chance. This feeding effect is likely a
minor component of variation in female mass, which is masked by the variation
in female mass already existing at the time of oviposition. When the mass to mass
ratio of RCM is calculated, the effect of variation due to post-oviposition feeding
is exposed. We conclude that P. viridans has a reproductive tactic of optimal

380

THE JOURNAL OF ARACHNOLOGY

reproductive effort regardless of the variation present in female mass, clutch size
and clutch mass.

Another parameter that is important in understanding reproductive tactics is
the energy allocation per spiderling. We utilized mean mass of young to estimate
this element. MMY is not correlated with any parameter discussed thus far.
Natural selection appears to produce an optimum MMY; in two of the three
years, it is the same. MMY also has the smallest C.V. of any parameter. This
leads to the conclusion that P. viridans, rather than having small females reduce
their allocation of energy per young so as to have more but smaller young, has
a reproductive tactic to produce fewer young of optimum size. Larger females
likewise produce young of optimum size, but more than them. This explains how
both clutch size and mass are strongly correlated with female mass.

Our findings are in agreement with those of Anderson (1978), who found that
variation in mass-specific energy content in spiders was less than variation in
clutch size, and there was no correlation between energy content per unit egg
mass and size of the female parent, egg size, or clutch size. In the light of these
findings and those of our study, RCM appears to be a valid estimate of
reproductive effort and may be a useful operational tool in studies that require
an indirect measure of the energy partitioned to reproduction by spiders.

In spiders, another quantitative reproductive parameter is important, namely
egg sac mass. This parameter varies in a manner similar to clutch size and mass.

Fig. 2. — Regression of clutch size against female mass of Peucetia viridans for three years, r^ is
the coefficient of determination.

KILLEBREW AND FORD— REPRODUCTIVE TACTICS OF PEUCETIA VI RI DANS

381

Fig. 3. — Regression of egg sac mass against female mass of Peucetia viridans for three years, r^
is the coefficient of determination.

The larger the female the greater the mass of the egg sac. However, when egg
sac mass per young is examined, it can be seen that this parameter has also been
optimized by natural selection (Tables 1 and 2). It is obvious that large clutches
require more silk for the same degree of protection, but the subtle point is that
small females tend to allocate the same amount of energy to egg sac construction
per offspring as do large females. Because the construction of an egg sac is a
surface to volume related phenomenon, allometric theory predicts the allocation
of less silk per young in larger clutches; since egg sac mass per young is constant,
egg sacs around larger clutches are likely thicker.

We examine now the broader implications of these life history parameters in
this species. Assume that a parameter is a constant from one year to the next
if in two of the three years it is not significantly different from the other years.
As seen in Table 1, clutch mass, clutch size, mean mass of young, relative clutch
mass and egg sac mass per young would be constants. Female size and egg sac
mass would therefore be the only variables by year. Under these conditions, it
appears that populations of R viridans tend to produce clutch sizes, clutch
masses, MMY’s, RCM’s, and MEY’s near optimal values. Female mass and the
mass of an egg sac are the parameters left as response variables, able to fluctuate
from year to year depending on environmental pressures. We propose the
following hypothesis as an explanation for these observations.

Peucetia viridans females protect their offspring in two ways: by watching over
the clutch and by covering the clutch with silk, the success of the former

382

THE JOURNAL OF ARACHNOLOGY

protective mechanism depends on the females’ endurance after oviposition; the
latter mechanism depends on her prey capturing ability before oviposition. For
a population to respond favorably to stochastic conditions in the environment
(prey and predator presence, temperature, moisture, etc.) some natural history
parameters may vary and others may remain constant, depending on the
particular selection pressures unique to a species. In P. viridans maintenance of
successful populations through time requires the tactic of adjusting female size
and all the attributes contingent on size. Trade-offs between other parameters
may be alternative reproductive tactics in other semelparous species, but in P.
viridans, female mass and egg sac mass vary in response to environmental
pressures to produce adequate numbers of optimal size offspring through optimal
reproductive effort, protecting their clutches with optimal egg-sac silk per young.

ACKNOWLEDGMENTS

L. Ross measured and counted the specimens taken in 1983. W. Green and M.
Smith prepared the figures. We thank J. Rovner and C. Dondale for their
suggestions and R Dunklin for typing the manuscript.

LITERATURE CITED

Anderson, J. F. 1978. Energy Content of spider eggs. Oecologia, 37:41-57.

Brady, A. R. 1964. The lynx spiders of North America, north of Mexico (Araneae: Oxyopidae). Bull,
Mus. Comp. Zool., 131:431-518.

Eberhard, W. B. 1979. Rate of egg production by tropical spiders in the field. Biotropica, 11:292-
300.

Enders, F. 1976. Clutch size related to hunting manner of spider species. Ann. Entomol. Soc. Amer.,
69:991-998.

Freund, R. J., and R. C. Littell. 1981. SAS for linear models. A guide to the ANOVA and GLM
procedures. SAS Institute. North Carolina.

Gertsch, W. H. 1949. American Spiders. Van Nostrand Co., Princeton.

Kaston, B. J. 1948. Spiders of Connecticut. State Geol. and Nat. Hist. Surv. Connecticut, Bull. No.
70.

Kessler, A. 1973. A comparative study of the production of eggs in eight Pardosa species in the field
(Araneida, Lycosidae). Tijdschr. Entomol., 116:23-41.

Peterson, B. 1950. The relationship between size of mothers and numbers of eggs and young in some
spiders and its significance for evolution of size. Experientia, 6:96-98.

Pianka, E. R. 1983. Evolutionary Ecology. Harper and Row, New York, 416 pp.

Riechert, S. E. and C. R. Tracy. 1975. Thermal balance and prey availability: bases for a model
relating web site characteristics to spider reproductive success. Ecology, 56:265-284.

Turner, M. 1979. Diet and feeding phenology of the green lynx spider, Peucetia viridans (Araneae:
Oxyopidae). J. Arachnol., 7:149-154.

Vitt, L. J. and J. D. Congdon. 1978. Body shape, reproductive effort, and relative clutch mass in
lizards: resolution of a paradox. Amer. Nat., 112:595-608.

Whitcomb, W. H., M. Hite and R. Eason. 1966. Life history of the green lynx spider, Peucetia
viridans (Araneida: Oxyopidae). J. Kansas Entomol. Soc., 39:259-267.

Manuscript received February 1985, revised June 1985.

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

NESTS AND NEST-SITE SELECTION OF THE CRAB SPIDER
MISUMENA VATIA (ARANEAE, THOMISIDAE) ON MILKWEED

Douglass H. Morse

Graduate Program in Ecology and Evolutionary Biology
Division of Biology and Medicine, Brown University
Providence, Rhode Island 02912

ABSTRACT

Adult female crab spiders Misumena vatia that had previously hunted in large milkweed clones
built nests on non-flowering milkweeds more often than predicted and used smaller than average
leaves for their nests. These results suggest that they actively select nest sites at both stem and leaf
levels. In commencing a nest, spiders first pulled the distal third of a leaf under the rest of the leaf,
secured it with silk, and then rested inside this space. A few days later they laid their eggs there,
sealed the edges of the leaf around the egg mass with silk and then guarded the nest, typically resting
on its lower side. Spiders always left the last stem upon which they hunted before nesting and usually
selected milkweed leaves on a stem several meters away from their hunting site for their nest site.

INTRODUCTION

The choice of a nest site is one of the most important decisions made by
animals that deposit eggs. Not only must these nests be placed where physical
conditions permit the development of eggs, and of hatchlings if they remain, but
the location must also be satisfactory for the adults, if they latter guard them.
Further, if the site is not guarded by parents, it is essential that if offer protection
from egg parasites or predators. A poor choice subjects the eggs to a wide range
of unfavorable factors. The stakes are particularly high if an individual lays many
or all of its eggs at a single site, for the probability of all-or-none success at this
stage of the life cycle is high. Not surprisingly, many species of animals exhibit
precise nest-site selection behavior (Hilden 1965, Partridge 1978, Morse 1980).

Although considerable general information exists on where spiders place their
egg sacs, much of it concerns web-building species (e.g., Comstock 1940, Bristowe
1958), and few quantitative data exist on egg placement for sit-and-wait spiders
such as crab spiders (see Enders 1977). Here I report on the characteristics of
nests and of nest sites selected by solitary, leaf-nesting crab spiders [Misumena
vatia (Clerck (Thomisidae)] that hunt in common milkweed {Asclepias syriaca)
immediately prior to nesting.

384

THE JOURNAL OF ARACHNOLOGY

METHODS

I gathered data on the nests of Misumena in a one-ha field in Bremen, Lincoln
Co., Maine, U.S.A., during July and August of 1980-1984. Although the
vegetation is composed primarily of grasses, a large clone of common milkweed
consisting of about 1500 flowering and non-flowering stems grows in the middle,
and three smaller clones occur elsewhere in the field. Almost all of the nests were
placed on milkweed. Several species of small forbs also grow throughout the
field.

I recorded the height of the nest, the height of the milkweed stem, and the
length and width of the leaf selected for the nest. If possible I recorded the
distances that spiders moved between their last hunting site and their nesting site.
I also measured stem height and leaf dimensions from random samples of
flowering and non-flowering stems in the study area. Data for most of the
variables came from 72 free-ranging spiders; however, I supplemented data on
leaf choice and nest height with information on 50 individuals confined at the
time of egg-laying to large screen cages placed over milkweed stems. Sample sizes
often differ because I did not gather data on all of the variables from each spider.

RESULTS

Distance between last feeding site and nest site. — Spiders typically placed their
egg-sacs several meters from their last hunting site, and many of them moved
toward the outside of the milkweed clone at this time. Individuals moving to the
periphery traveled over twice as far (9.6 ± 9.1 m S.D., N = 14) as did those
remaining within a clone (4.2 ± 3.2 m, N = 8) (P < 0.05 in a one-tailed t-test).
However, as indicated by the large standard deviations, distances moved were
extremely variable. Part of the difference in distances travelled could be a
consequence of milkweed stems, or other satisfactory nest-sites, being much less
dense around the periphery of the clone than in the center. If so, individuals
reaching the periphery would have to travel farther to reach the next stem than
would spiders in the center of the clone.

Characteristics of nest-sites. — Spiders laid their egg masses on the leaves of
both flowering and non-flowering milkweed stems. They selected stems that were
significantly shorter than those of the clone as a whole (Table 1), in large part

Table 1. — Stem choice of free-ranging Misumena vatia for nest sites on common milkweed.

Height of sample of Height of stems used P (Wilcoxon

stems in clone

by spiders

two-sample test)

Category

N

X + SD (cm)

N

X ± SD (cm)

Flowering and

143

' 68.0 ±17.2

65

62.6 ± 19.1

<0.01

non-flowering

Flowering

103

76.8 ± 10.4

25

84.4 ± 11.2

<0.01

Non-flowering

101

49.2 ± 15.3

40

51.1 ± 10.5

>0.3

'initially, I measured randomly-selected samples of 103 flowering and 101 non-flowering stems.
However, because the clone averaged 72% flowering stems and 28% non-flowering stems over the
study period, I produced the profile of flowering and non-flowering stems tested here by using the
entire sample of 103 flowering stems (72% of the sample) and 40 non-flowering stems randomly
chosen from the sample of 101 (the remaining 28% of the sample.)

MORSE— NESTS OF CRAB SPIDERS

385

the result of using significantly more non-flowering stems (40 of 65: Table 1) than
predicted from the numbers of flowering and non-flowering stems during this
study. Differences between the observed and predicted patterns of nest-site choice
ranged from P < 0.02 to P < 0.001, G = 6.61 to 23.50, respectively.)

