DISCUSSION OF SPECIES AND THEIR DISTEIBUTION. 163 



stance they are connate or confluent, but the degree of connection depends not so much 

 uirou tlieir proximity to each other as upon the extent of the connecting membrane in the 

 several forms, and we question whether the character can be so defined as to serve even for 

 specific distinctions. All the species examined by us have the peculiar modifications of the 

 posterior extremities of the maxillary, and the sharp spine more or less hidden by the fleshy 

 fold of the lips at the angle of the mouth on either side. 



KEY TO THE GENERA OF NOTACANTHID^ AND LIPOGENYID^. 



I. Jaws normal. Dorsal spines separated. Teetli in both jaws NoTACANTHiDiE 



A. Dorsal spines 6-12. Teeth in upper jaw compressed and obliquely triangular. Ventrals connate or 



confluent Xotacanthinw 



1. Origin of spinous dorsal far in advance of vent. Mouth lateral with lip contiuous.. Ventral fins 



connate or confluent Notacakthus 



2. Origin of spinous dorsal in vertical from vent. Mouth subinferior, crescentie, with lip absent 



in middle. Jaws each with 22 teeth. Ventral fins united Gigliolia 



B. Dorsal spines 27-38. Teeth in jaws erect, fine. Ventrals separated Polyacanthonotinw 



1. Snout proboscis-like. Dorsal and anal spines long, flexible, the latter not exceeding 30 in num- 



ber. Lateral line strongly arched Polyacanthoxotus 



2. Snout not very elongate. Dors.al and anal spines low and strong, the latter 50 or more in num- 



ber. Lateral line straight M acdonaldia 



II. Jaws modified to form a suctorial mouth with separated rami. Dorsal spines close together, united by 



membrane to form a high triangular fin. No teeth Lipogenyidjs 



A. Dorsal spines 5, with 5 soft rays. 



1. Lateral line obsolete behind Lipogenys 



NOTACANTHUS, Bloeh. 



Noiacanthus, Bloch, Abhandl. BiJUm. Gesellsch, 1787. — LACltpiiDE, Hist. Nat. Poiss., 1804. — GooDE, Proc, U. 



S. N. M.,1880, 535. 

 AcanlhaiiotKS, Bloch, Ichthyologia, xii, 1797, 113, pi. ccccxxxi. (No description separate from that of 



species -i. iiasits.) — Schneider, Bloch's Syst. Ichth., 1801, 390, pi. xlvii. 

 Campylodoii, FabriciuS (fide GOnther). 



Head and body much compressed, the body elongate, snout obtuse, rounded at its tip, 

 not proboscis lilie. The cleft of the mouth inferior. Dorsal fin almost rudimentary, con- 

 sisting of 12-15 very short, flexible spines, remote from each other and not connected by 

 a membrane. Anal liu very long, its origin close behind the vent, which is situated nearly 

 midway of the length of the body; its anterior jiortion is comiiosed of separate, flexible 

 spines, without membrane, resembling those of the dorsal; these gradually lengthen, 

 grading into the articulated branched rays. No caudal. Ventrals broad, with broad, 

 peduncle-like bases, clo.sely contiguous, separated only by a slight groove at the base, situ- 

 ated near the vent. Teeth acicular, in single row upon maxillaries, in a double row upon 

 mandibulars, villiform and in double row upon the palatines. Scales very numerous, of 

 moderate size, round, thin, flexible. Branchiostegals about 8. Gills, 4. 



NotdcanfJiHs sexspinis, Richardson, as figured by Giiuther in the Challenger Report, has 

 8 dorsal spines (the last with supplementary ray), instead of the indicated by its specific 

 name ; and this is the case also with the New Zealand specimen in the National Museum, 

 presented by Dr. Giinthei", and one from Cook Strait, New Zealand, in the Museum at Flor- 

 ence, which is 140 millimeters in length, and has a radial formula: D. 8; A. 13/12G; V. 12; 

 V. 1/8; C. 5. These specimens do not show the inflation of the cheeks, figured in Dr. 

 Glinther's plate. 



The results of Dr. Giinther's dissections seem to indicate that this species at least of 

 Notacanthus does not live at a very great depth. 



KEY TO THE SPECIES OF NOTACANTHUS AND GIGLIOLIA. 



I. Origin of dorsal considerably in advance of that of anal. Lip normal, continuous Notacanthus 



A. Body much higher over ventrals than over pectorals, and comparatively short. 



1. Lateral line following profile of back in front of dorsal spines, then sinking to median line of 

 body. D. x-xi. 



a. First dorsal spine behind vertical from axil of ventr.al. A. xiii-xv. (xvii?) N. tiasws 



b. First dorsal spine in front of vertical from insertion of ventral. A. xvn N. analis 



B. Body little higher over ventrals than over pectorals, and comparatively elongate. 



