Another suggestion made by Child,[157] is that there exists an axial gradient in the stem whereby the cells regenerate the more quickly the nearer they are to the oral pole. If this were correct, and we cut a long piece from the stem of a _Tubularia_ and bisect the piece, the oral pole of the anterior half should regenerate more quickly than the oral pole of the posterior half. According to the writer’s observations on a Tubularian (_T. crocea_) growing in the estuaries near Oakland, California, both oral ends regenerate equally fast in such cases. [157] Child, C. M., “Die physiologische Isolation von Teilen des Organismus,” Roux’s _Vorträge und Aufsätze_, Leipzig, 1911. 4. The phenomena of regeneration in _Cerianthus membranaceus_, a sea anemone, can be easily understood from the experiments on Tubularians, if we imagine the body wall of _Cerianthus_ to consist of a series of longitudinal elements running parallel to the axis of symmetry of the animal from the tentacles to the foot. The number of these elements may be supposed to correspond to the number of tentacles in the outer row of the normal animal. Each such element behaves like a Tubularian, with this difference, however, that the elements in _Cerianthus_ are more strongly polarized than in _Tubularia_, and that each one is able to form a tentacle at its oral pole only. This fact can be nicely illustrated in the following way: if a square or oblong piece (_a b c d_, Fig. 24) be cut from the body wall of a _Cerianthus_ in such a way that one side, _a c_, of the oblong is parallel to the longitudinal axis of the animal, tentacles will grow on one of the four sides only; namely, on the side _a b_.[158] (Fig. 25.) The other three free edges are not able to produce tentacles. If an incision be made in the body wall of a _Cerianthus_, tentacles will grow on the lower edge of the incision (Fig. 26). [158] Loeb, J., “Untersuchungen zur physiologischen Morphologie der Tiere.” [Illustration: FIG. 24] [Illustration: FIG. 25] [Illustration: FIG. 26]