Comprehensive Description
provided by Smithsonian Contributions to Zoology
Istiblennius edentulus (Schneider)
Blennius edentulus Schneider in Bloch and Schneider, 1801:172 [Society Islands, Tahiti; neotype, USNM 292529, designated below in nomenclatural discussion section].
Salarias quadricornis Valenciennes in Cuvier and Valenciennes, 1836:329 [type locality restricted to Isle-de-France and lectotype, MNHN A.2003, designated below in section on nomenclatural discussion].
Blennius truncatus Forster, 1844:231 [Society Islands, Tahiti; neotype, USNM 292529, designated below in section on discussion to be same specimen as neotype of Blennius edentulus].
Salarias melanocephalus Bleeker, 1849:18 [Pagotang, Javae; RMNH 6297, 14 specimens, probably syntypes, largest, 94.4 mm TL, possible holotype, designated lectotype below in section on nomenclatural discussion].
Salarias sumatranus Bleeker, 1851:256 [Sumatra occidentalis; location of holotype unknown].
Salarias diproktopterus Bleeker, 1857b:69 [Kajeli, Boeroe; holotype RMNH 4468].
Blennius cinereus Castelnau, 1875:26 [Queensland, One Tree Island; neotype, USNM 291700, designated below in section on nomenclatural discussion].
?Salarias atratus Macleay, 1882:361 [Port Moresby, New Guinea; 2 putative syntypes, MAMU F1047, apparently lost].
Salarias insulae Ogilby, 1899:741 [Lord Howe Island; holotype apparently lost].
Scartichthys enosimae Jordan and Snyder, 1902:460 [Misaki, Japan; holotype, CAS-SU 7068].
Salarias gilberti Bryan and Herre, 1903:135 [Marcus Island; holotype included in BPBM 2454].
Salarias marcusi Bryan and Herre, 1903:136 [Marcus Island; holotype, BPBM 2456].
Scartichthys basiliscus Fowler, 1904:552 [Padang, Sumatra; holotype, ANSP 27802].
Salarias azureus Seale, 1906:87 [Tubai, Austral Ids.; holotype, BPBM 783].
Salarias sindonis Jordan and Seale, 1906:427 [Pago Pago, Samoa; neotype, USNM 51793, designated below in section on nomenclature].
Salarias garmani Jordan and Seale, 1906:429 [Apia, Samoa; holotype, USNM 51792].
Salarias rechingeri Steindachner, 1906:1411 [Western Samoa, Upolu; neotype, USNM 293747, designated below in section on nomenclatural discussion].
Salarias fluctatus Fowler, 1945:70 [Saipan Island; holotype, ANSP 71603].
Salarias atrimarginatus Fowler, 1946:182 [Aguni Shima, Riu Kiu Islands; holotype, ANSP 72052].
Istiblennius edentululus.—Allen and Swainston, 1988:130 [Northwestern Australia; misspelling].
DESCRIPTION.—Dorsal fin (Tables 27 and 28). XII to XIV, 18 to 23 = 31 to 36 (XIII in 95% of specimens); segmented rays rarely 18 or 23; total elements rarely 31 or 36); mean number of total elements usually higher for males from any locality than for females from same locality (higher for 55 of 61 localities for which means for both sexes are available, statistically significantly higher for 14 of the 55 localities); membrane between spinous and segmented-ray portions notched deeper than half length of first segmented ray; fin membrane from posteriormost ray attaching to point on dorsal edge of caudal fin posterior to vertical through bases of central segmented caudal-fin rays in specimens >25 to 40 mm SL, attachment extending along edge up to 25% caudal-fin length; in general, attachment extending farthest posteriorly in largest specimens.
Table 27.—Frequency distributions for certain meristic characters of male specimens of Istiblennius edentulus from various localties. Underlining indicates significant differences between means of sexes from same locality (p ≤ .05).
