308 DK. J. E. DUERDEN ON THE 



indication of alternating large and small pairs. In this and the other figures the pairs 

 are numbered according to the order to which they belong, so far as this can be 

 determined with certainty. In the polyp whose mesenterial plan is represented by 

 PI. 26. fig. 11 a regularity of alternating complete and incomplete pairs is presented on the 

 leftside, but the right side is a region of irregularities. A mesentery also stretches across 

 from one part of tlie column-wall to the other, and encloses a rudimentary mesenterial 

 pair. In fig. 12 the left side j)ossesses perfect dicyclic hexameral symmetry, while 

 the development at the ujDper right-hand region is very irregular. As before, many 

 incomplete mesenteries occur at the upper right-hand corner. 



Fig. 14 is tbe completed mesenterial plan of the polyp of w^hicii fig. 13 represents 

 a little more than half. As before, the first and second cycles on the left side are 

 regularly develojied, and in addition three I'udimeutary pairs of the third cycle occur, 

 a pair in each of the primary systems. The new pair in the dorso-lateral chamber 

 is situated on the ventral aspect of the second cycle pair, but on the dorsal aspect in the 

 middle and veutro-lateral chambers. On the right side the ventro-lateral system 

 corres^TOnds with that on the left, but the middle and dorsal systems are irregular. 



Fig. 15 completes the series so far as I have been able to establish it. On the 

 left side of the directive axis all the pairs of the first, second, and third orders of a 

 hexameral polyp are represented. The second-order mesenteries (II. j, with the exception 

 of the dorso-lateral pair, have reached the stomodteum. One mesentery only is missing 

 on the right side from the number necessary to complete the hexameral plan, but at the 

 dorso-lateral region they are all incomplete and irregularly paired. 



From the conditions thus presented there can be no doubt that normally the 

 mesenteries should be arranged in alternating hexameral cycles, with two pairs of 

 directives, and that members of the second order should also reach the stomodseum. 

 The increase in the number of the mesenteries, how-ever, does not proceed throughout 

 the polyp according to the usual order established for the Hexactiniae. According 

 to this, the six primary pairs of mesenteries appear successively in a bilateral manner ; 

 then the members of the second cycle arise successively in unilateral pairs within the 

 six primary exocceles, as a rule on each side of the polyp i'rom the dorsal to the ventral 

 aspect ; the third cycle pairs appear as single unilateral pairs within one of the two 

 exocoeles in each of the six primary systems, and then later in the remaining exoccele 

 in each primary system. In fig. 14. the sequence of the thu-d cycle pairs seems to 

 follow the rule that a pair is developed in each of the primary systems before a second 

 pair arises to complete the radial symmetry. 



All the figures of BuHodeopsis indicate that on one side of the poly^D the increase 

 takes place in a fairly normal and regular manner, but that such is not the case for the 

 other side. In all the polyps a definite region of irregular growth occurs, both as 

 to the number of mesenteries which are present and also as to their paused relationships. 

 Evidently the hexameral cyclic regularity becomes more pronounced as the polyps grovi^ 

 larger and the number of mesenteries increases, but I have met with no specimen 

 in which perfect regularity was attained. Probably some of the polyps obtained are as 



