Other formats

    TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Tuatara: Volume 20, Issue 2, March 1973

Prey and Prey Capture in the Tunnel Web Spider Porrhothele Antipodiana

page 57

Prey and Prey Capture in the Tunnel Web Spider Porrhothele Antipodiana


Porrhothele antipodiana (Mygalomorphae, Dipluridae), is the large, orange-backed tunnel web spider commonly found under logs and rocks in the lower North Island and also in the South Island. Along with Aparua and Hexathele, the other two New Zealand genera in the Dipluridae, Porrhothele spins a tubular web which often leads into a burrow in soil or wood. All three genera contain large spiders, and P. antipodiana is no exception on this characteristic, some individuals of this species attaining a body length of 30mm. Their size, coupled with an aggressive manner and a covering of long black hairs on the legs and abdomen makes these spiders an impressive and readily identifiable sight.

Research on the New Zealand mygalomorphs has been mainly along taxonomic lines, although some observations have been made on their biology. Todd (1945) observed some of the habits of P. antipodiana but on the whole her paper is more extensive in its consideration of the biology of the Ctenizidae. Forster (1967), and Forster and Wilton (1968) included some interesting observations on the natural history of the New Zealand mygalomorphs; but ecological and behavioural data were still sparse at this date. Accordingly, the present investigation was undertaken to determine what prey P. antipodiana took and what particular adaptations this spider has for prey capture. The paper comprises three sections: prey types taken, the use of the web in prey capture, and the behaviour of the spider.

Prey Types Taken

P. antipodiana carries its prey into the tunnel of the web so this is the portion that must be examined to find out what the spiders have been capturing. Identification of prey is often difficult because small portions or segments are all that are found in some cases. Hard parts such as beetle elytra and the thoracic parts of most insects do survive. Slater and millipede segments usually become detached and identification in these groups is particularly difficult.

The following is a list of prey species recovered from 50 webs in the Wellington area:

Coleoptera: Ochosternus zealandicus, Oemona hirta, Odontria sylvatica, Cecyropa lineifera, Cecyropa sp., Philoneis aucklandicum, Cilibe sp., Holcaspis sp., Xyloteles humeratus, Xyloteles griseus, Artystona rugiceps, Navomorpha sp., Coptomma variegatum.

page 58

Homoptera: Scolypopa australis.

Hymenoptera: Vespula germanica, Amblyopone australis, Bombus sp., Salius monachus, Salius fugax.

Isopoda: Porcellio scaber.

Mollusca: Helix aspersa.

Araneae: Miturga sp., Dolomedas sp. Ixeuticus sp.

Various Unidentified: F. Hepalidae at least two species, Diplopoda at least two species, Isopoda at least one species, Diptera at least one species, Araneae: some crab spider fragments.

This list is by no means exhaustive. P. antipodiana seems to eat almost anything that comes near enough to its web and so it probably preys on the majority of beetles, millipedes and slaters that happen to be in the same locality as the spider. Of the hymenoptera, the Salius species found were probably the result of unsuccessful forages into the tunnel after the spider. V. germanica investigates most objects, so it is likely that it often gets caught in P. antipodiana webs; and from my own observations, the larger spiders have little trouble in dealing with this wasp once it has entangled itself in their web. Bumble bees are picked out of the web and carried back to the tunnel without difficulty too. In the case of V. germanica and the bumble bee, the spider invariably grasps them from behind the thorax thus avoiding the sting. The presence of land snails in the webs may be a surprise on the grounds that the spider would find difficulty in gaining access to the soft parts of the snail.

The New Zealand Ctenizidae seem to take similar prey to P. antipodiana. Todd (1945), in an examination of burrows of South Island trapdoor spiders reported finding beetles, a weevil, large numbers of Odontria remains, caterpillar remains and a small fly.