In spite of their apparent preference for non-flowering stems, spiders used a
large number of flowering stems for nest sites. Because the clone is a rough
rectangle of 20 x 30 m, and because most flowering stems are located in the
center and non-flowering stems about the periphery (Fritz and Morse 1985), the
relatively immobile, egg-laden spiders hunting in the middle of the clone may
often not travel far enough to find a non-flowering stem. None of the spiders
whose last hunting site was unambiguously known (N = 22) used that site for its
nest.

Flowering stems used by spiders were significantly taller than the non-flowering
stems that they used (Table 2). I then tested whether the spiders exhibited height
preferences within these two categories of stems. They used flowering stems
significantly taller than predicted by the heights of flowering stems within the
clone, but the heights of non-flowering stems that they used did not differ
significantly from the height predicted by the numbers of non-flowering stems
within the clone (see Table 1). Thus, although spiders showed clear preferences
for non-flowering stems, they also exhibited height preferences among flowering
stems.

Spiders also placed their nests higher on flowering than non-flowering stems
(Table 2). Egg masses were most frequently placed near the tops of milkweed
stems (Table 2), often on the uppermost leaf wide enough to enclose them.
Usually this leaf lay no more than 3-4 cm below the top of the terminal shoot
of the stem. In few instances (8 of 93: 8.6%) did the spider build its nest more
than 10 cm below the tip of the terminal shoot. In one such instance an egg sac
was placed 45 cm up a 75 cm stem in a small axillary leaf that had developed
subsequent to the loss of an original leaf.

On both flowering and non-flowering stems spiders used significantly shorter
and narrower leaves than predicted by the mean sizes of leaves on these two stem
types (Table 3). These results suggest that the spiders actively selected relatively
small leaves, although they might also select leaves at the top of the stems, since
most of the small leaves grow there. Further, they also used significantly longer

Table 2. — Characteristics of nest sites on common milkweed. N’s not all equal because not all
variables were measured each year.

Variable’

Flowering stems

N X ± SD (cm)

Non-flowering stems

N X + SD (cm)

Height of stem

25

83.4 ± 12.4

40

51.1 + 10.5

<0.0001

Height of egg mass

25

76.6 ± 15.3

90^

47.8 + 10.5

<0.0001

Length of opposite leaf

16

10.2+ 1.9

65'

7.4+ 1.5

<0.03

Width of opposite leaf

16

3.0+ 0.8

65'

3.1 + 1.8

>0.7

'No significant between-year differences occurred in any of these variables (P >0.1 in each Kruskal-
Wallis Test)

^Difference between flowering and non-flowering stems; one-tailed Wilcoxon two-sample tests
^Includes 50 nests on caged, non-flowering stems. These nests were included since they did not differ
from those of free-ranging individuals in height of stem (P > 0.7). Neither did they differ in height
of egg mass (P > 0.9), length of opposite leaf (P > 0.4), and width of opposite leaf (P > 0.4)
(Wilcoxon two-sample tests).

386

THE JOURNAL OF ARACHNOLOGY

Table 3. — Leaf choice of Misumena vatia for nest sites on common milkweed.

Size of sample of Size of leaves P (Wilcoxon

leaves in clone used by spiders two-sample

Category N x ± SD (cm) N x ± SD (cm) test)

Flowering stems, length

124

14.2 ± 2.2

16

10.2 ± 1.9

<0.0001

Flowering stems, width

124

5.3 + 1.9

16

3.0 ± 0.8

<0.0001

Non-flowering stems, length

135

11.7 ± 4.1

65^

7.4 ± 1.5

<0.0001

Non-flowering stems, width

135

4.5 ± 1.9

65"

3.1 ± 1.8

<0.0001

’Mean length and width of leaves from 15 randomly-chosen flowering stems and 15 randomly-chosen
non-flowering stems. Leaves are paired: only one leaf of a pair, randomly selected, was measured.
Some of the lower leaves had fallen from flowering stems by the time spiders laid their eggs.

^Includes 50 nests on caged, non-flowering stems. These nests were included since they did not differ
from those of free-ranging individuals in height of stem (P > 0.7). Neither did they differ in length
of opposite leaf (P > 0.4) or width of opposite leaf (P > 0.4). (Wilcoxon two-sample tests).

leaves on flowering stems than on non-flowering stems; however, these leaves
were not wider than those used on non-flowering stems (Table 2). This result
suggests that leaf width is a more important factor than leaf length in determining
leaf choice.

I observed three spiders unsuccessfully attempting to fashion a large leaf into
a nest. Each one subsequently used a smaller leaf as a nest site. Although nests
in this study were invariably placed on milkweed leaves, no other plants within
the study site had leaves or leaflets nearly as large as those routinely used by the
spiders. Common forbs and shrubs in the area included cow vetch {Vida cracca),
yellow-rattle {Rhinanthus crista-galli), goldenrod {Solidago spp.), red clover
(Trifolium pratense), pasture rose (Rosa Carolina)^ and meadow-sweet (Spiraea
latifolia). All were periodically searched carefully for Misumena nests.

General description of nests. — On milkweed, Misumena initially bent under the
distal tip of a leaf, securing it to the under side of this leaf with a few strands
of silk. Typically they remained inside the resulting folded leaf for a day or more
(2.1 ± 1.6 S.D. days, N = 56 nests censused each morning and evening) before
laying their eggs there and completing the nest. However, a few individuals left
this site and moved to another site (N = 9), where they subsequently laid their
eggs.

Spiders turned under about one-third of a leaf in making a nest on milkweed.
In seven nests that I measured, the turned-under tissue averaged 2.7 cm (± 0.3
cm S.D.) in length.

Spiders always laid their eggs at night (N = 122), and by morning had
completed the majority of their nest construction. Although I did not observe
them exhaustively at night, I saw two different individuals laying eggs, at 23:30
and 01:30, respectively. By the time I observed spiders the following morning
(05:30 - 09:00), almost all of them had pulled the distal part of the leaf tightly
to the under side of the rest of the leaf (fig. 1). However, in three of 122 instances
(2.5%), spiders had not completed this process by morning, and the filmy silk
surrounding the egg sac proper remained visible. Spiders that laid their eggs the
preceding night had invariably positioned themselves on the under side of the nest
by morning (Fig. 1). The three spiders that did not completely close their nests
during the first night finished the task on the following night.

MORSE— NESTS OF CRAB SPIDERS

387

Within two to three days several spiders secured their nests with strands of silk
to surrounding vegetation. With few exceptions they attached it to an adjacent
leaf of the stem bearing the nest, especially the one immediately under the leaf
in which the nest was placed. Fig. 1 illustrates a typical pattern. This extra silk
increases the stability of the nest, and pulling it closer to the lower leaf may
provide the attending female with a hiding place (Fig. 1). However, it also
facilitates access to the nest from the flat surface of the leaf below.

Numerous additional lines of silk were often subsequently produced in the
immediate vicinity of the nest (Fig. 1). They do not play a major role in
stabilizing the nest, but are probably a mere consequence of the attending spider

Fig. 1. — Nests of the crab spider Misumena vatia on common milkweed. Upper: A nest secured
by strands of silk (immediately to left of spider’s right forelimbs) to the leaf immediately under it,
with the postreproductive female attending it in a position often assumed on nests drawn close to
other leaves. However, many of the nests are made on leaves that do not grow closely to other leaves,
and are therefore not attached to other structures. Strands of silk near the stem are intermittently
laid down over the life of the nest and are unlikely to play a major role in stabilizing the nest. Lower:
Nest rotated 90° to illustrate the position most frequently assumed by Misumena while attending nest.
This extremely large postreproductive individual was 10 mm long (abdomen + cephalothorax) and
weighed 1 10 mg. Illustrations by Elizabeth Farnsworth.

388

THE JOURNAL OF ARACHNOLOGY

moving about near the nest. These post-reproductive individuals invariably left
lines behind them wherever they moved.

I have occasionally found egg sacs of Misumena on other plants. Species used
as nest-sites included pasture rose (Rosa Carolina), spreading dogbane {Apocynum
androsaemifolium), chokecherry {Prunus virginiana), and sensitive fern {Onoclea
sensibilis).

DISCUSSION

Leaf choice. — Although crab spiders do not select milkweed leaves randomly,
the mechanism governing this choice is not explicit. Spiders might either choose
the highest leaves possible, or the leaves that provide ideal characteristics for a
nest. However, a few data do suggest that leaf characteristics, rather than
location, serve as the major basis for selection. Twice I have found spiders using
small axillary leaves well down on the stem as nest sites. These leaves, which
follow the loss of a main leaf, occur infrequently on milkweeds. They are also
often the only medium-sized leaves below the tops of flowering plants.
Additionally, individuals that used extremely large leaves experienced difficulty in
manipulating them, probably because these leaves were thicker and larger than
the ones usually used. All free-ranging individuals subsequently selected medium-
sized leaves for nests. Only caged spiders confined to large leaves eventually
fashioned nests out of them (Morse, unpubl.). In no instance did spiders using
large leaves appose the two parts of the leaf closely, and, consequently, large
areas of their nests were protected only by silk, probably making them extremely
vulnerable to attack by parasitic insects (see Eason, Peck and Whitcomb (1967).

Problems of nest construction could also account for Misumena^s absolute
choice of milkweed leaves as nest sites within the study area. Although the sample
of free-ranging spiders I studied used milkweed leaves exclusively under natural
circumstances, spiders that I confined to cages containing only pasture rose,
spreading dogbane, or meadow-sweet fashioned nests of these leaves (Morse,
unpubl.). None of these nests incorporated as much leaf material into the
covering as did nests on medium-sized milkweed leaves; thus, larger areas were
covered only by silk than in typical nests built on milkweed. It would be of
interest to determine the spiders’ relative preference for these different leaves as
a function of the area of the resulting nests covered only by silk, and also as a
function of the level of parasitism that such nests would experience. These nests,
and nests on large milkweed leaves as well, could also be subject to greater
desiccation than tightly-constructed nests. Hatching success of crab spider eggs in
nests that I opened soon after laying was significantly lower than that of
unmanipulated eggs (Morse, in prep.).

Movement to nest site. — The movement of spiders from their last hunting site,
despite a mobility that was often inadequate to reach a different habitat, suggests
a strong selective pressure for leaving the hunting site. Moving may minimize
vulnerability to egg parasites. Egg parasitism by an ichneumonid wasp (Trychosis
cyperia) and a phorid fly (Megaselia sp.) was relatively constant between 1980
and 1984, averaging about 15% per year (Morse and Fritz, in press). Simply
moving away from the previously-used site might on average be advantageous.
Moving is unlikely to bring a spider into a yet higher density of flowering stems

MORSE— NESTS OF CRAB SPIDERS

389

and other spiders, and possibly higher concentrations of parasites, and it could
take the spider to an area of lower density.

However, because some individuals moved much greater distances than the
mean distance between the last hunting site and the nest site, conflicting
pressures, rather than absolute mobility, may constrain the length of their move.
Dangers to the adults may be played off against those associated with egg
parasitism and egg predation. I do not have quantitative data on spider mortality
at this time, but because relatively long movements require descending into the
grass year, spiders may become vulnerable to predators unlikely to disturb them
on the milkweed stems. Two likely predators sometimes common in the study
area are the meadow vole {Microtus pennsylvanicus) and the garter snake
{Thamnophis sirtalis). Both species regularly prey on arthropods (Zimmerman
1965; Hamilton 1951). Further, during the one year of this study in which a
Microtus outbreak occurred, twice as many spiders disappeared after leaving their
last hunting site, not to be found again, than in any other year (Morse, unpubl.)

ACKNOWLEDGMENTS

I thank J. Eccleston and R. S. Fritz for comments on the manuscript. H.
Townes identified the ichneumonid wasp, and E. Fisher identified the phorid fly.
R. Bartlett generously permitted the use of his property. E. Cravey, C. Duckett,
E. Farnsworth, D. Forbes, R. S. Fritz, J. Hoffman, E. K. Morse, and A.
Raynsford assisted in the field. E. Farnsworth drew Fig. 1. This work was
supported by the National Science Foundation (DEB80-08502, DEB81-18105, and
BSR81-18105-A01).

