Mobility of Spiders
On this page the probable mechanisms for walking and other activities of the limbs of spiders are examined.
There may be some limb or body movements that are peculiar to one or just a few spider species but the majority of spiders appear to operate their legs and other appendages in essentially the same manner and it is these limb movements that are the subject of this page.
What anatomical structures are spiders able to move?
Of obvious importance are the eight legs, which can be flexed, extended and rotated. The same is generally true for the palps and in adult males these must also be capable of collecting sperms from the male genital system and delivering them to the female's genitalia. Then there are the chelicerae and fangs which at least in araneomorph species display pincer-like actions for holding prey and partly macerating them prior to digesting them. Although the fangs and the teeth on the chelicerae may help a spider hold a struggling insect without continuous strong muscular contractions there is some need for muscle activity until the prey has been paralysed or killed. And finally, many spiders with relatively long spinnerets have been observed to spread them like the fingers on a human hand while extruding silk and it is reasonable to assume this also involves muscle contractions.
What physiological mechanisms do spiders use to make their appendages move?
In the human body limbs are flexed, extended and rotated by the actions of skeletal muscles, most of which exert their force via tendons attached to bones in the centre of each limb. Flexor muscles allow the limb to bend and extensors straighten it, the two kinds of muscles operating antagonistically so a limb can be moved to, and held in, whatever position we choose. Contractions of these muscles are under the control of the nervous system, a large part of which is devoted to regulating and synchronizing the contractions of individual muscles. But is the same true for spider legs? Well, until 1959 most biologists assumed spiders operated their limbs by much the same mechanisms as we use. This situation changed radically when Perry and Brown (see reference at the end of this page) published experimental results that indicated that spiders straighten their legs not by the use of extensor muscles and long tendons but by a hydraulic mechanism involving haemolymph.
Almost all spiders have walking legs composed of seven segments and the joints between most of these are hinge-like and have a soft membrane covering a fluid space on the flexion side (the underside) of the joint. This means the leg can bend in only one direction and pressurizing the haemolymph in the leg will extend it in the same way that pumping water into a deflated and flexed fireman's hose will straighten and stiffen it. Then when the haemolymph pressure is reduced and the surplus fluid flows back into the spider's body the elastic elements of the joint draw the leg back into a partly flexed posture, which is what is normally seen when a spider dies suddenly or decides to 'play dead' in order to deceive a predator.
Hydraulic extension of spider legs during walking was presumed by Perry and Brown to involve brief compression of the cephalothorax by its own internal muscles, perhaps even the same ones that operate the sucking stomach described in the "What spiders eat" page of this website. On most larger spider species it is possible to see the insertion points for at least some of these muscles, namely the fovea on the centreline of the carapace and (at least on mygalomorph species) the six pits called sigilla which form an oval on the sternum.
But does this hydraulic theory adequately explain the known leg movements of a typical spider? Supporting the theory is the fact that extending
legs as long and thin as those of a pholcid (daddy-long-legs) or tetragnathid spider by use of tendons appears likely to be much less
efficient than simply straightening them by hydraulic action. In addition, it has been shown that there are changes in the haemolymph
pressures within the cephalothorax whenever a spider becomes active. On the other hand, this increase in pressure reduces the ability of the spider's
abdominally located heart system to perfuse the cephalothorax with oxygenated haemolymph at a time when it might actually need more oxygen there to ensure
the nervous and sensory systems are adequately supplied.
However, we now have strong evidence that spider leg movements typically are NOT controlled purely by a hydraulic system but by a hybrid system that uses
muscles for leg flexion and hydraulics for leg extension. As the graphic below shows, it is not difficult to find muscles in the legs of large spiders such
as the Australian tarantulas and the orientation of these muscles across leg joints indicates that they will cause the leg to flex (and perhaps also to
rotate sideways in some cases) when they contract. Some researchers have also claimed that they have been able to make electrophysiological recordings
from nerves linking a spider's nervous system to its leg muscles. Other facts compatible with the involvement of neuromuscular mechanisms in
spider leg movements are as follows:
(1) If the legs are extended hydraulically by an increase in haemolymph pressure in the cephalothorax there presumably has to be some kind of valve
at the entrance to each leg to avoid having all legs extend at the same time when the spider is walking forward. And if this is the case then each valve must surely be opened
and closed in a sequential fashion by nerve impulses.
(2) Many spiders rest for long periods of time with all legs extended and this would have a high energy cost unless each extended leg has one or more valves to
prevent haemolymph backflow, and once again these valves would need some kind of neural control. Perhaps an extended leg can be held in place by the claws or
hairs at its end, these gripping onto any crevices on the surface on which the spider is resting, but then the spider would need a mechanism for
releasing these anchorages when it decides to move its legs.
(3) The only parts of the typical spider's leg that can be stretched during hydraulic leg extension are the flexible joint membranes but these are so
short they are unlikely to be able to exert enough force to fully flex a large leg that has been extended hydraulically. Indeed, if these elastic
elements of the leg were able to exert a strong flexing force this would have to be overcome by equally strong hydraulic action the next time the leg was
to be extended.
(4) Spiders need to be able to exert strong and sustained leg flexion force for such purposes as to be able to hold onto a twig during strong winds or to prevent the loss of a
relatively large insect that is struggling to escape from the spider's grasp. Flexor muscles can supply this need much more efficiently than passive
leg elasticity. Similarly, those species that can climb vertical surfaces or support threads need the flexing power of muscles in the first two
pairs of legs if their ascent is to be rapid and efficient.
