Competition is intensifying in the race to develop the next generation of industrial-strength fibers. These new fibers could be used to make special medical sutures or perhaps even body armor for law enforcement officers. The strongest of the new fibers might not be something man-made, but rather a natural material that's been around for a long time: silk. Scientists are debating whether silkworms or spiders are best equipped to help turn these materials into commercial products.

Most of the commercial silk developed today is harvested from silkworms, which have the scientific name of Bombyx mori, as they spin the cocoons from which they will later emerge as moths. The silk fibers from those cocoons can be used in clothing and other textiles. Since silkworms are friendly vegetarian insects, it is relatively easy to farm them and harvest their silk. Typically, silk produced by a silkworm is not extremely strong, but it is suitable for regular textiles.

Many scientists consider some types of spider silk, the glandular excretion the arachnids use to construct their webs, to be among the strongest natural fibers on earth. The problem is that spiders cannot be farmed like silkworms. Spiders are carnivorous, and any attempt to gather spiders in close quarters usually results in the spiders attacking and eating one another. This makes it nearly impossible to gather enough high-quality silk from them directly.

So scientists are looking for ways to genetically engineer artificial spider silk in the laboratory. One of these scientists is David Kaplan, a professor in the Department of Chemical and Biological Engineering at Tufts University in Boston. His technique involves growing bacteria to do the job for him.

"Either we build a synthetic gene, or we clone part of a gene from a spider, for example, and then we transfer that gene into a bacterium. The most common bacterium is E.coli. Then we place that bacterium in a vat of liquid that's conducive to growing large numbers of that organism. After a period of time, those bacteria are collected, the cells are broken open, and we basically purify the silk that's been produced from that bacterium. The silk in that case is in a powder form."

The spider silk powder is then mixed with water. From this liquid, Professor Kaplan explains, it's possible to start extracting fibers, essentially, silk proteins.

"If you take the liquid silk and put for example, a toothpick in there and then pull the toothpick out of that solution, you will form a fiber," he said. "It illustrates the point that these proteins in solution, in the gland of the silkworm and in the spider, they want to make something out of themselves. They're not just happy being like that. They want to make a fiber. So it's easy to make a fiber, it's hard to make a good fiber."

David Kaplan and his students at Tufts have experimented with extracting fibers at different rates of speed, but they say that at this point, spiders still create the best fibers.

And they note that their technique for manufacturing spider silk with E. Coli generally produces only a tiny pinch of material, enough for Professor Kaplan's research, but too little for commercial silk production.

That's where the Canadian biotechnology firm, Nexia, comes in. The Montreal, Quebec-based company bills itself as the world leader in recombinant, or genetically engineered, spider silk production. Like David Kaplan's approach, Nexia uses genetics to help produce spider silk proteins, but they derive it from a much more plentiful source.

"To get more spider silk, you need to go to the farm," said Jeffrey Turner is the President and CEO of Nexia Biotechnologies.

"And what we've done is successfully genetically modified or added one highly characterized spider silk gene to goats. The goats have a background of 70,000 goat genes. They look like goats, act like goats, but when they produce their milk as they started to a few months ago, they actually produce spider silk protein in their milk."

Mr. Turner won't say how much silk protein he can harvest from a liter of milk. But he does say that if demand were to require more silk than his existing facilities could handle, he'd simply breed more goats. Why use goats instead of cattle? Mr. Turner prefers goats because of their rapid growth and short reproduction cycle.

"Goats are a wonderful animal," he said. "They normally produce twins and their gestation period is five months. So basically in a normal dairy type of environment, you'd be able to have three kiddings every two years. Oh, let's say two to three offspring per female. So if you were looking for a thousand offspring, you'd basically need a thousand or so female goats and that's easily procured. We have well over one thousand goats at Nexia today."

After the goats are milked, the spider silk protein must still be extracted from the goat milk. Mr. Turner says that's easily accomplished with today's standard dairy equipment. Once the raw spider silk protein is separated, the big challenge is to properly extract or spin a fiber from that raw protein. Mr. Turner claims that his company is close to making industrial-quality fibers, but won't give any specifics yet. He will say that by carefully observing exactly how a spider spins its web, Nexia can learn to do the same.

In an interesting variation on this approach, other scientists have tried adding spider silk genes to traditional silkworms instead of goats. According to Fritz Volrath of Oxford University in England, initial experiments weren't successful, because he says the most important factor in creating quality silk is how that silk is spun or extracted. On that point, he and Nexia executive Jeffrey Turner agree. But Mr. Volrath argues that silkworms can already produce silk nearly as strong as a spider's.

"Like the spider silk, the spinning conditions affect it. Curiously enough, the silkworm silk that we reel from a cocoon is very much inferior to the silkworm silk that we reel from the silkworm, by holding the silkworm and pulling it straight out of its mouth onto a reeler, depending on the conditions. If we get the conditions right, then the mechanical characteristic of that silk is much better, not as good as spider silk, but it approaches it."

Mr. Volrath's findings were published recently in the journal Nature. If his initial trials are successful, his next step will be to observe whether silkworms can spin better fibers on their own, without the need to extract the silk by hand.