I will now describe the accident just mentioned, because it is closely connected with the main thread of my narrative. In the beginning of the Easter term, the third term of my training under Routh, I had caught up with my class and had spare time for outside reading. Niven was greatly impressed by my enthusiastic eulogies of Maxwell’s little book, “Matter and Motion,” and he suggested that I take up the reading of another of Maxwell’s little classics, “Theory of Heat.” It was written with the same elegant simplicity as his “Matter and Motion.” This little text-book on heat was the first to give me a living physical picture of the mode of operation by which heat is transformed into mechanical work, an operation which I had watched so often in the Cortlandt Street boiler-room. I had watched it, but I had never dreamed that the operation could be described as Maxwell described it. According to him it may be considered as the resultant action of non-coordinated activities of an immense number of busy little molecules, each of which, as far as human observers can tell, moves about lustily according to its own sweet will. But behold the miracle: the average activity of the countless crowd obeys with mathematical accuracy the fundamental law for heat transformations, the so-called second law of thermodynamics discovered by Sadi Carnot, the great French engineer. It was Maxwell’s little classic which also informed me that in all cases of very large numbers of individuals, whether they be active molecules or busy human beings, exhibiting as far as an observer can tell non-coordinated activities, we must apply the so-called statistical method of inquiry, that is, the method which statisticians employ in recording the activity of a nation. Newton’s dynamics, which at that time had been the food of Cambridge for two centuries, said nothing about that. It was a new idea in the heads of new men, who, under the leadership of Maxwell, were creating a new and far-reaching science. Up to that time Tyndall’s poetical description of “Heat as a Mode of Motion” was my gospel regarding thermal phenomena, but Maxwell’s plain and modest text-book, intended to stimulate the imagination of the inquisitive mind of the young student, was the first to assist me in forming my own judgment on the doctrines described by Tyndall and illustrated by beautiful experiments. Routh’s training-table of tripos athletes offered no such morsels of stimulating food, because these athletes were training for tripos examinations and not for research in physics. I will say now that it was in Maxwell’s theory of heat that I first saw the name of Willard Gibbs, and I heard from Niven that Maxwell held Gibbs in very high esteem. I must say also that Gibbs was the first in this country to write a splendid treatise on statistical mechanics. * * * * *