A(o\ nh Annals of the Eastern Cape Museums Volume 3, 2002 [February 2004] Contents AD Harrison A Contribution to the Taxonomy of Tanytarsini (Diptera: Chironomidae) of Sub-Saharan Africa, with a Description of a new Genus (Afrozavrelia) and five New Species from other Genera 1 AD Harrison Metriocnemus capicola, a Replacement Name for Metriocnemiis capensis Harrison 2002 1 5 HM Barber-James A Preliminary Investigation into the Influence of Turbulence on Larval Feeding in two species of Blackfly, SimuUum chutteri Lewis and SimuUum nigritarse Coquillett (Diptera, Simuliidae), from the Great Fish River, South Africa 16 Published by the Directorate of Museums and Heritage Resources of the East- ern Cape Province at the Albany Museum, Grahamstown, South Africa ISSN 1562-5273 ANNALS OF THE EASTERN CAPE MUSEUMS These Annals are the successors to the Annals of the Cape Provincial Museums, published until September 1997, Volume 19, Part 9 for the Natural History series and Volume 1 , Part 6 for the Human Sciences series. The Annals of the Eastern Cape Museums will be published in a single series with papers in both the natural sciences and humanities. Two volumes of papers may be published each year. The primary objective of these Annals is to disseminate the results of research work carried out by staff of Eastern Cape museums or by researchers whose re- search is based on material wholly or partially housed in any of these museums. Copies of these Annals as well as parts of the Annals of the Cape Provincial Muse- ums are obtainable from the Librarian, Albany Museum, Grahamstown. Editors for this volume I.J. de Moor & F.C. de Moor. Subject editors Specialist subject editors are selected from staff of the Eastern Cape's Museums. If these are not available, spe- cialists from other organizations will be approached to review and edit submissions. All correspondence to: The Editor Albany Museum Somerset Street GRAHAMSTOWN 6139 Back numbers from: The Librarian Albany Museum Somerset Street GRAHAMSTOWN 6139 Published by the Albany Museum on behalf of the Directorate of Museums and Heritage Resources of the Eastern Cape Province. Copyright © Trustees of Albany Museum These Annals are to be cited as Ann. E. Cape Mus. ISSN 1562-5273 Design and layout: Irene de Moor, Albany Museum. A CONTRIBUTION TO THE TAXONOMY OF TANYTARSINI (DIPTERA: CHIRONOMIDAE) OF SUB-SAHARAN AFRICA, WITH A DESCRIPTION OF A NEW GENUS (AFROZAVRELIA) AND FIVE NEW SPECIES FROM OTHER GENERA A D Harrison I'rcshwcUcr Research Unil. Deparliiieiu of Zoology, University tfCape Town. Sindh Ajririi E-inail: harrisa@iafriea.eoin ABSTRACT A brief overview is given of existing knowledge of the systematics of Tanytarsini in sub-Saharan Africa. A new genus, Afrozcivrelia, is established for Zavrelia kribiensis Kieffer and five new species (Rheotanytarsiis shebelensis, Tanytarsiis awashensis, Tcmytarsiis fhimiiieiis, Taiiytarsiis zimbabwensis and Virgatanytarsns aboensis) are also de- scribed. An additional note, on the diagnostic features of the larvae and pupae of two other species (Rheotanytarsiis fiisciis Freeman and Rheotanytarsiis iiioiitaniis Lehmann) and the pupa of Rheotanytarsiis guineensis Kieffer, is also provided. Key words; Chironomidae, Tanytarsini, Ethiopia, Zimbabwe, South Africa, new genus, new species, Diptera, systematics. INTRODUCTION There are a number of reliable publications on the systematics of the tribe Tanytarsini (Diptera, Chironomidae) in Africa south of the Sahara: Free- man ( 1958) described new species and revised the work of earlier taxonomists, Kieffer and Goetghe- buer. Kyerematen & S tether (2000) reviewed the Afrotropical Rheotanytarsiis, and Ekrem (2001) reviewed the Afrotropical Tanytarsiis. All these au- thors included useful keys in their publications. This does not exclude the likelihood that there are not more Tanytarsini to be described from the region, especially those of the mountain chain that extends from South Africa to Ethiopia in the north (see Ekrem 2001). This paper describes five new species from this mountain chain or its foothills and also sets up a new genus, Afrozavrelia, for Zavrelia kribiensis Kieffer, a species that does not fit into the genus Zavrelia Kieffer in any life stage. The descriptions of taxa in this paper are based on studies of museum collections, from specimens collected from Ethiopia, Zimbabwe, and South Af- rica (see ‘Methods’ for further details). THE TANYTARSINI IN SUB-SAHARAN AFRICA The 64 sub-Saharan Tanytarsini previously de- scribed by various authors can divided into two categories: 62 species in recognized or newly- described genera, and two species that cannot be placed in known genera. These are discussed be- low, together with references to source publications and short notes on the diagnostic features of some spe- cies. Species in recogniz.ed or newly-described genera Tanytarsiis: 26 species (Ekrem 2001). Rheotanytarsiis: 19 species (Kyerematen & S tether 2000). Cladotanytarsiis: 10 species. One species (C. c re- bus) described by Lehmann ( 1981); six species de- scribed by Ereeman & Cranston (1980); three spe- cies-C. irsaciis, "Tanytarsiis’ biikcmts and "Tanytarsiis’ congolensis - described by Lehmann (1979). {Note: Ekrem (1999) transferred T. bukaviis and T. congo- lensis to Cladotanytarsiis. ) Nidmirbia: one species, N. capicola, described by Siiwedal (1982). This was originally placed in Micropsectra by Ereeman (1958). Stenipellina. Two species; S. chanibiensis Goetghe- buer in Ereeman (1958); S. reissi described by Leh- mann (1981). Stenipellinella: one species. Although Freeman (1958) described "S. triincata ’ as a species of Stenipellina, this species fits more closely into the genus Steni- pellinella Brundin as it has a number of features more characteristic of the latter genus: in the adult male the anal point has longitudinal crests extend- ing onto the anal tergite with basal spinules be- tween; the pupa has a shagreen pattern similar to that of Stenipellinella brevis (Edwards) (Finder & Reiss 1986) and segment VIII has a single, robust sclerotized postlateral spine; the larval antenna is, however, of the Constenipellina-type with both Lauterborn organs being distal on segment 3 and a pedestal with a prominent apical spur. (This de- scription based on pupa, a pharate male and larval head capsule from the same sand grain case, Cat. ABLER 9J, in AM). Friederia villosa S tether & Andersen (Stether & Andersen 1998). Virgatanytarsns - three species; V. arduennensis (Goetghebeur) (= subreflexens Ereeman), V. nigri- cornis (Goetghebeur) (see Freeman 1958) and V. aboensis, described in this paper. Zavrelia kribiensis Kieffer {=Afrozavrelia sp.): one Annals of the Eastern Cape Musewns Vol. 3 (February 2004): 1-15 . species. In this paper this species is placed in a new genus, Afrozavrelia. Species that cannot be placed in known genera ‘‘Tanytarsus’ abnormis' (Lehrmann, 1981) Characteristic features: hairy eyes with no dorsal extension, superior volsella with digitus but no median volsella. ‘Tanytarsus’ saetosus (Lehrmann, 198!) Characteristic features: hairy eyes with no dorsal extension; costa ending well proximal to M3+4; anal point bare with a small knob-shaped ‘point’, but superior volsella and median volsella of Tany- tarsiis-type. METHODS Ethiopian specimens were collected by the au- thor while working in the cooperative progiamme described in the Acknowledgements. The Zimbab- wean material was collected by the author working in the Zoology Department, University of Zim- babwe (then Rhodesia) financed by the Rockefeller Foundation of New York. Specimens from the Western Cape Province were collected by the author or by members of the Freshwater Research Unit, Zo- ology Department, University of Cape Town; the rest of the South African material was collected by Dr F. C. de Moor and his team at the Albany Museum, Gra- hamstown. Eastern Cape. Pinned specimens were treated as follows: the wings were removed from the dried specimen and mounted directly in Canada balsam, then the rest of the specimen was macerated in 5% potassium hydroxide at room temperature for 24 hours; the KOH was removed by placing it in 70% alcohol for about 10 minutes, and then into 96% alcohol. It was then dissected and mounted in Canada bal- sam-dissolved in cellosolve-on the same slide as the wings. Specimens preserved in alcohol were dissected and mounted in the same type of balsam. Drawings were made by means of a drawing tube on a compound microscope. Measurements were made with an eyepiece micrometer in the com- pound microscope. Morphological terminology is according to S aether (1980) and the description of the males follows the style of Ciunston, Dillon, Pinder & Reiss (1989), using their generic definitions. The descriptions of females follows the style of S aether (1977). The holotypes and paratypes of all the species described here and other material used in the de- scriptions have been deposited in the Zoologische Staatssammlung, Munich, Germany (ZSM) or in the Albany Museum, Grahamstown, 6140, Eastern Cape Province, South Africa (AM). The catalogue numbers of the specimens are given in the text. Abbreviations AR antennal ratio. Ratio of length of apical flag- ellomereto combined length of basal flagel- lomeres. LR leg ratio. Ratio of length of tarsomere 1 to len- gth of tibia. SV ‘Schenkel-Schiene Verhaltnis’: Ratio of femur plus tibia to tarsomere 1 . BV ‘Beinverhaltnisse’. Combined length of femur, tibia and tarsomere 1 divided by length of tar- someres 2 to 5. ADH A D Harrison (Collector). TAXONOMIC DESCRIPTIONS Afrozavrelia gen. nov. Generic diagnosis ADULT MALE Size: small, wing length about 1 mm. Head: antenna with 10 flagellomeres; no frontal tu- bercles; eyes hairy with long dorso-medial exten- sions; palps normal, segment 3 with no subapical sensilla. Thorax: with antepronotal lobes widely separated overreaching pronotum, no tubercle; acrostichals 22, biserial, dorsocentrals 18, partly bi serial, prealars 2, supra-alar 1 , scutellars uniserial. Wings densely clothed with setae, costa not produced, R4+5 ending before tip of M3+4, Anal lobe not developed, squama bare. Legs: fore tibia with short, straight spur, other tibia with small combs well separated each with a thin spur. Pulvilli absent. Hypopygium: anal tergite bands weak not meeting centrally, about 12 anal tergite setae, anal point long, bare with rounded tip, superior volsella long with broad digitus, median volsella with few lamel- liform setae and no simple setae, inferior volsella long extending up to, or beyond tip of gonostylus which is narrow with two long setae at tip. ADULT FEMALE (based on one pharate specimen) Size: similar to male. Head: general structure similar to male; antenna with five flagellomeres. Thorax: similar to male. Wings and legs appear to be similar to male. Genitalia: sternite VIII not forming a distinct floor under the anterior part of the vagina; gonocoxapo- demes narrow and joined, gonopophysis of VIII simple, gonocoxite IX small with two setae, coxos- ternapodemes very small and curved; segment X with- out setae, postgenital plate pointed, cerci small; semi- nal capsules oval with short necks, ducts convoluted joining to a central bulb with common opening. Harrison: Tanytarsini of Sub-Saharan Africa PUPA Size: small, in straight cases composed mainly of diatom frustiiles and detritus. Cephalothorcvc cephalic tubercles low and rounded, frontal setae flat, almost taeniate, no antepronotals or dorsocentrals, thoracic horn long, pointed and down-turned, no spinules, minutely rugulose, three precorneal setae small, short and flattened. Abdomen: tergites I and II bare, II with a short hook row, m to VI with paired, elongate patches of points, simple S setae on segments II