746 
ME. G. J. ROMANES ON THE LOCOMOTOR SYSTEM OE MEDUSJE. 
secondary movements persist ; for instead of “ generally ceasing after a few moments ’’ 
or “hours,” they usually last for several days — in fact until the tissues begin to 
decompose. I am able, however, to agree with Dr. Eimer’s statement that, other 
things equal, the unmutilated tissues preserve their irritability longer than those 
from which lithocysts have been removed ; and I am able to agree with his state- 
ment that the rate of the natural rhythm has a tendency to bear an inverse pro- 
portion to the size of the individual, though I find that size is not the only factor 
in determining such rate. But his further statements with regard to the duration 
of the natural pauses bearing a direct proportion to the number and strength of the 
previous contractions, require, I think, to be taken with reserve ; for both in Aurelia 
and in Sarsia, although a general relation of this kind may be observed, I have not 
found it to be so precise as Dr. Eimer affirms. Similarly, although it is true that the 
smaller segments of a divided Aurelia show a general tendency to contract less frequently 
than the larger segments, I have not found evidence of the precise relation which Dr. 
Eimer describes. On the other hand, I have found evidence of two additional factors 
which in these segmentation experiments are as important in determining the rate of the 
rhythm as is the size of the segments. These factors are (a) the potency of the litho- 
cysts and ( b) the time that elapses after the segmentation has been performed ; for at 
first the smaller segments, if they contain prepotent lithocysts, have a quicker rhythm 
than the larger segments, but afterwards the rhythm of the former progressively slows 
more rapidly than does that of the latter. Although, however, the endurance is thus less 
in the case of a prepotent lithocyst in a small segment than is the endurance of less potent 
lithocysts in a large segment, the endurance of the former towards deprivation of oxygen 
is greater than is that of the latter; for, under such circumstances, the prepotent litho- 
cyst will continue to act rhythmically in water which is stale enough to cause the less 
potent lithocysts to act irregularly, or not to act at all. 
Other forms of mutilation — such as cutting off the polypite or portion of the swim- 
ming-bell — causes, first acceleration of the rhythm, and then a progressive decline to a 
certain point below the original rate. The rate then remains stationary at this point, 
but may again be made temporarily to rise and permanently to fall by removing another 
portion of the swimming-bell. In these experiments the rhythm, besides becoming 
permanently slowed, is also often rendered permanently irregular. Again, paring down 
the contractile tissues from around a single lithocyst has the effect, when the tissue is 
greatly reduced, of giving rise to enormously long periods of inactivity. During such 
a period, however, stimulation may initiate a bout of rhythmical contractions, to be fol- 
lowed by another prolonged pause. These facts tend to show that the apparently 
automatic action of the lithocysts is really due to a constant stimulation supplied by 
other parts of the organism. 
Temperature exerts a profound influence on the rate of the rhythm. This influence 
may be best observed within moderate limits of variation; for water below 20° suspends 
spontaneity and even irritability, while water above 70° permanently slows the rhythm 
