Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
SAF Archives  search ask the scientists in the classroom cool science
scientists from previous shows
cool careers in science
ask the scientists

Photo of David Clark David Clark as seen on Spiders!: Song and Dance

Click on David's photo to read a brief bio.

q Why do spiders dance before they mate? About how long do they dance? (asked by many viewers)

A In my opinion, the courtship dance of jumping spiders is one of the most delightful displays in the entire animal kingdom. These tiny animals display with a tremendous range of movements and postures. Most jumping spiders display with leg and pedipalp waving, others with abdomen twitching, some with body bouncing or tapping and males of most species use a combination of their body parts to court the female. The courtship display of jumper spiders can be vigorous and visually conspicuous or subtle and slow, depending on the species.

Since egg production by females appears to be directly related to her size and nourishment, female spiders have evolved a strategy of responding to males as prey, unless convinced otherwise by specific cues from the male (even then, the male may still be at risk of cannibalism). Consequently, male spiders have evolved conspicuous (to the female at least) species-specific displays that help distinguish them from prey. Since female spiders generally adopt a sit-and-wait foraging mode, and male spiders actively search for females, it follows that cues used for species recognition will be those which operate at a distance (such as, species-specific chemicals (pheromones), vibratory signals, visual displays) and allow discrimination between prey and a courting male.

Among spiders, the jumping spiders (Araneae: Salticidae) are distinguished by their large anterior eyes and acute vision, and are highly visually oriented in their predatory behavior (for more details about jumping spider vision, see my answer to the next question). Consequently, selection pressures from potentially cannibalistic females have resulted in the evolution of striking color patterns and active visual courtship behaviors in males. These elaborate leg waving and dance displays help males identify themselves and announce their presence to females at a safe distance. Generally, males of the dimorphic jumping spider, which I study, wait for a sexual receptivity signal from the female before moving in to mate with her.

The length of the courtship dance of jumping spiders is highly variable and depends on the species in question. The courtship display of the dimorphic jumping spider is relatively short, males display for an average of 8 to 10 seconds before the female either rejects the male or gives him a receptivity signal (usually a leg tap that signals her readiness to mate). Other species of jumping spider that I have studied have longer displays ranging from a minute or two up to 10 minutes of display before mating or rejection by the female.

q In addition to the dance, are there any other types of communication between males and females while mating? Students at WAS

A As a group, spiders have evolved several modes of communication, including acoustic/vibratory mechanisms, visual displays, and chemicals or pheromones. Like most animals, the exchange of signals between spiders serves many purposes. However, one of the most important forms of spider communication is courtship, where species and sex recognition are critical, and selection pressure from potentially cannibalistic females has shaped a diversity of male displays.

What makes the jumping spiders so unique among spiders is their well-developed visual acuity. Upon sighting a moving object in their anterior lateral eyes (ALE) and/or posterior lateral eyes (PLE) jumping spiders characteristically swivel and orient to the object. Once the long axis of the spider is aligned in this manner, the object is within the visual field of the anterior median eyes (AME) and scanned by a unique moveable retina. If the object is a potential prey item, the spider will assume a stalking posture and approach the prey while scanning with their AME. Finally, when within about 3 cm of the prey, the spider attempts to capture the prey by jumping on it and biting to subdue it.

In the presence of members of their own species, the jumping spider's behavior can be quite different. In aggressive encounters, after orienting to each other, they may assume a species-specific body posture, engage in a leg-wave display, escalate to contact and grappling, or retreat depending on the level of threat perceived, or the chance of winning. In male-female encounters, the female usually remains stationary while the male approaches using a zig-zag courtship display (or some variation of this display). While display behaviors are highly species-specific, the zig-zag movement is nearly universal. This common element of most species' courtship is thought to be a means of allowing the male to approach while remaining in the overlapping fields of view of the ALE and AME, thereby reducing the risk of being stalked by the female.

Jumping spiders are capable of resolving and discriminating images at considerable distances (and can recognize differences between conspecifics and heterospecifics based on visual stimuli alone). They have also been shown to respond to mirror images and cardboard models.

In addition to vision, jumping spiders also use acoustic/vibratory information when communicating. Depending on the species, they may thump their abdomen or tap their legs or pedipalps against the substrate to send vibratory signals. Such signals are usually given at relatively close distances when the two communicating individuals share the safe leaf surface or twig. Additionally, these signals are often coupled with visual elements, such as leg waving or body posturing. Finally, chemical signals may be used may be used in several ways, such as in silk threads that are left by spiders wherever they go, or given off into the air and detected basically as a smell or odor. Lastly, species-specific chemicals are likely on the spiders body parts and will be used as a signal when two individuals are very close to each other and touching.

q Why would a spider species evolve with two different types of males? (asked by many viewers)

A This is an excellent question, and I have been working on the answer for the past 10 years. Even after this amount of time, I still don't think I have all of the pieces of the puzzle (but that is the fun of doing scientific research). Before beginning to answer the question, I'd like to make a few comments about variation. Variation is the raw material of evolution and without variation between and within a species, evolution doesn't happen. For most animal species, there is at least some variation between males, whether it is their appearance (morphology) or their behavior, but rarely is it both within the same species and this is what makes this species of jumping spider so unique. Males of many species often vary in their physical make up (strength, stamina, endurance, etc.), their ability to secure resources (hold on to a territory, supply nourishment or shelter to mates and offspring, parental care, etc.) and their ability to acquire mates (which may be directly related to the aforementioned traits). Females often find a certain type of male more appealing than another and so certain males and their traits are favored (i.e. become more prevalent within a population).

