Enneanectes Jordan and Evermann, 1895

David G. Smith & Jeffrey T. Williams, 2002, History and status of the genera Enneanectes and Axoclinus (Teleostei: Blennioidei: Tripterygiidae)., Zootaxa 105, pp. 1-10 : 2-7

[[ Enneanectes Jordan and Evermann, 1895 ]]

History of the problem

The problem involves two distinct biological species, three nominal species, and three nominal genera. The names have been applied to different species at different times. The two biological species are characterized mainly by the configuration of the lateral line (Fig. 1). In the first species, which we will call Species 1, the lateral line is continuous, with an anterior section of tubed scales extending to approximately the origin of the third dorsal fin followed by a few notched scales. The lateral line descends gradually from the upper end of the opercle to the midlateral line, and there are 4-5 scale rows between the lateral line and the first dorsal fin. The dorsal rays are III + XII + 9-10, and the anal rays are II, 16-17 (Allen and Robertson, 1994: 219) In Species 2, the lateral line is discontinuous. The anterior section of tubed scales extends more or less horizontally from the opercle to approximately the posterior end of the second dorsal fin; it is followed by a series of notched scales that begins two scale rows ventral to the anterior section. There are three rows of scales between the lateral line and the first dorsal fin. Dorsal rays are III + XI-XII + 7-8, and anal rays are II, 14-15 (Allen and Robertson, 1994: 221).

The nomenclatural history begins with the description of Tripterygium carminale by Jordan and Gilbert (1882: 362), based on four specimens collected in a tide pool at Mazatlan, on the Pacific coast of Mexico. The lateral line is described as “extending to opposite last ray of soft dorsal, ascending anteriorly but without convex curve.” Fin ray counts are given as D. III - XI, 9, A. II, 17. Based on the form of the lateral line and the fin-ray counts, Tripterygium carminale agrees with Species 1. The holotype was given as USNM 28118, but neither it nor any of the three paratypes can be found today. It is assumed that they were destroyed in the fire at Indiana University in 1883 (Brittan, 1997: 234), as no mention of them exists after that time. Jordan often received USNM catalog numbers for specimens on which he and his group were working at Indiana; the specimens would be forwarded to the Smithsonian when the work was completed. If these specimens existed today, they would be either at the USNM or at the CAS, where the Indiana University collection was ultimately transferred. We confirm that the specimens are not present in the USNM collection, and they were not mentioned by Böhlke (1952) in his type catalog of the Stanford collection (which then held the former Indiana University collection and was subsequently moved to the CAS).

Thirteen years later, Jordan (1895: 501) placed the species epithet in a new genus, Enneanectes , based on a very brief diagnosis of the characters in which carminale differed from Tripterygion Risso (the correct spelling). Jordan stated that Enneanectes was “framed...by Jordan and Evermann,” suggesting that it was to be published in the upcoming“Fishes of North and Middle America” (Jordan and Evermann, 1898), which indeed contains a full description of the genus, attributed to Jordan and Evermann, 1895. It is significant that Jordan and Evermann did not use one of the types in their description, but rather a small specimen collected by Jordan and his colleagues during a later trip to Mexico. This is further evidence that the types were missing at least as early as 1895. They also expressed some doubt about the identity of their specimen, stating that it “...differs somewhat in the count of the fin rays; but the very small size of the specimen prevents us from being entirely sure of its correctness.”

Brock (1938: 131) collected two specimens that he identified as Enneapterygius carminalis from Cape San Lucas, Baja California. He also briefly discussed the generic problem, saying that the specimens agreed with Gillias jordani Evermann and Marsh, 1899 , “in the generic characters they used to separate Gillias from Enneanectes ....Even if Enneanectes be withdrawn from the synonymy of Enneapterygius , Gillias must still remain a synonym of Enneanectes .” Brock thus framed the nomenclatural question that would persist over the next 60 years. Two years later, Brock (1940: 33) revisited the problem in more detail. He decided that the specimens (Species 1) he had identified as E. carminalis were not the species described by Jordan and Gilbert, and he therefore described them as a new species, Enneapterygius storeyae (he continued to regard Enneanectes as a synonym of Enneapterygius ). He also discussed the confusion between the species called here E. carminalis and E. storeyae , asserting that the original description of carminalis could apply to either species. In order to resolve the ambiguity, and in the absence of a holotype, Brock designated a neotype for carminalis . The specimen (Species 2) he chose was SU 3854, the specimen Jordan and Evermann (in Jordan 1895) described as representative of the type species of Enneanectes . He selected this specimen on the perfectly rational assumption that Jordan would have been able to recognize the species he himself had described 13 years earlier. Brock also noted that the “American members of the genus Enneapterygius ” differ from the Indo-West Pacific species in the scalation of the head, and, if this character is agreed to be of generic significance, “ Enneanectes must replace Enneapterygius , and, of course, Gillias , for these American species.”

