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 Leticia Avilés Leticia Avilés as seen on Spiders!: Amazon Tales

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

q How big can the colonies of spiders get, and how many spiders can live in the web? Students at WAS

A That depends on the species we are talking about. Anelosimus eximius, the species shown in the show, produces perhaps the largest colonies of any social spider. I estimate that the largest nest I have seen (actually not quite the one shown in the show) measured over 25 feet in length, 6 - 8 feet in width, and 4 - 5 feet in height and may have contained over fifty thousand spiders. In other species, the colonies may grow to contain only a few dozen individuals.

One question one may ask is: What determines that maximum nest size in the various species? There are several factors that may play a role. These factors include the architecture of the webs, the type of vegetation the webs grow on or around (e.g. whether it is continuous or not), and the amount of food that can be caught at any one site given the surface to volume ratio of the web and the abundance of insects likely to get caught in it. Some of these factors may cause webs to fragment, others may affect the fitness of colony members. In the latter case, we may expect that individuals may develop a tendency to disperse when their colonies grow too large. Dispersal would then reduce population size and limit colony size.

q How many feet high and wide is the biggest spider web in the world? Jeremy

A I estimate that the refuge area of web shown in the show -- the part of the nest that surrounds the vegetation, which is where offspring are kept and spiders spend most of their time -- measured around 15 x 9 x 5 feet. In addition to this refuge area, the web consisted of a prey capture snare that extended over 20 feet from the refuge area towards the vegetation above. As I mentioned in the previous answer, however, occasionally I have seen A. eximius webs that are even larger.

q Why do some spiders live in groups and some don't? Davey

A An excellent and difficult question to answer. I will try my best, although this is a question spider biologists don't yet have an answer for.

There are at least two factors that need to be considered: (1) The type of spiders that have developed sociality (and those that haven't) and (2) the kind of environments they live in. You probably need the right combination of both of these for sociality to arise (or to be maintained, as sociality can also be lost).

In my answer to an earlier question, I listed the species that have a similar level of sociality as the species in the show. In these species, individuals live together for their entire lives, may mate with each other, and may take care of their offspring communally (we call these non-territorial permanent-social species). With only two exceptions (both in Australia), all these species make webs and the webs they make are irregular (sheet or tangle webs) rather than orbicular. Webs may be important because (1) they are costly to produce; (2) they are discrete physical structures that may hold individuals together in one place; and (3) they facilitate communication. Because webs are costly to produce, individuals may benefit by sharing a web. This may be especially true in habitats where webs are destroyed frequently by strong rains. Web sharing, however, may only be possible with irregular webs, as orbicular webs have been designed to be built and utilized by a single individual. Orbicular webs also may not be as costly to produce as irregular webs. Because web-sharing will also have its costs -- you may have to share your food not just your web -- what may really matter is the balance between costs and benefits of web-sharing. The denser (i.e. more costly) the type of webs produced, the stronger may be the benefits of sharing them. The more food there is available for nestmates at any one place, the lower may be the costs of sharing a web.

Species with the level of sociality as the one in the show are also usually related to species in which the mother takes care of her offspring for some period of time. This suggests that the various ancestral species from which different social spider species originated may have also displayed maternal care. This is an important possibility because in species with maternal care siblings live together in the same web for a fraction of their lives. So, part of the work that would be required to produce a species where individuals are together all their lives has already been accomplished. All that is needed is extending this period of sibling tolerance and cooperation through the adult lives of the spiders.

Irregular webs that can be shared and long-term maternal care, therefore, may be preadaptations that facilitated the evolution of permanent-social behavior in some spider lineages. In the absence of these two factors, sociality may not arise easily or not at all. This would put a lot of spider groups out of the sociality game. Lets now consider the other player in this game which is the environment.

The environment may play a role by either providing incentives for sociality or by placing insurmountable obstacles. As we already discussed, sharing your web can have the drawback of also having to share your food. It may be that in the temperate regions, where permanent-social spiders are absent, prey densities are not high enough to support more than single individuals or a group of young siblings sharing a web. In tropical areas, on the other hand, it may be that not only there are abundant prey year around, but also that much of that prey are large enough to require a group of individuals to capture them. These large prey, which are a wasted resource for solitary spiders in those regions, would provide an incentive for sociality. Another environmental factor already discussed are strong rains that may destroy webs frequently and, thus, select for web-sharing. Not all social spiders occur in rainforests, however. In some dry areas in Africa where social spiders exist, the selective pressure may be predation. John Henschel, a scientist living in Namibia, for instance, found that spiders living in groups are better able to defend themselves from ant raids than spiders of the same species living by themselves.

I am in the process of exploring some of the ideas discussed above. Most of these ideas have not been formally tested and many will not be easy to test. As in most scientific inquiries, putting the pieces of the puzzle together will require much creativity and insight to come up with plausible hypotheses, imagination to design appropriate tests for these hypotheses, a lot of dedication to carry them through, and the work of not one, but many scientists scattered around the globe and across the generations.