LITERATURE CITED

Bristowe, W. S. 1958. The world of spiders. Collins, London.

Comstock, J. H. 1940. The spider book. Revised and edited by W. J. Gertsch. Doubleday, Doran,
New York.

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

Enders, F. 1977 Web-site selection by orb-web spiders, particularly Argiope aurantia Lucas. Anim.
Behav., 25:694-712.

Fritz, R. S. and D. H. Morse. 1985. Reproductive success, growth rate and foraging decisions of the
crab spider Misumena vatia. Oecologia, 65:194-200.

Hamilton, W. J., Jr. 1951. The food and feeding behavior of the garter snake in New York State.
Amer. Midland Nat., 46:385-390.

Hilden, O. 1965. Habitat selection in birds. Ann. Zool. Fennici, 2:53-75.

Morse, D. H. 1980. Behavioral mechanisms in ecology. Harvard University Press, Cambridge, Mass.
Morse, D. H. and R. S. Fritz. In press. The consequences of foraging for reproductive success. In
Foraging Behavior 11. (A. C. Kamil et al., eds.). Plenum, New York.

Partridge, L. 1978. Habitat selection. In Behavioral Ecology: an Evolutionary approach. First edition
(J. R. Krebs and N. B. Davies, eds.). Sinauer. Sunderland, Mass.

Zimmerman, E. G. 1965. A comparison of the habitat and food of two species of Microtus. J.
Mammal., 46:605-612.

Manuscript received March 1985, revised June 1985.

• T f

^4;

■"■; 'j!i j" "voiMVi’ \

m

f '• ' r

, t:tl4

' • . i. ' ; 1®

-”■■ f<V

r # ,

■•skr

b

. tS'^V .‘ ■'

''i‘.'*Ji-<,;.'¥lr-?::fl;-:

- f** — . ■ • "’s:

^,.1*

■1 ^..’ • 1

’^* iiwi Ji.‘r y 4

^ . ;■ , ,,-„ rg i .■!

=.i^ ,

'■ Jtiiji. '■A:, 1 /'■■ ..;» .

■- ■ ' 4 X' ‘ ■•' ^

.y''y'’i5b,y

-■4:^ ,

f $ji\'

, '■ - .'1
V5' '-r^j ^ -

II

I a

^■■•‘ y'-., ,vy-. J'- ’•

w I!

.. ... ,. _ . - - :7 • .• y: .

H " ' '•*%■ "^^^ -*■'-- V'^ "^‘i . '<5'^ ^■' • • "1^ M <^'4 > '".r. ,0 /5 ■'.,•■■ fJ’f’ ,

':-r '■ ' y'i“'^

*.•>' 1 '

^'Otfticll*

^'rf i

► / .■: "i ■**

.■f-fr

! ’4V;.’ ■' T'^'W !

-I -o-f*

i

f.kp^ ■

.i>*« -4^ «*.

.1, ;in'- nj.'Cr '•i,^MJ^:.,,sif- -

■ ■•^. ’ . u^. -vH- % 4

«y‘*V

»:-?l

:,>v ■„ -''iFA'iy • » ' >7{.>0-»7:,;

v- r.,

r. c^r

i — -ik

i . ;

.■^ J-'

n

:i5.i'^ 'x ■ / v; ;. .

, I V

S * * ■U^.'*’ JKkjf^j^ - ' \ 't

>' V. < • •'/< •(

i ; V' - .. • *

e77!f;-;yv^pi

* • ■ r t’ !

vi .. U

a^**

■■-..] .'^4 ;“a

' > ♦ ^o«

•> .■■■ •< i f* '?■ y;' v

■ tr?/ • cW^i

! ~;(kj*T -1 I. .' <• • ,!»r>"v<V »di'.

^ i;a . ;’■* ^

•'- ‘p: :■.

is

i'. W ^ -j^

■ .* .■' , ■'* ' ** -'■!
f;-’ ,■< ■ ‘.4 ; I '■1L

. *■• -»

^»V' a®

. > I

^ -■. -'iJ - ‘;h>-^;-; ', t>i .,-i'.,^ .,< i-

•/I * I> 7» ; ; s

>l«^«!i<» -.V .-•:'v ";‘■I>^ -

Vvi *

‘ < -Si

* V7

f '‘‘J: " ‘ V-<C* .

-•. '.L ^■-

-•Jtl.

•’ ‘‘^ £7fl]

’' ' ’”* '. ■ y’^’ i 'vV'''^f. '“'Wf ■

' ‘ i ■ . ‘WW , , * ' •

1985. The Journal of Arachnology 13:391

RESEARCH NOTE

SUBMERGENT CAPTURE OF DOLOMEDES TRITON
(ARANEAE, PISAURID AE) BY ANOPLIUS DEPRESSIPES
(HYMENOPTERA, POMPILIDAE)

Wasps of the family Pompilidae prey exclusively on spiders (Evans and
Yoshimoto 1962). Female pompilids generally paralyze their prey and transport
it to a nest before laying a single egg on the spider. One North American member
of this family that is distinguished by its unusual mode of prey transport is
Anoplius depressipes Banks. This species is the only non-parasitic, aquatic wasp
in North America (Hagen 1978). Evans (1949) accumulated several fragmentary
published and unpublished observations that he attributed to this species and
concluded that A. depressipes transports its prey across the surface film of the
water and is capable of crawling into the water and running on the bottom. The
only known prey of A. depressipes are adult female fishing spiders of the genus
Dolomedes (Evans 1949, Evans and Yoshimoto 1962, Kurczewski and Kurczewski
1968). These spiders are capable of diving into the water and remaining
submerged for more than 30 min (Carico 1973). It has remained unknown
whether A. depressipes ever stings Dolomedes spiders underwater before
transporting them to its nest (Hagen 1978). Despite statements by McCafferty
(1981) implying submergent prey capture by A. depressipes, there are no
published accounts confirming this type of behavior.

The purpose of this note is to record an instance of submergent prey capture
by A. depressipes that I observed at a small, artificial pond on Powdermill
Nature Reserve, 3 km S of Rector, Westmoreland County, Pennsylvania. At
approximately 1425 h on 1 August 1983, I noticed a large female Dolomedes
triton on the water surface about 1 m from the shoreline. My attention was
drawn to this spider initially because its behavior seemed peculiar. It dove below
the surface and held onto a submerged plant stem for no apparent reason. In
retrospect, I believe that the spider had detected the presence of a nearby A.
depressipes. The spider resurfaced within a minute or two. During the following
minute I observed a pompilid wasp (later identified as A. depressipes) fly toward
the spider from farther out over the pond. The spider undoubtedly saw the wasp,
because it ran rapidly away from the wasp along the water surface for about 25
cm before diving again. This time the spider stopped amongst denser vegetation
{Potamogeton sp.) at a depth of 15-20 cm, near the bottom of the pond. The
wasp followed the spider into the water 2-3 sec later and did not seem to slow
down as it broke the surface (at the same point where the spider dove). The wasp
swam down on an angle directly toward the spider and stung it within 5 sec. The
spider attempted to evade the wasp a second time before being stung, but barely
managed to start its legs in motion, and did not progress more than 2 cm. The

1985. The Journal of Arachnology 13:392

wasp and paralyzed spider reappeared at the surface within 2 sec of the attack.
The wasp then proceeded to drag the spider across the surface film as it flew
toward an inactive muskrat (Ondatra zibet hicus) burrow in the bank. I
interrupted this behavior on several occasions as the wasp approached the
shoreline in an effort to capture it. The wasp flew away each time but returned
to its victim within a few minutes, and eventually was allowed to resume its
behavior without further disturbance. The wasp transported the spider across the
water surface a total distance of about 2 m from where the attack occurred to
a point along the shoreline even with the burrow entrance. The wasp then
dragged the spider backwards (by biting a hind leg) another 20 cm across grassy
vegetation, mud and an exposed root, up and into the burrow, where it
disappeared from sight at 1436 h. I captured the wasp at the burrow on the
following afternoon, but was unable to locate the spider despite a 15 min search
of the emergent portion of the burrow’s interior.

Subsequently, on both 16 and 19 August 1983, I observed several large, black
pompilids (presumably A. depressipes) walking on floating water lily (Nymphaea
odorata) leaves in two nearby ponds. These wasps seemed to be actively searching
for Dolomedes spiders hiding inside curled-up water lily leaves, because they
moved systematically from leaf to leaf, peering into those that were curled up.
At 1851 h on 29 July 1985 I observed another A. depressipes dragging a
paralyzed D. triton female across the water surface. Both specimens were
collected as the wasp approached the opening of a partially submerged plastic
drainage pipe.

I thank Howard E. Evans for verifying the identity of the initial wasp, and
James E. Carico, William G. Eberhard, Howard E. Evans and C. J. McCoy for
reviewing the manuscript. All specimens have been deposited in the entomological
collection of Carnegie Museum of Natural History. These observations were
made while I was engaged in research supported by the M. Graham Netting
Research Fund of Carnegie Museum of Natural History.

LITERATURE CITED

Carico, J. E. 1973. The nearctic species of the genus Dolomedes (Araneae: Pisauridae). Bull. Mus.
Comp. Zool., 144:435-488.

Evans, H. E. 1949. The strange habits of Anoplius depressipes Banks: a mystery solved. Proc.
Entomol. Soc. Washington, 51:206-208.

Evans, H. E. and C. M. Yoshimoto. 1962. The ecology and nesting behavior of the Pompilidae
(Hymenoptera) of the northeastern United States. Misc. Publ. Entomol. Soc. Amer., 3:65-119.

Hagen, K. S. 1978. Aquatic Hymenoptera. Pp. 233-243 In: An Introduction to the Aquatic Insects
of North America. (R. W. Merritt and K. W. Cummins, eds.). Kendall/ Hunt, Dubuque.

Kurczewski, F. E. and E. J. Kurczewski. 1968. Host records for some North American Pompilidae
(Hymenoptera) with a discussion of factors in prey selection. J. Kansas Entomol. Soc., 41:1-33.
McCafferty, W. P. 1981. Aquatic Entomology. The Fishermen’s and Ecologists’ Illustrated Guide to
Insects and Their Relatives. Science Books International, Boston.

Steven M. Roble, Section of Amphibians and Reptiles, Carnegie Museum of
Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213, U.S.A.

Manuscript received March 1985, revised May 1985.

1985. The Journal of Arachnology 13:393

A HAWAIIAN WOLF SFIBER, LYCOSA HAWAIIENSIS
SIMON FORAGING IN THE TOP OF A
METROSIDEROS POLYMORPHA TREE

Of the lycosid spiders, most Lycosa species are ground dwellers (e.g., Dondale
et al. 1971, Turnbull 1973, Bixler 1970). However, Lycosa rabida and L.
punctatum have been found in vegetation (Kaston 1948, Barnes 1953, Whitcomb
et al. 1963) and L. rabida has been noted on the lower branches of trees
(Kuenzler 1958).

Several endemic Hawaiian species of Lycosa (see Simon 1900, Suman 1964,
Gertsch 1973) are abundant in subalpine, alpine and aeolian zones of Hawaii’s
highest mountains, and in aeolian habitats on fresh lava flows on the geologically
active island of Hawai‘i (Howarth 1979, Howarth and Montgomery 1980). Very
little is known of their ecology and behavior. Their presence in barren lava
regions is maintained by windborne prey transported onto the flows from
adjacent vegetated areas. I have observed Lycosa on ‘a‘a lava flows at and above
treeline on the island of Maui. Typically, the spiders forage at ground level, but
may perch upon higher vantage points such as lava boulders and outcrops. I had
never previously observed the spiders in low native shrubs nor collected them by
sweeping vegetation.

