BASAL LAOPHONTID EVOLUTION 



59 



Caudal ramus (Fig. 5F) rectangular, without bulbiform expan- 

 sions; about 1 .6 times as long as wide; with medioventral cup-shaped 

 concavity as in 9; ventral ornamentation more elaborate than in 9. 



REMARKS. Lang (1948) synonymized L. bulbifera with E. 

 longicauda, alluding to the congruence in the female P5 

 baseoendopod between Gurney's (1927) description of L. bulbifera 

 and Edwards' (1891) original description off. longicauda (i.e. with 

 4 setae; Norman (1911) figured only 3), and in the number of 

 processes on the first antennulary segment between the males of L. 

 rhodiaca (cf. Brian, 1928a) and E. longicauda (cf. Edwards, 1891) 

 and the female of L. bulbifera (cf. Norman, 1911). He also referred 

 to Willey's (1935) discovery of L. bulbifera in Bermuda as addi- 

 tional zoogeographical evidence for this course of action. It is clear 

 however that ( 1 ) the morphological grounds for this synonymy only 

 prove generic identity and not conspecificity, (2) Willey's (1935) 

 record is both unreliable and unconfirmed, and (3) Gurney's (1927) 

 records from the Suez Canal in reality refer to another species E. 

 canalis sp. nov. (see below). 



Our redescription diverges from Norman's illustrations in only 

 two aspects: (1) the presence of 4 setae on the baseoendopod of the 

 9 P5, the innermost being overlooked by Norman as already sus- 

 pected by Lang (1948), and (2) the inner seta of P4 enp-1 which is 

 minute (checked against the holotype) instead of very well devel- 

 oped as figured by Norman. E. bulbifera can be differentiated from 

 its congeners on the basis of the following combination of characters: 

 antennule 9 indistinctly 7-segmented, PI enp-1 2.5 times as long as 

 basis, PI enp-2 3 times as long as wide, P2-P4 enp-1 with inner seta 

 (that of P4 minute), outermost seta of 9 P5 baseoendopod extending 

 to distal margin of exopod, caudal rami 9 distinctly bulbous. 



E. bulbifera is widely distributed around the British Isles with 

 reliable records from Ireland (Farran, 1913, 1915; Holmes & 

 O'Connor, 1990), the west coast of Scotland (Norman, 1911) and 

 Norfolk (Hamond, 1969). Moore's (1973) record of E. longicauda 

 from St. Abb's probably also refers to this species. It has not been 

 reported anywhere else in northwest Europe, however, its syn- 

 onymy with L. rhodiaca Brian, first suspected by Nicholls (1941/?) 

 and later confirmed by Lang (1948), has considerably extended its 

 distribution, including the Mediterranean, Gulf of Suez and Western 

 Australia. Nicholls based his conviction on similarities in the 

 antennule, antennary exopod, PI and P4 and the modified caudal 

 rami although he admitted that the latter were not bulbous in L. 

 rhodiaca. Brian's (1928a) original description, based on a single 

 male specimen from Rhodes in the Aegean Sea (Brian, 1928a-6), 

 shows very few discrepancies with our material from Ireland and 

 Norfolk. The caudal rami are somewhat longer in the Mediterranean 

 specimen, the inner seta on P4 enp- 1 is more developed, the antennule 

 shows an additional segment distal to the geniculation and small 

 proportional length differences can be noted in the antennulary 

 segments and P4 endopod. Lang (1948) had already pointed out that 

 Brian had overlooked one of the outer spines on the P2 exopod. We 

 regard these differences insufficient to warrant the reinstatement of 

 L. rhodiaca and tentatively regard it as a junior subjective synonym 

 off. bulbifera. Nicholls' (1945) few illustrations of a male from 

 Port Denison in Western Australia which he attributed to L. rhodiaca 

 do not contradict Brian's description. In the absence of information 

 on the swimming legs (except P3 endopod) and the female this 

 geographically widely separated record cannot be verified abso- 

 lutely. 



