Pi ar per WENQ J 5 % RX & YY 2 \e VAe\ th a oH RN S te es wi 3 Vip aa ee = NN EN SA re 2 SS = ‘AS HESS = ‘NosH» = = NOILALILSNI_ NVINOSHLINS S3I1YVUEIT LIBRARIES SMITHSONIAN INSTITUTION I ¥ NOILNLILSNI NOILNLILSN! LIBRARIES NOILALILSNI LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI 1uvudl NVINOSHLINS S3luvudiT | Saluvugi LIBRARIES SMITHSO NVINOSHLINS S3IYVYEIT LIBRARIES SMITHSO! wn ut oc = D » 2 A Z i es JiuvVUdIT_ LIBRARIES, SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHIINS SSIMVUGIT_ N NVINOSHLIWS jj NVINOSHLIWS \ S\ \S SMITHSONIAN NVINOSHLIWS NVINOSHLINS S31YVYGI7 LIBRARIES SMITHSONIAN INSTITUTION — LIBRARIES SMITHSONIAN ar) = a Y44y, = 4 = ty Yo fr [ ace: 7 “He c = Y fs = 5 a r ae = je) S = z JIYVUYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3IYVvVualT SZINVYSIT LIBRARIES SMITHSONIA Salyuvudi INSTITUTION INSTITUTION INSTITUTION ISTITUTION NOILALILSNI NYINOSHLINS S3IUVYAIT_ LIBRARIES SMITHSONIAN NVINOSHLINS S31YVYd!IT_ LIBRARIES Y Yjlu avs 4 NVINOSHLIWS SMITHSONIAN NVINOSHLIWS N ~ SMITHSONIAN GY LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS saluvygi WS" SMITHSONIAN aiuvudiy Sv, Sec Cn y z ~] ia) YY en SAN LNLILSNI NLILSNI RARIES ph eDe NOSnSS LILSNI RARIES RAR FES NLILSNI INOSRS SMITHSONIA NVINOSHLIV AN SMITHSONI NVINOSHLI SMITHSON| NVINOSHLI NVINOSHLIWS saluvuagl LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI_ NVINOSHLIWS 7p) S = i) 2 w 2 ey = my a ! « a ~ bee oc = ) < = < 2a < PA WW INSOE’ = fe e wc = : | = S 4 ay al N SS = ro) 23 5 2 re) ~'S os Be z a = 5 Z SMITHSONIAN _INSTITUTION NOILANLILSNI NVINOSHLINS S31Y¥vudiI ENE RAR ESE SMITHSONIA 4 ee | Oo . Z o a Cn a fh Li > E WEE EF LZ = = WY = ee : Yi yy rics -e ‘ = ke Ws Z 2 Z NOILNLILSNI NVINOSHLIW NVINOSHLINS S3INVUAIT LIBRARIES SMITHSONIAN SMITHSONIAN INSTITUTION NOILMLILSNI NVINOSHLINS Saluvagi7_LIBRARIES SMITHSONIAI a 4 NVINOSHLINS S3IYVustI ZY Zi NVINOSHLINS S31YVd SMITHSONIAN SMITHSONIAN Gp NVINOSHLIWS SMITHSONIAN saiuvugi7_LIBRARIES SMITHSONI. > SMITHSONIAN INSTITUTION NOILOALILSNI NVINOSHLIWS ? = 2) > ere Ww S w 4 Pi 27) ud ” SS i ” uw : . 2 Vd), 3 < =f .~ < a weg nt oy * o od he * ij 3 : 2 NS : a 4 | 2 5s S aie z Been NVINOSHLINS (SA1uVUGIT LIBRARI ES_ SMITHSONIAN” INSTITUTION | NOLLALILSNINVINOSHLIN = ° o = FS) ey, = Pe S Brie ak 0 A a ow — fp ao pam wo s " LN = x = 4, 2 = ra Lyg z Ya WS E > = Up fi > E > \ WS 0 WYra = Po) — Y e ary = Ps) KX = NO" 5 a to = n —_= wy " ~~ rm 2 m = a ASMITHSONIAN INSTITUTION | NOILNLILSNI NVINOSHLINS S31¥Vaa!7_ ‘ 3 a 2 we, < = y, = = 4 Zz 4 y UY Te 4 FA oH 2 3 Wh? j 3 & Z = 8,77 [Y & Z = Zz > = SS >” = > = X > 2 ” pr z ” z w ; z= “NVINOSHLINS S31YVYaIT LIBRARIES INSTITUTION NOILALILSNI_ NVINOSHLIV z WW z © Z ie Z a = 4 ~ v = 4 = We 5 > = e = = Re SS ye (ox < (ce (ex 4 WS = ce S oc = 3 = as re) a 5 2a re) Z a a 2 @F z mF Z i SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLIWS SAARI ES SMITHSON az E Aa = : Ks S o = 4 f; 1 = D WS - F) 5 (2 4 Gly * : NE : LY at = Ke te . E es = “LY NVINOSHLINS, SA1UVU EIT LIBRARI ES SMITHSONIAN _ INSTITUTION, NOILALILSNI_ NVINOSHII were =a v, = = pes . x a < 4G = = z =i Y) wz Xs Sj = a ONE! ae gl 3 a, A S SY = ie) py 28 UK O = 0 G, x E 2 Gy = SNS Zz = 2 “Wy * = > = STFS = > : = wn Pas 29) me w” Fas w W LOS ean Meee Nig 3 Le tee epi) | ENR YEN oO Sora 2 as } - , , ‘ f : ESTABLISH DATE Or lsoVe., FLEAS re Pee od a SNP DATE OF RECEIPT AND PLACE ee it ie IN YOUR LIBRARY. ; ‘MEMOIRS | - ‘OF THE a AMERICAN ENTOMOLOGICAL SOCIETY a i -NUMBER 34 PROCEEDINGS. poi es Can ‘cHIRONOMIDAE JACKSONVILLE, md ij e. ee JULY 2- 28, 1982 vb \-._ = Sane YT» toe Observations on and Descriptions of the Egg-Mass and Eggs of Buchonomyia thienemanni Fitt. (Diptera: Chironomidae). P. ASHE AND D.A. MURRAY Department of Zoology University College Dublin Belfield, Dublin 4, Ireland ABSTRACT. — The eggs, egg-mass and changes occurring throughout the development of the eggs of Buchonomyia thienemanni Fittkau are described and figured. Eggs were studied using the scanning electron microscope (SEM) and the light microscope. The most distinctive feature of the eggs is the presence of a large micropyle at the anterior end which is clearly visible under both the SEM and the light microscope. There are no other apparent pores and the remainder of the egg surface is smooth and structureless. The posterior part of the egg is rounded but the anterior end is slightly tapered and blunt. The egg-mass is tubular in shape, opaque, yellow in colour and contains approximately 320 eggs packed into the lumen. Development of the eggs from deposition to liberation of the 1st instar larvae takes about five and a half days. INTRODUCTION Adults of Buchonomyia thienemanni Fittkau were first recorded in Ireland from the River Flesk by Murray (1976) and since then attempts have been made to discover the immature stages. The pupa has recently been described (Murray and Ashe, 1981) but attempts to find the 4th instar larvae of this species in the river after several years searching have not been suc- cessful. Dr. H. Tichy (Ttibingen) suggested an alternative method of obtain- ing larvae by first capturing a fertile adult female and then provide suitable conditions for the female to lay eggs from which larvae should hatch. Male adults of B. thienemanni, which form distinctly recognisable swarms, were observed during August 1981 near the River Flesk in an attempt to obtain mated pairs falling from the swarm to the ground. On the 9th August a male and female adult of this species were captured in copulo. An egg-mass and eggs were subsequently obtained and are described here in detail. A brief ac- count of the development of the eggs is also given. The Ist instar larvae will be described in a subsequent paper. MATERIALS AND METHODS Techniques used to obtain eggs. — A male and female adult of B. thienemanni were captured in copulo falling from a swarm adjacent to the River Flesk, Killarney at 7:40 pm on the 9th August 1981. Mating, in the 3 MEM. AMER. ENT. SOC., 34 4 BUCHONOMYIA THIENEMANNI EGGS end to end position, lasted for 10 minutes. In the laboratory the female was placed in a 9.0 cm diameter petri dish containing river water. The lid of a smaller petri dish (diameter 5 cm) was placed on the water surface within the larger petri dish to provide a dry platform on which the female could rest. The petri dish was then left floating on the surface of water within a large tray in order to avoid major fluctuations in the temperature of the small quantity of water within the petri dish. The female was observed at half- hourly intervals throughout the night and the following day. At night between observations the female was kept in the dark. The female was deliberately placed on the water surface on three occasions during the night, at 9:00 pm shortly after capture, at 1:00 am and at 4:00 am and left on the surface for about two hours in each case in order to encourage it to lay an egg-mass. When on the dry platform the adult female was almost complete- ly inactive except for slight movement of the maxillary palps. On the 10th August at 1:30 pm the female was placed on the water surface to encourage it to lay and at 4:00 pm an egg-mass was laid. Immediately after laying the egg-mass the female became extremely active and flew around inside the petri dish whereas during the entire period since capture little or no activity was observed. The egg-mass was laid under water and attached to the bot- tom of the petri dish. Rearing the eggs. — The egg-mass was placed in a 9.0 cm diameter petri dish which was half full of river water. This petri dish was then placed float- ing inside a large tray containing water in order to avoid large fluctuations in temperature of the small quantity of water within the petri dish. The temperature of the water in the petri dish varied between 18.5°C and XV SAC, On the third day small clusters of an orange coloured fungus developed. Each day thereafter, until the eggs hatched, the egg-mass was placed in a clean petri dish and fresh water was added. This procedure ensured that the fungal colonies did not become very large, although they were clearly evi- dent by the following morning. Very few of the eggs were attacked by the fungus and only a few small colonies of the fungus had begun to grow on the egg-mass by the time most of the eggs had hatched. These precautions were necessary because of the known diverse flora of aquatic Phycomycetes which are parasitic on chironomid eggs causing high or even total mortality in egg-masses depending on the time of infection (Martin, 1981). SEM Technique. — Non-viable eggs and the larvae were preserved in 75% alcohol. For SEM work the eggs were gradually dehydrated to absolute alcohol. Attempts to clean surface particulate matter from the eggs by sonicating resulted in damage to the specimens so the SEM analysis was based on non-sonicated eggs. The eggs were dried in a Polaron Critical P. ASHE AND D.A. MURRAY Fic. 1. Egg-mass of B. thienemanni Fittkau. MEM. AMER. ENT. SOC., 34 6 BUCHONOMYIA THIENEMANNI EGGS Point Drier (Model E3000). A stainless steel boat, used to hold a flow- through timble (British-American Optical Company) containing the specimens, was first half filled with absolute alcohol so that the level of alcohol only reaches about half way up the timble. As a consequence of the small size of the eggs the timble was modified in the following way: a piece of teflon tubing with a filter paper at one end was inserted into the timble and a hole made in the timble lid. The eggs were pipetted into the teflon cap- sule within the timble, the excess alcohol passes through the timble because the porosity of the timble ensures an equal level of alcohol in the boat and timble. The level of alcohol in the boat was lowered so that the eggs, when dried, would be found near the bottom of the capsule. A small disk of filter paper was placed over the top of the teflon capsule to prevent the eggs escaping and the lid of the timble was then secured in position. After drying the eggs were removed from the timble by touching with a fine needle to which they readily adhered. After careful mounting on double sided sticky tape, which was affixed to the top of a stub, the eggs were sputter coated with gold using a Polaron SEM Coating Unit (Model E5100) and examined on a Jeol JSM 35C scanning electron microscope. OBSERVATONS AND DESCRIPTIONS OF THE EGG-MAss AND EGGS Description of the egg-mass. — The egg-mass of B. thienemanni is approx- imately 8.9 mm long, roughly tubular in shape and open at both ends (Fig. 1). It is yellow in colour, opaque and the entire surface is very viscous. The eggs are packed into the lumen of the egg-mass. A seam, through which some eggs could be clearly seen, runs along the entire length of the egg- mass. Approximately 320 eggs were contained within the egg-mass (this number was estimated by counting the number of non-viable eggs and the number of larvae that hatched). Of the 320 eggs which were present in the egg-mass 135 or 42% hatched into larvae and 185 or 58% of the eggs were not viable. Description of the egg. — The eggs of B. thienemanni are 230pu long and 92 wide (widest point). The posterior end of the egg is rounded but the anterior end is slightly tapered and blunt (Fig. 2). The most distinctive feature of the eggs is the presence of a large micropyle, 9.3 in diameter and 7.0u deep, at the anterior pole (Figs. 2, 3). There are no other apparent pores and the remainder of the egg surface is structureless and smooth (Fig. 4). The micropyle is clearly visible even under a binocular microscope. In- itially, after laying, the eggs are yellow and their contents and the develop- ing larva is clearly visible through the chorion. When the larvae hatch the P. ASHE AND D. A. MURRAY 7 Fic. 2. Egg of B. thienemanni Fittkau. Fic. 3. Micropyle at the anterior end of the egg of B. thienemanni Fittkau. chorion is clear and colourless. Some particulate matter is present on the surface (Figs. 3, 4) but attempts to clean the eggs by sonicating resulted in damage to the specimens. Development of the egg. — The development of the eggs of B. thienemanni was observed in a field laboratory using a low power binocular microscope which limited the amount of detail that could be seen. The eggs were not observed for more than a few minutes at a time for fear of damage resulting from the heat source in the microscope. Equipment and materials were not available at the time for the proper preparation of the eggs in various stages of development. The drawings of the development are based on sketches MEM. AMER. ENT. SOC., 34 8 BUCHONOMYIA THIENEMANNI EGGS Fic. 4. Side view of the egg of B. thienemanni Fittkau. drawn free-hand and the relative size and shape of stuctures may not be exact. The egg-mass was laid on the 10th August 1981 (Day 1) at 4:00 pm which is regarded as zero hour and the eggs hatched on the 16th August (Day 6) at 10:00 am or 138 hours later. Observations on the eggs are given day by day and the number of hours taken for the eggs to develop to the stages that are described and sketched are indicated. Day 1. (0 hour). The embryo at this stage appears undifferentiated and fills the entire space available within the egg (Fig. SA). (4 hours). The em- bryo skrinks back from the anterior and posterior ends leaving a clearly defined fluid-filled space at both ends of the egg (Fig. SB). Day 2. (21 hours). The egg has not changed shape but the embryo appears to undergo rapid cell proliferation. The fluid-filled spaces are not as clearly defined (Fig. SC) as in the previous stage. (29 hours). The dorsal surface of the eggs become slightly concave. A large area of the centre of the embryo becomes quite dense as well as a small area at the anterior end near the micropyle (Fig. SD). The only space visible within the egg is around the micropyle. Day 3. (45 hours). Little visible change is discernible (Fig. SE) since the previous observation. Day 4. (68 hours). The embryo now shows considerable differentiation and about 9-10 lines are visible on the ventral half of the egg which are the first indication of segmentation (Fig. 5F). Fluid-filled spaces are visible at P. ASHE AND D. A. MURRAY 9 E Fic. 5 A-K. Embryonic development in B. thienemanni Fittkau: A (0 hours); B (4 hours); C (21 hours); D (29 hours), lateral view; E (45 hours), lateral view; F (68 hours), lateral view; G, H (92 hours), lateral and dorsal view; I, J (116 hours), lateral and dorsal view; K (138 hours), lateral view. the anterior and posterior ends of the egg. The head region of the embryo is not yet discernible. Day 5. (92 Hours). In lateral view the head region is clearly distinguished from the rest of the body (Fig. SG). From the ventral aspect the head region cannot be clearly distinguished from the body although the pharangeal region and the intestine are visible (Fig. 5H). The spaces have decreased in size but are visible at the anterior and posterior ends. Day 6. (116 hours). The head is much more clearly defined and the eye- spots have developed (Fig. 5I, J). The shape of the eggs in lateral view have changed considerably but if examined in dorsal view they appear not to have changed since Day 1. The pharangeal region is clearly visible but the intestine is no longer visible. MEM. AMER. ENT. SOC., 34 10 BUCHONOMYIA THIENEMANNI EGGS Day 7. (138 hours). When the petri dish containing the egg-mass was ex- amined some 15 larvae had hatched. The larvae inside the eggs which had not yet hatched could be clearly seen moving the anterior parapods backwards against the chorion. Due to the consistency of the abdominal region and because the larvae were coiled within the eggs it was not possible to distinguish and draw the structures and segments of this region (Fig. SK). Prior to hatching the larvae of B. thienemanni begin to swell presumably by swallowing amniotic fluid. The larvae fill almost the entire egg except for a few fluid filled spaces at the anterior end. DISCUSSION The most detailed and comprehensive study of chironomid eggs and egg- masses to date is by Munsterhjelm (1920). The egg stage is a very much neglected part of the life cycle of chironomids and very little information on even the most basic aspects such as duration of development, hatching mechanism, embryology, etc. have been published. Most descriptions of chironomid eggs are very brief and usually little or no information other than size of the eggs is given. The earliest detailed account of the structure and development of a chironomid egg is given by Miall and Hammond (1900) for Chironomus dorsalis Auct. Photographs of various stages of development in the eggs of several species of Chironomus are presented in Strenzke (1950). The very large and distinct micropyle that is present in B. thienemanni is probably a very plesiomorphic feature and appears to be unique among the Chironomidae whose egg stage has so far been described. According to Miall and Hammond (1900) the micropyle in C. dorsalis is minute although its size in relation to the whole egg is not figured. In most descriptions of chironomid eggs no mention is made regarding the micropyle perhaps because of difficulties in observing it. Some eggs of Psectrocladius (Allop- sectrocladius) obvius (Walker) were examined in the present study but the micropyle could not be seen and it may be necessary to examine these eggs under the SEM. The most important factor affecting the duration of the egg stage is temperature. According to Chapman (1975) as temperature increases the time required for development decreases and conversely as temperature decreases the time required for development increases in a more or less linear manner except at the extreme ends of the range for development. In chironomids this linear relationship between temperature and duration of development has been demonstrated for the eggs of a Smittia species (Kalthoff, 1971). In Clunio tsushimensis Tokunaga the eggs hatch 80-90 hours after oviposition at 25°C (Oka and Hashimoto, 1959) but may take P. ASHE AND D.A. MURRAY 11 10-12 days to hatch at 15°C. B. thienemanni and P.(A.) obvius under similar conditions take about 5.5 and 3.5 days respectively to hatch at a. temperature of about 20°C. In Chironomus plumosus (Linn.) eggs hatched in 3 days at 24°C and in 14 days at 9° but eggs incubated at 8°C did not hatch and eventually decomposed (Hilsenhoff, 1966). For the majority of chironomids a temperature of between 9-25°C is required for development. The egg-masses of many chironomid species have been described and it is clear that there is considerable variation in size, shape and form (Munsterh- jelm, 1920; Branch, 1928, 1931; Morrow et al, 1968). A key to the egg- masses of some North American chironomid species is given in Morrow et al (1968). The eggs of most species are imbedded in a gelatinous material but the eggs of Paraclunio and Telmatogeton of the subfamily Telmatogetoninae are laid singly and not in a gelatinous mass (Saunders, 1928; Tokunaga, 1935). According to Hinton (1981) the eggs of Paraclunio are inserted into filamentous algae which explains why a gelatinous mass is not produced. In some terrestrial Orthocladiinae the eggs are also laid singly (Hinton, 1981). The egg-mass may contain only one egg or as many as a thousand and within the gelatinous mass specialised structures such as suspensory stalks, anchor chords, etc. may be found (Morrow et al, 1968). According to Hinton (1981) the function of the gelatinous material is to delay desiccation of the eggs when they are exposed above water. According to Chapman (1975) ‘‘Most insects force their way out of the egg, first swallowing the amniotic fluid so as to increase their volume and then pumping blood forwards by contractions of the abdomen so that the head exerts pressure against the shell’’. The chorion may split in an irregular manner or may split along a line of weakness or cuticular structures, termed egg bursters, usually present on the head, may be used. The drinking movements observed in Tanytarsus neoflavellus Malloch (sub Calopsectra neoflavellus) by Davis (1966a, b) probably refer to the up- take of the amniotic fluid as described by Chapman (1975). Before hatch- ing, the larvae of B. thienemanni are coiled within the chorion as is the case in Pseudosmittia gracilis Goetghebuer (Thienemann and Strenzke, 1940) and in Chironomus riparius Meigen (sub C. thummi thummi Kieffer) (Strenzke, 1959). In T. neoflavellus the larvae are not coiled prior to hatch- ing but lie straight within the chorion (Davis, 1966a). The larvae of B. thienemanni were observed moving the anterior parapods backwards and forwards against the chorion which may be an attempt by the larva to weaken the chorion. These movements of the anterior parapods were also noticed in T. neoflavellus by Davis (1966a:199) but he apparently did not regard this as being of any importance. Davis (1966a:200) states that the rupture of the egg in 7. neoflavellus was never accompanied by movement MEM. AMER. ENT. SOC., 34 12 BUCHONOMYIA THIENEMANNI EGGS of the larva and was initiated by internal pressure. Internal pressure alone does not explain why the eggs of 7. neoflavellus n 59 cases out of 60 burst on the ventral side of the egg just posterior of the anterior papapods. In all cases where larvae of B. thienemanni were seen breaking out of the egg the chorion split on the ventral side of the egg close to the anterior parapods. It seems plausible that in B. thienemanni and T. neoflavellus that the larva weakens the chorion with the anterior parapods and then internal pressure causes the eggs to rupture at this point. Alternative methods of hatching from the egg by chironomid larvae have been reported. Davis (1966a:201) states that in an unidentified chironomid larva, following a period of drinking movements, the chorion is split dorsal- ly towards the posterior border of the head, above a dark pointed structure which acted as an egg burster. Thienemann (1954) reports that chironomid larvae broke the chorion by using the mandibles. The description of the egg-mass of B. thienemanni given here should readily distinguish it from all other described chironomid egg-masses. The eggs and egg-masses of representatives of the subfamilies Podonominae and Aphroteniinae have apparently not yet been described. The form and con- struction of the egg-mass in different chironomid subfamilies may be a useful phylogenetic character. ACKNOWLEDGEMENTS A Praeger Grant from the Royal Irish Academy awarded to P. Ashe to defray field expenses during July and August 1981 is gratefully appreciated. Thanks are also due to the Office of Public Works and the staff at the Bourne Vincent National Park, Killarney, for permission to use laboratory facilities. We wish to thank: Dr. H. Tichy, Max-Plank-Institut fiir Biologie, Tiibingen, West Germany who suggested capturing mated adult females to obtain egg-masses; Mr. Barry Cregg, SEM Unit, Agricultural Science, U.C.D. for advice in preparing specimens for SEM and Mr. Robert French, Technician, Zoology Department, U.C.D. who carried out the scanning electron microscopy and prepared the photographic plates. REFERENCES BRANCH, E.H. 1928. Description and identification of some chironomid egg masses. Ann. ent. Soc. Am. 21:566-570. 