22 THE VOYAGE OF H.M.S. CHALLENGER. 



The carapace in all the species is rather large, almost covering the whole anterior 

 division of the body, not, however, being connate, only to a rather limited extent. 

 When viewed from above it exhibits anteriorly (see Pis. IL, III., IV., V. fig. 2) 

 a short linguiform area, sharply marked off behind by a curved line. This dorsal 

 area projects from the so-called stomachal region, and generally does not even 

 reach posteriorly the middle of the caraj^ace. Only within this restricted area, 

 and in front of it, does the carapace constitute the true body- wall ; behind it 

 covers, though very loosely, the trunk, all the segments of which appear well 

 developed in their whole circumference, and exhibit a rather uniform aspect (see 

 PI. VIII. fig. 17). Thus the greater part of the carapace, as in the genus Nehalia, would 

 appear to form, so to speak, merely a loose mantle arching the back and sides of the 

 trunk, and within which the body is freely movable ; a character also regarded by the 

 late Dr. v. Willemoes-Suhm as more particularly ■ distinguishing the present type. 

 This mantle-like portion of the carapace, likewise found, though generally less pro- 

 minently developed, in the other Lophogastrida^, as also in Eucopia and in the Mysidse, 

 exhibits in Gnathoj^ihausia on each side two more or less well-marked longitudinal keels, 

 dividing this part into five areas, the odd one in the middle, limited by both of the upper 

 keels, with a lateral area on each side, between the upper and lower keels, finally a much 

 smaller marginal area, extending between the lower keel and the free edge of the 

 carapace, the last being strongly inflected toward the ventral face. The lateral areas 

 occupying most of the sides of the carapace, are bounded anteriorly by an obliquely 

 descending elevated line, corresponding, in position, with the branchiostegal line of the 

 higher Podophthalmia, and terminating just above the liuccal area. Immediately below 

 this line, the carapace forms a more or less marked lateral expansion, jutting out in some 

 species into a strong, exteriorly pointing projection, which I regard as the branchiostegal 

 spine. 



Viewed from below (see PI. IV. figs. 3 and 4), the free margins of the carapace will be 

 found to exhibit on each side a deep, nearly semicircular emargination, into which 

 the exognath of the second pair of maxillse fits. Moreover, this emargination is 

 bounded anteriorly as well as posteriorly by an oljtuse linguiform lobe bent inwards ; 

 the anterior lobe partly overlaps the body of the mandible, whereas the posterior 

 lobe inclines toward the base of the maxillipeds. Immediately anterior to the above 

 mentioned lateral expansion of the carapace, a strong spine projects in most of the species, 

 pointing obliquely forward, which, as to its position on the side of the basal part of the 

 antennae, may properlj"- be regarded as the antennal spine. Finally, just above the eyes, 

 or on each side of the base of the rostrum, the frontal i>aYt of the carapace usually juts 

 out into a similar, and often rather strongly developed spine, undoubtedly corresponding 

 to the supraorbital spine in other Podophthalmia. 



The rostrum is generally rather elongate and slender, though sometimes very broad 