Of the fifty webs examined in the present investigation, the different prey groups comprised the percentages shown in table one.
Table 1. Percentages of Different prey Types Taken by Porrhothele antipodiana
Prey GroupNo. FoundPercent of total Prey
The proportions of the different prey taken does vary from area to area. For example, at Paremata, there are large numbers page 59 of millipedes and slaters found in webs. The same is true for Waikanae, Waitarere beach and Palmerston North. On the hill slopes of Wellington the prey are largely beetles and specially click beetles (Elateridae) and grass grub beetles (Scarabidae).
Fig. 1 Some variations in the web of P. antipodiana A. from under a log, Waikanae. B. from under a rock, Wellington. C. from a crevice in a bank, Paremata. D. from under a log, Wellington. Fine broken line represents area of web visible at edge of rock, log or out of crevice. All views are from the dorsal aspect.

Fig. 1 Some variations in the web of P. antipodiana
A. from under a log, Waikanae.
B. from under a rock, Wellington.
C. from a crevice in a bank, Paremata.
D. from under a log, Wellington.
Fine broken line represents area of web visible at edge of rock, log or out of crevice. All views are from the dorsal aspect.

The Web

The web of P. antipodiana, as in other spiders is an essential part of prey capture and has particular adaptations for this specific purpose. The basic pattern consists of a woven silk tunnel in which the spider shelters. In addition there may be a sheet of silk extending beyond the opening of the tunnel. The extent of the sheet is very variable and seems to be related to the local environment of the spiders. For example, webs found under logs or rocks have little if any development of the sheet; whereas webs situated on banks or in trees usually have an extensive sheet development. Thirty webs on a cliff face at Paremata showed a range of from 10cm to 40cm in the width of the sheet, measured at the widest point. Measurements taken of thirty webs from under rocks and logs on the hills around Ngaio (Wellington) showed that the sheet could be absent or up to 10cm in width. Fig. 1. gives some web shapes found in a variety page 60 of localities. The length of the tunnel also varies, webs from under stones having a tunnel ranging from 10cm to 30cm; while tunnels in crevices of banks are usually shorter, most being less than 15cm in length.

Of a hundred webs surveyed at Paremata, the following distribution of sites was found: in crevices of the bank 60; up trees 14; under logs or stones 26. An investigation of the hills around Wellington shows quite a different distribution of sites. Here the only sites available seem to be those under rocks and logs so almost 100% of the population is found in these sites.

Studies on mygalomorphs such as the Ctenizidae and the Atipidae have indicated that these spiders remain in the one burrow for the duration of their life (Forster and Wilton 1968, Savory 1926). This is not the case with P. antipodiana, for movement from one site to another seems to be common. In an investigation at Paremata lasting from March to August of 1971, 30 webs on an area of bank were pegged for identification and weekly inspections were made with a torch at night to determine if spiders were still in residence. This disturbance of the spider was minimal, causing them to retreat into their tunnels for 10-15 minutes at the most. At the end of the investigation, fifteen of the sites were no longer occupied, while ten new sites had been taken over by spiders. While this study did not identify which spiders moved where, it at least points out the mobility of the population.

Mobility is also evident in those populations living under logs and stones. Invariably there will be a number of web extensions to be seen when a log is lifted. Many are old and disused and the movements of the spider under the log can be traced by following the older tunnels to the newest, occupied part.

The Behaviour of the Spider

P. antipodiana, in its mode of dealing with prey, shows some adaptations not so apparent in spiders such as the orb web spiders (Epeiridae) and the common house spiders (Dictynidae). While it is common to see these spiders in their webs during daylight, it has not been my experience to see a single P. antipodiana either mending its web or waiting for prey while the light is still strong. P. antipodiana waits until dark and then comes out to the entrance of the tunnel. Occasionally, individuals may be found further out on the sheet web, presumably being engaged in mending the web. This very obvious difference between day and night behaviour of P. antipodiana is shown in table 2. which was based on the dropping of a slater into each of twelve webs in daylight and then repeating the procedure at night.

In the daylight part of the experiment, four spiders failed to appear at all so the mean figures given in the table are based on eight responses — not twelve as in the night responses. page 61
Table 2. Comparisons of P. antibodiana responses to a slater during day and night. (figures in seconds).
Mean time taken to reach prey29.02.0
Mean time taken to deal with prey1.58.0
Mean time taken to return to tunnel1.52.5

Interpreting the table, the most obvious feature is the time taken to reach the prey. The night spiders reached the slater in minimum time, running at high speed the moment it landed in the web. The daytime spiders — if they appeared at all, were very cautious and waited at the entrance of the tunnel for some time before making a rapid dash to the slater. During the day the spiders spent far less time dealing with the slater, whereas at night the spiders handled the slater in a leisurely manner, taking much longer to deal with it.

