Notice: Undefined offset: 3583 in /var/www/scholarpedia.org/mediawiki/includes/parser/Parser.php on line 5961
A spider's tactile hairs - Scholarpedia

# A spider's tactile hairs

(Redirected from Spider tactile hairs)
Post-publication activity

Curator: Friedrich G. Barth

Figure 1: Cupiennius salei (Ctenidae), a much investigated wandering spider from Central America; adult female with a leg span of ca. 12 cm [Photograph: FG Barth]

With more than 43,000 known extant species, some 500 new species currently described per year and an educated estimate of 150,000 or more existing species (Penney et al. 2012; Platnick 2013) spiders form a large group of animals, much larger than other groups traditionally receiving much more attention (e.g. birds with ca. 10,000 species, amphibians ca. 6,500 and mammals with ca. 5,500 species). They have existed at least since the carboniferous and their age of at least 320 mya is about twice that of mammals. From an evolutionary and ecological point of view spiders have been very successful to this very day. They have conquered practically all terrestrial biotopes and impress with their sheer number of individuals and their role as the main predators of insects. Undoubtedly, the spiders' success to a large extent is due to their rich repertoire of highly developed sensory systems. Among these the cuticular hairs are the most obvious ones. They also provide the input stage for a spider's tactile sense. As shown by studies in Cupiennius salei (Ctenidae), a Central American wandering spider (Barth 2002, 2008) (Figure 1), which has attained particular importance as an exemplary spider in a number of international laboratories, these tactile hairs are surprisingly well "designed" to serve their particular sensory purpose. This will be outlined in the following.

## The exoskeleton and its sensors

No other structure dominates the life of arthropods more than their exoskeleton. It not only protects them and provides the lever system needed for locomotion, but also carries a multitude of sensors at the interface between the organism and its environment. In addition to being the base for sense organs the exoskeleton is an auxiliary structure propagating the stimuli for a unique skeletal sense alien to us humans. Spiders in particular have an elaborate skeletal sense with up to some 3,500 sensory slits embedded into it. These represent sites of enhanced mechanical compliance responding to the slightest strain and deformation in the nanometer range, going along with stresses in the exoskeleton due to muscular activity, hemolymph pressure or substrate vibrations (Barth 2002, 2012a, 2012b; Fratzl and Barth 2009). Supported by many studies on the functional properties and behavioral role of its strain sensors we have to conclude that the spider is informed about the mechanical status of its exoskeleton in remarkable detail.

The sense of touch, however, is served by the other group of sensors, the hair sensilla. These are particularly numerous in hunting or wandering spiders which do not build webs to catch prey but roam around and much more rely on their tactile sense than do for instance orb weavers. Although counter to the intuition of most people (associating spiders with a web for prey capture) not only more than half of the known spider species, including Cupiennius, are such hunters, but there also seems to be an evolutionary tendency to abandon web building for the sake of free roaming hunting (Jackson 1986; Jocqué et al. 2013).

## Mechano-sensitive hair sensilla

Figure 2: The exoskeleton of Cupiennius salei is densely covered by cuticular hairs, the majority of which is innervated. (a) Ventral view of first two segments of walking legs (coxa, trochanter, trochanter/femur joint), (b) dorsal side of opisthosoma, and (c) dorsolateral aspect of proximal leg segments and prosomal tergum [Photographs: FG Barth]

