the Ichthyology of Australia. 209 



the colours of the recent fish, as well as with the markings of 

 cirrhosus recorded in the ^ Histoire des Poissons/ The prin- 

 cipal difference I can detect, on carefully going over Forster's, 

 Lesson^s and Cuvier's descriptions step by step with the spe- 

 cimen before me, is the very slight one of the granulations 

 of the plates on the head not being conspicuously arranged 

 in lines radiating from nine centres like so many stars (Lesson 

 and Cuvier). Forster use» the phrase ^ caput papillis crebris 

 scabymm ordinatis,^ 



The top of the head in the specimen is quite fiat from the occiput 

 to the end of the snout, and across between the temples, and from the 

 outer margin of one orbit to that of the other. The intermaxillaries 

 descend very slightly when protruded. The soft edge of the snout is 

 cut away in a shallow curve over the pedicles of the intermaxillaries, 

 and between the orbits there is a square membranous space. The 

 bony plates which cover the head are very irregular, and anastomose 

 so with each other, that it requires some attention to make out the 

 number mentioned by Cuvier, namely, two rows of four each, and a 

 single rounded occipital plate on the mesial line. The two outer- 

 most plates of the posterior row, and the middle pair of the anterior 

 row, show some granulated lines running forwards and radiating 

 from centres, but all the other plates are rough, with minute rounded 

 points crowded without order. The borders of the orbits are very 

 slightly raised, and the superciliary processes belonging to the middle 

 anterior pair of plates exhibit their granulations in lines. The first 

 suborbitar projects two acute points over the limb of the maxillary ; 

 the second and third are considerably broader, but cover only a third 

 part of the cheek. There is a plate of the same form with them, 

 lying just behind the orbit, and looking like a fourth suborbitar ; it 

 is required, with its fellow, to complete the number four of the an- 

 terior row of cranial plates. All these suborbitars are granulated 

 without order. The preoperculum is curved in the arc of a circle, 

 and is of equal breadth throughout, its upper and lower ends equally 

 obtuse being in the same vertical line. It is coarsely granulated on 

 its upper end, with some faint granular streaks lower down. The 

 operculum is more strongly marked by vertical granular lines, with 

 a few coarser granular points at its upper corner. The roughness 

 of the surface of these bones is concealed by the spongy integument, 

 when the specimen is soaked for a short time in water. The supra- 

 scapulars appear in form of oblong plates, densely granulated, and 

 sloping from the mesial occipital ridge, in conjunction with which 

 they form the boundary of the nape. The humeral bone emits a 

 strong spine, which is slightly curved at the point and not very pun- 

 gent : it is half an inch long, though the tip only protrudes through 

 the integument. The spongy skin of the recent fish will doubtless 

 nearly conceal it. Forster describes it as * spina valida,' and Cuvier 

 as ' tres-courte et presque cachee sous la peau ;' the discrepancy 

 arising, I presume, from Forster having dissected his fish. 



The lateral line icurves gradually from the outer end of the supra- 



