How a Spider’s Web Carries a Signal

Imagine a web stretched across an empty window frame. A tiny disturbance starts near one side. The silk shifts there, then a little farther in, and then at a junction closer to the centre. The thread does not have to travel across the frame for the disturbance to arrive somewhere else. Neighbouring parts of the silk move in succession. A vibration is movement that changes with time; here it is also a way for an event in one place to affect another.

Imagine looking at just the last junction. You might see a brief twitch. What caused it? Where did it begin? Did the junction move a little because the original event was small, or because much of its motion was lost along the way? The twitch gives you evidence, but it does not give you all the answers. To understand what the web can convey, follow the event from its source to the point where somebody could detect it.

Biologists can call a disturbance a signal when it conveys information to a receiver. That use of the word does not mean an insect, a gust of wind, or anything else disturbing the web intended to send a message. Nor does every movement produce an identifiable cue. Think of the web first as a physical connection. Questions about what an animal detects or does come after questions about what reaches it.

In a familiar kind of orb web, threads run outward from a central area called the hub. These radii are crossed by the capture spiral. A radial thread and the spiral that meets it give the web a connected structure. Not every spider builds an orb web, and real orb webs vary. We will stay with a few arrangements used by the species in two studies.

Picture one spot on the capture spiral where the silk is disturbed. At a junction the motion can affect a connected thread. If you draw a line from the spot to the hub on a photograph, you have measured a straight distance on the picture. You have not traced what happened to the silk. The route through a network and the straight distance across it answer different questions. A junction can also give the disturbance more than one route.

Location matters at the receiving end as well. A spider that monitors movement near the hub has contact with a different part of the web from one that waits in a sheltered retreat. In some orb webs a particular thread connects the hub and that retreat. You can now ask a specific question: how does the motion available on that connecting thread compare with motion available near the centre?

Hold the image of one thread still in your mind. If a point on it moves across the thread’s direction, that component of its motion is transverse. If it moves along the thread’s direction, that component is longitudinal. These words describe which way a bit of silk moves. The place where the disturbance is happening can advance along the thread even when individual points move across it. There are further distinctions in real web motion; these two are enough to show why a single label like “shaking” leaves out useful information.

There is another distinction worth keeping. Tension is the pull already acting along a stretched thread. Stiffness describes resistance to deformation. They are not interchangeable. In a web, both the condition of the threads and the pattern of their connections matter. It would be a mistake to take one result from a particular web and turn it into a rule that tightening any thread must make every movement at the far end stronger.

Mortimer and colleagues explored these properties in the orb web of the garden spider, Araneus diadematus, in 2016. They used a laser instrument to measure the motion of silk in their experimental arrangement. Those physical measurements concerned transverse movement; their computer models let them investigate longitudinal movement as well. Changes in tension and in the stiffness of the silk used for structural threads affected vibration amplitudes under the conditions they studied. What the instrument measured and what the model helped explain are both useful. Keeping them separate tells you how each conclusion was reached.

A 2018 study by Mortimer, Soler, Siviour and Vollrath compared two arrangements. Araneus diadematus generally monitors its orb web near the hub, although it can also use a retreat and a connecting thread in some conditions. Zygiella x-notata uses a signal thread from the hub to a silken retreat. Imagine the difference between standing at a junction and waiting at the far end of a route that leads away from it. Neither position lets you judge the result by distance alone.

The researchers applied short, controlled inputs at different web locations. A laser Doppler vibrometer measured silk motion at relevant receiving positions, including measurements made without the spider on the web. The team also used computer models to examine how disturbances travel. A measured change in silk motion is evidence about the structure. It is not a record of a spider’s decision or of how long it took the animal to respond.

Within the comparisons studied, monitoring via the signal thread showed no delay that the researchers considered biologically meaningful for propagation. That does not mean a disturbance travels instantly. In the comparison with the hub, transverse movement was also more attenuated at the signal thread: its amplitude, or size of displacement, was reduced. The result separates two questions. When does the motion reach a position? How large is it when it gets there? A small movement can arrive promptly. A prompt arrival says nothing by itself about whether the receiver detects it.

Use a made-up network with two routes to a measuring point. The figures here are for arithmetic practice. They are not measurements of silk or numbers from either study. Assume one constant speed along each route and start the disturbances at the same moment. Route P is 18 centimetres long at 6 centimetres per millisecond. Route Q is 28 centimetres long at 7 centimetres per millisecond. Divide distance by speed. P takes 3 milliseconds and Q takes 4 milliseconds. Q is faster per centimetre but still arrives later.

Keep the two lengths. Now change only the assumed speed for Q to 14 centimetres per millisecond. Q takes 2 milliseconds and arrives before P, which still takes 3. The longer route wins this race under the new assumption. If somebody tells you only that one route is longer, ask what else is known before deciding which disturbance will arrive first.

Give the notebook a pretend display as well. Suppose P produces a peak of 0.8 arbitrary units and Q produces 0.2. Suppose the display marks peaks of 0.3 or more. P is marked; Q is not. In the second timing example Q still arrives first in the model. It simply arrives below the display’s chosen cutoff. That cutoff belongs to an imaginary instrument. It is not a spider’s sensory threshold, and the exercise does not predict a real web. It shows why a blank screen can be a poor reason to conclude that nothing moved.

Return to the junction at the empty window frame. The next time you imagine that last twitch, try recording four things in separate columns: the event that might have started it, the route you can actually describe, the time at which motion appears, and the size of the motion at the point observed. Leave a column blank when you lack evidence. If you have only the final twitch, you know far less about the route and the source than a tidy diagram might suggest.

The toy example ends with one early arrival that stays below a display cutoff and a later one that makes a clear mark. The real comparison ends differently: it gives measured and modelled evidence about two ways of monitoring an orb web, with a difference in attenuation and no biologically meaningful propagation delay in the tested comparison. The two stories have a lesson in common. A useful account of a vibration needs both timing and strength, and it needs to say where each was observed.

You can stop there without claiming to know what a spider thought. The thread has finished moving. What the web carried was a change that could reach another place. What the receiver could make of that change is a new question, and now you know precisely where to begin it.

Mortimer, B., and colleagues (2016). Tuning the instrument: sonic properties in the spider’s web. Journal of the Royal Society Interface. https://pmc.ncbi.nlm.nih.gov/articles/PMC5046944/ — selected primary text for web structure, transverse measurements, modelled movement, tension and stiffness. DOI: 10.1098/rsif.2016.0341.

Mortimer, B., Soler, A., Siviour, C. R., and Vollrath, F. (2018). Remote monitoring of vibrational information in spider webs. The Science of Nature 105, article 37. https://link.springer.com/article/10.1007/s00114-018-1561-1 — primary study for the two monitoring positions and the difference between arrival time and attenuation. This article is licensed under Creative Commons Attribution 4.0 International: https://creativecommons.org/licenses/by/4.0/. The discussion here paraphrases selected results and adds an original teaching example. Neither study’s figures or tables are reproduced.

Selected primary text consulted on 7 November 2026, Bangladesh time. The window frame, notebook routes, speeds, amplitudes and display cutoff are imaginary. They are not a field experiment, a biological prediction or instructions for handling spiders or using laboratory equipment.