LEODICID^} OF THE WEST INDIAN BEGION. 81 



the body; 10 is the usual number in this region. The filaments are slender, arising 

 from a base which is at first stout but becomes smaller as each branch is given off, its 

 apex making the terminal branch. The gills have a brownish color, due largely to 

 pigment, though a part of their color is due to the contained blood. The gills are con- 

 tinued to almost the posterior end, the terminal ones consisting of a single slender filament. 



Anteriorly the aciculae have smooth ends, but posteriorly there are two ventrally 

 placed hooded aciculaB (text-figure 284). 



The maxilla (text-figure 285) is dark-colored, the carrier being black, the remainder 

 dark brown with black margins. The carrier is wider than long, its lateral wings rounded, 

 the forceps heavy. The proximal paired plates have 8 teeth on the right and 9 on the 

 left, the toothed margin being noticeably straight. The distal paired plates have each 

 6 teeth, the unpaired has 5 anterior larger and 4 posterior smaller ones. Lateral 

 to the distal paired plates are 2 smaller ones on either side. The mandible (text-figure 

 286) is very small as compared with the maxillae, its shafts slender, smooth, dark brown, 

 the beveled edge white with much brown marking. 



The animal lives in tubes having a basis of a parchment material. Most of the 

 tube is buried vertically in the sand, but a portion protrudes from the surface. The 

 latter is covered with small bits of shell or other debris from the beach (text-figure 287). 

 The tube is much longer than the animal, so that in case of danger it can retreat to a 

 considerable distance below the surface. A small specimen was observed to turn around 

 in its tube, and it is probable that larger ones are able to do the same, since the head 

 may be found at either end of any given tube. The eggs are laid in jelly masses and a 

 free-living ciliated larva results (plate 7, figure 5). These were obtained from specimens 

 kept in a live-car at the Dry Tortugas (the animals were brought from the Marquesas) 

 on July 15, 1914. I have no data as to the age of the larva figured. 



Onuphis magna occurs in considerable numbers at the Marquesas, in sandy mud. 

 In 1914 some were taken for study to Loggerhead Key in the Dry Tortugas and spawned 

 in a live-car on the west side of the island. In 1915 tubes were found on the east side 

 of the island and one was dug at Long Key. Since I had not seen it in the Tortugas in 

 earlier collecting, it seems possible that those collected in 1915 were descended from 

 those brought to the island the preceding year. 



I have compared these specimens with Andrews's type of Diopatra magna in the 

 II. S. National Museum, and find that they are identical. Having pectinate gills, they 

 belong to the genus Onuphis instead of Diopatra. Andrews (1891a, p. 287) states that 

 this is more common than Diopatra cuprea at Beaufort, North Carolina, and he thought 

 that it was this species which lays eggs in long, cylindrical masses of jelly. He records 

 it as measuring in some cases "upwards of 4 feet" in length. Andrews has also (18916, 

 pp. 113-120, plate vn, figures 1-7) described peculiar strings of cells attached to theovarian 

 eggs of this species, which he thought did not have any relation to the nourishment of 

 the ova, but possibly are mechanical suspensors holding the ova out into the nutritive 

 fluid contained in the ccelom. Some hitherto unpublished observations by the writer 

 on Diopatra cuprea at Woods Hole demonstrated a definite communication pore between 

 the ovum and the first cell of the chain, indicating that they are true "nurse" cells. 



Andrews (18916, p. 115, footnote) states that the larvae described by Wilson (1882, 

 pp. 288-291) as belonging to Diopatra cuprea are really of this species. Wilson, however, 

 says that the gills are at first simple and only later acquire the spiral character. It is 

 not clear whether Wilson saw this change in the gills or whether, knowing the gills of 

 the adult D. cuprea to be spiral, and assuming these to be the larvae, he inferred that 

 this change occurs. If the former, the larvae which he figured were not Onuphis magna; 

 if the latter, they might have been, and Andrews's description of the form of the 

 egg-string agrees with those I have seen in Florida. 



