DR. RICHARDSON'S DESCRIPTION OF AUSTRALIAN FISH. 91 



cheek is protected by scales, as well as the greater part of the operculum ; its posterior 

 border, and the ineiiibrane between the ridges, alone being naked. The scales even 

 spread over the union of the operculum and suboperculum, though in a scattered man- 

 ner ; the interoperculum is naked. These scales are not visible until the thick epider- 

 mis is dried and rendered transparent by varnish'. A small Scorpeena from Port Arthur 

 nearly resembles bufo in this respect, the scales however not covering quite so much of 

 the operculum. It is known, locally, by the name of " Soldier-fish," and is esteemed 

 for the table. Mr. Lempriere notes the specimen as having been of a " lively flesh- 

 colour" when recent, and as being considerably below the usual size of the species. The 

 colours have been destroyed by maceration in spirits, and the same cause has rendered 

 the existence of membranous processes on the head problematical. A second specimen, 

 rather larger, has more recently been received from the same quarter preserved in brine. 



Form in general that of Scorpeena bufo ; the body is less high and more compressed 

 than that of porcus, and the head is shorter, being less than one-third of the total 

 length, caudal included. The spines of the head are the same in number and similarly 

 situated with those of 6m/o, but are all more prominent and acute, particularly three which 

 occupy the upper margin of the orbit, and are large, hooked backwards, and trenchant. 

 These spines are more distinctly separated from each other by notches than those of 

 porcus. The under preorbitar spine is more slender and acute than in either of these 

 species, and the spiny ridge running backwards from it beneath the eye is more sharply 

 elevated. The interorbitar ridges are much more clearly marked than those of bufo. 

 being exactly similar to those of porcus, except that each terminates in an acute though 

 minute spine. The spines on the hinder part of the head, as well as those on the pre- 

 operculum and operculum, are more slender and acute than in porcus. The scapular 

 bone is armed as in bufo with two small spines, one over the other, and the clavicular 

 spine is the same as in porcus. The nasal spines are not dentated, and the opercular 

 ridges are perfectly simple, not fasciculated as nesoc/allica. 



Scales disposed on the body so as to present oblique rhombs, in a very regular quin- 

 cuncial order. They are larger than those of porcus, are marked on the base with from 

 eight to twelve furrows with corresponding marginal crenatures, and their free edges are 

 finely ciliated with slender, minute and fragile teeth, but they do not show the strong 

 radiating striae which characterise the scales oi porcus. There are about forty scales 

 on the lateral line, nearly every alternate one having a tubular ridge, which is free at 

 the tip. No other cirrhi or skinny filaments are visible on the two specimens. The 

 cheeks are covered by very thin, flexible, circular, oval, and in some places irregular 

 scales ; and the operculum is protected as low as the upper ridge by similar scales, 

 which form a patch running forwards to the orbit, and bounded above by the supra- 



I I could detect no scales on the cheeks or opercula of two specimens of Scorptcna porcus, nor on one of Sc. 

 scrofii, which I received from the Mediterranean. 



N 2 



