1854,] EGERTON — FOSSIL FISH FROM EGYPT. 375 



perfect condition. The anterior portion of the endo-skeleton is 

 displayed in several of the specimens, together with the dermal appa- 

 ratus for supporting the dorsal and anal fins. Unfortunately the 

 precise character of the dorsal fin is not recognizable, nor can the 

 soft rays of any of the fins be counted with accuracy. The branchio- 

 stegous rays are also indefinable, both as to form and number ; and 

 the vomer and palatine bones are not visible. 



These deficiencies render doubtful the determination of the family 

 or genus to which the species is to be referred. Enough, however, is 

 preserved to afford, if not conclusive evidence, at all events an 

 approximation to the truth. 



There is no question but that all the specimens belong to one and 

 the same species, and the crowded position of their remains warrants 

 the conclusion that it was a gregarious species. The characters of 

 the scales and fin-rays determine it to be a true Ctenoid of Agassiz, 

 or Acanthopteroid of Miiller, and belonging to one of the most 

 typical families of the Order. 



In the absence of the materials above alluded to, the preoperculum, 

 which is well seen in several of the specimens, is a feature of some 

 consequence. This bone, as in the true Perches and the Scieenoids, 

 is serrated on its outer margin ; a character which is not found in any 

 of the Sparoid family. These serrations on the posterior border are ' 

 very decided and regular, and have their points directed upwards. At 

 the angle of the bone are five or six strong diverging spurs, forming 

 a powerful armature to this portion of the opercular flap. These are 

 succeeded on the lower margin by a series of sharp points directed 

 forwards, and diminishing in size until they Ijecome obsolete at the 

 anterior angles. 



I have been unable to discover a similarly constructed preoperculum 

 in any genus of the Percoid family. In Perca the strongest spurs 

 are at the anterior angle of the lower border, and the serrations on the 

 posterior angle are small and irregular. In Cyclopoma and Labrux, the 

 same. Lates and Holocenti'um have a single spur at the posterior 

 angle. Enoplosus has fine serrations on the posterior margin, and 

 those on the lower margin, although stronger, are directed backwards. 

 In Serranus the serrations are fine and uniform. 



In the Scisenoid family the preoperculum is serrated, but not so 

 strongly as in the Perches. The Labroids have no serrations on the 

 opercular bones. 



It appears hence that the preoperculum in the Egyptian fish, 

 although more nearly resembling the forms of that bone prevailing 

 in the family of the Perches than in the Labroids and Scisenoids, has 

 nevertheless an appropriate form, differing from that found in any 

 known genus of that family. The operculum seems to have had a 

 similar outline to that of our common Perca fliiviutilis ; but, as the 

 margin is imperfect, it is uncertain whether the posterior angle was 

 so much produced. 



Assuming then that the fossil as regards the preoperculum is more 

 nearly allied to the Percoids than to any other family, we must con- 

 sider how far the other details will bear out that assumption. Tbe 



VOL. X. PART I. 2d 



