SYSTEM ATICS AND PHYLOGENY OF ZAUSODES 



121 



with Z. arenicolus, whereas in tree B it occupies a transitionary 

 position between the type species and the other Zausodes species. 

 Tree B has a lower f-value (118) than tree A ( 1 38), however, we have 

 selected the latter as the optimal one on account of the lower number 

 of convergences. In tree A Z. septimus and Z. arenicolus are clus- 

 tered on the basis of two apomorphies which are unique to these two 

 species (characters 4 and 20). This grouping is at the expense of 

 introducing convergences for characters 14 and 22, however, both 

 these characters already show convergence in other clades (Fig. 

 33A) and are known to be evolutionary labile. The justification for 

 grouping Z. septimus with the other Zausodes species in tree B is 

 based solely on the convergent evolution of characters 14 and 22, 

 thereby causing additional homoplasies for characters 4 and 20. 



The monophyly of the Zausodes complex and its sistergroup 

 relationship to Perissocope are confirmed. The complex is divided 

 in two lineages by a strongly supported basal dichotomy and each 

 lineage is composed of two clades. 



The biarticulatus-kirstenae-cookorum lineage is supported by 

 leg 5 characters such as the loss of the endopodal lobe and the 

 reduction of the innermost seta of the baseoendopod. Additional 

 apomorphies are the modification of the distal outer spine on the 

 male P2 endopod and the sexual dimorphism on the P3 endopod. A 

 peculiar character shared by these species (but not used in the 

 analysis) is the presence of a well developed hyaline frill on the 

 distal endopod segment of PI. Under the traditional light micro- 

 scope this frill resembles a tuft or fan of spinules sited at the distal 

 outer corner of enp-2 (Figs 22C, 29E). All three species have 

 retained the primitive segmentation and setation of the antennary 

 exopod. Within this lineage Z biarticulatus occupies the most 

 primitive position since it is the only species which has retained the 

 inner seta on P3 enp-2. The other species (cookorum, kirstenae) are 

 clustered on the basis of their unique sexual dimorphism on the P2 

 endopod, P3 endopod and P4 exopod. 



The monophyly of the second lineage which includes all other 

 species is supported by the 1 -segmented antennary exopod bearing 

 only 2 setae and the reduced armature on the P4 endopod. A basal 

 dichotomy divides the lineage into two distinct clades, the 

 arenicolus-c\ade and the sextus-c\ade. The former accommodates 

 the type species and Z septimus and is characterized by the pres- 

 ence of ornate setae on the 9 antennule and by the reduction of 

 particular setae on the 9 P5 baseoendopod. Both species have re- 

 tained primitive antennule characters such as the 8-segmented 

 condition in the 9 and the complete absence of modified elements 

 in both sexes. The sextus-clade, encompassing 6 closely related 

 species, is extremely well supported but largely unresolved. This 

 is partly due to the deficient descriptions of the Brazilian species 

 (limigenus, stammeri, paranaguaensis) for which it has proven 

 impossible to score all characters (Table 3). The clade is defined 

 by the 7(or 6 in areolatus)-segmented 9 antennule, the presence of 

 modified spines in both proximal and distal regions of 

 the 9 antennule and on segment 3 of thed"antennule, the reduced 

 armature on the maxillary praecoxal endite and the round shape of 

 the P5 exopod in both sexes. A subgroup, combining 

 shulenbergeri and Jakobi's (1954) species, can be recognized 

 within this clade and is characterized by the presence of only 1 

 inner seta on P2 enp-3. 