Cupiennius belongs to the hairy type of wandering spiders with several 100 thousands of hairs on its exoskeleton (Figure 2). The vast majority of these hairs is innervated and serving a tactile function, the major exception being the non-innervated short yellowish plume hairs of adult spiders. There are also some chemoreceptive hairs among them, however, which are mostly found on the distal sections of walking legs and pedipalps and recognized by the steeper angle to the cuticular surface and often a slightly S-shaped hair shaft. Among the 21 sensory cells of each of these hairs there are two mechanosensitive ones, responding to the deflection of the hair shaft (Harris and Mill 1973, 1977). The most interesting known function of pedipalpal chemosensory hairs is their response in Cupiennius to a pheromone (S-dimethyl ester of citric acid) attached to the female dragline which stimulates the male to start his vibratory courtship (Tichy and Barth 1992; Barth 1997, 2002; Gingl 1998; Papke et al. 2000). Different from chemoreceptive hairs the majority of the other hairs are unimodal, responding to mechanical forces deflecting the hair shaft by direct contact (Foelix 1985). The density of these hairs is up to about 400/mm2 (Eckweiler 1983; Friedrich 1998). This is remarkable, in particular when compared to our human skin, where the densely innervated glabrous skin of the fingertip has only 1.5 Meissner afferents per mm2, to give an example (Johnson et al. 2000). A lot still has to be learned to understand whether the spider uses the high spatial resolution potentially provided by such a dense arrangement and what for (Barth 2015).

Figure 3: Structural details of the long tactile hairs sticking out of the carpet of other hairs. (a) Scanning electron micrograph of a walking leg tarsus. Arrow points to one of the long tactile hairs dorsally. (b) Basal part of tactile hair indicated in (a) according to electron micrographs. CF connecting fibrils, DS dendrite sheath, HB hair base, HS hair shaft, JM joint membrane, S socket, SS socket septum, TC terminal connecting material. (c) Dendrite terminals in longitudinal section (transmission electron micrograph). TB tubular bodies, RL receptor lymph. (d) Example of projections of tactile hairs on the femur of the second left (A2) and fourth right walking leg (A4) into the prosomal (thoracic) ganglionic mass; dorsal view. Note longitudinal sensory tracts 4 and 5 [(a) – (c) from (Barth et al. 2004), (d) from (Ullrich 2000), unpublished]

At a closer look a large diversity of mechanosensitive hair sensilla can be seen. And it is to a large extent this diversity of the mechanosensory hairs which enriches the small sensory space of spiders which lack long distance senses like our vision and hearing. The tactile sense is a close-range sense par excellence. Its hairs all form first-order lever systems and follow the same basic Bauplan. Their outer hair shaft always protrudes from a cuticular socket and they are innervated by 1-3 (most typically three) sensory cells with dendritic terminals close to the end of the inner lever arm. Tactile hairs all have a so-called tubular body in their dendritic ends indicating their mechanoreceptive function. The primary afferent fibers into the central nervous system show a somatotopic organization in the longitudinal tracts of the subesophageal ganglionic mass. There are also projections into the brain proper, the supraesophageal central nervous system (Figure 3).

### Types of morphology and arrangement

Figure 4: Distal part of C. salei walking leg. (a) The tactile space of the spider’s walking leg (not considering active leg movement) determined by indicating the tips of the tactile hairs forming the outer borderline. Individually labeled hairs all belong to the particularly long type primarily exposed to tactile stimuli. Ta tarsus, Mt metatarsus, Ti tibia, Pa patella, Fe femur; encircled crosses mark the position of spines, scale bar 5mm. (b)Tarsus and metatarsus, showing the most intensively studied long tactile hairs. Note presence of such tactile hairs on the ventral side as well [(a) adapted from (Friedrich 2001),(b) photograph: FG Barth]

The most obvious difference among sensory hairs is that between hairs adequately deflected by direct contact and those deflected by medium flow,that is by the frictional forces contained in the slightest flow of air (Figure 4). The latter type are the so-called trichobothria, of which Cupiennius has about 100 on each of its legs. The trichobothria are easy to identify: Due to their outstanding mechanical sensitivity they can easily be seen moving under the dissecting microscope following the ever present air movement. Quite differently, the deflection of tactile hairs needs much larger forces (see below). Trichobothria have been the subject of numerous studies, including physical-mathematical modeling, physiological and behavioral experiments, analyzing their dominant role in the spider's capture of flying insect prey (Humphrey and Barth 2007; Klopsch et al. 2011, 2013; Barth 2014). According to studies in several laboratories and including the analog sensors (filiform hairs) of insects they are working down to energy levels close to that contained in thermal noise and for that reason mainly have received a lot of attention from physicists and engineers as well. For the biologist the trichobothria are an impressive demonstration of the functional changes achieved by varying and properly adjusting just a few parameters contained in the Bauplan common to all hairs. Thus the spring stiffness contained in the articulation of the airflow sensing hair and resisting its deflection is smaller by up to about four powers of ten than in a typical tactile hair, where it measures between 10-8 and 10-9 Nm/rad (Dechant et al. 2001; McConney et al. 2008; Barth 2014). Similarly, low mass and the effect of hair length on the frequency tuning are used in a very "clever" way closely related to fluid mechanics in trichobothria. Details are found in reviews by Humphrey and Barth (2007) and Barth (2014).

