Which leaves you in the uncomfortable position of speculating on a change that cannot happen unless you already know what it will be (or have the sort of time on your hands that natural selection has). That is the crux of the layman’s trouble in understanding evolution: it works on a time scale so far beyond personal experience that one must train oneself to think in eons, not decades. Here at Fortunate Fields, we have carefully defined objective criteria known in advance by which our animals may be measured for fitness; we know exactly which behaviors to select for. Therefore, though our progress must be slow, we are confident it will also be steady.

Since you insist on speculating, however, I will say this much. There are limits to what even the most rigorously scientific breeding program can accomplish-based not only on the foundation stock and the limits of precision we have for measuring the dogs, but on limits that come from within us-limits, in other words, of our own imagination, and of ourselves as conscientious human beings. In the end, to create better dogs, we will have to become better people.

And that, sir, is the last speculation you will hear from me along these lines.

There was more in the letter, an exchange of kennel techniques, clarification of older letters. What interested Edgar was how Alvin Brooks had signed the letter. To move such a formal man from that first outraged letter to one in which he closed with “Affectionately” must have taken dozens if not hundreds of exchanges. Why had these, of all the letters, been kept? Probably chance, he knew. But he dug back into the file cabinets to see what else he could find, opening letter after letter and setting it aside.

It all made him think about their records. The paperwork on a dog didn’t end when it left the kennel. At his father’s request, new owners sent back letters every few months describing how the dog they’d adopted was getting along. When a dog reached five years of age, his father had contacted the owners to fill out another form. And when the dog died, yet another form was filled out, recording the age, cause of death, behavior late in life, and so on. Edgar’s father sometimes even called the dog’s veterinarian. As a result, the file for every dog expanded over time until it was stuffed with notes, letters, photographs.

“A litter,” Edgar’s father once told him, “is like an x-ray of its parents and its parents’ parents, but an x-ray that takes years to develop, and even then it’s faint. The more x-rays you have, the better the picture you get.”

This made sense. A dog might sire half a dozen litters, each with six or seven pups. That meant forty or more pups who reflected the qualities of the sire. If, for an extreme example, cleft palates showed up in every litter-meaning pups that had to be put down-you knew that the sire carried a propensity toward cleft palate. (Of course, if a sire produced cleft palates more than once, they stopped breeding him, and a red slash was drawn across the dog’s folder.)

Just before a dog left the kennel, it was evaluated one last time. They called this “the finish.” There were no special tests in the finish, just the same things they’d always tested for, the same exercises, the same measurements. The difference was, this time the results were rolled up into a numerical score representing the dog in maturity. That finish score was the best indication of what its ancestors had passed on-the best x-ray.

Once a finish score was assigned, Edgar’s father recalculated all its ancestors’ finish scores, going back five generations. This was how evidence accumulated about how true the dog bred, how reliably it passed on its qualities, good or bad, to succeeding generations. A second number told them how many of a dog’s progeny had contributed to the master score-an index of confidence. When planning litters, the choice between two dogs with nearly identical finish scores favored the dog with the highest index of confidence-the one who had been tested most thoroughly. This was a system Edgar’s grandfather had worked out and refined, apparently in long discussions with Brooks, and which his father had practiced, modified, and improved.

It wasn’t a perfect scheme, of course. While the finish score gave an idea of how well a dog fared in testing, there were intangibles to consider. Not temperament, which they broke down into individual behaviors and assessed, and not physical qualities, which were easily measured, but how the dog combined all these things, for the whole of every dog was always greater than the sum of its parts. Some, for example, seemed capable of inspiration: they clued in on a new way of doing things more often than others. There was no way to measure this. And there was the dog’s personality, which was distinct from its temperament. A dog with a keen sense of humor would find ways to make jokes with you, and could be a joy to work with. Others were serious and contemplative, and they were good for other reasons.

Edgar’s father had sometimes grumbled that all he did was keep track of a dog’s faults, though what he meant was, even the best records in the world couldn’t capture the whole of the dog. They could record only what could be measured. And the measuring and testing of the dogs, the follow-up calls and letters, the reassessment of the ancestors of a placed dog, all served to remind his father of a dog’s total character. When it came time to plan a litter, the scores and numbers were only a guide. It hadn’t been unusual for him to select against the numbers based on intuition.

But his father’s complaint also pointed to the fact that the records mainly prevented bad pairings-breeding, say, two dogs that tended to produce weak fronts. That was the interesting thing about planning a cross. Two brilliant dogs couldn’t be bred if it risked a litter full of stifles so straight the dogs would be crippled by the time they were five years old. And so the first question about any potential pairing was not how great the offspring would be, but what problems it might produce.

Thinking about all this, Edgar began to understand what his mother meant when she’d claimed not to have the words to describe what made their dogs valuable. Partly it was the training. They spent long hours doing crazywalking, stays, releases, shared-gaze drills, and all the rest until the pups paid attention to where they were going and where they were looking; they learned that a certain expression on a person’s face meant that something interesting lay behind them, or in another room. He’d taken that for granted, but now that his mother had pointed it out, he saw how uncommon this was.

So a dog’s value came from the training and the breeding. And by breeding, Edgar supposed he meant both the bloodlines-the particular dogs in their ancestry-and all the information in the file cabinets. Because the files, with their photographs, measurements, notes, charts, cross-references, and scores, told them the story of the dog-what a dog meant, as his father put it.

Sometimes when Edgar got an idea, a whole series of other ideas clicked into place right behind it, as if they had logjammed somewhere in his mind, waiting for the way to clear. Suddenly, he saw how the training, the breeding, and the record-keeping worked together, how the training tested the dogs for their qualities, their ability to learn different kinds of work. That explained the training notes and why the Sawtelles had to raise the dogs to maturity: if they placed a pup, they wouldn’t know what kind of dog it became. But the Sawtelles could compare them because they trained every dog. So it made sense that a dog’s finish score could alter the scores of its ancestors, which in turn influenced the dogs used for the next mating. As if every dog had a voice in selecting the following generations.

Edgar closed his eyes and waited until he could hold it all in his mind, and once that happened, he wanted so badly to ask his father about it, to be sure he’d understood things correctly, that it nearly made him cry. But the only way left to him was through the records. And yet-he felt this, but couldn’t find the words for it-something else made the dogs valuable, too, something that hadn’t been among this sudden cascade of ideas. He wished he could read his grandfather’s side of the correspondence to understand what he’d meant by “the next dogs.”