Notes on Araneus diadematus

Araneus diadematus. Pregnant female Copyright: Chris Bowes
filter by region:

Add a Species Note

(you need to be logged on to do this)

Intentional Pollen Feeding in the Spider Araneus diadematus Clerck, 1757 by Claudia Ludy

From The Newsletter No. 101 November 2004

Spiders are almost invariably described as carnivorous. However, there are a few reports that spiders also ingest plant-derived nutrients such as nectar (Taylor & Foster, 1996; Jackson et al., 2001) and pollen (Smith & Mommsen, 1984). While hunting spiders probably actively forage pollen in times of prey scarcity (Vogelei & Greissl, 1989), orb-weaving spiders are believed to consume pollen accidentally by ‘recycling’ their webs (Smith & Mommsen, 1984). This unintentional pollen feeding of orb-web spiders was shown by Smith & Mommsen (1984) indirectly by demonstrating that spiders with a supply of pollen survive longer than starved spiders. Until now, there has been no direct proof of deliberate pollen uptake in orb-weaving spiders.

In this study, intentional pollen feeding in juvenile and adult garden spiders (Araneus diadematus Clerck, 1757) was directly proven by visual observation, and verified by a molecular biological method.

Pollen consumption by an adult garden spider was observed in the field. The spider was maintained in the laboratory under standardised conditions (temperature 20°°C, 10 h/10 h light/dark regime) for several weeks. It built its orb-web in a wooden frame (30 x 30 cm), and was exposed at a height of 80 cm for seven hours on a field margin covered with flowering plants. During field exposure, a pollen-carrying wild bee was caught in the spider web. The spider wrapped up the bee with silk, but the bee was eventually able to escape, leaving behind the spider’s silk wrapping including a mass of pollen. Later on, the spider took the silkwrapped pollen to the hub, and after a few minutes, fluid appeared on the cluster, and the pollen mass changed colour.

Pollen-feeding in a juvenile A. diadematus (approximately 8 weeks old with a length of about 3 mm) was observed in the laboratory. The spider was kept in a wooden frame (10 x 10 cm), where it built its orb-web. The spider was fed with fruit flies (Drosophila sp.) and was kept under the conditions described above for several weeks. Maize pollen (Zea mays) was placed onto the orb-web with a small brush. The spider immediately reacted by pulling on the radial threads. Subsequently, it moved towards a pollen package, touched the pollen package with its pedipalps, and then carried the pollen to the hub. Following this the pollen package was held between the chelicerae, and became darker in colour and coated with liquid.

The following laboratory experiment was carried out to prove pollen consumption directly: Juvenile garden spiders (A. diadematus) were kept under the above laboratory conditions, and were fed fruit flies.The webs of 11 spiders were dusted with pollen of conventional maize (control), and to another 20 spider webs pollen of genetically modified Bt-maize was applied (variety ‘Navares’, event Bt176). The Bt-maize produces a protein of the entomopathogenic bacterium Bacillus thuringiensis (Cry1A(b) protein), which can be detected by an enzyme-linked immuno-sorbent assay (ELISA). After the spiders had recycled their webs, the spiders were collected and frozen at –18°C to prevent possible degradation of the Cry1A(b) protein, and stored for six months. Subsequently, the spiders were defrosted at 5°C, and washed with water in order to remove any pollen possibly adhering to the spider (additionally, spiders were checked for pollen under a binocular microscope). Then, the gastrointestinal system of the prosoma was dissected and picked in cyclohexylaminopropane sulfonic acid buffer (CAPS buffer: 50 mM, pH 10.5). The dissected tissue was analysed for Cry1A(b) content with a commercial ELISA kit (EnviroLogix OuantiPlate™ kit for Cry1Ab/Cry1Ac from Adgen®). In 13 of the 20 spiders (65%), whose webs were dusted with Bt-pollen, but in none of the 11 spiders of the control group, the Cry1A(b) toxin was detected.

The behavioural observations showed that both juvenile and adult garden spiders can ingest pollen directly. Possibly, spiders recognise pollen as food by touching the pollen with taste receptors on the pedipalps. Fluid on the pollen package and concomitant colour change of pollen was likely due to the application of digestive enzymes from the spider’s midgut, indicating extraintestinal digestion. Further, the results of the ELISA with spiders whose webs were dusted with Bt-maize pollen, prove an uptake of pollen by the spider. The detection of the Cry1A(b) protein in the spider’s gastrointestinal system proved to be an effective method to verify pollen consumption. Thus, feeding Bt-maize pollen and a subsequent analysis by ELISA is a possible method to detect pollen feeding in both various spider groups and other animals.

