104 MKDT'S.K OF THE WOULD. 



V Tribe PAND.IIDI: 



With 4 or more well-developed tentacles. The gonails are interradial, corrugated, or folded ridges in the wall of the stomach 



4 radial-canals. 

 Pundta LESSON, 1843 (1837?). 4 interradial gonads in stomach-wall, not completely separated in the 4 principal 



radii. Hydntid unknown. 

 Conis BRANDT, 1838. Ocelli borne upon ends of special, short clubs which arise from bases of tentacles. Hydroid 



unknown. 



Turris LESSON, i^^=Tiara, LESSON -4- Catablema, HAECKEL. With 4 interradial crescent-shaped gonads in the ecto- 

 derm of the stomach-wall. Each crescent is composed of partially fused, swollen ridges. Hydroid: Clavttla ( ?), 

 Carnpaniclava ; or Perigonimus ( r) 



4. Tribe CAI.YCOPSIDI: 



With more than 4 simple, unbranched radial-canals. Adradial, transversely folded gonads. 



Cnhcopsis FEWKES, 1882. With 16 simple, separate, unbranched radial-canals, 4 radial, 4 interradial, 8 adradial. 



Haeckel, 1879, established the family Tiandae for Anthomedusa? with 4 wide, crenulated 

 lips; with 4 separated or 8 cleft gonads in the stomach-wall; with 4 wide, hand-like radial- 

 canals; and with simple, unbranched tentacles. 



Vanhoffen, 1891, showed that the gonads were interradial, not radial in position as was 

 supposed to be the case by Haeckel; and in 1892 Hartlaub gave important details of the 

 structure of the gonads, showing that Panden had simpler gonads than either Tiara or Turris. 

 For example, he showed that the gonads of Pandea conica consist of 4 interradial, horseshoe- 

 shaped, network-like, swollen regions in the ectoderm of the stomach-wall. The gonads of 

 T urns cceca, however, consist not only in the interradial network, hut chiefly in a double row 

 of fused longitudinal swellings in each interradial quadrant of the stomach. In Tiara pileata, 

 on the other hand, we find none of these network-like gonads, but instead a horseshoe-shaped 

 gonad in each interradial quadrant, the apex being upward and the sides of the horseshoe 

 being made up of a series of laterally-fused, horizontal swellings. Maas, 1904, loc. cit., gives 

 clear figures of these conditions and supports Hartlaub's observations. 



In the more complex and specialized Tiarinae, represented by the tribes Amphinemidi and 

 Pandaeidi, the gonads are thrown into complex folds or corrugations, and the lips have become 

 recurved, with folded edges. 



Vanhoffen concluded that Amplnnema and Codonorchls of Haeckel are equivalent to 

 Stomotoca L. Agassiz; and in this I concur. He also maintained that Pandea Lesson was 

 identical with Tiara Lesson, but in 1892 Hartlaub showed that the gonads of Pandea were 

 quite different in structure from those of Tiara; and Maas, 1904, supports this conclusion. 



Maas, 1904, calls attention to the fact that there are no definite distinctions between 

 Turns and Tiara. He shows that the gonads of Turris consist of 4 interradial horseshoes, the 

 sides of each horseshoe being made up of partially fused, transverse, branched, ectodermal 

 ridges and the upper apex of the horseshoe being composed of an open network of ridges. In 

 Tiara the gonads are horseshoe-shaped, but commonly lack the network-like ridges. The 

 transverse ridges do anastomose to some degree, however, in Tiara, so that a partial network is 

 often found. See Maas, 1904, lor. at., plate 2, fig. II. 



Maas describes the gonads of Catablema Haeckel as horseshoe-shaped and composed of 

 partially fused, vertical ridges. In the fully-grown medusa, however, I find that the ridges 

 tend to become transverse, as in Turris or Tiara. 



Maas, 1904, retains the genus Catablema of Haeckel to include medusae resembling 

 Turns or Tiara, but with blindly-ending, lateral diverticula upon their radial-canal and ring- 

 canal. As a matter of fact these diverticula are highly variable in different individuals of the 

 same species and are seen in a more or less well-developed condition in the majority of Tiarinae. 

 I therefore believe that confusion will be avoided it we combine the genera Turris, Tiara, and 

 Catablema to form a single genus. This should be called "Turris," for Lesson used this name 

 on page 283 of his Hist. Zooph. Acal., and on page 17 of his "Prodrome," 1837 Tiara he 

 defines later on page 286 of his "Histoire," 1843, and on page 20 of his " Prodrome, "1837. 



I propose, therefore, following the lead of Vanhoffen, Hartlaub, and Maas, to reduce the 

 13 genera of Tiarinae enumerated by Haeckel, 1879, to 8, as follows: Stomotoca, Modeeria, 

 Protiara, Heterotiara, Pandea, Turns, Corns, and Calycopsis. 



The Tiarinae are widely distributed, but are abundant only along continental coasts, for 

 in so far as is known, they arise by budding from Tubularian hydroids of the genera Pcngoni- 

 nius, Clavula ( ?), and Campaniclava ( r). Asexual budding or direct development in the 

 medusa-stage is unknown. 



