FISHERY BULLETIN VOL 77. NO 3 



ADDENDUM 



Fossil Fishes from California 

 Referred to Scomberesocidae 



We are uncertain of the synonymic status of the 

 nominal genus Scomberessus. based on a fossil 

 from the Miocene (Monterey) formations, intro- 

 duced by Jordan (1920). By context, Jordan pro- 

 posed Scomberessus as a new genus, as follows: 

 "Scomberessus Jordan, 571 [referring to the 

 same item in The Genera of Fishes \. orthotype 

 SCOMBERESOX ACUTILLUS J, & G. (fossil). Differs 

 from the living genus ScOMBERESOX in the much 

 larger dorsal, of 16 rays." But an examination of 

 the text and figures of the two fossil specimens 

 described by Jordan and Gilbert ( 1919: 37-38, pi. 

 XIV, fig. 2. and XVIII — Scomheresox acutilliis and 

 S. edwardsi ) indicate a serious confusion. The one 

 item of diagnosis (dorsal fin) was obviously drawn 

 not from the account and figure of Scorn beresox 

 acutillus Jordan and Gilbert ( 1919:37-38, pi. XIV, 

 fig, 2 [the paratype]), but from the description and 

 figure of Forfex hypuralis Jordan and Gilbert 

 (1919:36-37, pi. XIV, fig. 3). The description of S. 

 acutillus states only "dorsal obliterated," also, the 

 paratype (a complete skeleton examined by us) 

 shows no remaining trace of a dorsal fin. The de- 

 scription of F. hypuralis lists the dorsal rays as 

 "apparently I, 16 in number" and the figure shows 

 a long-based dorsal of approximately the stated 

 number of rays and beginning before the middle of 

 the body (without head). The juxtaposition of the 

 two figures on the plate presumably led the aged 

 master astray. Despite the nonapplicability of the 

 one stated character, the generic name Scom- 

 beressus must, we assume, rest on the designated 

 type-species, Scomberesox acutillus. 



Regardless, we are more concerned with the ref- 

 erence of these fossils to the family Scomberesoci- 

 dae. We have examined the paratype of Scom- 

 beresox acutillus (a complete skeleton but with 

 crushed head), and five essentially complete skele- 

 tons referable (presumably) to S. edwardsi (the 

 holotype is a head and anterior few vertebrae) and 

 have failed to find any finlets — a key character of 

 the family — this despite the listing by Jordan and 

 Gilbert (1919) of". . . traces of five finlets" for S. 

 acutillus (the paratype); under high magnification 

 these traces proved to be isolated scales. 



David ( 1943) may have inferred the presence of 

 finlets by listing counts forS. edwardsi of "Dorsal 

 fin 14. V; anal fin 18. VI . . . ."As Roman numerals 



have long been used to designate spiny or unseg- 

 mented rays, and as living scomberesocids and 

 related fishes all have segmented rays, we assume 

 that David was referring to finlets. However, on 

 examination of David's and other material la- 

 belled S. edwardsi, we find nothing to substan- 

 tiate a count including any "V" or "VI," particular- 

 ly for finlets. 



Each finlet of the Scomberesocidae and Scom- 

 bridae (mackerels and tunas) arises from a single 

 base (ray) that branches into a fanlike structure 

 that is much more robust than a slender, single 

 ray of the dorsal and anal fins proper. Since the 

 individual rays of these fins are distinctly evident 

 on some of these fossils, it is reasonable to expect 

 the heavier finlets also to be preserved or that an 

 imprint at least would have remained. 



The lack of imprint of finlets is substantiated by 

 the absence of any (or imprint) of the supporting 

 bones associated with them. In present scom- 

 beresocids these supporting bones are robust, flat- 

 tened laterally, and lie embedded somewhat 

 parallel to the surfaces of the caudal peduncle 

 rather than extending more or less vertically be- 

 tween the neural and haemal spines, as do those of 

 the rays of the main portions of the fins. Thus, 

 since the supporting rays of the main portions of 

 the fins are often visible in the fossils, it is reason- 

 able to expect such rays of the finlets also to be 

 visible, if present. 



In addition to the appai'ent lack of finlets on 

 these fossils (labeled as of Clarendonian stage), 

 there are notable differences in proportions in 

 lengths of anal bases and caudal peduncle between 

 them and present Scomberesox. In two fossils on 

 which the anal fins appear to be entire ( none have 

 complete dorsal fins) the length of this fin is 

 slightly shorter than the length of caudal peduncle 

 (23.7 vs. 26.5 and 28.5 vs. 33.2 mm). In present 

 Scomberesox the caudal peduncle is about 2.5 

 times the length of either the dorsal or anal fin 

 base, exclusive of finlets. In this regard the fossils 

 approach the condition found in the Belonidae, 

 wherein the length of the caudal peduncle is one- 

 half or less as long as the fin bases; in Ablennes 

 hians the peduncle is scarcely more than one- 

 fourth the length of these bases. Thus, among 

 known marine fishes with both jaws greatly pro- 

 longed into beaks, these fossils are about midway 

 between pre.sent belonids and Scomberesox in the 

 ratio of lengths of caudal peduncle to the base of 

 either dorsal or anal fins < exclusive of finlets in the 

 latter group). An additional difference is a notable 



560 