To my knowledge, this is the first published direct evidence that juvenile and adult orb-web spiders actively consume pollen. On the other hand, spiders sometimes refused offered pollen by dislodging the pollen actively out of their webs. This indicates that spiders sometimes regard pollen as a useless web load, perhaps affecting capture efficiency of the web. Spiders consuming pollen in this study were fed only with Drosophila flies, which are of poor nutritional value for spiders (Bilde & Toft, 2000), and perhaps therefore some spiders took the chance to utilise the extra protein portion of the pollen. As orb-webs can contain immense amounts of pollen in the field (Fig. 3), pollen-feeding may have a substantial significance for juvenile spiders.

Bilde, T. & Toft, S. (2000) Evaluation of prey for the spider Dicymbium brevisetosum Locket (Araneae: Linyphiidae) in single-species and mixed-species diets. In Gajdoš, P. & Pekár, S. (eds). Proceedings of the 18th European Colloquium of Arachnology, Stará Lesná, 1999. Ekol. Bratislava, 19 suppl. 3: 9–18.
Jackson, R. R., Pollard, S. D., Nelson, X. J., Edwards, G. B. & Barrion, A. T. (2001) Jumping spiders (Araneae: Salticidae) that feed on nectar. J. Zool., Lond. 255: 25–29.
Smith, R. B. & Mommsen, T. P. (1984) Pollen feeding in an orb-weaving spider. Science N. Y. 226: 1330–1331. Taylor, R. & Foster, W. (1996) Spider nectarivory. Am. Entomologist N. Y. Summer: 82–86.
Vogelei, A. & Gressel, R. (1989) Survival strategies of the crab spider Thomisus onustus Walckenaer 1806 (Chelicerata,Arachnida, Thomisidae). Oecologia, 80: 513–515.
Added by John Partridge at 15:19 on Fri 17th Feb 2012.

Araneus diadematus Responding to the Human Voice by Bruce Hoyle

From The Newsletter No. 101 November 2004

Whilst closely observing a specimen of Araneus diadematus in her web a few days ago, my wife called me. My response was a loud ‘no thanks’; at the same time the spider immediately responded to my voice by raising the first pair of legs in a threat posture. I repeated the word ‘no’ and again there was the same response. My daughter then tried and there was no response at all! Clearly my deeper voice frequency seems to mimic the vibration of a fly caught in the web, but as the vibration is airborne the spider has no idea in which direction it comes from—unlike a fly caught in the web. If I gently blew the web then she would usually oscillate, and if I blew in a very short burst then she usually flicked the web. She is obviously confused by a vibration coming from all directions at once and the threat posture seems to be her only response. Has anyone else observed this?
Added by John Partridge at 15:13 on Fri 17th Feb 2012.


Comparisons were made as follows: the side of a radius with the subsequent coiling around the hub; the side of a radius with the side of the subsequent radius; the coiling around the hub with the side of the subsequent radius; and the coiling around the hub with the coiling around the hub after placement of an intervening radius. The strongest correlation exists between the side of the radius and subsequent coiling: after a radius constructed on the left hand side, the spider will walk around the hub anticlockwise and, after a radius constructed on the right hand side, the spider will circle the hub clockwise (in 424 (69.3 per cent) of 612 cases, p less than 0.0001, binomial test). Weaker correlations exist between subsequent coilings (same coiling in 293 (58.5 per cent) of 501 cases, p=0.0002) and between subsequent sides of the radii (different sides in 305 (58.7 per cent of 520 cases, p=0.0001). No correlation can be found between the coiling and the side of the following radius (p=0.51) nor could any significant individual differences be found. Note that all these comparisons do not depend on the side from which the construction is observed. If you switch the observation side, left becomes right and clockwise becomes anticlockwise and vice versa. During the analysis of the moves of the spider, it was also noted that the coiling of the hub after the placement of the last radius is not always the same as the coiling of the auxiliary spiral: there was a U-turn in the hub on three occasions out of 104 webs.

Now we have again shifted the question back by one stage, at least partially, since the position of the last radius only determines the coiling of the auxiliary spiral with a certainty of about 70 per cent. We can say that the coiling of the auxiliary spiral is influenced by the position of the last radius, but other factors also seem to play a role. The question that we have to answer now is: on which side (left or right) does the spider place the ultimate radius. That is where I give up. I think that this is unpredictable, at least as long as we do not understand the exact mechanisms involved in determining the order of the radii placement.