Fowler (1944: 286) described a new species, Gillias sexmaculatus , from the Pacific coast of Panama. He described a discontinuous lateral line and three scales between the lateral line and the spinous dorsal-fin origin, characters that align it with Species 2. He emphasized its distinction from the description of carminalis Jordan and Gilbert and its resemblance to the Atlantic Gillias jordani Evermann and Marsh . Fowler was thus basing his interpretation on Jordan and Gilbert’s original description of carminalis (Species 1) and not on the specimen from Jordan (1895), which Brock had selected as the neotype (Species 2). Fowler described a new genus, Axoclinus , based on a new species A. lucillae . Axoclinus was distinguished mainly by the lateral line, described as “axial along side of body, incomplete, only extends as far as third dorsal.” Although Axoclinus lucillae is distinct from Species 1 in its scale and fin-ray counts, the lateral-line character agrees, and Axoclinus is the genus in which Species 1 would be placed.

Fowler’s treatment of the nominal species carminalis and storeyae is confusing and contradictory. On page 288, he accepted Tripterygium carminale Jordan and Gilbert as the “genotype” of Enneanectes , listing some characters from the description (i.e., Species 1, not from Brock’s neotype, which represents Species 2) that “are all out of harmony with Gillias sexmaculatus .” Aside from noting its distinction from his new species sexmaculatus , he did not treat it further. Also on page 288, he noted that “the Gillias storyae [sic] (Brock) ” also belongs in Gillias , although he further noted that it differed in having “5 rows of scales between the lateral line and the base of the first dorsal, though the course of the lateral line is not described” or illustrated. Without knowing the condition of the lateral line, it is difficult to understand how Fowler determined that storeyae belonged in Gillias . Then on page 289, he stated that his new species Axoclinus lucillae “approaches Enneapterygius storeyae Brock but [is] apparently quite different in coloration.” Here he seemed to be saying that storeyae belonged in Axoclinus rather than Gillias , where he had placed it on the previous page.

Schultz (1950: 268) added to the confusion by synonymizing the genera Enneanectes and Gillias with Tripterygion Risso and Axoclinus with Helcogramma McCulloch and Waite . He based this decision on the form of the lateral line, i.e., continuous in Axoclinus and Helcogramma , discontinuous in Enneanectes , Gillias , and Tripterygion . The species carminalis (i.e., based on the Jordan and Gilbert description, Species 1, not on the Brock neotype, Species 2) was therefore transferred to Helcogramma , and storeyae and lucillae were included as synonyms. The species sexmaculatus was placed in Tripterygion . Schultz also claimed to have found the holotype of carminalis , USNM 120946, which he removed from a syntypic lot of Gobiesox zebra Jordan and Gilbert . Although the specimen was collected at the same time and place as the type series, its type status is doubtful. In the first place, Jordan and Gilbert (1882: 363) said that the four types of carminalis were “each about 1 1/2 inches long,” USNM 120946 is barely one inch in standard length and 1 1/4 inches (33 mm) in total length. Second, why would Jordan and Gilbert have put the holotype back into a jar full of another species? It seems much more likely that the specimen was just a stray that was mixed in with the type lot of Gobiesox zebra ; in their description of G. zebra , Jordan and Gilbert (1882: 359) said only that “about 30 specimens” were collected, suggesting that they did not count them very carefully or even look at them all. If the holotype of carminalis really existed, as Schultz clearly believed, then Brock’s neotype designation would be invalid. Schultz followed this logic and stated that Brock’s “neotype ” (quotation marks used by Schultz) “does not belong to this genus and species [i.e., Helcogramma carminale as used by Schultz] but is a specimen of Tripterygion sexmaculatus (Fowler) .”