On 1 September 1982, 14:54 h, at 1700 m elev. in the Ko‘olau Gap of
Haleakala on the island of Maui, a mature female Lycosa hawaiiensis Simon (c.
2.75 cm body length) was observed stationed upon the apical tips of a
Metrosideros polymorpha tree (c. 2.4 m in ht.). The predominant vegetation was
subalpine scrub near the upper limit of Metrosideros in the gap. The nearest
neighboring tree was about 175 m away. Surrounding the tree were low, endemic,
xerophytic plants (Vaccinium, Coprosma, Deschampsia, Styphelia, etc.), growing
on largely unweathered ‘a‘a lava and cinder. The weather was mostly clear, with
convectional winds moving up the gap from the northeast, 5-10 knots. The
broken, rocky terrain provided habitat for an abundant Lycosa population. The
spiders were commonly seen running upon crags and between clinker. The
arboreal spider was collected after observations of her behavior and found to be
the same species as those seen on the ground.

The atypical location of the adult female spider above the substrate and in the
apical branches of a lone tree seemed unusual, so I stopped and observed her
behavior for about 15 minutes. The spider was perched at the terminal growth
of leaves, near several senescing inflorescences (about 35 cm from the nearest
cluster). The spider’s posterior three pairs of legs were clutched firmly upon the
apical leaf buds of the tree, which swayed about occasionally in the wind. The
front pair of legs were held free. The spider was motionless when first spotted,
but soon after observations began, several large sarcophagid flies arrived and flew
about the terminal branches of the tree in the vicinity of the inflorescences. When
the flies passed close to the spider, she darted after them, but failed to capture
any. I presented the knob-like end of a flowerstalk of the composite week,
Hypochaeris radicata, to the spider, and waved the flower bud close to her face.
The spider immediately lunged and grasped the bud in her chelicerae. Her grip
was tenacious enough so that she retained hold on the stalk when I released it.

It was clear that the spider was foraging in the tree, and had not been on the
terminal branchtips by chance. Vegetational perching is seen frequently among

1985. The Journal of Arachnology 13:394

the Thomisidae and Oxyopidae, and to a lesser extent, among the Salticidae.
Thomisids and salticids were represented in the area, but perhaps due to the more
open nature of the subalpine setting, lycosids were by far the most abundant
arachnids in the vicinity. Since the majority of the lycosid spiders seen at the site
were encountered on the ground, I initially suspected that the tree-inhabiting
individual was senescing or showing aberrant behavior. However, its vigorous
pursuit of flying insects and ready attack at a proferred object suggest otherwise.
Although previous surveys of canopy arthropods in Hawai‘i have not encountered
lycosid spiders (Gagne 1979), observation of a female lycosid on Dubuatia
menziesii (a sturdy, compact-leaved alpine shrub) in Haleakala (Gagne pers.
comm.) suggests that venturing upon vegetation during foraging activities may
not be unusual among the subalpine Lycosa in Hawai‘i. Further observations are
needed to clarify this point.

I wish to acknowledge critical comments on the manuscript by Wayne Gagne
and Francis Howarth of the Bernice Pauahi Bishop Museum, Department of
Entomology, Willis J. Gertsch in Portal, Arizona, and Phil S. Ward of the
University of California, Davis Entomology Department. Lawrence W. Pinter
kindly made the species determination of the voucher specimen (in the Bernice
Pauahi Bishop Museum).

LITERATURE CITED

Barnes, R. D. 1953. The ecological distribution of spiders in non-forest maritime communities at
Beaufort, North Carolina. Ecol. Monogr. 23:315-37.

Bixler, D. E. 1970. A study of wolf spider ecology in Grand County, Utah (Lycosidae: Araneae).
Southwest. Nat., 14:403-410.

Dondale, C. D., J. H. Redner, E. Parrel, R. B. Semple and A. L. Turnbull. 1971. Wandering of
hunting spiders in a meadow. Bull. Mus. Nat. Hist. Nat., 41:61-64.

Gagne, W. C. 1979. Canopy- Associated Arthropods in Acacia koa and Metrosideros tree communities
along an altitudinal transect on Hawaii island. Pacific Insects, 21:56-82.

Gertsch, W. J. 1973. The cavernicolous fauna of the Hawaiian lava tubes. 3. Araneae. Pacific Insects,
15:163-180.

Howarth, F. G. 1979. Neogeoaeolian habitats on new lava flows on Hawaii island: an ecosystem
supported by windborne debris. Pacific Insects, 20:133-144.

Howarth, F. G. and S. L. Montgomery 1980. Notes on the ecology of the high altitude aeolian zone
on Mauna Kea. Elepaio, J. Hawaii Audubon Soc., 41:21-22.

Kaston, B. J. 1948. Spiders of Connecticut. Connecticut Geol. Nat. Hist. Surv. Bull., 70:1-874.

Kuenzler, E. J. 1958. Niche relations of three species of lycosid spiders. Ecology, 39:494-500.

Simon, Eugene. 1900. Arachnida. Pp. 443-519, pis. 15-19, In: Fauna Hawaiiensis, (D. Sharp, ed.).
Cambridge Univ. Press, vol. 2.

Suman, T. W. 1964. Spiders of the Hawaiian islands: catalog and bibliography. Pacific Insects, 6:665-
687.

Turnbull, A. L. 1973. Ecology of the true spiders (Araneomorphae). Annu. Rev. Entomol., 18:305-
348.

Whitcomb, W. H., H. Exline and R. C. Hunter. 1963. Spiders of the Arkansas cotton field Ann
Entomol. Soc. Amer., 56:653-660.

Samuel M. Gon III, Animal Behavior Graduate Group, Department of
Entomology, University of California, Davis, California 95616. (Present address:
1604 Olalahina Place, Honolulu, HI 96817).

Manuscript received October 1984, revised March 1985.

1985. The Journal of Arachnology 13:395

ADHESIVE TECHNIQUE FOR CAPTURE
OF BURROW-DWELLING SPIDERS

The suitability of Geolycosa spp. for field studies has been recognized and
exploited in investigations of demography, wasp predation, activity patterns,
metabolism and thermoregulation (Humphreys 1974, 1975a, 1975b, 1978; Gwynne
1979; McQueen 1978, 1979, 1980, 1983; McQueen and Culik 1981). Adult spiders
of this group inhabit tubular burrows which extend straight down 20-40 cm,
terminating in a slightly enlarged side chamber at the bottom. Humphreys (1974)
employed traps to capture G. godeffroyi (L. Koch), but noted less than 100%
success in all populations studied. McQueen (1978) experimented with numerous
capture techniques for G. domifex (Hancock) ((= G. missouriensis (Wallace 1942))
and resorted to use of a medical otoscope for monitoring burrow occupants.

During an experimental field study of Geolycosa rafaelana (Ch.), I developed
a technique for capture of adults without damage to the spider or its burrow.
The capture device employed adhesive strips cut from the inner surface of Raid
Roach Traps (® 1980, S. C. Johnson and Son, Inc.). The adhesive was wrapped
around the end of a paper clip, which was suspended into the burrow with string.
A termite or ant was placed on the adhesive as bait, stimulating the spider to
grasp with its chelicerae, and thereby becoming firmly embedded in the adhesive.
Following extraction from the burrow, the spider was released by brushing a
drop of corn cooking oil against the adhesive surface.

This technique was used to capture more than 150 adult spiders for marking
and release during a two year period. Capture success was 100% except during
April-May, when females bearing egg cases were unreceptive to capture, and
would actively extrude the device from their burrows. Commercial adhesive for
trapping insects (Tangle-Trap) was found to be too thin for effectively embedding
the spider’s chelicerae. Passive traps employing adhesive strips placed within the
burrow entrance succeeded in snaring some individuals by one or more legs, but
leg loss (and resultant escape of spider) occurred when the strip was removed
from the burrow.

This study was supported in part by a grant from Sigma Xi, The Scientific
Research Society, and was conducted within an area designated for the purposes
of a National Science Foundation Long Term Ecological Research project at a
Chihuahuan Desert site near Las Cruces, New Mexico, USA. I thank Walter G.
Whitford and Walt Conley for advice and aid, Vince Roth and G. B. Edwards
for reviews of the manuscript, and June Peyton for manuscript typing.

LITERATURE CITED

Gwynne, D. J. 1979. Nesting biology of the spider wasps (Hymenoptera; Pompilidae) which prey on
burrowing wolf spiders (Araneae: Lycosidae, Geolycosa). J. Nat. Hist., 13:681-692.

Humphreys, W. F. 1974. Behavioral thermoregulation in a wolf spider. Nature, 251:502-503.

Humphreys, W. F. 1975a. The food consumption of a wolf spider, Geolycosa godeffroyi (Araneae:

Lycosidae), in the Australian Capital Territory. Oecologia, 18:343-358.

Humphreys, W. F. 1975b. The influence of burrowing and thermoregulatory behavior on the water
relations of Geolycosa godeffroyi (Araneae: Lycosidae), an Australian wolf spider. Oecologia,
21:291-311.

1985. The Journal of Arachnology 13:396

Humphreys, W. F. 1978. The thermal biology of Geolycosa godeffroyi and other burrow inhabiting
Lycosidae (Araneae) in Australia. Oecologia 31:319-347.

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

McQueen, D. J. 1979. Interactions between the pompilid wasp Anoplius relativus (Fox) and the
burrowing wolf spider Geolycosa domifex (Hancock). Canadian J. Zool., 57:542-550.

McQueen, D. J. 1980. Active respiration rates for the burrowing spider Geolycosa domifex
(Hancock). Canadian J. Zool., 58:1066-1074.

McQueen, D. J. 1983. Mortality patterns for a population of burrowing wolf spiders, Geolycosa
domifex (Hancock), living in southern Ontario. Canadian J. Zool., 61: 2758-2767.

McQueen, D. J. and B. Culik. 1981. Field and laboratory activity patterns in the burrowing wolf
spider Geolycosa domifex (Hancock). Canadian J. Zool., 59:1263-1271.

Wallace, H. K. 1942. A revision of the burrowing spiders of the genus Geolycosa (Araneae,
Lycosidae). Amer. Midi. Nat., 27:1-62.

Manuscript received October 1984, revised January 1985.

Marsha Reeves Conley, Department of Biology and Department of
Mathematics, New Mexico State University, Las Cruces, NM 88003.

FUNCTIONAL WEBS BUILT BY ADULT MALE
BOWL AND DOILY SPIDERS

It is generally accepted in the arachnological literature that adult female web-
building spiders build species-typical webs while adult males do no web-building
other than that required for courtship and sperm induction (e.g. Savory 1928,
Opell 1982), though there is at least one published report that adult male
uloborid spiders sometimes construct webs (Eberhard 1977). Immature bowl and
doily spiders (Frontinella pyramitela, Linyphiidae) of both sexes as well as adult
females have long been known to build relatively complex sheet webs consisting
of a bowl-shaped horizontal sheet, an underlying flat sheet (the doily), and a
barrier meshwork of silk that is above the bowl and doily.

We observed web-building by adult males while we were investigating the
behavioral effects of the chemical constituents of webs built by F. pyramitela
(Suter and Hirscheimer in press). In the course of those studies, we collected
spiders of all ages and both sexes from webs in Poughkeepsie and Millbrook,
New York, during May and June 1984. In the laboratory, the spiders were placed
on glass or wooden hexapods in 3.8 1 plastic jars where each could build a web
(for details of techniques, see Suter 1985). Usually within a day or two after a
spider built a web, the spider was removed from the web and placed on a new
hexapod, and the web was stored for subsequent testing. All spiders were
maintained in the laboratory on a diet of fruit flies {Drosophila melanogaster).

