approximately equal, and, like the cell throughout, are polygonal or hexagonal by reason of contact (Pl. A, Fig. 1). The incipient or expanding parts of cells are seen there as small tri-, or quadrangular openings at the angles between the older, larger cells. Crowding of the cells appears to prevent entire regularity in shape and size of each cell, and the walls are crenulate. In speaking of cell one includes for each the half of the bounding wall, although the walls originally are dense, amalgamated, calcareous structures, not double further than that they were built not only by increment upon their margins, but also on the two surfaces. Theoretically there is a boundary plane or division between cells midway in the wall. They frequently split midway in the wall when fractured. The walls in this species are thin and show little or none of their growth structure. At a slight depth within each cell the last growth increment or layer crosses the cell opening and forms a transverse partition or false bottom, the so-called tabula. Tabulæ occur successively in all cells more or less regularly, but not corresponding in neighboring cells. They are a little more numerous in the incipient part of each cell. The wall edge of one cell cannot extend above those of its neighbors, from which it never separates. The zoarium corresponds in structure to the following supposed manner of growth: It was covered when alive by numerous equal sized zooids which coalesced laterally, the lower part of each, however, extending into and secreting the walls forming the cell or zoœcium. Young zooids arose among them and, growing, extended relatively downward, building a new cell; the increase in number of zooids and, respectively, cells, being compensated by a necessary growth in radial length of cell wall to increase the surface of the inhabited zoarium. The tabulæ indicate successive planes where the bottom of the zooid rested between periods of necessitated self-extraction from the ever too long cell. E. O. Ulrich assumes that the cells are "Zoccia, directly superimposed upon one another so as to form long tubes 1 intersected by straight or curved partitions . . Variation in such species as this one would consist in the zoarium growing in one part more than another for some cause, being one-sided; or it is discoid because of perigene growth, i. e., the cell increase is greatest around the margin; or, again, it is acrogene. The surface upon which it grew affected its growth also. The cells remain nearly uniformly large, the number of young cells seen at the surface varying. The cell walls appear uniformly thin. Those characters are seen at the surface or in transverse thin section. Longitudinal section would show the tabulæ to vary, being fewest where cell length most rapidly increased, and the slightly differing rates of expansion of the cell initial. The atter character can also be estimated at the surface. Other species near to this one are in various ways more complex. Thus, Diplotrypa limitaris Ulr. has nearly the same manner of growth as Monotrypa magna, differing in smaller size of cells and shapes of their apertures at the surface (Pl. A, Fig. 3). Here the number of immature cells nearly equals that of the mature ones. Longitudinal section shows that the young or initial part of each cell is long, slowly expanding, or even for some length not increasing, then quite quickly becoming full-sized, mature (see Fig. 4). The tabulæ are closer in these initial parts, technically called mesopores. The plan of growth thus differs from that of Monotrypa magna in that in the latter new cells appeared only as fast or numerous as they were to develop into mature cells. In Diplotrypa the new cells appear too rapidly, so to speak, and await their turn to expand into full size; hence the more numerous small cells, mesopores, among the full-sized autocells at the surface. The calycal or open part of the cell is shallower in small cells or mesopores than in the larger ones, and that is true in all Monticuliporoidea. Also the autocells crowd the mesopores so that the former tend to become circular at the expense of the latter. Callopora multitabulata Ulr. began like a Mesotrypa, but acrogene growth obtained (Pl. A, Fig. 5), the zoarium being long, cylindrical, branched, arising from a basal discoid expansion. The tips, or apices, are only then like small Mesotrypa. The greatest growth of zoarium and cell increase was at the zoarial ends, and there the cells increase centrally, so that some were being crowded away and turned their apertures to the peripheral surface, i. e., away from the axis of growth (Pl. A, Fig. 6, c). The grown zoarium is thus composed of two regions, the axial or "immature" region of vertical cell part (Fig. 6, b, a), and the peripheral or "mature" of laterally directed cell (Fig. 6, c, d). In the peripheral region the cells grew slower in length, have thicker walls and more numerous tabulæ, and there are more mesopore cells. The apical parts also become finally slowgrowing, thick-walled, with many tabulæ and many mesopores, and it is evident that the zoarium grew rapidly to nearly full size; then a retarded or "mature" growth followed. Renewed rapid zoarial growth and a second retardation stage often occurs, wherefore the terms immature and mature regions are presumptious terms. Peripheral and axial regions are better, since they leave the degree of maturity to be described. Upon the thick-walled peripheral and apical part there occur at nearly regular intervals elevations called monticules. These are occupied by a small group of larger-sized, cells with mesopores or young cells. On the thin-walled or growing apex, and hence in the axial region, these are represented by a group of likewise slightly large-sized cells, with abundant mesopores, resembling less distinct cell groups occurring in Monotrypa, etc. Nicholson proved monticules to be points of greater cell increase, and while young cells may appear at any cell angle, their increase is greatest in the monticules or cell groups. In renewed rapid growth the peripheral monticules tend to develop into branches of the zoarium, but of course all could not. Thus monticules are similar to zoarial branches, but are not branches normally. Branching of the zoarium is due to double region of acrogene growth only. The tabulæ of Callopora multitabulata are thin, and the last one is near the cell aperture, and is said to be perforated at the center (vide Eastman, op. cit., f, 456 d). They They are nearly always solid. Right here is the chief supposed basis for the interpretation of Trepostomata as Bryozoa. According to E. O. Ulrich's definition (p. 271, Eastman), each tabula was the perforated top of one zoœcium and solid bottom of the next. In fact, the thickened walls here show only that the growth increments lengthening the wall continue on either side downward, thickening the wall, and thence as tabulæ across the cell opening. Further is not seen. Perforate last tabulæ may be incomplete ones. The characters for distinction of the species are, therefore, mostly visible on the exterior; the shape of zoarium and its branches, shape, size, and shallow depth of the cells, characters of the monticules, the number, size, and shape of mesopores, and thickness of the wall. All these characters vary, and the variation of all should be noted in learning the species. Extremes may be associated on parts of the same specimen. Prasopora simulatrix Ulr. grew upon some solid surface, at first lens-shaped, later conical (Pl. A, Fig 7), hemispherical, or irregular, expressing slight tendency to acrogene growth, united with moderate established perigene. A short finger-shaped or a branched sporadic acrogene growth occurs sometimes, and this usually at the center. If the colony died off in part, the remaining part then developed, overspreading the old. Even a symmetrical zoarium could develop from the irregular fragment of another. The cells radiate from the flat or concave lower |