Table Of Content2006. The Journal of Arachnology 34:422-426
THE PREY OF THE LYNX SPIDER OXYOPES GLOBIFER
(ARANEAE, OXYOPIDAE) ASSOCIATED WITH A SEMIDESERT
DWARF SHRUB IN AZERBAIJAN
Elchin Fizuli oglu Huseynov': Institute of Zoology, Azerbaijan Academy of
Sciences, Block 504, Passage 1 128, Baku 370073, Azerbaijan.
E-mail: [email protected]
ABSTRACT. The prey ofthe lynx spider, OxyopesglobiferSimon 1876, occurring on Salsola nodulosa
(Moq.) plants, was analyzed. In common with other cursorial spiders, the percentage offeeding specimens
in the population ofO. globiferwas low (5.5%). Males were observed feeding significantly less frequently
than females and immatures ofboth sexes. After oviposition, however, the records ofprey capture in egg-
guarding females also declined considerably. Oxyopes globifer is a polyphagous predator feeding on a
wide range of arthropods: insects of the orders Hymenoptera, Diptera, Lepidoptera and Homoptera, as
well as on several spider species. The primary food was worker ants, which accounted for 62.7% oftotal
prey. No other prey taxon was present in any considerable percentage. O. globifer captured prey ranging
between 22.7 and 243.8% (mean 88.8%) of its own body length. Most frequently taken were medium-
sized arthropods varying from 50-110% of spiders’ body lengths.
Keywords: diet, myrmecophagy, prey length, semidesert, Azerbaijan
Most lynx spiders (Oxyopidae Thorell 1870) are mating lasts throughout June. At the beginning of
typical cursorial hunters, which possess relatively July males disappear, while females starttoproduce
keen eyesight and do not use silk for prey capture egg sacs, which they attach to the branches of
(Kovoor & Munoz-Cuevas 1997). Instead, they ac- shrubs. Females attend their cocoons until the
tively pursue their prey and seize it with a short young emerge. Breeding season runs until mid Sep-
lunge {Oxyopes Latreille 1804) or attack it from tember and individual females usually produce sev-
ambush {Peucetia Thorell 1869) (Rovner 1980). eral egg sacs during this period (Huseynov unpubl.
In common with other cursorial spider groups, data).
the literature on prey of Oxyopidae is scarce. Ex- The investigation was carried out on the Apshe-
ceptions are the striped lynx spider, Oxyopes salti- ron Peninsula in Azerbaijan. The study site was lo-
cus Hentz 1845 (Lockley & Young 1987; Nyffeler cated near Yeni-Surakhany village (40°42'N,
et al. 1987b, 1992; Agnew & Smith 1989; Bardwell 49°95'E). This was anopen areaofephemeralsemi-
& Averill 1997), and the green lynx spider, Peu- desert covered with dwarfshrubs Salsola nodulosa,
cetia viridans (Hentz 1832) (Turner 1979; Randall Alhagipseudoalhagi (M.B.) and short grasses, pre-
1982; Nyffeler et al. 1987a, 1992; Quicke 1988), dominantly Calendulapersica C.A.M.,Seneciover-
both being from North America. Less extensive nalis Willd. & Kan, Medicago denticidata Willd.,
quantitative data are available on the natural prey Carduus arabicus Jacq., Erodium cicutarium (L.),
of Oxyopes apollo Brady 1964 from the USA (Ag- Pterotheca marschalUana (Rchb.), Poa bulbosa L.,
new & Smith 1989), and O. licenti Schenkel 1953 Anisanthea rubens (L.) andAegilops biuncialisVis.
