88 



BULLETIN 93, UNITED STATES NATIONAL MUSEUM. 



The labrum (fig. 17&) has about 12 teeth on each side of the notch. 

 Mandible as in B. amphitrite. Maxilla (fig. 17e), with six spines 

 between the upper and lower pairs. It is like some specimens of 

 B. amphitrite niveus, having the lower pair of large spines on a slight 

 projection. The labral palpus (fig. 17d) is like that of B. ampJiitrite 

 niveus, in having an external row of long spines, but the opposite face 

 (not shown in the figure) has a very serried band of short spines 

 directed toward the labrum, instead of a single row, as in amphitrite. 



The first cirrus has ranri of 19 and 14 segments, more unequal than 

 usual, the posterior ramus about three-fourths as long as the anterior. 

 Cirrus ii has slightly unequal rami of 12 and 10 segments. Cirrus iii 

 has 15, 14 segments, but the posterior ramus is but little more than 

 two-thirds as long as the anterior. The longer ramus has few long 

 spines on the lower 9 segments, but there are a few short ones on and 

 adjacent to the anterior margin (fig. 17/); also spines along the 

 posterior outline. The inner faces of these segments have very few 

 spines or none. The shorter ramus has armature like the longer on 

 the lower five segments, but their inner faces are densely covered 

 with long spines. The distal segments of both rami are densely 

 covered with long spines. Cirrus vi has five pairs of spines, with some 

 minute ones near their bases and along the posterior outline. 



Another individual of the same lot has cirrus i with 18 and 12 

 segments, posterior ramus four-fifths the length of anterior. Cirrus ii, 

 11 segments on both rami, anterior ramus longer by two. Cirrus iii, 

 14 and 11 segments. The labrum (fig. 11 a] has only jive teeth on cadi 

 side, so that at first I thought it could not be improvisus; but as all 

 other characters are exactly like the others of the same colony exam- 

 ined, 1 am reluctantly obliged to admit that what was thought to be 

 one of the most tangible differences between improvisus and amplii- 

 trite is not a wholly reliable criterion, though it holds in all other 

 specimens I have dissected. 