Monard's (1928) brief description of L. bulbifera from the Banyuls 

 area does not contain the level of detail to either confirm or deny his 

 identification. The setae on the P5 baseoendopod were probably not 

 drawn at their full length even though the outermost one appears to 



be exceptionally short, his spine formula would infer a 123 pattern 

 on P4 exp-3 and the size of his female specimens (0.8 mm) falls 

 outside our recorded range. Monard (1937) recorded the species a 

 second time from Algers but the specimens were apparently dis- 

 tinctly smaller (0.64 mm). 



The Croatian records of L. bulbifera from Rovinj and Split in the 

 northern Adriatic (Douwe, 1929; Klie, 1941) could not be con- 

 firmed. It is conceivable that Vriser's (1984, 1986) records of E. 

 longicauda from the Gulf of Trieste and Petkovski's (1955) record 

 from Montenegro refer to the same species. 



Esola galapagoensis Mielke, 1981 grad. nov. 



Esola longicauda galapagoensis Mielke, 1 98 1 



Type LOCALITY. Cabo Douglas, Fernandina (Galapagos). 



Wells & Rao ( 1 987) expressed reluctance about the subspecific rank 

 attributed to the Galapagos population off. longicauda. Although 

 Mielke (1981) acknowledged the reported variability and 

 cosmopolitanism of the latter to some extent, he considered the 

 differences exhibited by his material sufficient to warrant the recog- 

 nition of a distinct subspecies. Mielke diagnosed E. longicauda 

 galapagoensis on the basis of the following characters: (1) PI exp- 

 2 with 4 setae/spines, (2) PI enp-2 with remarkably short claw, (3) 

 P4 enp-1 without inner seta, and (4) P5 baseoendopod 9 with 

 strongly reduced outer apical seta. Additional diagnostic features 

 not mentioned by the author include ( 1 ) inner seta of P6 6* extremely 

 reduced, (2) outer setae of P5 exopod S naked, (3) outer spine of P2- 

 P4 enp-2 remarkably short, and (4) caudal rami very elongate with 

 conspicuous medial swelling in 9. Based on this suite of characters 

 we feel it justified to upgrade Mielke's form to full species rank as E. 

 galapagoensis. The species has thus far been recorded from two 

 localities in the Galapagos archipelago (Mielke, 1981). 



Esola canalis sp. nov. 



Laophonte bulbifera Norman, 191 1 sensu Gurney (1927) 



TYPE LOCALITY. Suez Canal, Port Taufiq (Egypt). 



TYPE material. Holotype 9 dissected on 10 slides (BMNH 

 1999.993); paratype 9 in alcohol (BMNH 1999.994); from material 

 originally registered as Laophonte bulbifera (BMNH 1928.4.2.1 16) 

 collected during the Cambridge Expedition to the Suez Canal in 

 1924; det. R. Gurney. 



Etymology. The species name refers to the type locality. 



Description. 



female. Body length from anterior margin of rostrum to posterior 

 margin of caudal rami 621 urn (n=2; range: 585-658 um). Maxi- 

 mum width ( 1 37 um) measured at posterior margin of cephalothorax. 



Body as in E. bulbifera; cephalothorax with paired cup-shaped 

 pores both anterodorsally and anteroventrally on either side of 

 rostrum, and with distinct transverse spinule row dorsally about 

 halfway down the cephalothorax length. 



Genital double-somite (Fig. 6A) wide and dorsoventrally flat- 

 tened; original segmentation marked by bilateral constriction and 

 spinule row arising from transverse surface ridge dorsally and 

 laterally; anterior (= genital) half with large cup-shaped pores 

 laterally; ventral surface without spinular ornamentation. First 

 postgenital somite with backwardly produced lateral angles, bearing 

 spinular tuft (Fig. 6A); without ventral ornamentation. Penultimate 

 and anal somites distinctly narrower; ventral posterior border with 

 spinules (Fig. 6C). Anal somite (Fig. 6B) with spinulose anal 

 operculum; spinules coarser than in E. bulbifera. 