and O. sertatus L. Koch 1877 from Japan (Furuta O. globiferwas abundant only on Salsola shrubs,
1977). therefore observations wereconcentratedexclusive-
The present paper is the first study ofthe prey of ly on this plant. Sixteen surveys were conducted
the Mediterranean lynx spider, OxyopesglobiferSi- from 20 May-9 September 1998 and took 27.5 h
mon 1876, which occurs in North Africa, Southern in total. All observations were made in daylight be-
Europe, Near East and Central Asia (Levy 1999). tween 12:00 and 18:00 h. During surveys, Salsola
In Azerbaijan, individuals of this species are fre- shrubs were thoroughly searched for spiders, and
quently found on dwarf shrubs Salsola nodulosa each individual O. globifer found was captured in
(Moq.),Artemisiafragrans (Boiss.), andNoaeamu- a transparent glass vial. In the vial the spider’s
cronata (Forssk.). O. globiferhas an annual life cy- mouthparts were inspected with a loupe of 4 X
cle. Adult specimens appear at the end ofMay and magnification to prevent small prey being over-
looked. Specimens with prey in their chelicerae
' Former surname: Guseinov were placed in separate vials containing 75% ethyl
422
HUSEYNOV—PREY OF OXYOPES GLOBIFER 423
alcohol and brought backto the laboratory forbody
length measurement and prey identification.
All spiders observed were classified into several
groups according to their age, sex and presence or
absence of egg sacs near females. When the inves-
tigation started, the vast majority ofspiders were in
the sub-adult stage. Thus, the following fourgroups
were delimited: (1) sub-adult males, which had
swelled, but not differentiated pedipalp tips; (2)
adult males, with distinctly developed palpal scler-
ites; (3) pre-reproductive females, including all spi-
ders without swollen pedipalp tips and without egg
sacs; (4) females guarding their egg sacs. During
each survey the numberofspiders with and without
prey was counted separately within each ofthe four
groups. Because the study area was large (ca. 500 Sizecategories
X 500 m) and successive surveys were conducted —
Figure 1. Distribution of prey in different size
in different parts ofthis area, it is highly likely that categories, which are body lengths of prey ex-
most of the O. globifer observed were different pressed as percentage of the body lengths of their
specimens. The log-likelihood ratio test (G, statis-
captors.
tic) was used forcomparison ofpercentage offeed-
ing specimens among different groups of spiders.
Voucher specimens of O. globifer and their prey specific. The latter was the male captured by the
items were deposited at the Institute of Zoology, female; i.e. sexual cannibalismprobably tookplace.
Azerbaijan Academy of Sciences. The remaining prey was a leafhopper (Cicadelli-
In total, 947 specimens of O. globifer were ob- dae).
served, 52 of which (5.5%) had prey in their che- Fifty prey items were measured. Their length
licerae. Among them 93 sub-adult males (6 with varied from 1.5-12.8 mm (mean ± SD: 4.9 ± 2.6
prey —6.5%), 153 adultmales (4withprey —2.6%), mm) and constituted from 22.7-243.8% (88.8 ±
431 pre-reproductivefemales (36 with prey —8.4%) 40.9%) ofthe length oftheircaptors (Fig. 1), which
and 270 females with egg sacs (6 withprey —2.2%) ranged from 4.7-9.0 mm (6.7 ±1.0 mm). Accord-
were recorded. The percentage of feeding speci- ing to their body length, all prey of O. globifercan
mens in pre-reproductive females was significantly be divided into three groups. Small prey, not ex-
higher than those in adult males (G, = 6.98, P < ceeding 3 mm in length, made up 14.9% of total
0.01) and egg-guarding females (Gj = 12.7, P < prey measured. This group included the salticid spi-
0.001). In contrast, there was no statistically signif- der, the dolichopodid fly, the leafhopper, Cardi-
icant difference between pre-reproductive females ocondyla and Plagiolepis ants, none of which ex-
and sub-adult males in this respect (Gj = 0.39, P ceededhalfthe sizeoftheircaptors. Thelargestpart
>
0.5). of the diet of O. globifer (62.5%) was medium-
One spider seen with prey escaped, so 51 prey sized arthropods (3-7 mm), consisting of all spec-
items were collected for dietary analysis. These imens of Cataglyphis aenescens, the halictid bee,
prey items were distributed among five orders of the braconid wasp, two spiders {Micaria rossica,
arthropods: four from the class Insecta (Hymenop- Oxyopes globifer) and one lepidopteran larva. Rel-
tera 68.6% of total prey, Diptera 11.8%, Lepidop- ative length of these prey items varied from 50-
tera 11.8%, Homoptera 2.0%), and one from the 110%. The third group consistedofarthropods larg-
class Arachnida (Araneae 5.9%). The dominant er than 7 mm, being from 110-250% of their
food component was worker ants, which accounted captors’ lengths. This prey amounted to 22.9% of
for 62.7% oftotal prey. Cataglyphis aenescens Ny- the prey as a whole and included a moth, a me-
lander contributed the bulk of ants (28 specimens), gachilid bee, bombyliid flies and most caterpillars.
