When half-disks were employed it was necessary to make two incisions through the muscles and nerves in such a manner that an endless labyrinth of these tissues would be formed (fig. 4), in which the circuit wave of contraction could be maintained.
These series will be referred to as the "active and activated" series.
In the third type of experiments "activated" specimens prepared in the manner just described were compared with "inactive" specimens (fig. 5).
In all instances where active and activated specimens were compared it was observed that the pulsation-rate of the activated specimens was at first about 3 times as great as that of the active specimens. The difference in rate became progressively greater for about the first 24 hours, both on account of an actual increase in that of the activated as well as a decrease in the rate of those with sense-organs intact, until at the end of this time the activated specimen was often pulsating 10 times as rapidly as its mate. If an entire disk or unmutilated half disk were used as the active specimen of any pair its pulsations appeared to be much more vigorous than its mate, in the muscles of which a circuit wave of contraction was maintained, because the bell margin was folded inward onto the body of the disk at each contrac tion. When, however, a similar labyrinth had been formed in the subumbrella tissues of the active specimen and the continuity of the sheet of muscles had been destroyed, the bell-margin was no longer folded over and its pulsations lacked the appearance of greater vigor than that shown by the activated specimens.
When simultaneous kymograph records were made of the pulsations of the halves of a disk, one active and the other activated, each with an endless labyrinth formed in its subumbrella tissues, the amplitudes of contraction were found to be equal, showing that the apparent character of the contraction was a function of the normal spatial rela tionship of the muscle tissue and that when similar series of cuts were made in the active tissues of each half-disk of any pair the rate of pulsation was a true measure of the work done.
Besides a control series carried along as a check for each experiment, a large series of half-disks were prepared, so that each member of any pair (i. e., halves of the same disk) had been subjected to the same operation. Fifty pairs of such half-disks of the three operative types used were recorded for rate of regeneration and loss of weight during starvation, and 10 pairs for total metabolism, with the result that in every instance the similarly prepared halves of each disk gave results which were within the limits of error of the unit of measurements employed in each set of experiments. It is evident from these results that the medusa disks can be safely considered as physiological units which may be subdivided, with the expectation that like areas or masses of tissue of any one disk will respond in an equal degree to experimental changes in environmental conditions.