1931. Identification of chironomid egg masses II. Trans. Kans. Acad. Sci. 34:151-157. P. ASHE AND D.A. MURRAY 13 CHAPMAN, R.F. 1975. The insects structure and function. The English University Press Ltd. London, 819 pp. Davis, C.C. 1966a. A study of the hatching process in aquatic invertebrates. XVI. Events of eclosion in Calopsectra neoflavellus Malloch (Diptera, Tendipedidae). XVII. Hatching in Argulus megalops Smith (Crustacea, Branchiura). Hydrobiologia 27:196-207. 1966b. Hatching processes in the eggs of aquatic invertebrates. Verh. int. Verein. theor. angew. Limnol. 16:1685-1689. HiseNHorF, W.L. 1968. The biology of Chironomus plumosus (Diptera: Chironomidae) in Lake Winnebago, Wisconsin. Ann. ent. Soc. Am. 59:465-473. Hinton, H.E. 1981. Biology of insect eggs, Vol. II, 475-778 pp. Pergamon Press Ltd, Oxford. KalttHorF, K. 1971. Temperature effects on embryogenesis in Smittia spec. (Diptera, Chironomidae): Qjo-values of normal development and frequency of ‘‘double abdomens”’ after UV-irradiation. Wilhelm Roux Arch. EntwMech. Org. 168:85-96. Martin, W.W. 1981. Couchia circumplexa, a water mold parasitic in midge eggs. My- cologia 73:1143-1157. MIALL, L.C. AND A.R. HaMMonpD. 1900. The structure and life-history of the harlequin fly (Chironomus). Clarendon Press, Oxford, 196 pp. Morrow, J.A., J.L. BATH AND L.D. ANDERSON. 1968. Descriptions and key to egg masses of some aquatic midges in Southern California (Diptera: Chironomidae). Calif. Vector Views 15:99-108. MUNSTERHJELM, G. 1920. Om Chironomidernas Agglaggning och Aggrupper. Acta Soc. Fauna Flora fenn. 47:1-174. Murray, D.A. 1976. Buchonomyia thienemanni Fittkau (Diptera, Chironomidae), a rare and unusual species recorded from Killarney, Ireland. Entomologist’s Gaz. 27:179-180. Murray, D.A., AND P. AsHE. 1981. A description of the pupa of Buchonomyia thienemanni Fittkau, with notes on its ecology and on the phylogenetic position of the subfamily Buchonomyiinae. Spixiana 4:55-68. Oxa, H. anp H. HasnHimoro. 1959. Lunare Periodizitét in der Fortpflanzung einer spezifischen Art von Clunio. Biol. Zbl. 78:545-559. SAUNDERS, L.G. 1928. Some marine insects of the Pacific coast of Canada. Ann. ent. Soc. Am. 21:521-545. STRENZKE, K. 1959. Revision der Gattung Chironomus Meig. I. Die Imagines von 15 norddeutschen Arten und Unterarten. Arch. Hydrobiol. 56:1-42. THIENEMANN, A. 1954. Chironomus. Leben, Verbreitung und wirtschaftliche Bedeutung der Chironomiden. Binnengewdasser 20:1-834. THIENEMAN, A. AND K. STRENZKE. 1940. Terrestrische Chironomiden III-IV: Zwei parthenogenetische Formen. Zool. Anz. 132:24-40. ToxunaGa, M. 1935. Chironomidae from Japan (Diptera). IV. The early stages of a marine midge, Telmatogeton japonicus Tokunaga. Philipp. J. Sci. 57:491-511. MEM. AMER. ENT. SOC., 34 A Biological Appraisal Of Water Quality In The Ohio River Using Macroinvertebrate Communities, With Special Emphasis On The Chironomidae Davip C. BECKETT! Department of Biological Sciences University of Cincinnati Cincinnati, OH 45221, USA and JOHN L. KEYES Ohio River Valley Water Sanitation Commission Suite 900, 411 Walnut St. Cincinnati, OH 45202, USA ABSTRACT. — Macroinvertebrate communities were sampled using multiplate samplers at 15 locations on the Ohio R. and 7 tributary stations in late summer of 1978. The depauperate fauna in the upper Ohio R. pools supports the contention that aquatic habitats subjected to toxic wastes generally possess communities characterized by low numbers of taxa and low den- sities. In the upper Ohio R. the number of taxa collected per sampling station was inversely correlated with the number of industrial dischargers over the segment of river preceding each of the sampling sites. Interestingly, a mussel survey over the same area in 1979 produced a ranking of the upriver pools in terms of existent mussel species which coincided perfectly with our ranking of numbers of macroinvertebrate taxa collected on multiplates. Species richness was especially useful in assessing toxic effects, while evenness of distribution of individuals among species was a useful parameter in discerning organic waste input. Chironomid species such as Cricotopus bicinctus, C. intersectus gr., and Dicrotendipes ner- vosus Type II were very important in assessing the nature and severity of pollutional impacts. Historically, despite indications of improvement in the Kanawha R., water quality problems apparent in a mid-1960’s macroinvertebrate study of the Ohio River Valley still persist. INTRODUCTION Early studies of pollutional effects on macroinvertebrate distribution in North America focused on a large-river system (the Illinois R.) (Forbes and Richardson 1913; Richardson 1928). Since that time the majority of North ‘Present address: Environmental Laboratory/AHG, U.S. Army Engineer Waterways Ex- periment Station, P.O. Box 631, Vicksburg, MS 39180, USA. 15 MEM. AMER. ENT. SOC., 34 16 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY American freshwater pollutional work encompassing macroinvertebrate community response has dealt with lake sytems (Carr and Hiltunen 1965; Brinkhurst 1969; Mozley and Howmiller 1977; Saether 1979) or small stream or small river systems. For example, Gaufin and Tarzwell’s classical papers (1952, 1955, 1956), using the indicator species concept to assess pollutional impact, had a small midwestern stream (Lytle Creek) as a study site; the use of the highly popular Shannon-Wiener diversity index was developed by Wilhm (1970) and Wilhm and Dorris (1968) from stream or small river work; a third means of pollutional appraisal, the biotic index (Chutter 1972), has been used most extensively in North America by Hilsenhoff (1977). Hilsenhoff’s biotic index was developed, and works most effectively, in streams or small rivers possessing a definite riffle com- ponent. The scarcity of large-river pollutional work is due largely to the physical difficulties involved in sampling large rivers and to the inadequate develop- ment of taxonomic treatments dealing with large-river macroinvertebrates. Mozley (1979) recently pointed out the lack of information concerning the larval Chironomidae (Diptera) of large rivers. The larval chironomids are an extremely important component of aquatic macroinvertebrate com- munities; the number of chironomid species often makes up at least 50% of the total macroinvertebrate taxa present in aquatic habitats (Coffman 1978). Fortunately, recent taxonomic treatments of the larval Chironomidae (Oliver et al. 1978; Simpson and Bode 1980) have vastly im- proved the state of large-river larval chironomid taxonomy. In addition, re- cent appraisals of pollution in large-river systems using macroinvertebrates collected on artificial substrates (Beckett 1978, Simpson 1980a, b, c, d) have shown promise in elucidating macroinvertebrate response to toxic and organic wastes in large rivers. This study investigated macroinvertebrate responses to pollutional in- fluences in the Ohio R., a large navigable river in the east-central United States, and some of its principal tributaries in the highly industralized Ohio R. Valley. Results are compared with: 1) fairly extensive chemical sampling performed by the Ohio River Valley Water Sanitation Commission (ORSANCO) over the macroinvertebrate study area; 2) a mussel survey (Taylor 1980) over the upper Ohio R.; 3) a mid 1960’s macroinvertebrate study of the Ohio R. (Mason et al. 1971); and 4) other pollutional studies, including Simpson and Bode’s (1980) assessment of chironomid response to pollution in large rivers in New York. Questions which are addressed in this study include: will large rivers which are subjected to a complex of pollu- tional inputs register an interpretable response; and if so, how will this response be expressed at the community and population levels? D. C. BECKETT AND J. L. KEYES 17 Z z, N.Y Sa pas NY hy Kir eave | - | PA | | a | IND. : | OHIO Dy? | | Spe Pittsb ILL. | | oT | | ig 47 et oats a aaa | | ras of t. og eine 6 Is NS at) W. VA . {) iS R——™ NS \ of) 9 "8 £ y, Ge 20 13 Wo o% Ig I AW ouisv. pie 12 AO el KY. pS aed 7 VA 22 Lo Fic. 1. Map of 1978 Ohio River Valley macroinvertebrate sampling stations. Sampling sta- tions indicated by solid circles; numbers near the circles are station numbers. Triangles indicate the location of dams on the Ohio and Kanawha Rivers while open squares are the large cities of Pittsburgh, Cincinnati, and Louisville. METHODS Study Area — The Ohio R. Basin drains a total of 422,170 square km (ex- cluding the Tennessee R. drainage) and includes portions of 11 states. Its primary river, the Ohio, has its origin in Pittsburgh, Pa. at the confluence of the Monogahela and Allegheny Rivers and flows 1578 river km in a southwesterly direction, joining the Mississippi R. at Cairo, Ill. With the ex- ception of the Mississippi, the Ohio R. carries the largest amount of freight of any waterway in the United States — 148 million tons in 1976 (Ohio River Basin Commission 1978). A system of 20 locks and dams over the length of the river maintains a navigational channel for year-around transport. At higher discharges the dams do not impede river flow and the river’s current velocity is close to that of natural conditions. At lower flows, however, the dams transform the river into a continuous series of ‘‘pools”’ with current much reduced from natural conditions (United States Army Corps of Engineers 1980). Because of river turbulence thermal stratification MEM. AMER. ENT. SOC., 34 — co BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY 65 60 RIVER DISCHARGE 50 40 RELATIVE 30 20 DISCHARGERS ~~ Pittsburgh — Cincinnati — Louisville NO. OF INDUSTRIAL / 100 300 500 700 900 1100 1300 1500 RIVER KM Fic. 2. Number of industrial dischargers per river discharge for segments of the Ohio R. preceding each locks and dam macroinvertebrate sampling site. Segment here is that length of river between sampling sites. Number of dischargers per segment was divided by a “‘dilution factor’’ which was equal to the mean river discharge in Sept. 1978 at the respective sampling site (see Table 1) divided by the mean river discharge in Sept. 1978 at the Ohio R. origin (Pitts- burgh). The locations of Pittsburgh, Cincinnati, and Louisville are indicated on the abscissa. does not occur in the river, and water temperatures are identical above and below the dams. The Ohio R. is subjected to a complex of discharged effluents as a result of its location in the highly industrialized east-central United States. Ap- proximately 200 industrial dischargers release their effluents directly into the Ohio R. Many of these dischargers have multiple discharges. Many other industries (in addition to the 200) send their effluents to municipal sewage treatment plants where the effluents receive primary or secondary treatment and are then released into the river. Approximately 115 municipal sewage treatment plants discharge effluents into the Ohio. Macroinvertebrate Sampling. — Macroinvertebrate communities were sampled in late summer of 1978 using Hester-Dendy multiplate samplers at 15 locations on the Ohio R. and seven tributary stations (Fig. 1 and Table D. C. BECKETT AND J. L. KEYES 19 1). Artificial substrates are of great utility in comparative pollutional studies since they present a uniform substrate for colonization at each sampling station (United States Environmental Protection Agency 1973; Beak et al. 1973). The multiplate samplers were placed in their respective locations in mid-August and collected in late Sept., giving a six-week col- onization period as suggested by the United States Environmental Protec- tion Agency (1973). Samplers were suspended ca. 1 m below the surface. Each sampler consisted of 14 masonite plates separated by a varying number of masonite spacers (spacing varied within the sampler — all samplers were of identical configuration and spacing). Varied spacing was employed since many macroinvertebrates demonstrate a preference for cer- tain microhabitats produced by variable spacings (Mason et al. 1973). Each of the 14 plates per sampler measured 65 mm x 65 mm with a thickness of 3.0 mm. After collection macroinvertebrate samplers were placed in plastic con- tainers and ethyl alcohol was added as a preservative. In the laboratory samplers were disassembled and the plates ‘‘cleaned’’ of macroinvertebrates with a soft-bristled brush. Macroinvertebrates were amassed by pouring and rinsing the slurry through a U.S. Standard no. 60 sieve. Samples were then stained with rose bengal (Mason and Yevich 1967) to facilitate the sort- ing of organisms from the organic detritus which had accumulated on the plates. Larval chironomids were prepared for identification using the pro- cedure of Beckett and Lewis (1982). Three samplers were collected from all sites with the exception of the Ohio R. at Addison (four samplers), the Kanawha R. at London Dam (four samplers) and the Ohio R. at Evansville, Ind. (two samplers). Since species diversity and number of taxa are often dependent on sampling effort (Sanders 1968; Simberloff 1972) three of the four recovered samplers were picked at random at four-sampler sites to report chironomid and total taxa levels, therefore giving a uniform number of three samplers/station. I em- pirically determined that two samplers had a mean of 80% of the total number of species of three sampler collections. The number of taxa col- lected at Evansville (two samplers recovered) was therefore adjusted (for both total taxa and number of chironomid species) to an approximate three- sampler level. Comparisons of number of taxa/station for the Ohio R. sampling sites were done separately for each of the two ‘“‘types’’ of sampling stations: those at dams (always located on the upstream, slow-water side of the dam) and those stations not near dams. Beckett and Miller (1982) showed large differences in macroinvertebrate distributions between above-dam and below-dam sites in the Ohio R., thereby necessitating these two separate comparisons. MEM. AMER. ENT. SOC., 34 20 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY TABLE 1. September 1978 Ohio R. Valley macroinvertebrate sampling sites. Discharge data are the mean discharges for Sept. 1978 at the respective sampling locations. Station Station River River Discharge No. Mile Km (m?/sec) O.R.' at South Heights, Pa. 1 15.2 24.5 294.5 O.R. near East Liverpool, Oh. 2 43.0 69.2 339.8 O.R. at Pike Island Dam 3 84.2 135.5 351.2 O.R. at Willow Island Dam 4 161.8 260.3 360.0 O.R. at Belleville Dam 5 203.9 328.1 490.0 O.R. at Kyger Cr. 6 260.0 418.3 490.0 O.R. at Greenup Dam 7 341.0 548.7 702.3 O.R. at Meldahl Dam 8 436.2 701.8 784.5 O.R. at Markland Dam 9 $31.5 855.2 1005.4 O.R. at McAlpine Dam 10 606.8 976.3 1155.5 O.R. at Mill Cr. 11 625.9 1007.1 1164.0 O.R. at Cannelton Dam 12 720.7 1160.0 1246.1 O.R. at Evansville, Ind. 13 791.5 1273.5 1382.0 O.R. at Uniontown Dam 14 846.0 1361.2 1387.7 O.R. at Joppa, Ill. 15 952.3 1532.2 3260.0 Monongahela R. at Pittsburgh, Pa. 16 4.5* U2 133.1 Beaver R. at Beaver Falls, Pa. 17 5.3* 8.5* 37.7 Muskingum R. near Lock & Dam #2 18 5.8* 9.3* 93.5 Kanawha R. at London Dam 19 82.8* 133.2* NA Kanawha R. at Winfield Dam 20 31.1* 50.0* 138.8 Green R. near Seebree, Ky. 21 41.3* 66.5* 104.8 Tennessee R. near Paducah, Ky. 22 6.0* 9.7* NA ‘O.R. = Ohio River * = distance from confluence with Ohio R. NA = data not available RESULTS AND DISCUSSION Macroinvertebrates in the highly industrialized areas of the upper Ohio R. below Pittsburgh and nearby Steubenville and Wheeling are subjected to the effluents of a large number of industrial discharges (Fig. 2). Cyanide and phenolics, both waste products of industrial processes (Hach 1975; ORSANCO 1980), showed proportionately higher percentages of elevated levels in the upriver areas (Fig. 3). Although cyanide and phenolics may not have affected macroinvertebrates directly they served as indicators of in- dustrial effluent impact. In addition, this area of river is near its origin and river discharge is much less than downstream stations (see Table 1), with the river’s assimilative capacity thereby lessened. The depauperate fauna collected at sampling stations 3 and 4 (Fig. 4) sup- ports the contention that aquatic habitats subjected to toxic wastes general- D. C. BECKETT AND J. L. KEYES 21 e—_e PHENOLICS > | ug/I CYANIDE > 5 ug/I CYANIDE > 5ugQ/! PHENOLICS > lug/I % OF SAMPLES WITH: 200 400 600 800 1000 1200 1400 RIVER KM Fic. 3. % of 1978-1979 water samples collected at ORSANCO’s Ohio River monitoring stations in which cyanide exceeded levels of 5 »g/1. Also shown is the % of water samples in which phenolics exceeded 1 ng/1. Lines connecting points are for visual continuity and do not indicate percentages at intermediate locations. Data source is ORSANCO (1980). ly possess communities characterized by low numbers of taxa and low den- sities (United States Environmental Protection Agency 1973; Hocutt 1975; Lenat et al. 1980). Station 3, the locks and dam sampling station which had the greatest number of industrial dischargers over the preceding segment of the river/river discharge (Fig. 2) (and the highest percentage of elevated cyanide and phenolic levels — Fig. 3) possessed both the fewest total number of taxa and the fewest number of animals per sampler (Fig. 4). Sta- tion 4, another upriver station subjected to a relatively large number of dischargers (Fig. 2), also had low numbers of taxa in low densities (Fig. 4). A striking change in the total number of species and number of individuals per sampler was apparent for the macroinvertebrate collection at station 5 vs. those at stations 3 and 4 (Fig. 4). The conspicuous increase in both the total number of species and the number of individuals per sampler at station 5 is probably due to a marked reduction downriver in the number of in- dustrial dischargers (Fig. 2) and the dilution of upstream waters by the addi- tion of water from the Muskingum and Little Kanawha Rivers. Average river discharge for Sept. 1978 increased by only 2% between stations 3 and 4 (Table 1). Discharge increased by 36% between stations 4 and 5, however, MEM. AMER. ENT. SOC., 34 22 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY 30 25 q = 20 4 Ee uo °o #615 ing WwW a = 5 10 Zz 200 400 600 800 1000 1200 1400 RIVER KM (L&D STATIONS) 50 100 150 20) MACROINVERTEBRATES / SAMPLER 250 OF 300 NUMBER Fic. 4. Number of taxa/sampling site and mean number of individuals/ sampler for Ohio R. locks and dam macroinvertebrate sampling stations. Length of the solid bar denotes the total number of chironomid species collected and the mean number of chironomid individuals per sampler at each site; length of the entire bar denotes the total number of species collected and the mean number of individuals per sampler for all macroinvertebrate taxa at each site. Number above each bar indicates the station number. D. C. BECKETT AND J. L. KEYES 23 due chiefly to the confluence of the two tributaries with the Ohio. The relatively high numbers of taxa and individuals collected at station 5 in- dicate that toxicity problems have greatly diminished at this point. A 1979 study of the mussels of the upper Ohio R. (Taylor 1980) presents an interesting comparison with this 1978 macroinvertebrate — artificial substrate study. Although mussels are ‘‘macroinvertebrates’’, they are un- common on suspended artificial substrates; Taylor therefore studied the same system, over a different year, using a completely different subset of aquatic organisms. He found the Belleville Pool (Belleville Dam is our sta- tion 5 — Figs. 1 and 4) “‘to be the most productive of . . . the pools in total numbers and diversity of [mussel] species’’ (23 species), while the Greenup Pool (Greenup Dam — our station 7) ‘‘had fairly active mussel populations’’ (14 species). He found the upstream Willow Island Pool (Willow Island Dam — our station 4) to have ‘‘only a very scanty mussel population’”’ (3 species), while the Pike Island Pool (Pike Island Dam — our station 3) is ‘‘essentially devoid of mussel life’’ (no species). Taylor’s rank- ing of the upriver pools in terms of number of mussel species present coin- cides perfectly with our ranking of numbers of macroinvertebrate taxa col- lected at the upstream sampling sites (see Fig. 4). This shows an encourag- ingly (scientifically) similar response to environmental conditions has occur- red within both communities. Interestingly, the ranking of number of in- dustrial dischargers/river discharge (Fig. 2) is the exact converse of the ranking of the number of macroinvertebrate or mussel species collected, i.e., the upriver sampling station having the highest number of species (sta- tion 5 — Fig. 4) had the lowest number of industrial dischargers, the sampl- ing station having the second highest number of taxa (#7) had the second lowest number of dischargers, etc. This inverse correlation suggests that in- dustrial dischargers are exerting a strong influence on macroinvertebrate communities in the upper Ohio. Downstream sampling stations’ total taxa levels did not drop below that of the far upriver stations in either the locks and dam comparison (Fig. 4) or for the non-locks and dam stations (Fig. 5), indicating that the toxicity problems apparent upstream did not recur with such severity for the re- mainder of the river. The largest numbers of taxa collected on the artificial substrates for both Ohio R. comparisons (Figs. 4 and 5) were at the stations furthest downstream. This is probably due to the reduction in both in- dustrial (Figs. 2 and 3) and organic waste inputs (Fig. 6) and the increased size (and therefore increased assimilative capacity) of the river (Table 1). Water quality of the Ohio R. is also improved by the addition of water from downstream tributaries such as the Green and Tennessee Rivers (Table 3). MEM. AMER. ENT. SOC., 34 24 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY 30 25 I 20 OF TAXA a NUMBER 200 400 600 800 1000 1200 [oXo} RIVER KM Fic. 5. Number of macroinvertebrate taxa collected at Ohio R. sampling stations which were not located at dams. Length of the solid bar denotes the total number of chironomid taxa collected at each site while the entire bar refers to the total number of all macroinvertebrates taxa collected per site. Number above each bar indicates the station number. In addition to industrial inputs, 115 municipal sewage treatment plants discharge organic wastes into the Ohio R. Figure 6 shows that the greatest organic inputs, based on 5-day BODs, are from the large metropolitan areas of Pittsburgh, Cincinnati, and Louisville, with Cincinnati having the greatest impact. Testing for fecal coliforms in 1978-1979 showed that the greatest percentages of high coliform levels occurred below Cincinnati and Louisville (Fig. 7). Causes of the high levels of fecal coliform bacteria in- clude inadequately treated sewage, combined sewer overflows, and urban runoff (ORSANCO 1980). In Sept. 1978 (the month of the macroinverte- brate collections) low dissolved oxygen levels reflected this organic input with dissolved oxygen frequently observed below 5.0 mg/1 from river km 788.4 (just below Cincinnati) to Cannelton Dam at river km 1160 (ca. 200 river km below Louisville) (ORSANCO 1980). The most frequent violations occurred at Markland Dam (station 9 — ca. 100 river km below Cincinnati) at which Dicrotendipes nervosus Type II (see Simpson and Bode 1980) was the most abundant chironomid larvae collected (Table 2). Simpson and D. C. BECKETT AND J. L. KEYES 25 Bode (1980) state that ‘‘this is often the dominant species in rivers contain- ing both sewage and toxic wastes, and is usually not found in relatively clean waters.’’ In a study of the Great Miami River system (Beckett 1978) D. nervosus Type II was almost completely absent from both relatively un- polluted areas and an area in the plume of a chlorinated effluent from a large sewage treatment plant. However, in downstream areas subjected to both toxic and organic wastes D. nervosus Type II was the dominant species. The United States Environmental Protection Agency also found this species to be a dominant on artificial substrates placed in the lower Kanawha R. in the late 1960’s and early 1970’s (P.A. Lewis, personal com- munication). At that time the lower Kanawha R. was subjected to heavy toxic and organic waste loadings (Mason et al. 1971). The results of these Ohio R. Valley studies therefore agree with those of Simpson and Bode’s (1980) in regard to the occurrence of this species. A comparison of numbers of taxa collected at the various Ohio R. tributary sampling sites (Fig. 8) shows the upstream tributaries near Pitts- burgh, the Monongahela and Beaver Rivers, had fewer numbers of taxa col- lected than the other tributary stations. Macroinvertebrate results from the Beaver R. were of particular interest since this river is degraded by pollution along its entire length (United States Army Corps of Engineers 1980). Table 3 shows that this river had a greater percentage of elevated cyanide and phenolic levels in 1978 and 1979 than even upriver Ohio R. water quality monitoring stations. This tributary supported the lowest number of taxa among the tributary stations sampled (Fig. 8), further supporting the high toxicity — low number of taxa conjecture. Individuals of the species Cricotopus bicinctus, Nanocladius distinctus, and Thienemannimyia gr. made up 95% of the total number of chironomids (63%, 19%, and 13% respectively) collected at this site. Dominance of the chironomid community by the combination of C. bicinctus, N. distinctus, and a species belonging to the Thienemannimyia gr. was also apparent in multistressed areas of the Great Miami R. (Beckett 1978). Simpson and Bode (1980) have indicated that these species are prevalent in areas of New York rivers subjected to tox- ic wastes. Cricotopus bicinctus has been found to be abundant in lotic systems receiving electroplating wastes (Surber 1959), chronic stress from copper (Winner et al. 1980), oil contamination (Rosenberg and Wiens 1976), and chlorinated sewage treatment plant effluents (Beckett 1978). Simpson (1980d) found this species to be a dominant in areas of the Niagara R. below industrialized areas. Although C. bicinctus is a widespread species and is found both in unpolluted and polluted waters, its tendency to be found as a numerical dominant over a spectrum of toxic conditions makes it a useful indicator of toxic effects. MEM. AMER. ENT. SOC., 34 26 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY 3 1x10 Cincinnati Pittsburgh DISCHARGE Louisville ix1o4 RIVER 1x10 BOD5s (kg/day) / RELATIVE STP EFFLUENT lo} 200 400 600 800 1000 1200 1400 RIVER KM Fic. 6. 5-Day BODs of major (> 1000 kg BOD/day) sewage treatment plant (STP) ef- fluents along the length of the Ohio R. during 1978. At each site BOD input has been divided by the relative river discharge. The relative river discharge is equal to the mean river discharge (Sept. 1978) at that point divided by the mean river discharge in Sept. 1978 at the Ohio R. origin (Pittsburgh) (Table 1). The effluents from the principal STP at Pittsburgh, Cincinnati, and Louisville are indicated. Note that the ordinate is on a logarithmic scale. The two Kanawha R. sampling stations both supported relatively high numbers of taxa (Fig. 8) but presented an interesting contrast in species composition. Cricotopus intersectus gr., averaging 118 individuals/sampler and making up 61% of the chironomid community (Table 2) at Winfield Dam, downstream of Charleston, W. Va., was not represented by even a single individual at London Dam, located 83 km upstream of Winfield Dam. The Charleston sewage treatment plant released an organic-rich ef- fluent in the summer of 1978 (D. Fisher — Dept. of Natural Resources, Charleston, W. Va., personal communication) and daily average dissolved oxygen levels at Winfield were less than 5.0mg/1 for 41.7% of the days in the Sept. 1978 macroinvertebrate sampling period (ORSANCO 1980). Hirvenoja (1973) found C. intersectus to be characteristic of eutrophic TABLE 2. Station No. D. C. BECKETT AND J. L. KEYES ay Dominant (by number) taxa at September 1978 Ohio River Valley macroinverte- brate sampling sites. Two most numerous non-chironomid taxa and two most numerous chironomid species shown. Chironomids Non-chironomid taxa 1 2 Gammarus sp. Nais bretscheri Dero sp. Cyrnellus fraternus Gammarus sp. Dero sp. Cyrnellus fraternus Gammarus sp. Cyrnellus fraternus Potamyia flava Gammarus sp. Cyrnellus fraternus Stenonema integrum Gammarus sp. Cyrnellus fraternus Argia apicalis Cyrnellus fraternus Gammarus sp. Cyrnellus fraternus Stenonema integrum Cyrnellus fraternus Stenonema integrum Cyrnellus fraternus Hydra sp. Cyrnellus fraternus Physa sp. Cyrnellus fraternus Stenonema integrum Hydropsyche orris Stenonema integrum Dugesia tigrina Dero sp. Dero sp. Mooreobdella microstoma Cyrnellus fraternus Caenis sp. Dugesia tigrina Gyraulus sp. Dugesia tigrina Nais bretscheri Cyrnellus fraternus Hydroptila waubesiana Cyrnellus fraternus Tricorythodes sp. Nanocladius distinctus Orthocladius sp. Nanocladius distinctus Orthocladius sp. Cricotopus intersectus gr. Nanocladius distinctus Ablabesmyia parajanta Nanocladius distinctus Cricotopus intersectus gr. Polypedilum illinoense Polypedilum illinoense Orthocladius sp. Polypedilum illinoense Cricotopus intersectus gr. Dicrotendipes sp. Nanocladius distinctus Dicrotendipes nervosus IT Polypedilum illinoense Dicrotendipes sp. Nanocladius distinctus Ablabesmyia parajanta Stenochironomus sp. Polypedilum illinoense Tribelos sp. Tribelos sp. Dicrotendipes nervosus Glyptotendipes sp. Polypedilum illinoense Nanocladius distinctus Parachironomus frequens Nanocladius distinctus Dicrotendipes sp. 2 Cricotopus bicinctus Nanocladius distinctus Dicrotendipes nervosus Tribelos sp. Dicrotendipes neomodestus Orthocladius sp. Cricotopus intersectus gr. Orthocladius sp. Polypedilum illinoense Dicrotendipes nervosus Ablabesmyia parajanta Dicrotendipes nervosus MEM. AMER. ENT. SOC., 34 28 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY 100 2000/|00 ml COLIFORMS > % OF SAMPLES WITH FECAL 200 400 600 800 1000 1200 1400 RIVER KM Fic. 7. % of 1978-1979 water samples collected at ORSANCO’s Ohio R. monitoring sta- tions in which fecal coliforms exceeded 2000 counts/ 100 ml of water. ‘‘Peaks’’ at river km 788 and 1007 are at monitoring stations located a short distance downriver from Cincinnati and Louisville, respectively. Lines connecting points are for visual continuity and do not indicate percentages at intermediate locations. Data source is ORSANCO (1980). lakes, existing in a wide range of dissolved oxygen concentrations, while Simpson and Bode (1980) often found this species group in sluggish areas of rivers and canals subjected to severe organic and toxic waste loadings. The Kanawha R. results therefore add evidence that this species becomes abun- dant in slow-water areas subjected to fairly high organic loadings. In- terestingly this organic loading did not markedly reduce the total taxa levels (Fig. 8). This suggests that, unlike small streams in which organic wastes tend to reduce the number of species present (Bartsch 1948; Tarzwell and Gaufin 1953; Lenat et al. 1980), moderate organic pollution in large rivers tends to reduce evenness and increase the total number of individuals (as a consequence of domination by species such as C. intersectus gr.) rather than decreasing the total number of taxa present. Although problems with organic loadings were manifested at Winfield Dam, the macroinvertebrate collections indicate that, historically, a con- siderable improvement in water quality has occurred at this site. In a mid-1960’s study Mason et al. (1971) found this area to have ‘‘odors of hydrogen sulfide and other chemicals ...in the area of the locks. D. C. BECKETT AND J. L. KEYES 29 35 KL TAXA OF NUMBER 0 200 400 600 800 1000 1200 1400 RIVER KM Fic. 8. Number of macroinvertebrate taxa collected at tributary sampling stations. The solid bar indicates the total number of chironomid taxa collected per site while the entire bar refers to the total number of all macroinvertebrate taxa collected. M= Monongahela R., B = Beaver R., Mu = Muskingum R., K, = Kanawha R. at London Dam, Ky = Kanawha R. at Winfield Dam, G = Green R., T = Tennessee R. The rivers are placed along the abscissa at their points of confluence with the Ohio R. Undecomposed leaves, coal fines, and tar-like deposits covered the bottom’’ (D. nervosus Type II was the most abundant chironomid collected at this site at that time, as mentioned earlier in this paper). Mason et al. (1971) concluded that their artificial substrate samplers (barbecue baskets) “contained mostly worms’’ and that ‘“‘low dissolved oxygen concentrations and toxic substances in the water below Charleston greatly reduced the variety of macroinvertebrates.”’ Despite indications of water quality improvement in area such as the Kanawha R. near Winfield Dam, this study and that of Taylor’s (1980) in- dicate that water quality problems evident in the mid-1960’s study (Mason et al. 1971) of the Ohio R. Valley still persist. Mason et al. (1971) sum- marized their long term study by stating that ‘‘macroinvertebrate popula- tions in the industrialized upper Ohio River area were sparse throughout the years sampled. The fauna was characterized by pollution tolerant and MEM. AMER. ENT. SOC., 34 30 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY TABLE 3. Percentages of elevated phenolic, cyanide, and coliform levels for 1978 and 1979 at the Ohio River tributary macroinvertebrate sampling stations. Data source is ORSANCO (1980). Station Phenolics Cyanide Fecal Coliforms (%) (%) (%) >lyg/l >Spg/l1 >Spg/l >25yg/1 >2000/100 ml Monongahela R. 86 35 89 32 32 at Pittsburgh, Pa. Beaver R. at 88 51 79 2 34 Beaver Falls, Pa. Muskingum R. near 79 21 03 02 12 Lock and Dam #2 Kanawha R. at 61 18 03 00 08 Winfield Dam Green R. near 03 00 00 00 07 Seebree, Ky. Tennessee R. near 24 00 00 00 00 Paducah, Ky. facultative organisms. There was a noticeable increase in the number and variety of benthic organisms in the middle and lower reaches of the Ohio River as compared to the upper reach.’’ Such trends are apparent in this 1978 study as well as in the mid-1960’s. This study has shown that, even in a complex, large-river situation, macroinvertebrate communities can exhibit an interpretable response to pollution. In this study, species richness has been very useful in assessing pollutional effects, especially in response to toxic wastes. Evenness of species distribution was a useful parameter in discerning organic waste in- put, especially when coupled with an ‘‘indicator species’’such as C. intersec- tus (gr.). Although the indicator species concept has recently received con- siderable criticism it has been a valuable tool in this study when coupled with community parameters such as species richness and evenness. Chironomid species such as C. bicinctus, C. intersectus gr., D. nervosus Type II, etc. were very important in assessing the nature and severity of pollutional impacts in this large-river study. Because of their great abun- dance and great species diversity the Chironomidae should be used exten- sively in pollutional research and water quality monitoring. ACKNOWLEDGMENTS A number of state and federal agencies cooperated in the planning of this study, which was coordinated by ORSANCO. We are grateful to Denise D. C. BECKETT AND J. L. KEYES 31 Bamber O’Brien for her exceptional assistance in the laboratory processing of the macroinvertebrates. LITERATURE CITED BartscH, A.F. 1948. Biological aspects of stream pollution. Sewage Works J. 20:292-302. BEAK, T.W., T.C. GRIFFING, AND A.G. AppLeBy. 1973. Use of artificial substrate samplers to assess water pollution, p. 227-241. In J. Cairns, Jr. and K.L. Dickson [ed.] Biological methods for the assessment of water quality, ASTM STP 528. American Socie- ty for Testing and Materials, Philadelphia, Pennsylvania. BECKETT, D.C. 1978. Ordination of macroinvertebrate communities in a multistressed river system, p. 748-770. In J.H. Thorp and J.W. Gibbons [ed.] Energy and environmen- tal stress in aquatic systems. National Technical Information Service, Springfield, Virginia. CONF-771114. AND P.A. Lewis. 1982. An efficient procedure for slide mounting of larval chironomids. Trans. Am. Microsc. Soc. 101:96-99. AND M.C. MILLER. 1982. Macroinvertebrate colonization of multiplate samplers in the Ohio River: the effect of dams. Can. J. Fish. Aquat. Sci. 39:1622-1627. BrINKHuRST, R.O. 1969. Changes in the benthos of Lakes Erie and Ontario. Bull. Buffalo Soc. Nat. Sci. 25:45-65. Carr, J.F. AND J.K. HILTUNEN. 1965. Changes in the bottom fauna of western Lake Erie from 1930-1961. Limnol. Oceanogr. 10:551-569. CHUTTER, F.M. 1972. An empirical biotic index of the quality of water in South African streams and rivers. Water Res. 6:19-30. CoFFMAN, W.P. 1978. Chironomidae, p. 345-376. In R.W. Merritt and K.W. Cummins [ed.] Aquatic insects of North America. Kendall/Hunt Publishing Co., Dubuque, Iowa. ForsBEs, S.A. AND R.E. RICHARDSON. 1913. Studies on the upper Illinois River. Bull. Nat. Hist. Surv. 9:481-574. GauFIN, A.R. AND C.M. TARZWELL. 1952. Aquatic invertebrates as indicators of stream pollution. Public Health Rep. 67:57-64. 1955. Environmental changes in a polluted stream during winter. Am. Midl. Nat. 54:78-88. 1956. Aquatic macro-invertebrates as indicators of organic pollution in Lytle Creek. Sewage Ind. Wastes 28:906-924. Hacu CHEMICAL CoMPpANy. 1975. Hach water and wastewater analysis procedure manual. Hach Chemical Co., Ames, Iowa. HitsENHOFF, W.L. 1977. Use of arthropods to evaluate water quality of streams. Wis. Dep. Nat. Resour. Tech. Bull. 100:1-17. Hirvenosa, M. 1973. Revision der Gattung Cricotopus van der Wulp and iher Verwandten (Diptera, chironomidae). Ann. Zool. Fennici 10:1-363. Hocutt, C.H. 1975. Assessment of a stressed macroinvertebrate community. Water Resour. Bull. 11:820-835. LenaT, D.R., L.A. SMock AND D.L. PENROsE. 1980. Use of benthic macroinvertebrates as indicators of environmental quality, p. 97-112. In D.L. Worf [ed.] Biological monitor- ing for environmental effects. D.C. Heath and Co., Lexington, Massachusetts. Mason, W.T., JR., P.A. LEwis AND J.B. ANDERSON. 1971. Macroinvertebrate collections and water quality monitoring in the Ohio River Basin 1963-1967. Water Quality Office, United States Environmental Protection Agency, Cincinnati, Ohio. 52 p. MEM. AMER. ENT. SOC., 34 32 BIOLOGICAL APPRAISAL OF OHIO RIVER WATER QUALITY Mason, W.T., JR. AND P.P. YEvicH. 1967. The use of phloxine B and rose bengal stains to facilitate sorting benthic samples. Trans. Am. Microsc. Soc. 86:221-223. Moztey, S.C. 1979. Neglected characters in larval morphology as tools in taxonomy and phylogeny of Chironomidae (Diptera). Entomol. scand. Suppl. 10:27-36. AND R.P. HowMILLER. 1977. Zoobenthos of Lake Michigan. Environmental status of the Lake Michigan region (volume 6). ANSL/ES-40. Argonne National Laboratory, Argonne, Illinois, 148 p. Ourto RIVER BASIN Commission. 1978. The Ohio River Basin. Ohio River Basin Com- mission, Cincinnati, Ohio. 66 p. OLIVER, D.R., D. MCCLYMONT AND M.E. RoussEL. 1978. A key to some larvae of Chiro- nomidae (Diptera) from the Mackenzie and Porcupine River Watersheds. Can. Fish. Mar. Serv. Tech. Rep. No. 791:1-73. Orsanco. 1980. Assessment of water quality conditions Ohio River Mainstem 1978-9. Ohio River Valley Water Sanitation Commission, Cincinnati, Ohio. 39 p. RICHARDSON, R.E. 1928. The bottom fauna of the Middle Illinois River 1913-1925; its distribution, abundance, valuation, and index value in the study of stream pollution. Bull. Ill. Nat. Hist. Surv. 17:387-475. ROSENBERG, D.M. AND A.P. WIENS. 1976. Community and species responses of Chir- onomidae (Diptera) to contamination of fresh waters by crude oil and petroleum pro- ducts, with special reference to the Trail River, Northwest Territories. J. Fish. Res. Board Can. 33:1955-1963. SAETHER, O.A. 1979. Chironomid indicators as water quality indicators. Holarct. Ecol. 2:65-74. SANDERS, H.L. 1968. Marine benthic diversity: a comparative study. Am. Nat. 102: 243-282. SIMBERLOFF, D. 1972. Properties of the rarefaction diversity measurement. Am. Nat. 106: 414-418. Simpson, K.W. 1980a. Macroinvertebrate survey of the Buffalo River System — 1976. N.Y. State Dep. Health, Environ. Health Rep. No. 8:1-31. 1980b. Macroinvertebrate survey of the Allegheny River, New York — 1975. N.Y. State Dep. Health, Environ. Health Rep. No. 9:1-39. 1980c. Macroinvertebrate survey of the Mohawk River — Barge Canal System, 1972. N.Y. State Dep. Health, Environ. Health Rep. No. 10:1-43. 1980d. Macroinvertebrate survey of the Niagara River — 1976. N.Y. State Dept. Health, Environ. Health Rep. No. 11:1-29. Simpson, K.W. AND R.W. BopE. 1980. Common larvae of Chironomidae (Diptera) from New York State streams and rivers. N.Y. State Mus. Bull. No. 439:1-105. SURBER, E.W. 1959. Cricotopus bicinctus, a midgefly resistant to electroplating wastes. Trans. Am. Fish. Soc. 88:111-116. TARZWELL, C.M. AND A.R. GAuFIN. 1953. Some important biological effects of pollution often disregarded in stream surveys. Purdue Univ. Engineer. Bull. — Proc. 8th Ind. Waste Conf. 1953:295-316. TAYLOR, R.W. 1980. A survey of the freshwater mussels of the Ohio River from Greenup Locks and Dam to Pittsburgh, Pa. Report to U.S. Army Corps of Engineers, Huntington District, Huntington, West Virginia. 71 p. UNITED STATES ARMY CorPs OF ENGINEERS. 1980. Ohio River navigational project opera- tion and maintenance. U.S. Army Corps of Engineers, Cincinnati, Ohio. 229 p. UNITED STATES ENVIRONMENTAL PROTECTION AGENCY. 1973. Biological field and laboratory methods for measuring the quality of surface waters and effluents. United States Environmental Protection Agency, Cincinnati, Ohio. D. C. BECKETT AND J. L. KEYES 33 WituM, J.L. 1970. Range of diversity index in benthic macroinverterbrate populations. J. Water Pollut. Control Fed. 42:R221-R224. WituM, J.L. AND T.C. Dorris. 1968. Biological parameters of water quality. Bioscience 18:447-481. WINNER, R.W., M.W. BoESEL AND M.P. FARRELL. 