followed by two Cardiocondyla sp. and two Pla- There was no positive relationship between spi-
giolepis sp. Otherhymenopteransincludedtwobees der length and prey length (Correlation analysis, r
Coelioxys argentea Lepeletier (Megachilidae), Hal- = 0.1588, P = 0.271), and no difference between
ictus sp. (Halictidae) and one parasitic wasp (Bra- spidersexes in theprey length taken (ANOVA,F445
conidae). Diptera were represented by five bom- - 0.3020, P = 0.875).
byliids {Villa sp.) and one dolichopodid fly, and The percentage offeeding specimens in the pop-
Lepidoptera comprised one unidentified moth and ulation ofO. globifer studied was low (5.5%), as is
five unidentified caterpillars. Among the spiders usual with cursorial spiders (Nyffeler & Breene
captured were one gnaphosid {Micaria rossicaTho- 1990) and lynx spiders in particular (Nyffeler et al.
rell 1875), one unidentified salticid and one con- 1987a, b, 1992). Moreover, the percentage offeed-
424 THE JOURNAL OF ARACHNOLOGY
—
Table L Length of prey of different oxyopid species.
Length of spiders Length of prey Length of prey
mm mm %
Spider species n range mean range mean range mean Source
— —
Oxyopes salticus 64 2.6-8.0 0.6-5.6 2.61 10.0-110.0 Nyffeler et al. 1987b
Oxyopes salticus 63 L9-8.0 4.24 0.5-5.8 2.41 8.0-129.0 5—6.0 Nyffeler et al. 1992
Peucetia viridans 25 8.2-12.7 10.96 1.6-16.5 5.90 14.0-130.0 Nyffeler et al. 1987a
Peucetia viridans 31 4.5-16.5 10.08 1.3-13.6 7.04 26.0-136.0 68.0 Nyffeler et al. 1992
Oxyopes globifer 50 4.7-9.0 6.70 1.5-12.8 5.90 22.7-243.8 88.8 Present study
ing specimens among adult males was significantly may be unusual, since these insects, possessing ef-
lower than among pre-reproductive females. Labo= fective defensive equipment, are not palatable to
ratory investigations on feeding in other oxyopids m.ost cursorial spiders (Nentwig 1986). However,
have also revealed that males feed less often than myrmecophagy is apparently a common phenome-
females (Liegren et al. 1968; Furuta 1977). A sim- non within the family Oxyopidae. Workerants were
ilar tendency has been reported injumping spiders found among the prey of all lynx spiders, where
(Salticidae) by Jackson (1977) and Givens (1978). data on natural prey is available; O. salticus (Nyf-
Both investigators attributed this fact to the specific feler et al. 1987b), O. apollo (Agnew & Smith
life style of salticid males, which emphasizes mat- 1989), O. scalaris (Mclver 1989), O. licenti and O.
ing and only opportunistically involves feeding. In sertatus (Furuta 1977), F. viridans (Nyffeler et al.
contrast, females, which need a high intake offood 1992), Does O. globiferprefer ants to other prey or
for yolk production, spend much of their time is their predominance in its diet due to the abun-
searching or waiting for prey. It seems reasonable dance ofthese insects on Salsola nodulosal Exper-
to apply this speculation to lynx spiders too. In this imental laboratory investigations arerequiredto an-
context it is worthwhile to note that the percentage swer this question. However, extensive field
of feeding specimens among sub-adult males of O. observations on the prey of O. salticus in cotton
globifer was comparable to that among pre-repro- agroecosystems in the USA have shown the pro-
ductive females. Apparently the life style of im- portion ofants in the diet ofthis species to behigh-
mature males does not differ significantly from that ly variable. While worker ants were the dominant
of females and changes drastically only after final food component of spiders occurring in Houston
molt. County, Texas (Nyffeler et al. 1987b), their per-
After oviposition, lynx spider females normally centage was quite insignificant in the diet ofO. sal-
attend their cocoons (e.g., Cutler et al. 1977). Such ticus inhabiting cotton fields in Burleson County,
behaviorfacilitates protection ofeggs and increases Texas (Nyffeler et al. 1992), and they were entirely
survival of the offspring (Fink 1986), but it is dis- missing among prey of spiders observed in Sun-
advantageous for parent spiders themselves, since flower County, Mississippi (Lockley & Young
this restricts their hunting activity. Although egg- 1987). These data suggest that the proportion of
guarding oxyopid females have been observed eat- ants in the diets of Oxyopes species may depend
ing prey (e.g., Willey & Adler 1989), this is prob- significantly on their abundance in the spider’s hab-
ably not frequently the case, because spiders have itat.