We recorded the data of construction of all webs and the molt dates of every
individual spider. These data allowed us to determine the last day on which male

1985. The Journal of Arachnology 13:397

spiders definitely could build webs. It did not enable us to determine what the
first day was on which they could not build webs. These two measures might well
have been different, since a spider may not have built a web on a given day
though he still had the capability to do so. Our method of measurement, the last
day on which males definitely could build webs, is actually the more conservative
of the two, and indeed male spiders may have had the ability to build webs on
several subsequent days.

Of the six immature males that reached maturity in the laboratory, two never
built species-typical webs as adults. Instead they built rudimentary platforms that
consisted of a diffuse and very small (1-3 cm diameter) horizontal sheet supported
by a few strands of silk below. These structures resembled crude versions of the
normal “doily.” Each of the remaining four adult males built a fully functional
and structurally complete species-typical web soon after the final molt. One male,
brought into the laboratory as an adult, also built functional and species-typical
webs. The last day on which we could ascertain that adult males built species-
typical webs ranged from 2 to 7 days after the final molt (Table 1). The last web
built by each male tended to be truncated (i.e. the bowl was relatively flat and
the vertical dimension of the entire web was reduced relative to the heights of
webs constructed by adult females and juveniles), but still had an identifiable
“bowl” and “doily.” These webs were as functional as those of other sex and age
classes in that prey items could still be captured on them.

Because adult males of this species emerge in the spring at the same time as
females (Suter 1985), adult males are not dependent upon their own web-building
abilities to capture food. During most of their active adult lives, males make use
of females’ webs for predation. This is rare among spiders because in most species
the males take no food during adulthood (Bristowe 1985). Male bowl and doily
spiders feed frequently while cohabiting on females’ webs and capture about 37%
of the prey that hit the web despite competitive activities of the females (Siiter
1985). Of what benefit then, might it be to adult males to expend energy and
nutrients in building webs when they could expend the same resources in search
of females?

When a male comes upon a female’s web that already has a resident male on
it, the two males engage in an an agonistic interaction which the larger (heavier)
male wins most of the time (Austad 1983, Suter and Keiley 1984). And when a
male comes upon a female’s web with no resident male on it, the male’s size will
likewise determine the outcome of contests with subsequent intruder males. Thus
male size is probably closely tied to male reproductive success and it will be
advantageous to adult males to be as large as possible when leaving their own

Table 1. — Construction of functional webs by adult male spiders.

Spider

Date of final

molt

Adult age (days) when
last web was built

1

5/26

7

2

5/28

3

3

5/30

6

4

5/31

2

5

♦

>5

♦Captured as an adult on 5 / 14.

1985. The Journal of Arachnology 13:398

webs. Because the mass of a spider, one measure of size, is directly related to
the amount of food that it consumes, a male may benefit by trying to capture
as many prey as he can for as long as he can, before setting off to find females.

According to Austad (1982), first male sperm priority is operative in bowl and
doily spiders; that is, the first male to inseminate a given female will fertilize
about 90% of her eggs. Finding virgin females, then, is crucial to the reproductive
success of a male, and any delay in beginning the search reduces the probability
that the male will find a virgin female.

Thus males have to weigh the benefits of capturing more prey against the costs
of delaying their search for unmated females. The optimum behavior for a male
depends upon at least three variables: nutritional status, season, and size. The
spider’s nutritional status determines what his energy reserves are and therefore
how long he can spend searching for females before starving. The time of year
is an important variable because the intensity of intermale competition varies with
male and virgin female densities which vary seasonally (Suter 1985). And size is
important because it is the primary determinant of male success in intermale
agonistic interactions. [Size can be measured both as lengths of body parts and
as mass. The latter is a good predictor of the outcome of intermale interactions
in bowl and doily spiders (Suter and Keiley 1984)].

Therefore, if a male were small at maturity, early encounters with females
might prove futile because of competition with other males, and the addition of
mass would be advantageous. In contrast, if the male were large, additional mass
might be far less advantageous than early encounters with females. This putative
relationship between mass at maturity and cessation of male web-building is
easily tested. We plan to test the relationship during the 1985 season.

We thank William Eberhard and Steven Austad for their helpful reviews of this
manuscript.

LITERATURE CITED

Austad, S. N. 1982. First male sperm priority in the bowl and doily spider, Frontinella pyramitela
(Walckenaer). Evolution, 36:777-785.

Austad, S. N. 1983. A game theoretical interpretation of male combat in the bowl and doily spider
{Frontinella pyramitela). Anim. Behav., 31:59-73.

Bristowe, W. S. 1958. The World of Spiders. Collins, London.

Eberhard, W. G. 1977. The webs of newly emerged Uloborus diversus and of a male Uloborus sp.
(Araneae, Uloboridae). J. Arachnol., 4:201-206.

Opell, B. 1982. Post-hatching development and web production of Hyptiotes cavatus (Hentz)
(Araneae, Uloboridae). J. Arachnol., 10:185-191.

Savory, T. H. 1928. The Biology of Spiders. The MacMillan Company, New York,

Suter, R. B. 1985. Intersexual competition for food in the bowl and doily spider, Frontinella
pyramitela (Araneae, Linyphiidae). J. Arachnol., 13:61-70.

Suter, R. B. and A. J. Hirscheimer. In press. Multiple web-borne pheromones in a spider, Frontinella
pyramitela (Araneae, Linyphiidae). Anim. Behav.

Suter, R. B. and M. Keiley 1984. Agonistic interactions between male Frontinella pyramitela
(Araneae, Linyphiidae). Behav. Ecol. Sociobiol., 15:1-7.

Andrea J. Hirscheimer and Robert B. Suter, Biology Department, Vassar
College, Poughkeepsie, NY 12601.

Manuscript received January 1985, revised June 1985.

1985. The Journal of Arachnology 13:399

ON THE CHILEAN SPIDERS OF THE FAMILY
PALPIMANIDAE (ARACHNID A, ARANEAE)

Only two species of palpimanid spiders have previously been recorded from
Chile. Anisaedus pellucidas Platnick is known only from Antofagasta and
Atacama provinces in northern Chile, and (judging by the presence of a
translucent embolar flange) seems most closely related to Anisaedus rufus
(Tullgren) of northwestern Argentina (Platnick, N. I. 1975. Amer. Mus. Novit.,
2562:1-32). Two localities can now be added to the only previous records of A.
pellucidas from Atacama province (Platnick, N. I. 1977. J. Arachnol., 3:203-205):
40 km S Copiapo (700 m), 23-25 October 1983 (L. E. Pena G.), 1 male, and
between El Transito and Pinte (1100-1600 m), 25-27 October 1980 (L. E. Pena
G.), 1 male, 1 female, all in the American Museum of Natural History (AMNH).

The second species, Otiothops lanus Platnick, is known only from the
Quebrada de la Plata in Santiago province. It differs from all other previously
known Otiothops in having relatively small posterior median eyes separated by
almost twice their diameter; this discrepancy was perhaps responsible for the
original description of the species as a Fernandezina (Zapfe, H. 1961. Invest.
Zool. Chilenas, 7:141-144). It was of particular interest, therefore, to discover
(among material recently sent to the AMNH by Dr. Luis E. Pena G.) a male
of a second Chilean species of Otiothops from Maule province. The new species
resembles O. lanus, rather than other Otiothops, in eye pattern. Although

Figs. 1, 2. — Left male palp of Otiothops maulensis: 1, ventral view; 2, retrolateral view.

1985. The Journal of Arachnology 13:400

outgroup comparison with the Stenochilidae indicates that the separated posterior
median eyes are plesiomorphic, the distolaterally originating and distally bifid
embolus suggests that the new taxon is indeed the sister species of O. lanus.

This work was supported by grants BSR-8312611 and BSR84-06225 from the
National Science Foundation. The illustrations are by Dr. M. U. Shadab.

Otiothops maulensis, new species
Figs. 1, 2

Type . — Male holotype from W of Cauquenes, Maule, Chile (May 1984, L.
Irarrazaval), deposited in AMNH.

Etymology. — The specific name refers to the type locality.

Diagnosis. — Males can be distinguished from those of all other Otiothops
except O. lanus by the widely separated posterior median eyes, and from those
of O. lanus by the proximally bulging embolus (Figs. 1, 2).

Male. — Total length 3.28 mm. Carapace 1.55 mm long, 1.15 mm wide. Femur
I 1.19 mm long, 0.54 mm wide. Cephalic area moderately elevated. Posterior
median eyes separated by twice their diameter. Abdomen with scattered brownish
purple patches on pale brown background. Claw tufts reduced to few setae
surrounding onychium. Palp with globose tibia, long narrow cymbium, and
retrolaterally prolonged bulb almost entirely occupied by reservoir. Embolus
originating retrolaterally near tip of bulb, distally bifid, with proximal lobe
translucent, distal lobe sharply pointed (Figs. 1, 2).

Female. — Unknown.

Material Examined. — Only the holotype.

Norman I. Platnick, Department of Entomology, American Museum of
Natural History, New York, New York 10024.

Manuscript received April 1985, accepted May 1985.

LABORATORY INFECTION OF GARYPUS CALIFORNICUS
(PSEUDOSCORPIONIDA, GARYPIDAE) WITH NEOAPLECTANID
AND HETERORHABDITID NEMATODES (RHABDITOIDEA)

Entomogenous nematodes of the genera Neoaplectana and Heterorhabditis are
being commercially produced as biological control agents for use against a variety
of insect pests (Poinar, Jr., G. O. 1983. Proc. Tenth Int. Congr. Plant Prot.
2:751-758). Tests are being conducted to examine the ability of these nematodes
to infect non-insect representatives of the Arthropoda. The present study was
conducted to determine if members of these nematode genera could infect
representatives of the Pseudoscorpionida under laboratory conditions.

1985. The Journal of Arachnology 13:401

The nematodes used in this study were the 42 strain of Neoaplectana
carpocapsae Weiser and the NC strain of Heterorhabditis heliothidis (Khan,
Brooks, and Hirschmann) which had been reared on wax moth larvae in the
laboratory.

Fig. 1. — An adult G. californicus killed by N. carpocapsae. Mature nematodes removed form the
pseudoscorpion surround the abdomen of the dead host.

Fig. 2. — Adults of N. carpocapsae removed from the abdomen of an infected G. californicus.

1985. The Journal of Arachnology 13:402

Specimens of the pseudoscorpion Garypus californicus Banks were collected at
Bolinas Point, California, on September 23, 1984. They were brought to the
laboratory and placed individually in 5 cm diameter petri dishes lined with a filter
paper disc which had an area of 18.1 cm^. One cc of infective stage nematodes
in water was added to the filter paper in each dish. This resulted in 12 x lO'*
nematodes/ dish for N. carpocapsae and 11 x 10^ nematodes/ dish for H.
heliothidis. Ten G. californicus were placed in dishes containing N. carpocapsae:
eight were challenged with H. heliothidis and four served as controls. Control
dishes had 1 cc of water only added to the filter paper and were maintained
similar to the treatments. A small piece of cotton gauze was placed in each dish
as a refuge for the pseudoscorpion, and adult Drosophila were added as a source
of food. The experiments were run at room temperature and extended for one
week.

At the time of death, a drop of hemolymph was removed from the
pseudoscorpion and plated out on a culture plate of Tergitol 7 agar plus TTC
(triphenyltetrazolium chloride). The symbiotic bacteria (Xenorhabdus spp.)
carried by the nematodes produce a characteristic color reaction on this medium.
The presence of the bacterium in a host’s hemolymph indicates that the
nematodes were able to infect and enter the body cavity of the host. Dissections
were performed at regular intervals after the pseudoscorpions died in order to
follow nematode development.

By the end of the second day after initial contact, all treated hosts were dead.
The control specimens remained alive for the duration of the experiment.
Samples of hemolymph removed from the dead hosts revealed the presence of the
nematodes’ symbiotic bacteria {Xenorhabdus spp.).