Anal fin (Tables 27 and 28). II,20 to 24; mean number of segmented rays usually higher for males from any locality than for females from same locality (higher for 55 of 61 localities for which means for both sexes are available, statistically significantly higher for 20 of the 55 localities); posteriormost ray usually split to base, (posterior element of split ray often represented only by a basal nub of bone visible only in radiographs or cleared and stained specimens); posteriormost ray usually not bound by membrane to caudal peduncle (bound in 0 to 10% specimens examined for character from any locality). Skin covering anal-fin spines and distal half of several rays expanded and wrinkled along lateral edges in large, apparently sexually mature males (similar to condition in mature males of Entomacrodus species; Springer, 1967, especially pl. 24f).
Pectoral-fin rays 13 or 14 (14 bilaterally in 90 and unilaterally in 4 of 94 specimens examined for character).
Pelvic-fin segmented rays 3.
Caudal fin. Dorsal procurrent rays 6 to 8 (7 in 78% of 278 specimens examined for character), ventral procurrent rays 6 to 8 (7 in 71% of specimens), combination of 7 dorsal with 7 ventral procurrent rays in 58% of specimens; segmented rays 12 or 13 (12 in <1% of specimens).
Vertebrae (Tables 27 and 28). 10 or 11+26 to 30 = 36 to 40 (modally 11 precaudal vertebrae at most localities; however modally 10 in Red-Sea specimens; 10 also common or modal count at several Indian Ocean localities and Okinawa, Philippines, and Trobriand Islands, but more data needed for these localities); mean number of total vertebrae usually higher for males from any locality than for females from same locality (higher for 53 of 61 localities for which means for both sexes are available, statistically significantly higher for 14 of the 53 localities); posteriormost pleural rib on 11th to 13th from anteriormost vertebra (predominantly on 12th vertebra at all localities, except commonly on 11th in Red Sea specimens); posteriormost epineural on 14th to 22nd from anteriormost vertebra (Table 29), modal occurrence varying among localities from on 16th to on 20th vertebra.
Table 28.—Frequency distributions for certain meristic characters of female specimens of Istiblennius edentulus from various localties. Underlining indicates significant differences between means of sexes from same locality (p ≤ .05).
Cirri. Nape cirrus present on each side, relatively long (compared with that of other species of Istiblennius or Blenniella), usually simple, occasionally with short side branch or ragged edges, particularly in large specimens. Orbital cirrus slender, transversely flattened, tapering distally, usually simple, occasionally with 1 to 8 long or short branches along edges or at tip, particularly in large specimens or specimens in which cirrus appears deformed and/or regenerating after damage. Nasal cirri usually 3 to about 20 in specimens >35 mm SL, number generally increasing with increasing SL.
Lateral line (based on 181 specimens ≥30 mm SL; Table 30). Continuous canal anterodorsally with simple pores (no vertical pairs of pores; rarely, vertical pair near lateral-line origin), extending posteriorly to point between verticals from 9th dorsal-fin spine and 3rd segmented ray (to or beyond vertical from 11th spine in 85% of specimens), usually descending to or near midline of body, then continuing posteriorly along midline as series of 2 to 18 short, disconnected, horizontally bi-pored canals (tubes) in skin (more than 5 tubes in 89% of specimens); posteriormost bi-pored tube at point between verticals from bases of 13th dorsal-fin spine and 15th segmented ray (posterior to vertical from 3rd segmented ray in 93% of specimens; mean position of posteriormost tube tending to increase with increasing SL, Table 30).
Mandibular pores 5 or 6 (bilaterally 6 in 88%, and at least unilaterally in 98% of specimens examined for character).
Five to 7, predominantly 5, sensory pore positions between 1 o'clock and 5 o'clock on postorbital margin; 0 to 5 positions with pairs or multiples of pores (see Table 31 and discussion of infraorbital pores under section on geographic variation).
Posterior dentary canines absent.
Ventral margin of upper lip and dorsal margin of lower lip entire (for exception, see subsection “Western Australian specimens” in section on geographic variation).
Males with fleshy, blade-like crest on head; females with or without crest (see geographic variation section). Sexual dichromatism present or absent (see geographic variation section).
