THE CH.ETOGNATHA. 67 



15 hauls which closed between 300 and 50 fathoms, and in every haul with an open net 

 in which the greatest depth touched was less than 350 fathoms. S. macrocephala was 

 therefore undoubtedly mesoplanktonic. 



Sagitta zbtesios, sp. n. (PL 5. figs. 22-27.) 



Characters. — Head jiroportionately of medium size, larger than the body immediately 

 behind it, thiis exhibiting a marked neck. Body fairly stout (stouter than bipimctata), 

 tapering from the middle gradually towards each end ; a collarette present; lateral fields 

 rather narrow, longitudinal muscles well developed ; tail 25 to 33 per cent, of the total 

 length. 



Posterior fin triangular, slightly more on the body than on the tail, its widest jMrt at 

 or slightly in front of the septum, not extending backwards to the vesiculae seminales. 

 Anterior fin well separated from the posterior, longer and narrower than the posterior, 

 wider behind than in front, not quite or only just reaching the hinder end of the ventral 

 ganglion. Tail-fin truncate. 



Jaws very slender and strongly curved. 



Teeth very long and slender, light-coloured (unlike macrocephaUi) ; the anterior teeth 

 increasing in size ventrally ; the posterior teeth at first increasing in length inwards, 

 then decreasing again in the few innermost. 



Vestibular ridge well developed, but covered by a thin cuticle only, very slightly 

 mammillated by numerous small and irregu.lar prominences, not extending for the 

 whole length of the row of jiosterior teeth. 



Corona ciliata not observed. 



This is a fairly well-marked species. The lai'ge numbers of posterior teeth are enough 

 to separate it fiom all but serraloclentata, bipunctata, Bedoti, hispida, and elegans. 

 Prom serratodentata the absence of serration on the jaws is enough to distinguisli it. 

 As compared with bqnmctata, the head is larger, the body stouter, the lateral fins 

 wider, the teeth very much longer. Bedoti (Beraneck), of which I have sjiecimens 

 from the ' Siboga ' Expedition, has a much slenderer outline, and zetesios may be 

 distinguished readily by the larger number of jaws and smaller numbers of anterior 

 and posterior teeth in specimens of the same size. S. elegans (Verrill) has been 

 insu^fficiently described and figured, so that it is not easy to point to salient characters 

 for differentiation ; but at least the proportion of the tail-segment to the total length and 

 the number of jaws are easily checked. >S'. hispida (Conant) has a quite different form: 

 the absence of a neck, the length and shape of the posterior fins, and the slender body 

 all mark it off from zetesios, though the numbers of jaws and teeth are almost identical. 



The following table has been summarized from 38 specimens. The smallest specimen, 

 8 mm. long, is possibly referable to some other species. 



The specimen of 32 mm., which is at the head of the table, is represented in fig. 23. 

 I have no doul)t that it really belongs to this species, but that some curious action of 

 the preserving-fiuid has swelled out the epidermis in front of the anterior fin in a way 

 which is common in my specimens oi f areata and some other species. This was the 

 only specimen in which the vesiculie seminales projected noticeably. The vestibular 



