28 AMERICAN HYDROIDS. 



colony. Iii most of the Statoplea, however, the sarcostyles are provided with nematocysts which 

 are very large, long, and slightly arcuate, differing greatly from those found elsewhere. They 

 are arranged in a regular bundle just inside of the opening of the nematophore, from which they 

 are seldom, if ever, protruded to any considerable distance. When the thread is extended it is 

 seen to be extensively barbed in its proximal portion, the remainder consisting of an exceedingly 

 delicate filament of great length (tig. 86). It is hard to see the utility of barbs situated as these 

 are, or to understand how they can be brought into play to aid the penetration or lacerating 

 action of the threads when used, as the supposition is, for the defense of the colony. 



The most formidable nematocysts that I have seen among the Phiniularida- occur in the genus 

 Lytocarpus (fig. 70), several species of which are capable of causing severe irritation to the human 

 skin, which is not usually pervious to the attack of the nettling cells of the Hydroida, with the 

 exception of the millepores which probably iind their rightful place in this group. 



The researches of Ilamann, Jickeli, and others seem to demonstrate with a fair degree of 

 certainty that the nematocysts are not strictly histological cells, but an accessory part or product 

 of cells. The latter writer goes so far as to regard them rather in the light of a secretion.' 



As to the morphological significance of the sarcostyles, all of the more recent authorities, 

 except Jickeli,- regard them as degenerate individuals of the colony, or as " fighting persons." 



That they are individuals or "persons" is a matter hardly admitting of doubt; but it may 

 well be questioned whether they are ilci/cncrate persons or not, and an argument might be con- 

 structed which would go to show that instead of being degenerate individuals they are in fact 

 very highly specialized persons. Specialization is indicated when the structure has departed 

 from the original type in order to become adapted to more definite and exclusive function. It 

 would seem that the sarcostyles have done this very thing departed from the original type 

 (Protohytlnt?), and become morphologically differentiated into individuals having the definite 

 function of defense, in most cases, and of prehension by means of adhesive cells in others. 



Defensive persons are more widely distributed among the hydroida than is usually supposed. 

 They are, as we have seen, universally found among the Plumularida', and they are also to be 

 met with in Opliiodes paraaitica Sars, Ophiotles mirabilis Sars, Lafonna, tennis Sars, Oplorhiza 

 parvula Allman, Pertsiphonia filicaulis Allman, Dlplovyatlim dichotomtifs Allman, Halecium gorgon- 

 ide Sars, and Hydractinia wlihuttti Fleming. 



An interesting new family of hydroids was described by Prof. W. Baldwin Spencer,- 1 which, 

 among other novel features, is characterized by numerous cylindrical tubes inclosing defensive 

 zooids which consist of a solid endodermal axis surrounded by an ectodermal layer. The distal 

 end forms a round knob with a number of large nematocysts which greatly resemble those of the 

 Plumularida'. The whole structure appears to be almost identical with the sarcostyles ofLafoi : ina 

 ten ii is Sars. 



This family (Ilydroceratinidu-) shows distinct relationship to the Hydrocorallime in the 

 arrangement of the cu'iiosarcal tubes and their connection with the hydranths. 



Finally, there are many points of resemblance between the sarcostyles of the Plumularida- and 

 the dactylozooids of the Milleporidse. A fairly satisfactory line of intergradation between the 

 former and the latter may be traced through Lafoeina parasitica, Halecium gorgonide, JlyiJrncliiiin 

 cehiiHitn, and Clathrozoon wilsoni, the only known species of Hydroceratinidae. 



So far as I am aware, none of the defensive zooids in other groups exhibit the great 

 extensibility and the pseudopodial movements found in the sarcostyles of the Plumularidse, 

 and it is doubtless true that these latter form in themselves a very distinct type of defensive 

 zooids, a type which differs more from any other known form than the remaining types do from each 

 other. It must be remembered, however, that only part of these latter have been studied with care 

 and the use of modern facilities, and there is thus a possibility that a more perfect intergradation 

 between the sarcostyles of the plumularians and the defensive zooids of other groups may yet be 

 demonstrated. 



1 The iiuthor hoyes in present, ;i discussion of the nematocysts iu connection with the introductory part of 

 thin work. 



J As before mentioned, this writer seeks to establish a homology between the sarcostylcs and the tentacles. 



:1 A New Family of Ilydmide:i, together with a Description of the Structure of a New Species of riunmluria; 

 Transactions i>f the 1,'uval Society of Victoria., 1890. Professor Spencer very kindly furnished the author wilh a 

 specimen for study. 