The speed of the spider is probably significant in prey capture so timings were carried out to determine the spider's maximum speed. Webs with long sheets were selected, and a slater was dropped in at a point approximately 30cm from the tunnel entrance, and the time taken for the spider to reach the slater was estimated using a stop watch. Some spiders reached the slater in 1.5 seconds which meant they were running at a speed of at least 0.2metres-sec. For comparison, slaters and millipedes were timed over flat level ground to get an idea of the maximum speed they might be capable of running at. The fastest slaters ran at a speed of 0.1 metre-sec.; while 0.02metre sec. was the fastest speed recorded for the millipedes. These speeds are a generous estimate of the speed that slaters and milipedes could move at in a P. antipodiana web, but are the best that can be offered because of the difficulties experienced in getting prey species to move in a straight line in a web.

At night the spider has to contend with the stimuli provided by leaves, small pieces of wood, etc., dropping into the web. Initially, these are treated as prey — the spider approaching and assuming a position with the palps touching the object. This affords at least a tactile investigation of the object and possibly also a gustatory or an olfactory investigation. Blumenthal (1935) considered that the palps had an olfactory function, but this has been disputed. Inanimate objects are left alone after a few seconds investigation. Sometimes a prey animal will stop moving as the spider approaches and if this happens, the spider ‘tests’ it with the palps and within a short time it usually stabs the animal with its chelicerae. The stabbing of the prey is done more than once; this probably ensures deep penetration of the venom and would help in dealing with hard-bodied forms such as the beetles where the spider may have to manoeuvre the fangs until they slide into a joint or a soft spot. The chelicerae, along with the palps serve to hold the prey for carrying it back to the tunnel where it will be eaten.

page 62
Fig. 2: Some prey-seeking manoeuvres shown by P. antipodiana. In A-D, the fine, broken line and arrows represent the path followed by the spider to and from the entrance to the tunnel, which is shown at the top of each diagram. The object in the web represents the slater used in the experiment. All manoeuvres were observed on moonlit nights.

Fig. 2: Some prey-seeking manoeuvres shown by P. antipodiana. In A-D, the fine, broken line and arrows represent the path followed by the spider to and from the entrance to the tunnel, which is shown at the top of each diagram. The object in the web represents the slater used in the experiment. All manoeuvres were observed on moonlit nights.

At night the spider does not always find the prey on its first attempt. It is common for the spider to overshoot the prey or to misjudge the direction of the prey. Fig. 2. contains some prey-seeking movements of P. antipodiana that were observed during this study.


The ability of P. antipodiana to accept a wide variety of prey types must be of significance in its success as a predator. The presence of millipedes as prey may surprise some biologists, on account of the unpalatable secretions of these animls, which apparently deter some predators (Forster, 1971). Petrunkevitch (1952) has recorded that the tarantula Cyrtopholis portoricae feeds on insects and millipedes, so it may be that many of the mygalomorphs can exploit the millipedes as a source of food. Land snails in the tunnels of P. antipodiana present a problem of interpretation; it may be that the spider carries them there or it may be a similar case to that of worms in the burrow of the purse web spider Atypus. Savory (1926) recorded that partly devoured worms were found in the tunnel of Atypus; however Bristowe (1958) discounts records of Atypus feeding on worms. He observed that worms taken into the page 63 web were always partly eaten only. The suggestion is then that the snails in P. antipodiana tunnels, like worms in Atypus tunnels are intruders that may or may not be killed by the spider.

