In “Solar Wars,” David Biello discusses the impact of the rooftop solar energy “boom” in the U.S., including electric utilities' worries about lost revenue.

He omits the question of the future stability of large-scale power systems, which will be radically affected by increasing local energy generation. New types of load imply a greater risk of feedback-induced oscillatory behavior.

Ensuring the stability of the grid while creating a system that can deal with two-way local power flows and less predictable generation requires effective technical coordination across multiple parties. A one-system approach rather than one that looks at only parts of the power network system in isolation would seem necessary.

University of Lancaster, England

The article had one important fact missing: a household solar power system that is connected to the electrical utility network cannot serve as backup source of power if the utility service fails unless the homeowner has arranged for battery backup.

Tucson, Ariz.

Using solar power panels for air conditioning, as Doug Cox is described as doing in the article, is hardly making “the bucolic lifestyle of the suburbs sustainable.” Apart from the required regular maintenance, the panels had to be constructed (using petroleum products), assembled on a line and transported to his house. Dumping any environmental benefits gained from using solar panels back into air-conditioning seems perverse.

via e-mail


In “The Case for Kill Switches” [Forum], Jonathan Zittrain argues that implementing ways to remotely disable military weapons if they fall into the wrong hands deserves serious consideration.

One of the most valuable uses for such kill switches in weaponry would be in land mines. It is well known that leftover mines remain hazardous for years to innocent civilians, animals and vehicles. To the extent that the military insists that these mines are an essential weapon, they could be equipped with kill switches that would be activated remotely or even automatically after a given period of time.

via e-mail

Kill switches should become a compulsory but secret add-on for all weapons. Once everybody has become dependent on their imaginative and efficient ways of carnage … hit the kill switch.

Hemhofen, Germany


“Mind of the Meditator,” by Matthieu Ricard, Antoine Lutz and Richard J. Davidson, really captured my interest because I have been a member of a Zen sangha for about five years now.

But I was disappointed that the question of selection bias in the meditation studies the authors refer to was not discussed. It appears to me that most, if not all, of the subjects observed in most of the studies had been regular meditators or at least would have shown a positive disposition toward meditation. So, the question arises: To what degree are identified differences in brain structure, brain dynamics, behavior and temperament that correlate with varying levels of “meditation practice” a cause and to what degree an effect?

Montclair, N.J.


“Virus Therapy for Cancer,” by Douglas J. Mahoney, David F. Stojdl and Gordon Laird, describes research on using engineered viruses to infect and destroy human cancers. According to the authors, one of the reasons that such viruses are effective is that they damage the tumor cells' camouflage from the immune system, allowing it to do its work and finish the job.

I have thought that if I were unlucky enough to have an incurable cancer, I would ask for a graft of a tumor, as similar as possible to mine, from another person. My body would start to reject the grafted tumor because it was foreign but in the process might notice the damaged cells were also cancerous. Once my immune system was primed in this way, it might move on to start destroying my own cancer cells. Is there any mileage in this idea?

Devizes, England

MAHONEY REPLIES: What you have described is similar to an approach called allogeneic cancer cell vaccination, in which cancer cells from one patient are irradiated with ultraviolet light (to kill them) and then grafted into another patient in an attempt to elicit an antitumor immune response. It has shown some promise in clinical trials and—like other cancer vaccines—will almost certainly benefit by rational combination with some of the new immune-targeting drugs recently approved by the U.S. Food and Drug Administration.


In “Do No Harm—And No Drugs” [Science Agenda], the editors argue that random drug testing should be mandatory among health care workers, citing the substance abuse rates of physicians.

Perhaps the reason drug abuse is prevalent among physicians is that drugs are perceived as performance enhancers: no doctor wants to be seen as drowsy or moody.

I worked in the construction business for years, and we never used drug tests. A supervisor can tell when an employee has a hangover, is lacking sleep or has cut a board short, but a drug test cannot. Drug tests do not test for impairment and cannot substitute for proper supervision.

Phoenix, Ariz.


There is an interesting slipup in “Killer Chairs,” by James Levine [The Science of Health]. The article states that “when you combine all causes of death and compare any group of sitters with those who are more active, sitters have a 50 percent greater likelihood of dying.” The active group members have a 100 percent chance of dying during their lifetime, so this would mean that the sitters have a 150 percent chance of dying.

Obviously, that is nonsense, so there must have been some time factor omitted.

Amherst, N.H.

THE EDITORS REPLY: Indeed, that sentence should have mentioned a time period. The overall differences in mortality between sitters and more active people were observed after both groups had been followed for about 8.5 years.


“Inside the Audience Studio,” by Katharine Gammon [Advances], incorrectly refers to McMaster University as being in Toronto. It is in Hamilton, Ontario.

“The Evolution of Architecture,” by Rob Dunn, incorrectly refers to Jesse N. Weber and Hopi E. Hoekstra as creating a backcross of hybrid mice with oldfield mice. The hybrid mice were crossed with deer mice.