no opportunity to choose optimal foraging sites The range and mean value of relative length of
when they brood their egg sacs. One would there- prey of O. globiferwere greater than those ofother
fore expect a substantial reduction in feeding fre- lynx spiders studied (Table 1). However, this dif-
quency in O. globiferfemalesduringtheegg-guard- ference is not pronounced. There was a marginally
ing period. In fact, the percentage of feeding significant positive relationship between spider
specimens among egg-guarding females was signif- length and prey length across species (Correlation
icantly lowerthan among solitary ones. In wolfspi- analysis, r = 0.853, P = 0.066). Larger species (F.
ders the feeding frequency of maternal females is viridans) tended to catch larger prey, while smaller
also significantly lower than that of non-maternal species (O. salticus) tended to catch smaller prey
females (Nyffeler 2000). than did O. globifer.
This investigation has shown that O. globifer is Laboratory tests on prey-size preference of spi-
a polyphagic predator feeding on a wide range of ders have shown that most cursorial spiders do not
arthropods. Lynx spiders are generally known to capture prey larger than 150% oftheirbody length,
have broad diets (see Nyffeler 1999). The heavy with a most preferred range of 50-80% (Nentwig
prevalence of worker ants in the diet of O. globifer & Wissel 1986). Oxyopes globifer caught several
HUSEYNOV—PREY OF OXYOPES GLOBIFER 425
prey items exceeding 150% ofits length. However, Levy, G. 1999. The lynx and nursery-web spider
the proportion of such prey in its diet was low. families in Israel (Araneae: Oxyopidae and Pi-
Moreover, only one prey item was more than two- sauridae). Zoosystema 21:29-64.
fold larger than its captor (243.8%). This is prob- Liegren, P.D., R.L. Ridgway & S.L. Jones. 1968.
ably the largest prey recorded for oxyopid spiders Consumption by several common arthropod
(in terms ofprey/predator size ratio). All otherprey predators of eggs and larvae of two Heliothis
have not exceeded 170% inrelative length, and me- species that attack cotton. Annals of the Ento-
dium-sizedarthropodsprevailedamongthem. Thus, mological Society of America 61:613-618.
one can conclude that the prey-size spectrum of O. Lockley, TC. & O.P. Young. 1987. Prey of the
globifer is, in general, comparable to the common striped lynx spider Oxyopes salticus (Araneae,
pattern found among cursorial spiders, although it Oxyopidae), on cotton in the delta area of Mis-
is somewhat biased to large prey. sissippi. Journal of Arachnology 14:395-397.
ACKNOWLEDGMENTS Mclver, J.D. 1989. Protective resemblance in a
community of lupine arthropods. National Geo-
I thank Dr. Halid Aliev (Institute of Zoology of graphic Research 5:191-204.
AS ofAzerbaijan) foridentifyingthehymenopteran Nentwig, W. 1986. Non-webbuilding spiders: prey
insects. I am also grateful to Mr. Graham Wishart specialists orgeneralists?Oecologia69:571-576.
