Jordan and Ercrinann. Fishes of North America. 929 



almost imperceptible degrees opposite lobe of anal. Greatest depth of 

 the arch less than diameter of pupil ; the length of straight part less than 

 that of the curve. Plates developed only on the posterior third of the 

 straight part ; the plates small, with low keels, their spines little promi- 

 nent ; 15 to 18 plates developed, including small ones, in front of which 

 are about 40 ordinary scales on the straight portion of the lateral line. 

 Spinous dorsal very small, of 3 weak spines slightly connected by mem- 

 brane, the highest spine not longer than diameter of pupil (these spines, 

 probably, more numerous and larger in young examples). Soft dorsal 

 long and low, with slender rays; a well-developed scaly basal sheath 

 anteriorly. Elevated rays in front a little more than i the base of the 

 tin, a little more than half length of head ; anal shorter than dorsal, its 

 anterior lobe equally high, and with a similar basal sheath. Free anal 

 spines obsolete in typical specimen. Caudal lobes moderate, equal, as 

 long as head, their length equal to the depth of the fin from tip to tip. 

 Pectoral fin falcate, its tip very slender, reaching eighth ray of anal, its 

 length 2| in body. Ventrals short, 2 in head. Coloration in spirits, 

 smutty olivaceous, everywhere irregularly clouded with darker, the belly 

 scarcely paler than the back ; opercular spot obsolete. Dorsal, anal, pos- 

 terior border of caudal, and tips of ventrals blackish ; fins otherwise 

 dull olivaceous. Length 16 inches. Revillagigedo Islands; one specimen 

 known; the species is possibly identical with the one called Caranx 

 fcrdau by Giinther (Fische der Siidsee, 134, 1876), from the Pacific 

 islands. It is, however, certainly different from the original Scomber 

 fcrdau of Forskal, from the Red Sea. (6p0of, straight; ypapw, line.) 

 Caran.c ortfoyraMNMM, JORDAN & GILBERT, Proc. U. S. Nat. Mus., 226, 1881, Sulphur Bay, 



Clarion Island, Revillagigedos. (Type, No. 28345. Coll. Lieut. Nichols.) 

 Carangoides orthogrammus, JORDAN & GILBERT, Proc. U. S. Nat. Mus., 202, 1883. 



422. CITULA, Cuvier. 



Cilula, CUVIER, Regno Animal, Ed. i, 315, 1817, (armaia). 

 Alropus, CUVIER, Regno Animal, Ed. i, 1817, 324, (atropus). 



This genus differs from Carangoides mainly in having one or more of the 

 anterior rays of the soft dorsal produced in filaments. The body has not 

 the distorted form seen in Alectis, but is more like that of Caranx. The 

 young is more like Caranx in form and appearance than the adult,* its 



* Compared with the central species of Caranx (as Caranx laim), Cilula dorsalis differs strikingly 

 iu many respects, among others the following: The greater compression and elevation of the 

 body, the greater length and sharpness of the breast, the much greater depth of the preorbital 

 region, tin greater prominence and ebarpness of the frontal arid occipital keel, the more oblique 

 position of the bones of the head, the feebleness and uniformity of the teeth, the approximation 

 below of the mandibular mini, the reduction in size of the scales and scutes, and the increase of 

 the naked areas, the reduction or loss (in the adult) of the dorsal and anal spines, the increase in 

 length of the lobes of the dorsal, anal, and caudal, and, finally, the increase in the difference 

 between the old and the young. Not one of these features is, however, well adapted for generic 

 distinction, as they increase by easy transitions from species to species. On the other hand, 

 comparing Citiilu ilf>r.*<ilis with S> -Iviu- i-<>ni<r, we find that in almost every one of the differences 

 above noted, the latter species carries these characters to a still greater extreme, and differs from 

 ('i/iil'i dorsalis in just the respects in which the latter differs from <'nr<tn.r Intu*. Voiimr s-v/y/ /,//< 

 is, in some regards, a species still more extreme than Selene vomer, although it has not quite lost 

 the lateral scutes. As a matter of fact the <\ir<ut<jhi;> (excepting J\Iri/<tl<t.*)itx, herujiterug, and pos- 

 sibly Trachnrns) form a continuous, almost unbroken series, only divisible into genera for con- 

 venience sake, beginning with Trachurus on the one hand and ending with Selene on the other. 



F. N. A. 60 