to IV, taeniate S setae on V to VIII; segment VIII with a few small spines at posterolateral comer; anal lobes with moderately de- veloped fringe of taeniate setae; no pedes spurii. LARVA Size: small, in straight, portable cases consisting mainly of diatom frustules and detritus. Head: with five-segmented antennae, of about 86% of head capsule length, on prominent pedestal, AR about 0.6, long basal segment with ring organ near base and seta subterminally, blade shorter that segment 2, accessory blade present, Lauterborn or- gans large, terminal on segment 2, sessile, style present. Labrum with SI palmate, bases fused, SII plumose on tall pedestal. Sill and IV seta-like, labral lamella well-developed, pecten epipharyngis consisting of three separate, slender distally pointed scales; premandible with two teeth and well- formed brush; mandible with dorsal tooth with three inner teeth, seta sub-dentalis long and curved, seta interna small and plumose, pecten mandibularis consisting of about 10 long lamellae. Mentum with median tooth rounded, six lateral teeth, the second being shorter then the first and third; ventromental plates fan-shaped, widely sepa- rated. Body: anterior and posterior parapods with simple hooks, procercus with long anal and lateral setae. NOTES Afrozavrelia can be distinguished from Neo- zavrelia and Zavrelia in all life stages (Table 1). In some particulars - such as the anal point struc- ture, the presence of a digitus, and the terminal Lauterborn organs - the former genus resembles Neozavrelia. In other particulars - such as the hairy eyes, the antennal tergite bands not meeting, and the more complete anal fringe of the pupa - it resembles Zavrelia. Nevertheless, in many other particulars - such as the strong dorsal extension of the eye, the very naiTOw gonostylus, the lack of a floor to the anterior vagina, the taeniate frontal setae of the pupa, the long antennae of the larva (especially Table 1: Morphology of Afrozavrelia contrasted with that of Neozavrelia and Zavrelia Afrozavrelia Neozavrelia Goetghebuer Zavrelia Kieffer Eyes hairy with long dorso-medial exten- sion Eyes bare or with short pubescence, no dorsomedial extension Eyes hairy, no dorsomedial extension Front tubercles absent Front tubercles absent Small frontal tubercles w -J Anal tergite bands weak, not meeting Anal tergite bands fused, forming V-shape Anal tergite bands not meeting < Anal point bare dorsally Anal point bare dorsally Anal point with crests and spinules Digitus present Digitus present No digitus Median volsella with very few larnelliform setae Median volsella with numerous setae, some slender Median volsella with numerous se- tae, some simple Gonostylus very narrow Gonostylus broad Gonostylus broad FEMALE No floor under anterior vagina Floor under anterior vagina (Cranston 1998) Floor under anterior vagina Frontal setae taeniate Frontal setae normal Frontal setae normal < (X Thoracic horn without spines or apical teeth Thoracic horn without spines, with or without apical teeth Thoracic horn with spines D CL Posteriolateral comb weak, of a few small light teeth Posterolateral comb strong, of numerous dark teeth Posterolateral comb with single or double dark teeth Anal lobe with fringe of about 22 taeniae Anal lobe fringe of 5-14 taeniae, or re- duced Anal lobe with fringe of 17-20 taeniae < > Antenna long, 86% length of head capsule, basal segment longer than flagellum Antenna short, basal segment about as long as flagellum Antenna short, basal segment about as long as the flagellum < -J Lauterborn organs terminal atid sessile Lauterborn organs terminal and on pedi- cels One Lauterborn organ sub-terminal, both on pedicels Annals of ilie Eastern Cape Museums Vol. 3 (February 2004): 1 -15 . 3 the basal segment), and the short antennal blade It resembles neither of the above two genera. Saether & Andersen ( 1 998) describe another African member of the subtribe Zavreliina (that was erected by Saether (1977)): Friederia villosa Saether & Anderson, from Ghana. The genus Friederia differs very markedly from Afrozavrelia, however, as it has bare eyes, an anal point with setal tufts, superior volsella with no digitus, and median volsella reduced to a small tubercle with one, simple seta. The female and immatures are unknown. Type species Zavrelia kribiensis Kieffer. Afrozavrelia kribiensis (Kieffer) (Figs 1-13) Zavrelia kribiensis Kieffer, 1923 Also described in Freeman (1958) and Freeman & Cranston (1980). The description of the male, given above (in the generic diagnosis) and below, is more detailed than that of Freeman (1958) and the female, pupa and larva are also described. The female and imma- tures were associated from pupal cases with pharate males and females. ADULT MALE (N == 4 mounted) Body: length 1.1 mm. Wing: length 0.9 mm; body colour very light brown, halteres dark tipped. Head (Fig. 1); No frontal tubercles; antenna AR 0.6, 10 flagellomeres, eyes hairy with long dor- somedial extensions. Head setation: six inner verti- cals, two outer verticals, 30 clypeals. Length of palp segments, 15, 21, 24, 60, 111 pm; no subapical sensil- lae on segment 3. Thorax. Setation: lateral antepronotals nil, acros- tichals 22 biserial, dorsocentrals 18 biserial anteriorly, posterior prealars 2, supra-alar 1 , scutellars 4 per side. Wings (Fig. 2): broad, anal lobe not developed; densely clothed in setae; squama bare; R4+5 ending before tip of M3+4, costa not extended. Legs: fore tibia with short, straight spur, other tibia with small combs, well separated each with a thin spur; LR fore 1 .5, mid 0.6, hind 0.7; SV fore 1 .7, BV fore 2.5. Hypopyginm (Figs 3 & 4): anal tergite bands weak and not meeting, about 12 anal tergite setae, anal point long, bare with rounded tip, superior volsella long with broad digitus, median volsella (Fig. 4) with one, or possibly two, lamelliform setae, no simple setae, inferior volsella longer than the com- bined length of the gonocoxite and gonostylus; gonostylus narrow with two long setae at the tip. ADULT FEMALE (based on a pharate specimen, mounted). Body: length 1 .6 mm Head: general structures as male, antenna with five flagellomeres. Thorax. Setation: acrostichals 22 biserial, dorsocen- trals 14, posterior prealars 2, supra-alar 1, scutellars 5 per side. Wings and legs appear to be similar to those of the male. Genitalia (Eigs 5 & 6): stern ite VIII not forming a distinct floor under the anterior part of the vagina, gonocoxapodemes narrow, light, joined; gonopophy- sis simple but with ventrolateral enlargement; gono- coxite IX small with two setae; coxosternapodemes very small, light, curved; segment X without setae; postgenital plate pointed; cerci small, 27 pm; semi- nal capsules (Fig. 6) oval with short necks, 42 pm long, with neck; ducts convoluted joining to a cen- tral bulb and with a common opening. PUPA (N = 5 mounted) Remain in larval cases, made mainly of diatom frustules and detritus. Exuviae: almost colourless, transparent. Cephalothorax: cephalic tubercle low and rounded, frontal setae flat, almost taeniate; dorsum minutely pebbled; no antepronotals or dorsocentral setae; thoracic horn (Fig. 7) long, pointed and down- turned, no spinules, minutely njgulose; precorneal setae small, short and flattened; wing sheath with nose. Abdomen (Fig. 8): tergites I and II bare, II with short hook row of small rounded “hooks”; tergites III-VI with paired elongate patches of points; S setae (simple)- I none, II one. III and IV three; S setae (taeniate)-V, VI, VII four, VIII five; segment VIII with a few small spines at posterolateral corner (Fig. 8); no pedes spurii. Anal lobes moderately well developed with a complete fringe of about 22 taeniate setae per side. LARVA (N = 1 complete specimen, mounted; also based on five larval exuviae in pupal cases.) The larva lives in a straight, portable case made mainly of diatom fnjstules and detritus. Colour: yellowish. Head: capsule 168 pm long, dorsal surface smooth. Antenna (Fig. 9) 144 pm long, with five segments on pedestal 84 pm high with a prominent apical spur. AR 0.6. Long basal segment with ring organ near base and seta subtenninally, blade shorter than segment 2, accessory blade half the length of blade, both Lauterbom organs terminal, sessile almost as long as segment 3, style present, segment five fine and pointed. Labrum (Fig. 10): SI palmate, bases fused, SII 4 Harrison: Tanylarsini of Sub-Saharan Africa plumose situated on tall pedestal, Sill and SIV seta-like; labral lamella well-developed; pecten epi- pharyngis consisting of three separate, slender, dis- tally-pointed scales. Premandible (Fig. 1 1) with two teeth and we 11 -formed brush. Mandible (Fig. 12): dorsal tooth light yellow, api- cal tooth and two inner teeth light brown; seta sub- dentalis long and curved, not extending beyond the apex; seta interna small and plumose; pecten man- dibu laris consisting of about 10 long lamellae. Mentum (Fig. 13); all teeth light brown, median tooth rounded, six lateral teeth, second tooth shorter than the first tooth, then regularly decreas- ing in size; ventromental plates fan-shaped with anterior margin smooth and widely separated. Body, anterior and posterior parapods with simple hooks; procercus broader than long with long anal and lateral setae. SPECIMENS EXAMINED All from Western Cape Province, South Africa, 6', Molenaars River, 33'’43'S 19“l0'E, 27 iv 96 Figs 1-13. Afrozavrelia kribiemis. 1-4, adult male; !, head; 2, wing; 3, hypopygium; 4, median volsclla. 5-6, adult Icmale: .3, geni- talia; 6, seminal capsules and ducts. 7-8, pupa: 7, thoracic horn and frontal setae; 8, abdominal tergites II-IV. 9-13, larva: 9, antenna: 10. labium; 1 1, premandible; 12, mandible; 13, mentum. Annals of the Eastern Cape Mnsenins Vol. 3 (February 2004): 1-15 5 (cat. SAC.45R) (collector ADH) also 3 SS, Eer- ste River, 33°56’S 19°10'E (Cat. E1Q2 24R, S2Q2 7M, S2Q2 lOR) (collector Denise Schael, 1998); with 1 $ (Cat. RTDU.219), 2 pupae (Cat. RTDU. 282 with pharate (S and 148H with pharate $, cases contained larval remains, used for associa- tion), 3 larvae (Cat. RTDU, 294, 282 and 148H) (collector Rebecca Tharme, 1993-95). All material in AM. NOTES Habitat Preferences: The South African speci- mens were found in montane rivers with soft, un- buffered water. The larvae live in straight portable cases of diatoms and detritus and they appear to be detritivores or scrapers. Distribution: Cameroon, Kribi; South Africa, West- ern Cape Province. Rheotany tarsus shebelensis sp. nov. (Figs 14-17) ADULT MALE (N = 3 mounted) Close to the generic definition in Cranston et al. (1989) except for the structure of the medium volsella that has no plate. Body length: 1.92 mm. Wing length: 1.3 mm. Colour: Head and antennae brown; thorax and legs brown, vittae, preepisternum and postnotum dark brown; abdomen brown with no obvious markings. Head: AR 0.43-0.47; palp segments: 24-30, 27-30, 60-63, 78-81, 144-150 /tin; no subapical sensillae on palp segment 3. Head setation: temporals 7, clypeals 15. Thorax: no scutal tubercle; setation: lateral ante- pronotals nil, dorsocentrals 9, posterior prealars 1, scutellars 4 per side. Wings: most of the surface covered with setae. R2+3 absent. Setation of veins: R 22, R| 26-29, R4+5 50, seta also on other veins. Legs: foretibia with short spur on scale; all combs with spurs. Leg measurements and ratios are given in Table 2. Hypopygium (Figs 14, 15, 16, 17): anal tergite bands transverse or slightly V-shaped, not meeting; anal point without spine patches and downturned; supe- rior volsella club-shaped without microtrichia (Fig- 14); median volsella (Fig. 16) with distal filamen- tous setae long extending beyond apex of inferior volsella, almost reaching the tip of the gonostylus, without median bend; distal setae narrow with no plate-like structure; inferior volsella normal, gonosty- lus tapering gradually. Figure 14 illustrates the Eth- iopian specimen (the holotype) and Figs 15 & 16 a South African specimen with a narrower anal point and a somewhat smaller superior volsella. Some South African specimens have superior volsellae of intermediate size. Fig 17 illustrates the apodemes of the type. SPECIMENS EXAMINED 1(5' found drowned, Wabe Shebele (river), Ethiopian Highlands, 07°01’N 39"03'E (ETC.34F) 1984.1.4, collector ADH (ZSM); 2 SS from light trap. Little Mooi River, KwaZulu-Natal, 29*^13'S 29^53E, (MOI 56CD & CH) and 2 SS from light trap, Kleinmooi River. KwaZulu-Natal. 