MUSEUM OF COMPARATIVE ZOOLOGY. 107 
cavity is less marked than iu the first protovertebra, but the cells are 
quite uniformly arranged into the two mesodermal layers. It may be 
noted, in passing, that while the pericardial and body cavities are con- 
tinuous at an early stage, they are not developed simultaneously ; that 
the pericardial cavity is well advanced before any trace of the corre- 
sponding split, or body cavity, occurs in the pectoral region. 
All the protovertebre caudad from the second are distinctly cut off 
from the lateral layers. In the region of the fifth (Fig. 14) and for some 
distance caudad from it, the lateral layers (so’plu., spl’plu.) end more 
abruptly at their distal margins, while in the proximal region there is a 
tendency on the part of the cells — due apparently to rapid prolifera- 
tion—to become aggregated, and thus to render somewhat obscure 
the boundary between the two lateral layers; meanwhile the proximal 
margin of the lateral mesoderm (Fig. 14) shifts its position ventrad along 
the side of the protovertebre. 
Still farther back in the region of the sixth or seventh protovertebra, 
the proximal region of the lateral mesoderm is distinctly cut off from the 
more lateral portion (Fig. 15, c/-mo. im.) ; as will be seen later, it has a 
very interesting and significant fate. At this stage the nuclei of the cells 
of the complete protovertebra are clearly arranged in a peripheral 
position (Fig. 14, pr’vr.). Rarely nuclei are observed in the interior. 
If, however, the plane of the section is not parallel with the transverse 
dorso-ventral plane of the protovertebra, or chances to pass through its 
anterior or posterior wall, the section shows nuclei which are apparently 
in the middle of the protovertebra. These are in many instances the 
nuclei of cells forming its anterior or posterior wall; but in other cases 
they do not represent cells in the peripheral walls, for the condition in 
which all the cells have this position is very transitory. Immediately 
following this stage (Plate III. Figs. 17, 18), nuclei are observed which 
are unquestionably in the middle of the protovertebra. The nuclei 
which are thus found in the interior of the protovertebra no doubt have 
their origin in the peripheral cells, and usually are undergoing division 
rapidly at this stage. Not only do the peripheral cells thus contribute 
to the interior of the protovertebra, but also to the increase in the super- 
ficial extent of the peripheral wall, and in many cases it is not difficult 
to determine from the direction of the nuclear division whether the con- 
tribution is to the wall or to the interior. In Figure 18 (Plate IIT.), on 
the left side of the embryo (right in the figure) in the dorsal region of 
the protovertebra, is an example of nuclear division (ni.'), in which one 
of the daughter cells is destined to pass into the interior; in the proto- 