The nematodes developed to the adult stage and produced progeny inside the
dead pseudoscorpions (Figs. 1 and 2). However, foreign bacteria rapidly entered
the host cadavers and greatly lessened the conditions for nematode development.
As a result, few infective stages were produced.

This is the first report describing the ability of neoaplectanid and
heterorhabditid nematodes to infect pseudoscorpions. It indicates that these
nematodes are not as restricted in their parasitic habits as originally thought.

The present results show that G. californicus is highly susceptible to these
nematodes under laboratory conditions with all deaths occurring 1-2 days after
initial contact. However, this arachnid is a poor developmental host since bacteria
associated with the host enter the cadaver and inhibit establishment of the
nematode’s symbiotic bacteria which are required for parasite multiplication.

In a program involving the placement of these nematodes on the soil surface
of agricultural or horticultural land (Poinar, 1983, op. cit.), most pseudoscorpions
would avoid contact with the parasites by the cryptic nature of their physical
habitats (e.g. under bark of trees, in moss, under debris on the beach, etc.).

George O. Poinar, Jr. and G. M. Thomas, Department of Entomological
Sciences, University of California, Berkeley Calif. 94720, and Vincent F. Lee,
Department of Entomology, California Academy of Sciences, Golden Gate Park,
San Francisco, Calif. 94118-9961.

Manuscript received November 1984, revised February 1985.

1985. The Journal of Arachnology 13:403

AN UNUSUAL SPECIES OF TYRANNOCHTHONIUS
FROM FLORIDA (PSEUDOSCORPIONIDA, CHTHONIIDAE)

The genus Tyrannochthonius is tropicopolitan in distribution, including
tropical America (Hoff 1959, Mahnert 1979, Muchmore 1973, 1977); and though
it has long been known to occur in the United States (Chamberlin and Malcolm
1960), no U.S. species has ever been described. Now with our increasing interest
in and knowledge of West Indian species (Muchmore 1984, 1986), it seems
appropriate to describe a unique epigean form from Florida and Alabama.

Nearly 40 years ago J. C. Chamberlin recognized as new two individuals of
Tyrannochthonius found in northwestern Florida. He began a description of the
form but, unfortunately, never completed and published it; an illustration of the
palp was published, without a name, by Chamberlin and Malcolm (1960: Fig.
lA). Later, as part of a broad survey of the American species of
Tyrannochthonius, the senior author restudied Chamberlin’s specimens and
expanded the description. More recently the junior author has received and
studied new material from Florida and Alabama and has brought the description
to its final form.

Tyrannochthonius floridensis, new species
Figs. 1-4

Material. — Holotype male (WM 1664.01004) and 20 paratypes (6 males, 8
females, 6 nymphs) from litter in a sinkhole 3 mi. NW Marianna, Jackson
County, Florida, 8 September 1968, S. Peck; one paratype male from Florida
Caverns State Park, Jackson County, Florida, 9 September 1968, S. Peck; 2
paratypes (1 male, 1 female) from Rock Bluff, Liberty County, Florida, 27 April
1927, [T. H.] Hubbell; one paratype male from Moundville State Park,
Moundville, Hale County, Alabama, 7 July 1967, S. Peck and A. Fiske. All types
are in the Florida State Collection of Arthropods, Gainesville, except those from
Rock Bluff, which are in the Cornell University Insect Collection, Ithaca, NY.

Diagnosis. — A small, epigean species of Tyrannochthonius characterized by the
presence (usually) of a small seta at the base of the apical projection of the coxa
of leg I, and the absence (usually) of a preocular dwarf seta on the carapace;
tergites 1-4 each with 4 setae and tergites 5-8 each with 6 setae.

Description (based mainly on 12 mounted specimens). — Male and female very
similar, though female usually a little larger and with slightly more robust palps.
Carapace subquadrate; epistome a low, rounded lobule, barely evident between
the 2 median setae (Fig. 1); with 4 conspicuous, corneate eyes; chaetotaxy usually
4-4-4-2-2, though two paratypes have a single preocular dwarf seta on one side
or the other. Coxa I with a broad, rounded apical projection, which usually bears
a small seta (m) at its base mesially (Fig. 2); complete coxal chaetotaxy 2-2-1 :m-
3-0:2-lor2-CS:2-3:2-3; each coxa H with an oblique row of 6-10 terminally incised
spines (CS).

Abdomen typical. Tergal chaetotaxy of holotype 4:4:4:4:6:6:6:6:7:4:T2T:0;
others similar. Sternal chaetotaxy of holotype (male) 10:[4-4]:(3)13-12/
6(3):(3)6(3):8:8:8:8:8:9:0:2; other males similar; female usually 10:(3)6(3):(3)6(3):-.

1985. The Journal of Arachnology 13:404

1

Figs. 1-4. — Tyrannochthonius floridensis, new species: 1, epistome and flanking setae on anterior
margin of carapace; 2, anteromedial part of left coxa I, showing apical projection and setae; 3, dorsal
view of right palp; 4, lateral view of left chela.

Chelicera about Ya as long as carapace; hand with 5 setae; flagellum of 6-7
irregularly pinnate setae; fixed finger with one large distal tooth followed
proximally by 7-9 much smaller teeth; movable finger with about 15 small teeth;
galea a low elevation in both sexes.

Palp as shown in Fig. 3; femur 4.4-4. 6, tibia 1.8-1.95, and chela 4.65-5.45 times
as long as broad; hand 1. 7-2.0 times as long as deep; movable finger 1.6-1.85
times as long as hand. Trichobothria as shown in Fig. 4; on movable finger sb
distinctly nearer to b than to st. Hand with one heavy spinelike seta on medial
side near base of fingers. Fixed finger with 16-20 widely spaced, prominent
macrodenticles and 12-16 microdenticles interspersed distally; movable finger with
7-10 macrodenticles distally and 6-9 interspersed microdenticles, and 8-10 very
low, rounded teeth basally. Sensillum at dental margin, usually nearer to
trichobothrium st than to sb.

Legs typical: leg IV with entire femur 2.25-2.5 and tibia 4.1-4.55 times as long
as deep; a prominent tactile seta on the proximal third of telotarsus IV.

Measurements (mm). — Figures given first for the holotype followed in
parentheses by ranges of the 11 mounted paratypes. Body length 1.31 (1.00-1.40).
Carapace length 0.39(0.335-0.445). Chelicera 0.29(0.26-0.325) long. Palpal femur
0.415(0.375-0.465) by 0.085(0.08-0.105); tibia 0.185(0.17-0.205) by 0.095(0.085-
0.11); chela 0.60(0.53-0.69) by 0.11(0.11-0.14); hand 0.22(0.20-0.26) by

1985. The Journal of Arachnology 13:405

0.115(0.105-0.15); movable finger 0.375(0.355-0.43) long. Leg IV: entire femur
0.385(0.355-0.42) by 0.155(0.155-0.18); tibia 0.28(0.245-0.31) by 0.065(0.06-0.07);
metatarsus 0.13(0.115-0.15) by 0.05(0.045-0.055); telotarsus 0.26(0.23-0.295) by
0.035(0.03-0.04).

Etymology.— The name originally selected by J. C. Chamberlin is retained,
floridensis referring to Florida where the first specimens were found.

Remarks. — T. floridensis is apparently unique in the genus in having a small
seta on the apical projection of coxa I. Such a seta (sometimes 2 or 3) is regularly
present in many chthoniid genera, but not in Tyrannochthonius or the related
Lagynochthonius and Paraliochthonius (cf. Muchmore 1984, 1986). It remains to
be seen whether its occurrence is of any phylogenetic significance.

In addition, T. floridensis is unusual in having the epistome very low, broad,
and rounded. The usual form of the epistome in epigean species of
Tyrannochthonius is distinctly triangular, closely flanked by 2 setae (cf.
Muchmore 1984). Some cavernicolous species of the genus have low, rounded
epistomes (unpublished observations), but in few if any are they as insignificant
as in T floridensis.

We acknowledge the first recognition of the species by the late J. C.
Chamberlin and are greatly indebted to S. B. Peck for providing the more recent
material.

LITERATURE CITED

Chamberlin, J. C. and D. R. Malcolm. 1960. The occurrence of false scorpions in caves with special
reference to cavernicolous adaptation and to cave species in the North American fauna (Arachnida
— Chelonethida). Amer. Midi. Nat., 64:105-1 15.

Hoff, C. C. 1959. The pseudoscorpions of Jamaica. Part I. The genus Tyrannochthonius
(Heterosphyronida: Chthoniidae). Bull. Inst. Jamaica, Sci. Ser., 10(1): 1-39.

Mahnert, V. 1979. Pseudoscorpione (Arachnida) aus dem Amazonas-Gebiet (Brasilien). Rev. suisse
Zool., 86:719-810.

Muchmore, W. B. 1973. A second troglobitic Tyrannochthonius from Mexico (Arachnida,
Pseudoscorpionida, Chthoniidae). Assoc. Mexican Cave Stud. Bull., 5:81-82.

Muchmore, W. B. 1977. Preliminary list of the pseudoscorpions of the Yucatan Peninsula and
adjacent regions, with descriptions of some new species (Arachnida: Pseudoscorpionida). Assoc.
Mexican Cave Stud. Bull., 6:63-78.

Muchmore, W. B. 1984. Pseudoscorpions from Florida and the Caribbean area. 13. New species of
Tyrannochthonius and Paraliochthonius from the Bahamas, with discussion of the genera
(Chthoniidae). Florida Entomol., 67:119-126.

Muchmore, W. B. 1986. New species of Tyrannochthonius and Lagynochthonius from caves in
Jamaica, with discussion of the genera, (in preparation).

David R. Malcolm, Pacific University, Forest Grove, Oregon 97116, and
William B. Muchmore, Department of Biology, University of Rochester,
Rochester, New York 14627.

Manuscript received August 1984, revised November 1984.

1985. The Journal of Arachnology 13:406

EFFECTS OF METHOD AND TIME OF PRESERVATION ON
VOLUMETRIC MASS ESTIMATES OF SPIDERS (ARANEAE)

Various techniques have been used to back-estimate the live masses (i.e. the
masses prior to killing and preservation) of preserved spiders (see Hagstrum 1971,
Rogers et al. 1977, and Greenstone et al. 1985a, for literature reviews). We have
recently described a technique in which live masses were regressed on an estimate
of volume derived by treating the specimen as a cylindrical solid of uniform
density having height equal to total length and diameter equal to the mean of
greatest widths of the carapace and abdomen. A regression containing such data
for 101 animals of the families Lycosidae, Salticidae, Araneidae, Oxyopidae, and
Thomisidae explained 95.7% of the variation in mass and was superior to
traditional methods using total length or carapace width rather than volume as
the estimator (Greenstone et al. 1985a).

Although the technique employed spiders which were preserved directly in 70%
ethanol, it was intended for estimating live masses of spiders which had been
sticky-trapped using the adhesive Tack Trap'^“(Animal Repellents Inc., Griffin,
Georgia). Such specimens must be soaked for four days each in paint thinner and
toluene before final preservation in 70% ethanol. This series of solvent changes
could conceivably affect their shape and size and, hence, volume estimate. In
order to determine whether the volume-mass regression would be different using
animals trapped in this fashion, and also to determine whether the time since
preservation affects the regression of mass on the volume estimate, we performed
the following experiment.