29'^13'S 29^53E, 4.iv. 1995 (Cat. MOI. 65BT & CB), col- lectors F. C. de Moor and team (all in AM). Holotype ETC. 34F; paratype S MOI 65BT. NOTES This species falls into the Rheotanytarsiis pel- liicidiis group of Kyerematen et al. (2000) but it can be taken no further than couplet 3 of their key as it has filamentous not foliate setae on the median volsella. If this is neglected it would key out at cou- plet 5. It can be distinguished from the two species keyed there as follows: from Rheotanytarsiis pellu- cidiis (Walker) by not having the anal tergite bands clearly V-shaped, with the median volsella without Table 2. Rheotanytarsiis shebelensis leg measurements in pm and ratios fe tl ta. ta2 ta3 tar ta. LR BV SV Pi 528 288 588 324 240 168 98 2.0 1.7 1.4 P2 540 420 204 132 84 72 48 0.5 2.9 4.7 P3 600 504 300 204 168 120 72 0.6 2.5 3.7 Legend: fe = femur; ti= tibia; ta,. ta^, ta,, taj tas= tarsomeres 1-5; LR = leg ratio (length of tarsomere I: length of tibia); BV = 'Beinverhaltnisse’ = combined length of femur, tibia and tarsomere 1 divided by length of tarsomeres 2 to 5; SV=‘Schenkel-Schiene Verhalt- nis’ = ratio of femur plus tibia to tarsomere 1 . 6 Harrison: Tanytarsini of Sub-Saharan Africa Figs 14-17. Rheolanylarsiis shcbelcnsis, adult male: 14, hypopygium (Ethiopian); 15, liypopygium (South African); 16, median vol- sclla; 17, apodemes. a median bend and lacking microtrichia on the su- perior volsella. It can be distinguished from Rheo- tanytarsus huculicaudus Kyerematen & S tether by not having the anal tergite bands fused and with the gonostylus not abruptly but gradually narrowed (rather like that of R. pellucidus). In the key in Kyerematen and S aether (2000) it keys to R. hucii- licaiuliis but differs from it as above. Etymology: shehelensis referring to the Wabe She- bele (river) Ethiopian Highlands. Habitat preferences', the larva has not yet been identified but it is assumed that, like most other members of the genus, it lives in the current in cases constructed for filter feeding. The adults were all collected drowned in or near rivers. Distribution: Ethiopian Highlands and KwaZulu- Natal, South Africa. Rheotanytarsiis fiiscus Freeman (Eigs 1 8-20). Tanytarsns ( Rheotanytarsiis ) fuse us Ereeman 1954 Also described in Ereeman ( 1 958) and Ereeman & Cranston ( 1980). The adult male is described by Freeman ( 1 954 & 1958) and the adult male and pupa by Kyerematen & S aether (2000) but the pupal specimen irsed by the lat- ter authors lacked the thoracic honi. Scott (1967) described the distinctive characteristics of the pupa and larva and also gave a detailed account of the biology and behaviour of this species. Certain fea- tures of the pupa and larva have not yet been de- scribed and are added here. PUPA (N = 2 mounted) Scott (1967) illustrated the cephalothorax from the lateral and dorsal view; showing no frontal se- tae and none could be detected on the specimens examined here. In lateral view the cephalic horn is shown to be of the usual Rheotanytarsiis type with a broad base and a down-turned distal section bear- Annals of the Eastern Cape Museums Vol. 3 (February 2004): l-LS . ing small points. One long precorneal seta extends as far as the bend of the cephalic horn as shown in Fig. 1 8 (one of the mounted specimens), the other two being very short. Figure 19 shows details of the pupal anal spur on VIII. Otherwise these specimens conform to the descriptions of Kyere- maten & S aether (2000) and Scott ( 1967). LARVA (Numerous mounted and unmounted specimens). Body length: 2-4 mm, depending on trophic con- ditions. Colour: Scott notes; “In life the head is bright red- dish brown and the abdomen greenish. In alcohol the head is brown and the abdomen yellowish white.” Antenna: as per Scott, Lauterborn organs reach base of segment 5. Labruni: similar to R. curtistylus Goetghebuer. Illustrated by Pinder & Reiss (1983). Mandible: as per Scott (1967), all teeth are dark. Mentuni: all teeth are dark, width of ventromental plate 0.95 x width of men turn. Maxilla (Fig. 20); the lacinal chaetae are well- developed, there is no pecten galearis and there are two setae maxillaris. Body: Scott (1967) illustrated the posterolateral bifid setae on segments III to VI. One branch lies anteriorly, the other posteriorly, both are plumose and are about as long as one third of the segment. These are best .seen on unmounted specimens as the mounting medium may make them transparent and difficult to see. The anal setae are long and dark brown and the anal papillae are short with rounded tips. SPECIMENS EXAMINED 1 pupa; Cecilia Ravine, Table Mountain, 33°60'S I8“25'E, i.98 (cat. ABLCR.7T), collector Deni.se Schael; larvae from small waterfall. Silvermine River, 34"05'S 18°25'E (cat. SAC. 37K) 24 ix 95, collector ADH (all in AM). NOTES Habitat preference: stony torrents. Distribution: in permanent mountain streams in southern Africa. Rheotanytarsus guineensis Kieffer (Figs 21-25) Rheotanytarsus guineensis Kieffer, 1918 Tanytarsus (Rheotanytarsus) guineensis Freeman, 1958 Also described in Freeman & Cranston (1980) and Kyerematen & Ssether (2000). The male hypopygium is described by Freeman (1958) and the hypopygium, female genitalia and pupa are described by Kyerematen & Saether (2000). More details of the pupa and a description of the larva are given here. One larva was taken from a pupal case with pupa containing pharate male. PUPA (N=:2, mounted) Cephalothorax: surface finely mgose, two minute pairs of dorsocentral setae, thoracic horn and pre- corneal setae as in Fig. 21, horn downturned, very transparent with minute small points on distal half, discernable under high power magnification; pre- corneal setae also very transparent, one long and two short. Abdomen: paired spine patches on tergites II-IV, as per Keyermaten & S tether; anal spur (Fig. 22) simple. LARVA (N=6 mounted) Colour greenish in life; length c. 4 mm. Head capsule: light brown; length 312 pm. Antenna (Fig. 23): length 120 pm, AR 0.4; Lauter- born organs reach base of segment 5. Labruin: similar to that of R. curtistyliis Goetghe- buer (Pinder & Reiss 1983). Mandible (Fig. 24): all teeth dark brown, similar to the generic definition (Pinder & Reiss 1983) in- cluding the large seta interna consisting of four plumose branches. Mentiim: all teeth dark brown, width of ventromental plate 0.91 X width of mentum, specimen somewhat flattened on slide. Maxilla (Fig 25): similar to that of R. fiiscus but differing in small details such as the shape of the lacinal chaetae. Body: claws of parapods all simple; procercal setae long and dark; anal tubules short with rounded tips. Bifid posterolateral setae are present on at least some of the middle segments. These are plu- mose but transparent and difficult to discern on mounted specimens. SPECIMENS EXAMINED Two mounted pupae and one larva from the con- fluence of the Mpisini and Manzamnyama Rivers near Lake Mzingazi, 32“09'S 28°42'E(Cat. MpMan. 8/96 (1)) viii 96, collector Petra Vos, (all in AM). NOTES Habitat preference: running water; tolerates slower flow than some other species of the genus. Distribution: tropical and sub-tropical Africa. Rheotanytarsus montanus Lehmann (Figs 26-33) Rheotanytarsus montanus Lehmann, 1979 Also described in Kyerematen & Saether (2000). Lehmann (1979) described the adult male and the pupa; Kyerematen & Stether (2000) described the adult male. The adults are dark brown to almost black flies, apparently more heavily chitinized and more compact than other African members of the genus. The species is found in upper mountain re- gions where streams are torrential. South African pupae and Ethiopian and South African larvae are described here. PUPA (N = 4 mounted) Colour: mostly colourless; abdominal spurs brown. Cephalothorax: cephalic tubercles small with short frontal setae; thoracic horn (Fig. 26) long and slen- der, the distal, pointed third section more strongly chitinized than the rest and with a few small points; wing sheath with prominent nose; setae: two small antepronotals, three precorneal s (Fig. 26), one much longer and darker than the other two; two pairs of small dorsocentral s, close together. Abdomen (Fig. 27): no shagreen, tergites II-VI with anterior pair of point patches, very wide on II and III but width decreasing progressively so that those on V and VI are only as wide as long; patches of small spines just before and after hook row on tergite II; L setae very small or absent in some positions; three LS setae on V-VIII; fringe on anal lobe well developed; hook row on II small and undivided; pedes spurii A and B absent, but a few small transparent spines at the usual position of A on IV; posterolateral spur large with accessory points on ventral surface (Fig. 28). LARVA (N= 16 mounted) Head capsule: length 325 pm. Colour: in fresh specimens the head and antennae glossy black, body greenish, claws light, and anal setae dark brown; in mounted specimens the head and antennae dark brown. Some South African specimens medium brown. Head: dorsal surface, frontal apotome and labral 8 Harrison: Tanylarsini of Sub-Saharan Africa sclerites 1 and 2 all finely granular. Antenna (Fig. 29): AR 2.6. Segment 2 very short, blade extends to tip of segment 3; Lauterborn or- gans prominent. Labriinr. similar to that of R. curtistylus Goetghe- beur (Finder & Reiss 1983). Premandible with two teeth and dense brush (Fig. 30). Mandible (Fig. 31): all teeth brown, pecten man- dibularis well-developed. Menlum (Fig. 32): median tooth with lateral notches except in worn specimens, all teeth dark, width of ventromental plate the same as width of men turn. Maxilla: very similar to that of R. fusciis (Fig. 20). Body: apart from some simple setae, segments IV-VIII bear lateral setae with two main branches each bearing about six curved smaller branches (Fig. 33), in some cases the two main branches are adpressed so that the structure is not obvious; claws simple; anal tubules short with rounded tips. SPECIMENS EXAMINED Larvae from upper Kechene River, 09“ 04'N 38°45'E (Ers. 6K), 13 ix 83; Danka River, 07°05'N 39°46'E (Ers. 34F), 20 i 84, collector ADH (in ZSM); South African specimens: from the Eastern Cape Province, one larva from KuKowa Stream, tribu- tary of Slang-Mbashe River, (ECR 52) 7. xii 90; numerous larvae and four pupae from the Wilde- bees River near Gleneig, 31°13'S 28°04E (ECR 1 12C 1 & 2) 26 iii 93; from KwaZulu-Natal, one larva from Bushmans River above Tugela conflu- ence, 28“46'S 30"l0'E (Bus 57Z (2) 05 x 2000. Collectors F.C. de Moor and team (in AM). Figs 18-33. Rheotcmytarsus spp.: 18-20, R. fiisais. 18-19, pupa: 18, thoracic horn and frontal setae; 19, anal spur on Vlll. 20, larva, maxilla. 21-25, R. guineensis. 21-22, pupa: 21, thoracic horn and frontal setae; 22, anal spur on VIII. 23-25, larva: 23, antenna; 24^ mandible; 25, maxilla. 26-33, R. montanus. 