Figure 5: Tactile hairs with one sensory cell only. (a) and (b) hair plate on the anterior side of a walking leg coxa of C. salei; scale bars 100 μm and 20 μm, respectively [from (Seyfarth et al. 1990)]. (c) long smooth hairs (length from 40 to 1000 μm) on the coxae of two neighbouring walking legs (R2 and R3); Pl pleural sclerite of prosoma; asterisks indicate areas of smooth hairless cuticle opposing both groups of hairs [from (Eckweiler et al. 1989)]

A simple way to classify spider tactile hairs is the number of sensory cells attached to them. The exceptions to the rule of 3 bipolar cells mentioned above (Foelix and Chu-Wang 1973) are hairs supplied by only 1 sensory cell, as shown for the short and stout hairs of the coxal hair plates (Seyfarth et al. 1990) and most likely applying to the two hair plates more recently found on the chelicerae. Another type of tactile hairs with one sensory cell only are the so called long smooth hairs on the coxae as described by Eckweiler et al. (1989) (Figure 5).

When going into more morphological detail (like shape of hair socket, hair length, shape of hair shaft), a further classification of tactile hairs turns out to be difficult and its value is still doubtful. Intriguingly, the distribution of hair types characterized by a certain combination of these parameters is conservative, possibly indicating a relation to particular stimulus patterns, being kind of templates of them (Friedrich 1998; Barth 2015).However, in order to turn this speculation into reliable data it still needs a lot of research. Hair sockets measure between 3 µm and 15 μm in diameter and vary in shape and their degree of openness, which strongly affects the mechanical directional characteristics of hair deflection.

### Proprioreceptive vs. exteroreceptive

Tactile spider hairs represent both proprioreceptors and exteroreceptors, the difference being that the first ones are stimulated by self-generated stimuli whereas the latter respond to stimuli from an outside source. Proprioreceptive stimulation amply occurs during locomotion when hairs located at joints are deflected by joint movement or when a joint membrane rolls over a field of coxal hair plate sensilla. The long smooth hairs are stimulated when two neighboring coxae are approaching each other, most likely measuring the distance between them (Eckweiler et al. 1989; Seyfarth et al. 1990; Schaber and Barth 2014).

### Well matched micromechanics

A number of electrophysiological experiments as well as computational studies of their micromechanical properties have revealed surprisingly "clever" details of tactile spider hair properties, reflecting the properties of the stimuli they have to cope with. Examples are the following.

