MR. Gr. J. ROMANES ON THE LOCOMOTOR SYSTEM OF MEDUSAE. 
727 
such apparently complete physiological harmony is not coordination properly so called, 
i. e . is not due to special nervous connexions between the ganglia ; for, if such were the 
case, perfectly synchronous action of this kind ought to be the rule rather than the excep- 
tion. I am therefore inclined to account for these cases of perfectly synchronous 
action by supposing that all, or most, of the ganglia require exactly the same time for 
their nutrition ; that they are, further, of exactly equal potency in relation to the 
resistance (or excitability) of the surrounding contractile tissues; and that, therefore, 
the balance of force being exactly equal in the case of all, or most, of the ganglia, their 
rhythm, though perfectly identical, is really independent. I confess, however, that I 
am by no means certain regarding the accuracy of this conclusion, as it is founded on 
negative rather than on positive considerations. That is to say, I arrive at this conclu- 
sion regarding the cases in which such apparent coordination is observable only because 
in other cases such apparent coordination is not observable, and also, I may add, because 
my experiments in section have not revealed any evidence of nervous connexions capable 
of conducting a stimulus-wave with greater rapidity than a contractile wave. I there- 
fore consider this conclusion an uncertain one ; and its uncertainty is perhaps still fur- 
ther increased by the result of the following experiments. If a covered-eyed Medusa 
be chosen in which perfectly synchronous action of the ganglia is observable, and if a 
deep radial incision be made between each pair of adjacent ganglia, the incisions being 
thus eight in number and carried either from the margin towards the centre or vice 
versa , it then becomes conspicuous enough that the eight partially divided segments 
no longer present synchronous action ; for now one segment and now another takes the 
initiative in starting a contractile wave, which is then propagated to the other segments. 
And it is evident that this fact tends to negative the above explanation ; for if the dis- 
charges of the ganglia are independently simultaneous before section, we might expect 
them to continue so after section. It must be remembered, however, that the form of 
section we are considering is a severe one, and that it must therefore not only give rise 
to general shock, but also greatly interfere with the passage of contractile waves, and, in 
general, disturb the delicate conditions on which, according to the suggested explana- 
tion, the previous harmony depended. Besides, we have before seen that for some reason 
or other segmentation of a Medusa profoundly modifies the rate of its rhythm. In view 
of these considerations, therefore, the results yielded by such experiments must not be 
regarded as having any conclusive bearing on the question before us. And as these or 
similar objections apply to various other modes of section by which I have endeavoured 
to settle this question, I will not here occupy space in detailing them. It seems 
desirable, however, in this connexion again to mention a fact briefly stated in my former 
paper, namely, that section conclusively proves a contractile wave to have the power, 
when it reaches a lithocyst, of stimulating the latter into activity ; for “ it is not diffi- 
cult to obtain a series of lithocysts connected in such a manner that the resistance offered 
to the passage of the waves by a certain width of the junction-tissue is such as just to 
allow the residuum of the contractile wave which emanates from one lithocyst to reach 