26-28, pupa: 26, antenna and frontal setae; 27, tergites ll-V; 28, anal spur on VIII 29-33, larva: 29, antenna; 30, premandible; 31, mandible; 32, mentum; 33, lateral body seta. Annals of the Eastern Cape Miiseinns Vol. 3 (February 2004): 1-15 9 NOTES The South African pupa is very similar to that described by Lehmann (1979) notably in the wide point patches on tergites II & III. Habitat preferences-, the larvae were found among stones in very rapid currents in upper mountain streams; they seemed to be better adapted to faster currents than other species of Rheotanytarsus. The larval cases are similar to those of other species of the genus but the arm-like extensions of the cases, that support the silk strands, are comparatively much shorter. Distribution: Ethiopian Highlands, the Kivu dis- trict mountains of the Congo (ex-Zaire) and KwaZulu-Natal and Eastern Cape Province (Drak- ensberg Mountains), South Africa. Tanytarsus awashensis sp. nov. (Figs 34-37) Material e.xaminecb. this description is based on two drowned and damaged specimens. ADULT MALE (N = 2 mounted) Close to generic definition of Cranston et al. (1989). Wij}g length'. 0.98-1.17 mm. Colour, head with palps and antennae light brown; thorax and legs light brown, vittae, preepisternum and postnotLim brown, abdomen light brown with no obvious markings. Head-. AR 0.71-0.76; frontal tubercles about five times as long as the width of the base; palps (Fig. 34): 195 jttm long; segments measuring 16, 31, 40, 46, 62 jum (rather short); two subapical sensillae on segment 3. Thora.v. no scutal tubercle; setation: lateral ante- pronotals nil, dorsocentrals 6, posterior prealars 1, scutellars 2 per side. Wings-, setae in the apical third of 14+5 and mi +3. Wings were in too poor a condition to determine the setation of veins. Legs', scale of fore femur with long straight point; combs of other legs all with spurs. Tarsi were miss- ing in the specimens, so are not described. Hypopygiuin (Figs 35, 36 & 37): anal tergite bands almost transverse and not meeting; anal point with setae but no spine patches, downturned; superior volsella broad, almost square with short beak; digi- tus reduced; median volsellae short, brush-shaped and meeting centrally (Fig. 36); inferior volsella parallel- sided; apodemes as in Fig. 37. SPECIMENS EXAMINED 2 d'(5' found drowned in Lake Busata, a fresh- water lake near Awash railway station, Ethiopia 8^^05'N 4(y’30'E, xi 84, both the holotype and para- type on the slide ETC.67A (the holotype being the specimen near the label); collector ADH (in ZSM). NOTES Habitat preference', must have bred in the Lake Busata as there was no other water nearby. Distribution: Ethiopian Rift Valley. Gejieral note on systematics: T. awashensis is dis- tinguished from all other known African species of Tanytarsus (Freeman 1958; Ekrem 2001) and all the Palaearctic species described by Reiss & Fittkau (1971), by the combination of anal point with no spine patches; almost square superior volsella with a short beak and reduced digitus and short, brush- shaped median volsellae that meet centrally. Seg- ment 5 of the labial palps is short. Following the key in Ekrem (2001), T awsahensis reaches couplet 7 but differs markedly from the two species keyed there: from T. pallidulus Freeman in the shape of the superior volsella (not roughly oval) and its me- dian volsella (not rounded) (Freeman 1958), and from T. atrocincus Goetghebuer, which has an L- shaped superior volsella, a distinctly rounded me- dian volsella and an almost club-shaped inferior volsella (Freeman 1958). Etymology: from the Awash River, Ethiopia. Tanytarsus fliimineus sp. nov. (Figs 38-43) ADULT MALE (N= 4 mounted) Close to generic definition of Cranston et al. (1989). Body length: 2.7 mm. Wing length: 1.8 mm. Colour: whole body yellowish when preserved in alcohol Head: AR 0.52-0.61; eyes with narrow dorsal ex- tension; no frontal tubercles; head setation with seven verticals; palp segments measuring 30, 33, 96, 105, 192 jttm, no subapical sensilla on segment 3. Thorax: no scutal tubercle; setation: lateral ante- pronotals nil, dorsocentrals 9 uniserial, posterior prealars I, scutellars 4 per side. Wings: no anal lobe; membrane setae dense around wing tip from costa into mj+4, sparser over most of the rest of the wing; vein seta: brachiolum I, R 22, Ri 36, R4+5 50, numerous on other veins. Legs: tarsi are not described as most were missing from the material available; spur on foretibia short and curved. LR mid 0.7; sensilla chaetica on tar- somere I, midleg 6. Hypopygiuin (Figs 38, 39, 40, 41 & 42 ): anal tergite bands not meeting; few small median anal setae; anal point downturned with large crests with long spicules between (Fig. 39) with a few apical anal setae on either side; superior volsella short with at least 20 dorsal setae and a patch of microtrichia at lateral base; protruding digitus with no seta, but with a patch of microtrichia on its base (Fig. 40), 10 Harrison: Tanytarsini of Sub-Saharan Africa median volsella (Fig. 41) broad and long with short setae; apodemes as in Fig. 42. ADULT FEMALE (n=I mounted) Body length: 2.0 mm. Wing length: 1.4 mm. Colour, similar to male. Head: AR 0.3. No frontal tubercles, eyes like male. Setation: verticals 7. Palp segments 30, 36, 99, seg- ments 4 and 5 missing. No siibapical sensory sen- silla on segment 3. Thorax: setation: lateral antepronotals nil, dorso- centrals 1 1, posterior prealars 1, scute liars 2. Wings: membrane setae similar to male. Vein seta- tion: R 13, R| 28, R4+5 55. Legs: fore tibia with scale and spur like male. All tarsi missing from the available specimen so not described here. Genitalia (Fig. 43): S VIII forming a small floor under anterior part of the vagina; gonopophysis VIII divided into rounded dorsomesal lobe (Fig. 43a), large ventrolateral lobe (Fig. 43b) and small apodeme lobe (Fig. 43c); gonocoxapodemes nar- row and light in colour and joined; coxosternapode- mes broad and light in colour; notum long; gono- coxite IX closely applied to body with two to four setae; segment X without setae; postgenital plate triangular; cerci small (48/xm); seminal capsules ovoid, large (63/xm); seminal ducts with curves. central portion glandular, with common opening. SPECIMENS EXAMINED 10 tmd 1 9 from Little Mooi River, KwaZulu-Natal, 29°13'S 29*’53'E, 4. iv. 95, light trap (Cat. MOI 73AC, AD 1-4); {S holotype MOI 73 AD, S paratype MOI 73 AD 3; $ paratype MOI 73 AD 4). I S Bushmans River above water- fall, 28°46'S 3()°I()'E, 17 viii 99, (cat. BUS 48L (3)); collectors F.C. de Moor and team (in AM). NOTES Diagnostic features are the very large crests and long spicules between on the male hypopygium that distinguish this species from all other species in the genus described by Freeman (1958), Reiss & Fittkau (1971) and Ekrem (2001). This species does not fit in with features described in Ekrem’ s key apart from the main division in couplet 1 and the division in couplet 8 that leads to the group without a seta on the digitus. The female genitalia differ from those described for the genus by S aether (1977) as the floor under the an- terior part of the vagina is much smaller than he de- scribes, also the species he examined had a simple gonopophysis VIII not divided into three paits as in this species. In discussing the genus he notes: ‘There are several shaip differences between the female geni- talia of the species examined”, so the female genitalia do seem to be very variable in this genus. Etymology: ‘fhiniineus ’ Latin (Ovid) riverine. Figs 34-43. Tanytarsiis spp. 34-37, Tany tarsus awashcnsis, adult male: 34, maxillary palp; 35, hypopygium; 36, median volsella; 37, apodemes. 38-43, Tanytarsiis fliimineiis. 38-42, adult male: 38, hypopygium; 39, anal point, lateral; 40, superior volsella; 41, median volsella; 42, apodemes. 43, adult female, genitalia. 1 1 Annals of the Eastern Cape Museums Vol. 3 (February 2004): 1-15. Habitat preferences: the adults were caught along- side rivers so, presumably, the larvae lived there. Distribution: known only from KwaZulu-Natal, South Africa. Tanytarsiis zinibabwensis sp. nov. (Figs 44-46) (Note: the specimen was originally pinned). ADULT MALE (N = 1 mounted) Close to generic definition of Cranston et al. (1989). Body length: 1.7 mm. Colour: pinned specimen with head light brown, thorax light brown, vittae darker, abdomen yel- lowish. Wing length: 1.2 mm. Head: AR 1.0. Eyes with short parallel-sided dor- sal extension; no frontal tubercles; palp segments damaged but appear to be normal for the genus. Thorax setation: lateral antepronotals nil, dorso- centrals 10, posterior prealars 1, scutellars 2. Wings: no anal lobe; setae on membrane: dense patches of setae at tips of i'4+5and mi+2 with an ir- regular row in the rest of these cells, dense patch of setae at tip of 1113+4 and a few in anal cell; vein setation: R 16. Ri 12. R4+5 22, other veins all with setae. Legs: mid and hind tibia each comb with a short straight spur. (Note: in the specimen examined the forelegs were missing and the other legs in poor condition). Hypopygium (Figs 44, 45 & 46): anal tergite bands separate; seven anal tergite setae and three apical anal tergite setae per side; anal point with no cen- tral ridge, crests or spines, bare and rounded at the tip; superior volsella dog’s-head shaped with long, protruding digitus; median volsella (Fig. 45) short with lamelliform setae; gonostylus small and nar- row. The apodemes are illustrated in Fig. 46. SPECIMEN EXAMINED 1 (S (holotype) bred out in laboratory from stream from granite dome Ngoma Kuriru, Chin- domora, Zimbabwe, 17°35'S 31°10'E, 25. ii. 1964. (CCA.96C). Collector ADH (in AM). NOTES The following features are of diagnostic impor- tance: the bare anal point with no central ridge, the dog’s-head shaped superior volsella; the distinctive median volsella and the small and narrow gonosty- lus. These distinguish this species from those de- scribed in Freeman (1958), Reiss & Fittkau (1971) and Ekrem (2001). This species keys out to couplet 6 in Ekrem’ s key but the distinctive structures of the superior and media volsellae and the narrow gonostylus easily separate it from the species keyed there - T. atomariiis Kieffer and T. paUidissimus Kieffer. Habitat Preferences: the reared larvae were col- lected from stones in rapids in a small, swift- flowing stream. Distribution: known only from Zimbabwe. Virgatanytarsus aboensis sp. nov. (Figs 47-50) Material examined: this description is based on males from Ethiopia and one from Zimbabwe. De- scriptions of the Zimbabwean specimen are given in brackets. ADULT MALE (N = 1 1 mounted) Close to generic definition of Cranston et al. (1989). Body length. Ethiopia (N=9): 4.25-3.65 mm; Zim- babwe (N=2): 3.1 and 2.7 mm. Wing length. Ethiopia (N=9): 2. 9-2. 5 mm; Zim- babwe (N=2): 2.0 and 1.9 mm. Colour: head with antennae and palps brown; tho- rax brown, vittae, preepisternum and postnotum dark brown; wings with brownish tinge; legs brown; abdomen and hypopygium brown. Head: AR 0.88-0.98 (0.90); frontal tubercles pre- sent; palp segments measuring 34, 53, 136, 140, 248 (34, 50, 124, 146, 248) ptm; two subapical Figs 44-46. Tanytarsiis zinibabwensis, adult male: 44, hypopygium; 45, median volsella; 46, apodemes. 12 Harrison: Tanytarsini of Sub-Saharan Africa Figs 47-50. Virgalanytarsiis aboensis, adult male: 47, hypopygium; 48, superior volsella; 49, median volselia; 50, apodemcs. sensillae on segment 3. Thorax: no scutal tubercle; setation: lateral ante- pronotals nil, dorsocentrals 9-10, posterior prealars 1, scLitellars 5 per side. Wings: most of the wing extensively covered by setae, denser distally, but only present distally in cu and an. Setation of veins: brae hiol urn 1, R 28, R| 25, R4+5 25; other veins with setae except subcosta. Legs: LR fore 2. 3-2. 