You can find additional information on social spiders in the following article: Aviles, L. 1997. Causes and consequences of cooperation and permanent-sociality in spiders. Pages 476-498 in: J. Choe and B. Crespi, eds., "Evolution of Social Behaviour in Insects and Arachnids," Cambridge University Press, Cambridge.

q What is the name of the species of spiders you are studying in colonies in Ecuador? Were you the first person to discover them? (asked by several viewers)

A The species is Anelosimus eximius, in the same family (but a different genus) as the black widow. No, I was not the first to discover them. The species was described by a famous French arachnologist, Eugene Simon, at the end of last century. However, during the time I have worked in the rainforest of Ecuador, I have discovered a few social spider species that either were not known at all (e.g. Tapinillus sp.) or that were not known to be social (Aebutina binotata, Theridion nigroannulatum). When you work in a place with such great diversity as the rainforest, and where so few people are working, you are almost certain to discover things that nobody has seen before.

q In the episode "Amazon Tales" I heard that the spider you are studying is one of seventeen species that lives in colonies, but you never mention any of the other species. Could you tell me the name of some of the other spider species that live in colonies and where they are located? Mark

A All the species that have a similar level of sociality as Anelosimus eximius are tropical. There is one species in Mexico (Mallos gregalis), six in Ecuador and other regions of the Amazon (Anelosimus domingo, Anelosimus rupununi, Theridion nigroannulatum, Aebutina binotata, Tapinillus sp), one in Paraguay (Anelosimus lorenzo), four in Africa (Agelena consociata and Agelena republicana in Gabon and Stegodyphus dumicola and Stegodyphus mimosarum in southern Africa), one in India (Stegodyphus sarasinorum), two in New Guinea (Achaeranea wau and Achaearanea vervortii), and two in Australia (Diaea socialis and Delena cancerides).

As you see, these species belong to different genera and most of these genera belong to different spider families. Species that belong to different genera, and certainly those that belong to different families, probably represent different independent origins of social behavior. In fact, it appears that these 17 species represent at least 12 independent derivations of this type of social behavior in spiders.

q How did you come to love spiders so much and decide to spend your career studying them? I love spiders too. I just always have been fascinated with them. Katie

A When I was an undergraduate in Ecuador, I had to observe a social spider colony as part of a project in a field biology course. As I did this, I realized that the spiders were very easy to work with, but very little studied (a very large book on social organisms I had just read contained only a page and a half on social spiders).

At this time I also became intrigued with why there were ten times as many females as males in the social spider colonies. From reading the same book on social organisms, I realized that this could be very interesting from the point of view of evolutionary biology. However, for this to be interesting, females had to be produced in greater numbers, rather than males just living for a very short time. I showed that indeed more females were produced as part of my undergraduate thesis project.

At the same time, I came up with a possible explanation as to why they did this. It turned out that the explanation was just as interesting and intriguing as the original question. So, I decided to develop a computer model to show that my explanation would actually work. This I did as part of my doctoral dissertation at Harvard University, while continuing to study the spiders in the field (to validate the model and test its predictions).

Since then, as I continue to work with the spiders, every time I come up with an answer, I find more questions, every one as interesting as the first. So, I am never done. In fact, I now have so many questions that I have students help me answer them. That is what is so exciting about science. There is always something new to discover and something new to explain. It never gets boring.

q What do the spiders you study eat in the forest? And what eats them? Jetin

A The spiders usually eat insects that fly through their webs and get caught in them. The spiders may be eaten by other spiders and by insects such as spider-eating wasps, ants, or praying mantises. Perhaps some vertebrates, like monkeys or some birds, may also occasionally prey on them.

q The spiders that live in the same colonies mate within the colony, and you said they are all family. How come they don't have birth defects like humans or other species mating within family? Jen

A Very good question! When humans and many other species mate with relatives, their offspring tend to exhibit genetic defects. The reason for this is that relatives may have identical copies of defective genes that could be passed on to their offspring (organisms usually have two copies of each kind of gene, one inherited from each parent). When these individuals mate, some of their offspring may inherit the defective copy from both parents. These offspring would express the genetic defect. This is the usual situation.

In the inbred social spiders, however, we believe that the problem has been reduced significantly because they have been mating with relatives for thousands or millions of generations. At the beginning, when the spiders just became social and started to mate with relatives, offspring with genetic defects would have been produced. Those offspring, however, would have died or left fewer descendants. As a result, every generation there would have been fewer and fewer individuals carrying copies of the defective genes. Eventually, the defective genes would have been almost completely eliminated from the population. So, nowadays, offspring carrying two copies of some defective gene are unlikely to be produced.

Inbreeding (i.e. mating with relatives) and lack of mixing among colonies, however, may still be a problem for the social spiders. Inbreeding results in individuals that carry two identical copies at most of their loci (plural for locus, the site where a gene coding for a particular trait sits). Lack of mixing among colonies results in individuals within a colony being very similar to each other. With so little genetic variability, the individuals and colonies should be more vulnerable to disease and parasites than in regular (i.e. outbred) species. Also, with so little variability the social spider populations may be unable to evolve fast enough to respond to environmental change. Perhaps for these reasons there are so few social spider species, even though this type of behavior has arisen independently at least 12 times in spiders.


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