1980. Insect community and structure as an index of heavy-metal pollution in lotic ecosystems. Can. J. Fish. Aquat. Sci. 37:647-655. MEM. AMER. ENT. SOC., 34 hi, Qe aa. patly abd ie i eA te ‘gy linia! ie® Ties) i} ‘, : 1 A preliminary Study of the Chironomidae (Diptera) from a Stream in Northern Nigeria BY A.E. BROWN Department of Biological Sciences Bayero University P.M.B. 3011 Kano, Nigeria ABSTRACT. — Male imagines were collected weekly during the dry season, October to June, from a stream below a large reservoir. The stream is perennial and is derived from a piped outlet of hypolimnetic water from the reservoir. The fast flowing, relatively cool water makes this an unusual habitat in the Sudan savanna zone of West Africa. Forty species were recorded, 19 were new records for Nigeria and 11 were new records for West Africa. A number of cosmopolitan species found throughout Africa were also recorded. Cricotopus scottae was the predominant species (69%) and emerged throughout the dry season. Cladotanytarsus pseudomancus (6%), Harnischia curtilamellata (4%) and Cricoptus verbekei (3%) were the next most abundant species in male imagine collections. A preliminary investigation of the distribution of species on the main substrate types is included. INTRODUCTION The taxonomy of chironomid male imagines from the Afrotropical region is comparatively well known mainly as a result of the extensive revi- sions by Freeman (1955, 1956, 1957, 1958). A number of significant con- tributions have been made since that time for example by Brundin (1966), Dejoux (1968, 1969, 1970a, 1970b), Harrison (1970, 1971, 1978) and Lehmann (1979, 1981). However, despite this, comparatively little is known about the geographical distribution and ecology of chironomids from the region, particularly those associated with lotic habitats. This study is a preliminary investigation of the chironomid fauna from an unusual lotic habitat in northern Nigeria. Stupy AREA The stream site is located in the south-west corner of Kano State in northern Nigeria and lies in the Sudan savanna vegetation belt (Fig. 1). The site is below the dam wall of a reservoir, Tiga Lake, which was created in 1974. The climate of the area is characterised by distinct wet and dry seasons (Fig. 2). The wettest months are June, July, and August. No rain falls from November to March. This period is cooler due to the effect of the a5 MEM. AMER. ENT. SOC., 34 36 NORTHERN NIGERIA CHIRONOMIDAE ) 200km (—_____} Fic. 1. Location of stream site in Nigeria. Sudan savanna zone shaded. Harmattan wind and dust that it carries. The last. rain in 1981 was in September and the first rain in 1982 was at the end of May. The stream is derived from a piped outlet of hypolimnetic water from the reservoir. The water is fast flowing and relatively cool. From February to May 1982 the mean minimum temperature was 20°C and the mean max- imum temperature was 21.5°C. The temperature of the surface water of Tiga Lake during this period was 25°C. The stream is perennial whilst similar natural streams in the area are seasonal. Chemical analysis of the stream water was carried out on 16-4-82 using Hach DL-EL/4 equipment. units) TOTAL RAINFALL (arbitrary A. E. BROWN ia AUN IVYddWsil | Diese WeeAy aioe) uke SOlsNo) >) — (=p) NUMBER OF MALE IMAGINES Ow © NO © Selle Deki oF ie APM 1981 1982 Fic. 3. Variation in the total number of male imagines collected. The number of species recorded is also given. MEM. AMER. ENT. SOC., 34 NORTHERN NIGERIA CHIRONOMIDAE 40 e e e e e e e e xX 90 Z| ds sadipuajoiiq *61 e e e e e e e x e 10 [ (doJJory) vjoaugisy Sadipuajosiq “81 e e e e e e e e x Z0 | (weulsely) si1odsos0481u SnwmoUuosYyI0JAKID * LI e e e e e e e e xX Z0 | (wewieer4) Djooljiuoau snuouosyI0jddig “9 e e e e e e e xX e 1 a0) (UeWI9e1,J) Maupul] SnuIOUOMIYIOIAKID “S} e e e e e e e xX x L0 (JaJJary) spsaoip snuouosmyoojddiy “yt furmouolly) aAVNINONOWIHO e e e e e ° x e x 80 ZT (tangoy3}90H) jj JIA 14] DjjaluuDWeaUaIYT “€] e x e x e e e x xX 1 i ZT { (ueWIseI,J) SNQojiIa4 DIIWSOpNasd “TZ e e e xX e e e e e 10 Z| ds snippjz0uvNy “1 TUIWIOUSOLNOIN e e xX xX x e e e e V2 Z{ “ds snippjr0yoioLvg ‘Ol x x x x x x 4 x x €7% Z| (tanqoys}20H) Ifjnmap snippjrouavydvivd “6 x x x x x x x x x 4 (ueUulselq) 1ayaqsaa SndojoouD °8 e e x xX e xX x xX e TI (uewisel4) sndiuppns sndojoou) °L ».4 x x x x x x x x 7 69 (uewlisel4) apjj09s sndojooiD °9 4 x xX xX e e e e e el ] (sanqay3}o05) sisuanjupsiy sndojoouD °S e e e e x e xX x e €0 Z‘{ (ueulsely) snuvoiufp snippjro1Ipivy “py TulIpefsoyuO AVNITIGVTOOHLYO e e e x e e xX e e 70 (ueUulaelq) snjpiu0s sndduvjojiIn “€ e e e e e e x e e 70 Z[ (songoy3}20H) Sisuaininysjns DISIDT “TZ e e e e e x e e e Z0 (JafJory) snusdo vidojadpyauoyD “| TulInouejUsg AVNIGOdANV.L W Vv W 4 f d N O S % “eOyV “A\ 10} MoU—Z ‘BLIOBIN] 10J MOU—] “po1INd90 Asy} YOIYA UL syJUOUI oY) pu PoLied ay} J9A0 saroads Jo aouaIININO a8eIUIIINg =°7 ATEV 41 BROWN A. E. a ee a <~eeee = e e e e e e e e e e e e ».¢ e x e e e e x e e x< e e e e Cy e e e e e e e e e e e e e x e e e e e e e e e e e e e e x x x x V W A x eoxeee xeeoerwooonoenoe Ko XK 0 0 ee Xe e a a ee ee ee a er a a a fa) eoeee x a a a ee ci a a er er ee Z. xexe x mK KKK Ke eo Keo ew ew ew oe (e) xe eo x xx eo KK Ke eo oo eo eo ew oO n v0 07 £0 c0 LA) 9s 10 10 v0 c0 €°0 9°0 v0 £0 £0 v0 v0 80 a) LY Set % “ds xojduios piyssiusvzy “Op soiseds pooeidup, (UBWIO01J) SnyoJoUOaDIpDdS snsunDjAUuD], “6€ (ueUdeI4) Snjnpijjod snsivjduny ° gE Z| (ueweely) snasnf snsuvjkuvjoayy ° LE ] (uewlse14) snjonpas snsunjdunjopvjD “9€ (19ngeY3130H) snauDWOpNasd snsinjduvjopvj]D “S€ TuIsIeyAue |, (tongsy3190H) appunsn snuouomyIOUaY “PE I (J9J Jory) Snsojasi4] SnuouosmyIOUaXKX *EE ZT “ds winpipaddjod “7 (UeWOsIy) DUINId UNjIpaddjod “1¢ (JoJJory) winpiydounjau winjipaddjod °O¢ (JoJJory) snsoisuo] unjipaddjod *67 ZT (longays}20H) asuayngiy uinjipaddjod *8Z (Jonqoy3}90H) Yjnmap wnjipadajod ° LZ (1ajJory) aviuissdqn winjipaddjod ~97 (saJJory) vosnf sisopojin’ “SZ (19JJory) Sidjodiaasq winsopojiN’ “yZ (19JJoTy) VIONGIA|Ig LINIOPOJIN’ “€Z [ (doJJony) 4aua) snuouosmyr0IIIP *7Z (Ueda Iq) Jafij1Js SNUOUOLIYIOINIP * {TZ (Yoo]TeW) MiDIjawunjNAnd DIYOSIUIDFT “07 penunuoy) “7 AaTdv yi MEM. AMER. ENT. SOC., 34 42 NORTHERN NIGERIA CHIRONOMIDAE PERCENTAGE NUMBER SAND GRAVEL Fic. 4. Proportions of the major taxa recorded in each substrate. Stippled — Polypedilum, Clear — Cricotopus, Diagonal lines — Tanytarsini, Black — others. ing some of the smaller fluctuations there does appear to be a trend from a large number of species and high numbers in September and October to a low number of species and low numbers in November and December, and then an increase again. The declining in numbers and species diversity from September was the result of a decline in Cryptochironomus species, Microchironomus species, Nildorum species and Cladotanytarsus pseudomancus (Table 2). C. pseudomancus did occur later in the study period but only in relatively low numbers. Larger numbers of Orthocladiinae were recorded than Chironominae. The dominant species throughout the dry season was Cricotopus scottae (Table 2). It was recorded on every sampling occasion. Cricotopus verbekei, Paraphaenocladius dewulfi and Harnischia curtilamellata were recorded every month but not on each sampling date. A. E. BROWN 43 TABLE 3. Species recorded from different substrate types by mass rearing. Species are ar- ranged in order of abundance. SUBSTRATE SPECIES a rN a ee es ee ALGAE (madicolous habitat) Larsia rutshuruiensis Cricotopus scottae Cricotopus verbekei VEGETATION Cricotopus sudanicus (marginal Gramineae) Polypedilum melanophilum Rheotanytarsus fuscus Polypedilum dewulfi Paraphaenocladius dewulfi Cladotanytarsus pseudomancus Cladotanytarsus reductus SILT Cricotopus verbekei Cricotopus scottae Dicrotendipes sp. Harnischia complex sp. SAND Nilotanypus comatus Nanocladius sp. Polypedilum longicrus GRAVEL Polypedilum pruina Cricotopus verbekei Cricotpus sudanicus Cricotopus verbekei STONES Polypedilum dewulfi Cardiocladius africanus Nilotanypus comatus A total of 40 species was recorded from the site (Table 2). 19 species were new records for Nigeria and 11 species were new records for West Africa. Geographical distributions were based on data in Freeman and Cranston (1980). A preliminary investigation of larval occurrence on five major substrate types was carried out. The upper reaches of Tiga stream drop rapidly pro- ducing a turbulent flow of water. Here the substrate is composed of large stones and Cricotopus larvae comprised 86% of the chironomid fauna (Fig. 4). As the gradient of the stream decreases the current slows and the substrate changes to gravel, sand and silt, with marginal vegetation com- posed mainly of grasses. Polypedilum and Cricotopus larvae occurred in MEM. AMER. ENT. SOC., 34 44 NORTHERN NIGERIA CHIRONOMIDAE approximately equal numbers in the gravel substrates (Fig. 4). Polypedilum larvae were predominant on vegetation, silt and sand substrates (Fig. 4). Tanytarsini larvae were present in all substrate types, except sand, and were most abundant in silt (Fig. 4). Mass rearing from each substrate type revealed the actual species com- position. Only one species, Larsia rutshuruiensis, was reared from algae collected from a madicolous habitat (Table 3). From vegetation only two species of Polypedilum were reared and from sand and silt substrates none were reared. Polypedilum species were expected in mass rearing from sand and silt substrates on the basis of the larvae collected from these substrates. Three species of Cricotoptus predominated the reared adults from vegeta- tion, two species of Cladotanytarsus predominated the reared adults from silt and a species belonging to the Harnischia complex was the most abun- dant species from the sand substrate (Table 3). Fewest adults were reared from the sand and gravel substrates. DISCUSSION Two streams studied by Lehmann (1979, 1981) in Zaire, central Africa, show a number of similarities with this study. The streams were perennial and showed no major water level or temperature changes throughout the year. The total number of species were comparable although the number of species within each tribe differs (Table 4). The larger number of Chironomini species recorded at the Tiga stream may have been due to col- lection of imagines that emerged from Tiga lake. This is supported by the fact that several Chironomini species recorded as imagines were not reared from larval material collected from the stream. Comparison with the fauna of the two streams in Zaire suggests that further sampling and rearing may yield a larger species list, particularly for the Orthocladiinae. Of the seven Orthocladiinae genera recorded from the Tiga stream all but one, Paraphaenocladius, were recorded from the river Kalengo by Lehmann (1979). However, the actual species differed. The River Kalengo was dominated by Chironomini, mainly due to Microtendipes numerosus, whilst the Tiga stream was dominated by Orthocladiinae, mainly Cricotopus scottae. All three streams possessed a number of Polypedilum species, several of the same species occurring. Most species recorded in the River Kalengo by Lehmann (1981) emerged throughout the year. In the Tiga stream Cricotopus scottae, Cricotopus verbekei and Paraphaenocladius dewulfi were found as imagines throughout the dry season. Several other less abundant species may also emerge throughout the dry season, for example, Cricotopus kisantuensis and Pseudosmittia rectilobus. Cryptochironomus, Microchironomus and A. E. BROWN 45 TABLE 4. Number of species in each tribe. Data for Simisimi stream and the river Kalengo from Lehmann (1979, 1981). NUMBER OF SPECIES Tiga stream Simisimi stream River Kalengo Nigeria Zaire Zaire Macropelopiini 0 0 1 Pentaneurini 3 5 6 Orthocladiini 7 13 6 Metriocnemini 3 11 5 Chironomini 22 10 13 Tanytarsini 5 7 7 Total 40 46 38 Nilodorum species were all recorded in September and October immediately after the rainy season but were not recorded subsequently. Species recorded both in this study and in Zaire by Lehmann (1981) show similar substrate preferences. However, in the Tiga stream considerable overlap was observed in the substrates utilised by species. For example, Cricotopus verbekei was reared from substrate samples of silt, gravel, stones and vegetation. Further work is required before any typical species assemblages can be predicted for a particular substrate type. ACKNOWLEDGEMENTS I would like to thank Safianu Rabiu, Magnus Nzerue and Isaac Uyouko for assistance in the field and Dr. P.S. Cranston for checking identifica- tions. REFERENCES BRUNDIN, L. 1966. Transantarctic relationships and their significance evidenced by chir- onomid midges. With a monograph of the subfamilies Podonominae and Aphroteniinae and the austral Heptagyiae. K. Svenska Vetensk Akad. Handl. 11:1-472. Deyoux. C. 1968. Contribution a l’etude des premiers etats des Chironomides du Tchad. Hydrobiologie 31:449-464. 1969. Contribution a l’etude des premiers etats des Chironomides du Tchad (2e note). Description de Tanypus fuscus et Tanypus lacustris. Bull. Mus. natn. Hist. nat. Paris 41:1152-1163. 1970a. Contribution a l’etude des premiers etats des Chironomides du Tchad (Insectes, Diptera) (3e note). Description conparee des nymphes de Chironomus (Nilodorum) brevibucca, Ch. (N.) brevipalpis et Ch. (N.) fractilobus. Bull. Mus. natn. Hist. nat. Paris 42:175-184. . 1970b. Contribution a l’etude des premiers etats des Chironomides du Tchad (4e note). Cah. O.R.S.T.O.M., ser. Hydrobiol. 4:39-59. MEM. AMER. ENT. SOC., 34 46 NORTHERN NIGERIA CHIRONOMIDAE FREEMAN, P. 1955. A study of the Chironomidae (Diptera) of Africa South of the Sahara. Part I. Bull. Br. Mus. nat. Hist. Ent. 4:285-368. . 1956. A study of the Chironomidae (Diptera) of Africa South of the Sahara. Part II. Bull. Br. Mus. nat. Hist. Ent. 4:285-368. 1957. A study of the Chironomidae (Diptera) of Africa South of the Sahara. Part III. Bull. Br. Mus. nat. Hist. Ent. 5:323-426. 1958. A study of the Chironomidae (Diptera) of Africa South of the Sahara. Part IV. Bull. Br. Mus. nat. Hist. Ent. 6:261-363. AND P.S. CRANSTON. 1980. Family Chironomidae. In Crosskey, R.W. (ed.) Catalogue of the Diptera of the Afrotropical region. Trust. Brit. Mus. (Nat. Hist.) 175-202. Harrison, A.D. 1970. The Tanypodinae (Chironomidae) of Africa south of the Sahara. Key to known genera and some species. News Lett. Limnol. Soc. South Afr. 15:62-73. 1971. A conspectus of the Nacropelopiini and Pentaneurini (Tanypodinae: Chironomidae) of Africa south of the Sahara. Can. Ent. 103:386-390. 1978. New genera and species of Tanypodinae (Diptera: Chironomidae from Africa south of the Sahara. J. Ent. Soc. Sth. Afr. 41:63-80. LEHMANN, J. 1979. Chironomidae (Diptera) aus Fliessegewassern Zentralafrikas (Syste- matik, Okologie, Verbreitung und Producktionsbiologie) Teil I. Kivu-Gegiet, Ostzaire. Spixiana, Zeitschrift fur Zoologie, Munchen, Suppl. 3:1-143. 1981. Chironomidae (Diptera) aus Fliessgewassern Zentralafrikas. Teil II. Die region um Kisangani, Zentralzaire. Spixiana Zeitschrift fur Zoologie, Munchen, Suppl. 5:1-85. SALTER, P.J. AND J.B. WititAMs. 1967. The influence of texture on the soil moisture characteristics of soils. J. Soil Sci. 18:1-8. Factors Influencing the Chironomid Community of a Nearshore Sand Area C.E. CARTER AND R.W.G. CARTER New University of Ulster Limnology Laboratory Traad Point, Drumenagh, Magherafelt Co. Derry, and Environmental Science The New University of Ulster Coleraine, Co. Derry BT52 ISA, Northern Ireland. ABSTRACT. — An intensive study of the chironomid fauna in the sandy littoral zone (< 1m depth) of Lough Neagh revealed the important interactions between facies assemblages and community structure. The migratory nearshore bedforms exposed to breaking waves had a more restricted community, both in terms of species and numbers, than their sheltered water counterparts. Some species were more tolerant of sand disturbance than others. These spatial patterns are related to physical factors, for example, wave energy and bottom shear stress. INTRODUCTION This small-scale study forms part of a larger survey of the littoral fauna of Lough Neagh. Sand is one of the major littoral habitats (approximately 30% of the shoreline) and a small beach area has been studied since 1980 in order to describe the chironomid community and the life cycles of the species involved. Coincidentally, work has been done on sand bedform movements in the same area. It became apparent that the morphology of the lake bed is an important factor in the distribution of the chironomid community, and this paper attempts to show how different species react to changes in morphology, and to discuss some of the physical factors that cause these changes. An appreciation of these factors is obviously impor- tant in the design of a sampling programme for such an area. StuDy AREA Traad Beach comprises the northern 100 m of a sandy embayment, Ballyronan Bay, in the north-west corner of Lough Neagh (Fig. 1). The beach opens to the south and is exposed to waves generated by prevailing winds. Waves are generally 0.2 to 0.5 m in height, but may reach over 1.0m during onshore storms. Wave periods commonly fall between 2 and 4 seconds. Water level in the lake is artifically controlled and varies only a few tens of centimetres over a year. However seiches, set up by winds blowing 47 MEM. AMER. ENT. SOC., 34 48 NEARSHORE CHIRONOMID COMMUNITY IRELAND NUU Limnology Fic. 1. Location of the study area. over the longest fetches (25-30 km) may alter the local water level by 10-15 cm over 30-60 minutes (Carter 1983). The gradient on the nearshore shoal- ing slope is very low (< 1°) and incident waves break into a dissipative, spilling domain. Breaker position is strongly influenced by the position of the nearshore bedforms, consisting of three or four shore-parallel sand bars at depths ranging from 0.1 to 1.2m. According to wave height, waves break and reform over these bars; energy dissipation due to breaking being be- tween 75-95% of incident wave energy (Carter and Balsillie 1983). The bar bedforms, which have a slow response time relative to wave frequency, show a tendency to migrate landward only during storms or as the lake water level falls. Some of these migrations are quite rapid, up to 0.3m/day, C.E. CARTER AND R.W.G. CARTER 49 23 September 1980 Breaking Breaking Breaking = IRL I RPL 30-1 WAVE ENERGY 20-5 0 1505 BOTTOM SHEAR STRESS a all ee ee foe] ara T rack T 1 0 50 100 150 m Log energy joules/mx 10 Maximum bottom shear stress cm/s Fic. 2. Wave energy conditions and bottom shear stress in the study area. and occur across 10-20 m of the littoral zone. Sediment is returned lakeward by return currents due to wave set-up. The large-scale bedforms are covered by smaller forms — dunes and ripples related to flow regime. Shear stress across the shore increases rapidly around the bar crests, falling away in the bar troughs (Fig. 2). The beach is bounded by a rocky headland (Traad Point) to the north and a Phragmites stand to the south. On the northern side of Traad Point is a gravel shore which changes to sand at a depth of 0.5 m. Some samples were taken from this for comparison. METHODS The lake bed was surveyed by Quickset level for the longer profiles, and by measuring the depth below the still water level at regular intervals when the intensive sampling was being done. Sediment cores were taken with a modified Kajak corer and the depth of disturbance of the sand at the sampl- ing points (apparent as a colour difference between the layers of sand) was recorded in the field. In the long term survey, five cores were taken at each site but during the intensive, daily sampling only two cores per site were MEM. AMER. ENT. SOC., 34 50 NEARSHORE CHIRONOMID COMMUNITY 4 April 82 | Depth of sand disturbance 204 5 April 82 10 6 April 82 404 7 April 82 - Pe € S Ee = a § 8 April 82 s z 20 304 405 9 April 82 10 F J 20 30 404 10 April 82 15 April 82 10 204 30 4 404 T T a Th alee al i= a3 al alee 0 4 8 12 16 20 24 28 32 36 40 Distance from waters edge in m Fic. 3. Nearshore morphology profiles in April 1982. taken for logistical reasons. The intensive sampling period mainly referred to in this paper was seven consecutive days and one set of samples four days later, in early April, 1982. C.E. CARTER AND R.W.G. CARTER 51 0 0 20 30 40 50 60 70 80 90 100 10 120 130 140 yt 4 —_ 1 Jt 4 —— April 1980 — —- August 1980 ------- January 1961 —-—-- June 1981 Apri 1982 s o go? € t TRAAD BARS Fic. 4. Variation in nearshore morphology over 2 years. RESULTS The nearshore morphology profiles for the period of intensive sampling are shown in Figure 3. There were two bars present at that time in the area sampled. Bar I, 7-8 m from the water’s edge, varied between 0.1 to 0.18 m below water level, with a depth of sand disturbance on its crest of about 0.15 m (range 0.13 to 0.26 m). Bar II, 20-23 m from the water’s edge was from 0.02-0.12 m below the water surface, with sand disturbance 0.25-0.3 m. There was very little apparent sand movement during the period of study, a time of exceptional calm when the wave height at breaking on Bar II barely exceeded 0.1 m. Figure 2 is a stylized diagram showing the stresses to which bars are subject in terms of wave energy and bottom shear stress. It is particularly evident that the latter is much higher on the crest of the bars. Figure 4 indicates the degree to which the sand bars can move, being near- shore profiles measured over a period of two years. Bar I was more often absent than present, Bar II, when present, varied horizontally over a distance of c. 3 m and vertically about 0.15 m. There are futher bars, not sampled during this study, for which the range of movement was greater. The total chironomid fauna of the various inshore zones is shown in Table 1, where results from several sites have been grouped together. The nearshore area was the richest in species terms, with some e.g. En- dochironomus, Microtendipes, which did not occur elsewhere on the transect. The inner bar (1) had quite a variety of species but no Cladotanytarsus or Tanytarsus, both of which reappeared between the bars. The outer bar (II) had the poorest fauna, followed by its lakeward side. The densities of the five most common species are shown (Fig. 5) but it must be remembered that confidence limits of there estimates are large, because of the small number of samples. Paracladopelma and Stictochironomus were generally the most abundant species, and the most widely distributed, although Paracladopelma was generally absent from Bar II and its im- MEM. AMER. ENT. SOC., 34 NEARSHORE CHIRONOMID COMMUNITY 2 3B a WAV elas Worsvers PN wyvs ; VIVISIN I Fic. 5. The density of the five most common species in April 1982. C.E. CARTER AND R.W.G. CARTER 53 125 Lake Position of transect over Bar I (11 April 82) Water depth and depth of sand disturbance at sampling Points along transect € 10 oO & x= & 30 vw (a) 71) I T T T isa T tal au 1 1 2 3 4 5 G 2 8 2 @ i P Sampling points Distribution of fauna along transect. ~ = 7) c co o ace c °o 2 2 E) a ° a Sampling points N Chironomus cingulatus Paracladopelma ll Stictochironomus sticticus is Cladotanytarsus Fic. 6. Morphology and fauna of Bar II. mediate lakeward slope. Cladotanytarsus was generally confined to sites 1-3, the most sheltered. Chironomus cingulatus was for no obvious reason, never found at site 1, but in at least small numbers at all other sites. Monodiamesa was present at sites 1-6, but very infrequent at other sites. MEM. AMER. ENT. SOC., 34 NEARSHORE CHIRONOMID COMMUNITY 54 DISIDJON SnIpvjIOYJAO DsaUpipouow puyadoppjI01g SNUOUOMYIOJINS SNIDINBUID “D (11-6 Sous) Jeg JojnO JO preMayeT pujadoppja nog SNWOUOMYIOIINS SNIDINSUID “D (8 2118) leg 10InO JO 1sdID DISIDJON DIISDYJJOd psaupipouo~ snsipjaun J, snsipjdAunjopojd, wnpipeddjog pulyadoppjap.ing SNWOUOMYIOJINS SnjojnsUuid “dD (L-¢ seus) sleg us9M10g pisvyjjOd DSAaUIDIPOUO- wuinpipadajog pujadopvjap.i0g SNWOUOMYIOJINS sadipuajojaajp Snjpjnsuid “dD (p ous) Jeg IJouU] JO 1saID DIISDYJIJOd SniIpDjIOYJ4AOC DSAaUIDIPOUOP, snsinjadun J, SnsipjduvjopvjD wnjipaddjog Dujadoppja0.vg Sadipuajoiiw SNWOUOLYIOJINS snwmouomyoopugq sadipuajojda]/5 SNIDINSUID SNUOUOMYD (€-1 Sets) Jeg Iouuy Jo s10Ysu] “sQUOZ SIOYSUI SNOLIeA sy] JO eUNe; pruloUOITYyD “| alavy C.E. CARTER AND R.W.G. CARTER 55 30 January 81 (3000) (4800) = HI Population density 100 re) Water level m above OD 7 R 6 —____— t t ! t i 2 30 5 10-15 8 Depth of sand disturbance in cm 11 March 81 a 6 0 J ( al a ‘= Water level 2] 3 a Sas 1 = i t t t 2 6 2 20 4 Depth of sand disturbance in cm (3700) (7000) 13 August 81 i> a c © +) c © 3 > a ° Y a a re o 3 g = = Water level gu 3 € 1 t t t ! 