Size. Largest male, 144 mm SL; largest female 132 mm SL; specimens > 120 mm uncommon, mostly females. Ophioblennius-stage specimens unknown, but smallest specimen examined, 17.6 mm SL (UF 31309, Flores, Indonesia), perhaps recently metamorphosed: pectoral fins with distinctive pattern of black pigment distally (Figure 6b), patch of moderate-sized melanophores distally in membrane between 2 ventralmost rays, none between next 2 rays, then melanophores at tips and/or in membranes between next 9 rays, no melanophores dorsal to these rays; proximal of fin bears sprinking of very fine melanophores, mostly restricted to fin rays (these melanophores possibly not present in unmetamorphosed specimens). Next smallest available specimens, 17.8 to 19.2 mm SL (various localities), lack black pigment distally on pectoral fins.
Table 29.—Frequency distributions for location of posteriormost epineural relative to a vertebra, counting anteriorly to posteriorly, in specimens of Istiblennius edentulus from various localities (modes are underlined).
Table 30.—Frequency distributions for certain lateral-line characters in specimens of Istiblennius edentulus ≥30 mm SL.
GEOGRAPHIC VARIATION.—We address here why we treat I. edentulus as a single species rather than as a complex of species with allopatric distributions. In summary, the decision is mainly an arbitrary and pragmatic one, not necessarily consistent with the way we treat other putatively monophyletic groups in Istiblennius and Blenniella. Our decision was influenced by the fact that males from the various I. edentulus populations are, in general, indistinguishable, with differences manifested mainly by shifts in modes and means for meristic characters (Tables 27 to 29). Female specimens from some localities, however, are, or are almost, completely separable from those from some other localities based on color pattern, development of a fleshy head crest, number of infraorbital pores, or meristics, or on some combination of these characters; however, more than one type of female can occur at a particular locality, but only one type of male.
The development of the female head crest and color pattern is complex and may vary individually and apparently ontogenetically among specimens from a single locality. Method and state of preservation of museum specimens also may strongly affect the appearence of specimen color pattern, which is conceivably influenced as well by season, sexual maturity, and general coloration of the bottom on which the specimens were inhabiting when collected. In the discussion that follows, we may have biased our findings by excluding from consideration numerous specimens that we believe are faded or otherwise artifactually discolored.
In spite of the abundance of representative specimens from a great number of localities, we did not have enough specimens of the proper size or condition on which to fully characterize the populations from each of the localities represented by our samples.
Following is a summary discussion of some of the main population variations represented by our material. The discussion concentrates first on color pattern, followed by discussions of crest development, meristics, and infraorbital pores.
Table 31.—Frequency distributions for number of infraorbital pore positions occupied by more than one pore (IPPMP) between 2 and 4 o'clock on postorbital margin in specimens of Istiblennius edentulus from various localities.
Color Pattern. At all localities, except the Line Islands, where I. edentulus occurs, the color pattern of females smaller than about 50 mm is indistinguishable from that of males. The pattern on the body consists mainly of paired, dark bands. This pattern persists throughout life in males (Figures 40, 41), but may change in females. At most localities, females 55 to 60 mm SL begin to exhibit a pattern of small, dark spots posteriorly on the body (e.g., Figure 36a,b). With increase in SL, the distribution of spots expands to include the dorsal and anal fins, posterior half of the body, and, occasionally, the base of the caudal fin (e.g., Figures 36c, 37a). The first spots appear distally in the segmented-ray portion of the dorsal fin. Prior to the appearance of these spots, the dorsal fin appears to be uniformly blackish or exhibits oblique, blackish, streak-like markings.
Grant (1987:335, fig. 715) provides an excellent color photograph of a large, fresh female specimen from Queensland that exhibits a fully developed adult color pattern. Yoshino (in Masuda et al., 1984, pl. 268: fig. J) provides an excellent color photograph of a large, fresh male specimen from Japan (probably Okinawa) with fully developed adult color pattern (compare also Grant's figure of the female with Yoshino's, pl. 268: fig. K). Myers (1989:211, fig. 5c) presents a good black-and-white halftone of a male and female from Guam.
