6 AMERICAN HYDROIDS. 



tube is anterior in the branches on the upper part of the stem and becomes immersed in the lower 

 or proximal portion of the colony. In only one species, Aglaophenopsis kirsuta, could this tube in 

 full-grown specimens be traced as a superficial tube clear to the origin of the stem. 



There appears to be some difference of opinion as to the origin of the branches of the fascicled 

 stem. Bale 1 says: " The branches spring not from the.jointed stem [hydrocladiate tube], but from 

 the supplementary tubes which grow up in contact with it." Prof. Baldwin Spencer, another 

 Australian writer, in describing a new species, Plumularia procumbens, speaking of the tubes of 

 the fascicled stem, says: 2 "The central one gives origin to all of the branches, passing out into 

 the pinna- and hydrocladia." 



In order to obtain additional light on this question, I have made a number of dissections of 

 fascicled stems of various genera, and find that the hydrocladiate tube gives origin to the branches 

 in the following species: Plumularia profunda, Calvinin mirabilis, Cladocarpus paradisea, Theco- 

 iarpua benedict i, and Antennopsis species. In the two following species the accessory tubes give 

 origin to the branches: Phintulttria dendritica and Lytocarpus darker. Thus, out of seven species 

 examined, five were characterized by branches which spring directly from the hydrocladiate tube 

 and only two had branches formed entirely from accessory tubes. It seems, therefore, that there 

 is no consistent arrangement, and that the branches may spring either from the hydrocladiate or 

 accessory tubes; more frequently from the former among American species. 



Professor Spencer says that the hydrocladiate tube in /'. procumbens is not divided into 

 internodes except in its distal free portion. In all of the American species examined the inter- 

 nodes could plainly be discerned throughout after the hydrocladiate tube had beeu dissected 

 away from the others so as to admit of satisfactory examination. 



The accessory tubes vary in number from one (A<jl<u>iilir-nia longicornis) to scores or even hun 

 dreds (Plumularin dendritica). The individual tubes are, in general, parallel to the hydrocladiate 

 tube, but are often more or less sinuous, especially in the basal portions of thick and woody stems. 



Each tube terminates distally in an open end, and it can often be traced downward where it 

 is found to end in a rootlet or to become connected with the hydrocladiate tube at the point from 

 which the original hydrocladia sprung. Ordinarily the accessory tubes communicate with each 

 other by means of lateral tubular processes passing from one tube to the other. These cross com- 

 munications are very prominent in some species of Cladocarpus (fig. 4) and Thecwarpus. They are 

 minute in Calvinia mirabilis, and I am unable to make them out in Plumularia dendritica. In 

 some species each of the superficial tubes bears a double row of cauline uematophores (Cladocarpus 

 paradisea), while in others these nematophores are very minute (Lytocarpus clarkei), consisting of 

 sarcostyles without sarcotheca 1 (Plumularia procumbent), or are absent (Aglaophenopsis Jiirttiita). 



As to the homology of the accessory tubes, authorities seem to agree that they are modified 

 hydrorhizal elements. Bale says on this point: "As regards the origin of the combined stem, it 

 is obvious that the plumularia tubes are hydrorhizal elements. Monosiplionic species sometimes 

 occur with a few irregular tubes, which, springing from the hydrorhiza, have attached themselves 

 to the basal part of the stem instead of to a foreign body." :! Spencer comes to the same conclu- 

 sion. 4 Some of my own observations would seem to contradict these authorities. In dissecting 

 the fascicled stem of Gladocarpus paradisea I found that the accessory tubes were sometimes given 

 off from the hydrocladiate tube. Indeed, they seemed to grow from the old stumps of hydro 

 cladia, which they had apparently replaced. The tubes originating in this way could in no wise 

 be distinguished from the ordinary accessory tubes, and so, morphologically speaking, they must 

 be regarded as modified hydrocladia. The same condition of affairs was found in Plumularia 

 dendritica. 



The best demonstration of the hydrocladial origin of the accessory tubes was afforded by a 

 dissection of a gigantic specimen of Thecocarpus secured in Naples, in which, throughout the 

 immersed portion of the hydrocladiate tube, the accessory tubes had their origin from the stumps 

 of the old hydrocladia. 



1 The Genera of the Hniiiulariickr, with observations on various Australian H.vdroids, Melbourne, 1S86, ]i. ">. 

 -A New Family of Hydroida, together with a Description of the Structure of a New Species of Plumularia, 

 Transactions of the Royal Society of Victoria, 1890, p. 133. 

 'The (ieuera of the Plumulariidae, ji. 5. 

 1 A New Family of Hydroida., Transactions of the Royal Society of Victoria, 1890, p. 133. 