Given that there is variation between males of most species, it is important to stress just how different the two male morphs of Maevia inclemens are from one another. As you saw in the show, the two male morphs differ dramatically in both morphology and courtship behavior. The tufted morph is entirely black in body coloration, has black colored pedipalps, white colored legs and three tufts of setae located on its anterior cephalothorax. The gray morph, is black and white striped in body coloration with a prominent white colored stripe above the anterior eyes, it has striped colored legs, it has bright orange colored pedipalps and it lacks the tufts. In addition to morphological differences, the males differ in behavior during the initial phase of courtship: the tufted morph stilts up, waves its first pair of legs back and forth and swings its abdomen side to side, all from an average distance of 9 cm from the female; the gray morph crouches in a prone posture, pointing its first two pair of legs forward in a triangle-like configuration and sidles back and forth in front of the female from an average distance of only 3 cm.

When I first started this project, I thought that females might have a mating preference for one male type over the other. Along with my colleague, Dr. George Uetz of the University of Cincinnati, we conducted a female choice experiment to determine if females had a mating preference. Females were presented with both males simultaneously in a choice test and we determined that females were not choosing males based on how they looked, but simply on which male got their attention first. We hypothesized that the different courtship behaviors evolved as different strategies for capturing female attention at different distances from the female.

I think that much can be explained about the evolution of these two morphs by understanding why the males initiate courtship the way they do and from where they do. It turns out, that even though the gray male and the tufted male look very different to us when courting, that as a function of their distance from the female, the two males actually appear similar in size. Even though stilting up makes the tufted male absolutely bigger, being at a greater distance makes him appear smaller and from a female's perspective, it appears to be the same size as the gray male courting at only 3 cm from the female. Additional studies have shown that the contrasting colors of black and white of the tufted males probably evolved to help make them conspicuous to females at greater courtship distances. The bright orange pedipalps of the gray male are useful as a signal to the female at close distances. So, males may be colored the way they are to help them signal to the female from different courtship distances. The female mating bias may also help explain why males court the way they do. Because females prefer to mate with the first male that attracts their attention, each male morph may be exploiting this response by courting the female from different distances where each has the highest probability of mating. Secondly, while male courtship behavior might be attractive to females, it could also be attractive to predators lurking in the vegetation or circling above. I have recently completed a set of experiments that suggest that the gray morph display is actually more visible when viewed from above than the tufted male. Furthermore, females themselves can be predators of the males and recent data suggest that the tufted morph is at a greater risk of being preyed upon by females when it initiates courtship close to her. While this may seem counter intuitive, it does help explain why gray males start courtship close to females and why tufted males can get away with courting at further distances from females. So, the evolution of these two males seems to be related to the benefits of attracting female attention from two different distances and the low risk of predation at these different distances from the female.

q I thought it was astonishingly bizarre that there were two highly distinct versions of the male Maevia inclemens. (Easily the most amazing thing I saw on this program.) It was mentioned in the show that half the males are of one type and half are of the other, but it wasn't clear to me: Does "half" represent a species proportion or a brood proportion. In other words, will each kind of male only produce similar males (and the female chooses each type half the time), or will the offspring of each kind of male contain both kinds of males? Nicholas

A Nicholas - This is a wonderful question. I have conducted several years of mating experiments in the lab and found that a virgin female mated to one morph is capable of producing both male morphs in the offspring. The ratio of male morphs is not always 50:50 (sometimes a tufted produces a few more gray sons than expected and visa versa), but when the morph ratio of several females are pooled, the population ratio does approximate 50:50. Likewise, field observations on these spiders have revealed that the males are nearly always found in a ratio of 50:50. Currently, I am investigating the genetics of the male dimorphism and conducting a long term field study to determine if morph ratio varies over the course of the breeding season. Female choice experiments have shown that females mate readily with either male type (and with both if given the opportunity). Females do not appear to have a morph bias as females mate with the first male that attracts their attention, regardless of male morphlogy.

q Did you have do a lot of experimentation to determine what kind of video images a female spider would respond to? Does the video image have to be a certain size, clarity, color, etc.? Mary

