518 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



mesenteries!. Four to eight spermaria may be present on a single mesentery in section; but 

 the number of ova is much less, usually one or two, rarely reaching four or rive. In the 

 hermaphrodite mesenteries the ova are invariably situated toward the fixed end of the 

 mesentery, and the spermaria, nearer the free edge, that is. more central. 



I'',iri<i fragum is likewise monoecious, and in this species also the two kinds of sexual cells 

 may occur on the same mesentery, though more often they are on different mesenteries within 

 the one polyp. In one large colony examined, the cavity of practically every polyp was crowded 

 with larva', all at the same stage of development. In addition to these the mesenteries were 

 swollen with spermaria and ova, all apparently at a similar stage toward maturity — the ova 

 having a large nucleus and nucleolus, and the spermaria with the tails of the spermatozoa 

 already developed. 



On a colony of Pontes clavaria most of the individual polyps contained several free larvae, 

 again all at about the same stage of development. In this instance, ova only were present ou the 

 mesenteries, usually one large egg to each, and of numerous polyps examined from the same 

 colony none bore male sexual elements. In transverse sections of a polyp of Madrep&ra pal/mata 

 a single ovum occurred on three of the four lateral complete mesenteries, but not on any of the 

 directive mesenteries. 



Female gonads were moderately developed in a colony of Tsophyllia dvpsacea. In transverse 

 sections some of the mesenteries exhibited three or four ova, restricted in their distribution 

 toward the insertion of the mesentery in the body wall. Most of the mesenteries of a polyp of 

 OrbiceMa radluta also contained ova, again situated toward their insertion in the polypal wall; 

 likewise polyps of Sidt rastrsea siderea. In the last the ova were greatly distorted and irregular 

 in form, as if adapting themselves to the very narrow interseptal loculi. 



From the above examples, it is impossible to say how far sexual differentiation has proceeded 

 within Madreporarian polyps. In all instances where a unisexual character would be indicated, 

 only female cells have been found, and then somewhat sparsely. Spermaria have never been 

 found alone, but always in association with large numbers of ova." It may be that coral species 

 are mainly monoecious, but that ova are first developed (protogynous), and later spermaria, 

 either on the same or different mesenteries. Further, the ova are, as a rule, restricted toward 

 the fixed or peripheral margin of the mesentery, while in Actinia? generally they are disposed 

 about the middle of the radial length of the mesentery. 



Apparently very little importance can be attached to the particular mesenteries on which 

 gonads may occur, for where present in numbers they are found on practically all the mesen- 

 teries, both complete and incomplete. In Orbicella radiata a,uy of the mesenteries of the three 

 cycles may bear ova; as shown on PI. IX, fig 68, ova occur on a small incomplete mesentery, 

 while they are not developed on the larger complete member. In Actinian studies the distribution 

 of the gonads on particular mesenteries is considered to be of some importance for systematic 

 purposes, but manifestly the production of a few sexual cells on certain mesenteries can not be 

 of much significance, when in riper polyps they may possibly be found on all the organs. 



Within its lifetime a polyp may give rise to more than one series of ripe sexual cells, for 

 individuals charged with fully developed larvaj also contain numbers of nearly ripe eggs, still 

 within the mesoglcea of the mesentery. 



Viviparity would appear to be the rule among corals, though Wilson records the extrusion 

 of eggs and semen from Manioina, and I have also found this to happen in Fa via. It may be 

 doubted whether such occurrences take place under normal conditions, as in both instances 

 highly developed larvae have also been found within the gastric cavity. Similar promiscuous 

 extrusions of sexual cells are likewise found to take place among anemones. From Rhodactis 

 sancti-Thoiittt', Aulactinia fitdloides, and others unfertilized eggs and semen have been found 

 to be extruded in abundance, while from both species mentioned larva' at an advanced stage of 

 development have also been obtained. 



" Miss I'ratt ( 1900 ) describes Xinlielin as probably monoecious, spermaria only being present. 



