Gatfy Marine Lahoratorrj, St. Andrews. 11 



below the main fang — curves forward over the rounded 

 prow. The projection of the base posteriorly is narrow, 

 but somewhat shorter than in Potamilla reniformis. Each 

 is accompanied by a short, broad, cuspidate or penniform 

 bristle. The hooks behind the anterior region (in the 

 fragmentary example) do not materially differ, though tiie 

 neck is longer, the base souiewhat stouter, and their size 

 less. 



The small number of the anterior segments, so unusual in 

 the group (though this number has been found in P. torelli), 

 raises the question as to its relationship to the latter, from 

 which it dilfers in the terminal processes of the branchiae 

 and in the narrower web in the anterior bristles, but further 

 investigations may clear up the divergencies. Like other 

 Sabellids it is acid to litmus paper. In the example from 

 St. i*etcr Port, Guernsey, the tube is composed of a trans- 

 lucent horny secretion, somewhat like that of Potamilla 

 reniformis. Some examples have nearly ripe ova. 



Potamilla torelli, the third form, is closely allied to the 

 foregoing, and could scarcely be distinguished by the bristles 

 and hooks. The general aspect of the cephalic plate, when 

 the branchise are removed, agrees M'ith that of its allies, and 

 in the preserved examples some have a prominent T-shaped 

 projection formed by the developing bases of the branchise — 

 a condition not observed in other forms. The collar has a 

 narrow slit dorsally, then it extends on each side laterally 

 and ventrally with an even edge to the ventral laniellie, 

 which are reflected in protrusion and somewhat triangular 

 in outline, and are separated from each other by a deep 

 V-shaped notch. No eyes are visible in the spirit-prepa- 

 rations from Britain or from Canada. 



The branchial filaments are of moderate length and are 

 pale in the preparations. The structure of each filament is 

 typical, and it ends in a short thick process distally. The 

 puuue are of average length, and it is only at the tip of the 

 organ that shorter forms occur, the last ten or twelve 

 giadually diminishing to end in a short papilla-like rudiment 

 at the base of the distal process. The number of the fila- 

 ments appears to be from twelve to fourteen in each fan. 



When the oral region is in a state of expansion a fold 

 passes on each side from the ventral lamellae u})ward, and 

 Its end fuses with the middle of each branchial fan, and, 

 indeed, appears to be the only representative of the tentacle 

 of other forms. Malmgren states that it is very short, 

 broad, and subcircular. This fold is quite separate from the 



