It may here be observed that to produce these results, both carbonic acid and oxygen must be considerably diluted with air, for otherwise they have the effect of instantaneously inhibiting all response, even to the strongest stimulation. When this is the case, however, irritability returns very soon after the tissue is again exposed to air or to ordinary sea-water. But I desire it to be understood that the results of my experiments on the influence of oxygen, both on the natural and on the artificial rhythm, have proved singularly equivocal; so that as far as this gas is concerned further observations are required before the above results can be accepted as certain. I have still one other observation of a very interesting character to describe, which is closely connected with the current views respecting ganglionic action, and may therefore be more conveniently considered here than in any other part of this treatise. I have already stated that in no case is the manubrium of a Medusa affected as to its movements by removal of the periphery of the swimming-bell; but in the case of Sarsia a very interesting change occurs in the manubrium soon after the nectocalyx has been paralyzed by excision of its margin. Unlike the manubriums of most of the other Medusæ, this organ, in the case of Sarsia, is very highly retractile. In fresh and lively specimens the appendage in question is carried in its retracted state; but when the animals become less vigorous--from the warmth or impurity of the water in which they are confined, or from any other cause--their manubriums usually become relaxed. The relaxation may show itself in various degrees in different specimens subjected to the same conditions, but in no case is the degree of relaxation so remarkable as that which may be caused by removing the periphery of the nectocalyx. For the purpose of showing this effect, it does not signify in what condition as to vigour, etc., the specimen chosen happens to be in; for whether the manubrium prior to the operation be contracted or partially relaxed, within half an hour after the operation it is sure to become lengthened to a considerable extent. In order to show the surprising degree to which this relaxation may proceed, I insert a sketch of a specimen both before and after the operation. The sketches are of life size, and drawn to accurate measurement (Figs. 29 and 30).[29] [29] I may here mention that the fact of the manubrium of Sarsia undergoing this extreme elongation after the removal of the marginal ganglia, serves to render the artificial rhythm of the organ under the influence of injury, as previously described, all the more conspicuous. [Illustration: Fig. 29.] [Illustration: Fig. 30.]