spread over a layer of warm air, resting quietly upon it with the help of an intervening diaphragm, and the latter be removed, we all know that the cold air must descend and the warm air rise—a process of overturning such as is occurring every day in the atmosphere. The mechanical conditions or mechanical theory of this upsetting were recently worked out by Margules, and his views, with some important modifications, are developed by Professor Bigelow in such a way that a certain conclusion is inevitably reached. This overturning takes place not merely in a small way, as in thunderstorms, but on the grandest scale in tropical hurricanes. Now the question has been discussed pro and con for a hundred years as to whence comes the energy involved in the production of the rapid rotary winds of hurricanes. Espy maintained that in thunderstorms this energy was derived from gradients due to the condensation of aqueous vapor and the evolution of heat in the clouds. I thought it due also largely to the sun's heat acting on the top of the cloud. Professor Bigelow shows that while these are true causes, yet for hurricanes they are entirely insufficient, and that the energy of these great storms is mainly derived from the gradients produced by the overturning of layers of cold air flowing from northern latitudes over the warm air that is flowing from southern latitudes; by the descent of this cold air to the ground the force of gravity gives it great velocity and momentum. In other words, we must not look upon a great storm as a symmetric cyclone with a center of warm rising air and an inflowing pericyclone of cold air, as was taught by former meteorologists, but we must face the problem of a simple overturning in the lower strata of the atmosphere below the level of the general west wind that is flowing a few miles above us. The ideal cyclone and anticyclone probably do not exist in the atmosphere. This conclusion gives precision to an idea that Ferrel fully acquiesced in, namely, that the atmosphere has no simple circulation, cyclonic or anticyclonic, but is a complex mass of interlacings of currents; so that the progress made by himself in studying ideal types must sooner or later be replaced by researches that adhere more closely to the actual phenomena of nature.

CONCLUSION.

The resolution of problems bearing on the mechanics of the earth's atmosphere is stimulating the efforts of the world's best men, and illustrates the stage to which meteorology has attained in its progress toward being an exact science. Some portions of meteorology are already as exact as our knowledge of chemistry, optics, physics, or astronomy can make them; other parts are still in an unsatisfactory condition, which, of course, is also true of every branch of knowledge. We must congratulate our colleague on the contributions that he has made along lines of research that will help the next generation of students to a more thorough knowledge of laws that will eventually become the basis of satisfactory long-range forecasts. It will always redound to the credit of the Weather Bureau to have encouraged and published such work in this difficult field.

Equally creditable to America is the conception and establishment by the Chief of the Weather Bureau of a special research observatory at Mount Weather, where for the first time in the history of meteorology the researcher has been separated from the observer; and a special institution provided for him. This seems like the realization of an idea contained in a paragraph in my address at Indianapolis in 1890: "Why found new colleges and universities to teach what is already taught elsewhere? Exploration is the order of the day. Give us first the means to increase knowledge, to explore nature and to bring out new truths. Let us perfect knowledge before we diffuse it among mankind, so that what we teach may with every coming year be nearer and nearer the eternal truth of God's creation."

This exhortation is as applicable today as then. Meteorology is not yet properly recognized in our colleges, nor as a postgraduate course in our universities. The science has progressed, but the universities have not kept up with it. Laboratories have been provided for chemistry, physics, psychology, wonderful observatories for astronomy and elaborate establishments for mechanical engineering, but a laboratory for the experimental study of the motions of the atmosphere has not yet been provided, although the men who could conduct it are ready and anxious to begin the good work that they see before them.