Female specimens from Papua New Guinea and neighboring islands may be heavily spotted by 55 mm SL. Females from Christmas and Caroline islands (both Line Islands) are unique in having a heavily spotted pattern manifested by 43 and 38 mm SL respectively (Figure 38d).
At McKean Island (Phoenix Islands), large, comparably sized females may exhibit either spotted female or nonspotted male-type patterns (Figures 38a,b). Other Phoenix Islands’ females and those at more remote Howland and Baker islands exhibit normal female spotting.
Large females from Penrhyn, Cook Islands (Figure 38c), most localities in Japan north of about 32°N (36c), and the Bonin Islands have male-like body coloration. An extended discussion of the Japanese specimens is presented at the end of this section.
Females 49 to 74 mm SL from the Marquesas Islands may exhibit spotting, which, on the body, is usually restricted to the ventral half of the caudal peduncle (Figure 37c). Although larger females, up to 91 mm, were available, they were from a collection in which the color pattern of all specimens, both males and females, was uniformly dark. We are uncertain whether this pattern accurately reflects the pattern of these specimens when fresh, or if it is an artifact, the result of poor preservation in isopropyl alcohol.
Females from Elat (Gulf of Aqaba-north end Red Sea), Sudan coast (middle Red Sea), and Difnein Island (southern Red Sea) either have a male-like pattern or bear a few relatively large spots on the caudal peduncle (Figure 39a,b). Randall (1983, figure on page 154) illustrates a fresh Red Sea female specimen of unspecified size that shows a complete absence of body spots.
Two small (52 to 56 mm SL) females (only ones available) from Mait, Somalia (in the Gulf of Aden, into which the Red Sea opens), exhibit well-developed spots on the caudal peduncle. Two females, 63 and 85 mm SL, from Mahé (Seychelles, Indian Ocean) have a few indistinct spots on the caudal peduncle, and another, 72 mm, is unspotted and male-like in pattern, whereas a female, 71 mm, from nearby Curieuse, has well-developed spotting (and a fleshy crest—see next section). Females from the Comoros Islands, including one of the largest specimens of I. edentulus examined, 132 mm SL, have indistinct spots on the caudal peduncle (as do the 63 and 85 mm Seychelles females) and, variably, in the dorsal fin. At similar sizes, females from Aldabra, Agalega, and St. Brandons (islands scattered throughout the western Indian Ocean) are typically very pale, but the darker specimens among these appear to have patterns closely similar to the most common Pacific female pattern (Figure 39c,d). Females, 51 to 88 mm SL, from Mauritius and Reunion, however, have a male-like in color pattern. Fleshy Head Crest. With rare exception, male specimens 30 mm SL and larger from all localities have either well-developed crests or clear indications of a ridge-like crest precursor.
At all but a few localities (i.e., Marquesas Islands, Bonin Islands, main islands of Japan, Red Sea, Enderbury, Mangareva, Marcus) where sizable numbers of females are available, females of all sizes usually lack any indication of a crest or its ridge-like precursor. When a crest or its precursor is present in females, it is usually restricted to specimens larger than about 70 mm SL. In particular, a precursor ridge or crest was not present in any Gulf of Thailand female, including many large specimens, that we examined. An inconspicuous crest precursor, at most, is present in large Indian Ocean females, with the notable exception of a 71 mm spotted female from Curieuse, Seychelles, which has a low but distinct crest (0.8 mm high).
At the exceptional localities, all or almost all females at or larger than a certain size bear a crest or precursor ridge, and at all these localities, except Japan, the size (length, height) of the female crest is always smaller than that of males of comparable SL. In Japan, the female crest may be equal to or larger than the crest of males of comparable SL.
All females longer than 37 mm SL from the Marquesas, longer than about 57 mm SL from Mangareva, longer than about 70 mm SL from Marcus, and longer than 35 mm from the main islands of Japan, have crests or definite ridge-like precursors. At the Bonin Islands (= Ogasawara), the smallest females available were about 50 mm SL (also includes data reported by Fukao, 1985), and all had small crests. Most females longer than about 60 mm SL from the Red Sea (including Gulf of Aqaba), 70 mm SL from Mangareva (Gambier Islands), and 65 mm SL from Enderbury (Phoenix Islands) have ridge-like precursors or low crests.