On June 12, 1984, we collected spiders by sweeping in native tall grass prairie
at the Tucker Prairie Preserve, 27 km east of Columbia in Callaway Co.,
Missouri. In order to minimize variability only araneids were used. The animals
were returned alive to the laboratory and weighed on a Mettler AE 160 electronic
balance. Following this they were ranked from lowest to highest mass and then
assigned serially and alternately to either of two treatment groups to ensure that
both groups covered approximately the same range in masses. The first group
(hereinafter referred to as “direct-ethanol”) was preserved directly in 70% ethanol.
The second (hereinafter “sticky-trapped”) was placed on a previously prepared
12.7 mm (1/2") mesh hardware cloth sticky trap (Greenstone 1985b) with the exact
location of each animal on the trap recorded. The trap was then placed in the
field for a week as per our normal protocol. Following this the animals were
removed from the trap and placed for four days each in paint thinner and toluene
before final preservation in 70% ethanol.

To determine the effect of preservation time on the volume estimate the
animals in each set were measured five times at set intervals. We anticipated that
the most rapid changes would happen in the early phases of preservation and
therefore made our first two measurements at weekly intervals. To minimize the
possible adverse effects of excessive handling of the specimens we made the
remaining three measurements at bi-weekly intervals. Overall, then, the
measurement period covered a total of eight weeks following preservation.

Sign tests (Siegel 1956) were performed on the volume estimates at the
beginning and end of each interval to determine whether significant increases or

1985. The Journal of Arachnology 13:407

Fig. 1. — Regression lines, 95% confidence bands, and data for volume-mass regressions for original
direct-ethanol data set (Greenstone et al. 1985a) (Fig. lA) and for sticky-trapped set (Fig. IB). Fig.
lA shows only that portion of the original data set covering the same volume range as the sticky-
trapped set.

decreases had occurred during the interval. Sample sizes for each comparison
were fifteen for the direct-ethanol set and eighteen for the sticky-trapped set. In
the direct-ethanol there was a significant (P <0.025) decrease amounting to 19%
(arcsin transformation of original data) in the volume estimate in the first week
of preservation. None of the succeeding intervals showed significant change
although there was an almost-significant (P <0.10) increase of 22% between the
first and fourth week estimates. The sticky-trapped set showed a significant (P
<0.02) increase of 9.5% during the first week interval and a significant (P <0.02)
decrease of 6.0% during the second; all subsequent intervals were non-significant
(P >0.60). In both samples, then, following one or two weeks of alternating
increases or decreases in volume estimate there were no significant differences
between the initial (one week) estimate and subsequent estimates following not
more than six weeks preservation. This appears to be ample time to wait before
making measurements to be sure that further changes will not occur.

To determine whether the regressions of live mass on that of direct-ethanol
preserved and of previously sticky-trapped spiders differ, the sticky-trapped
sample may be compared with either the simultaneously preserved direct-ethanol
fifteen animal set of araneids or the original 101 animal direct-ethanol set
(Greenstone et al. 1985a). Figure 1 shows the data for the sticky-trapped set (Fig.
IB) and that portion of the 101 animal set which covers the same range in
volume (Fig. lA), and the 95% confidence bands for the complete data sets. The

1985. The Journal of Arachnology 13:408

variances of these two samples are significantly different (P <0.01, Bartlett’s Test
for Homogeneity of Variances, Sokal and Rohlf 1969). Therefore t-tests on slopes
and intercepts with unequal variance were performed (Snedecor and Cochran
1967). Both of these were non-significant (t = 0.690 and t = 2.880, respectively,
P >0.50 in both cases). Failure to reject is not due to the added variance in the
101-animal set due to inclusion of non-araneids, because comparison of the
sticky-trapped set with the simultaneously direct-ethanol preserved araneid set is
not significant (t = 1.1223, P >0.20, t = 0.1002, P >0.50, slope and intercept,
respectively).

There is therefore no evidence that prior sticky-trapping followed by passage
through paint thinner and toluene alters the relationship between estimated
volume and live mass for ethanol preserved spiders.

We thank Charles Dondale, Herb Levi, Norm Marston, George Uetz, and Jan
Weaver for reviews of the manuscript. A.-L. H. was supported by a U.S. De-
partment of Agriculture Research Apprenticeship.

LITERATURE CITED

Greenstone, M. H., C. E. Morgan, and A.-L. Hultsch. 1985a. Ballooning methodology: equations for
estimating masses of sticky-trapped spiders. J. Arachnol., 13:225-230
Greenstone, M. H., C. E. Morgan, and A.-L. Hultsch. 1985b. Spider ballooning: development and
evaluation of field trapping methods (Araneae). J. Arachnol., 13:337-346.

Hagstrum, D. W. 1971. Carapace width as a tool for evaluating the rate of development of spiders
in the laboratory and field. Ann. Entomol. Soc. Amer., 64:757-760.

Rogers, L. E., R. L. Buchsaum, and C. R. Watson. 1977. Length-weight relationships of shrub-steppe
invertebrates. Ann. Entomol. Soc. Amer., 70:51-53.

Siegel, S. 1956. Nonparametric statistics. McGraw Hill, New York. 312 pp.

Snedecor, G. W. and W. G. Cochran. 1967. Statistical methods. Sixth, Ed. Iowa State University
Press, Ames. 593 pp.

Sokal, R. R. and F. J. Rohlf. 1969. Biometry. W. H. Freeman and Co., San Francisco. 776 pp.
Manuscript received December 1984, revised January 1985.

Matthew H. Greenstone, Anne-Lise Hultsch, and Clyde E. Morgan, U. S.

Department of Agriculture, Biological Control of Insects Research Laboratory,
R O. Box 7629, Research Park, Columbia, Mo. 65205.

1985. The Journal of Arachnology 13:409

THE JOURNAL OF ARACHNOLOGY

Instructions to Authors

GENERAL COMMENTS

Manuscripts are acceptable in English, French, Portuguese, and Spanish, and must be TYPED
DOUBLE OR TRIPLE SPACED THROUGHOUT. Use only paper 8.5 by 1 1 in. in size, but not
erasable bond. Leave margins at least 1.5 in. (4 cm) on the left and 1 in. (2.5 cm) on the top, bottom,
and right, and do not hyphenate any words at the right margin. To facilitate prompt review by two
or more referees, send the Associate Editor the original manuscript and two good copies, together
with copies of the illustrations, for mailing to referees. Follow the recommendations of Council of
Biological Editors (CBE) Style Manual, Third Edition, unless indicated otherwise below. Do not edit
your own manuscript: italics are permitted only to indicate scientific names; only the TITLE,
PRIMARY HEADINGS (e. g. INTRODUCTION, etc.) and RUNNING HEAD should be typed in
capital letters.

Manuscripts longer than about 1500 words (five double-spaced typewritten pages in Elite, six
double-spaced typewritten pages in Pica) should be prepared as Feature Articles, shorter papers as
Research Notes.

Manuscripts and all related correspondence must be sent to Dr. William B. Peck, Associate Editor,
Department of Biology, Central Missouri State University, Warrensburg, Missouri 64093, U.S.A.

FEATURE ARTICLES

Arrange the various parts of feature articles in the following sequence: (1) mailing address, (2) title,
(3) by-line, (4) abstract, (5) body of text, (6) acknowledgments, (7) literature cited, (8) figure legends,
(9) footnotes, (10) running head, (11) tables with legends, (12) figures. Put only numbers 1, 2 and
3 above on page 1, and start page 2 with number 4 above, numbering ali other pages consecutively.

Mailing address. — Include the complete address and the telephone number of that author with
whom all correspondence with the Editor should be handled.

Title. — When using common or scientific names in the title be sure to include in parenthesis the
appropriate higher taxa (order, family, . . .). Include footnote indication if appropriate (e. g., to
acknowledge grant support), but type footnote separately (see number 9 above). Title should be typed
in capital letters and may not exceed 48 characters and spaces per line.

By-line. — Include name(s) of author(s) as customarily written (less titles) and complete address(es)
including zip code or other postal designation. Include footnote indication(s) if appropriate (e. g., to
indicate change of address), but type footnote separately (see number 9 above). Leave ample space
between title and by-line, and four spaces between name(s) and address(es).

Abstracts. — Abstracts should be a summary of the basic findings, should not exceed 2 to 3 percent
of the text in length. Papers in a language other than English must be accompanied by an English
abstract as well as an abstract in the language of the text.

Body of text. — Use whatever form seems best to accommodate the necessary description of
research. Be concise. Conform to CBE Style Manual in all abbreviations. Use the metric system on
all measurements (the English system is acceptable only when transcribing locality data accompanying
museum specimens), and note that abbreviations of metric units of measurement are not punctuated
(e. g., mm and km, but ft. and mi.). Citations should be in the form: Bellrose (1950), or Bellrose
(1950:33), or (Bellrose 1950). The complete scientific name of a species or genus, including author(s),
must be given the first time they are mentioned in the text. Use single-line notation for fractions
(e. g., 1/4 and not \, (4-12)/3 and not 4-12).

4 3

The following special directions apply to authors of taxonomic papers:

(a) . Do not use abbreviations to indicate that a new name or a new combination is being proposed
in a primary heading (e. g., A-us x-us, new species, rather than A-us x-us n. sp. or comparable
abbreviations).

(b) . Keys must be typed as follows:

1. Use Arabic numerals to designate the leading entry of a couplet 2

Do not designate the second entry of a couplet, either by means of numbers, letters, or other
marks 3

1985. The Journal of Arachnology 13:410

2. Type numbers flush to left margin, and start entry on fifth space. The second, and subsequent,
lines of one entry must also be indented five spaces .species

(c) . Synonymies must follow the abbreviated style shown below:

A-us x-us Jones 1930:3, 1935:9; Russell 1945:453; Smith 1954a: 16, 1954b:678; Cooper and Lim
1955:18 (in part).

A-usy-us Bates 1932:18, fig. 4. NEW SYNONYMY.

A-us z-us: Miranda 1948:98 (misidentification); Harris 1951:3 (in part ?). (nee A-us z-us Zimmer).

(d) . Lists of specimens examined of a given taxon must be the last item typed in the treatment of
that taxon as they will be set in smaller type. Adhere to the following style for listing specimens
examined: Country: state or comparable political subdivision; county or district, detailed locality
(elevation), 14 July 1945 (collector), 2 males, 5 females (acronym of institution where specimens are
deposited), next detailed locality within that county, and so forth; next county in the same state; and
so forth: next state in the same country: and so forth. Next country: and so forth. Punctuation rules
are very simple. Use a period to separate countries, colon to separate states, semi-colon to separate
counties, and commas to separate specific localities.

Acknowledgments. — Avoid overlooking persons who have in some substantial way assisted with the
work. Authors of taxonomic papers should spell out the name, and indicate parenthetically the
acronym, of institutions where specimens studied are deposited.

Literature cited. — Include only those publications to which reference is made in the text. Adhere
to the CBE Style Manual or refer to a previous issue of The Journal of Arachnology for style. Do
not abbreviate place names in journal citations. Repeat name(s) of author(s) in case of multiple
entries.

Figure Legends. — Provide one legend for each illustration to be reproduced singly, or for each
“plate” consisting of several illustrations. Adhere to the following styles:

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

Figs. 27-34. — Right chelicerae of species A-us from Timbuktu: Figs. 27, 29, 31, 33. — Dorsal views;
Figs. 28, 30, 32, 34. — Prolateral views of movable finger; Figs. 27-28: A-us x-us, holotype male; Figs.
29-30: A-us w-us, male; Figs. 31-32: A-us z-us, holotype male; Figs. 33-34; A-us t-us, male. Scale =
1.0 mm.

Type all figure legends consecutively on same page(s), using double space within each legend and
leaving 4 spaces between legends. Keep in mind that 85 characters and spaces represent one printed
line; for each line subtract 4 mm from the maximum length permissible on full-page illustrations.

Footnotes. — Footnotes will be permitted on page 1 when it is appropriate to acknowledge grant
support and to indicate change of address, etc. No footnotes will be permitted on any other page.
Type footnotes on a page by themselves.

