96 



bution of the scales on the gill-covers, and in some other minute 

 particulars. Iruttaceus is said to have the intcroperculum and subo- 

 perculum entirely naked, and only a few scales on the operculum 

 itself {" quelques ecailles sur sa surface"). In C. salar there are 

 five rows of pretty large scales on the operculum, one row on the 

 suboperculum, covering surfaces of both these bones, and a row of 

 smaller scales on the intcroperculum, clothing its upper half only. 

 As these scales are very easily detached, and the gill-plates remain 

 hard and silvery, after they are removed with the epidermis, it must 

 be difficult to distinguish an injured specimen from truttaceus ; whose 

 description in other respects exactly accords with salar, except that 

 the latter has the suborbitar very faintly denticulated, and two rays 

 fewer in the soft dorsal. 



3. Aplodactylus arctidens. Aplo. de7itibus oris trieuspidatis, 

 superioribus in serie octuplici, inferioribus in serie qicintuplici 

 dispositis ; cacis pylori quatuor. 



Radii:— Y>x. Q-& ; "P. 9 e< 6 ; V. 1,5; D. 16-1, 17 ; A. 3, 8 ; 

 C. 16i. 



This species differs from A. punctatus of the Chilian seas (the only 

 species j)reviously known) in its dentition, but lesembles it so much 

 in external form, colours, and markings, as well as in anatomical 

 structure, that it cannot be placed in a separate genus. In the 

 Histoire des Poissons the teeth of dentatus are described as follows : 

 " Les dents sent dispostes sur trois rangees a la machoire superieure 

 et sur deux a I'infcrieure : elles sent aplaties et out Icur bards arrondis 

 et dentelrs en petits festons ; elles sont tres-semhlahles H celles des cre- 

 nidens, on en compte quatorze de chaque cole a la mdchoire siipdrieure 

 et treize a V inferieure . Derriere ces j'angtes anterieures il y a des 

 petites dents grenues sur une hande etroite H chaque mdchoire." In the 

 Van Diemen's Land fish, the teeth stand in eight or nine crowded 

 ranks in the upper jaw, and in five or six in the lower one, those of 

 the interior rows being verj' much smaller in all their dimensions, but 

 otherwise shaped exactly like the teeth of the exterior rows, which 

 resemble those of punctatus. Their points show three small lobes, 

 the middle lobe being largest and most prominent. The species 

 further differs from punctatus in having four cseca, but its food ap- 

 pears to be similar, the intestines having been found filled with large 

 fragments of sea weed, apparently Ulva vmhilicalis. 



4 and o. Two new species of gurnard were then mentioned as the 

 first of the genus that have been brought from the Australian coasts, 

 though one species (Trlgla kumu) is known to inhabit the seas of 

 New Zealand. They were stated to agree with that species, with 

 several Indian ones, and with Trigla paciloptera of the Mediterra- 

 nean, in their large pectoral fins being ornamented with eye-like 

 marks similar to those on the wings of some lepidopterous insects. 

 One of them, Tkigla polyommata, has minute cycloid scales on 

 the body, an unarmed lateral line and the dorsal plates confined to 

 the first dorsal, there being no dilation whatever of the interspinous 

 -bones of the second dorsal. All the spines of the head are stiletto- 



