A man stands at the front of a lecture hall with a container of talcum powder. He removes the cap, shakes the container a few times, and tiny particles of powder disperse visibly in the air. Then he invites members of the audience to inhale. Do you do take him up on that offer? Or do you decline, on the offchance that those aren't just talcum powder particles, but rather, something potentially more harmful?

According to Physics Today's Jim Dawson, this is one of Andrew Maynard's favorite visual demonstrations when he talks about the health risks and concerns potentially associated with nanoparticles. (If you have a subscription to the magazine, you can read Dawson's September 2006 article here. UPDATE: There was also a good article on Wired.com in October about a recent FDA conference on nanotechnology.) Maynard works for the Woodrow Wilson Institute, and while he didn't use this particular demonstration in this morning's talk on managing potential risk and protecting human health and the environment in the age of nanotechnology, the point it is intended to illustrate was front and center: namely, that we need to develop a nanotechnology oversight policy built on a firm scientific basis, to best protect ourselves from any potentially harmful impacts of nanotechnology without unnecessarily impeding further developments in this booming field.

It's easier said than done because of the sheer complexity and scope of the field: do we regulate consumer products like stain-resistant pants, shoe insoles containing nanoparticles, iPods, or giant magnetoresistant hard drives? How about food additives or novel water purification systems? Is it even possible to devise a general model or strategy for regulating this chaotic research area when it's so incredibly diverse?

Maynard seems to be saying. "Well, yes and no." On the one hand, he insists that any approach to nanotechnology oversight must be discussed within the context of specific applications. On the other, he does seem to think that some general principles can be stated that can then be extrapolated to the wide range of nanotech applications. For one thing, he thinks it possible to set a few boundaries for oversight along two basic criteria: oversight is needed for (1) nanomaterials capable of entering or interacting with the body in potentially harmful ways; and (2) nanomaterials which exhibit biological activity that is dependent on its nanostructure.

These two criteria incorporate such things as nanoparticles (aerosols, powders, suspensions and slurries), and also agglomerates or aggregates of nanoparticles that may (or may not) retain their unique small-scale structural properties even when grouped. Maynard is particularly concerned about the potential impact of unintentional use: a stain-resistant nanotech tie might be perfectly safe as a piece of apparel, but what if a toddler decides to start sucking on the tip of it? Does this enable nanoparticles to pass into the body?

Research into potential health and environmental risks of nanotechnology is still in the early stages, but Maynard was able to cite a few case studies involving lab rats. For instance, there is some compelling preliminary experimental evidence that nanoparticles of titanium oxide and barium sulfate -- which are normally quite chemically inert --can cause inflammation of the lungs in rats. There's also been some consumer concern about the use of nanoparticles in sunscreens and make-up: is it possible for tiny nanoparticles to penetrate the protective dermis of the skin? Experiments suggest that the skin is actually a pretty effective barrier. But Maynard cited a recent experiment involving quantum dots which suggests that altering the surface chemistry of the dots in very specific ways can increase the chances of penetration. His point: the size of nanoparticles isn't the only feature that matters when it comes to assessing potential health risks.

The truth is, there's so much that we just don't know about nanomaterials and how not just their size, but their structure, surface area, surface reactivity, and other properties influence both their behavior and impact. So how can we even begin to realistically assess risk and take responsible and appropriate action? Maynard sees promise in a model used by the pharmaceutical industry -- which frequently must assess potential risk on the basis of incomplete knowledge -- called "control banding." If it were possible to create an "exposure index" based on, say, the "dustiness" of a given nanoparticle and the amount used, as well as an "impact index" taking into account the bulk hazards, surface area, shape, size, and surface activity of that same nanoparticle, then these two indices could be plotted to provide some kind of objective assessment tool for determining potential risks.

Several audience members expressed skepticism regarding Maynard's suggested approach, on the rationale that no conceptual or qualitative approach can be successfully deployed in an area that -- by Maynard's own admission -- pretty much requires that risk assessments be made on a specific case-by-case basis. So the jury is still out on that front.

Maynard isn't the only person concerned with health risks, of course. He was followed by Michele Ostraat of DuPont, who described the Nanoparticle Occupational Safety and Health Consortium (NOSH), launched in 2004. DuPont is a founding member, as is Procter & Gamble, Dow, and Intel. Today, there over 16 companies and organizations worldwide participating in the consortium. Its purpose is to begin filling in the "knowledge gaps" via a collective effort.

There are numerous stated goals and "deliverables", most notably the performance of aerosol chamber experiments on silicon dioxide nanoparticles to determine things like the rate of particle diffusion and coagulation and the life cycle of the nanoparticles (i.e., how long do they pose a risk, if at all). That aspect of characterizing the aerosol behavior of a well-understood nanoparticle over time is nearing completion.

More challenging is the planned development of a portable air sampler device capable of detecting and measuring the presence of aerosol nanoparticles in the atmosphere. Such a device would need to be low cost, simple to operate, handheld (easily portable), and robust -- i.e., it must function not just in a controlled laboratory environment to detect carefully engineered nanoparticles, but also in the "real world," which most likely has a strong "background" of pre-existing airborne nanoparticles that will make reliable detection and measurement much more challenging.

About the only thing that is clear is that some assessment of potential risks should be undertaken -- because scientific research, particularly in the nanotech area, simply doesn't take place in a vacuum. By definition, it has a societal impact. We can take a lesson from history: the discovery of X-rays at the turn of the last century. X-rays found immediate application in medical imaging - indeed, the technology is still used today, with appropriate safeguards -- but initially users ignored early warnings that perhaps there were risks associated with this wondrous invisible form of ionizing radiation.

People embraced the new technology without fully understanding it. A dean at Vanderbilt University lost all his hair after sitting for a radiograph of his skull in February 1896, while a Scottish researcher who routinely used a fluoroscope to test the quality of X-rays by holding his hand between the tube and the fluorescent screen developed tumors and lost both his hands. And in 1904, Thomas Edison's assistant, Clarence Dally, died from over-exposure to X-rays -- carefully documenting his burns, serial amputations and diseased lymph nodes for posterity so that others need not suffer the same fate.

Hopefully history won't repeat itself when it comes to the nanotechnology revolution.