#### Hair shaft or outer lever arm

1. The tactile hairs studied in detail are the long hairs (length up to ca. 3.2mm) dorsally on the walking leg tarsus and metatarsus, forming the outer boundary of the spider's tactile space (Figure 4) (Albert et al. 2001). They are stimulated from above when the spider is moving around in small spaces and hitting obstacles while wandering at night (Schmid 1997; Barth 2015). As already mentioned the stiffness of the hairs' articulation is larger by up to four powers of ten than that of the trichobothria. As a consequence the forces (which are in the μN range) needed to deflect these and similar tactile hairs on other body parts (Figure 2) not only deflect but at the same time bend the hair shaft. Accordingly, Young's modulus E and the second moment of inertia J along the bending hair shaft dominate the hair's mechanical behavior during stimulation. Inertial forces due to the mass of the hair shaft and highly relevant in case of the trichobothria may be neglected in case of the tactile hairs. An important consequence of the bending of the hair shaft, its cross sectional heterogeneity and the increase of J by roughly 4 powers of ten from the tip of the hair towards the hair base is that the point of contact of the stimulus moves closer and closer towards the hair base with increasing stimulus force from above. From this it follows that at the same time the effective lever arm decreases. The bending moment therefore increases more and more slowly until it saturates at ca. 4 ✕ 10-9 Nm (Dechant et al. 2001).This in turn implies protection against breaking, an increased working range (as compared to a non-bending rod) and higher mechanical sensitivity for small deflections (forces ca. 5 ✕ 10-4 N/°) and the initial phase of a stimulus than for large stimuli (forces ca. 1 ✕ 10-5 N/°).
2. According to Finite Element Analysis the hair shaft may be considered a structure of equal maximum strength underlining its mechanical robustness. Axial stresses do not exceed ca. 3.2 ✕ 105 N/m2 (Dechant et al. 2001).
3. A seemingly perfect match between stimulus and hair micromechanics is in addition found in what was identified as a "second joint" within the socket. There the deflected hair shaft bends even before it contacts the socket (Barth et al. 2004).

#### Inner lever arm and dendrites

As it seems the hair base and the inner lever arm of the tactile hairs (on the inner side of the axis of rotation) likewise are structures favoring a combination of high sensitivity and mechanical protection of the dendritic terminals from being overloaded and damaged (Barth 2004). The inner lever arm is only ca. 3.5 μm long, that is at least ca. 750 times shorter than the outer one. This implies a considerable scaling down of the hair tip movement and a corresponding scaling up of the force close to the dendrites. When the hair is deflected by 10°, which is close to the maximum deflection under biological conditions, the torque counteracting the stimulus measures ca. 10-8 to 10-9 Nm and the displacement close to the dendrites ca.0.5 μm. At the physiologically determined threshold stimulus of 1° ("slow cell", see below) the latter value is only 0.05 μm (Albert et al. 2001).

#### Directionality

Apart from effects of the asymmetry of the socket structure (as clearly seen from above in some tactile spider hairs) there are directional dependencies of the torques resisting hair deflection before any contact with the socket. The most pronounced case so far known of such a mechanical directionality are the hairs at the joint between walking leg tibia and metatarsus. For the natural direction of stimulation the torsional restoring constant S is smaller by a factor of about 100 as compared to all other directions (Schaber and Barth 2014). Dechant et al. (2006) provide a quantitative mathematical description of any cuticular hair, based on the stiffness of its articulation in the preferred direction and transversal to it.

### Physiological responses of sensory cells

Figure 6: Electrophysiological characteristics of long tactile hairs. (a) Responses of the fast and the slow cell to a ramp and hold stimulus (hair deflection). (b) Characteristic curves of slow and fast cells of tarsal and metatarsal tactile hairs [from (Albert et al. 2001)]

The tactile hairs sticking out of the carpet of hairs dorsally on the walking legs of Cupiennius have also been subjects of electrophysiological experiments (Albert et al. 2001). Like in Ciniflo (Harris and Mill 1977) for unknown reasons extracellular recordings were only possible from two of the three sensory cells, one of which is much larger than the others. Tactile hair sensory cells consistently show phasic response characteristics, answering to the dynamic stimulus phase, that is to the velocity of hair deflection. Using biologically relevant stimulus velocities the maximum response is seen with a latency of 1 to 2 ms only, a very short time typical of many mechanoreceptors and implying high temporal resolution. Adaptation time to static deflection varies; consistently, a "slow" and a "fast" cell were found, the latter being much less sensitive in terms of the deflection velocity threshold than the former (ca. 30°/s vs. <0.1°/s) (Figure 6).When exposed to ramp-and-hold stimuli the action potential frequency follows a simple power function $$y(t) = a \times d \times t^{-k}$$ in both cases. Here y is the impulse rate, t is the time, a a constant representing the amplification, d stimulus amplitude and k a receptor constant describing how quickly the response to a maintained stimulus declines. In the present case k-values are around around 0.5, implying properties in between that of a pure displacement receiver (k=0; response independent of frequency) and that of a velocity sensor (k=1;differentiator of first order). Lowest threshold deflection angles are in the range of 1°. The characteristic curves (impulse rate vs. angular velocity of hair shaft deflection) are saturation curves for both the slow and the fast cell. However, the slow cell saturates at much lower velocities than the fast cell (Figure 6). The corresponding values for the tarsal tactile hairs (TaD1 and TaD2; Figure 4) are 250° s-1 and 650° s-1, respectively. For the slow cell threshold deflection angles are independent of deflection velocity, whereas they highly depend on it for the fast cell for which the minima occur at ca. ≥100° s-1 (Albert et al. 2001). Importantly, the cells do not provide information about the exact time course of hair deflection but only about its presence and onset. Whereas the "fast" cell is working like a mere quasi-digital "event detector", the "slow" cell is suggested to serve the analysis of the texture and shape of surfaces actively scanned by the spider and to be well adapted to that function, similar to the SAI tactile units in the vertebrate glabrous skin (Albert et al. 2001; Johnson 2001; Barth 2015).