Running head. — At the top of each right hand page will be printed the author(s)’ last name(s) and
an abbreviated title. This running head may not exceed 55-60 characters and spaces, must be typed
in capital letters, and a hyphen must separate the name(s) from the title.

Tables with legends. — Prepare all tables precisely as they are to be typeset. Construct tables as
simply as possible. Include legend at the top of each. Make marginal notations in the text which
clearly indicate the appropriate point of insertion of each table. Note that in these instructions
“graphs” are regarded as “figures,” not “tables.”

The size of the printed page imposes a limit on the size of tables than can be accepted. On a
normal, upright table these limits are 99 characters and spaces per line and 56 lines long, including
legend. Sideways tables are difficult to typeset and present problems during paste-up; therefore, they
will be accepted in exceptional cases only, as most tables that are “too wide” can usually be re-
arranged to fit into the more desireable, upright position. Any table that exceeds the dimensions given
above will not be accepted for publication. Tables must be typed double spaced throughout, and the
legend must be in the following style:

Table 1. — Incorporate into legend all information necessary to make the data presented clearly
understandable. Do not resort to use of footnotes.

Figures. — Figures are the only parts of the paper for which the author is responsible for preparing
camera-ready material, and they must be done with great care. Drawings should be done with black
ink on high quality, smooth, white, heavy paper. Drawings done in some manner not employing black

1985. The Journal of Arachnology 13:41 1

ink (e. g., shaded pencil drawings) are difficult to print, more so than photographs. When such a
technique is used, the drawings should be fairly dark, have good contrast, and be well executed. The
largest dimensions on the printed page to which illustrative material (with legends) will be reduced
is 127 by 200 mm. Decisions as to size of originals, placement of several smaller drawings on a single
page, etc., should be based upon this format. Where several smaller drawings are to appear on a
single page, it is the responsibility of the author(s) to furnish an assembled “plate” precisely as it
should appear in print. The “plate” must be prepared to allow for inclusion of the legend beneath
the drawings. Number all drawings consecutively using Arabic numerals, do not designate illustrations
with letters and never designate plates. Labelling should be done neatly, using either pressure sensitive
characters or a lettering aparatus employing black ink. Photographs should have good contrast, be
glossy or semi-glossy, and printed slightly larger (up to 2X) than anticipated published size.
Illustrative material of any kind larger than two times the anticipated size will not be accepted. Larger
illustrations are difficult to handle and get easily damaged or lost in the mail. Authors preparing
larger illustrations should consult local photo shops and printing firms about the availability of
PMT’s (PhotoMechanical Transfer) for reducing their illustrations to acceptable size. PMT’s are high
contrast, direct positive-to-positive reproductions on semi-glossy paper which will also give the
author(s) a good idea of how the illustrations will look when printed (The Journal of Arachnology
does not provide proofs of illustrations). PMT’s can be obtained at a reasonable cost in a number
of sizes, usually 8 by 10 in. or 9 by 12 in., with any amount of figure reduction desired within that
space. Sending PMT’s rather than originals avoids the loss of art work in the mail, and if the PMT’s
get lost or damaged they can be easily replaced. Illustrations, PMT’s, and photographs must be
mounted on hard white mounting board (Scotch Magic Transparent Tape can be used). On the back
of each plate include author’s name(s), abbreviated title of manuscript (or running head), figure
numbers included, and orientation (= topside). Illustrative material that is poorly prepared, too large,
or not mounted properly will be returned to the author.

RESEARCH NOTES

Arrange the various parts of your research notes in the following sequence: (1) mailing address,
(2) title, (3) body of text, (4) by-line, (5) figure legends, tables with legends, and illustrations. Follow
instructions given above for feature articles unless otherwise indicated below. Do not include
footnotes anywhere, references to grant support and all other acknowledgments should be made in
a statement in the body of the text. If an indication of change of address is desired, it should be
included parenthetically after the credited institution as “present address.” If five or fewer references
are cited in the text, these should be made parenthetically where they appear in the text following
style set for feature articles but omitting the title of the article cited (e. g., Jones, J. 1967. J.
Arachnol., 2:199-214). If more than five citations are to be made, however, separate them from the
body of the text and include them in their own section as in feature articles. The by-line must be
typed in paragraph form after the body of the text, or the literature cited when one is present.

GRANTS-IN-AID FOR RESEARCH

Grants-in-Aid for research on Arachnida (excluding Acarina) and Myriapoda
are made available to students and researchers through the Exline- Frizzell Fund
for Arachnological Research^' of the California Academy of Sciences.
Applications, which will be evaluated by the American Arachnological Society
and the Department of Entomology, California Academy of Sciences (Golden
Gate Park, San Francisco, California 94118-9961, phone [415] 221-5100), may
be submitted to the latter. Application forms may be obtained upon request.
Awards will be made upon the approval of the Academy’s Director shortly after
March 1 and September 1 yearly. Grants will normally not exceed $750. The
Exline- Frizzell Fund may be used for fieldwork, museum research (including
travel), expendable supplies, and costs of publications (including artwork).

Deadlines for applications are 31 December for March awards and 30 June for
September awards.

1985. The Journal of Arachnology 13:412

THE AMERICAN ARACHNOLOGICAL SOCIETY

President:

Jerome S. Rovner (1985-1987)

Department of Zoology
Ohio University
Athens, Ohio 45701

Membership Secretary:

Norman I. Platnick (appointed)

American Museum of Natural History
Central Park West at 79th Street
New York, New York 10024

Secretary:

Brent D. Opell

Department of Biology
V.P.I.S.U.

Blacksburg, Virginia 24061

The American Arachnological Society was founded in August, 1972, to
promote the study of Arachnida, to achieve closer cooperation between amateur
and professional arachnologists, and to publish The Journal of Arachnology.

Membership in the Society is open to all persons interested in the Arachnida,
Annual dues are $25.00 for regular members, $15.00 for student members.
Correspondence concerning membership in the Society must be addressed to the
Membership Secretary. Members of the Society receive a subscription to The
Journal of Arachnology. In addition, members receive the bi-annual newsletter
of the Society, American Arachnology.

American Arachnology, edited by the Secretary, contains arachnological news
and comments, requests for specimens and hard-to-find literature, information
about arachnology courses and professional meetings, abstracts of papers
presented at the Society’s meetings, address changes and new listings of
subscribers, and many other items intended to keep arachnologists informed
about recent events and developments in arachnology. Contributions for
American Arachnology must be sent directly to the Secretary of the Society.

Manuscripts for THE JOURNAL OF ARACHNOLOGY are acceptable only
from current members of the Society, and there are no page charges. Detailed
instructions for the preparation of manuscripts appear in the Fall issue of each
year, and can also be obtained from the Editor and the Associate Editor.
Manuscripts that do not follow those instructions will be returned to the
author(s) without benefit of review. Manuscripts and all related correspondence
must be sent to Dr. William B. Peck, Associate Editor, Department of Biology,
Central Missouri State University, Warrensburg, Missouri 64093, U.S.A.

Presiden t- Elect:

William A. Shear (1985-1987)
Department of Biology
Hampden-Sydney College
Hampden-Sydney, Virginia 23943

Treasurer:

Norman V. Horner (1985-1987)
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308

Directors:

Frederick A. Coyle (1985-1987)

G. B. Edwards (1984-1986)

Susan E. Riechert (1985-1987)

Research Notes

Submergent capture of Dolomedes triton (Araneae, Pisauridae) by

Anoplius depressipes (Hymenoptera, Pompilidae), Steven M. Roble 391

A Hawaiian wolf spider, Lycosa hawaiiensis Simon foraging in the top of a

Metrosideros polymorpha tree, Samuel M. Gon III 393

Adhesive technique for capture of burrow-dwelling spiders,

Marsha Reeves Conley 395

Functional webs built by adult male bowl and doily spiders,

Andrea J. Hirscheimer and Robert B. Suter 396

On the Chilean spiders of the family Palpimanidae (Arachnida, Araneae),

Normal I. Plat nick 399

Laboratory infection of Garypus californicus (Pseudoscorpionida,

Garypidae) with neoaplectanid and heterorhabditid nematodes
(Rhabditoidea), George O. Poinar, Jr., G. M. Thomas and

Vincent F Lee 400

An unusual species of Tyrannochthonius from Florida (Pseudoscorpionida,

Chthoniidae), David R. Malcolm and William B. Muchmore 403

Effects of method and time of preservation on volumetric mass estimates
of spiders (Araneae), Matthew H. Greenstone, Anne-Lise Hultsch and
Clyde E. Morgan 406

Others

Instructions to authors 409

Grants-in-Aid for Research 411

CONTENTS

THE JOURNAL OF ARACHNOLOGY

VOLUME 13 Feature Articles NUMBER 3

Prey capture and stinging behavior in the Emperor scorpion, Pandinus

imperator {¥.och) (Scorpiones, Scorpionidae), Gary S. Casper 277

Two new species of Ixamatus Simon from eastern Australia (Nemesiidae

Mygalomorphae, Araneae), Robert J. Raven 285

Ballooning mygalomorphs: Estimates of the masses of Sphodros and

Ummidia ballooners (Araneae: Atypidae, Ctenizidae), Frederick A. Coyle,

Matthew H. Greenstone, Anne- Lise Hultsch and Clyde E. Morgan 291

Laboratory infection of spiders and harvestmen (Arachnida: Araneae and
Opiliones) with Neoaplectana and Heterorhabditis nematodes

(Rhabditoidea), George O. Poinar, Jr., and G. M. Thomas 297

Uptake of leucine and water by Centruroides sculpturatus (Ewing) embryos

(Scorpiones, Buthidae), Eric C. Toolson 303

Opiliones from the Cape Horn Archipelago: New southern records for

harvestmen, James C Cokendolpher and Dolly Lanfranco L 311

Stage-biased overwintering survival of the filmy dome spider

(Araneae, Linyphiidae), John Martyniuk and David H. Wise 321

Evidence of insemination of multiple females by the male black widow
spider, Latrodectus mactans (Araneae, Theridiidae),

Robert G. Breene and Merrill H. Sweet 331

Spider ballooning: Development and evaluation of field trapping
methods (Araneae), Matthew H. Greenstone, Clyde E. Morgan

and Anne-Lise Hultsch 337

Leiobunum lineatum: A synonym of Leiobunum cretatum (Opiliones,

Gagrellidae), James C. Cokendolpher and William F Rapp 347

A new species of Heteronebo from Jamaica (Scorpiones, Diplocentridae),

Scott A. Stockwell 355

Habitat use by colonies of Philoponella republicana (Araneae, Uloboridae),

Deborah R. R. Smith 363

Reproductive tactics and female body size in the green lynx spider,

Peucetia viridans (Araneae, Oxyopidae), Don W Killebrew and

Neil B. Ford 375

Nests and neset-site selection of the crab spider Misumena vatia (Araneae,

Thomisidae) on milkweed, Douglass H. Morse 383

(continued on back inside cover)

Cover photograph, Phrynus marginemaculatus Koch, by Scott A. Stockwell
Printed by the Texas Tech Press, Lubbock, Texas
Posted at Lubbock, Texas, in December 1985

^ '

* )

f

■ :>'

I

I

’4

C/)

S3

in

</)

avyan libraries Smithsonian institution

INIAN INSTITUTION

NOlIDiliSNI

NVINOSHilWS S3 I HVaa IT

in

LIBRARIES SMITHSONIAN INS

if)

iiiiNS S3idVHan

LIBRAR

ES SMITHSONIAN
z
o

INSTITUTION

NOIlflillSNI

NVINOSHillAIS S3
2:

INIAN

NSTITUTION

NOlinillSNI

NVINOSHillAIS

» ^

avaan

LIBRAR

ES SMITHSONIAN
2: \

if)

z

m

if)

m

if)

m

if)

LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVlNOSHillNS SBiavaaiT LIBRA

iRARIES

;MITHSON

3 9088 01059 4224