DEVONIAN FISHES OF IOWA 103 



Family PLEURACANTHIDAE. 



Body slender, but slightly depressed; mouth terminal; tail 

 diphycercal. Dorsal fin elongate, low, continuous along the back 

 from a point shortly behind the head ; slender interneural carti- 

 lages more numerous than the neural spines. Pectoral fin with 

 biserial arrangement of cartilaginous rays. (Woodward.) 



Genus pleiiracanthus Agassiz. 



The characters of this, the best known genus, have been briefly 

 summarized by Smith "Woodward as follows : 



The fish is elongated and round-bodied, with a straight and 

 tapering (diphycercal) tail. Fixed upon the occipital region of 

 the cranium is a long and slender, bilaterally-symmetrical der- 

 mal spine, which is turned backwards when not erect, and does 

 not appear to have supported a fin. The spine is hollow and is 

 armed with a double longitudinal series of denticles or barbs. 

 The foremost few neural arches are bent forwards, but from the 

 origin of the dorsal fin backwards their direction is normal ; the 

 right and left halves of each remain separate. Ribs are rudi- 

 mentary or absent; but in the caudal region the haemal arches 

 are as large as the corresponding neurals. The branchial arches 

 are believed to have been seven in number, though there may 

 be only five. [Jaekel is positive that the latter number is cor- 

 rect.] The teeth are tricuspid, but the middle denticle is com- 

 paratively minute, and the name Diplodus, commonly applied 

 to the isolated fossil teeth, is thus appropriate. 



Very delicate dermal filaments (actinotrichia) are clustered 

 round the supporting cartilages in all the fins. In the paired 

 fins all the fringing cartilages are segmented ; in the pectoral fin 

 those of the anterior (preaxial) border are much more numerous 

 and better developed than those of the hinder (postaxial) bor- 

 der. The dorsal fin is borne by a close series of slender carti- 

 laginous suports, which are exactly twice as numerous as the 

 neural arches to which they are apposed; each support is seg- 

 mented into three pieces, of which the proximal element is short, 

 the middle one very long, and the outermost again short, while 

 the latter projects into the fin-membrane. The dorsal portion 

 of the caudal fin is similarly supported, but its cartilages are 

 equal in number to the neural arches. The ventral portion of 

 the caudal fin has no supports beyond the haemal spines. The 

 anal fin is curiously subdivided into two portions, of which 

 the supports are crowded, partially fused together, and thus in 