The reader interested in insect mechanoreceptive hairs is referred to a review article by Keil (1997) and to a paper by Theiß (1979) who found a spring stiffness S in fly macrochaetae very similar to that described here for the spider case. Of particular interest may be the remarkable lack of a standing transepithelial potential in spiders. Such a potential was found to be fundamental for the primary processes (transduction) in insect sensilla but is absent from spiders according to all knowledge currently available (Thurm and Wessel 1979; Thurm and Küppers 1980; Barth 2002).

### Behavioral roles

Figure 7: The body raising behavior of C.salei .(1) The spider approaches a 10 mm high wire obstacle from the left.(2)It raises its body as soon as tactile hairs ventrally on the proximal leg and the sternum touch it.(3) Having passed the obstacle the spider returns to its undisturbed walking position. [from (Seyfarth 2000)]
Figure 8: (a) Action potentials of tactile hair sensory cell (SN) and coxal muscle c2 (Myo) responding to the deflection of 10 tactile hairs on the hind leg coxa. (b) Cross sectioned neuromer of leg 4 showing the incoming tactile hair ending and the motor neuron supplying coxal muscle c2.There are local and plurisegmental interneurons which are not shown here. Numbers 1 to 5 indicate the centrally located sensory longitudinal tracts; CT, CL, and VL represent the central, centro-lateral and ventro-lateral tract, respectively. [from (Milde and Seyfarth 1988)]
Figure 9: (a) Excitatory motor neurons innervating coxal muscle c2. Above: Motor neuron activity (MN) and the corresponding muscle response (Myo) following tactile hair stimulation. (b) Plurisegmental interneuron involved in body raising behavior of C. salei. Above: Stimulation of right hind leg (R4) induces reflex muscle activity in the same leg (R4, Myo) and also activity of an interneuron on the contralateral side (L4,IN) [from (Milde and Seyfarth 1988)]

A lot still needs to be done to better understand why Cupiennius and other wandering spiders are so extremely well equipped with tactile hairs. A particular research requirement is the analysis of complex stimulus patterns like those seen during active tactile probing in the dark and courtship and copulation (Barth 2002). Upon simple stimulation like the experimental deflection of a few neighboring or individual tactile hairs the spider withdraws the stimulated body part or turns away from the source of stimulation. Often this behavior follows the deflection of a single hair only, like raising the opisthosoma ("abdomen"), lowering the prosoma ("cephalothorax") or withdrawing the spinnerets (Seyfarth and Pflüger 1984; Friedrich 1998; Barth 2015). Interestingly, the long tactile hairs forming the outposts of the sense of touch are not only found dorsally on the leg, but also ventrally on the tarsus (Figure 4) and among all tactile hairs so far studied they were the ones most easily deflected (smallest values for elastic restoring constant S; 5.9 ✕ 10-11 Nm/° for distad deflection). They are assumed to provide sensory feedback information during locomotion.

