LARVAE OF SERGESTES 7 



hitherto confounded with 5. cornicuhtm, so that it may be correct to speak of the 

 Elaphocaris of ortmanni type as representing a corniculum group. Apart from the 

 presence of a dorsal spine the Acanthosoma of S. corniitus very closely resembles that 

 of S. corniculum and not its close ally S. atlanticiis. 



Of the two species S. pectinatiis and sargassi only the Acanthosoma of the former is 

 known with certainty, and this seems to have no near relationship with the type 

 characterizing the other species of Hansen's group II. If we are right in the identifica- 

 tion of the Acanthosoma of S. sargassi their relationship seems to be with S. corniculum. 

 Burkenroad (1937, p. 320) has already pointed to the close resemblance of S. sargassi 

 to S. corniculum. 



The petasma offers the most reliable basis for separation of species and should also 

 serve to distinguish groups within the genus. The evidence here shows very clearly that 

 the vigilax-edwardsi group is well founded, but that S. pectinatus and S. sargassi fall 

 outside it, as their larvae also show that they do. It is also very clear that S. atlanticus 

 and S. cornuius are most closely related, but in this case the larvae do not appear to 

 corroborate. Within the remainder of the species it is difficult to distinguish any clearly 

 defined groups on the basis of the petasma. All alike tend to have elongated median 

 stems and the same form of uncinate process. The simplified form of the terminal lobes 

 place S. pectinatus and S. sargassi together and somewhat apart from the rest, while 

 S. arcticus and S. corniculum are both distinguished by the elaborate armature of the 

 ventral process, though the resemblance is not such as necessarily to imply relationship. 

 The remaining species — tenuiremis, robustus, grandis, crassus, splendens, mollis and 

 prehensilis — correspond to the robustus group, in which the Elaphocaris is, or may be 

 expected to be, of the hispida type. 



Burkenroad (1937, p. 317) attaches much importance to the presence or absence of 

 the organ of Pesta. This organ, he says, is present in all species except the S. mollis, 

 S. tenuiremis, S. robustus and S. challengeri superspecies of Hansen's " Group I ", and he 

 also says that S. arcticus, of all species which have this organ, comes nearest to those 

 which lack it. 



Of the remaining genera of the Sergestidae, Petalidium, Lucifer, Acetes and Sicyonella, 

 the development oi Lucifer has long been known, and Soejima (1926) and Menon (1933) 

 have now described fully that of Acetes. Acetes, like Lucifer, hatches as a Nauplius, and 

 the Elaphocaris is of the same hispida type as that of S. robustus, for example, though 

 differing in the much shorter thoracic processes, and shorter telson arms, but there are 

 great differences in later stages. Thus there is only one Acanthosoma stage, which lacks 

 all the spines so characteristic of Sergestes, and the fifth leg does not appear at all. 

 Furthermore, the mouth parts are, to a large extent, degenerated. Another remarkable 

 feature is the appearance of pleopods i and 2 when those behind are absent, and even 

 in Mastigopus i there are only three pairs. Pleopod 4 appears as a non-setose appendage 

 in Mastigopus 2 and pleopod 5 in Mastigopus 3. 



The development of Petalidium and Sicyonella is uncertain. Larvae attributed to 

 Petalidium have been described by Gurney (1924), but we are informed by Mr Burken- 



