424 MEMOIKS OF THE NATIONAL ACADEMY OF SCmNCES. 



pi'oti'uded, while the rest are iiiturncd. The proeess of inti'over.sion and .subsequent extension 

 has been aetually observed on the expanded polyp. The movements of the tentacular walls, 

 inwardly or outwardly, take place so slowly that they can be readily followed, the process some- 

 what resembling that seen when a glove finger is indrawn and afterwards pushed out. The 

 alternations ma}' be continued for some time. Likewise on i-etraction of the polyps the tentacles 

 of Pariti'x are, as a rule, introverted, instead of remaining merely exposed or covered by the 

 column wall; and on preserved colonies very small a])iM'tures can lie detected with a lens at 

 the place of introversion. In longitudinal sections through such polyps the apical knob of the 

 tentacle is deepest within the gastro-ctelomic cavity, and is directed outwardly, while with regard 

 to the walls themselves, the ectoderm is internal and the endoderm external, a reversal of the 

 ordinary condition. These conditions are clearly shown in tig. 40, representing a vertical section 

 through a polyp of J'orttes astriBotdi'g. No overfolding of the column wall has taken jilace on 

 retraction, so that the tentacles communicate directly with the surface of the colony. Three 

 introverted tentacles (f) are present; the one to the left is divided radially, so that the section 

 includes its aperture of communication with the exterior, while the two to the right are tangential 

 sections, and therefore do not display the external opening. Again, in fig. 30, representing a 

 transverse section through the stomodieai region of the same species, seven introverted tentacles 

 are seen in section, almost completely occupying the mesenterial chambers, and exhibiting 

 a reversal of the ordinary relations of ectoderm and endoderm. In other polyps sectioiiized a 

 variable number of introverted tentacles has been met with. The ape.x of the introverted 

 tentacle may extend as far inwardly as ])elow the inner termination of the stomod.eum. so that 

 accompanying the introversion very little diminution in the length of the stem has taken ])lace. 



Among the living expanded polyps of Mudripom also complete tentacular introversion is 

 often observed, in both apical and radial polyps. In the former six slight opacities around the 

 margin of the transparent disk remain to indicate the tentacuhir area; later, th(> tentacles may l»e 

 observed to ])rotrude, either all together or successively. 



During the retracted condition of the tissiparous genera\7^i( ?'*'«, Manicina, Mpeandrlna^ and 

 Is<ij)IiyJI!a it is sometimes impossible to discover any tentacles externally. When sections are 

 made, however, the organs are found to l)e introverted, occupying both the entoccelic and 

 exoccelic mesenterial spaces. 



In addition to actual introversion, in whicli all parts of the tentacles are still determinable, 

 a condition is often presented in which the stem wholly disappears, becoming a part, as it wei'e, 

 of the discal wall. 



Retracte<l tentacles of Si(J(r<istrH>it and Ai/aricid, for example, are usually repi'esented by only 

 a slight tubercidai' (devation of the disk, which is the knob or swollen apex, while the stems have 

 wholly disappeared in the disk (PI. XXIII, figs. 1.54-, 1.5.5, and PI. XXIV, fig. 163). In micro- 

 .scopic sections the former are displayed as mere ectodermal thickenings, charged with nemato- 

 cysts, and no ditlerentiated can l)e found whereby the tentacular stem can be distinguished from 

 the discal wall. 



Among the fully expanded polyps of OrhiceHn iiiiiii/lni'ls the two cycles of short tentacles 

 often wholly disappear. Here, again, it appears as if the tentacular tissues were not introverted, 

 but rather have become involved in the greatly expanded margin of the disk; slightly i-aised, 

 triangular areas, repi'esenting the apical swellings, are all that can l)e observed of the organs. 

 On full extension of the adult polyps of Mniiielna areoJatu the tentacles likewise may be 

 wholly wanting, their walls having become part of the expanded disk. Thicker, more opaque 

 discal spots, which are tin- only evidence of their former presence, represent the nematocyst- 

 bearing capitulum. In the young polyp of Manlrinu display(>d in section on PI. XIX, tig. 137, 

 the tentacle appears only as a thickened, nematocyst-bearing area of the ]ioiyi)al wall. Occa- 

 sionally in Pofltcii astrimldes tentacular disappearance, as contrasted with tentacular introversion, 

 may be also ot)s(>rved. 



Of previous o))servers. Fowler (1888, \). 11) has described and figured the introversion of 

 the tentacles in Sei't<it<>pom xHhuJttfa. A'on Ileider (ISSO. p. 158) has described in Astratdt'S 

 (idycidurls the opposite condition, in which the intertentacular portions of the disk have been 