A full neuroethological analysis of a tactile behavior comes from the work of E-A Seyfarth and his associates who examined "body raising" behavior in Cupiennius, which is also found in salticid and theraphosid and probably many other spiders (Figure 7)(see review by Seyfarth 2000). This behavior must be very helpful when the spiders walk around on structurally complex terrain and have to avoid and cope with all sorts of mechanical obstacles. Seyfarth and co-workers successfully traced the information flow from the sensory receptors to the central nervous system and the motor output. The deflection of long tactile hairs on the sternum or ventrally on the proximal leg first activates the coxa levator muscle of the stimulated leg. Thus a primary local response pulls the coxa against the prosoma while the distal leg joints are extended hydraulically. Coxal muscle activity in turn leads to the stimulation of internal joint receptors at the tergo-coxal joint, which triggers a second, the pluri-segmental response: The muscles of the remaining seven legs contract almost simultaneously and the legs are extended. By intracellular recording and staining the neuronal correlates of "body raising" behavior could be identified from the stimulated tactile hair and its sensory afferents to inter- and motor neurons (Figure 8, Figure 9).

Another, more recent analysis, asked how well adapted tactile hairs ventrally at the tibia-metatarsus joint might be to a proprioreceptive function, monitoring the movement of the joint (Schaber and Barth 2014). The results are much in favor of such an adaptedness.

1. Hairs opposing each other on the tibia and metatarsus side of the joint, respectively, deflect each other by some 30° (mean) during locomotion, with the microtrichs covering the hair shafts interlocked.
2. For both hairs the torque resisting deflection during locomotion (ca. 10-10 Nm rad-1) is smaller by up to two orders of magnitude than that in the opposite direction.
3. Action potential frequencies recorded from individual tactile hair sensory cells follow the velocity of hair deflection within the naturally occurring range of step frequencies between 0.3 and 3 Hz. Obviously then, these joint hairs are well suited for a proprioreceptive function.

There is much left for future research on the tactile sense of spiders. Do wandering spiders use tactile information on form, size and texture of an object's surface? When watching their smooth and elegant way of moving around on geometrically complex structures like bromeliads or other plants and watching the females producing and manipulating their egg sac one is strongly inclined to assume that they do use it. The same applies to the capture and handling of their prey. The next question then is: How are they doing it? The active tactile probing of their immediate environment (Schmid 1997) should be examined in more detail as should be the handling of prey and the sexual partner during precopulatory and copulatory communication (Barth 2002, 2015). Such knowledge would now allow us to predict the respective stimulation patterns of the sensory periphery and to hypothesize on the information theoretically available to the central nervous system. What the central nervous system is doing with it still largely is in the dark but the neuroethological analysis of the body raising behavior of Cupiennius (Seyfarth 2000) nicely shows what can be done and hoped for.