3 10 3 2 Depth of sand Key disturbance in cm Oo Chironomus cingulatus O Stictochironomus sticticus i | Microtendipes ee Paracladopelma [_ Potypiditum f& Cladotanytarsus Monodiamesa Fic. 7. Fauna of the study area on selected dates in 1981. Total abundance was generally greatest at sites 1-3, then, in order of decreasing abundance, 5-7, 4(Bar I), 9-11, 8(Bar II). To investigate further the effects of a sand bar on the fauna, a transect was taken across Bar II, consisting of single cores taken at 1 m intervals (Figure 6). This shows very clearly the difference in sand disturbance be- tween crest and trough, 20-25 cm on the crest and 5-7 cm in the trough. Most of the fauna was confined to the trough, only C. cingulatus and Stictochironomus appearing more than once on the crest. MEM. AMER. ENT. SOC., 34 56 NEARSHORE CHIRONOMID COMMUNITY To set the fauna in a longer term context, Figure 7 shows three profiles with associated fauna from 1981. In January, number of species was restricted but abundance of some was quite high. By March, the fauna was very depleted (emergence was unlikely to have been a factor in this) but the picture was quite different in August when most species had entered a new generation. Judging by the depth of the disturbed layer, sand disturbance had been minimal for a period so, presumably, allowing young larvae to settle. A further indication of the potential variety of the fauna in a sand habitat is shown by the deeper water sites off the beach and on the more sheltered north side of Traad Point (Table 2). Several species, e.g. Cladotanytarsus, reappear at 1 m depth and can be very abundant, in- dicating that they are not depth limited on the inner depth profile. Microtendipes is present in the deeper samples and is most abundant on the sheltered side of the point, while Stictochironomus is the opposite, more abundant on the beach side. DISCUSSION The chironomid fauna of this sandy area of Lough Neagh is similar to that found in other lakes, e.g. Loch Leven (Maitland & Hudspith 1974). However, in an exposed place like Traad Beach, the fauna of the area from the inner bar (I) to the lakeward slope of Bar II is restricted both in species and numbers, and it is in this area that wave energy has its greatest effect on the lake bed, as can be seen in the profile of bottom shear stress. Different species react differently to this. Cladotanytarsus is restricted to the low energy zone inshore (sites 1-3), reappearing at depths of > 1 m where wave energy no longer causes sand disturbance. C. cingulatus distribution is not obviously related to sand disturbance; it is present at most sites along the profile, although it appeared to be less common at sites on the immediate lakeward slope of Bar II (9 & 10). Possibly, sustained disturbance is an adverse factor at these sites. Stictochironomus also shows no statistical rela- tionship to depth of sand disturbance and is able to tolerate conditions at all sites on the transect, often attaining peak numbers on the crest of one of the bars. As it appears to be less successful when other species are present, lack of competition on the crest of the bars could be a factor. Paracladopelma and Monodiamesa can to some extent be considered together as their abun- dance is strongly correlated (correlation coefficient 0.51, significant at 1% level). Both give significant inverse relationships when regressed against depth of sand disturbance (D). Paracladopelma = 6.6842 - 0.2666D Monodiamesa = 2.8854 - 0.1186D TABLE 2. Point. C.E. CARTER AND R.W.G. CARTER 1m. 1 Chironomid fauna and density in 1-2 m zone of Traad Beach and north Traad October 1981 C. cingulatus Stictochironomus Microtendipes Glyptotendipes Endochironomus Limnochironomus Paracladopelma Cryptochironomus ‘defectus’ Cladotanytarsus Tanytarsus Monodiamesa Natarsia Orthocladius C. cingulatus Glyptotendipes Microtendipes Endochironomus Cryptochironomus ‘defectus’ Cladotanytarsus Monodiamesa Orthocladius Potthastia longimana March 1982 C. cingulatus Glyptotendipes Stictochironomus Microtendipes Endochironomus Paracladopelma Cryptochironomus ‘defectus’ Cladotanytarsus Monodiamesa Orthocladius C. cingulatus Stictochironomus Endochironomus Cryptochironomus ‘defectus’ Cladotanytarsus Monodiamesa Orthocladius MEM. AMER. ENT. SOC., 34 Beach 1324 407 407 102 305 305, 102 7140 204 305 204 509 1018 102 1018 10200 611 102 102 815 407 102 102 305, 204 204 407 509 No m7” Point 3366 204 3060 1428 2958 102 305 14586 102 407 407 815, 204 204 611 102 713 102 815 102 1730 204 10608 305 204 204 102 305 102 611 204 58 NEARSHORE CHIRONOMID COMMUNITY Both these lines intersect the X axis at about 25 cm, the average value found on the crest of Bar II, both species seeming able to tolerate the degree of disturbance on Bar I. Another species of note is Microtendipes, generally a common inshore species and occasionally found off Traad Beach (Figure 7) but usually confined to the deeper sites (Table 2) and perhaps unable to tolerate sand disturbance. Similar factors to these have obviously been considered in relation to lotic fauna (Petran & Kothe 1978, Wiley 1978), but less often for lakes (but see Maitland 1979). Wiley (1981) did some experimental work on factors in- fluencing the penetration of chironomid larvae into particular substrates and found that the burrowing behaviour of Stictochironomus and Para- cladopelma was different. The former employed a ‘‘shovelling’’ motion and the latter a ‘‘jackhammer’’ approach, which reduced penetration time by 20%. He also found that body size affected penetration time. Work on similar lines might help to explain some of the distribution differences in lake chironomids. It is also possible that depth of occurrence in the sedi- ment could also be important in tolerating sand disturbance. Wiley (1981) found this to be so for Stictochironomus; it avoided stream bed erosion by burrowing deeper into the sediment. Preliminary results with divided cores from Lough Neagh are inconclusive, but this is another possible line of enquiry. In conclusion, it is obvious that the energy input on this small beach area is an important factor determining the distribution of the fauna, and that different species can tolerate different degrees of disturbance. Secondly, a point not much emphasized so far in this paper but nonetheless important, in exposed areas like this one, where there is a lot of sand movement, it is essential to understand the changing morphology of the lake bed when sampling, particularly whether samples are from bar or trough; otherwise interpretation of consecutive results from one site may be very difficult. ACKNOWLEDGMENTS C.E.C. would like to thank the Garfield Weston Trust for the grant which made this work possible. We would also like to thank Hugo McGrogan for field assistance, Mairead Diamond for typing the manuscript and Shirley Tinkler for drawing the diagrams. REFERENCES CARTER R.W.G. 1983. The geology, hydrology and sedimentology of Lough Neagh and its surrounding catchment. In Wood, R.B. and Smith, R. (eds.) Lough Neagh Monogr. Biol. (in press). C.E. CARTER AND R.W.G. CARTER 59 CARTER R.W.G. AND J.H. BALsmLLIE. 1983. On the amount of wave energy transmitted over nearshore sand bars. Earth Surf. Proc. Landforms 8: MalITLAND, P.S. 1979. The distribution of zoobenthos and sediments in Loch Leven, Kin- ross, Scotland. Arch. Hydrobiol. 85:98:125. MAITLAND, P.S. AND P.M.G. HupspiTH. 1974. The zoobenthos of Loch Leven, Kinross, and estimates of its production in the sandy littoral area during 1970 & 1971. Proc. Roy. Soc. Edin. 74B:219:239. PETRAN, M. AND P. KotTHE. 1978. Influence of bedload transport on the macrobenthos of running waters. Verh. Internat. Verein. Limnol. 20:1867:1872. Wiey, M.J. 1978. The biology of some Michigan trout stream chironomids (Diptera: Chironomidae). Mich. Acad. 11:193:209. 1981. An analysis of some factors influencing the successful penetration of sediment by chironomid larvae. Oikos 36:296:302. MEM. AMER. ENT. SOC., 34 ony Bee, 7 a" sere rpartea he 0H ln . . ; | Thoracic Chaetotaxy of Chironomid Pupae (Diptera: Chironomidae) WILLIAM P. COFFMAN Pymatuning Laboratory of Ecology Department of Biological Sciences University of Pittsburgh Pittsburgh, Pennsylvania 15260 ABSTRACT. — Thoracic chaetotaxy was studied in pupae of 285 Chironomidae species belong- ing to 158 genera and nine subfamilies. The purpose of the study was to observe patterns of chaetotaxy present in the higher taxa and to determine if such patterns are indicative of trends throughout the family. The analysis included the setal groups present, the number of setae per group and the longitudinal placement of the Dcs and Sas. Intraspecific variation was found to be very low, both within and between populations. Sexual dimorphism was not apparent. In- trageneric variations were found but, basically, the species of a genus have the same groups and number of setae and the same relative placement. Although variations were considerable within some subfamilies, each subfamily appears to have a basic plan. Observed trends through the family include: (1) the movement of the supraalar seta to the position of the fourth dorsocentral; (2) the loss of metanotal setae; and (3) the development of anterior and posterior pairs of dorsocentral setae. INTRODUCTION As a continuation of my study of character states in chironomid pupae (Coffman, 1979), I am presenting in this paper the results of a comparative analysis of the number and distribution of thoracic setae. Although many authors have included thoracic setae in descriptions of pupae, the character states of most lower taxa are unknown and general patterns associated with most higher taxa have not been described in detail. The purpose of this study is to determine, through the examination of a large number of species and genera, if patterns exist which are typical of higher taxa and if those patterns may be related by trends throughout the family. METHODS AND MATERIALS Slide mounted pupal exuviae representing 285 species, 158 genera and eight subfamilies were used in the analysis (Table 1). Character states for Aphroteniinae were determined from Brundin (1966). When possible, 61 MEM. AMER. ENT. SOC., 34 62 CHIRONOMID PUPAL THORACIC CHAETOTAXY TaBLE 1. Number of Chironomidae genera and species examined for chaetotax of the thorax. TAXA NO. OF GENERA NO. OF SPECIES TELMATOGETONINAE 1 1 TANYPODINAE 24 31 COELOTANYPODINI (1) (1) MACROPELOPIINI (7) (11) NATARSINI (1) (1) PENTANEURINI (14) (16) TANYPODINI () (2) PODONOMINAE 7 8 BOREOCHLINI (4) (4) PODONOMINI (3) (4) BUCHONOMYIINAE 1 1 DIAMESINAE 13 20 BOREOHEPTAGYIINI (1) (1) DIAMESINI (8) (15) HEPTAGYINI (3) (3) PROTANYPODINI (1) (1) PRODIAMESINAE (3) (4) ORTHOCLADIINAE 54 118 CHIRONOMINAE 55 102 CHIRONOMINI (42) (70) PSEUDOCHIRONOMINI (1) (6) TANYTARSINI (12) (26) TOTAL 158 285 several specimens of each species were examined to verify the chaetotaxy. The terminology for the groups of setae analyzed is slightly modified from Saether (1980): MAps-median antepronotal seta(e); LAps-lateral antepronotal seta(e); Pcs-precorneal seta(e); Pas-prealar seta(e); Dcs- dorsocentral seta(e); Sas-supraalar seta(e); Mns-metanotal seta(e) (Fig. 1). The number of setae in each group was determined and the position of the Des and Sas on the longitudinal axis measured with an occular micrometer. The longitudinal position of these setae was determined as the distance to each of the setae from the anteromedial angle of the antepronotum along a line extending to the most posterior part of the thorax (Fig. 1). Dorsoventral position was not measured and differences in size (diameter and length), branching and pigmentation were not systematically studied. RESULTS AND DISCUSSION Intraspecific variation—The validity of this analysis would be severely restricted if the number and/or location of setae varied greatly at the species W. P. COFFMAN 63 TABLE 2. Intrapopulation variation in the longitudinal position of the Dcs of Tvetenia cf. calvescens. All specimens from Linesville Creek, Pa. (position of setae expressed as percent of thorax length from the anteromedian angle). SEX DATE Des1 Des2 Des3 Des4 female 9 July 70 32.5 35.2 56.9 57.4 female 21 July 70 33.7 35.3 SWoll 59.1 male 17 Aug. 70 33.3 36.1 58.8 59.4 male 17 Sept. 70 33.4 36.1 56.1 56.5 male 17 Oct. 70 33.2 34.3 57.5 59.7 female 30 Nov. 70 33.7 34.9 Sc 58.9 female 14 Apr. 71 34.8 35.8 54.2 54.8 female 15 May 71 31.3 33.5 53.7 54.7 female 15 June 71 37.6 39.1 58.6 59.0 male 6 July 71 35.5 38.5 60.7 61.5 XK = 33.9 35.9 SA 58.1 S.D.+ 1.7 7 2.1 Dep TABLE 3. Interpopulation variation in the longitudinal position of the Dcs of Tvetenia cf. calvescens. (Position expressed as in Table 2). LOCALITY Desl Des2 Des3 Des4 Alaska 34.5 36.9 58.3 58.6 Maine 33.0 34.3 56.3 56.3 Michigan 32.1 34.2 56.6 57.6 New York 35.4 38.2 58.0 58.0 Oregon 1 34.6 37.2 56.4 57.7 Oregon 2 33.2 35.5 56.1 58.4 Pennsylvania 1 31.6 32.9 54.4 55.2 Pennsylvania 2 33.5 34.7 55.9 56.9 Wyoming 1 32.1 34.6 56.4 57.2 Wyoming 2 31.7 34.0 53.9 56.3 XK = 33.2 35.3 56.2 57.2 S.D. + 1.3 1.7 1.4 1.1 level as a function of sex, size or season. To test for such variation, ten ex- uviae of the common and widespread Tvetenia cf. calvescens, including males and females of different sizes collected in eight different months from Linesville Creek, Pennsylvania were analyzed for number of setae in each group and position of the Dcs. There was no variablility in the groups of setae or number of setae per group. The data for the position of the Dcs demonstrate that regardless of sex, size or season these setae are constant in MEM. AMER. ENT. SOC., 34 64 CHIRONOMID PUPAL THORACIC CHAETOTAXY TABLE 4. Intrageneric variation in the longitudinal position of the Dcs. (Data are mean values for position, S.D. in parentheses after each value; position expressed as in Table 2). TAXON # OF SPP. Des1 Des2 Des3 Des4 Cricotopus s.\ 9 40.6(2.1) 44 .4(4.0) 55.8(3.0) 58.0(3.6) Eukiefferiella 11 37.7(3.8) 42.6(5.7) 57.0(3.4) 61.7(3.5) Orthocladius s.1 22 40.7(2.0) 47.4(4.3) 53.3(3.2) 56.1(3.2) Pseudochironomus 6 33.7(3.2) 34.4(3.3) 57.0.1) 57.8(3.0) Cryptochironomus 5 38.6(6.1) 40.8(6.9) 52.3(6.3) 55.3(5.3) position (Table 2). An additional ten specimens of 7. cf. calvescens, col- lected from ten locations spanning 20 degrees of latitude, 80 degrees of longitude and 3000 meters in elevation, were analyzed to determine if there is a significant geographical component to thoracic chaetotaxy at the species level (Table 3). The groups of setae present and the number of setae per group did not vary in any of these specimens and the mean positions of the Des were essentially identical to the Linesville Creek values. It may be infer- red from these results, which are similar to those from other species ex- amined, that the chaetotaxy of the thorax is extremely constant at the species level. Intrageneric variation—As. might be expected, the number and location of setae does vary among the species of a genus. This may be illustrated us- ing the genera Cricotopus, Eukiefferiella, Orthocladius, Pseudochironomus and Cryptochironomus (Table 4). Most of the variations in position of the Des result from longitudinal shifts of all four or, at least,most of the four, thereby retaining their relative positions. Some species may have additions to or deletions from the ‘‘typical’? number in the other groups of setae. Basically, however, the species of a genus have the same groups of setae, the same number per group and the same relative position of the Dcs. Varia- tions in chaetotaxy may, in many instances, prove to be of taxonomic im- portance for species and generic level taxa. Variations in higher taxa—Considering the over-all ecological and struc- tural heterogeneity characteristic of most tribes and subfamilies (at least the larger ones), it is somewhat surprising to find that most members of a par- ticular taxon demonstrate a remarkable degree of similarity in groups of setae present, number of setae in each group (Table 5) and the longitudinal positions of the Dcs and, when present, Sas (Table 6). The over-all pattern of chaetotaxy for these higher taxa and the variations encountered within often appear to have taxonomic and phylogenetic significance. Trends throughout Chironomidae—Discussion of trends at the family level is based on the data contained in Tables 5 and 6. W. P. COFFMAN 65 TABLE 5. Chaetotaxy of chironomid pupae by subfamilies and tribes. (Values are number of setae per group). TAXON MAps_ LAps Pes Pas Des Sas Mns TELMATOGETONINAE 1 0 2 0 3 1 2 TANYPODINAE 1 AE Oe 0 2 1 1 Anatopyniini® 1 of 0 0 2 1 1 Coelotanypodini 1 2 0 0 D) 1 1 Macropelopiini 1 2(1) 0 0 2 1 1 Natarsini 1 2 1 (0) 2 i 1 Pentaneurini 1 2 1 0 m 1 1 Tanypodini 1 04 0 0 2 1 1 PODONOMINAE 1-2 2(3) 3(2) 0(1) 3 0-1 4(3) Boreochlini 1-2 2(3) 3 0(1) 3 0-1 4(3) Podonomini 2 2; 3(2) 0 3 0-1 4(3) APHROTENIINAE® 1 2 1(2) 0 3 1 2 BUCHONOMYIINAE 3 2 3 0 1 1 2 DIAMESINAE 1-2 1(0,2) 3(2) 0(1,2) 3(0,1,2) 1) 1(0,2,5) Boreoheptagyiini 1 2 3 2, 3 1 1 Diamesini 2-1 1 3(2) 0(1) 1-2-3 1 1-2 Heptagyini 1 1-0 2 0 0 0 0 Protanypodini 6 il? 3 0 3 1 5 PRODIAMESINAE 2 2 3 1 4 0 0 ORTHOCLADIINAE 2(3) 1-2(3) 3(2,1) 0(1,2) 4(3) 0 oc)! CHIRONOMINAE 1-2 0-1(2) 2-3 0 4(2) 0 0 Chironomini 1 0-1 2(3) 0 4(2) 0 0 Pseudochironomini 2 0 3 0 4 0 0 Tanytarsini 1 1(2) 3(2) 0 4 0 0 = Values in parentheses indicate a rare character state. Values following a dash indicate additional common character states. These are given in order of frequency of occurrence. = Data taken from the literature. = Values given may not be correct. = Protanypus possesses an additional group of setae near the LAps. = I have seen an Mns only in species of Chaetocladius. Ss » ll | mm 0 AO MAps—There is no obvious trend in the number of MAps. Considered as a whole one MAps seems to be the basic number, although there are many exceptions to this. Tanypodinae never have more than one and it is always slightly posterior of the antepronotum, a condition which appears to be apomorphic for that subfamily. Three MAps are encountered in a few Orthocladiinae and in Buchonomyiinae. Protanypus is unique with six. The number in all Chironomini and Tanytarsini examined was one, apparently representing a reduction from the presumed ancestral condition of at least two MAps found in all Prodiamesinae and Orthocladiinae. It is interesting MEM. AMER. ENT. SOC., 34 CHIRONOMID PUPAL THORACIC CHAETOTAXY 66 ‘uor}Isod psoq oY} UI SI eI9S sNOZOjOUIOY ay) A[qeUINsoid ‘seg ON “IBdS [BIOS JO UOTIISOg Il Il S90 070 “‘uontsod ,jeoidAj,, ul ses ‘uontsod , .jesidAy,, Jo 10110}Uv si seg Il ss ema 3(€'1)6'SS (7 'S)evS (S°S)6'Or (I'r)9°LE 3(8'7)9°9S (L'S)Z'€S (7'9)8 br (8°P)L6E 3(9°S)S°6S (pS) TES (L'1)0'br (€"p)7 6E q0'r)1 69 (O'r)8'Is (0'S)I°sp (Z'€)S "SE 99 $9 €°9S pe SP p6 OF Ppoinseoul 10U q(9'0)8'L9 (O'9)E°SS (€°9)0°8r (7'9)r Eb ql€ S)E99 ouou (ASYAIVS (S°€)9°9r 20'S 9°8r (Gna 7 6E SeS-psoq €soq 789d 1S9q gs AVNINONOUIHO vs AVNIGVTOOHLYO t AVNISHNVIGOUd Ol AVNISANVIG I AVNITANONOHONA = AVNIINA LOYHdV L AVNINONOGOd v7 AVNIGOdANVL I AVNINOLADOLVIN TAL VAAN) AO # NOXVL == lllaa=a_—_—=—_——__——EEEe————— ee eee ‘(p e192 Ul SB possaidxe a1¥ pu SUOTIISOd IOJ USAT 1B SUBIUI 1JOUDD) “Sar[IWeJQns prwoUO TY UI soq Jo BuIoeds [eUIPN}ISUO] pue JaqUINN “9 aTaV], W. P. COFFMAN 67 to note that Pseudochironomini is the only tribe of this subfamily to have two MAps. Given the present understanding of phylogenetic relationships — within Chironominae, this would imply either the parallel reduction to one MAps in Chironomini and Tanytarsini or a reduction to one in the ancestor of the Chironominae and a secondary increase to two MAps in Pseudo- chironomini. LAps—The number of setae in this group varies widely throughout the family and there is no obvious trend. Protanypus is unique in having an ad- ditional group of setae posterior to the LAps. Pcs—The basic number of setae in this group is two or three but a number of exceptions occur, especially in Tanypodinae which never have more than one. There is no obvious trend. Pas—The most common condition is for Pas to be absent. This is presumably the case for all of the Telmatogetoninae, Tanypodinae, Aphroteniinae, Buchonomyiinae and Chironominae. The only Podonominae that I have seen with a Pas is Trichotanypus. One Pas occurs in all Prodiamesinae and one or two Pas are found sporadically in Diamesinae and Orthocladiinae. Apparently no Chironominae have Pas. Des-Sas (Figs. 2-12)—If the most ventral seta of Telmatogetoninae is a Sas, although it is anterior of the ‘‘typical’’ position, then all Telmatogetoninae (Fig. 2), Tanypodinae (Fig. 3), many Podonominae (Fig. 4), all Aphroteniinae (Brundin, 1966; Figs. 493, 494, 515, 517), Buchonomyiinae (Fig. 5) and most Diamesinae (Fig. 6) possess a Sas. This seta is absent in all Prodiamesinae (Fig. 7), Orthocladiinae (Fig. 8) and Chironominae (Figs. 9-12). The fact that Sas never occurs in taxa with four Des is considered as evidence that Sas is homologous to Dcs4. The trend that would seem to be evident involves the movement of Sas of the plesiomorphic subfamilies into the position of Dcs4 in the more apomor- phic subfamilies. Another trend in the longitudinal distribution of the Des is the develop- ment of an arrangement of these setae into anterior and posterior pairs. This pattern is found in all Prodiamesinae (Fig. 7), many Orthocladiinae (Fig. 8) and all Chironominae (Figs. 9-12). Although the paired pattern is common in Orthocladiinae, other patterns are not rare, for example: (1) three anterior and one posterior; (2) one anterior and three posterior; (3) all four evenly and widely spaced; (4) all four grouped closely together. The two pair pattern is most developed in the Chironominae (Figs. 9-12). All of the Tanytarsini (Fig. 9) and Pseudochironomini (Fig. 10) that I have observ- ed not only have the two pair arrangement but, the members of the pairs in- sert contiguously. This condition is also found in almost all of the genera of the Chironomini of Saether (1977, Fig. 62) from Paratendipes through MEM. AMER. ENT. SOC., 34 68 CHIRONOMID PUPAL THORACIC CHAETOTAXY Fig.5 Fig.6 \ es Fics. 1-6: Fic.1. Left half of generalized chironomid thorax with groups of setae located (see text for terminology of setae and measurement technique). A = anteromedian angle of thorax; B = posterior angle of thorax; A’ = point of origin for measurement of longitudinal position of seta C: Fics. 2-6. Left half of thorax illustrating chaetotaxy: 2. Telmatogeton japonicus; 3. Macropelopia sp.; 4. Trichotanypus sp.; 5. Buchonomyia thienemanni; 6. Pagas- tia sp. (Des1-4 indicated, Sas, when present, labeled as Dcs4). Stenochironomus. It would seem, therefore, that the two pair pattern in which members of the pairs are contiguous is an apomorphic trait of Chironominae and that the secondary separation of the pair members in the genera of the Chironomus and Harnischia complexes is apomorphic for those groups. W. P. COFFMAN 69 Fig. 11 \ Fics. 7-12. Left half of thorax illustrating chaetotaxy: 7. Prodiamesa sp.; 8. Chaetocladius sp.; 9. Rheotanytarsus anomalous gr. sp.; 10. Pseudochironomus sp.; 11. Polypedilum sp.; 12. Parachironomus sp. (Dcs1-4 indicated). Mns—At least one Mns is, apparently, present in all Telmatogetoninae, Tanypodinae, Podonominae, Aphroteniinae and Buchonomyiinae. It is also present in all Diamesinae except members of the tribe Heptagyini which lack most thoracic setae. In subfamilies Prodiamesinae, Orthocladiinae and Chironominae I have seen a Mns only in species of the orthoclad genus Chaetocladius. P. Cranston (pers. communication) has seen one in a species of O. (Eudactylocladius) but I have not been able to verify this with Nearc- tic material. This trend is strengthened by the observations that the location of the Mns in Telmatogetoninae through Buchonomyiinae is on that part of MEM. AMER. ENT. SOC., 34 70 CHIRONOMID PUPAL THORACIC CHAETOTAXY the metanotum that forms the base of the haltere sheath. In Diamesinae the Mns are usually more medially located on the flat surface of the metanotum. Occasionally in Diamesinae and in Chaetocladius the Mns are much dislocated medially and have a different shape. The latter may not be homologous to the Mns of other taxa. CONCLUSIONS 1. At the species level, thoracic chaetotaxy is remarkably constant in groups of setae present, number of setae in each group and longitudinal positions of the Des. 2. At the generic level, variations in chaetotaxy may be significant from One species to another but, in general, the pattern within a genus is constant and has taxonomic value. 