6 PERCY SLADEN TRUST EXPEDITION 



specimen is available. Foi-tunately most of the earliest types of the species discussed in 

 this paper are still existent in the various museums visited. Regarding the species of the 

 American authors I have referred only to those whose characters could be made out either 

 from descriptions or from figures. 



In the systematic part of this work the plan followed has been to base species and 

 genera on the characters of the soft ' as well as of the hard parts by a comparative 

 examination in every possible case of poljrps from different colonies and of as many hard 

 specimens as were available. Thus it has been possible in many cases to study the varia- 

 tion that the corallum undergoes, the principal forms being shown in the photo-plates. 

 Owing to lack of polyp-material, only the probable specific rank of some species can be 

 indicated by recording their skeletal characters. Notes on the specimens examined in the 

 various museums are also given under each species. 



Finally I would remark that there should be little cause for complaint in the task of 

 cataloguing examples of known species in a museum. In my visits to museums I found 

 that any species whose limits had once been settled by the study of both polyps and hard 

 parts could be recognised later from the hard parts alone. Only in the case of undeter- 

 mined species should it be necessaiy to examine the polyps. 



II. ANATOMY OF THE POLYPS (Plates 1—10). 



Method. The material had been in the first place fixed in a saturated solution of 

 corrosive sublimate and preserved in 90 °/^ pure alcohol ; some were killed first in formic 

 aldehyde poured into seawater. From each specimen two pieces were chi^sped off, one to 

 be decalcified for the soft parts and the other to be macerated for the hard parts ; the 

 decalcification was done in weak solutions of nitric acid. Polyps to be sectioned were then 

 embedded in the usual way*, and sections cut to thicknesses of 12 /a, 10 /a, 8 /z , 6/a, 4ju.. 

 After repeated experiments with various stains, I found that Haidenhain's iron heematoxylin 

 followed by eosin gave the best results ; the former stains the nuclei, muscle-fibres and 

 nematocyst threads dark, and the latter colours the cytoplasm light pink and the mesogisea 

 deeper pink. All the polyps whose histology I have studied were treated by this method. 



Body-layers. From an examination of sections of coral polyps I am led to think 

 that the ectoderm and endoderm are not composed of definite units known as cells. The 

 usual appearance presented by each layer is that of a sheet of protoplasm with nuclei 

 either closely aggregated as in the ectoderm, or somewhat scattered as in the endoderm ; 

 an exti-eme case is that of the stomodseal ectoderm, in which the nuclei are so numerous 

 and so closely packed together that it is difficult to see how thei'e could be cell-limits. 



* In connection with the Meandroid Astrseids, T cut some sections in gelatin with Aschoff's COg freezing- 

 microtome, a method recently employed by Mr J. F. Gaskell ("A Method of cutting frozen Sections by embedding 

 in Gelatin," Journ. of Pathology and Bacteriology, vol. xvii. No 1, p. 58, 1912). The advantages of this method 

 consist in the avoidance of absolute alcohol, xylol or paraffin, which usually shrink and distort the internal 

 organs of the polyps, and in the better chance of preserving the histological condition of the tissues as the object 

 has not to be subjected to a temperature above that of the room, but there is great difficulty in making serial 

 sections and no method has yet been devised of dissolving out the gelatin from the sections. These sections in 

 general confirm the account of polyp cytology given subsequently. 



