360 CUMIKGS AND GALLOWAY ^EORPPIOLOGY OF TREPOSTOMATA 



sphndens) , a member of the Integrata, the grov/ing edge of the wall is 

 very thin, and the wall becomes gradually thicker farther in, finally re- 

 ceiving a secondary deposit, the cingulum. Because of this method of 

 growth the laminte of the wall have a very steep pitch, and the bend they 

 make in tlie axis of the wall is sharp. On the other hand, the growing 

 edge of the wall in Dckayia (figure r)!) is smoothly rounded, and the 

 laminae pass across from zooecium to zoct'cium Avitli a regular curve. For 

 some reason, wlicrever the wall laminre of tlie trepostomes are sharply 

 beiit the material appears dark. Tliis is true also of sharp bends of dia- 

 phragms (figures 1, 9, 11, 44, 45, 4G, 47). It is probable that the size 

 and arrangement of the minute graiiules of wliicli the wall laminte are 

 composed differ slightly at such points fTom the normal size and ar- 

 rangement in other parts of the wall. In fact, in well preserved material 

 and in very thin sections it can he shown that this is actually the case. 

 Usually the granules are too small to show under the magnifications in- 

 dicated in these figures. In some species, as in Bytliopora gracilis, 

 Heterotrypa prolifica, etcetera, the granules can l)e distinctly seen under 

 a 4 mm. objective and 10 X ocular. Figure 42 shows very clearly their 

 appearance in Bythopora. In the latter genus, and in the Amalgamata 

 generally, these larger granules are distributed in more or less concentric 

 bands from face to face of tlio wall, or they are distributed in a broad 

 zone in the central portion of tlie ^\i\\\. In the Integrata commonly, and 

 occasionally in the Amalgamata. tliey are more closely concentrated in 

 the axial region of the wall, and when bands of granules from either side 

 of the wall are present they are often offset instead of continuing unin- 

 terruptedly across the median region of the wall. Such an arrangement 

 can be seen in figures 44 and r)(i. The granules of the lighter appearing 

 portions of the walls are so minute tliat they can scarcely be seen under 

 any feasible magnification. It may l)e remarked that these minute gran- 

 ules of the trepostome wall probably represent each an individual cell of 

 the original ectosarc, since we have good reason to believe that the cal- 

 careous deposits of the zooecial walls of Bryozoa are made intracellularly. 

 There is no good evidence that either type of wall described above was 

 double except in the sense that any Ijryozoan wall is double, namely, be- 

 cause it is secreted bv the juxtaposed or coalesced ectosarcs of two ad- 

 joining in(li\i(liials. Xor does the analogy of the Cyclostomata, tlie order 

 nearest related to the Trepostomata, lead us to expect a double wall in 

 the latter order. To say that the boundary between adjoining zooecia is 

 obscured in the Amalgamata does not go far enough. There is no such 

 boundary. The wall was one and single and common to two zocecia, as 

 it is in the Cyclostomata (Levinsen, 20). On the other hand, such ap- 