## References

• Albert, J T; Friedrich, O C; Dechant, H-E and Barth, F G (2001). Arthropod touch reception: spider hair sensilla as rapid touch detectors. Journal of Comparative Physiology A 187(4): 303-312. doi:10.1007/s003590100202.
• Barth, F G (1997). Vibratory communication in spiders: Adaptation and compromise at many levels. In: Orientation and communication in arthropods. M. Lehrer Birkhäuser, Basel. pp. 247-272. 9783764356934
• Barth, F G (2002). A spider’s world: senses and behavior. Springer, Berlin Heidelberg New York. pp. 394. 9783540420460
• Barth, F G (2008). Cupiennius (Araneae, Ctenidae): Biology and sensory ecology of a model spider. In: Natural and cultural history of the Golfo Dulce region. 88 A. Weissenhofer, W. Huber, V. Mayer, S. Pamperl, A. Weber, G. Aubrecht (Hrsg.) Stapfia., . pp. 267-276.
• Barth, F G (2012). Spider strain detection. In: Frontiers in sensing: from biology to engineering. F.G. Barth, J.A.C. Humphrey, M.V. Srinivasan Springer, Wien New York. pp. 251-273. 9783211997482
• Barth, F G (2012). Arthropod strain sensors. In: Encyclopedia of nanotechnology. Bushan Bharat Springer, New York. pp. 127-136. 9789048197521
• Barth, F G (2014). The slightest whiff of air: Airflow sensing in arthropods. In: Flow sensing in air and water: behavioral, neural and engineering principles of operation. H. Bleckmann, J. Mogdans, S.L. Coombs Springer, Berlin Heidelberg New York. pp. 169-196. 9783642414466
• Barth, F G (2015). A spider’s sense of touch: What to do with myriads of tactile hairs? In: The ecology of animal senses. Matched filters for economical sensing. G. von der Emde, E. Warrant E Springer, Berlin Heidelberg New York. IN PRESS
• Barth, F G; Nemeth, S S and Friedrich, O C (2004). Arthropod touch reception: structure and mechanics of the basal part of a spider tactile hair. Journal of Comparative Physiology A 190(7): 523-530. doi:10.1007/s00359-004-0497-4.
• Dechant, H-E; Rammerstorfer, F G and Barth, F G (2001). Arthropod touch reception: stimulus transformation and finite element model of spider tactile hairs. Journal of Comparative Physiology A 187(10): 851. doi:10.1007/s00359-001-0255-9.
• Dechant, H-E; Hößl, B; Rammerstorfer, F G and Barth, F G (2006). Arthropod mechanoreceptive hairs: modeling the directionality of the joint. Journal of Comparative Physiology A 192(12): 1271-1278. doi:10.1007/s00359-006-0155-0.
• Eckweiler, W (1983). Topographie von Propriorezeptoren, Muskeln und Nerven im Patella-Tibia- und Metatarsus-Tarsus-Gelenk des Spinnenbeins, Diploma thesis. Fachbereich Biologie, University of Frankfurt um Main.
• Eckweiler, W; Hammer, K and Seyfarth, E-A (1989). Long smooth hair sensilla on the spider leg coxa: sensory physiology, central projection pattern, and proprioceptive function. Zoomorphology 109(2): 97-102. doi:10.1007/bf00312315.
• Fratzl, P and Barth, F G (2009). Biomaterial systems for mechanosensing and actuation. Nature 462(7272): 442-448. doi:10.1038/nature08603.
• Friedrich, O C (1998). Tasthaare bei Spinnen: Zur äußeren Morphologie, Biomechanik und Innervierung mechanoreceptiver Haarsensillen bei der Jagdspinne Cupiennius salei Keys. (Ctenidae), Diploma thesis. Faculty of Life Sciences, University of Vienna.
• Friedrich, O C (2001). Zum Berührungssinn von Spinnen. Doctoral thesis. Faculty of Life Sciences, University of Vienna.
• Gingl, E (1998). Nachweis der Rezeptoren für das Kontaktpheromon an der Seide der weiblichen Jagdspinne, Diploma thesis. Faculty of Life Sciences, University of Vienna.
• Harris, D J and Mill, P J (1973). The ultrastructure of chemoreceptor sensilla in Ciniflo (Arachnida, Araneida). Tissue and Cell 5(4): 679-689. doi:10.1016/s0040-8166(73)80053-9.
• Harris, D J and Mill, P J (1977). Observations on the leg receptors of Ciniflo (Araneidae, Dictynidae) I external mechanoreceptors. Journal of Comparative Physiology 119(1): 37-54. doi:10.1007/bf00655870.
• Humphrey, J A C and Barth, F G (2007). Medium-flow sensing hairs: biomechanics and models. In: Advances in Insect Physiology 34, Insect mechanics and control. 34 J. Casas, S.J. Simpson Elsevier Academic Press,  . pp. 1-80. 9780123737144
