as to the rehabiUty of the latter that gave planktonologists so much concern during the last quarter of the nineteenth century. It seems pertinent, however, to remind the reader that a tow net, of the usual conical form, and of mesh appropriate for the capture of planktonic organisms, filters somewhat less than the amount of water that would pass through a simple hoop of equal diameter, the loss (i.e., the amount regurgitated) depending on the shape of the net, on the proportionate areas occupied by the threads and by the spaces between the latter, and on the pressure, i.e., on the velocity with which the net is drawn through the water. The slower the towing, the more complete filtration, as every planktonologist knows from observation. In the case of fine meshed silk (e.g., #20) , the loss may be so serious at ordinary towing speeds as to necessitate special types of net, to increase the filtering surface relative to the mouth opening. And even with silk as coarse meshed as #0 (38 meshes per linear inch), such as used in the fore parts of our nets, only about 30% as much water would pass through a given area, drawn transversely at 2 knots, as through an open cyhnder of the same area, according to the relationship between pressure and filtration in Hensen's (1895) experimentsT his loss is so minimized by the increase in straining area relative to mouth opening, resulting from the conical form, that it probably averaged less than 10%, at the usual towing speed for our nets, though experiments at various hands have proved that the catches made in parallel hauls with similar nets may vary as much as 10-30% or even more (see Winsor and Walford, 1936, p. 190, for a recent discussion). Neither is clogging likely to be serious, for nets such as those used, except on rare occasions when diatoms or Phaeocystis may swarm, or when the gelatinous bodies of ctenophores, appendicularians, etc., may block the meshes. Much more serious is the uncertainty as to what relation the fraction of the plankton that is adequately sampled bears to the remainder that the net fails to capture, it being common knowledge that no one type of net will equally well sample the various size categories of planktonic animaLs. The nets used in the present studies (meshes, 38-54 per linear inch) being comparatively coarse, the failure of small copepods such as Oithona, or the young of others, to figure more largely in the following volumetric lists, does not necessarily mean that ' Pressure at the given velocity calculated from the formula, .