Ito (1979) already remarked that Z biarticulatus occupied a 

 separate taxonomic position within Zausodes and highlighted par- 

 ticular similarities with the genus Perissocope. Lang (1965) on the 

 other hand favoured a subdivision of the genus Zausodes but was 

 reluctant to do so on the basis of P4 endopod segmentation. Our 

 analysis has revealed marked intrageneric differences in the sexual 

 dimorphism of all three swimming legs (P2-P4), the setation and 



armature of the antennary exopod, and the form and modification of 

 antennulary elements in both sexes. Such variability has not been 

 recorded for any of the other 8 genera in the family, suggesting that 

 the Zausodes complex combines distinct lineages which - in accord 

 with the generic concept currently applied in the Harpacticidae - 

 would deserve generic status. The genus Zausodes is therefore 

 redefined to include only Z arenicolus and Z. septimus, and three 

 new genera (Archizausodes gen. nov., Mucropedia gen. nov. and 

 Neozausodes gen. nov.) are proposed, reflecting the basic topology 

 illustrated in Fig. 33A. 



Status of Zausodes cinctus Krishnaswamy, 1954 



The taxonomic position of this species from off the Madras coast 

 (India) is enigmatic since Krishnaswamy's (1954) description is 

 erroneous in many aspects. We attempted but were unable to obtain 

 the type specimens from the Zoological Survey of India in Calcutta. 

 The strongly elongated P5 exopod is unique within Zausodes sensu 

 lato and leaves little doubt that Z cinctus is a distinct species. 

 However, the numerous deficiencies in the original figures make it 

 impossible to allocate this species to one of the four genera recog- 

 nized herein. For example, Krishnaswamy (1954) claims that the PI 

 exopod is only 2-segmented and sexually dimorphic, the male 

 having only 2 claws and 1 seta on the distal segment and no outer 

 seta (corresponding to exp-2). The endopod of this leg is reminiscent 

 of the laophontid type, bearing only one strong claw on the distal 

 segment. There is no doubt that the author has overlooked elements 

 on both rami and that his report of sexual dimorphism is based on 

 this oversight. Similarly, there is considerable confusion over the 

 armature formula of the endopods of P2-P4. According to 

 Krishnaswamy the distal endopod segment of P2-P4 has 1 terminal 

 and 3 inner setae which Lang (1965) translates as a [21 1] formula, 

 implying that an outer spine is present. The latter is invariably short 

 in Harpacticidae, however, Krishnaswamy's figures show only a 

 long plumose seta which is outwardly directed. We speculate that 

 this unusual orientation of the outer apical seta (perhaps as a result 

 of imperfect mounting) has obscured the outer spine (cf. Ito's (1979) 

 drawings of A. biarticulatus) and that the armature formula of P2- 

 P4 enp-3 is more likely [221] as in Archizausodes and Mucropedia. 

 If this assumption is correct then Z cinctus displays the most 

 primitive swimming leg armature within Zausodes sensu lato since 

 no other species possesses an inner seta on P4 enp-2 (and A. 

 biarticulatus being the only other species to exhibit an inner seta on 

 P3 enp-2). In this context we point out the possible homology 

 between the latter seta and the proximal inner seta of P4 enp-2 in A. 

 biarticulatus which we - by reference to the 1.0.221 pattern in 

 closely related M. kirstenae (Table 1 ) - have interpreted as originat- 

 ing from enp-3. 



Another remarkable feature is the presence of only 4 elements on 

 the9P5 exopod. Krishnaswamy's illustration shows a distinct gap 

 between the proximal and distal outer spine which corresponds with 

 the position of the vestigial seta on the P5 of A. biarticulatus (cf. Ito, 

 1979: Fig. 5-1). It is conceivable that a similarly reduced seta is 

 present in Z. cinctus. Both species, coincidently the only Asian 

 representatives of the Zausodes complex, also share the absence of 

 the endopodal lobe and show a similar arrangement of the endopodal 

 setae (with Z cinctus having an additional long seta). 



The male P2 endopod of Z. cinctus was neither described nor 

 illustrated by Krishnaswamy (1954). Sexual dimorphism in the P2 

 endopod is always present in the Zausodes complex, so it is conceiv- 

 able that it was overlooked. Pending the re-examination of 

 Krishnaswamy's types or topotype material we rank Z cinctus as 

 species incertae sedis in the Harpacticidae. 



