HQQ THE VOYAGE OF H.M.S. CHALLENGED. 



•broader than long, both margins strongly convex, with long apical spines ; there is a long 

 spine also on the inner surface near the hind margin, and a longitudinal groove in front 

 of the middle of the outer surface ; the fourth joint shorter and narrower than the third, 

 with long feathered spines at each apex ; the fifth joint shorter than the second, longer 

 .than the third or fourth, with a row of four setiform spines on the slightly concave hind 

 margin, three or four on the rounded apex of the front, with a spinule higher up ; the 

 finoer more than half as long as the fifth joint, tapering at first abruptly, then gently, 

 with an opening on the tip. 



Second Perseopods. — Side-plates very broad and very shallow, with a little front 

 lobe carrying a couple of feathered spines at the lower front corner. The branchial 

 vesicles long oval, broader than the preceding pair, and not much apically narrowed. Mar- 

 supial plates like the preceding pairs. The first joint of the same character and size as in 

 the first perseopods, but of different outline, broader below than above, the front margin 

 evenly convex, unangled ; the second joint twice as long as broad ; the third joint longer 

 than the second, longer than the fourth and fifth united, with some marginal spinules 

 besides the apical spines ; the remaining joints much as in the first perseopods. 



Third Perseopods. — The side-plates attached to the lower border of the long segment 

 for almost its whole length, forming a small lobe in the rear, but for the most part of 

 considerable and nearly uniform depth, the slightly crenulate margins armed with setse 

 of moderate length, twenty -four in number, the series beginning about the middle of the 

 front margin and continued nearly to the hinder lobe. From the appearance of this 

 pair of side-plates it may be supposed that they fulfil the function of marsupial plates, 

 dispensing with the necessity for a separate pair of those appendages, and, if this be so, 

 it will help to explain the peculiarity which Professor S. I. Smith has already noticed in 

 regard to the kindred species, Cerapus tubidaris, in which he says the ovigerous lamella? 

 are " only three pairs, and these are borne upon the coxa? of the second pair of 

 gnathopods and of the first and second perseopods." In the figure prp. 3 it is not the 

 proper side-plate of this limb, but the torn and dislocated side-plate of the next segment 

 that appears. The branchial vesicles similar to those of the preceding pair, or a little 

 shorter. The first joint of the limb scpxared, a little wider above than below, with 

 one or two spines near the apex of the slightly convex front margin and some 

 spinules on the lower curve of the hind margin, which projects beyond the short 

 broad second joint ; the third joint is longer than any of the others except the first, 

 and distally nearly as broad as long, the front margin convex, with two little setules 

 near the produced rounded apex, the hind margin sinuous, forming with its rounded 

 apex a narrow lobe produced more than the front, tipped with four very long plumose 

 setse, and a spinule ; the fourth joint is almost embedded in the third, none of the short 

 front margin free, the hind margin convex, thickly furred with adpressed cdia and tiny 

 spines, the rounded decurrent apex carrying a single feathered spine ; the fifth joint 