6; mid 0.6; hind 0.64-0.74. Sen- silla chaetica on tarsomere 1, on midi eg 5-7, on hindleg nil. Table 3 shows leg measurements and ratios. Hypopygium (Figs 47, 48, 49, 50): anal tergite bands separate but appear to continue posteriorly as darkly pigmented stripes that almost join Just be- fore the base of the anal point that has a short reflexed rod between the crests; superior volsella (Fig. 48) with at least 15 dorsal setae and digitus protruding in most specimens; median volsella (Fig. 49) with tip wide and flattened, edges tending to distort upwards, lamellae pointed; inferior vol- sella with a large, rather flat process (crista dor- salis?) without microtrichia, extending dorsome- dially or dorsally. The apodemes are shown in Fig. 50, the pair of rods immediately ventral to the sternapodeme are present in all mounted speci- mens and may be part of the phallapodeme. Etymology: V. aboensis from the Abo River in Ad- dis Ababa and Ethiopia. SPECIMENS EXAMINED I c5' Addis Ababa, Ethiopia, ()9°()0'N 38°47'E, (ETC.5J) xi. 82; 6 c5'6' from Abo River, Addis Ababa, Ethiopia, 09°04'N 38°47'E (ETC. 42F & G) 8 ix 84, I (ETC.49H) v.85, 2 (ETC. 51M1 & M2) V. 85, I (ETC. 60J) ix. 85, I (ETC. 6IC) ix. 85; 1 6' from Kosso River Ethiopia, 09°43'N 39°39'E, (ETC. 26K) 12 i 84; 3 c5'(5' from Chindomora, Zim- babwe, I7°36'S 3I°08'E (CCA. 24C & E, 96G) 10. ii.63 (collector ADH). All deposited in ZSM ex- cept CCA. 96G (that is in AM). Other material: SS from Tugela Estates below Blauwkranz River confluence, upstream of Bush- mans River, KwaZulu-Natal, 28°45’S 30°09’E (TUGI27 AE6-I0) 18 viii 99 (collectors F.C. de Moor and team). This material was in poor condi- tion and was not used for description; deposited in AM. Holotype S ETC. 51M1; paratype S ETC. 42G. Table 3: Virgatanytarsus aboensis, leg measurements in pm and ratios fe ti ta. taz ta.-, taj ll\s . LR BV SV c c C3 0 ■a. s Pi 750 500 1200 500 400 350 175 2.4 0.96 1.6 .2 ‘u jC ^ £> .5 a 0 P C- P2 750 650 500 250 175 100 75 0.58 0.26 3.0 N P-3 850 850 625 350 300 200 75 0.74 0.37 2.5 LEGEND: fe = femur; ti= tibia; ta,. tan, 133 104 tas= tarsomeres 1-5; LR = leg ratio (length of tarsomere I: length of tibia); BV = ‘Beinverhaltnisse’ = combined length of femur, tibia and tarsomere I divided by length of tarsomeres 2 to 5; SV=‘.Schenkel-Schicne Verhiiltnis' = ratio of femur plus tibia to tarsomere 1 . Annals of the Eastern Cape Mnseinns Vol. 3 (February 2004): 1-15 . NOTES Differential diagnosis: anal point with short re- flexed rod between crests, inferior volsella with a large flat process, without microtrichia, extending dorsomedially or dorsally. Only two further spe- cies were known previously from sub-Saharan Af- rica: Vi rgatanytarsiis ardnensis Goetghebuer and V. nigricornis Goetghebuer. The first is widespread in the tropics and the second in the tropics south to Kwa-Zulu-Natal. E aboensis is easily distinguished from these by the shape of the inferior volsella. The specimens from Zimbabwe came from a pristine mountain stream running off a large gran- ite inselberg and the Ethiopian specimens came from a mountain stream immediately above Addis Ababa that was somewhat organically enriched; this could account for the larger specimens from that stream. The KwaZulu-Natal specimens were drowned in the river and unsuitable for measurements. Habitat Preferences: this species appears to breed in mountain streams, even in stony runs. Distribution: Ethiopian Highlands, Zimbabwe and KwaZulu-Natal. ACKNOWLEDGEMENTS The Ethiopian part of this study was part of a programme of co-operative research on fisheries and limnology, developed between Addis Ababa University, Ethiopia, and the University of Water- loo, Ontario, Canada, and aided by the Canadian International Development Agency. I wish to thank Dr. H. B. N. Hynes and Ato Tes- faye Berhe for specimens, and other colleagues at the Biology Department, Addis Ababa University, for their help with transport and field work. The author also thanks the following for speci- mens and other material used in this study: Dr Eerdy de Moor and his team from the Albany Museum, Gra- hamstown. South Africa, and Rebecca Tharme and Denise Schael of the Ereshwater Research Unit, University of Cape Town. REFERENCES Cranston, P.R. 1998: The Australian species of Neozavrelia Goetghebuer (Diptera: Chironomidae: Tanytarsini). Aus- tralian Journal of Entomology 37: 107-1 12. Cranston, P.R., Dillon, M.E., Pinder, L.C.V. & Reiss, F. 1989. The adult males of Chironominae (Diptera: Chi- ronomidae) of the Holarctic region - keys and diagnoses. In: T, Wiederholm (ed.) Chironomidae of the Holarctic region. Part 3. Adult males. Entomologica Scandinavica Snpplenient 34: 353-502., Ekrem, T. 1999. Cladotanytarsns bnkavns (Lehmann, 1979) comb. n. and C congolensis (Lehmann, 1979) comb.n. from Central Africa (Diptera: Chironomidae). Annales limnologiae. 35: 185-191. Ekrem, T. 2001. A review of Afrotropical Tanytarsns van der Wulp (Diptera: Chironomidae). Tijdsclirift voor Ento- niologie 144: 5-40. Freeman, P. 1954. Chironomidae (Diptera) from Western Cape Province - III. Proceedings of the Royal Entoniologi- clal Society of London B. 23: 17-24. Freeman, P. 1958. A study of the Chironomidae (Diptera) of Africa south of the Sahara. Part 4. Bulletin of the British Mnsenin (Natural History.) Entomology 6: 263-363. Freeman, P. & Cranston, P.S. 1980. Family Chironomidae. In: Crosskey, R.W. (ed.) Catalogue of the Diptera of the Afrotropical Region. London: British Museum (Natural History): 175-202. Kieffer, J. J. 1918. Chironomides d’Afrique et d’Asie conserves au Musee National hongrois de Budapest. Annales de Miisee de Histoire-naturelle national hongrois. 16: 31-139. Kieffer, J. J. 1923. Chironomides de I’Afrique Equatorial. 3® partie. Annales de Societe entomologicpie franqais. 92: 149-204. Kyerematen, R.A.K., S aether, O.A. & Andersen, T. 2000. A review of the Rheotanytarsus pellucidus group (Diptera: Chironomidae). In: O. Hoffrichter (ed.) Late 20''' Century Research on Chironomidae: An Anthology from the 13''' International Symposium on Chironomidae, Freiburg, 5-9 September 1997. Aachen: Shaker: 147-170. Kyerematen, R.A.K. & Saether, O.A. 2000. A review of Afrotropical Rheotanytarsus Thienemann & Bause, 1913 (Diptera: Chironomidae). Tijdsclirift voor Entoniologie 143: 27-69. Lehmann, J. 1979. Chironomidae (Diptera) aus Fliessgewassern Zentralafrikas. Teil I: Kivu-Gebiet, Ostzaire. Spixiana Supplement. 3: 1-144. Lehmann, J. 1981. Chironomidae (Diptera) aus Fliessgewassern Zentralafrikas. Teil II: Die region um Kisangani, Zen- tralzaire. Spixiana Supplement. 5: 1-85. Pinder, L. C. V & Reiss, F. 1983. The larvae of Chironominae (Diptera: Chironomidae) of the Holarctic region - Keys and diagnoses. In: T. Wiederholm (ed.) Chironomidae of the Holarctic region. Part I larvae. Entomologica Scand- inavica Supplement 19: 293-435. Pinder, L.C.V. & Reiss, F. 1986. The pupae of Chironomidae (Diptera: Chironomidae) of the Holarctic region - Keys and diagnoses. In: T. Wiederholm (ed.) Chironomidae of the Holarctic region. Part 2 pupae. Entomologica Scand- inavica Supplement 28: 299-456. 14 Harrison: Tanytarsini of Sub-Saharan Africa Reiss, F. & Fittkau, E.J. 1971. Taxonomic und Okologie europaische verbreiteter Tcuiytarsiis-Avlen (Chironomidae, Diptera). Arcliiv fiir Hydrobiologie Supplement. 40: 75-200. Saether, O.A. 1977. Female genitalia in Chironomidae and other Nematocera; morphology, phylogenies, keys. Bulletin of the Fisheries Research Board of Canada 197: 21 I pp. Srether, O. A. 1980. Glossary of chironomid termim)logy (Diptera: Chironomidae). Entomologica Scandniavna Sup- plement 14: 1-51. SaTher, O.A. & Andersen, T. 1998. Friederia, a new Afrotropical tanytarsine genus (Diptera: Chironomidae). Eiito- mologica scandinavica 29: 29-37. Sliwedal, L. 1982. Description of Nidnnrbia n. gen. with notes on the distribution of Micropsectra Kieffer (Diptera: Chironomidae). Entomologica Scandinavica 13: 317-320. Scott, K. M. F. 1967. The larval and pupal stages of the midge Tanytarsns (Rheotanytarsns) fnscns Freeman (Diptera: Chironomidae). Journal of the Entomological Society of South Africa. 30-2: 174-184. SHORT COMMUNICATION METRIOCNEMUS CAPICOLA, A REPLACEMENT NAME FOR METRIOCNEMUS CAPENSIS HARRISON 2002 AD Harrison Freshwater Research Unit. Department of Zoology, University of Cape Town, Roncleboseh, Cape Town Harrison (2002) named a new species of Metriocueiiuis, 'M. capensis, overlooking the fact that the name was preoccupied. Freeman & Cranston { 1 980) listed another species, Parametriocnemus cap- ensis (Freeman), but Freeman (1954) originally named that species Metriocnemus capensis. Hani son’s spe- cies name is therefore invalid and must be considered a homonym, Metriocnemis capensis Hairison homonym. A replacement name, Metriocnemus capi- cola iiomen novum is proposed for Harrison’s (2002) species. REFERENCES Harrison A. I). 2002. Chironomidae (Diptera) m the Al- bany Museum Part 1 . Annals of the Eastern Cape Mu- seums 2: 9-18. Freeman, P. 1954. Chironomidae from the Western Cape Province-IV. Proceedings of the Royal Entomo- logical Society of London Series B 23: 172-180. Freeman, P & P. S. Cranston 1980. Family Chi- ronomidae. In: Crosskey, R.W. (ed.) Catalogue of the Diptera of the Afrotropical Region. London Brit- ish Museum (Natural History): 287-368. Annals of the Eastern Cape Mnseiinis Vol. 3 (Febiiiary 2004): !-15 . 15 A PRELIMINARY INVESTIGATION INTO THE INFLUENCE OF TURBU- LENCE ON LARVAL FEEDING IN TWO SPECIES OF BLACKFLY, SIMULIUM CHUTTERI LEWIS AND SIMULIUM NIGRITARSE COQUILLETT (DIPTERA, SIMULIIDAE), FROM THE GREAT FISH RIVER, SOUTH AFRICA H M Barber- James Department of Freshwater Invertebrates, Albany Museum, and Department of Zoology li Entomology, Rhodes University, Grahamstown , South Africa. ABSTRACT Since 1977, larvae of the pest species Simiiliiini cluitteri Lewis have largely replaced Simiiliiun nigritarse Coquillett as the dominant species of blacktly in the Great Fish River in the eastern Cape. This is thought to be due to a change from intermittent to continuous water flow, since the completion of an interbasin water transfer scheme from the Orange River to the Great Fish River. Changes in turbulence of the flow may be one of the factors responsible for this shift in species composition, and in this study, the effect of turbulence on the feeding of larvae was investigated under laboratory conditions. Turbulence was measured as Reynold’s number, with larvae feeding on algae under differ- ent conditions of turbulence created in glass tubes. Feeding activity was measured by measuring algal (Clilorella) cell counts and chlorophyll a concentrations from gut contents after feeding, and by cephalic fan activity. The results indi- cate that, under conditions of higher turbulence, S. cluitteri feeds more efficiently than does S. nigritarse. Keywords: blackfly larvae, turbulence, flow rates, Reynold’s number, feeding behaviour INTRODUCTION Similiiim cluitteri Lewis is well known in its adult stage as a pest, as the females imbibe blood from livestock (Chutter 1968). Car & de Moor (1984) quote farmers reporting losses in stock pro- duction and occasional deaths in young animals through excessive blood feeding by this species. Studies by Chutter (1972) and Scott et al. (1972) showed that between 1970 to 1971, Siniiiliiim nigritarse Coquillett and Simiiliiim adersi Pomeroy were the dominant blackfly species in the Great Fish River at Carlisle Bridge (33W55"S 26M3' 45"E), although S. cluitteri was present in low numbers (O’Keeffe & de Moor 1988). Since 1977, the Great Fish River and its suiroundings (Fig. 1) have been plagued by increasing numbers of S. cluitteri, which replaced S. nigritarse and S. ad- ersi as the dominant blackfly species (de Moor & O’Keeffe 1987). The change in species composi- tion occurred after the completion (in 1977) of an interbasin transfer scheme whereby water was channelled