What drew me to study whales and elephants was in each case an interest in their vocal behavior. The songs of humpback whales-long, complex, and beautiful to human ears-are fascinating because they change progressively with time, with all whales in a population adhering to the latest version of the song. Whales in different ocean basins sing different songs, but all have, broadly speaking, the same basic structure-the same rules of form, and the same rules of change. The changes proceed fastest in the height of the breeding season, when singing is most intense, suggesting that song changing may be an aspect of courtship (as changing fashions are for people); we know that the whale songs function in competition between males, as in many birds. During periods when a population's song is complex (lasting sometimes several years), the song contains rhyme-like material, but not when the song is simple, suggesting that rhyming may help whales remember what they hear, as is the case in humans.

I've been helped in this research by Roger Payne, Peter Tyack, and Linda Guinee. The work involved time at sea, recording and watching whales, and about 15 years poring over tapes and spectrograms. The songs provided quick clues to the structure and migrations of widely dispersed humpback whale populations. There is conservation value in knowing about these things, and also in realizing the extent to which whales are acoustic animals, animals whose normal lives may be threatened by human-generated noise in the ocean. There is reason to be especially concerned about the lowest-voiced whales, the fin and the blue whales, as their life pattern appears to involve communication over very long distances.

In 1990, I discovered, with W. L. Langbauer and E. Thomas, that elephants use infrasound-sound below the frequencies that human beings can hear-in communication. Our discovery was sparked by an observation of mine in the zoo. I noticed a shuddering in the air, every now and then, near the elephant cages. Years earlier I had noticed the same sensation when standing near an organ, and had realized that I was feeling the organ's lowest notes better than I was hearing them. This led me to guess that elephants might be making sounds too deep for me to hear.

Infrasonic communication had not been recognized before in land mammals. The finding was especially interesting because very low frequency sounds travel exceptionally far. We now have some evidence that elephants use their deep powerful calls to coordinate mating and ranging behavior over rather long distances. The organization of these enormous herbivorous animals into a complex widespread society-the ability of males and females to find each other for mating, and their ability to coordinate the movements of their large groups in such a way that they do not exhaust their resources-appears to depend on their capacity for producing and hearing low-frequency sound.

The elephant studies have taken various associates and me to Kenya, Namibia, and Zimbabwe. The Zimbabwean project, led by Dr. Bill Langbauer, enabled us to show some relations between elephants' infrasonic communication, genetic relationships, and uses of the land. We worked with African associates, whose knowledge of wildlife was humbling to scientists taught to depend on instruments and numbers more than on observation. It has been inspiring to live among people who know the inter-dependence of nature and humanity from immediate experience, and who base many aspects of their sacred, as well as their secular, behavior on this knowledge.