Mike often wonders out loud whether Toby really likes it when we pet him, or whether he just acts that way to get food.

An evolutionary biologist would say Toby's behaviors -- wriggling, wagging his tail, leaning up against us -- evolved over eons because they proved to be effective ways of getting humans to share their resources.

Maybe so. But it feels good to Toby, too.

J.S.J. Odendaal and R.A. Meintjes of Pretoria's Life Sciences Institute measured the blood levels of endorphins, oxytocin, prolactin, B-phenylethylamine and dopamine -- the chemicals of relaxation and pleasure -- and cortisol, the stress hormone, in both people and dogs before and after they interacted affectionately.

They compared levels of the neurochemicals three times: after people petted their own dogs, petted unfamiliar dogs and after they sat quietly and read a book.

In both humans and dogs, those chemicals of pleasure rose after five to 24 minutes of petting. In addition, the people's cortisol levels fell as they enjoyed their pets. The dogs' cortisol remained the same, probably because they found this new environment interesting and fun.

The increase in oxytocin was higher in the experimental group where people interacted with their own dogs.

Pet lovers will find this study especially interesting, because it lends scientific support to what we feel is true: Our dogs love us back.

When you define emotions as physiological states instead of psychological processes and then identify a similar state in human and animal, it only makes sense to say that man and dog feel the same.

Odendaal and Meintjes believe that oxytocin can be used as a measure of  "social attachment on an intraspecies basis."

In fact, buried in one of their papers is this rather revolutionary statement: "It becomes clear that some emotions are universal among vertebrates."

Emotions that we call love, that is.