Mature catch tentacle
Note that the percentages of small and large holotrichs (respectively) in relation to the cross section cnidom were nearly identical among the six mature catch ten tacles used in this study (i.e., the standard deviation was small—seeFigure 9). Since the mature catch tentacle cnidom is made up almost entirely of small and large holotrichs, it is possible that the total number of large holotrich cnidocytes that differentiate from interstitial cells in the latter stages of catch tentacle development is related to the total number of small holotrich cnidocytes that were produced in
CATCH TENTACLE DEVELOPMENT 517
“¿ early― stages of catch tentacle development (i.e., in a fixed ratio of 1.0 large holo trichs produced per 1.3 small holotrichs). However, since the distribution ofthe two types of holotrichs is spatially distinct, with small holotrichs lining the epithelial surface and large holotrichs occupying the space beneath them, this ratio of large to small holotrichs might simply reflect optimal densities of each cnida type inde pendent of the other. Catch tentacles are particularly suited for demonstrating pos sible regulation of this type (or maximization of space utilization) because normal tissue depletion in catch tentacles includes tentacle tip autotomy (an intermittent, all-or-none phenomenon).
Catch tentacle regression
The removal ofcatch tentacle cnidae (small and large holotrichs) from regressing catch tentacles is concurrent with the appearance of feeding tentacle cnidae. The holotrichs are not transported to the tentacle coelenteron (to become “¿ gastrodermal cnidae―)during catch tentacle regression, as feeding tentacle cnidae are during catch tentacle development. Instead, large holotrichs move from the middle epithelium to the epithelial surface alongside the small holotrichs, indicating that both types of holotrichs are probably expelled from these tentacles externally. The source of the mature feeding tentacle cnidae that appear in the tips ofregressing catch tentacles is not yet known, since cnidoblasts were absent from the tips of the regressing catch tentacles used in this study.