On Spider Webs

The influence of structural form.

The last time I tackled a topic from the animal kingdom I looked at Stegosaurus, which is a rather large animal, so on this occasion I have decided to look at something much smaller. As before I still know nothing about animals and have no expertise whatsoever in the fields of biology, zoology and so forth. I shall be looking at the topic from the perspective of a structural engineer and will likely be making all sorts of terminological and other obvious errors. 

My prepared defence against basic dinosaur errors was based around stegosaurus being extinct and therefore nobody really knowing for sure. This time around I don’t have that luxury. Instead I shall base my rebuttals on that most modern phenomena of ‘getting my message out there’. In this mode of thinking errors are acceptable so long as the direction of travel is correct and ones motivation is honourable. 

Now that my new excuse has been set out, lets jump right in.

There is a cliche about spider silk being stronger than steel, which is obligatory to every discussion of Spider’s webs. I thought that I would get it out of the way early. It is perhaps less well known that spider silk is a non-linear material who’s stiffness varies depending on the applied load; it has both a slack and a stiff phase. This unusual property helps webs absorb impact from captured prey.

Although interesting the properties of spider silk are not actually the topic of this post. I shall be taking them as a given. Rather, I would like to take a brief look at the influence of structural form on spider’s webs.

In the hypothetical scenario of Sir Attenborurgh stumbling across this post and deciding to read it he would no doubt want to point out that there are many species of spider and consequently there are many kinds of web to contend with. For simplicity I shall be sticking with the common variety that most people, including me, are familiar with.

It seems to me that webs consist of several different types of member, which exist within a distinct structural hierarchy. In the first instance there are a series of threads which anchor the web to its surroundings, we shall call these the moorings.

The moorings are connected to the corners of an outer primary frame, which encloses the web. At each corner a secondary frame joins both sides of the outer frame together, but without touching the corner.

A series of radial threads extends from the centre of the web onto the primary and secondary frames. Together these members give the web its overall shape.

A spiral thread winds from the centre of the web towards the outer frames. Unlike those discussed thus far the spiral thread is made of a sticky silk, which is thinner than the other members, and is intended to catch the spider’s prey.

Before we consider how this rather spindly arrangement of threads manages to resist the impact of spider prey, and the force of strong winds, it is worth explaining a key structural principle.

The theory of elasticity dictates that when a load has several different load paths to choose from it will always prefer the one which has the greatest stiffness. In simple terms load is distributed between members according to their stiffness with the stiffest parts attracting the most load.

The importance of this principle may be illustrated by considering what would happen if spider’s designed their webs a little differently. Let us suppose that the hypothetical Institution of Web Safety, were to decree that secondary framing was no longer permitted and therefore radial threads must connect directly to the outer primary framing on all sides. Connecting directly to the primary frame, sounds like a simplification of the structural load-path. That must be good, right?

A further implication of the IWS’s directive would be manifest at the corners of the frame; the radials would now be connected directly to the moorings. Again, that must be a worthwhile safety improvement, because load is directed straight to the point of support.

We can test our theory by imagining a hypothetical fly careering into the web. We want the web to absorb the impact without breaking; that would be bad for the spider’s prospect of lunch.

This means that we want the web to spread the impact force across as many structural members as possible. The more members mobilised to resist the applied load the smaller the load each will carry.

Immediately after the fly strikes the sticky spiral the web’s load-path swings into action. The spiral is connected to lots of radials and begins to share its load. But now something has gone wrong, the load has reached a radial which is connected directly to the moorings. Being connected to the point of support this radial is much stiffer than adjacent radials, which are attached to the outer frame, which has started to flex. Load is immediately attracted out of the radials connected to the outer frame in favour of the stiffer pathway. Soon the radial connected to the mooring is carrying nearly all of the load and is stressed to breaking. Our spider’s lunch is about to escape.

Surprised by the evidence of systemic failure in radial web members the IWS takes the decision to withdraw its directive and reinstates secondary framing, which once again must be connected to the outer frame either side of corners.

Soon an unfortunate fly finds itself bumping into a web with newly reinstated secondary framing. Once again the spiral members spring into action and begin to transfer load into the radials, but this time something different happens. Instead of load being directed straight to the moorings it finds itself being directed into the secondary framing which begins to flex before re-directing load back towards the middle of the outer framing and away from the stiff corners.

The outer framing begins to flex and in doing so starts to engage other radials to which it is connected, before long much of the web is flexing and load sharing is being maximised. On this occasion there is going to be no failure. It looks increasingly like the fly is doomed and the spider will be enjoying lunch.

It seems to me that the web’s structural arrangement is designed to avoid stress concentration. This key feature maximises load spread and minimises the stress in individual members. Another consequence of the redistributive process is that the web becomes less vulnerable to local damage; because load can by-pass those areas.

Thus spider webs have a highly efficient structural form optimised for absorbing impact and for ensuring spiders remain well fed.