The presence of precursor ridges or crests in females from other localities is considerably variable, but a ridge is not usually present in females under 60 mm SL and a crest in females under 80 mm SL (but note 72 mm SL female from Palmerston Island, Figure 37b).
Dorsal- and anal-fin elements. Means and modes for dorsal-and anal-fin elements and vertebrae (Tables 27, 28) vary considerably and somewhat irregularly throughout the distribution of Istiblennius edentulus. Notable are the very low means for Gulf of Aqaba and Red Sea specimens and the very high means for western Indian Ocean specimens (particularly St. Brandons Shoals, Mauritius, Reunion), and specimens from Bonin, Marcus, McKean, Niue, Mangareva, Rapa, Penrhyn, and Pitcairn islands, Western Australia, and the more northern Japanese localities. Several of these localities were noted above for their variations in color pattern and possession of a head crest in females. Most notably absent from the localities where specimens have either low or high means are the Marquesas Islands, where means, although intermediate, are somewhat lower than those for specimens from the nearby Tuamotus.
Epineurals. Means for the serial number of the posteriormost vertebra bearing an epineural (Table 29) also vary considerably geographically and usually parallel means for the other meristic characters. Mean number of epineural ribs for specimens from northern Japan and Marcus Island, however, are low, and those for specimens from NW Madagascar are high, whereas means for the other meristics of specimens from these localities are just the opposite.
Infraorbital pores. Although not studied for all populations, there is evidence that geographic variation exists in the nature of the infraorbital sensory pore positions (Table 31). This variation, which is independent of sex, is complex, masked by ontogeny, and may not be obvious in specimens less than 30 mm or more than 80 mm SL.
Between 1 o'clock and 5 o'clock on the postorbital margin, there are normally 6 (rarely 5 or 7) regularly placed sensory pore positions adjacent to the margin. In almost all specimens over 25 mm SL, the third from dorsalmost pore (at 3 o'clock) is connected posteriorly by a horizontal subcutaneous canal to one or more pores. Depending on SL and population, each of the other pore positions is occupied by 1 or more pores, with multiple pores at a position connected by a subcutaenous canal, which may be quite short and directed horizontally, obliquely, or vertically (when 2 pores are present at the 2 o'clock position, the canal is very short and may be vertical). With the exception of Marquesas Islands specimens, it is uncommon to find multiple pores at more than 2 pore positions in specimens less than 60 mm. At more than 59 mm, it is common to find multiple pores at 3 or more positions.
The Marquesas specimens are distinctive in usually having more than 2 multipored positions (up to 4) at sizes less than 60 mm SL, and all size classes tend to have more multipored positions than those of other populations. In the largest size class, ≤90 mm, specimens from Marcus, Marquesas, Rangiroa, and Mangareva tend to have relatively more individuals with 4 or 5 positions with multiple pores than do specimens from other localities.
From the information presented on variation in meristics, color pattern, head crest, and pores, several distinctive populations can be discerned from among those studied: Marquesas Islands, main Islands of Japan, Bonin Islands (= Ogasawara), Marcus Island, Red Sea, St. Brandon's-Mauritius-Reunion, northwest Madagascar, McKean Island, Mangareva, etc. For most of these populations, a male specimen is usually inadequate for recognition of the specimen's locality of origin. If the specimen is a female, the chances for recognizing its provenance are better, but, even so, the specimen must be of a certain size and its color pattern well preserved. Even with these conditions satisfied, assignment of most individual females would probably be uncertain. Building a satisfactory key to the I. edentulus populations is, therefore, not possible.
There are many localities from which adequate numbers of specimens of I. edentulus are unavailable or, if available, were not studied (see “Material examined”). These localities might also harbor populations with reasonably distinctive characteristics. Even for those populations studied in most detail, other taxonomic characters probably exist that would extend the number of recognizable populations. At the risk of erring on the side of conservatism, we have chosen to recognize only one species-group name for the I. edentulus complex of populations.
