Web damage during prey capture in Hyptiotesparadoxus ( C . L . KOCH 1834 ) ( Uloboridae )

Web damage during prey capture in Hyptiotes paradoxus (C.L.KOCH 1834) (Uloboridae) H. p8radoxus well known for its characteristic triangular web has frequently been described to always completely collapse its web when catching prey. The aim of the present article is to show that this is not the case, and to discuss how the myth of the obligate complete collapse of the web has arisen and why it survived so well.


INTRODUCTION
We are often faced with the problem thatwe can not research everything ourselves and we therefore have to rely on published descriptions by others.Unfortunately, these descriptions are sometimes not very accurate or even incorrect, especially descriptions that have been copied from yet other descriptions.Repeated copying inevitably leads to errors -we are all familiar with the telephone game where children sit in a circle and the first child whispers a phrase into the ear of the second one, and thi~ one repeats what it has understood into the ear of the third one, and so on I,mtil finally the last one says aloud what it has understood -usually something not even remotely resembling what the first child had started with.In science we have to deal with similar problems; WALTER (1999) describes nicely how the originally inaccurate description of the prey spectrum of Eresuscinnaberinus was altered and became even less correct over time with repeated copying.
The aim ofthe present article is to show the similar fate ofthe description of the web damage during prey capture in Hyptiotes paradox us, to do away with the misconception that H. paradoxus will always completely collapse its web when catching prey, and to discuss why nevertheless many descriptions refer to such a complete collapse.
H. paradox us is one of only two orb-weaving uloborid spiders in Central and Northern Europe (HEIMER & NENTWIG 1991).It is usually found on dry twigs and branches of spruce where it builds its characteristic triangular web (cf.Fig. 1).

THE LEGEND IS FORMED
Probably the first descriptions of the prey capture in a Hyptiotes species can be found in WILDER (1874;1875).Describing the prey capture in Hyptiotes cavatus (Hentz), the American sister species of H. paradox us, he notes that "generally, an entire net is destroyed in making a single capture" (WILDER 1874, p. 270).KEW in his paper on the "snares or snap-nets" of Hyptiotes (KEW 1900) simply reviews and quotes WILDER on this subject.GERHARDT (1924) is the first to describe the prey capture of H. paradox us.
He writes that the web becomes useless with almost every prey capture and has to be rebuilt ("Es ist eine ausgesprochen unokonomische Einrichtung, dass auf diese Weise das Netz fast jedesmal beim Fang einer Beute unbrauchbarwird und neu angefertigt werdenmuss", p. 116).WIEHLE (1927) refers to GERHARDT's description but writes that the web has become useless with every prey capture (" ... denn bei jedem Fang ist das Netz unbrauchbar geworden", p; 524).He is thus the first to write that the web is destroyed with every prey capture.Descriptions written in the following years do not repeat this mistake.REUKAUF (1931) writes quite correctly that the web which has been damaged more or less during prey capture will be renewed in the following night ("Das beidem Fang mehr oder weniger schadhaft gewordene Netz wird in der nachsten Nacht erneuert", p. 695).NIELSEN (1932) describes one prey capture in detail where "the web was completely destroyed" (p.63) but he adds that this is probably not always the case.PETERS (1938) explicitly corrects WIEHLE by writing that the web does not become useless with every prey capture, on the contrary, it can be used several times for the capture of small prey animals (" ... das Netz nicht etwa bei einmaligem Gebrauch stets unbenutzbar wird, sondern dass es zum Fang kleinerer Beutetiere mehrmals verwendet werden kann", p. 57).This statement left arachnologists with conflicting descriptions of the two great experts on spider webs of that period, WIEHLE and PETERS.
Interestingly, almost all subsequent descriptions of the prey capture in Hyptiotes that describe the fate of the web refer to an• inevitable complete collapse of the web.This error is even printed in the otherwise generally reliable books by WITT etal. (1968) "This web has to be rebuilt after each prey capture" (p.34) and FOELlX (1996) "Obviously, the web becomes so damaged during the capture that a new one has to be constructed after each catch" (p.131).It is therefore not surprising that most popular books written in the second half of the 20th century (CROMPTON 1950;SAUER & WUNDERLlCH 1985;BAEHR & BAEHR 1987;BELLMANN 1992) erroneously describe an obligatory complete collapse of the web.The only exception is SAVORY (1952) who cautiously writes "Of some species of Hyptiotes it is said that the web can be used only once in this manner before it is renewed."(p.137).