A Mary - Developing the video playback techniques has been one of my favorite parts of doing these experiments. I first got started doing this when I was a graduate student working with Dr. George Uetz, at the University of Cincinnati. After accidentally discovering that spiders responded to motion picture images, we conducted several experiments to determine how well jumping spiders see and respond to video images. Test spiders were presented with video images of prey (small crickets on small TV screens), video images of a predator (a larger species of jumping spider) and video images of members of their own species (live females were shown video males, etc.). Results of these experiments showed that jumping spiders were responding to each type of stimulus appropriately; they hunted and attacked the video images of crickets; they displayed aggressively and ran away from the video predators; and, females displayed receptivity behavior to video images of courting males and males courted video images of females. Conceptually, video playback experiments are relatively straightforward. However, obtaining an effective stimulus (i.e. one that gets an appropriate responses from test animals) requires fine tuning of lighting conditions; elimination of background noise; proper camera control; appropriate depth of field; and artistic skills when drawing or creating an image with a computer assisted animation program. Furthermore, video devices are designed for human visual sensitivity and video images may appear very different to animal test subjects, which may have different sensitivity ranges.

The ability to perceive video images as continuous motion, rather than a series of static pictures, is related to the critical flicker fusion frequency (CFF) of the eye of the animal. Critical flicker fusion frequency is determined by the relationship between the intensity of the source of illumination and its frequency; for humans, CFF frequencies range between 16 and 55 Hz, depending on mean luminance and contrast modulation. In addition to CFF values, video image perception may also be influenced by the range of spectral sensitivity, or colors that the eye and brain are sensitive to. Jumping spiders have a CFF value of approximate 40 Hz, so flicker at least should not be a problem. It is possible that the colors of the video image do not match the same colors that we perceive on a television, but my data indicate that the jumping spiders respond appropriately to color images of members of their own species.

Video playback is useful for testing questions related to visual perception, because it gives the experimenter complete control over the test stimulus. Although there are limitations, there are many advantages for using video playback as a tool for investigating questions of animal perception and behavior. Some of the most exciting possibilities of this technology is that, video images can be altered with desktop computer systems. Using computer animation and paint software programs, specific characteristics of a stimulus can be individually manipulated to determine what they mean to test subjects. Additionally, identical stimuli can be presented to all test subjects, thus eliminating the variability that may bias natural experiments with live test stimuli.

q I work at an imports store in Washington State and a few days ago I found a spider that looked exactly like the one on the show. He was small, with a black body and white legs. I have 3 questions...What is the name of that spider (is he poisonous)? Where is his natural habitat? And can he survive outside his natural habitat? Thank you. Galatea

A Galatea - Thank you for your questions. Without actually seeing the spider that you found, it is difficult for me to identify it based on your description. Most jumping spiders are small and many species have black bodies with white legs. Given your location, it seems unlikely to me that the spider you found is the same kind that I study. However, since you say you work at an import store, it is possible that you have found one of the tufted males that were shown on the show. These jumping spiders are commonly called the dimorphic jumping spider, and their scientific name is Maevia inclemens. They occur naturally in the northeastern USA, east of the Mississippi River, and their preferred habitat is old fields, generally near the woods. If the environmental conditions are suitable, that is, they aren't over-heated or chilled down and they have access to food and water, I am sure they could survive out of their habitat for a period of time. However, as adults, most spiders have a fairly short life expectancy and male Maevia live only about 12 weeks and females a bit longer. Like all spiders, Maevia is venomous, meaning that it has the ability to inflict a poisonous bite. However, they are not at all harmful to humans and in over 10 years of handling them, I have never been bitten (and even if I had been, it would not have been noticeable).

q Can you please provide some of your recent publications so I can learn more about your research? (asked by several viewers)


Here are some relevant publications:

Clark, D.L. and K. P. Stephenson (In Press) Response to video and computer-animated images by the Tiger barb (Punteus tetrazona). Environmental Biology of Fishes.

Clark, D.L., J. M. Macedonia and G. G. Rosenthal. 1997. Testing Video Playback to Lizards in the Field. Copeia, 1997(2): 421-423.

Clark, D.L. 1994. Sequence analysis of courtship behavior in the dimorphic jumping spider, Maevia inclemens. Journal of Arachnology, 22:94-107.

Clark, D.L. and G.W. Uetz. 1993. Signal Efficacy and the Evolution of Male Dimorphism, Proceedings of the National Academy of Sciences, 90, 11954-11957.

Clark, D.L. and G.W. Uetz. 1992. Morph Independent Mate Selection in a Dimorphic Jumping Spider: Demonstration of Movement-bias in Female Choice Using Video Controlled Courtship Behaviour. Animal Behaviour, 43, 247-254.

Clark, D.L. and G.W. Uetz. 1990. Video Image Recognition by the Jumping Spider Maevia inclemens. Animal Behaviour, 40, 884-890.

Roster, N., D.L. Clark and J.C. Gillingham. 1995. Prey catching behavior in frogs and toads using video-simulated prey. Copeia, 1995(2): 496-498.


Scientific American Frontiers
Fall 1990 to Spring 2000
Sponsored by GTE Corporation,
now a part of Verizon Communications Inc.