32 PERCY SLADEN TRUST EXPEDITION 



in the genus. In a later paper (51) he further suggested that coral polyps are male when 

 young, becommg female as they get older, that all the members of a colony are either 

 male or female at the same time. I have looked over his sections of Flahellum and have 

 found the cell-groups he has described to be identical in appearance with those in the 

 polyps of Cylicia and Prionastrea abdita already referred to. If Professor Gardiner's 

 suggestion be confirmed, then a species like his P. abdita will have to be regarded as 

 hermaphrodite, since ripe ova were present in the same mesenteries along with the 

 testicular masses. If, on the other hand, the bodies I have termed spermatozoa really 

 represent the final stage in spermatogenesis and not simply artefacts*, then the follicles in 

 Flabellum would not be testes but groups of primitive sex-cells from which ova were 

 ultimately to develop. Indeed, the stringed arrangement of these groups and of the ova 

 are similar. Fresh ova would develop from these groups as the ripe ones were discharged. 

 When all the possible ova have thus been formed, the follicles would altogether disappear, 

 as happened to be the case in the two largest polyps of Flabellum which Gardiner 

 examined. All the remaining polyps which had only the follicles would then be immature 

 females. On such an interpretation none of Gardiner's polyps could be regarded as 

 protandrousf. However, on questions like these, no definite conclusions are possible till 

 fresh polyps are actually examined. 



The similarity between the nuclei of the germ-cells described above and those of the 

 calicoblastic layer of ectoderm has already been noted. Around some of the calicoblast- 

 nuclei a thin coat of cytoplasm appears to be differentiated. Certain appearances further 

 suggest that these cell-elements migrate into the mesenterial endoderm and ultimately 

 into the mesoglsea to form the above-mentioned follicles;|;. If so, this phenomenon, 

 together with some of my observations on the possible migration of nematocysts from the 

 ectoderm into the endoderm, will have considerable theoretical importance§. 



Zooxanthellse (PI. 1, figs, l, 4, and PL 6, fig. 51, etc). All the polyps I have examined 

 contain the so-called zooxanthellas in greater or less abundance, but they are invariably 

 restricted to the endodermal layer. Each of these algge is round, the protoplasm staining 

 pink, the nucleus excentrically placed and granular in appearance. In addition, there is, 

 in most algge, a homogeneously dark-stained body — in all probability a pyrenoid — with a 

 transparent ring round it. Algae are most abundant in the edge-zone and the peristome, 

 often completely filling the endoderm, and to a less extent in the tentacles, i.e. in the 

 exposed regions of the soft parts. In a species like Galaxea musicalis, in which the 

 stomodfeum seems to be imperfectly functional, the zooxanthellse are most numerous. This 

 fact may be taken as an additional indication of the symbiotic nature of these algse. In 

 the mesenteries they are more numerous in the non-pleatal than in the pleatal sides, which 

 may be accounted for by the action of the strong musculature on the pleatal sides. 



* The definite shape of these bodies and their massed arrangement in the mesenterial mesoglsea (in 

 exactly the same position as the ova) do not warrant their being regarded as artefacts. 



t Duerden's account of the gonads in the West Indian coral species which he studied is practically the 

 same as Gardiner's, but he thinks that ova are developed first and spermaria later. His PI. 20, fig. 140, is 

 similar to the cases I have recorded. 



\ Boulenger (10) has traced the origin of sex-cells in craspedote medusse to interstitial cells of the ectoderm. 



§ The sex-cells in coral polyps have hitherto been regarded as arising from the endoderm. 



