LARVAE OF SERGESTES 5 



Elaphocaris. When a single number is given it is to be understood that this is taken from 

 one typical specimen ; but we have found the number to be variable, though only withm 

 narrow limits. Any considerable departure from the number given would render 

 identification very doubtful. 



The structure of the mouth parts and thoracic appendages throughout the genus seems 

 to be so uniform that we have omitted all reference to them, apart from the mandibular 

 palp, which appears earlier in some species than in others. They have been fully 

 described and excellently figured by Hansen. 



We feel that it is of importance to establish the specific characters of the larvae of the 

 Sergestidae since they are so common in the plankton of the warmer oceanic areas. 

 There is some evidence, though not sufficiently precise to be brought forward, that the 

 Sergestid larval fauna of the waters fished from the Bermuda Biological Station varies 

 with season and year. Since nearly all the larvae can now be identified, at all events in 

 freshly caught material, these changes can now be studied, and it seems probable that 

 they may contribute to throwing light upon seasonal changes in the oceanic currents. 

 It is desirable that the bathymetrical distribution of the larvae in relation to that of the 

 adult should be studied. Nothing is as yet known with certainty as to the oviposition 

 and hatching of the eggs, but it is unlikely that this knowledge can be obtained except 

 by taking plankton at night at depths exceeding 200 metres. 



MOVEMENTS OF THE LARVAE 



Systematic observation of the swimming habits of the larvae have not been made, 

 but the following points have been noted. The Protozoea, like that of the Penaeidae, 

 swims by slow beating of the antennae in a forward direction, generally back upwards. 

 We have not seen vertical movement such as is described by F. MuUer (1863, p. 9) m 



Penaeiis. 



The Acanthosoma swims in exactly the same way as most Carid larvae, namely, back 



down and tail first. 



The swimming of the Mastigopus is difficult to observe, since the confinement 

 necessary under laboratory conditions probably upsets its normal habit. As it is not 

 until stage 2 or 3 that the antennal flagellum is fully developed, stage i , at all events, must 

 have different habits from those of later stages. It would seem that the normal position, 

 at least of some species, is with head upwards, abdomen flexed sharply backwards so 

 that it is almost horizontal and the pleopods widely spread. In this position the legs 

 are held forwards and slightly to either side so that, with the long setae which fringe 

 their two sides, they form an open basket-work perhaps for catching food. The enor- 

 mously long flagella, armed at each segment beyond the "kink" with a pair of curved 

 setae, the points of which nearly meet, are most efficient organs for suspending the body. 

 It was noticed that, when a specimen jumped backwards by sudden ffexure of the 

 abdomen, it was jerked back again by the drag of the flagella. 



