506 



MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



Were the separation of the disk and column wall to be completed at this stage, across the plane 

 of fission, it is clear that two similar daughter polyps would be produced. In each case one 

 moiety of the middle mesenterial pair of one side would form with the corresponding mesentery 

 from the other side a pair, attached to the side of the stomodseum opposite the directives. But 

 the longitudinal muscles of the two mesenteries in each pair would be arranged so as to face one 

 another, instead of turning from one another, as in the case of the directives; in other words, 

 the complete mesenteries of each polyp would include only one pair of directives, and five pairs 

 in which the retractor muscles are vis-a-vis. The six pairs of complete mesenteries attached to 

 each stomodseum would then constitute a first cycle, the six large alternating pairs a second, the 

 twelve next a third cycle, and the odd pairs would represent the commencement of a fourth 

 cycle. 



The almost perfect regularity in the number, arrangement, and extent of development of the 

 mesenteries found in the above example appears to be rather exceptional, for other specimens 

 of J/" niri Hit sectionized reveal many irregularities. Fig. 145 represents the mesenterial arrange- 

 ment in a second polyp, also bearing two oral apertures. The details are those presented by a 

 single section toward the lower termination of each stomodseum. The hexameral regularity has 

 been altogether lost, or perhaps was never reached. But here again the plane of fission passes 

 through the entoccele of two opposite pairs of complete mesenteries, at right angles to the direc- 



D 



Fig Uh. 



Manteina areolata. — Another polyp with two oral apertures. The mesenteries have entirely lost their hexameral cyclic regularity, and 



of the pr< itocnemes only the directives at opposite ends can be determined with certainty. 



tive plane, and one of the two pairs of primary directives is found at each extremity. More than 

 six pairs of complete mesenteries are united with each stomodanun. and only in two or three 

 regions of the polypal wall are any members of the second and third cycles developed; the hex- 

 ameral cyclic plan will be entirely lacking in the daughter polyps as in the original, and two 

 or more complete mesenterial pairs may occur in succession, without any alternating incomplete 

 members. In the sections higher than the one represented other pairs of small mesenteries 

 occur, but do not in any way assist towards the production of the hexameral regularity. At the 

 left end of the polyp two single large mesenteries occur, without any corresponding member to 

 complete the pairs; higher sections demonstrate that one moiety has simply disappeared in 

 advance of the other. 



In a third bi-oral polyp studied the alternation of first, second, and third cycles of mesen- 

 teries was a little more regular than in the last example, but was by no means perfect all round; 

 one stomodseum was provided with six pairs of complete mesenteries, while to the other eight 

 pairs were attached. 



Young polyps of Mcmici/na with three oral apertures are not so plentiful as specimens with 

 two or four apertures. It seems not, unlikely that after the first division into two, each stomo- 

 dseum is again divided at about the same time, and thus the stage with three apertures would 

 rarely occur. 



