in their article &dquo;Swelling Capacities of Fibers in Water: Part II: Centrifuge Studies&dquo; [9] described an interesting new freezing technique for measuring moisture gradients.They also discussed some of the data given in a recent paper of ours [7]. On the basis of this discussion they arrived at some deductions which call for comment. The theoretical treatment of Welo et al.has points of similarity with ours. It differs principally in considering the capillary water among fibers to be present as &dquo;aggregates,&dquo; the volume of which is proportional to the cube of the linear dimensions of the air-water-fiber line at which surface tension acts. This consideration, among others, leads to two deductions which differ from ours. It introduces a two-thirds exponential into the effects of the centrifugal acceleration and, more important, it eliminates any consideration of a length factor independent of the dimensions of the air-water-fiber line which supports the capillary water. In our treatment we considered this independent length factor, which was effectively the dimension of the sample of fibers in the direction of the centrifugal acceleration. Our treatment starts from the assumption that in assemblies of hydrophilic fibers there are continuous water columns throughout. The basis for this assumption is well established. The earliest reference to it was made by Socrates in Plato's Symposium [3] . It is a traditional method of siphoning and filtering liquids from vessel to vessel [2]. An ordinary cotton cloth left hanging over the side of a bucket of water demonstrates the presence .of continuous columns of water in the cloth; whether the cloth is crumpled or straight, after a short time the contents of the bucket are on the floor. In a rerent paper [6] it was shown that suction forces are capable of maintaining continuous columns of water even when opposed by forces equivalent to very large centrifugal fields. There is an elementary physics experiment which shows that capillary rise is independent of the straightness or crookedness of the capillary in the direction of the gravitational field, as shown in Figure 1. It is thus irrelevant whether the fibers, which have capillary channels between them, are convoluted or straight. The only important dimensions are: (1 ) the effective length of the air-water