3. Variations in chaetotaxy at the tribe and subfamily levels may be con- siderable but each of the higher taxa has a typical pattern. These patterns have taxonomic and possible phylogenetic significance. 4. At the family level the following trends are evident: a. the movement of Sas of Telmatogetoninae, Tanypodinae, Aphro- teniinae, Podonominae, Buchonomyiinae and Diamesinae to the Dcs4 position in the other subfamilies. b. the development of anterior (Dcs1 and Dcs2) and posterior (Dcs3 and Dcs4) pairs of Dcs in Prodiamesinae, Orthocladiinae and Chironominae and, in particular, the very close proximity of the members of these pairs in Tanytarsini, Pseudochironomini and the plesiomorphic Chironomini. REFERENCES CITED COFFMAN, W.P. 1979. Neglected characters in pupal morphology as tools in taxonomy and phylogeny of Chironomidae (Diptera). In: O.A. Saether, (ed.): Recent developments in chironomid studies (Diptera: Chironomidae). Ent. scand. Suppl. 10:37-46. BRUNDIN, L. 1966. Transantarctic relationships and their significance as evidenced by chironomid midges, with a monograph of the subfamilies Podonominae and Aphrote- niinae and the austral Heptagyiae. K. Svenska Vetensk Akad. Handl. 11:1-472. SAETHER, O.A. 1977. Female genitalia in Chironomidae and other Nematocera: morphol- ogy, phylogenies, keys. Bull. Fish. Res. Bd. Can. 196:209 pp. 1980. Glossary of chironomid morphology terminology (Diptera: Chirono- midae). Ent. scand. Suppl. 14:51 pp. Chironomid Haemoglobins: Their Detection and Role in Allergy to Midges in the Sudan and Elsewhere. By P.S. CRANSTON Department of Entomology British Museum (Natural History) Cromwell Road, London SW7 5BD. ROSEMARY D. TEE Department of Allergy and Clinical Immunology Cardiothoracic Institute, Brompton Hospital Fulham Road, London SW3 6HP. P.F. CREDLAND Department of Zoology, Bedford College University of London Regent’s Park, London NWI 4NS. A.B. Kay Department of Allergy and Clinical Immunology Cardiothoracic Institute, Brompton Hospital Fulham Road, London SW3 6HP. ABSTRACT. — Studies on allergic reactions of humans to chironomids, particularly to nuisance midges in the Sudan, are reviewed. Evidence is presented that chironomid haemoglobins are im:portant allergens, a finding which indicates that chironomid midges should be seen as significant environmental allergens. INTRODUCTION Massive swarms of non-biting midges (Diptera: Chironomidae) emerge from the Sudanese Nile in the winter months (Lewis, 1956; Cranston ef al., 1981), and, according to Lewis (/oc. cit.), appear to have done so since the 1920s. The swarms which are a most serious nuisance to man and livestock, restrict outdoor activity and may, in extreme cases, cause asphyxia. These swarms consist largely of Cladotanytarsus lewisi (Freeman, 1950), although other species of Chironomidae may also occur in large numbers (Lewis, 1956; Wiilker, 1963; Cranston ef a/., 1981). In order to understand the biology of these nuisance midges, Lewis (1956, 1957) and Lewis ef al., 71 MEM. AMER. ENT. SOC., 34 72 CHIRONOMID HAEMOGLOBINS (1954) studied aspects of Nilotic chironomid life histories and fluctuations in numbers of adult midges attracted to lights. Attempts to control the midges using DDD and DDT as larvicides were reported by Brown et al. (1961), but fish mortality was high and the subsequent reduction in midge numbers cannot be interpreted as effective insecticidal control without detailed knowledge of normal population fluctuations. This need for “thorough baseline studies’’ was observed by Wiilker (1963) when in- vestigating biological control as a possible alternative mechanism for alleviation of the midge problem. Apart from the nuisance caused by midge swarms, the association of human allergic reactions, such as asthma and allergic rhinitis with seasonal midge emergence has long been suspected (Kirk, 1952, 1953). In more recent years detailed investigations into the nature and extent of chironomid allergy in the Sudan have taken place. Particular emphasis has been given to the relationship between the epidemiological, entomological and im- munological aspects of the problem. The nature and results of these multidisciplinary investigations are reviewed below. These studies, together with those of other researchers, indicated that chironomid haemoglobins are important allergens for humans. Therefore, the second part of the paper details our investigations into the role of these haemoglobins in the Sudanese midge allergy and discusses the belief that chironomid midges are more than a local Sudanese problem, being potentially a world-wide cause of allergy. BACKGROUND Strong evidence for the association between midges and seasonal human allergic reactions was provided by Kay ef al. (in press a), who compared the prevalence of allergic symptoms in two Sudanese villages. An epidemiological survey was made of the population of Kalakla, a Nilotic village with midge problems, and contrasted with a similar survey of Umm Dawa Ban, a desert village some 40 kms east of Khartoum, distant from the Nile and without midge problems. The results indicate that allergic rhinitis occurred at a rate of 6.7% in Kalakla and 1.5% in Umm Dawa Ban. The percentage of those surveyed with asthma in addition to allergic rhinitis, was four times greater in Kalakla than in the control village. The sufferers’ own assessment of the provoking agents indicate that winter seasonal ex- posure to chironomid midges was a major aetiological factor in asthma and rhinitis in Kalakla. Kay ef al., (in press) concluded that repeated exposure to chironomids results in a very high incidence of allergic rhinitis, as well as in- creasing significantly the indigenous asthmatic population. P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY U3; The geographical distribution of midge induced allergy was considered by Lewis (1956) who recorded the problem in Khartoum and Wadi Halfa, and by Satti & Abdel Nur (1974) who suggested that problems of nuisance midges and allergy might occur as far north as Lake Nasser (Lake Nubia or Lake Aswan). In an attempt to obtain further information on the geographic extent of hypersensitivity to chironomid midges, Cranston et al. (in press) performed skin tests with unfractionated (‘crude’) C. lewisi ex- tracts on asthmatic subjects living close to the Nile in Sudan and Egypt. Hypersensitive individuals were found on the White Nile as far south as Kosti, on the Blue Nile as far east as Sennar, and on the Nile as far north as Aswan, Luxor and Qena in Upper Egypt (Cranston ef al., in press). The problem appears not to occur in middle and Upper Egypt (Dr. Soliman Diaa el Din, pers., comm.; ms in prep.). The evidence of the epidemiological survey of Kay ef a/. (in press) and the geographical distribution studies indicate that the exposed population numbers hundreds of thousands, and the number expected to suffer from midge related allergic problems must be in the tens of thousands. C. lewisi is always the dominant species of chironomid in samples of midges collected in the areas of Sudan most affected by nuisance midges. Four species, Dicrotendipes fusconotatus (Kieffer), Conchapelopia cygnus (Kieffer), Procladius noctivagus (Kieffer) and Nanocladius vitellinus (Kief- fer), were also present in all catches. Although these subdominant species underwent similar daily changes in abundance in light-trap catches as did C. lewisi the greater the total daily catch the greater was the proportion of C. lewisi until in the largest catches of over a quarter of a million individuals per trap per night, the proportion of C. /ewisi was over 95% of the total catch (Cranston et a/., 1981). Lewis (1956, 1957) made the only detailed study of the biology of C. lewisi and observed that the midge nuisance seemed to be associated with the construction of dams and the subsequent increase in lacustrine condi- tions in the Nile. Cranston ef al. (1981) confirmed and strengthened this hypothetical relationship and suggested that the summer seasonal rains in the catchment areas of the White and Blue Niles caused a natural eutrophication of the river by washing in of plant nutrients, particularly nitrates, phosphates and silicates. After the period of maximum flow in the river, turbidity decreases in the extensive slow-flowing areas caused by natural and man-made damming of the Nile. As a result, the increased light and high plant nutrient levels allow abundant algal and diatom growth which provides a food source for many chironomid larvae. C. /ewisi larvae, shown to be benthic grazers of diatoms and algae and presumed to have a rapid life cycle, make maximum use of this seasonally abundant food MEM. AMER. ENT. SOC., 34 74 CHIRONOMID HAEMOGLOBINS resource. This results in a subsequent large increase in adult midge numbers. Cessation of midge emergence in the Spring seems to coincide quite closely with the crash in algal numbers following nutrient depletion. Further evidence supporting this theory is seen when altered hydrological conditions occur, such as reduced or sporadic rainfall or alterations in the flow regime of the Nile. These factors which occurred in the period 1980-82 were associated with a reduction of the duration and severity of the midge season. It is significant that the factors which lead to the development of very high populations of midges in the Sudan are not unique to this ecosystem, but occur in many bodies of water throughout the world. Further evidence for the association between chironomid midges and allergic reactions in Sudanese people has been provided by immunological investigations. Comparison of patients’ symptomatology (severity of bron- chial asthma and/or allergic rhinitis) with daily numbers of midges assessed by light-trap catches, indicated a relationship. During periods of minimal midge emergence allergic symptoms tended to be reduced, but increased during moderate to large emergences of up to 400,000 midges per trap per night. A massive emergence comprising over 99% C. /ewisi during December 1979, not sampled quantitatively but estimated to contain the equivalent of well in excess of half a million midges, was clearly associated with severe signs and symptoms of immediate-type hypersensitivity, in- cluding bronchospasm and rhinitis (Kay ef al., 1983). Kirk (1952) showed that a number of Sudanese bronchial asthmatics gave a high proportion of strongly positive responses when skin tested with crude extracts of midges. This skin test reactivity was confimed by Kay ef al. (1978) who also showed that sera from C. /ewisi-sensitive Sudanese caused passive sensitisation of lung fragments which led to IgE (immunoglobulin E) — mediated release of histamine and SRS-A (slow reacting substance of anaphylaxis). The development of a radioallergosorbent test (RAST) by Gad El Rab & Kay (1980) allowed quantitative measurement of specific IgE present in serum, and was used to demonstrate a relationship between RAST scores and the severity of patients’ symptoms. In order to test if C. Jewisi alone was responsible for midge hypersensitivi- ty in the Sudanese, patients previously shown to be hypersensitive to ex- tracts of pure C. /ewisi were skin ‘prick’ tested with extracts of seven of the subdominant species of Nilotic Chironomidae. The results indicated that C. lewisi was the most important species, but that there was limited cross- reactivity with some other species, particularly Dicrotendipes fusconotatus, Procladius noctivagus and Conchapelopia cygnus (Cranston ef al., in press). P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 75 Confirmation of the aetiological role of chironomid midges, particularly C. lewisi, in allergic problems in the Sudan, led to attempts to characterise the major allergens involved (Gad El Rab, Thatcher & Kay, 1980; Tee et al., in press). Fractions from Sephadex G100 gel filtration and ion exchange chromatography were assayed for antigenicity by skin testing hypersensitive individuals. These techniques indicated that a major proportion of the allergenic material was associated with a molecular weight of 15-20,000 daltons and a pl of 4.3. More definitive techniques, still using Sephadex G100 gel filtration but including RAST inhibition and autoradiography with '**I-anti-IgE, showed that the ‘major peak’ of allergenicity was associated with molecules of approximately 17,000 daltons and with a pl range of 3.5 to 5.5. The major allergens from C. /ewisi therefore appear to be a group of closely related acidic peptides. Contemporaneously with the revival of interest in midge allergy in the Sudan, Baur and his colleagues in West Germany were investigating the in- cidence of hypersensitivity and respiratory allergy amongst workers occupa- tionally exposed to freeze-dried larvae of Chironomus riparius (cited as Chironomus thummi thummi (CTT), but see synonymy by Credland, 1973). Using RAST and RAST inhibition, Baur ef al. (1982) showed that the an- tigenic determinants were sited within peptide sequences in some of the 11 polymorphic forms of haemoglobin present in CTT, and that specific an- tibody against haemoglobins accounted for a proportion of the total IgE of hypersensitive individuals. The first indications that there might be antigens common to both the German occupational allergy and to the Sudanese en- vironmental allergy came from Baur (1982) and Baur ef al. (1982). They found that the sera of Sudanese hypersensitive to C. /ewisi gave positive RAST results in tests with CTT larvae and adults, with isolated total CTT haemoglobin and with one of the antigenic peptide sequences. The implications of an allergen or group of allergens common to two genera of chironomids which are phylogenetically distantly related, are con- siderable, in view of the ubiquity and abundance of the Chironomidae. Thus it was important to establish whether haemoglobin was one of the ma- jor allergens in the Sudanese midge allergy. It became necessary to extend the range of diagnostic skin tests on C. /ewisi sensitive individuals to try to test these ideas. Extracts were prepared from larvae and pupae of C. /ewisi and from larvae and adults of Ch. riparius (= CTT) and used in skin testing together with a haemoglobin extract from CTT provided by Dr. X. Baur, and with four allergenic fractions derived from Sephacryl S200 by gel filtra- tion and rechromatography from C. /ewisi extract. The methods of prepara- tion of the extracts, results of skin tests and interpretation of the results are presented below. MEM. AMER. ENT. SOC., 34 76 CHIRONOMID HAEMOGLOBINS Since the results are dependent on quantification of skin ‘prick’ test responses, it was necessary to establish the relationship between these responses and the actual levels of specific anti-C. /ewisi IgE present in patients’ sera. RASTs were performed on the serum from each patient following a skin test with unfractionated C. /ewisi extract and the relation- ship examined. MATERIALS AND METHODS Radioallergosorbent test (RAST). — A relationship between the percentage binding of '?*I-anti-IgE to allergen polymer complex (the RAST value) and the severity of clinical symptoms of C. /ewisi hypersensitive individuals was demonstrated by Gad El Rab & Kay (1980). In order to demonstrate whether there was a similar relationship between the severity of skin test response and the specific IgE directed against C. /ewisi antigen(s), RAST tests were performed on the sera of 40 Sudanese individuals previously shown to be skin test sensitive to unfractionated C. Jewisi. A group of 24 skin test negative individuals from the United Kingdom, who had not been exposed to C. /ewisi, were selected as controls, and their sera were similarly subject to RAST testing. The basis for this method of allergen determina- tion is given by Wide ef al. (1967). (i) Preparation of Allergen Polymer Complex (APC). — Ten mg of lyophilised, unfractionated C. /ewisi extract was coupled with each gram of CNBr-activated Sepharose 4B (Pharmacia). (ii) RAST assay. — Optimum conditions for the RAST assay were deter- mined as two 16 hour incubations with SOul of serum and 100 pul of APC (6.25% concentration) in the first incubation, and 50 pl '?*I-anti-IgE (Phar- macia) in the second. Cord blood sera with no demonstrable IgE were used in all assays as negative controls. Skin ‘prick’ testing. — Patients shown previously to be hypersensitive to unfractionated C. Jewisi extracts by skin test or RAST test, or both, and who had been recruited for clinical trials in Khartoum, Soba or Kalakla clinics, were selected for two further series of tests. In one group 16 patients were tested with unfractionated extracts of C. /ewisi adults, pupae and lar- vae. In another group, 26 different patients were tested with unfractionated C. lewisi adult extract, four S200 fractions of adult C. /ewisi, three unfrac- tionated extracts of Chironomus riparius and with Ch. riparius haemoglobin. Skin ‘prick’ tests were performed on the volar region of the forearm, and maximum and minimum diameters of any resultant weals were measured 15 minutes after the test. All patients selected gave a P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 77 response of at least 4mm by 4mm to the histamine control and of at least 2mm by 2mm to the unfractionated C. /ewisi. Patients who gave any signifi- cant response to the negative control were not selected for further tests. All skin ‘prick’ test responses were calculated as a weal ‘area’ by multiplying the maximum and minimum diameters of each weal and sub- tracting any weal ‘area’ provoked by the negative control. In 23 cases from the group of 26, skin ‘prick’ tests were duplicated in tiie reverse sequence on the opposite arm; in these cases the average response to each antigen was calculated. Antigenic material for the skin tests was obtained as follows: (i) Collection of midges. — Adult midges were collected by light-trap at Kalakla in the Sudan and identified as comprising over 99% C. lewisi. These were dried at 26°C and stored at 4°C in sealed plastic bags. Larvae and pupae of Nilotic Chironomidae were collected by aquatic drift nets par- tially immersed in the White Nile at Khartoum. The samples, comprising up to 75% C. lewisi, were dried on filter paper at 26°C and stored at 4°C. Larvae and adults of Chironomus riparius were collected from cultures maintained by Dr. P. Credland and dried as before. A commercially available tropical fish food containing predominantly Ch. riparius larvae, killed by ultraviolet radiation, packed by ‘‘Gamma Foods’’, was purchased from an aquarists’ shop and dried as above. The provenance of these larvae is unknown. (ii) Preparation of extracts. — A known weight of each sample of adult, pupal or larval midge was defatted in three changes of ether over 24 hours. Extractions were performed in Coca’s solution (Sg sodium chloride, 2.75g sodium bicarbonate and 4g phenol, made up to 1 litre with distilled water) for 48 hours at room temperature. The midges were removed by filtration and the filtrate centrifuged at 18,000g for 40 minutes. The supernatant was passed through a 0.45 Millipore filter (Millipore Ltd., Bedford, U.S.A.), dialysed against six changes of distilled water over two days and lyophilised. The material used in skin testing was prepared at 1 mg ml™ in 50% Coca’s solution: 50% glycerol, and passed through a 0.45 Millipore filter into a glass bottle with plastic cap and applicator (Bencard Ltd, Worthing, England). Histamine at 1 mg ml"! was used as a positive control solution and 50% Coca’s solution: 50% glycerol as a negative control. Crude Ch. riparius (CTT) total haemoglobin sent by Dr. X. Baur, had been extracted from larvae following the technique of Baur ef a/. (1982), and was prepared at 0.1 mg ml”. (iii) Preparation of allergenic fractions. — Lyophilised C. /ewisi extract, prepared according to the method outlined above, was applied in four separate batches of 200 mg to a Sephacryl S200 (Pharmacia, Uppsala, Sweden) column (90cm x 2.6 cm diameter, in 0.05M NH, HCOs, at 29.5 ml MEM. AMER. ENT. SOC., 34 78 CHIRONOMID HAEMOGLOBINS >10°K 67K 43K 25K 12-4K 14K | ahicles) ones aaa! 1 1 1 it 3) Pe x Heise ade | eve ie eso | ( ! iol tae! H o N >) + 7) Cc § 2 =, (67-6) | uv 4 [jas fos) (d) 0 10 20 30 40 50 60 70 80 90 100 110 120 Fraction number FIGURE 1. Optical densities of C. /ewisi extracts before and after rechromatography of fractions: a) 200 mg of unfractionated C. /ewisi extract; (b), (c), (d) & (e) rechromatography of fractions 1, 2, 3 & 4 respectively from four fractionations of (a). Numbers in parentheses show relative magnitude of skin test weal response as indicated in the text. [Sephacryl S200 gel filtra- tion, column calibrated with markers of known molecular weight from 1.4K to 10° daltons.] Hr“'). The column was calibrated with blue dextran (> 10°K), bovine serum albumin (67K), ovalbumin (43K), chymotrypsin (25K), cytochrome C (12.4K) and vitamin B12 (1.4K). Four fraction areas were identified by O.D. profile at 280 nm (Fig. 1a), which demonstrated that Sephacryl S200 gave better resolution than Sephadex G100. The Kav of the two median fraction peaks (O.D. at 280 nm) (labelled 2 and 3 in Fig. 1) were 0.307 and P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 79 C. lewisi hypersensitive U.K. controls 1251 anti -IgE binding (%) FicuRE 2. Comparison of RAST and skin test weal response to C. /ewisi. A control skin test with Coca’s solution was performed at the same time as the C. /ewisi skin tests. Any resul- tant weal ‘area’ was deducted from that produced by the test extract. 