Caribbean species of Eiconaxius (Decapoda: Axiidea: Axiidae)

The type status of specimens of three species of the axiid genus Eiconaxius Bate, 1888 from the Caribbean Sea is clarified. Eiconaxius agassizi Bouvier, 1905, E. borradailei Bouvier, 1905 and E. caribbaeus (Faxon, 1896) are diagnosed and illustrated. Axius ( Eiconaxius ) communis Bouvier, 1905, Axius ( Eiconaxius ) rotundifrons Bouvier, 1905, and Axius ( Eiconaxius ) caribbaeus carinatus Bouvier, 1925, hitherto treated as valid species, are synonymised with E. caribbaeus . Lectotypes are selected for Eiconaxius agassizi Bouvier, 1905 and Eiconaxius borradailei Bouvier, 1905.


Introduction
The axiid genus Eiconaxius Bate, 1888 comprises more than 30 species confined to deep water that are, as far as is known, associates of sponges (Komai & Tsuchida 2012).
The US Coast Survey Steamer Blake surveyed the Gulf of Mexico and the Caribbean Sea between 1877 and 1880. Faxon (1896) described the first species of Eiconaxius from this collection but the bulk of the reptant decapod collection was studied by the French zoologist Eugène Louis Bouvier (1856Bouvier ( -1944 at the Muséum nationale d'Histoire naturelle, Paris (MNHN). Bouvier (1905) described four species of Eiconaxius and one subspecies by means of a dichotomous key but listed no material. In 1917 he completed his study and listed stations from which his specimens came. This work was published eight years later, like his first, in French (Bouvier 1925). Bouvier apparently returned most of the Blake collection to the Harvard Museum of Comparative Zoology, Cambridge (MCZ) but representatives were retained at the MNHN.
Among the many species of decapods found during the KARUBENTHOS 2015 expedition to Guadeloupe Poupin & Corbari (2016) tentatively identified only E. antillensis Bouvier, 1905. This record has been confirmed by Poore & Dworschak (2018). Other specimens of this genus proved more difficult to identify using Bouvier's works. His descriptions and illustrations are sometimes difficult to interpret, features described for one species are not described for another and many of the supposed differences appear very subtle. In fact, Bouvier (1925) himself was rather tentative in differentiating his taxa. He synonymised one of the species he had described in 1905 with E. caribbaeus, reconsidered another as a variety and added a second variety to E. caribbaeus. Further, he noted that one specimen listed in the material of E. caribbaeus "… est le passage à la variété rotundifrons." Nevertheless, Bouvier's taxa have been recently listed as full species (Kensley 1996;Komai & Tsuchida 2012;Sakai 2011).
In this contribution, only three species are recognised in addition to E. antillensis. The three are partially reillustrated and the type status of all known material is established. The distribution of all four species is mapped.

Material and methods
Size is expressed as carapace length, including rostrum, in mm. As part of this study and continuing discovery of new species in the Indo-West Pacific diagnoses have been prepared for all species of Eiconaxius and coded into a DELTA database (Dallwitz 2010). This database was used to generate the diagnoses presented here; character states in italics diagnose each species in at least two respects from every other species.
In the absence of specific designation of type specimens by Bouvier (1905) ICZN Article 72.4.1.1 is invoked to treat Bouvier's (1925) publication as "evidence … to determine what specimens constitute the type series".
The MCZ catalogue (https://mczbase.mcz.harvard.edu/) was used as a source for material held but these specimens were not examined. The list of Blake stations (Smith 1888) was consulted for problematic localities. The MNHN material was examined in detail and listed with both its original registration numbers (Th prefixes) and current catalogue numbers (IU-) prefixes. The on-line catalogue of the National Museum of Natural History, Smithsonian Institution, (https://collections.nmnh.si.edu/search/iz/) was searched and records incorporated into distributions and depth ranges. Representative specimens of one species are retained in Museums Victoria, Melbourne (NMV). Photographs of the types are available at https://science.mnhn.fr/institution/mnhn/collection/ iu/item/search. Bouvier (1925) provided general localities and Blake station numbers for all his material; depths were given in "brasses", translated from fathoms in Smith's (1888) lists, and converted to metres here.
The map was prepared from all available records but reflects, at best, collecting effort.

Taxonomic account
Eiconaxius agassizi Bouvier, 1905 (Figs 1, 4) Eiconaxius agassizi Bouvier, 1905: 803. Iconaxius agassizi.-Balss 1925. Diagnosis. Rostrum twice as long as wide. Major cheliped merus lower margin with prominent tooth at midpoint, fixed finger about half as long as upper margin of palm, cutting edge with proximal blade ending in prominent sharp tooth; dactylus cutting edge with basal molar, notch at midpoint, and straight beyond. Minor cheliped palm upper margin significantly shorter than greatest width; distolateral margin with prominent triangular toothed lobe at base of dactylus.
Remarks. Most of Bouvier's (1925) syntypic material can be accounted for in the MCZ; two lots were retained in the MNHN. Bouvier referred to two "exemplaires types" from Blake stn 2 on page 460 and to the "Type mâle" in the legend for plate 9. His account does not meet the conditions of ICZN Article 74.5 for lectotype designation. The illustrated male (MCZ CRU-11966) is here designated as the lectotype and is assumed to be the one remaining from this station in the MCZ.
Eiconaxius agassizi is notable for the narrow acute rostrum, median carina simply bifurcating on the gastric region, a strong notch on the dactylus of the major cheliped opposing a prominent tooth on the fixed finger, and a bifid distolateral tooth on the propodus of the minor cheliped. The species is distributed along the margins of the Caribbean Sea and into the Straits of Florida. Diagnosis. Rostrum tapering more over distal third than proximal, with rounded apex in adult (acute in juvenile), 1.2-1.3 times as long as wide. Sublateral gastric carinae present, diverging widely from base of median carina. Major cheliped merus lower margin with single denticle. Major cheliped palm upper margin distally bicarinate, ending in 2 blunt teeth; fixed finger about half as long as upper margin of palm, cutting edge with 2 blunt teeth (overlapping); dactylus cutting edge with basal molar, notch proximal to midpoint, and concave beyond. Minor cheliped palm distolateral margin with sharp spine at base of dactylus (small); fixed finger cutting edge with sharp tooth near base. Pereopods 3 and 4 dactyli ovate, with row of marginal plus up to 3 facial rows of spiniform setae. Male pleopod 1 present, simple.
Distribution. Cuba, Guadeloupe, Barbados; Caribbean Sea; 194-627 m depth. Remarks. Bouvier referred to a "femelle Type" in the legend for plate 9, almost certainly the female remaining in the MCZ from what is probably Blake stn 57 (the first listed but not numbered in Bouvier's list of material. Again, his account does not meet the conditions of ICZN Article 74.5 for lectotype designation. The illustrated female (MCZ CRU-11959) is here designated the lectotype. The incomplete female from Blake stn 259 (227 m, Milligan's Key) is lost.
Diagnosis. Rostrum tapering more over distal third than proximal, with rounded apex in adult (acute in juvenile), 1.5-2.0 times as long as wide; lateral margins smooth or denticulate. Major cheliped fingers about half as long as upper margin of palm; fixed finger cutting edge with broad blade over proximal half, irregular beyond; dactylus cutting edge with basal molar, notch and straight beyond. Minor cheliped palm upper margin as long as greatest width; distolateral margin with prominent triangular toothed lobe at base of dactylus; fingers almost as long to longer than upper margin of palm. Uropod endopod anterolateral apex acute, with 1 or few small teeth.