OWN OBSERVATIONS
I first observed prey capture in Hyptiotes sp.during an excursion to Corsica some years ago, when I failed to demonstrate tomyfellow students thatthe spider would completely collapse its web when catching prey.We tried various prey sizes, but we could never observe a complete collapse.In 1998, I collected H. paradoxus from the wild near Basel and brought them to the laboratory where four of them (1 c:! and 3 Q ) constructed webs in perspex frames (30 cm x 30 cm x 5 cm).In my laboratory feeding experiments,l used exclusivelyfruitflies Drosophila sp. for my observations and again,l could never observe a complete collapse of the web.On several occasions; I fed one fruit fly after another into one web.Figure 1 shows such a sequence where a female H. paradoxus caught three fruit flies With the same web.The web gets progressively more damaged with each prey capture, but even after having caught three fruit flies, the web could still be used to catch more ,flies.
During my work with Hyptiotes; I have never observed a complete collapse of a. Hyptiotes web.However, I can not exclude that this may happen occasionally, especially with bigger or more struggling prey.Nevertheless, . 1 can exclude with certainty that Hyptiotes always completely collapses its web.

DISCUSSION
Why has the inaccurate description that Hyptiotes will always completely collapse its web become so widespread during the last decades?Is it be.causeWIEHLE -who was the first to publish this mistake -is generally considered to be accurate?I think it is more than this.DAWKINS (1989) introduced the term "meme" to describe a thought or concept that is passed on from one• individual to the next (orton'lany others).Memes can be thought 10cm Fig. 1: Sequence of web condition after web construction (A) and successive prey captures (8-0) in a Hyptiotes paradoxus web in the laboratory.Arrows indicate the positions where the fruit flies were caught.Note the thin trailing thread the spider left behind on its way to and from the prey.A new fly was given when the spider had finished feeding on the previous one, which took about 2.5 hours.to be the cultural equivalents of genes.As with genes, there are mutations of memes and invasions of new memes (ideas) into an existing population.Some memes come to fixation, others disappear and are replaced by others.Memeslike genes -have differential survival values (i.e. have different probabilities to survive in the meme pool).We can consider scientific concepts like OORER's rhinoceros, the prey spectrum of Eresus cinnaberinus (WAL TER 1999) or the fate of the web of Hyptiotes after prey capture to be such memes.When we analyse the survival values of these memes, we find -in accordance with the prediction by OAWKINS (1989)that memes with a high sensational value (e.g."Eresus cinnaberinusfeeds on fast and strong beetles" or "the web of Hyptiotes always collapses completely") survived better than their less spectacular alternatives.
When reviewing the publishing history of the fate of the Hyptiotes web after prey capture, it can be seen that the number of incorrect descriptions has increased overtime.Before 1950, all but one descriptions I could locate were correct, whereas almost all descriptions published since then are incorrect.Why is there this increase in incorrect publications?
It seems likely that researchers in the last century had to rely on their own observations because it was known that printed descriptions could not generally be trusted.As an example, the oldest description of the web of Hyptiotes known to me (AUSSERER 1867) describes the web to consist of three or four radii where it in fact a/ways consists of four radii.In contrast, today's research is characterised with an ever increasing complexity of the subject.This requires more use of literature data than ever before which in turn -together with a high pressure to publish -is probably the cause for the observed increase in falsely copied descriptions.

Fig. 1 :
Fig.1: Sequence of web condition after web construction (A) and successive prey captures (8-0) in a Hyptiotes paradoxus web in the laboratory.Arrows indicate the positions where the fruit flies were caught.Note the thin trailing thread the spider left behind on its way to and from the prey.A new fly was given when the spider had finished feeding on the previous one, which took about 2.5 hours.Fig. 1: Zustand des Netzes von Hyptiotes paradoxus nach dem 8au (A) und nach dem Fang von einer (8), zwei (C) und drei (D) Fruchtfliegen.Die jeweiligen Fangpositionen der Fliegen sind durch Pfeile gekennzeichnet.Die dOnnen Faden sind die Wegfaden der Spinne zur 8eute hin und zurOck zur Warteposition.Eine neue Fliege wurde jeweils in das Netz gebracht, wenn die Spin ne die vorherige vollstandig verzehrt hatte, was etwa 2.5 Stunden dauerte.