On June 10, 1880, it was noticed that the fresh water in the large tank of the lily-house of the Royal Botanical Society, Regent's Park, was swarming with a small and active species of Medusa, previously unknown to science--it being, indeed, at that time unknown to science that any species of Medusa inhabited fresh water, although it was well known that some of the other Hydrozoa do so. Examination showed that the new species belonged to the order Trachomedusæ, and the Petasidæ of Haeckel's classification--its nearest known relative, according to Professor Ray Lankester, being the genus Aglauropsis, which occurs on the coast of Brazil. The Medusa was called Limnocodium ([Greek: limnê], a pond, and [Greek: kôdôn], a bell) sorbii by Professors Allman and Lankester. I am indebted to the kindness of Professor Allman for permission to reproduce his drawing of the animal. (Fig. 31.) It is remarkable that, although this Medusa has reappeared every June in the same tank, no one has yet succeeded in tracing its life-history. Nor is it known from what source the tank first became impregnated with this organism. No doubt the germs must have been conveyed by the roots or leaves of some tropical plant that at some time was placed in the tank; but the Botanical Society has no record of any plant which can be pointed to as thus having probably served to import the organism. [Illustration: Fig. 31.] I shall now proceed to give an account of my observations on the physiology of this interesting animal, by quoting _in extenso_ my original paper upon the subject (_Nature_, June 24, 1880). Before doing so, however, I may state that Professors A. Agassiz, Moseley, and others have since informed us that sundry species of sea-water Medusæ have been observed by them living and thriving in the brackish waters of estuaries--a fact which strongly corroborates the inference at the end of the present paper.