The most readily recognizable population in the complex is that from the Marquesas, which is prominent in having an increased number of multipored infraorbital pore positions (in both sexes), particularly in small specimens, and in having a head crest in all females over 37 mm SL. The main islands of Japan are the only other areas where females exhibit evidence of crests at sizes less than 40 mm SL (as small as 35 mm).
As a final comment, females with well-developed head crests are almost entirely restricted to the Red Sea, Japan (including Bonin Islands), and Pacific-plate islands; male type color pattern in large females is restricted to the same localities and Mauritius and, adjacently, Reunion.
THE Istiblennius enosimae PROBLEM
Istiblennius enosimae has long been recognized by students of the Japanese biota as distinct from and allopatric to I. edentulus. The characters given for distinguishing these two forms are that female I. enosimae have a head crest and color pattern like those of males, whereas female I. edentulus lack a head crest and have a different color pattern from that of males. For Japan, we have seen specimens of the enosimae-form female only from the main islands, from Chiba Prefecture, Honshu (~35°30′N), south to Tanegashima (~32°30′N), and from the Bonin Islands. Until recently, the edentulus form had not been reported to occur north of Tanegashima.
Yoshino (in Masuda et al., 1984:300) reported that I. edentulus occurs in Japan as far north as Wakayama Prefecture (~34°N), but delimited the distribution of I. enosimae only generally: southern Japan. Fukao (1985:104–108) reported two types of female I. edentulus in his collections from Shirahama, (~33°N), Wakayama Prefecture: those that resemble males in color pattern (unspotted) and have a crest (enosimae-form) and those that differ from males in having a spotted color pattern and lack a crest (edentulus-form). He had 17 enosimae-form females, but only 5 edentulus-form females. The mean number of dorsal-fin elements of his edentulus females was 0.82 less (.005>p>.002) and the mean number of segmented anal-fin rays 0.51 more (.10>p>.05) (Table 28) than for the enosimae-form females.
Fukao could find no basis for recognizing more than one type of male in his Shirahama specimens. The mean number of dorsal-fin elements for the Shirahama males (both forms together?) was much lower than that of the Shirahama enosimae-form females and slightly higher than that of the Shirahama edentulus-form females. The mean number of anal-fin elements for the males was, conversely, much higher than the mean for the enosimae females, but slightly lower than the mean for the edentulus females. Males of the I. edentulus species complex almost always have higher mean numbers of fin elements (and vertebrae) than do females. The Shirahama males and Shirahama edentulus females have fin-element means closely similar to those of “pure” I. edentulus females from Okinawa.
Based on the means of “pure” enosimae-form specimens from other Japanese localities, and “pure” I. edentulus-form specimens from Okinawa (compare in Table 28), it appears that Fukao's Shirahama males are probably assignable to the edentulus form, rather than the enosimae-form, as he suggested. This conclusion is only partially supported by data from our Shirahama specimens. The mean for dorsal-fin elements for the males we examined is very similar to the means of males from more northern Japanese localities, and is significantly different (.05>p>.02) from that of Fukao's males (Table 27). The segmented anal-fin ray mean of the Shirahama specimens we examined, however, is identical to that of both Fukao's males and males from more northern Japanese localities, and, thus, is neutral in its bearing on our conclusion.
It may be, as Fukao (1985:106) proposed, that the Shirahama edentulus-form specimens originated as the pelagic larvae of southern populations and were transported north into enosimae-form territory by the Kuroshio current. The edentulus form is not known otherwise from further north than about 27°N in the chain of islands extending from Okinawa north to Chiba Prefecture. On the other hand, the enosimae form is known at least as far south as 32°30′N. It would be of interest to know if the two forms are reproductively isolated, or if not, whether “swamping” may mask expression of the edentulus female form in enosimae-form territory.
Fukao (1985) was ambivalent about whether to recognize I. enosimae, and also mentioned that the Bonin Islands’ specimens appeared to represent a third form. We appreciate his difficulties.