from the Orange River to the Great Fish River. The water transfer resulted in a change in the Great Fish River from intermittent flow to per- ennial flow (O'Keeffe & de Moor 1988). A similar change in species composition has been reported with damming of rivers in northern Alberta and Saskatchewan, whereby Siuuilium Juggeri Nicholson and Mickel replaced Siuuiliiim articum Malloch as the dominant species, after changes in flow regimes, following impoundment (Wood 1985). Coetzee (1982) found that in the up- per reaches of the Great Fish River, upstream of the outlet of Orange River water, S. nigritarse larvae 16 were still more abundant than those of S. cluitteri. An unpublished survey earned out by the author and colleagues during 1985 indicated that S. nigritarse was a dominant macroinvertebrate species in the upper reaches of the Great Fish River. Although regulated by numerous small farm dams, the flow in these reaches continued to be seasonal. More recent observations (pers. obs. and communication from local farmers) indicate that water in these reaches remains clear for most of the year. Condi- tions may, however, become turbid during spates after rainfall. The water in the lower reaches is gener- ally more turbid. O’Keeffe & de Moor (1988) investi- gated flow patterns and water chemistry for the Great Fish River, both upstream and downstream of the out- let, prior to, and after, the introduction of Orange River water. They found that there had been marked changes in flow regimes and water chemistry below the outlet, compared to the situation prior to the intro- duction of Orange River water. Downstream of the Orange River water outlet, relative abundances of both S. nigritarse and S. adersi declined, while those of S. cluitteri increased greatly subsequent to the inflow of Orange River water (O’Keeffe & de Moor 1988). Current veloc- ity and water volume are known to be important in controlling habitat suitability for various Siiiniliiiiu species (de Moor 1994). S. cluitteri is normally found in fast-flowing, large rivers, while S. nigritarse and S. adersi are generally found in small streams to medium sized rivers (de Moor 1989; Palmer & de Moor 1998). Palmer & de Moor (1998) have also indicated that, although S. cluitteri has been re- corded in low numbers from several small, clear rivers in the eastern Cape, it only achieves high Barber-James: Turbulence and Lairal Feeding in Blackfly Fig. 1: The Great Fish River, indicating the position of Carsile Bridge. The dashed line reirresents the tunnel connecting the Orange and Great Fish river systems. population densities in large, turbid rivers, such as the Vaal River. During a period of low flow fol- lowing a severe drought (from 1980-1982), when the turbidity in the lower Vaal river was noted to be unusually low, the normally-dominant sinuiliid species, S. chufteri, was replaced by S. hargreavesi (Car & de Moor 1984; de Moor 1994). The avoid- ance of clear water by S. chulteri is further con- firmed by the observation by Palmer & O’Keeffe (1990) that numbers of S. chulteri drop signifi- cantly downstream of impoundments where clearer water is released into normally-turbid rivers. The effect of changing flow conditions on feed- ing activity has been documented for several simu- liid species. For example, distinct differences in fan adduction activity relating to increased cuiTent speed have been shown for nine European Siimi- liiim species (Schroder 1980, 1988). Studies on feeding behaviour, rates of ingestion and selectiv- ity of food particle size have also been carried out for several Nearctic species, for example, Kurtak ( 1978), Craig & Chance (1982), Ciborowski & Craig (1989), Hart & Latta ( 1986), Hart et al. ( 1991 ). For this study, laboratory experiments were set up to investigate the effects of increasing tur- bulence, defined by Reynolds number, on the feeding ability of S. chutteri and S. nigritarse. MATERIALS AND METHODS The calculation of Reynold’ s nnnihcr Reynolds number is the ratio of inertial forces to viscous forces in liquids (Smith 1975; Vogel 1981 ) and defines the properties of fluid motion. Low Reynolds numbers (below 500) indicate laminar flow, while Reynolds numbers approaching 2000 or more indicate turbulent flow. Reynold's number (Re) is calculated according to the equation: Ul p U1 Re = — = — p u where: U is the mean velocity (cm s ') I is the tube diameter (cm) p is the density of water at a specified temperature y is the dynamic viscosity of water u is the kinematic viscosity of water Annals of the Eastern Cape Museums Vol. ?< (February 2004): 16-24. Collection of samples and measurement of larvae Larvae were collected from the Great Fish River at Carlisle Bridge (Fig. 1), during late May (winter) and again in late October into early No- vember (summer) for both species. As it was not possible to do all the experiments at the same time, summer populations of larvae of both species were used for the haemocytometer-analysed results and for fan adduction counts, while winter populations were used for those analysed by lluorometry (see analysis of feeding below). As final instar larvae do not feed (de Moor 1982a), sixth instar larvae were selected for all experiments. These differ from final instar larvae in that the respiratory histo- basts are not yet fully developed and the cervical sclerites are not yet fully separated from the preoc- ciput (de Moor 1982 c & b). Recently-moulted lar- vae (those with pale yellow head capsules) were used, as the rate of feeding may differ in older lar- vae of the same instar (de Moor 1982b). Larvae of each species from both populations were measured to compare sizes, in case one group should be big- ger than the other, which could also influence feed- ing. The total body lengths of 75 typical sixth in- star lai'vae of each species were measured using a dissecting microscope with a graticule. Differences in winter and summer populations have been pre- viously documented (de Moor 1982a, 1982b), and can be explained in terms of reduced water tem- peratures resulting in slower growth rates, longer development periods and, hence, the development of larger larvae. No attempts were made to investi- gate sexual difference between larvae in these ex- periments, and although de Moor (1982a) found that females of later stage larvae were larger than the equivalent males, Elsen et al. (1978) have indi- cated that sex has no influence on larval feeding in S. danmosum. Acclimatization Before initiating the experiment, the larval ali- mentary tracts had to be cleared of food. A section of plastic guttering, connected to a water inlet at one end and with an outlet at the other end (Fig. 2), was set up as a channel to allow gut clearance. A continual flow of clean water could pass through this channel when required. Strips of hard, clear plastic measuring 5 mm by 100 mm, with cotton loops at each end, were suspended lengthwise down the channel, into the flowing water, from wooden rods, which rested across the channel (Fig. 3). Larvae collected from the field were placed in the channel, and some of the larvae at- tached themselves to these plastic strips, which would later be transferred, bearing larvae, into the experimental tubes. The larvae were retained in the channel in algae-free water to clear their alimen- tary tracts before starting the feeding experiments. The gut retention time for Siiiniliiim species has been shown to be highly variable, ranging from 0.5 Figs 2 & 3. Apparatus used for gut clearance of blacklly larvae. 2, photograph of the apparatus showing diagonal bars from which |)lastic strips were suspended. 3, Diagrammatic representation of a section of the rigid plastic strips on which larvae settled. Arrows indicate the direction of water How. Ikirber-James: Turbulence and Utrval Feeding in IHackfly hours to 4 hours (Kurtak 1978; Ladle & Hansford 1981; Fredeen 1964; Chance 1970). Preliminary investigations were carried out to establish the re- quired clearance times for S. chutteri and S. uigri- tarse. This was done by removing larvae from the channel every 30 minutes and investigating the contents of the alimentary tracts on a slide under a microscope. These tests revealed that 2 hours was a sufficient time period for gut clearance in both S. chutteri and S. nigritarse. Quantitative meas- urements were not done for this. The larvae used for these experiments were acclimatized to 21°C in the laboratory, irrespective of season collected, and the water temperature in the experimental tank was maintained at 21°C. Winter water temperatures would have been cooler, but it was decided to approximate summer temperatures for the experiments as feeding may have been reduced and would, therefore, be harder to measure at lower water temperatures. Apparatus and Experimental Procedure The experimental apparatus for this investiga- tion was modified after Noble (1970). It consisted of a glass tank with a capacity of 15/, filled with a suspension of Chlorella, in which a system of in- terlocking, open-ended glass tubes (Figs 4 & 5A) was submerged. The upper tube was wider (internal diameter 25 mm) than the lower tube (internal diameter 10 mm), since tubes with smaller diameters produce more turbulent condi- tions. The upper tube was attached to a water pump, which could be adjusted to vary flow rates. The flow of water from the pump through the tubes was controlled using a rheostat. Velocities at dif- ferent rheostat settings were determined by intro- ducing drops of 1.25M potassium permanganate solution (used as an indicator dye) through a port on the upper side of the tube near the mouth (Fig. 5A), into the middle of the stream of water passing through the tube. The nucleus of the droplet moved forward, with a thin trace of colour trailing behind. The progress of the droplet was timed over a fixed distance. This was repeated twenty times for each rheostat setting to estimate the velocity so that Reynold’s number could be calculated. Laboratory tests involved a three-step process: field collection and measurement of larvae; labora- tory acclimatization, and feeding experiments. Feeding tests After clearing the alimentary tracts of food im- bibed in the natural river environment (as de- scribed above), the plastic strips, bearing larvae, were carefully transfen'ed to the centre of one of the glass tubes in the tank (Fig. 4). The strips were positioned longitudinally down the center of a tube and attached by the cotton loops to hooks within the tube (Fig. 5B). In this way, different groups of larvae could be subjected to one of six different conditions of turbulence (i.e. six values for Rey- nold’s number). A maximum of about 10 larvae were introduced to a tube each time, since it was found that, at higher densities, feeding of individ- ual larvae was often interrupted by movements of neighbouring larvae. Once the larvae had settled in the tubes (2 min- utes acclimatization allowed each time), the flow Fig. 4: The experimental apparatus within an aquarium (with water level lowered to show inner apparatus more clearly) showing tubes in which turbulence can be varied. LEGEND: GT = glass tubes; H = heat stirrer; 1 = inlet; O = outlet; P = part of pump. Annals of ihc Eastern Cape Museums Vol. 3 (February 2004): 16-24 19 was switched on, creating the required level of tur- bulence. Feeding was allowed to continue for 10 minutes before switching off the flow and remov- ing the larvae from the tube. Larvae were placed in labelled vials containing 5% formalin, or in cold insect saline, and immediately refrigerated, de- pending on the subsequent method of analysis of algal content (see below for a description of these methods). Oxygen concentration of the water in the tank was frequently