430 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



G. ('. Bourne (1887). in his paper: "The anatomy of the Madreporarian coral Fungia." 

 refers to the disposition of the tentacles in that genus. His figure oiFungia (PI. XXI11) shows 

 a wide interspace between the different cycles of tentacles, as is found to be the case, in the much 

 smaller polyps of Siderastrsea and Agaricia. Evidently, the character may he taken as of some 

 diagnostic importance within the Section Fungacea. Bourne casts suspicion upon the accuracy of 

 Dana's description and figures of Fungla (Zoophytes. Wilkes Exploring Expedition, and 

 Corals and the Coral Islands), which represent an irregular distribution of the tentacles at 

 intervals over the whole of the large disk, as does also the figure of Quoy and Gaimard in 

 Voyage de la corvette P Astrolabe. The results from Siderastrsea and Agaricia, detailed 

 above, prove that an irregular appearance in the disposition of the tentacles is by no means 

 uncommon in the Fungacea. The regular cyclic disposition, when really present, can often be 

 established only after a long acquaintance with the forms, and under favorable conditions of 

 expansion or retraction. 



The figures of the fully expanded polyps of Siderastrsea, accompanying Agassiz's Florida 

 Reefs (1880, PI. XV., tigs. ti. 7). indicate an irregular tentacular arrangement in both cases, 

 and such would probably he assumed by any observer on a casual acquaintance with the polyps. 

 The appearances given the tentacles in Agassiz's figures were rarely met with in Jamaican 

 specimens, but the dimorphism is clearly shown on some of the members, and is referred to by 

 Pourtales in "Deep Sea Corals" (1871). 



ORDER OF APPEARANCE OF PROTOTENTACLES. 



In corals whose development has been studied sufficiently far, the first tentacles are found to 

 make their appearance within a few days after the fixation of the larva. The number of tentacles 

 first to arise corresponds as a rule with the number of internal mesenterial chambers already 

 established, the tentacles being outgrowths from them. Generally, in coral larva', the twelve 

 primary mesenteries, with their corresponding chambers, are developed either at the time ol 

 fixation or shortly after, and the twelve primary tentacles appear either simultaneously, one 

 from each mesenterial chamber, or one cycle may arise in advance of the other. In the latter 

 • ass the inner cycle of entocoelic tentacles usually appears first, and the exoccelic members next, 

 but in Siderastrcea radians this order is reversed (p. 533). 



The establishment of the tentacles serves to delimit for the first time the-larva into two 

 regions — disk and column; and with this the larva may be considered to have become the polyp. 

 The part of the polypal wall hearing the tentacles and mouth is the disk, and the region outside 

 or below is the column. The former becomes more or less flattened, and constitutes the free oral 

 extremity of the polyp, as opposed to the fixed or basal aboral extremity, while the column is 

 vertical and remains more or less cylindrical. 



The actual appearance of the primary tentacles has been observed as follows: The larvae of 

 Astroidt s calycularis, examined by Lacaze-Duthiers ( L873), presented twelve tentacular prominences 

 at a very early stage after fixation. During the development of Garyophyllia cyatltus, G. von 

 Koch (1897) found that in most cases the two primary cycles of tentacles appeared simultaneously, 

 though some of his observations seemed to indicate a successive origin. Von Koch's figure 

 (p. 760) «>f the young polyp, at the stage when the prototentacles are all developed, represents the 

 members of the inner entocoelic cycle as smaller than those of the outer exoccelic cycle, but in the 

 text the author states that they are larger. Lacaze-Duthiers (1897), in Ins recent paper on the 

 corals of the Gulf of Lyon, gives many figures illustrating the early development of Balanophyllia 

 regia. From flic beginning two alternating cycles of large and small tentacles are indicated, and 

 no reference is made to any intermediate stage. The same author I L894) mentions six tentacles 

 as occurring at an early stage in the development of Flabellum anthophyllum, and later figures 



the complete twelve. 



Young polyps of Manicina <rr< olata, which I was able to rear to the stage with twelve tent aides. 

 were also characterized by the simultaneous development of these organs. When first defi- 

 nitely recognizable under the microscope, after a period of fixation of about fourteen days, two 

 cycles were present, nearly equal in size (PL XIX. tig. f:;.">). In two orthree young polyps, from 