WESTERN AUSTRALIAN SPECIMENS OF Istiblennius edentulus
We have 11 specimens of Istiblennius edentulus from Western Australia (see material examined section): 2 males (73 and 98 mm SL) from Cape Range National Park, 2 km S Yardie Creek, 22°21′S, 113°48′E; male (44 mm) and 2 females (76 and 108 mm), Rosemary Id, Dampier Arch.; 5 females (85 to 113 mm) and 1 male (111 mm), Point Quobba, 24°29′S, 113°25′E.
The 2 males from Cape Range National Park, have well-branched (4 to 6 free tips) cirri on each eye. Of the other 9 specimens, 2 females (1 each from Rosemary Island and Point Quobba) have branched cirri (2 or 3 tips) on each eye; the other 7 specimens have simple cirri. Additionally, the larger male from Cape Range National Park has very fine, but noticeable, crenulations on the ventral margin of the upper lip at the comers of the lip. The 11 specimens are otherwise unremarkable for the species. The females are spotted and 2 have low, but distinct, crests and one has a low ridge on the head dorsally.
Although branched orbital cirri occur occasionally in Istiblennius edentulus, particularly in large specimens, we find the proportion of specimens (4 of 11) from Western Australia with branched cirri unusual. Additionally, we have not found lip crenulae in any other specimen of the species, although fine, fleshy pads may be present on the hidden, buccal surface of the upper lip.
COMPARISONS AND RELATIONSHIPS.—We have no intuitive opinion about the interrelationships of I. edentulus. If I. enosimae should prove to be valid, we would believe that it and I. edentulus would be sister groups. Our phylogenetic analysis indicates that I. edentulus forms a clade with I. rivulatus, I. meleagris, and I. dussumieri, but this clade is only weakly supported and we are not convinced of its monophyly.
Most recently, Strasburg (1955) suggested that I. zebra was the Hawaiian form of I. edentulus. He treated the 2 species as 1 in discussing the effects of temperature and latitude on meristics of the combined species. Although I. edentulus does not occur in Hawaii, we believe Strasburg combined the wrong 2 species, even though his conclusions may have been correct.
Strasburg (1955), who lacked data for Japanese specimens of I. edentulus (the I. enosimae of authors), noted that I. zebra differed from I. edentulus mainly in lacking nape cirri. He missed some other important characters: the crenulate lower lip, shorter continuous portion of the lateral line, and the often simple nasal cirrus of I. zebra. We believe that these characters indicate that I. zebra is more probably the sister species of I. bellus, which occurs sympatrically with I. edentulus, and, as does I. edentulus, has lower meristics than I. zebra (hence. would have supported Strasburg's correlations of temperature and latitude with meristics).
DISTRIBUTION (Figure 67, based almost entirely on specimens we examined; a few localities entered based on literature).—Istiblennius edentulus is distributed along most of the warm, rocky shores of the Indo-west and central Pacific oceans, including the Red Sea and Persian Gulf, as far north as about 35°N on the coast of Japan, south to about 30°S on the east coast of South Africa and 33°51′S on the east coast of Australia, east to Pitcairn Island, but excluding Hawaiian, Johnston, and Easter islands, where it is notably absent.
Common on rocky flats in intertidal areas (a type of habitat notably absent at Johnston Island). Istiblennius zebra, which is not a sister group of I. edentulus, occupies the ecological niche in the Hawaiian Islands that I. edentulus would occupy. Istiblennius zebra is also absent from Johnston Island. According to Grant (1987:435), I. edentulus is often found in the company of I. meleagris, an Australian endemic, in the intertidal area on the Queensland cost.
NOMENCLATURAL
- bibliographic citation
- Springer, Victor G. and Williams, Jeffrey T. 1994. "The Indo-West Pacific blenniid fish genus Istiblennius reappraised : a revision of Istiblennius, Blenniella, and Paralticus, new genus." Smithsonian Contributions to Zoology. 1-193. https://doi.org/10.5479/si.00810282.565
