[Illustration: FIG. 39½.--_A._ Aboral view of _Gonionemus vertens_. _A¹._ Side view of same. Dotted line in each indicates where jelly-fish was cut into halves. _B, B._ New individual from a half. As seen from above and from the side. _C, C¹._ New individuals from a ¼ piece. As seen from above and from the side. _D._ New individual from a piece less than ¼. It contained a part of one of the radial canals. A new proboscis with mouth regenerated in all pieces, but no new canals or tentacles.] assume the form of a bell, but the missing parts are not replaced.[60] I have worked on the same form and obtained substantially the same results. If the jelly-fish is cut in two, as indicated by the dotted line in Fig. 39½, _A_ and _A¹_, each half closes in and assumes the form shown in _B, B_. Each new jelly-fish has only the two original radial canals that each half had when separated from the other. A faint line along the region of fusion of the pieces seems to represent a new radial canal,--it is not represented in the figures,--and each half-proboscis has completed itself. There are not formed any new tentacles, except perhaps one, or a few more, where the cut-edges meet. Thus there is actually very little regeneration, although the typical jelly-fish form is assumed by the half-piece. If a jelly-fish is cut into four pieces, each piece containing one of the radial canals, the pieces also assume the bell-like form, as shown in _C, C¹_. A new proboscis develops from the proximal end of the old radial canal, and since this end is often carried to one side during the closing in of the piece, the new proboscis lies not at the top of the sub-umbrella space, but, as seen in the figure, quite to one side. Pieces even smaller than these one-fourth jelly-fish will assume the bell-like form, especially if they contain a bit of the margin of the old bell and a part of one of the radial canals, as shown in Fig. 39½, _D_. Although I have kept these partial medusæ for several weeks, and have fed them during this time, I have found that the missing organs do not come back. That these pieces do undergo a certain amount of regeneration is shown by the formation of a new proboscis, and, in certain cases, a new radial canal. Even the tentacles may be partially regenerated, as Hargitt has shown,--especially, as I have found, if the margin of the bell is cut off very near the base of the line of tentacles. Small knobs appear along the cut-edge, but the pieces die before regeneration goes very far. If, however, the margin is cut off in only one quadrant, new tentacles may be produced along the cut-edge. CHAPTER VII PHYSIOLOGICAL REGENERATION. REGENERATION AND GROWTH. DOUBLE STRUCTURES.