determined using the Winkler method, to confirm that it was never in short sup- ply. Because the water was rich in algae, Alster- berg’s modification of the Winkler titration for water rich in organic matter was used (in Macker- eth et al. 1978). ANALYSIS OF FEEDING Three methods were used to compare feeding in the two species. Larval gut contents (equivalent to the amount of algae ingested) were estimated using haemocytometer counts of algal cells, and fluorometric measurements of chlorophyll a. A third estimate involved counting the number of ad- ductions of a blackfly's cephalic fans over a fixed period of time, to see if the behavioural response was affected by changing turbulence. Haemocytometer analysis After feeding, the larvae were immediately pre- served in 5% formalin for later dissection. The entire gut content of a larva was dissected and dispersed in LOO ml of distilled water and sub- samples of this were examined under a Neubauer Bright Line Haemocytometer. The haemocytometer was standardized to hold exactly 0.0009 ml of sam- ple, and all of the algal cells present in this volume were counted. Five separate counts of alga! cells were recorded for each 1 ml sample (i.e. for each larva), and this was done for the gut contents of 15 larvae at each of the six levels of turbulence {Re values). The mean number of ceils per 0.0009 ml sample from the gut contents was taken to repre- sent the amount ingested for each larva, and these values were compared for the two species. Com- parisons between the feeding of the two species was done using a t-test at each level of turbulence. A control experiment was run to see if the same quantities and size ranges of cells were available to both species. Algal cell counts and size measure- ments were done on water samples taken directly from the 15/ tank prior to each set of experiments. Comparisons of the size of algae in the tank to the size ingested by the larvae of each species were analysed using t-tests. Fhiorometry analysis Immediately after feeding, larvae were placed in a vial containing cold insect saline and refrigerated at 4°C. Refrigeration was chosen rather than storage in a preservative so that the chlorophyll a P X B Y Fig. 5. A. Diagrammatic representation of interlocking, open-ended glass tubes used to measure turbulence. Drops of potassium pemianganate are introduced through a small aperture (P) to estimate cuixent speeds (and hence, turbulence). B. Detail of section X-Y (from A) illustrating lawae resting on plastic strips. Aitows indicate direction of water Row. 20 Barber-James: Tiirbidenve and Ijin’al Feeding in Biarkfly would not be chemically altered. Dissection was undertaken as soon as possible thereafter. Alimen- tary tracts were dissected from ten larvae for each of the six turbulence levels, gut contents were then pooled from each group of ten and placed in a vial containing 10 ml of 90% acetone. This was carried out three times for each of the six turbulence lev- els. Chlorophyll a concentration was measured us- ing a Turner Fluorometer model 111. This was set at its greatest sensitivity and calibrated using stan- dard chlorophyll a of known concentrations and 90% acetone as a blank. The feeding of the two species at different turbulence levels was compared using a t-test. A control test of the chlorophyll a concentration of the water in the tank was measured at the onset of each set of experiments to check that conditions remained constant. Observations of cephalic fan adductions In blackfly larvae, particles from flowing water are trapped in the cephalic fans and transfeired to the mouth by means of fan adductions, combined with a series of movements of other mouthparts. Fan adduction rate therefore gives an indication of the feeding rate (Craig 1977; Craig & Chance 1982; Hart & Latta 1 986; Currie & Craig 1987; Palmer & Craig 2000). The number of cephalic fan adductions per minute were noted for three individual larvae at each value of Reynold’s number for both S. chulteri and S. nigritarse. Craig & Chance ( 1982) found that only the primary fans of simuliids are directly involved with capturing particles from the water, and only these fans were observed. While the animals were in the experimental tubes, each individual was observed three times, for a minute each time. A dissecting microscope was set up against the tank to facilitate observa- tions. The contractions of only one fan were counted for each larva. Comparisons between the fan adduction rates of the two species was done using a t-test for each turbulence level. RESULTS Length measurements Body length measurements of 75 larvae at the chosen stage of development (sixth instar) show that there were no significant differences between the sizes of larvae within each species (p<0.05), and that the sixth instar larvae of the two species had a similar size range. The winter populations of both species were, however, significantly (p<0.005) larger than the summer populations. Information on the sizes of larvae in the two populations is given below. Annals of the Eastern Cape Museums Vol. ?i (February 2004): 16-24. SUMMER POPULATIONS S. chiitteri range = 3.1 5-4.35 mm; mean = 3.65 mm ± 0.34. S. nigritarse range = 3.22-4.04 mm; mean = 3.58 ± 0.03. WINTER POPULATIONS S. chiitteri range= 4.72-6.20 mm; mean = 5.36 ± 0.43. S. nigritarse range = 4.49-6. 1 5 mm; mean = 5.27 ± 0.52. Feeding experiments Measurements of algal cell sizes revealed that similar sized cells were available in the experi- mental tank for both species at the onset of experi- ments using haemocytometer measurements (Table 1). Wotton (1973) found a significant posi- tive correlation between particle sizes in simuliid guts and in stream water, showing feeding to be unselective, and Chance (1970), despite finding selective feeding in the laboratory, found that in the field, feeding was un,selective. For both S. chiit- teri and S. nigritarse, the mean size of the cells ingested was, however, found to be smaller than the mean size of the cells available, indicating that both species selectively ingested a slightly smaller range of cells sizes than the mean range available (Tables 1 & 2). A 10-minute feeding period was found to be long enough to allow a measurable amount of inges- tion. Haemocytometric analysis of the gut contents of larvae which fed under different conditions of Table 1. Size ranges of algal cells available in tank prior to feeding experiment. Standard deviations are indicated in pa- renthesis (n = .2.S) Size range (pm) Mean (pm) S. clmlleri 200- 1.200 (260) 68.2 S. nigrilarse 2.20-1 100(220) 550 Table 2. Size ranges of algal cells ingested at each turbulence level (Reynold's number) for the haemocytometry experi- ments. Standard deviations indicated in parenthesis (n = US) Turbu- lence (Re) S. chiitteri S. tugritar.se Range ingested (pin) Mean (pin) Range ingested (pin) Mean (pm) 174 2.20-1224 (24.2) 474 2.20-816 (182) 447 481 204-1097 (292) 292 2.20-1020(229) 492 152 2.20-1020(192) 412 2.20-1071 (260) 442 1487 2.20-99.2 (196) 412 2.20-1020(222) 47.2 16.28 220-89.2 (201) 4.29 2.20-1097 (2.26) 202 2004 240-8.26 (188) 408 2.20-791 (182) 4.21 21 turbulence showed a significant increase in the number of cells ingested by S. chutteri at higher tur- bulences {Re values of 1638 & 2004). The opposite was seen for S. nigritarse, with a decrease in num- bers of cells ingested at higher turbulences (Fig. 6A). Comparing the two species using t-tests, highly significant differences at various turbulence levels (at Re values of 481, 752, 1487, 1638 and 2004: p<0.005) were observed, while only at Re values of 174 were cell counts not significantly different (p>0.2) (Fig. 6A). For the fluorometric experiments, chlorophyll a concentrations in the tank were similar at the onset of experiments with each species (35.9 pg/l in the ‘A chutteri’ tank; 36.3 pg/l in the ‘5. nigritarse’ tank). The results from these analyses showed a similar trend to the haemocytometer-measured experiments, but the distinction was less clear (Fig. 6B), and at a turbulence of 1487 Re, algal ingestion by S. nigritarse actually increased, as opposed to the dramatic drop observed in the haemocytometer counts. A significant difference (p<0.05) in chlorophyll a levels determined from the gut contents of each species was seen at each turbulence level, although this was greatest at the highest turbulence (/?e=2004). Although fan adductions ceased at turbulence values of 2004 Re for S. nigritarse (Fig. 6C), chlo- rophyll a was measured from the guts of the larvae subjected to this level of turbulence (Fig. 6B). It is possible that some algae were ingested during the 2-minute acclimation period prior to the onset of the experiments. This was not taken into consideration, and was assumed to be the same for all larvae, since they were subjected to similar conditions. Fan adduction observations showed that S. chut- teri had a higher adduction rate than S. nigri- tarse for all turbulence levels, but this became more pronounced at higher turbulences (Fig. 6C). Significant differences (p<0.005) in fan adduc- tions between the two species were observed at all but the lowest turbulences (Re=l74). In moder- ately turbulent conditions {Re=2004), or condi- tions approaching high turbulence (Re=752, 1487 & 1638), the rhythmic fan adductions of S. ni- gritarse, which characterized feeding at lower turbulences (Re=l74 & 481), became less regu- lar. The fans of many individuals remained closed throughout the runs at the highest turbu- lences (Re=2004), or flicked quickly without opening properly. Others seemed unable to adduct their fans, which remained continually open. In S. chutteri the adduction pattern was little affected by increasing turbulence. These observations suggest that under high turbulent conditions, larvae of S. nigritarse were unable to feed properly. Turbulence as Reynold’s number (Rc) Fig. 6. Results of feeding experiments: (a) Relative numbers of algal cells counted from the gut contents of simuliid larvae; n=10.. (b) Levels of chlorophyll a measured from the gut con- tents of simuliid larvae; n=30. (c) Fan adduction rates per min- ute for simuliid larvae; n=3. Larvae were subjected to flow conditions of increasing turbulence {Re values). Standard eiTors are indicated by vertical lines. o=S. chutteri; • S =nigritarse. DISCUSSION A potential problem with interpreting the data obtained during these experiments is that the larvae in the experimental tubes may be responding directly to changes in current velocity rather than to the changes in turbulence. However, since the calculation of Reynold’s number takes current ve- locity into account, the responses measured are in relation to Reynold’s number and therefore as- sumed to be due to changes in turbulence. Another limitation of this experimental work is that higher turbulences, above Re values of 2004, could not be attained with the apparatus used. This means that most of the studies fell into the flow range that is transitional between laminar and turbulent flows. In the field, the natural turbulence would be much higher than that measured in these experiments. Finally, an increased number of replicates would have provided better confidence in the statistical analysis, but the time limitations of the project did 22 Barber-James: Turbulence and Larval Feeding in Blackfly not allow this. Despite these shortcomings, the dif- ferent responses of the two species to changes in turbulence was clearly demonstrated using three different methods. It is recommended that further work be done on the feeding of these two species to corroborate these results. A number of conditions besides turbulence have changed in the Fish River since 1977, all of which may have influenced the invertebrate