0.418 respectively, equivalent to molecular weights of approximately 32,000 and 17,000 daltons. Each of the four fraction areas from each of the four fractionations of extract were pooled individually and purified further by rechromatography on the S200 column as shown in Fig. 1b, c, d, & e respec- tively. After harvesting, these four fractions were lyophilised and used for skin testing. RESULTS Radioallergosorbent test (RAST). — After establishing the optimum condi- tions for the maximal binding of IgE in the RAST, the percentage of 5T-anti-IgE binding was determined for the serum of each of the 40 hypersensitive Sudanese and 24 United Kingdom controls. The relationship MEM. AMER. ENT. SOC., 34 80 CHIRONOMID HAEMOGLOBINS Weal areas (sq.mm.) mm a ean C. lewisi extracts (img/ml) Controls FicurE 3. Skin test weal response to adult, larval and pupal extracts of C. Jewisi. Responses to the negative (Coca’s solution) and positive (histamine) controls are shown. between this percentage of isotope binding and the skin test reactivity of each individual, with the skin test response aggregated into three size categories, is illustrated in Figure 2. Twenty-four United Kingdom skin test negative controls gave between 0.6 and 2.6% isotope binding. Fifteen skin test reactive Sudanese with weal ‘areas’ (maximum xX minimum weal diameters) of 1 - 20 mm? gave RAST values from 4 to 21% binding, and pa- tients with stronger skin test responses of 21 - 80 m? had RAST values of between 17 and 27% binding. Eight individuals with very strong skin test reactions from 81 to 180 mm? did not give RAST values higher than those individuals giving intermediate (21 - 80 mm?) size weals. All values were Statistically highly significant: 1 - 20 mm? weals versus 0 mm? weals (p<0.005), 1-20 mm? weals versus 21-80 mm? weals (p<0.025), 21 - 80 mm? weals versus 81 - 180 mm? weals (not significant). Skin ‘prick tests. — The skin ‘prick’ test responses, as weal ‘areas’ of each of 16 Sudanese patients in the group tested with adult, pupal and larval ex- tracts of C. lewisi, Coca’s glycerol negative control and histamine positive control, are illustrated in Figure 3. The skin test responses, as weal ‘areas’, of the other group of 26 Sudanese patients tested with the four rechromatographed S200 fractions of C. lewisi are shown in Figure 4, and the responses of the same individuals to four ex- tracts of Ch. riparius (= CTT) are shown in Figure 5. The responses of these 26 individuals to the Coca’s glycerol negative control and to the P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 81 histamine positive control were similar to those of the 16 patients of the first group shown in Figure 3. The range of responses to the negative control is shown in Figure 5, but omitted entirely from Figure 4. Each skin ‘prick’ response to these eight antigens (as measured in 26 patients) was stand- ardised by conversion to a proportion of the response to unfractionated C. lewisi adult extract. In order to investigate possible differences between responses to antigens tested, a Fisher-Yates analysis of variance test was ap- plied to these data. The results, ranked from high to low response, are as follows: The response to allergen fraction 2 is significantly greater than to allergen fraction 3 (p< 0.0001), which is significantly greater (p< 0.0001) than to Ch. riparius adult, laboratory and commercial larval extracts, none of which differ significantly from each other. Ch. riparius responses are significantly greater than to the high molecular weight allergen fraction 1 (p<0.005). The response to the low molecular weight fraction 4 is significantly lower than to any other allergen tested (p< 0.0001). In another analysis the weal data from C. Jewisi extract and the four rechromatographed $200 allergen fractions were ranked and mean ‘scores’ rescaled to a mean value of 50. The Fisher-Yates analysis was applied where, with rescaled data, a difference of 10 is significant at the 0.1% level. The ‘score’ for unfractionated C. /ewisi was 62.3, and the scores for S200 rechromatographed fractions 1, 2, 3, & 4 (Fig. 1) were 37.1, 67.6, 59.3 and 27.1 respectively (shown in Fig. 1 in parentheses under each peak). Thus the response to fraction area 2 is significantly greater than to un- fractionated C. /ewisi and fraction area 3 (p< 0.005), which are significantly greater than to the high molecular weight fraction 1 (p< 0.0001), which in turn is significantly greater than to the molecular weight fraction 4 (p< 0.0001). DISCUSSION The responses of Sudanese individuals hypersensitive to C. /ewisi, when tested with ten different chironomid antigenic extracts, give a strong indica- tion that haemoglobins are major allergens in chironomid allergy in the Sudan. The four rechromatographed antigenic fractions elicited different responses which help to elucidate the nature of these allergens. The allergen area 2, with a molecular weight of approximately 32,000 daltons, elicited a significantly greater response than did the unfractionated extract. The allergen area 3, equivalent to a molecular weight of about 17,000, provoked a response equivalent to that of the unfractionated extract. The major allergenic activity is clearly associated with these two fractions. The higher molecular weight fraction 1 (molecular weight 66,000 and above) and the lower molecular weight fraction each elicited smaller responses, none higher MEM. AMER. ENT. SOC., 34 82 CHIRONOMID HAEMOGLOBINS 5 150 125 100 Weal areas (sq.mm.) un 50 25 C. lewisi extract $200 fractions FicurE 4. Skin test weal response to ‘crude’ (unfractionated) and four rechromatographed fractions (Fig. 1 b, c, d & e) of C. /ewisi. Any response to negative (Coca’s solution) control is deducted. than the response to unfractionated extract. However, in some individuals, there were weak but positive responses, particularly to the higher molecular weight fraction. Although the precise natures of fractions 2 and 3 are not fully elucidated, it is significant that the molecular weights of these frac- tions coincide quite closely with those reported for dimeric and monomeric P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 83 Skin prick tests 26 patients 300 : 200 2 : og 77) e A #4 e se ee eal we Ss 100 e ) e rs ° ° ° : = ~ ese °° of? eee $ o%e 8.8 8 oc e ° x ofe $ M te e ane o5° 8 1 (Vises Et Be ee eee, Laman || cates ---23f areas C.lewisi Ch.riparius Chriparius Ch.riparius Ch. riparius adult adult larva (lab) larva (comm) pp.Hb. 1mg/ml img/ml 1mg/ml 1 mg/ml 0-1 mg/ml FicurE 5. Skin test weal responses to unfractionated C. /ewisi and extracts from Ch. riparius. Abbreviations: lab.-laboratory culture; comm.-commercial, ‘‘Gamma Foods’’; pp. Hb.-partially purified total haemoglobin (from Baur). Note: Haemoglobin is 1/10 dilution of other extracts. The interrupted lines delimit the range of responses to the Coca’s negative con- trol. haemoglobins. The high molecular weight fraction contains some molecules with a molecular weight similar to that of tetrameric haemoglobin. Biochemical studies by Tee ef a/. (1983, in press) on the structure of the allergens are not in conflict with this identification. Further indications that haemoglobins are major antigens in the C. /ewisi allergy come from skin tests performed with extracts of Ch. riparius and of the immature stages of C. lewisi. Ch. riparius has been shown to contain po- tent allergens for people occupationally exposed to freeze dried larvae used by aquarists. The antigenic activity has been identified as belonging to specific peptide sequences within the different haemoglobins present in the haemolymph of the larvae, and apparently present also in the adult midges (Baur ef al., 1982). Our tests reported in this paper indicate that there is an MEM. AMER. ENT. SOC., 34 84 CHIRONOMID HAEMOGLOBINS appreciable degree of cross-reactivity between the antigens of Ch. riparius and those of C. /ewisi, evidenced by skin reactivity to extracts of both larval and adult Ch. riparius. Since Ch. riparius is a species unknown in the Afrotropical region and it is inconceivable that Sudanese sensitized to C. lewisi could have encountered Ch. riparius prior to skin testing, their skin test reactivity to Ch. riparius extracts is most likely to be due to the presence of similar antigens in these distantly related species. The presence of one or more antigens common to unrelated species of Chironomidae may be the explanation for the cross reactivity observed between subdominant species of Nilotic Chironomidae reported by Cranston ef al. (in press). That these antigens include haemoglobins is confirmed by the response of C. lewisi sensitised Sudanese to tests with the Ch. riparius haemoglobin supplied by Dr. X. Baur. Although this antigen was prepared at 1/10 the dilution of all others tested, a strong positive response was elicited in several of the pa- tients tested, and weaker responses in many others. Since the occupational sensitisation reported by Baur was to antigens shown to be present in larval Chironomidae, it was important to test whether the immature stages of C. /ewisi were antigenic. That numerous in- dividuals did show reactivity to extracts of larvae, and to a lesser extent to the pupae, can be explained most parsimoniously as being due to the presence of similar antigens in all stages of this holometabolous insect. This conclusion is strengthened by Baur’s (1982) discovery that both larvae and adults of Ch. riparius contain similar antigenic determinants, namely haemoglobins. Further evidence for the close similarity between the antigens present in Ch. riparius and C. lewisi comes from RAST tests. Baur (1982) reported that the sera of C. /ewisi sensitised patients contained high titres of Ch. riparius haemoglobin specific antibodies. Conversely, the serum of one Ch. riparius sensitised individual contained antibodies specific to extract of C. lewisi. However, Tee (R.D. Tee, unpubl. obs.) found that there was variable inhibition of the C. lewisi RAST by Ch. riparius total haemoglobin provided by Baur, indicating that although haemoglobins are clearly signifi- cant in the problem, they may not account for the total allergenicity in all C. lewisi sensitive individuals. The considerable amount of variability between individual responses to skin tests may similarly indicate that the antigens of C. lewisi may not be completely identical to those of Ch. riparius. Whether one would expect total inhibition of the RAST by haemoglobins from distantly related species of Chironomidae is an open question. Although high immunological cross-reactivity between haemoglobins of different species of Chironomus was demonstrated by Baur ef al. (1982) and Tichy et al. (1982), the species examined belong to the same genus, while C. Jewisi P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 85 and Ch. riparius are in different tribes, although belonging to the same sub- family. ; If, as seems likely, chironomid haemoglobins are important allergens in the Sudanese midge allergy, as in the Ch. riparius occupational allergy, ques- tions remain unanswered concerning the mechanism by which the haemoglobin is available as a rapid acting allergen associated with adult C. lewisi. Susceptible Sudanese individuals may have an asthma attack trig- gered by contact with a single midge and the onset of the attack may be im- mediately on contact with the whole fly. The limited amount of information available, based on studies of one or two species of the genus Chironomus, suggests that the larval haemoglobin which is present in the haemolymph of many Chironomidae, may not be fully broken down during metamor- phosis, but may persist in the pupa and, to a lesser extent, in the adult (Schin et al., 1974). Laufer & Poluhowich (1971), studying Chironomus pallidivittatus, showed that haemoglobin, and what appeared to be breakdown products of haemoglobin metabolism, were present in the meconium of the newly emerged adult midge. Further investigations are re- quired to establish whether haemoglobins and related antigenic molecules are present in the meconia of the newly emerged (and short-lived) adult C. lewisi and whether this might be a significant mechanism for the dispersal of the antigens. Although haemoglobins are believed to be present in many Chironomidae, particularly in the subfamilies Chironominae and Tanypodinae, precise information on the distribution within the family is missing. Detailed understanding of the structure of chironomid haemoglobins is restricted almost entirely to the genus Chironomus. The in- adequacy of our knowledge concerning the importance and prevalence of these antigenic molecules restricts our ability to speculate on the extent of the problem of chironomid allergy. However, in reporting their findings on the cross-reactivity between Ch. riparius and C. /ewisi, Baur and his col- leagues (1982) suggest that ‘‘the increase in asthmatic diseases reported by several authors in water-rich regions such as river basins during the chironomid season is predominantly caused by sensitisation to haemoglobin molecules in this insect family.’’ Our investigations tend to support this sug- gestion, and, in view of our field-based studies, indicate the potentially world-wide nature of this environmental allergen. We have suggested that the environmental conditions which allow the development of huge popula- tions of adult midges along the River Nile are not unique to this region. In- deed, Ali (1980), in documenting nuisance midge outbreaks and their con- trol, lists eight countries which have been afflicted with midge outbreaks serious enough to have been documented in the literature. Our own obser- MEM. AMER. ENT. SOC., 34 86 CHIRONOMID HAEMOGLOBINS vations, together with those of colleagues studying chironomids, show that problems of nuisance chironomids are steadily increasing as waters become more eutrophic, and man lives closer to such habitats. Ali reviews in some detail the midge nuisance in central Florida, and now, for the first time, there is evidence (through responses to skin ‘prick’ test) that some in- habitants of this region have become hypersensitive to chironomid midges (Cranston, unpubl. obs.). In view of these findings, chironomids must be seen as more than a source of world-wide nuisance causing economic problems through the defacing of buildings and paintwork, traffic hazards and restriction of outdoor activity, but should be seen as significant environmental allergens. ACKNOWLEDGEMENTS The authors are grateful to Professor Hashim H. Erwa, Dr. M.O. Gad El Rab and Mr. Abdel Moneim of the Clinical Immunology Unit, Department of Medical Microbiology & Parasitology, University of Khartoum, Sudan, for encouraging this study, allowing access to facilities and giving skilled assistance in the Sudan. We also wish to thank Ms. M. Rehahn, Group Statistician in the Cardiothoracic Institute, Brompton Hospital, for her ad- vice on statistics and for undertaking the statistical analyses. This work was supported by the Wellcome Trust. REFERENCES Au, A. 1980. Nuisance Chironomids and Their Control: a Review. Bull. ent. Soc. Am. 26:3-16. Baur, X. 1982. Chironomid hemoglobin — a major allergen for humans. Chironomus 2(3):24-25 Baur, X., M. DeEwair, G. FRUHMANN, H. ASCHAUER, J. PFLETSCHINGER, AND G. BRAUNITZER, G. 1982. Hypersensitivity to chironomids (non-biting midges): localisa- tion of the antigenic determinants within certain polypeptide sequences of hemoglobins (erythrocruorins) of Chironomus thummi thummi (Diptera). J. Allergy Clin. Immunol. 68;66-76. Brown, A.W.A., D.J. McKINLEY, H.W. BEDFORD AND M. QUTUBUDDIN. 1961. Insecti- cidal operations against chironomid midges along the Blue Nile. Bull. ent. Res. 51:789-801. CRANSTON, P.S., M.O. GAD Et RAB AND A.B. Kay. 1981. Chironomid midges as a cause of allergy in the Sudan. Trans. R. Soc. trop. Med. Hyg. 75(1):1-4. CRANSTON, P.S., R.D. TEE, M.O. GAD Et RAB AND A.B. Kay. (in press). Immediate-type skin reactivity to extracts of the ‘green nimitti’ midge, (C/adotanytarsus lewisi) and other chironomids in asthmatic subjects in the Sudan and Egypt. Ann. trop. Med. Parasit. CREDLAND, P.F. 1973. The taxonomic status of Chironomus riparius Meigen and Chirono- mus thummi Kieffer (Diptera: Chironomidae). J. nat. Hist. 7:209-216. P.S. CRANSTON, R.D. TEE, P.F. CREDLAND AND A.B. KAY 87 FREEMAN, P. 1950. A species of Chironomid (Diptera) from the Sudan suspected of causing asthma. Proc. R. ent. Soc. Lond. (B) 19:58-59. Gap Ex Ras, M.O. AND A.B. Kay. 1980. Widespread immunoglobulin E-mediated hyper- sensitivity in the Sudan to the ‘green nimitti’ midge, Cladotanytarsus lewisi (Diptera: Chironomidae). I. Diagnosis by radioallergosorbent test (RAST). J. Allergy Clin. Im- munol. 66(3):190-197. Gap Ext RaB, D.R. THATCHER AND A.B. Kay. 1980. Widespread IgE-mediated hyper- sensitivity in the Sudan to the ‘green nimitti’ midge, Cladotanytarsus lewisi (Diptera: Chironomidae). II. Identification of a major allergen. Clin. Exp. Immunol. 41;389-396. Kay, A.B., M.O. Gap Et Ras, J. STEWART AND H.H. ErRwA. 1980. Widespread IgE- mediated hypersensitivity in Northern Sudan to the chironomid Cladotanytarsus lewisi (‘‘green nimitti’’). Clin. Exp. Immunol. 34:106-110. Kay, A.B., M.O. Gap Et Ras, P.S. CRANSTON, C.M.U. MACLEAN AND R.D. TEE. 1983 Clinical, Entomological and Epidemiological studies on Allergy to a Non-biting midge (Cladotanytarsus lewisi) (green nimitti). Clin. Allergy Suppl. Proc. XI Int. Congr. Allerg. & Clin. Immunol. pp. 463-467, Macmillan, Basingstoke 1983. Kay, A.B., C.M.U. Macrean, A.H. WILKINSON AND M.O. Gap Er Rap. (in press). The prevalence of asthma and rhinitis in a Sudanese community seasonally exposed to a potent airborne allergen (the ‘green nimitti’ midge, Cladotanytarsus lewisi). J. Allergy Clin. Im- munol. Kirk, R. 1952. Report of Medical Services, Ministry of Health, Sudan Government [ Khartoum]. 1950/1:44-45. : 1953. Report of Medical Services, Ministry of Health, Sudan Government [Khartoum]. 1951/2:43. LAUFER, H. AND J. PoLtuHowicH. 1971. A factor controlling the concentration of hemo- globins of Chironomus during metamorphosis. Limnologica, Berlin 8:125-126. Lewis, D.J. 1956. Chironomidae as a pest in the Northern Sudan. Acta trop. 13:142-158. 1957. Observations on Chironomidae at Khartoum. Bull ent. Res. 48:155- 184. , A.J. HENRY AND D.N. GRINDLEY. 1954. Daily changes in the numbers of chironomid midges at Khartoum. Proc. R. ent. Soc. Lond. (A) 29:124-128. Satt1, M.H. AND O.M. AspeL Nur. 1974. Chironomidae as a cause of nuisance and allergic manifestation. Sudan med. J. 12:77-80. Scuin, K.S., J.J. PoLuHowicu, T. GAMo AND H. LAuFER. 1974. Degradation of haemo- globin in Chironomus during metamorphosis. J. Insect Physiol. 20:561-571. TEE, R.D., M.O. Gap Et Ras, P.S. CRANSTON AND A.B. Kay. 1983. Allergens of the ‘‘green nimitti’’? midge. Clin. Allergy Suppl. Proc. XI Int. Congr. Allergy & Clin. Immunol. pp. 541-544. Macmillan, Basingstoke, 1983. Tee, R.D., J.L. LoncBortoM, P.S. CRANSTON AND A.B. Kay. (in press). Partial characterisation of allergens associated with hypersensitibity to the ‘‘green nimitti’’ midge (Cladotanytarsus lewisi, Diptera: Chironomidae) Clin. Allergy. Ticuy H., T. KLEINSCHMIDT AND G. BRAUNITZER. 1981. Studies on the Evolutionary Relationships Between Hemoglobins in Chironomus pallidivittatus and C. tentans. I. Isolation and Immunological analysis of Monomeric and Dimeric Hemoglobins. J. mol. Evol. 18:9-14. WipE L., H. BENNICH AND S.G.O. JoHANssSoN. 