MUSEUM OF COMPARATIVE ZOOLOGY. 133 



tentacular knobs of which had two lateral terminal filaments, while 

 Stephanomia had but a single filament of this kind, although the last 

 two genera have a biserial row of nectocalyces and an involucrum. 



There are certain other characteristics of this genus which are not so 

 well marked as those already given by Huxley. I refer to the character 

 of the tentacles, and more especially to the position of the sexual organs. 

 Tentacles such as we find in Agalma do not seem to exist in the genus 

 Halistemma, but the tentacular knobs have very long pedicels, longer 

 than in other Physophoridse, which allow the knob to project so far be- 

 yond the covering scale as to resemble tentacles very closely. Accord- 

 ing to some observers true tentacles do exist in the genus Halistemma. 

 For instance, Leuckart says that Kolliker missed the true tentacle, and 

 mistook the pedicel of the knob for a tentacle itself* Kolliker's figure 

 of Agalmopsis punctata, which is the same thing as Halistemma ruhrumy 

 shows the absence of the tentacles very plainly. My observations on the 

 tentacle agree with Kolliker's, yet his figure of the animal is not com- 

 plete, in that he failed to represent the sexual system. The female 

 sexual organs I shall later describe. (PI. I. figs. 3-5.) Leuckart* fig- 

 ures a true tentacle in Halistemma. What Claus describes as Hali- 

 stemma tergestinumf does not seem to belong to Halistemma in the sig- 

 nification given to the generic name by its founder, Huxley. It belongs 

 rather to Huxley's genus Stephanomia in all its structure, but especially 

 in the character of its tentacular knobs, a feature of greatest importance 

 in the classification of the Physophoridse. 



Haeckel (^Entwickelungsgeschichte der Siphonophoren) proposes a di- 

 vision of the Agalmidse which has some advantages, although to use 

 the trifid character alone of the tentacular knob as a basis of his sub- 

 family Crystallodacea separates those with an involucrum, and places 

 Agalmopsis {Stephanomia, Huxley) with Forskalia and Halistemma. 

 These last have no involucrum in the tentacular knob, and the former 

 has sexual organs arising at the base of a polyp, while the latter has 

 these same structures midway between two tasters. There does not 



* Leuckart, Zur Nahern Kenntniss der Siphonophoren von Nizza, Taf. XII. fig. 15. 

 When I studied Halistemma, I did not know of this difference of observation by 

 Kolliker and Leuckart. 



t 1. Metschnikofif, Proc. So. Fr. Nat. Moscow, Vol. VIII.; Studien der Medusen 

 und Siphonophoren, Zeitsch. f. Wiss. ZooL, Bd. XXIV. 



2. Claus, C, Ueber Halistemma Tergestinum, &c., "Wien, 1878. Mittheilungen 

 liber Siphonophoren und Medusen Fauna Triests, Zool. hot. Gesell. Wien, Tom. XXVI. 



3. Eschscholtz characterized the genus Agalma, ** Tentacula ramulis clavatis : clava 

 apice bicuspidata. " 