community, and which may have played a part in the observed change in species dominance. There have been changes in water chemistry (O'Keeffe & de Moor 1988) and the high turbidity and strong flow condi- tions associated with high numbers of S. chiitteri larvae (de Moor 1994) now prevail in the Great Fish River. Certain behavioural responses such as oviposition are known to affect the success of simuliids. de Moor et al. ( 1986) found that females of S. cluitteri scatter eggs in slower-flowing water upstream of rapids. The small larvae colonize the slower flowing reaches, while the more mature lar- vae drift and establish themselves in the rapids, unlike the coexisting species, S. nigritarse and S. culersi, which restrict themselves to the slower flowing reaches of the river. SimuUiim nigritarse lays its eggs in patches below the water surface on partly submerged stones (Chutter 1972). Ovi- position and larval establishment are therefore aspects of the life history of these species that are directly affected by flow. CONCLUSIONS Although other factors may innuence the inver- tebrate population structure, the results of this study do indicate that turbulence is one of the fac- tors favouring the change in simuliid species com- position in the Great Fish River. After quantifying the effects of increased turbulence on the feeding of S. cluitteri and S. nigritarse, it is not surprising that changes in the community structure have oc- curred, favouring the pest species S. cluitteri. ACKNOWLEDGEMENTS Thanks are due to Professor B.R. Allanson for initiating the project in my honours year, while 1 was a student at Rhodes University, and to Dr J.H. O'Keeffe and Dr F.C. de Moor for advice during the project. Special thanks also to Dr F.C. de Moor, Ms I.J. de Moor and Dr R.W. Palmer for constructive criticism of this manuscript, and to my late friend and colleague. Dr B.P. Boden, for some useful comments. This work was done at Rhodes University, Grahamstown, South Africa, as pai1 of a BSc (Hons) project. The Albany Museum is acknowledged for providing the facilities that en- abled me to produce this paper. REFERENCES Car, M. & de Moor, F.C. 1984. The response of Vaal River drift and benthos to Simiilium (Diptera: Neinatocera) con- trol using Bacillus tlniringiensis var israelensis. Oiuler.stepoort Journal of Veterinary Research ( Pretoria ) 51: 155-160. Chance, M.M., 1970. The functional morpliology of the mouthparts of blacktly larvae (Diptera: Simuliidae). Qnaestio- nes Entonwlogicae 6: 245-284. Chutter, F.M. 1968. On the ecology of the fauna of stones in the current in a South African river supporting a very large Sinniliiiin (Diptera) population. Journal of Applied Ecology 5: 55 I -561 . Chutter, F.M. 1972. Notes on the biology of South African Simuliidae particularly Sinnilinin (Eiisiimilinin) nigritarse Coquillett. Newsletter of the Limnological Society of South Africa 18: 1 0- 1 8. Ciborowski, J.J.H. & Craig, D.A. 1989. Factors inlluencing dispersion of larval black Hies (Diptera: Simuliidae): Effects of current velocity and food concentration. Canadian Journal of Eisheries and Aquatic Sciences 46: 1329- 1391. Coetzee, A.N. 1982. The population structure and dynamics of Sinudiiini in the Great Fish River. Unpublished MSc project report. Department of Zoology and Entomology, Rhodes University, Grahamstown: 37 pp. Craig, A, 1977. Mouthparts and feeding behaviour of Tahitian larval Simuliidae (Diptera: Neinatocera). Qiiaestiones Entonwlogicae 13: 195-218. Craig, A. & Chauce, M.M. 1982. Filter feeding in larvae of Simuliidae (Diptera: Culicomorpha): aspects of functional morphology and hydrodynamics. Canadian Journal of Zoology 60(4): 7 1 2-724. Currie, D.C & Craig, D.A. 1987. Feeding strategies of larval black Hies. In: Ke Chung Kim & Merritt, R.W. (eds.) Blackflies - Ecology, Population Management and Annotated World List. The Pennsylvania State University: 1 55- 170. de Moor, F.C. 1982a. A Community of Simulium species in the Vaal River near Warrenton. PhD thesis. University of the Witwatersrand, Johannesburg. 2 Vols: 317 pp. de Moor, F.C. 1982b. Determination of the number of instars and size variation in the larvae and pupae of Simulium cluitteri Lewis 1965 (Diptera: Simuliidae) and some possible bionomical implications. Canadian Journal of Zool- ogy 6^)-. 1374-1382. de Moor, F.C. 1989. Alternative life-history styles in Simuliidae (Insecta, Diptera). In: Bruton, M.N. (cd.) Alternative Life-History Styles of Animals. Kluwer Academic Press, Dordrecht: 293-316. Aiuutls oj the Eastern Cape Museums Vol. 3 (February 2004): 16-24. 23 de Moor, F.C. 1994. Aspects of the life history oi' Siimiliuin chiitteri and 5. hovis (Diptcra; Simuliidae) in relation to changing environmental conditions in South African rivers. Verlianclliingen dcr lutemationalen Vereiniguug fiir Theoretische imd Angewcmdte Limnologie 25: 1817-1821. de Moor, F.C., Chiitter, F.M. & de Moor, I.J. 1986. Drift behaviour and microhabitat selection in the preimaginal stages o\' Sinmlitmi chiitteri (Diptera Simuliidae). Hydrohiologia 133: 143-154. de Moor, F.C. & O'Keeffe, J.H. 1987. The Influence of Flow Modification on Invertebrate Connnnnities in the Great Fish River, Eastern Cape. Poster paper p.94, The Hydrological Sciences Symposium Rhodes University, Grahams- town. South Africa. Programme and Abstracts: 96 pp. Elsen, P., Quillevere, D. & Hebrard, G. 1978. Le transit intestinal chez les larves du complex Sinniliiiin dainnosuin (Diptera, Simuliidae) en Afrique de I'Ouesl. I. Inlluence du sexe et de I'espece. Annales de la Societe Beige de Medecine Tropicale (Antwerpen) 58: 209-217. Fredeen, F.J.H. 1964. Bacteria as food for blackfly larvae (Diptera: Simuliidae) in laboratory cultures and in natural streams. Canadian Journal of Zoology. 42: 527-548. Hart, D.D. & Latta, S.C. 1986. Determinants of ingestion rates in filter-feeding larval blaekllies (Diptera: Simulii- dae). Freshwater Biology 16: 1-14. Hart, D.D., Merz, R.A., Genovese, S.J. & Clark, B.D. 1991. Feeding postures of suspension-feeding larval black- llies: the conllicting demands of drag and food acquisition. Oecologia 85: 457-463. Kurtak, D.C. 1978. Efficiency of filter feeding of black lly larvae (Diptera: Simuliidae). Canadian Journal of Zool- ogy 56: \(m-\ 623. Ladle, M.D & Hansford, R.G. 1981. The feeding of the larvae of Siinnlinni aiisteni Edwards and Sinmlinin (Wilhelmia) spp. Hvdrobiologia 78: 17-24. Mackereth, F.J.H., Heron, J. & Tailing, J.F. 1978. Water analysis. Freshwater Biological Association Scientific Publication 36: 120 pp. Noble, R.G. 1970. Relation between tolerances and distribution of two species of Ephemeroptera. National Institute for Water Research Limnological Project Reports 3(3): 490 pp. O'Keeffe, J.H. & de Moor, F.C. 1988. Changes in the physico-chemistry and benthic invertebrates of the Great Fish River, South Africa, following an interbasin transfer of water. Regulated Rivers: Research and Management 2: 39-55. Palmer, R.W. & O’Keeffe, J.H. 1990. Downstream effects of a small impoundment on a turbid river. Archiv fiir Ilydrobiologie Supplement Band 119(4): 457-473. Palmer, R.W. & de Moor, F.C. 1998. Annotated records of blacklly (Diptera: Simuliidae) distribution in southern Africa. African Entomologv 6(2): 223-25 1 . Palmer, R.W. & Craig, D.W. 2000. An ecological classification of primary labial fans of filter- feeding black By (Diptera: Simuliidae) larvae. Canadian Journal of Zoology lU: 199-218. Sebrdder, P. 1980. On the nutritional biology of the larvae of Odagmia ornata Meigen (Diptcra: Simuliidae). I. The filtering activity as inlluenced by current-velocity, water-temperature and food-concentration. Archiv fiir Hydrobi- ologie Supplement Band 59( 1 ): 43-52. Schroder, P. 1988. Filter feeding activity of blacklly larvae (Diptera: Simuliidae) in relation to stream velocity and food supply. Archiv fiir Hydrobiologie Supplement Band 77(2): 161-182. Scott, K.M.F., Allanson, B.R. & Chiitter, F.M. 1972. Orange River Project Working group for ORP Hydrobiology of the Eish and Sundays Rivers. C.S.l.R. Research Report 306. Pretoria, South Africa. 61 pp. Smith, l.R. 1975. Turbulence in Lakes and Rivers. Ereshwater Biological Association Scientific Publication 29: 79 pp. Vogel, S. 1981. Life in Moving Eluids. The Physical Biology of Plow. Willard Grant Press: 352 pp. Wood, D.M. 1985. Biting Plies attacking Man and Livestock in Canada. Publication 1781/E. Agriculture and Agri- Food, Canada. Minister of Supply and Services Canada 1985. http://res2.agr.gc.Cci/ecorc/diptera/bfl I -dp I l_e.htm Wotton, R.S. 1973. The size of particles ingested by moorland stream blacklly larvae (Simuliidae) Oikos 29: 332-335. 24 Barher-Jennes: Turbulence and Iximd Teedinf> in Blackfly INSTRUCTIONS TO AUTHORS LAYOUT of recent issues should be examined for details of acceptable style and formal. MANUSCRIPTS should be submitted on disk in IBM format (MS Word or Wordperfecl) accompanied by three hard copies, with 1 Vi or 2 line spacing. Manuscripts may also be emailed in RTF. Typed manuscripts will not be accepted. Footnotes should be avoided as far as possible (see References below). All measure- ments should be metric. Systematic papers should follow the rules of the International Code of Zoological No- menclature or the International Code of Botanical Nomenclature as applicable. PAGE CHARGES of R65.00 ($ l().()0/£7.()0) per set page may be levied and are due once the author's proofs have been approved. STYLE AND MARKINGS. Only generic and specific names should be italicized (not underlined). Ail other marking up must be left to the Editor. Do not use any imbedded codes or styles for indentations and paragraphing. ILLUSTRATICINS should bear the author's name, title of the paper and illustration num- ber in pencil on the reverse side. Diagrams, line drawings and black and while photo- graphs are all treated as figures and should be termed as such and numbered in a single series. Colour photographs are treated as plates and special arrangements for these must be made with the Editor. Authors are responsible for the cost of the colour plates. Illustrations must be submitted in the original plus one photocopy. Diagrams and line drawings should be reducible to a size not exceeding 170 mm wide x 245 mm high, in- cluding caption. Black and white or good quality colour prints of high contrast and printed on glossy paper maybe submitted for black and white figures. These should be of final size, and not exceeding the dimensions above. Only good quality transparencies are accepted for colour plates. Scales should be given with illustrations or magnification indicated in the captions. A list of captions should be submitted on a separate sheet. TABLES should be submitted on separate sheets. Eoldouts should be avoided. REEERENCES should be listed alphabetically at the end of the paper and must be cited as the examples below. Note that periodical titles must not be abbreviated. Alexander, J.M. 1969. The indirect evidence for domestication. In: Ucko, P.J. & Dimbley, G. (eds) The Domestication of Plants and Animals, Duckworth, London: 123-129 Dc Winter, B. 1969. Plant taxonomy today. South African Journal of Science 65 (8): 229- 242. Jubb, R.A. 1 967. Freshwater Fishes of Southern Africa. HH Balkema, Cape Town: 248 pp. PROOES will, where practicable, be sent to authors for checking printer's errors. Adjust- ments of language, style and punctuation and any other amendments of the original manuscript, if made at this stage, will be charged to the aulhor(s). AUTHOR'S COPIES (25) are issued free of charge to the senior/author. Additional cop- ies will be supplied at cost plus cost of postage and handling. ' (.® * fV- ? ' ’i- ' ■ .- ' '' Grocotte Pulishers and Printers 046-6227222