• Jackson, R R (1986). Web building, predatory versatility and the evolution of the Salticidae. In: Spiders: webs, behavior and evolution. W. Shear Stanford University Press, Stanford. 232-268 978-0804712033
• Jocqué, R; Alderweireldt, M and Dippenaar-Schoeman, A (2013). Biodiversity: An African perspective. In: Spider research in the 21st century. D. Penney Siri Scientific Press, Manchester. 9780957453012
• Johnson, K O; Yoshioka, T and Vega-Bermudez, F (2000). Tactile functions of mechanoreceptive afferents innervating the hand. Journal of Clinical Neurophysiology 17(6): 539-558. doi:10.1097/00004691-200011000-00002.
• Klopsch, C; Kuhlmann, H C and Barth, F G (2012). Airflow elicits a spider’s jump towards airborne prey. I. Airflow around a flying blowfly. Journal of The Royal Society Interface 9(75): 2591-2602. doi:10.1098/rsif.2012.0186.
• Klopsch, C; Kuhlmann, H C and Barth, F G (2013). Airflow elicits a spider’s jump towards airborne prey. II. Flow characteristics guiding behaviour. Journal of The Royal Society Interface 10(82): 20120820. doi:10.1098/rsif.2012.0820.
• McConney, M E et al. (2008). Surface force spectroscopic point load measurements and viscoelastic modelling of the micromechanical properties of air flow sensitive hairs of a spider (Cupiennius salei). Journal of The Royal Society Interface 6(37): 681-694. doi:10.1098/rsif.2008.0463.
• Milde, J J and Seyfarth, E-A (1988). Tactile hairs and leg reflexes in wandering spiders: physiological and anatomical correlates of reflex activity in the leg ganglia. Journal of Comparative Physiology A 162(5): 623-631. doi:10.1007/bf01342637.
• Papke, M; Schulz, S; Tichy, H; Gingl, E and Ehn, R (2000). Identification of a new sex pheromone from silk dragline of the tropical hunting spider Cupiennius salei. Angewandte Chemie 112: 4517-4518.
• Penney, D; Dunlop, J A and Marusik, Y M (2012). Summary statistics for fossil spider species taxonomy. Zoo Keys 192: 1-13. doi:10.3897/zookeys.192.3093.
• Schaber, C F and Barth, F G (2014). Spider joint hairs: adaptation to proprioreceptive stimulation. Journal of Comparative Physiology A 201(2): 235-248. doi:10.1007/s00359-014-0965-4.
• Schmid, A (1997). A visually induced switch in mode of locomotion of a spider. Zeitschrift für Naturforschung 52c: 124-128.
• Seyfarth, E-A (2000). Tactile body raising: neuronal correlates of a 'simple' behavior in spiders. Proc 19th Eur Coll Arachnol. In: Europ Arachnol 2000. S. Toft, N. Scharff Aarhus University Press, Aarhus. pp. 19-32.
• Seyfarth, E-A and Pflüger, H-J (1984). Proprioreceptor distribution and control of a muscle reflex in the tibia of spider legs. Journal of Neurobiology 15(5): 365-374. doi:10.1002/neu.480150506.
• Seyfarth, E-A; Gnatzy, W and Hammer, K (1990). Coxal hair plates in spiders: physiology, fine structure, and specific central projections. Journal of Comparative Physiology A 166(5): 633-642. doi:10.1007/bf00240013.
• Theiß, J (1979). Mechanoreceptive bristles on the head of the blowfly:mechanics and electrophysiology of the macrochaetae. Journal of Comparative Physiology A 132(1): 55-68. doi:10.1007/bf00617732.
• Thurm, U and Wessel, G (1979). Metabolism-dependent transepithelial potential differences at epidermal receptors of arthropods. Journal of Comparative Physiology A 134(2): 119-130. doi:10.1007/bf00610470.
• Thurm, U and Küppers, J (1980). Epithelial physiology of insect sensilla. In: Insect biology in the future VBW80. M. Locke, D.S. Smith Academic Press, New York. pp. 735-763. 9780124543409
• Tichy, H and Barth, F G (1992). Fine structure of olfactory sensilla in myriapods and arachnids. Microscopy Research and Technique 22(4): 372-391. doi:10.1002/jemt.1070220406.
• Ullrich, N D (2000). Zum Berührungssinn von Spinnen: Feinstruktur und zentrale Projektion von Tasthaaren bei Cupiennius salei Keys.(Ctenidae). Diploma Thesis. Faculty of Life Sciences, University of Vienna. NOT FOR PRINT