REPORT ON THE ANT1PATHARIA. 51 



characters, which when fully investigated will undoubtedly prove of great interest. 

 The exceptional mode in which the ultimately round and densely spinose horny axis is 

 produced is especially worthy of note. 



Savagliid.e ( = Gerardiid.e). 



The single species constituting the genus Savaglia, Nardo {Gerardia, Lacaze 

 Duthiers), was separated by Lacaze Duthiers in 1864 from the Antipathidae on account 

 of the structure of the polyps, which very closely resemble those of Hexactinise. His 

 observations refer to living specimens, and bring out many points showing the most 

 interesting relations of the genus. The following is a summary of his results : — 



The sclerenchyma in fresh specimens has a bronzed coppery black colour ; wdien dry 

 it is jet-black. The mode of branching in young colonies, which are always parasitic, 

 varies with the species of Gorgonidse, which acts as its support ; usually the branches ex- 

 tend in one plane without touching or fusing. Later, with greater development, bridges 

 are often thrown across from one branch to another, and fusions take place without regu- 

 larity. These fusions are produced by fractures or abrasions, and cannot be considered 

 characteristic of the species. Lacaze Duthiers points out that it was want of knowledge 

 on this point which led Haime to consider the mode of branching as a specific character. 

 The base is often very large, sometimes " as thick as a man's leg," whilst the branches are 

 only 1 to 2 dcm. long. This is an abnormal growth due to the constant working of the 

 coral fishers over the ground, by which the branches are repeatedly broken off. Speci- 

 mens which have been allowed to grow undisturbed, such as were brought to Lacaze 

 Duthiers from a bank not previously worked, are very large, fine, and much branched, 

 without such a great base. In such cases, where the sclerenchyma extends beyond the 

 Gorgonid basis, the growth becomes bushy. Very old specimens frequently have a number 

 of anastomosing branches, sometimes descending from a superior part to fuse with one 

 below, at others very long branches may unite with those of the opposite side of the 

 sclerobasis and form transverse connections, the origin of which is due to a primary frac- 

 ture and subsequent fusion. The branches are not cylindrical, but flattened on one side, 

 along which there is a well-marked groove. The branchlets are usually swollen at the 

 tip, and are never thread-like or pointed as in the Gorgonidse. Tubercles often occur 

 on the sclerobasis, which indicate the point of origin of new branches, or the bases of 

 broken ones. Under a low magnifying power the sclerobasis is seen to be covered with 

 very small mammiform tubercles, with a depression in the centre. Lacaze Duthiers 

 describes one specimen parasitic on a grey, flexible, striated axis, undoubtedly that of 

 Muricea placomus. In this specimen the zoanthodema consisted solely of sarcosome 

 on a borrowed axis, and no sclerenchyma had as yet been deposited. In another 

 specimen of the same Gorgonid two or three patches of Gerardia were present on the 



