In the summer of 1965, a female Dalmatian was stolen from a farm in Pennsylvania. Her story changed America.

A quiver went through Clayton's whole body as I took in this sight, and then a stream of liquid erupted from his groin, gradually building like a fountain that's just been switched on. An arc of urine splashed against my shoulder—and suddenly the monkey room was bedlam. Macaques began to throw themselves against the walls in a cacophony of shrieks and crashes. One animal in the upper tier started doing back flips; his neighbor stepped toward the front of his cage, turned in my direction, and started urinating, too.

I placed Clayton's small body into his cage, locked the door, and retreated to the safety of the hallway. The postdoc who had assigned me this task smiled as I peeled off my wet lab gown and T-shirt. "Don't worry," he said, as if it happened all the time.

Just as zookeepers rarely share the names of their animals with the public, so are laboratory monkeys left anonymous in the science literature. If I'd had the opportunity to publish the results of my work with Clayton, we would have called him Monkey C, in accordance with journal etiquette; other mammals, like mice, rats, and kittens, are almost never identified, even in code.

That hasn't always been the case. Ivan Pavlov called his surgical dogs by name in published lectures. Among his most successful subjects was a collie-setter mutt named Druzhok, "Little Friend." The anti-vivisectionist movement was much stronger in the United States than it was in Russia, though, and American physiologists were soon hiding the more sentimental details of their work from the public. In 1914, the chair of the Council on the Defense of Medical Research, Walter Cannon, warned journal editors to excise from their manuscripts any "expressions which are likely to be misunderstood" or turned against them by animal activists. Historian Susan Lederer has traced the expansion of this policy over 25 years at the nation's top biomedical research journal. Starting in the 1920s, she writes, a slew of technical jargon was systematically inserted into the pages of the Journal of Experimental Medicine. The word starving was replaced by fasting, bleeding by hemorrhaging, poison by toxicant; full-body photographs of lab animals were removed, and the pronoun it was subbed in for any use of he or she to describe them. Authors who referred to their animals by given names were instructed to use a string of letters and numbers instead.

That doublespeak (by now having become a matter of habit) obscures some of the incidental cruelties of animal research. But it hides just as well the attention and care that are essential to working in the lab. An experimental macaque costs about $8,000 and may require months or years of training before it can start producing useful data. That is to say, its continued health is of extraordinary value both to the professor who paid for it and to the graduate student whose dreams of a thesis depend on its well-being. It was my job to nurture Clayton so he would perform in my experiments as best he could. Given the constraints of the lab—a cage, a chair, a metal head post—I wanted him to be as happy as a monkey could be.

Outright negligence might have affected the quality of my data, as an animal in distress is likely to deliver skewed results. That idea, so obvious in retrospect, dawned on physiologists only near the turn of the 20^th century, according to historian Otniel Dror. Researchers began to notice how fear or anxiety could be expressed as physiological phenomena—changes in blood sugar, for example, or digestive function. A fearful rabbit might "blush," wrote one scientist, and yield false measurements of blood pressure. While journal editors of the 1920s worked to strip emotional phrases from the scientific literature, scientists learned how to control emotion in the lab. Walter Cannon, whose letter in 1914 inaugurated the era of science-journal jargon, remarked that he could alter the gastric motility of a cat by "reassuringly" stroking her fur.

I've also experimented on cats—kittens, really—by probing their exposed brains with an electrode to see where tiny shocks might palpitate their feet. (We were studying neuroplasticity and how behavioral training affected the development of the motor map.) I spent time with the animals every day, teaching them to grab morsels of meat from a plastic container with their little paws. Like Walter Cannon, I stroked their bellies, too, and scratched under their chins. But there's no mention of those affections in the published results of the study. (Kittens "were trained to reach through the aperture to grasp the beef from a narrow cylindrical food well (3.2 cm inside diameter; 5 cm deep) using their preferred limb only," we wrote.) Nor did we mention that the animals—some as young as 3 months old—were euthanized at the end of each "intracortical microstimulation" experiment.

It's easy to see why we used this furtive language. Any sentimentality over the cats would have suggested a lack of scientific rigor, and a frank description of the killings would only invite anger from animal rights groups—and alienate the taxpayers who paid for the study (and my graduate student stipend). But it seems to me the pressure to keep the laboratory door shut comes from both sides. The public acceptance of animal research, and the biomedical breakthroughs it engenders, has always come with the understanding that no one will divulge too many of the gory details—we put up with animal sacrifice only so long as we don't have to think about it.

On Sept. 11, 1981, police officers in Montgomery County raided the two-story Institute for Biological Research in Silver Spring, Md., and found there a gruesome, filthy holding room for experimental macaques "who were in such physical and mental stress that they appeared to have bitten off their fingers and arms, or whose cages were locked together so that they injured each other." With the help of a young animal rights activist named Alex Pacheco, the officers seized 17 of the animals, and the lab's director, Edward Taub, was charged under state law with more than a dozen counts of animal cruelty.

Taub's trial began in October 1981, and as happens in nearly every case of alleged laboratory animal abuse, the ugliness of invasive research became a defense in itself. Should the condition of Taub's monkeys have been taken as evidence of abuse on its own terms or in the context of how research monkeys were treated everywhere else? Expert witnesses debated every detail of the case along these lines, from the question of how filthy a monkey lab might reasonably become to whether it made sense to bandage the wound of a deafferented animal. Taub was found guilty on six counts, but five of them were overturned in a second trial the following year; he was acquitted of the sixth in 1983.