58 DR. W. G. RIDEWOOD ON THE CRANIAL [May 3, 



the parietal bones during evolution must be borne in mind ; and if 

 the Berycoid fishes are to be I'egarded as the parental stock of the 

 Acanthopterygians genei'ally, a view which is supported by both 

 comparative anatomy and palaeontology, all instances of contact 

 between the parietal bones among the Acanthopterygian fishes 

 (e. g. Cyttidse, Scorpsenidse, Triglidse) are to be looked upon as 

 secondary. In support of the plausibility of the hypothesis may 

 be mentioned the parallel instance of union of epiotic bones in the 

 middle line in such forms as Lophius, Regalecus, and Luvarus, 

 this union being without doubt of secondary origin. 



In Chanos there exists a condition which is calculated to make 

 one pause before concluding that even within the limits of the 

 Malacopterygian fishes the meeting of the parietals necessarily 

 indicates the retention of the primitive condition ; for in this genus 

 the parietals are widely separated in the young, but by subse- 

 quently fusing with the scales of the commissural section of the 

 sensory-canal system, they come to meet above the supraoccipital 

 bone *. They do not meet in an extensive suture ; indeed, they 

 leave a considerable portion of the supraoccipital exposed both in 

 fi'ont and behind ; but the condition is just sufficient to make it 

 advisable to trace the development of the roofing-bones of the 

 cranium in those forms in which the meeting parietal bones are 

 thin, and beneath which bones the supraoccipital extends a con- 

 siderable distance forward. This last relation, it may be observed, 

 is the rule rather than the exce]3tion. Boulenger has already 

 alluded to it in the case of the Salmonidse (Proc. Zool. Soc. 1895, 

 p. 300), and I gather that he regards the union of the two 

 parietal bones as secondary if the supraoccipital bone can be 

 shown to extend beneath them so as to touch the frontal bones. 

 But if this be so, the condition found in those primitive genera 

 Elops and Alhula must be secondary, for in them the supra- 

 occipital touches the frontals beneath the united parietal bones ; 

 a fact evidently overlooked by Boulenger when he drew up his 

 synopsis of the families of Teleostean fishes (Ann. & Mag. Nat. 

 Hist. 1904, vol. xiii. p. 164). 



In Clupea, Chatoessus, and Chirocentriis the cranial roof is 

 deficient towards the front of its upper surface, and a fontanelle 

 occurs between the frontal bones and the mesethmoid. Perhaps 

 this tendency for the frontals to remain apart points to some 

 affinity with the Characinidse and Cyprinidse. In Citharimos such 

 a fenestra extends the whole length of the frontal bones and 

 involves also the parietals, while in Alestes a fenestra is found 

 between the parietals and the hinder part of the frontals. A 

 condition similar to the last occurs in such Oyprinoids as Cato- 

 stomiis, Ci/p?'inuSj and Cohitis (see Sagemehl, Morph. Jahrb. x. 

 and xvii.). 



The interfrontal suture is obliterated in Gonorhynchus, but it 

 does not appear that much weight need be attached to this 



* Cope (Trans. Amer. Phil. Soc. ii. s. xiv. 1871, p. 455) mentions "parietals 

 united" as one of the primary distinctive features of the family Lutodiridas. 