Rosenblatt (1960: 3) resurrected Enneanectes , pointing out its differences from both Tripterygion and Enneapterygius (which Schultz had also synonymized with Tripterygion ). He then summarized the nomenclatural history of Enneanectes , emphasizing again that the specimen of “ carminalis ” used by Jordan and Evermann (in Jordan 1895) represented a different species (Species 2) from the one originally described by Jordan and Gilbert (Species 1) and that Brock’s storeyae represented “the true T. carminale .” He outlined two alternative scenarios that would flow from this analysis and gave his reasons for choosing one of these scenarios:

It would seem that the most logical course is to recognize as the type of Enneanectes the actual specimens upon which Jordan and Evermann based the generic description, namely Gillias sexmaculatus Fowler . The alternative to this [i.e., to recognize the nominal species that Jordan and Evermann had misidentified their specimen as] would be to replace Axoclinus with Enneanectes , and refer the species here placed in Enneanectes to Gillias . This would mean that not one of the species of Enneanectes would then agree in any generic character with the original description of the genus. This latter alternative, while admissible nomenclatorially, would create a zoological paradox. It is for this reason that Gillias sexmaculatus is here recognized as the type species of Enneanectes . (Rosenblatt, 1960: 5)

Rosenblatt, however, was still under the impression that a holotype existed for Tripterygion carminale and that the name carminale applied to the species actually described by Jordan and Gilbert (1882). In his view, Brock’s (1940) neotype designation was invalid, as evidenced by his use of quotation marks around the word “neotype.” As we have explained above, however, Schultz’s “holotype” is not a holotype, and Brock’s neotype designation stands.

Almost all subsequent authors have followed Rosenblatt’s interpretation and assigned species to Enneanectes according to his concept of that genus. We note in particular Allen and Robertson (1994: 218), which has become a standard reference to fishes of the eastern tropical Pacific. Those authors dealt with all the eastern Pacific species of Tripterygiidae and explained the lateral-line feature that characterizes Enneanectes and Axoclinus . A prevailing consensus has thus been achieved on the definition of these two genera and on the species that are assigned to each.

This consensus has been challenged by Fricke (1997: 564). He recognized Brock’s neotype designation (though incorrectly attributing it to Schultz) for carminalis as fixing the name to Jordan and Evermann’s (1895) species (i.e., Species 2, not the one originally described under that name by Jordan and Gilbert, Species 1). However, he incorrectly included storeyae as a junior synonym of carminalis instead of treating it as the next available name for the original carminalis of Jordan and Gilbert (Species 1). He correctly included sexmaculatus as a junior synonym of carminalis of Brock (Species 2). Then he incorrectly synonymized Axoclinus with Enneanectes and transferred the species currently assigned to Axoclinus to Enneanectes , along with carminalis . The remaining species currently in Enneanectes were transferred to Gillias . Fricke did not explain why one of the current Enneanectes species ( carminalis ) remained in his concept of Enneanectes while all of its congeners were moved to Gillias .

Conclusion

The resolution of this case rests upon whether Brock’s designation of a neotype for Tripterygion carminale Jordan and Gilbert is accepted. If it is, then the specific name carminalis is applied to Species 2 and assigned to the genus Enneanectes . If the neotype is not accepted, then carminalis is applied to Species 1, and Enneanectes goes with it. It is fairly clear from the description that Tripterygion carminale as described by Jordan and Gilbert refers to the taxon we call Species 1. Jordan and Evermann (in Jordan, 1895) misidentified the specimen they chose as the type species of Enneanectes ; it actually belongs to the taxon we call Species 2. Article 70.3 of the ICZN addresses the problem of misidentified type species.

If an author discovers that a type species was misidentified..., the author may select, and thereby fix as type species, the species that will, in his or her judgment, best serve stability and universality, either 70.3.1. the nominal species previously cited as type species, or 70.3.2. the taxonomic species actually involved in the misidentification. If the latter choice is made, the author must refer to this Article and cite together both the name previously cited as type species and the name of the species selected (ICZN, 1999: 74).

We hereby invoke Article 70.3.2 and select “the taxonomic species actually involved in the misidentification,” i.e., the species represented by the specimen designated by Brock (1940) as the neotype of Tripterygion carminale Jordan and Gilbert, 1882 . This maintains prevailing usage by fixing the genus Enneanectes to the species that have been almost universally assigned to it since 1960. Enneanectes carminalis (Jordan and Gilbert, 1882) becomes the valid name of the taxon we call Species 2, with Gillias sexmaculatus Fowler, 1944 as a junior synonym. The taxon we call Species 1 takes the next available name and becomes Axoclinus storeyae (Brock, 1940) . Table 1 summarizes the names applied to the two species by the various authors. Table 2 summarizes the historical and current placement of the species included in the two genera Axoclinus and Enneanectes .