216 DONALD WALTON DAVIS 
yet none can be seen between the members of pairs of incom- 
plete mesenteries. One of my specimens indicates this. It was 
described, while living, as a triglyphic individual with three old 
orange stripes and eleven on the boundaries, or within, the new 
area (fig. 41). Sections show normal positions for the three old 
stripes. The sections also demonstrate that the regenerated 
area is composed of an older and a newer part. The three 
division planes are all in complete endocoels. Including the 
bounding endocoels there are in the new tissue eight complete 
endocoels. In the older regenerated part there are three pairs 
of incomplete mesenteries, all of the first grade, no mesenteries 
of a lower grade being present anywhere in the regenerated 
portion. In the most recently formed part there are four pairs 
of incomplete mesenteries. The relative position as well as the 
number of the orange stripes makes it probable that the eleven 
new orange stripes occupied all of the complete endocoels and 
the incomplete endocoels of the original piece and of the older 
regenerated area only. It is probable that new orange stripes 
would later have appeared in the incomplete endocoels of the 
first grade of the newest region. Another regenerating specimen 
showed externally ten orange stripes, Whereas sections revealed 
six pairs of complete mesenteries and six pairs of incomplete 
mesenteries of the highest grade. It is probable that two of the 
second grade of incomplete endocoels in the new region lacked 
orange stripes at the time of observation. I have no clue as to 
which of the endocoels lacked the stripes. 
The formation of orange stripes first in the complete and later 
in the first grade of incomplete endocoels would correspond with 
the order of development of the mesenteries of the regenerating 
region. This order of development of the stripes is in harmony 
with the statement of Davenport ('03, p. 143, and fig. 2) that 
new orange stripes appear between old ones. It cannot be too 
positively stated, however, that this formation of new stripes, 
as well as the production of new complete mesenteries, is con- 
fined to regenerating regions, and that the process is strictly a 
determinate one. After the formation of the group of complete 
mesenteries heretofore described (pp. 185 to 192) and of the first 