The web characteristics must rate highly as a factor in P. antipodiana's success too. The extensive sheet webs of those living on exposed sites on banks or trees contrasts strongly with the often complete lack of development of the sheet in webs found under logs or stones. The reason for this is probably to be found in the distribution of prey species; for instance, under logs the spider is right amongst its prey — the beetles, slaters, etc., that shelter and feed there too. Whereas on the exposed sites the spider must rely on those animals that are moving over the surface of the bank for its food. There is no concentration of prey here on the bank as there is under the log. Consequently the spider on the exposed site requires a more extensive web to cover a wider area and so increase its access to prey. A similar situation has been described by Main (1957) in a study on the trapdoor spiders of South Western Australia. She found that those forms inhabiting damper, richer litter having an adequate prey supply, seized prey that were within reach of the burrow entrance. Those forms living in dry environments lay out lines of silk to act as indicators of when prey animals are passing by. These spiders would rush out several inches from their burrows to grab prey — an obvious advantage where prey are scarce.

The speed at which P. antipodiana is capable of running must be important too, for in cases of an insect wandering across the edge of the web, it is probably the speed at which the spider chases the insect that will determine whether or not a capture is made. The majority of prey species do seem to be able to cross a P. antipodiana web without becoming entangled; and it is here that comparisons can be made between P. antipodiana and the aran-eomorphs such as the orb-web spider, Aranea. The sticky webs of the latter, by holding prey, make prey capture so much easier; while P. antipodiana has to have the extra speed and hunting ability to ensure that captures are made.

Migratory behaviour probably gives a more efficient utilisation of prey resources, in that the spider goes to the prey and not vice-versa as is the case with the trapdoor spiders. Like the building of extensive sheet webs, mobility functions to cover a wider area of possible prey movement.

The behaviour of the spider in daylight must have considerable survival value. That few individuals give an immediate response to the presence of prey in the web probably reduces the susceptibility of P. antipodiana to predation by the Pompilid wasps Salius monachus and S. fugax (as well as other possible predators).

In conclusion it is recognised that a paper on prey capture could well include reference to other features such as the toxicity page 64 of P. antipodiana's venom; the postures assumed by the spider and its sensitivity to stimuli from the web. It is hoped that these features will form the substance of another paper at a later date.


P. antipodiana, commonly known as the tunnel web spider, feeds on a wide variety of prey types. The list includes the Isopoda, Hymenoptera, Coleoptera, Lepidoptera, Homoptera, Diplopoda, and other spiders as well. The web of P. antipodiana figures strongly in prey capture and shows considerable variation in differing environments. Mobility from site to site is a feature of this spider's behaviour. A strong photo-negativism is evident in its behaviour; which when coupled with the speed of running results in minimum exposure to predators. The manner in which P. antipodiana subdues its prey is described.


My thanks are due to Dr R. R. Forster, Otago Museum, for confirming the identification of P. antipodiana; and to Miss Logan Hudson. Dominion Museum, who identified the prey remains, read the manuscript and offered valuable criticisms of it.


Blumenthal, H. 1935: Untersuchungen uber das ‘Tarsalorgan’ der Spinnen. Zeits. Morph. Okol. Tiere, 29 (5) 667-719.

Bristowe, W. S. 1958: The Spider's World. London: Collins.

Cloudsly-Thompson, J. L. 1968: Spiders Scorpions Centipedes & Mites, Oxford: Pergamon Press.

Forster, R. R. 1967: The Spiders Of New Zealand, Part 1 Otago Mus. Bull. 1.

——, 1971: Personal communication.

——, and Wilton, C. L. 1968: The Spiders Of New Zealand, Part 2. Otago Mus. Bull. 2.

McKeown, K. C. 1963: Australian Spiders (rev. N. L. Roberts) Sydney: Sirius Books.

Main, B. Y. 1957: The Biology of Aganippine Trapdoor Spiders (Mygalo-morphae: Ctenizidae) Austral. Journ. Zool., 5 (4).

Myers, J. G. 1927: Ethological Notes on Some New Zealand Spiders, N. Zeal. Journ. Sci. Tech. 9: 129-136.

Petrunkevitch, A. 1952: The Spider and The Wasp. Sci. Am. 187 (2) 20.23.

Savory, T. H. 1926: British Spiders (Their haunts and habits). Oxford: Clarendon Press.

——, 1952: The Spider's Web. London: Warne.

Todd, V. 1945: Systematic and Biological Account of the New Zealand Mygalomorphae (Arachnida). Trans. Roy. Soc. N. Zeal. 74 (4) 375-407.