III. Pictet converted D’Orbigny's families into tribes, and introduced some additional genera created by Giebel King, etc.; and, except in the description of new genera and species by Reuss, Roemer, Salter, Eichwald and others, the subject remained very much where D’Orbigny left it until M. de Fromentelle proposed a new arrangement, based upon what he terms the “organs which serve for the nutrition of the sponge,”—viz., the tubule, oscules, pores, etc. Like D’Orbigny, he divides the sponges into two orders: 1st, the Spongitaria, which comprises only recent genera; and, 2nd, the Spongitaria, which contains all the fossil genera, with the exception of the doubtful group, the Clionidæ. The second order is further divided into three sub-orders: 1, those sponges which have one or more tubules (the Spongitaria tubulosa); 2, those that have oscules, but no tubule (Spongitaria osculata); and 3, those that have neither tubule nor oscules (Spongitaria porosa). Each of these suborders is further divided thus: the tubular sponges into those in which the tubule is solitary, and those in which it is grouped, and also into those with oscules and those without oscules. The oscular sponges are similarly subdivided, according to form, disposition of the oscules, and presence or absence of an epitheca. Lastly, the porous sponges are divided into those that are more or less regularly cup-shaped, and those that assume some other form.