The conclusions we can draw from the results are that the coiling of the auxiliary spiral is only determined at the very end of the construction of the radii and that it is therefore generally unpredictable. This implies that it is unlikely for the spider to have a preferred coiling, or a handedness in this respect.

I thank the Swiss National Science Foundation for financial support and Dr Fritz Vollrath for his academic support of my research.

McCook, H. C. (1889) American Spiders and their Spinning Work, 1. Published by the Author, Philadelphia.
Peters, H. M. (1937) Studien am Netz der Kreuzspinne (Aranea diadema). II. Uber die Herstellung des Rahmens, der Radialfaden und der Hilfsspirale. Z. Morph. Okol. Tiere, 33: 128-150.
Peters, H. M. (1939) Uber das Kreuzspinnennetz und seine Probleme. Naturwissenschaften, 27: 777-786.
Zschokke, S. (1993) The influence of the auxiliary spiral on the capture spiral in Araneus diadematus Clerck (Araneidae). Bull. Br. arachnol. Soc. 9: 169-173.
Zschokke, S. (1994) Web construction behaviour of the orb weaving spider Araneus diadematus Cl. Ph.D. Thesis, Universitat Basel.
Added by John Partridge at 11:12 on Thu 9th Feb 2012.

Coiling of the Spirals in the Orb Web of Araneus diadematus Clerck by Samuel Zschokke

From The Newsletter No. 74 November 1995

Which way round do spiders build their orb webs: clockwise or anticlockwise? This is a question one regularly hears from interested lay persons. There is of course no simple answer, because it depends which side you observe web construction from. Preliminary studies (unpublished) using a standardised side of observation revealed no preference for building the capture spiral clockwise or anticlockwise. Here I try to answer the question: when and how is the direction of the spiral coiling in the webs of Araneus diadematus Clerck determined?

Zschokke (1993) showed that the coiling of the capture spiral in the webs of Araneus diadematus is determined by the coiling of the auxiliary spiral. But this does not answer the original question, it simply changes it to: which way round does Araneus diadematus build the auxiliary spiral? Since the auxiliary spiral is usually built without U-turns (or reverses), we have to focus on the very beginning of the auxiliary spiral construction, namely hub construction and construction of the last radii. There is no information on this in the literature, but a number of authors have attempted to answer the related question: in which order do the spiders build the radii? McCook (1889) described it as alternating from one side to the other. Peters (1937, 1939) confirmed this, but added that there are exceptions to this rule. Let us consider how Araneus diadematus builds the last few radii. When it has almost finished the construction of the radii, it circles the hub to find a gap between two radii where an additional radius is required (Peters, 1937). When it has found such a gap, it walks out along the upper existing radius (the exit radius) to the frame, attaches its dragline to the frame below the exit radius and returns to the hub, laying the definitive radius. As soon as it reaches the hub, it attaches the definitive radius and continues circling the hub to find another gap requiring a radius. When the radii are complete, the hub is usually circled a few more times and then, without interruption, the auxiliary spiral is started by widening the distance to the previously laid thread. The coiling of the auxiliary spiral therefore depends on the direction the hub is circled after the construction of the last radius. We have now again successfully shifted the question: which way does the spider circle the hub after the construction of a radius? To try to answer this question the construction of 104 webs built by 14 spiders (between 2 and 21 webs per spider) was observed. The methods used (Zschokke, 1994) allowed the moves of each spider to be analysed in detail. The side of the construction (left or right) of the last six radii and the subsequent coiling around the hub was noted. Radii were classified as on the left or right hand side, according to the relative positions of the exit radius and the newly laid radius: if the radius was shifted clockwise compared to the exit radius, then this radius was classified as on the right hand side of the web; if shifted anticlockwise from the exit radius, then it was classified as on the left hand side of the web. This classification corresponds to their real position in the web, except for radii at the top or bottom of the web. Continued in next note
Added by John Partridge at 11:05 on Thu 9th Feb 2012.