(5) When walking horizontally many spider species use the first two pairs of legs to drag the spider's body forward by flexing, only the fourth pair,
and to a lesser extent the third pair, pushing the spider forward by hydraulically induced extension. So once again Legs 1 and 2 are better able to exert
flexing force by muscle contraction than by interruption of hydraulic leg extension.
Over the last 50 years there have been a number of attempts to determine spider walking patterns with the aid of high speed film or video recording. In a 1967 research paper (see below) involving the walking actions of a tarantula species D.M. Wilson stated that it can be assumed that the basic stepping sequence for a spider is 4-3-2-1, these numbers referring to Legs 4, 3, 2 and 1, the spider's first pair of legs being L1 and R1. He also claimed that the cycle on the two sides of the body normally is in opposite phase so at the same time as L1 is extending forward R1 will be flexing backwards. However, Wilson did admit that even during steady walking there could be phase drifts between legs and that contralateral antagonism is not strongly fixed but is most nearly consistent in the fourth pair of legs. Even in 1967 these walking patterns were believed to be controlled by nerve ganglia in the spider's body and that proprioceptors associated with the limbs play a significant part in varying the walking pattern of tarantulas. More information about these ganglia and preoprioceptor systems can be found on the nervous and sensory systems page of this website.
The following silhouette graphic is an image taken from a video sequence recorded while a European araneomorph specimen was running forward. The legs are numbered with R for the right side an L for the left. The graphic also is labelled to show whether each limb was extending or flexing at the moment the image was captured. If you study this image carefully you should be able to see that each leg is about a half-cycle out of phase with the equivalent leg on the other side of the spider's body. It also appears that the actions driving/releasing/lifting-away/preparing-to-drive do not take exactly the same amount of time, namely a quarter of the cycle, but the sequence of driving does appear to follow the 4-3-2-1 pattern.
However, after even a relatively brief period of carefully watching a variety of spider species putting their legs to good use most people will probably conclude that it is simplistic to say that every kind of spider uses its legs only in the 'classic' 4-3-2-1 pattern. Spiders use their legs for a variety of purposes and in a variety of circumstances and need to be able to vary their sequencing to suit each particular situation. Even slow walking as distinct from fast running is likely to change the leg sequence pattern to some extent. Jumping spiders (Salticidae) usually perform their jumps using only Legs 3 and 4 and both pairs are extended hydraulically at the same time to execute the jump, the front two pairs of being extended upwards and forwards perhaps in readiness to grasp prey the spider has spotted.
Similarly, those spiders that use webs to capture insect prey characteristically hold the prey and/or the web with their front legs and use their rear legs in an alternating fashion to draw silk out of their spinnerets and wrap it around the prey to secure it. In addition, some spiders support the body on the rear legs while using at least the first pair of legs for a tactile exploration of an object in front of them that they cannot actually see because their eyes do not face forward. Similarly, the male and female of mygalomorph species such as the funnel-web Hadronyche infensa rear up and lock their front legs during mating, the male using this stratigem to mate without being eaten by the female. But the spectacularly coloured male peacock spider, Maratus volans has a very different use for its legs during mating. The male waves both left and right Leg 3 vertically and extends an elaborately patterned fold in its abdominal skin in an attempt to attract a favourable mating response from the female.
Sudden changes of direction, twisting to enter a retreat or burrow, or tweaking a web with just one leg to attract prey or a mate are just a few more of the many reasons why spiders sometimes need to temporarily change their leg sequencing. But one of the most common changes in leg sequence is that seen when, as the next graphic shows, a spider runs into a large obstruction, or just decides to rest. It is extremely common for the members of most spider families to come to rest with all legs extended or at least with the matching legs on each side of the body in the same position. This is certainly true when the spider is resting on a flat surface but is even a normal practice of species such as Argiope keyserlingi and Leucauge granulata that spend almost their entire lives in the middle of their webs.
The leg sequencing mechanisms so far described apply to spiders that have prograde legs which by definition are ones that point fore and aft and
flex in a vertical direction. But what about those sparassid and thomisid spiders which have laterigrade legs that are rotated sideways and move in a crab-like
fashion? Well, there is no reason why their sequencing should be all that much different except that Legs 3 and 4 may not point so far backwards when driving the
spider forward. Flexing of the legs can still occur but this will be more nearly sideways than vertical. This arrangement may be less efficient than is the case for prograde spiders and this might at least partly explain why in the
Thomisidae Legs 1 and 2 are often much larger and more robust than Legs 3 and 4. The laterigrade leg configuration may allow spiders to be relatively flattened so they can
slide into narrow crevices but it tends to impair fast running and it is probably significant that the faster members of the Sparassidae such as Heteropoda and Neosparassus species tend to have only
partially laterigrade legs.
Thus in summary, while there clearly are variations in the ways particular spider species move their legs the presently available data indicates that in general
The pages on what spiders eat, haemolymph circulation, and silk production contain some information that is related to what is covered in the above paragraphs. In addition, the following are worth reading:
Perry D.A. and Brown R.H.J. (1959) "The hydraulic mechanism of the spider leg" J. Experimental Biology, 36, 654-664
Zentner L., Petkun S. and Blickhan R. (2000) "From the spider leg to a hydraulic device" Technische Mechanik, 20, 21-29
Weihmann T., Gunther M. and Brickham R. (2012) "Hydraulic leg extension is not necessarily the main drive in large spiders" J. Experimental Biology, 215, 578-583
Wilson D.M. (1967) "Stepping patterns in tarantula spiders" J. Experimental Biology, 47, 133-151
Email Ron Atkinson for more information. Last updated 21 April 2015.