(Associate ofthe Australian Museum, Sydney), Dr. Nentwig, W. & C. Wissel. 1986. A comparison of
PaulaCushing (DenverMuseum) and Dr. DanMott prey lengths among spiders. Oecologia 68:595-
A & M
(Texas International University) for correc- 600.
tion ofthe Englishtext andto Dr. SoerenToft(Uni- Nyffeler, M. 1999. Prey selection of spiders in the
versity of Aarhus) and two anonymous reviewers field. Journal of Arachnology 27:317-324.
for their comments on the manuscript. Special Nyffeler, M. 2000. Do adult female lycosids feed
thanks are extended to Dr. Jordi Moya-Larano during the period of maternal care? Bulletin of
(Universitat Autonoma de Barcelona) for his statis- the British Arachnological Society 11:388-390.
tical aid. Nyffeler, M, & R.G. Breene. 1990, Evidenceoflow
LITERATURE CITED daily foodconsumptionby wolfspiders in mead-
owland and comparison with other cursorial
Agnew, C.W. & J.W, Smith. 1989. Ecology of spi- hunters. Journal ofAppliedEntomology 110:73-
ders (Araneae) in a peanut agroecosystem. En- 81.
vironmental Entomology 18:30-42, Nyffeler, M., D.A. Dean & W.L. Sterling. 1987a.
Bardwell, CJ. & A.L. AverilL 1997. Spiders and Predation by green lynx spider, Peucetiaviridans
theirprey in Massachusetts cranberrybogs. Jour- (Araneae: Oxyopidae), inhabiting cotton and
nal ofArachnology 25:31-41. woolly croton plants in east Texas. Environmen-
Cutler, B., D.T Jennings & M.J. Moody. 1977. Bi- tal Entomology 16:355-359.
ology and habitats of the lynx spider Oxyopes Nyffeler, M., D.A, Dean & W.L. Sterling, 1987b.
scalaris Hentz (Araneae: Oxyopidae). Entomo- Evaluation of the importance of the striped lynx
logical News 88:87-97. spider, Oxyopes salticus (Araneae: Oxyopidae),
Fink, L.S. 1986. Costs and benefits ofmaternal be-
as apredator in Texas cotton. Environmental En-
haviour in the green lynx spider (Oxyopidae,
tomology 16:1114-1123.
Peucetia viridans). Animal Behaviour 34:1051- Nyffeler, M., D.A. Dean & W.L. Sterling. 1992. Di-
1060.
Furuta, K. 1977. Evaluation of spiders, Oxyopes ets, feeding specialization, and predatory role of
two lynx spiders, Oxyopes salticus and Peucetia
sertatus and O. badius (Oxyopidae) as a mor-
viridans (Araneae: Oxyopidae), in aTexascotton
tality factor of gypsy moth, Lymantria dispar
agroecosystem. Environmental Entomology 21:
(Lepidoptera: Lymantriidae) and pine moth,
1457_1465.
Dendrolimus spectabilis (Lepidoptera: Lasio-
campidae). Applied Entomology and Zoology Quicke, D.L.J. 1988. Notes on prey taken by some
12:313-324. North American spiders. British Journal of En-
Givens, R.P. 1978. Dimorphic foraging strategies of tomology and Natural History 1:107-110.
a salticid spider (Phidippus audax). Ecology 59: Randall, J.B. 1982. Prey records of the green lynx
309-321. spider, Peucetia viridans (Hentz) (Araneae, Ox-
Jackson, R.R. 1977. Prey ofthejumping spiderPhi- yopidae). Journal of Arachnology 10:19-22.
dippusjohnsoni (Araneae, Salticidae). Journal of Rovner, J.S. 1980. Adaptations for prey capture in
Arachnology 5:145-149. oxyopid spiders: phylogenetic implications. Pp.
Kovoor, J. & A. Munoz-Cuevas. 1997. Comparative 233-237. In Verhandlungen 8th Internationaler
structure of the visual system of lynx spiders Arachnologen-Kongress, Wien (J. Gruber, ed.).
(Oxyopidae) and its relation to habitat and be- H. Egermann, Vienna.
haviour. Zoologischer Anzeiger 235:133-145. Turner, M. 1979,, Diet and feeding phenology ofthe
426 THE JOURNAL OF ARACHNOLOGY
green lynx spider, Peucetia viridans (Araneae: maternal care. Journal of Arachnology 17:275-
Oxyopidae). Journal ofArachnology 7:149-154. 284.
Willey, M.B. & RH. Adler. 1989. Biology of Peu-
cetia viridans (Araneae, Oxyopidae) in South Manuscript received 23 March 2004, revised 15
Carolina, with special reference to predation and October 2004.
