288 J- Playfaik Mc Mubrich, 



palmate variety, althoug'h it is in reality unilateral and in some 

 individuals may be reduced to a comparatively small number of 

 mesogloeal processes, arising for the most part directly from tlie 

 column wall. The longitudinal muscles of the tentacles and the 

 radiating muscles of the disc are moderately developed and ecto- 

 dermal in position. Zooxanthellae occur in considerable numbers 

 throughout the endoderm and are especially abundant in the tentacles. 



The stomatodaeum is longitudinally ridged and possesses two 

 well developed siphonoglyphs. The mesenteries are arranged in 

 three cycles, with a partial development of a fourth in some in- 

 dividuals. As a rule they are arranged hexamerously, but in one 

 individual only twenty-one pairs occurred, eight pairs intervening 

 between the two directive pairs on one side and eleven on the other. 

 In some individuals two cycles of mesenteries were perfect, but in 

 others only one cycle reached the stomatodaeum. In all both the 

 first and second cycles were provided with well developed muscle 

 pennons and mesenterial iilaments, while the third cycle lacked them. 

 The mesogloea of the mesenterial fllaments contained numerous 

 Zooxanthellae. The longitudinal muscles have the form shown in 

 Fig. 81 and the parieto-basilars form a well-marked fold. 



As a rule the mesenteries of the first cycle, including the di- 

 rectives, were all fertile and here and there reproductive cells were 

 observed in individual members of the second cycle; the third cycle 

 was sterile in all cases. In one of the specimens examined each 

 fertile mesentery contained both ova and spermatozoa, and in two 

 others, one from Antofogasta and one from Puerto Montt embryos 

 in various stages of development were found in the body cavity. In 

 the individual from Puerto Montt a few scattered ova occurred in some 

 of the fertile mesenteries, but spermatozoa were much more abundant, 

 and in the Antofogasta specimen only spermatozoa were present. 

 It would seem in this last instance either that embryos had made 

 their way from the exterior into the body cavity of a male indi- 

 vidual, or eise that the specimen was a protogynous hermaphrodite. 

 For many reasons the latter supposition is preferable. 



I have identified these specimens with Carlgren's Bunodes 

 Jiermaphroditicus (1899) from Talcahuano on account of their marked 

 similarity in every particular except the number of perfect and 

 fertile mesenteries. Caelgeen finds three cycles of these, instead 

 of the one or two described above. This diiference may be due to 

 the fact that Carlgeen's specimens were apparently about twice as 



