HYDROMEDUSAE, SIPHONOPHORES, AND CTENOPHORES. 291 



Station 5125, off Nogas Island, Sulu Sea, vicinity of southern 

 Panay, 411 fathoms; 1 specimen, 20 mm. in diameter, very well pre- 

 served. The height of the specimen is only 12 mm., but it is obvi- 

 ously so much flattened that the measurement represents much less 

 lhan the normal proportion to the breadth. 



Ma as has given an excellent account of the general organization 

 >f G. geometrica. The reasons for hesitancy in identifying this 

 specimen as geometrica are the relative numbers and succession of 

 tentacles and canals and, more important, the presence of ocelli at 

 the base of the tentacles. 



The relation of the canals to each other and to the stomach, of the 

 same general type as Maas has described it for geometrica, is as 

 follows: each of the four primary radial canals, after leaving the 

 stomach, from which it is not sharply defined, gives off " branches " 

 (the word being used in a descriptive, not an ontogenetic sense), 

 alternatively right and left (pi. 40, fig. 5). In the present series the 

 numbers in each of the four groups are 5, 6, 5, 6. All of these, a 

 total of 22 canals, run to the margin. In each group one canal joins 

 the radial trunk so very close to its juncture with the stomach that 

 it might be spoken of as running to the stomach. But as there is 

 no definite morphologic limit to the stomach this point is not a very 

 important one. There are no blind canals, centripetal or centrifugal. 

 At the margin every canal is connected with a tentacle. Of the 

 22 seventeen are large, evidently fully developed ; five, in every case 

 associated with one of the two outer members of a group of canals, 

 are small and spurlike (pi. 41, fig. 2). There are also three small 

 tentacles unconnected with any canals. At the base of every large 

 tentacle, on the outer side, there is a dense group of red pigment 

 granules (pi. 40, fig. 7) or "ocellus." 



Manubrium and gonads. — These structures agree very well with 

 Maas's account. The mouth is surrounded by a simple quadrate lip. 

 The gonads are purely interradial; each consists of a double series 

 of about 19 regular transverse folds (pi. 40, fig. 6). These folds are 

 opaque and brownish-yellow in the preserved condition. 



On comparing the specimen with Maas's description we find the 

 following differences: in his one specimen there were 32 canals: 

 but though all of these run to the bell margin there were only 16 

 tentacles. When we compare this with the condition in the Philip- 

 pine specimen it is difficult to derive either one from the other by 

 assuming progressive development, because the evidence afforded 

 by the margin — that is, the presence of young tentacles without cor- 

 responding canals — shows that the order of succession is tentacle- 

 canal, just as it is in Calycopsis typa (Bigelow, 1909&). On the 

 other hand, the fact that half the canals have no corresponding 

 74841°— 19— Bull. 100, pt. 5 2 