Pollen-eating spiders: Araneus diadematus Clerck by Lawrence Jones-Walters

From The Newsletter No. 43 July 1985

In a fascinating recent paper, Smith & Mommsen (1984) showed that microscopic organic matter may be the main food of orb-weaving spiderlings, with insects providing only a dietary supplement Speculations about the possible benefits of web eating were raised after young orb-weavers were observed to spin and dismantle several successive webs without apparently capturing any insect prey. In a remarkable piece of research it was found that pollen, which is caught on the sticky spirals of Araneus diadematus orb webs, doubles the life expectancy of second instar spiderlings and alters their web-spinning behaviour so that they spin more frequently than fasting controls. Interestingly, none of the spiderlings fed solely on pollen were able to moult but lived as long as 36 days. Although the cause of death was unknown, the observation that some of these individuals outlived aphidfed spiderlings suggested that they might have died from failure to moult rather than starvation. It was suggested that this was due to some nutritional deficiency of the pollen, perhaps tyrosine, an amino acid essential for the formation of new cuticle in insects and spiders. Spiderlings fed a single aphid at the beginning of an experimental period were able to moult, even when subsequently maintained on an insect-free diet after their first meal. If the spiders thrive on pollen, they must have an enzyme in their pharyngeal fluid which can break down the resistant exine coat of pollen grains. The existence of such an enzyme in animals is supposed to be rare, and the challenge for arachnological researchers is obvious.


Smith, R. B. & Mommsen, T. P. (1984) Pollen feeding in an orb-weaving spider. Science 226: 1330-1332.
Added by John Partridge at 19:51 on Tue 10th Jan 2012.

Life-cycle of Araneus diadematus Clerck, Argiopidae by D. S. Bunn

From The Newsletter No. 35 November 1982

Despite its widespread abundance, there still seems to be uncertainty as to whether Araneus diadematus takes one season or two seasons to reach maturity. Locket and Millidge (1953) quote Wiehle (1931) who stated that it takes two 'years' on the continent, and they considered it likely to be the same in this country; while Bristowe (1958) believed that there was insufficient time for young spiders dispersing in May to attain maturity the same season. I can now throw some light on the actual situation.

For many years I had observed the species in the northwest of England and in Worcestershire and had noted that in late summer both final instar and young specimens were in evidence, indicating a two-year cycle. However, in 1977 I spent two holidays in Dorset — a fortnight spanning late June and early July, and in the third week of September: these holidays put a quite different complexion on the matter. I made most of my observations at Lodmoor near Preston, Weymouth, and there found that in June and July there were only very tiny, recently hatched spiders present, while in September there were only modestly-sized adults. Lodmoor is a sunny open area and the above observations clearly pointed to a temperature factor, linked with an abundance, or otherwise, of prey, and is supported by my discovery in September of two generations of spiders on nearby Portland (a more exposed habitat), the adults here being exceptionally large.

Additional observations of the species were made in July 1974, August 1978 and September 1981 at La Coruna, north-west Spain, where I found a situation compatible with that at Lodmoor: there was only one age-group. In July and August they all appeared to be in the penultimate instar and in September they were in their final instar. These results are consistent with a one- or two-year life-cycle, the spider attaining maturity 4 or 15 months from leaving the egg-sac in May, dependent upon the local weather conditions. In order to confirm my conclusion, I observed the progress of young spiders hatched from eggs in May 1978 in my (unheated) greenhouse in Lancashire: these made smallish adults four months later, as at Lodmoor. I then reasoned that the very large specimens I saw on Portland were those in a situation where the conditions are not warm enough to promote a one-year cycle but are nevertheless better than average. In May 1979 I therefore collected a small spider, which would be about a year old, from Worcestershire and introduced it into my greenhouse. By August (some 15 months from hatching) it had developed into a large specimen.

In my home district of Blackburn, Lancashire, an area noted for its dull, cloudy weather, A. diadematus usually reaches a very small adult size even after two summers, whereas some of those in La Coruna in September 1981 were very large, demonstrating their capacity to grow to a large size even in one season, albeit a much longer one, than anywhere in Britain. I have not been able to ascertain whether the number of ecdyses varies according to the size of the adults.

Bristowe, W. S. (1958). The World of Spiders. Lend.
Locket, G. H. and Millidge, A. F. (1953). British Spiders, Vol. II. Ray Soc., Lond.
Wiehle, H. (1931). Spinnentiere oder Arachnoidea V1 27. Familie: Araneidae in Tierwelt Deutschlands, Jena, 1931, 1-136.
Added by John Partridge at 19:42 on Tue 10th Jan 2012.

Mature spider currently living under outer lip of garden waste bin. Often constructs web with anchoring strand up to 0.5m away from bin on concrete path. Seems to gather up web and eat it? before constructing a new one
Added by Philip Gage at 09:45 on Tue 13th Sep 2011. Return to Summary for Araneus diadematus