information on.the operation of gas absorption equipment is at present quite inadequate to permit proper estimates and designs for new operations. Further progress in this field can be made most effectively through a study of the mechanism of absorption phenomena and of the factors which control them. When the basic principles have been definitely established, they can then be applied to specific problems by the engineer in the same manner as he now employs heat transfer coefficients, friction factors, and the like.The object of this paper is to outline certain of the more fundamental conceptions which have been developed from the available data and from the general laws of reactions between two phases.

No abstract

Film coefficients of heat transfer have been determined for three petroleum qils and for water flowing through the tube of a horizontal double-pipe heat exchanger. The experiments cover both heating with steam and cooling with water under similar conditions. The flow was in the turbulent or semiturbulent regimes. Linear velocity varied from 1 to 20 feet per second, viscosity ran from 0.5 to 55 centipoises, and the heattransfer coefficients on oil covered a range from 10 to 700 B. t. u. per hour per square foot per degree Fahrenheit.The data for oils and water can be correlated on the basis of Nusselt's theoretical equation, which states that 9he heat-transfer coefficient is a function of the velocity, * . . . . . LTHOUGH the heating and cooling of liquids in pipes are fundamental to many industrial operations, A scarcely any investigations of the heat-transfer rates have been published which consider liquids other than water. Most of the data on the heating of water have been correlated by McAdams and Frost* in the form of an equation which adequately handles this important case. The chemical engineer, however, deals with many other liquids and isinterested in cooling as well as heating them. In the petroleum industry, especially, the rational design of equipment such as pipe stills, heat exchangers, and coolers demands a reliable method for predicting heat-transfer coefficients for oil flowing in pipes. The work described in this paper was undertaken to supply a basis for a general determination of such coefficients. I n order to make sure that the experimental method and apparatus gave results in accord with those of others, a few runs were made heating water, but the greater part of the work was with petroleum products having a wide range of viscosities and in half of the experiments the oils were cooled. Theoretical ConsiderationsThe modern treatment of heat flow through a series of resistances by evaluating each resistance independently of the others is now so generally accepted that its discussion here appears unnecessary. Following this principle, the problem of heat transfer with liquids in pipes devolves into a study of the coefficient between the liquid and the pipe itself, disregarding the particular heating or cooling phenomena which occur on the other side of the pipe. Coefficients of this sort are often designated as "film coefficients," a term derived from the convenient concept of fluid films at the boundaries of a fluid.The variables which might be expected to govern the film coefficient for liquids flowing in turbulent motion in pipes are the diameter of the pipe, the density, specific heat, viscosity, and thermal conductivity of the liquid, and the velocity of flow. In a given piece of experimental apparatus the first is fixed. Furthermore, the four physical properties of the fluid are not mutually independent but are determined by the nature of the particular liquid and by its temperature.

To the many science teachers who have been active in developing general science from its early beginnings to its present state, the authors make full acknowledgment for much help and inspiration. Acknowledgment for illustrative material is made in the text. The drawings were nearly all made by Mr. F. M. Wheat, of the George Washington High School, New York. The following teachers have carefully read the entire proof and made many valuable suggestions :

No abstract