1967. Diagnosis of allergy by an in-vitro test for allergen antibodies. Lancet 2:1105. WULKER, W. 1963. Prospects for biological control of pest Chironomidae in the Sudan. W.H.O./E.B.L. Series 11, Vector Control 42:1-23. MEM. AMER. ENT. SOC., 34 mr vii ee Seat A Description of Two New species of Tanypodinae (Diptera: Chironomidae) from North Africa. COLETTE DOWLING Zoology Dept. University College Dublin, Ireland ABSTRACT. — The male imagines of a new species of the Genera Thienemannimyia and Rheopelopia are described. The species were collected in North Africa. The pupal exuvium of the Rheopelopia species is known. The species are illustrated. INTRODUCTION The Tanypodinae of the ““Thienemannimyia series’’ can be distinguished by the presence of a simple, but distinct lobe on the inner edge of the gonocoxite (Fittkau, 1962). Material collected in North Africa, which was made available to the author by Dr. E.J. Fittkau and Dr. F. Reiss, was found to contain a number of specimens from this series. A description of two new species from the collections is presented in this paper. The ter- minology used follows that of Saether (1980). Thienemannimyia choumara sp.n. Imago male (n=3). Length 3.75-4.00 mm. Head: AR 1.4-1.5: Antenna with 13 flagellomeres. Proximal end of last flagellomere darkened. Palpal segments pale, whitish-yellow. Clypeus dark yellow. Thorax: Color yellow; scutal stripes mid-brown; Wing: — length 2.3 mm; membrane very pale, cross-vein RM and FCu with yellow markings. Legs, pale yellow with darker ring at tip of femora. Leg measurements as follows (um):— Fe Ti Ta Ta, Ta; Ta, Ta; LR P, 930- 820 970 680 360 — — — 0.7 Pi 830- 800- 430- 230- 200 150 100 0.2-0.3 850 820 440 240 — — — Pin 78- 104- — _— — _— — — 80 107 — — — _ — — Abdomen: Color, pale yellow. Tergites II-VI with darker oral band consisting of a median and two lateral spots linked together. Tergite VII and VIII darker, VII with a pale ym, anal band. 89 MEM. AMER. ENT. SOC., 34 90 NEW NORTH AFRICAN TANYPODINAE Fic. 1. Thienemannimyia choumara sp. n. — male hypopygium. Hypopygium: (Fig. 1). Gonocoxite 158 um, twice as long as broad. Gonostylus 115 pm, slightly curved. Gonocoxal lobe 56 um, approximately one third the length of gonocoxite. Female imago, pupa and larva; Unknown. Type locality: Morocco. Holotype: 1 & labelled ‘‘Thienemannimyia choumara sp.n. Morocco Sp. 1 M6.”’ Leg. Choumara. Det. C. Dowling, 1982. Type mounted on microscopic slide deposited in Zoologische Sammlung des Bayerischen Staates, Munich. Paratype: 1 & labelled ‘‘Thienemannimyia Sp. 1. Morocco 1397 68u:1070 68 M7, T. choumara’’ Leg. Choumara. Det. C. Dowling, 1982. (1397.68 = Environs of Berkane, Monts Beni, Shassen, 1968. 1070.68 = Environs of Marrakesh, Souk des Judais, 1968. Two different samples, broken at C.A. DOWLING 91 Fic. 2. Rheopelopia murrayi sp. n. (a) Male — wing. (b) Crown of setae on tip of tarsal segment 111, leg 11. (c) Claws and pulvilli. transport and mixed — information supplied by Dr. F. Reiss). Mounted on microscopic slide deposited in Zoologische Sammlung des Bayerischen Staates, Munich. Comments: Members of the Genus Thienemannimyia can be separated by the various markings on the wings and legs, as well as the structure of the male genitalia. On these criteria 7. choumara sp. n. appears similar to T. geijskesi (Goetgh.), having no markings on the wing and a brown ring at the tips of the femora. 7. choumara can be distinguished from T. geijskesi by darkened scutal stripes, smaller leg ratios and the structure of the genitalia. T. tinctoria (Freeman), the only other member of the Genus recorded from Africa (Freeman, 1955) can be differentiated by the presence of a brown ring on the base of the tibiae. Rheopelopia murrayi sp. n. Imago male (n=2): Length 4.0-4.5 mm. Head: AR 1.50-1.55. Antenna with 13 flagellomeres, the proximal end of the last flagellomere darkened. Palpal segments, pale brown; Clypeus, pale brown. MEM. AMER. ENT. SOC., 34 92 NEW NORTH AFRICAN TANYPODINAE Fic. 3. Rheopelopia murrayi sp. n. — male hypopygium. Thorax: Color pale brown, scutal stripes mid-brown. Wing, length 2.5 mm, membrane pale with markings as in Fig. 2. Legs, pale yellow with ring at tips of the femora and paler ring at bases of the tibiae. Crown of setae on P,, Ta ,,; (Fig. 2). Claws trifid; pulvilli present, less than half the length of the claw, but not clearly visible on all legs (Fig. 2). Fe Ti Ta, Ta, Ta; Ta, Tas LR P} 900- 1080 740 400 300 210 140 0.68 950 [Py 120- 110- 980 970 500 270 220 170 160 0.50 leva 930 1210 780 410 320 220 140 0.64 Abdomen: Tergites II-VI pale, with darker oral band: VII and VIII completely brown. C.A. DOWLING 93 Hypopygium: (Fig. 3). Gonocoxite 178.5 um, approximately half as long as wide. Gonostylus 127.5 ym, curved. Gonocoxal lobe 61 pm, just greater than one third gonocoxite length. Female imago; Unknown. Pupa. (n=1). Color yellow. Length, Abdomen 4.5 mm. Cephalothorax 1.5 mm. Thoracic horn, similar to R. ornata, longer than the first abdominal segment. LS setae on abdominal segment VIII well developed, longer than the length of the segment. LS, situated on the middle of the segment edge. Anal lobe as long as broad. Anal lobe tooth slightly less than half the length of the lobe. Anal lobe setae longer than the lobe, the hind seta situated on the middle of the lobe edge. Larva; Unknown. Type locality: Morocco. Holotype: 1 & labelled ‘‘Morocco 2 354.67 M10. Rheopelopia murrayi sp. n. Tata, Moyen Dra. 1967’’. Leg. Choumara. Det. C. Dowling, 1982. Type mounted on microscopic slide deposited in Zoologische Sammlung des Bayerischen Staates, Munich. Tip of a second abdomen mounted beside the legs. Paratype: 1 o& labelled ‘‘F 28A Sahara. Rearings from a small stream, Zousfana, Algerien 19.3.1955. Rheopelopia murrayi’’. This specimen was obviously dissected from the pupal exuvium also mounted on the slide. Mounted on microscopic slide deposited in Zoologische Sammlung des Bayerischen Staates, Munich. Leg. E.J. Fittkau. Det. C. Dowling, 1982. Comments: The relationship of Rheopelopia murrayi sp. n. with other members of the Genus is difficult to determine. The pupal features would suggest that it belongs to the ‘‘ornata-group”’ (Fittkau, 1962). In fact on the basis of the characters listed, it is difficult to separate the pupa from that of R. ornata (Meig.). On the specimen examined, the thoracic horn is bent, but it appears to be somewhat longer and narrower than that of examined specimens of R. ornata. The adult characteristics suggest that R. murrayi is closer to R. eximia of the ‘‘maculipennis-group’’. The pulvilli are present but they are small and are not clearly visible. R. eximia is the only other species with a brown ring at the base of the tibiae. The long oval shape of the gonocoxal lobe in R. murrayi is clearly different from that of other European species. ACKNOWLEDGEMENTS Sincere thanks are extended to Dr. E.J. Fittkau and Dr. F. Reiss for mak- ing the material available and supplying included information. Thanks are also due to Dr. D.A. Murray and Mr. P. Ashe for their invaluable help and support and for their comments on the manuscript. MEM. AMER. ENT. SOC., 34 94 NEW NORTH AFRICAN TANYPODINAE REFERENCES FiItTKAU, E.J. 1962. Die Tanypodinae (Diptera, Chironomidae). Die Tribus Anatopyni, Macropelopinii und Pentaneurini)-Abh. Larvalsystem Insekten: 1-453. FREEMAN, P. 1955. A Study of African Chironomidae. Part 1. Bull. Br. Mus. Nat. Hist. (Entomology) 4(1):1-67. SAETHER, O.A. 1980. Glossary of Chironomid morphology terminology (Diptera, Chir- onomidae). Ent. Scand. Suppl. No. 14. 51 pages. Morphological and Ecological Remarks on the Larva of Chernovskiia macrocera (Chernovskii) (Diptera: Chironomidae) By U. FERRARESE Museo Civico di Storia Naturale Verona, Italy ABSTRACT. — The larva of Chernovskiia macrocera (Chernovskii) is redescribed on the basis of specimens collected in the Po River (northern Italy) and its ecology is reported. An overall agreement with Chernovskii’s description is established and certain differences are discussed. Since 1974 the National Electricity Board of Italy has carried out a biological survey to determine the effects of the nuclear power plant at Caorso (northern Italy) on the aquatic environment of the Po River. During this survey, several groups of macroinvertebrates living on the macrophytes of the riverside or on the sandy bed of the central part of the river, were in- vestigated from a systematic standpoint. Among these groups the chironomids played an important role, by having the greatest number of species. Several of these chironomids were previously unknown in Italy. The occurrence of Robackia demeijerei (Kruseman), Beckidia zabolot- skyi (Goetghbuer), and Chernovskiia macrocera (Chernovskii) was somewhat of a surprise. The occurrence of Chernovskiia macrocera is of particular interest. Since Chernovskii’s first description of larvae from the River Volga, I believe no further specimens have been collected. The extreme rarity of this species is also demonstrated by the fact that only 5 of approximately 50,000 chironomids that I examined were Chernovskiia macrocera. The larvae I collected corresponded in their measurements and in almost all morphological details with Chernovskii’s description (1949) (Fig. 1). In my specimens the number of antennal segments varied from 7 (Fig. 2) to 8 (Fig. 3), while Chernovskii speaks only of 8 segments (Fig. 1). Although this might demonstrate a difference between the two descriptions, this is not necessarily so. In Chernovskii’s description the first antennal segment is divided into two parts. Thus, his 8 antennal segments result from counting each of these division as a segment. I found some specimens to have 7 95 MEM. AMER. ENT. SOC., 34 96 LARVA OF CHERNOVSKIIA MACROCERA antennal segments, because the first segment (or the first division of this segment), is much less sclerotized than the rest of the segment and possibly retracted into the head in some way. The following character should be added to Chernovskii’s description: premandible with 4 teeth (Fig. 2, Pm). Chernovskiia orbicus has 3 pre- mandibular teeth. The mentum (M) in my specimens also differs from Chernovskii’s description. In my specimens (Fig. 2, 3, M) the anterior margin is deeply concave, with 1 median tooth and 5-6 pairs of flat, light brown, lateral teeth. On both sides of the mentum 2 lobes can be seen, with striae visible only with difficulty. These lobes probably represent the ventromental plates (Fig. 2, 3, VmP). Chernovskii did not note these lobes, and because of the lack of the ventromental plates, placed his larvae in the subfamily Orthocla- diinae and provisionally named them Orthocladiinae gen. ? 1 macrocera. Saether (1977) revised the Harnischia complex and erected the genus Chernovskiia, in which he included Chernovskii’s larvae together with C. orbicus (Townes), although he hypothesized that they could represent the larval stage of C. amphitrite (Townes). Saether only had Chernovskii’s drawings available, and it was natural for Saether to attempt to interpret what particulars of Chernovskii’s drawings could be the ventromental plates. The first ‘‘true’’ larvae of C. macrocera that Saether could have ex- amined were those collected by myself and forwarded to him by A.M. Nocentini at the end of 1981. Thus, Saether wrote (according to a personal communication with J.E. Sublette) that what appears to be the lateral teeth of the mentum, could in fact be ‘‘...the collapsed ventromental plates . . .”’ This is a very suggestive interpretation, but, on the basis of my observations, I believe it is not well-founded. I feel that my observations have demonstrated that certain aspects of the larval morphology of C. macrocera and similar species are far from being completely clarified. I feel that further studies are necessary, perhaps using more refined microscopic techniques than those at my disposal. The environmental characteristics of the collecting site correspond to those reported by Chernovskii. The larvae were collected from a sandy and muddy substrate in the central part of the Po River, some distance from either bank. At the collecting site at Metapotamal, the river is 9 m deep and about 400 m wide. During collecting visits the flow varied from 800 to 1800 m?/sec. Oxygen concentrations remained at a reasonably high level Fic. 1. Chernovskiia macrocera larva (after Chernovskii, 1949). (A) antenna. (CAS) caudal abdominal segments. (HI) head, lateral view. (Hv) head, ventral view. (LS) labral sensillum. (M) mentum. (Md) mandible. (MP) maxillary palp. OF U. FERRARESE MEM. AMER. ENT. SOC 98 LARVA OF CHERNOVSKIIA MACROCERA Fic. 2. Chernovskiia macrocera larva from the River Po, with seven segmented anten- na. (A) antenna. (L) labrum. (LS) labral sensilla. (M) mentum. (Md) mandible. (MP) maxillary palp. (Pm) premandible. (VmP) ventromental plates. U. FERRARESE 99 CAS Fic. 3. Chernovskiia macrocera larva from the River Po, with eight segmented anten- na. (A) antenna. (CAS) caudal abdominal segments. (Hv) head, ventral view. (L) labrum. (LS) labral sensilla. (M) mentum. (Md) mandible. (MP) maxillary palp. (VmP) ven- tromental plates. MEM. AMER. ENT. SOC., 34 100 LARVA OF CHERNOVSKIIA MACROCERA (7.3-10.5 ppm) and the other 45 chemical and physical values contributed to defining a mildly polluted environment. Cases of acute pollution were only occasionally verified. Cheronovskiia macrocera has not recently been found in other large West European rivers possibly because of their more serious pollution problems. LITERATURE CITED CHERNOVSKIL, A.A. 1949. Opredelitel lichinok komarov semeistva Tendipedidae. Izd. Akad. Nauk, SSSR 31:1-186. 1961. Identification of larvae of the midge family Tendipedidae. (Trans- lation E. Lees, Editor K.E. Marshall). National Lending Libr. Sci. Tech., Boston Spa, Yorkshire. 300 pp. SAETHER, O.A. 1977. Taxonomic studies on Chironomidae: Nanocladius, Pseudo- chironomus, and the Harnischia complex. Bull. Fish. Res. Bd. Canada 196:1-143. Interdigitating Broadscale Distributional Patterns of some Kansas Chironomidae LEONARD C. FERRINGTON, JR. State Biological Survey of Kansas 2045 Ave. ““A’’, Campus West University of Kansas Lawrence, Kansas 66044 ABSTRACT. — Preliminary investigations of the Chironomidae species occurring within the state of Kansas has revealed taxa which indicate the faunal composition of the state is in- fluenced by species exhibiting six generalized distributional patterns. In addition to cosmo- politan species and eastern species with western distributional limits in the central plains, six species common to the southwest or to states west of the Rocky Mountains have been collected in western Kansas. Two species with more northerly or Rocky Mountain distributions have also been collected in western Kansas. Two species with more southerly or southwestern distributions are common to eastern and central Kansas. Six species common throughout the state are considered as having a central plains distribution. In addition a distinctive but subor- dinate Ozarkian Plateau influence is implicated. These statewide patterns of broadscale inter- digitating distributions appear to be consistent with those found for other aquatic insect taxa inhabitating the state. INTRODUCTION In 1974 the staff of the State Biological Survey of Kansas initiated a long term study of the organisms occurring in aquatic environments of the state of Kansas. The stated goals of this study are to determine on a county-by- county basis the species composition of aquatic invertebrates and to characterize the statewide distribution, abundance and populational nature of these organisms. The results of this ongoing project have been and will continue to be disseminated in an annual publications series entitled ‘Technical Publications of the State Biological Survey of Kansas,”’ copies of which are available upon request. Early in this study it has become evident that major distributional break- points for certain aquatic taxa occur within the political boundaries of Kan- sas (Coler and Slater, 1982; Gelhaus, 1982; Gilbert, 1980, 1979; Hamilton and Shuster, 1980, 1979; Huggins, 1981, 1978; Liechti, 1982, 1981, 1980, 1979, 1978; Liechti and Huggins, 1977; May, 1982a, 1982b; Shuster and Hamilton, 1978; Slater, 1982, 1981, 1980, 1979; Stewart and Huggins, 1977). A very distinctive east-west distributional demarcation is evidenced 101 MEM. AMER. ENT. SOC., 34 102 DISTRIBUTIONAL PATTERNS OF KANSAS CHIRONOMIDAE by certain species, particularly of the orders Hemiptera, Coleoptera, Ephemeroptera, and Plecoptera. Within these orders the westernmost distributional limits of more easterly distributed taxa are roughly coincident with the westernmost extensions of Oak-Hickory deciduous forests in eastern Kansas. In addition, the easternmost boundaries of several western or southwestern taxa appear to be roughly coincident with the easternmost edge of the high plains in western Kansas. For Odonata there also exists a north-south demarcation within the central plains states for species that have approximate east-west transcontinental distributions (Huggins, 1978). In August of 1980 I became affiliated with the State Biological Survey and began to determine the species of Chironomidae collected within the state. At that time I felt that the chironomid fauna of the state would be composed principally of taxa that are cosmopolitan in their distribution, along with a lesser number of taxa of more easterly distribution which reach their westernmost distributional limits in the Great Plains. Owing to the predominance of major river courses with shifting sand substrates within the region, it was felt that numerous taxa associated with this substrate type, primarily ‘‘Harnischia Complex’’ species, would also be commonly encountered. The intent of this paper, however, is to discuss recently published (Ferrington, 1981, 1982) and as yet unpublished data which sug- gest that some broadscale interdigitating of distributional patterns similar to the type found for other aquatic insect orders also occurs with the Chironomidae. Preliminary distributional data are presented and potential geographic areas of influence are proposed based upon comparisons with existing known species distributions. METHODS Light trap samples of Chironomidae in the collection of the State Biological Survey were the principal source of material for this paper. These samples were taken throughout the state and include collections from springs, small to large rivers, reservoirs and farm ponds. The geographic location and ecophysiographic regions of the state are indicated in figures la and 1b, respectively. Figure 2 illustrates the major rivers and county boundaries. All identifications of species in this paper were based upon determina- tions of adult male specimens. Keys used for identification were Malloch (1915), Townes (1945), Roback (1971), Hansen and Cook (1976), and Oliver (1981). Original descriptions by Saether (1971) and Sublette (1960) L. C. FERRINGTON, JR. 103 Be ee ee ee a a rai San INN jerome Leeman Tete Trt Tries Tsuna Catv t ; | i l } | i i i i ! | | ee Sree | ee ee eel eee ‘SHERMAN Trrowas TSneroan res [aooxs q t i | | i 4 1 | | i i i ee io > eh. Bile Agus Sian ie Sines : panes HIGH PLAINS alr i i i i i : 5 H hen ee | i i Se 1 (eke ‘ | z i ! aia Trnannn } -—-— —-—-—- ee ep jaa ad 1 ! i GREELEY rem Tscorr Te Lane p ‘sg on BM vente ae i i p i | Se i i ! ! ee ee ee eee | ' i i i L.—.—i-.—.-. | i | (ISUTCER i ' i i fom | Si [ews ieee 1 a gs (| ise pecs V STANTON ieee tes | | Nc i ! oe... eg hce. ee t ee, — leave VeLark y MORTON | 1 SaaS EG i ‘ i | | i | | CHAUTAUQUA HILLS CHEROKEE LOWLANDS 1b OZARKIAN PLATEAU Fic. 1. Geographic location (1a) and ecophysiographic regions (1b) of Kansas. MEM. AMER. ENT. SOC., 34 104 DISTRIBUTIONAL PATTERNS OF KANSAS CHIRONOMIDAE NIDNNVYS __.—-—- 4 "]) Hivom | Ananya) | NOSH3I43C., NOSIHOLV Zs > ‘© NOSHOV? | “Sesuey UIYIIM SIOATI JOfeu pue saliepunog AjuN0D 7 ‘o1g 3HONVWOO } uzeuvecty— a ae aL j | | pelbon = NVW3900H ; ¥ 3NNOGSO | SNOOU | qicasouiiar ou j TIVHSUNA oienday | mj L. C. FERRINGTON, JR. 105 were consulted for determinations of Lenziella cruscula and Cryptotendipes ariel respectively. Specimens of all species herein included are deposited in . the collection of the State Biological Survey of Kansas. RESULTS As is the case with certain other aquatic insects, there appears to be four principal geographic influences contributing to the species richness of Kan- sas Chironomidae. The principal influences can be defined in general terms as east-west and north-south. In addition subordinate Ozarkian Plateau and Central Plains influences are also implicated. Species arranged according to geographic areas of influence are presented in tables 1-4. Western or Southwestern Elements — The most surprising species records for the state have been those for species that were previously con- sidered to occur only in the southwest or to be limited to the western side of the Rockies. These species are given in table 1. Ferrington (1981) previously reported the occurrence of Thienemannimyia barberi (Coquillett) and Tanypus grodhausi Sublette from collection sites along the South Fork of the Republican River near St. Francis in Cheyenne Co. Subsequent collec- tions of T. barberi from the Arkansas River in Hamilton and Barton coun- ties indicate that this species is widespread in the western half of the state, apparently being common to the high plains ecophysiographic region. Cryptotendipes ariel (Sublette) has also recently been collected from the South Fork of the Republican River at a site where 7. barberi and T. grodhausi occur. Paralauterborniella subcinctum (Townes) and Polypedilum isocerus Townes also occur in counties within the high plains ecophysiographic region. Stictochironomus naevus (Mitchell) was collected from the inlet stream at Coldwater Lake in Comanche County, along the northern edge of Red Hills ecophysiographic region. All of these records represent the easternmost known occurrences of the species. Northern Elements — This category contains two species within it, Ablabesmyia pulchripennis (Lundbeck) and Diamesa_ heteropus (Coquillett), see table 2. Distributional records for A. pulchripennis (see Roback, 1971) indicate that it is common across the Canadian shield region, extending from Alberta north to Greenland. To the south it has been recorded from Rocky Mountain areas in Colorado and Washington and from two miles north of Spearfish in the Black Hills region of South Dakota. The Hamilton County record in western Kansas is the southern- most record for this species. MEM. AMER. ENT. SOC., 34 106 DISTRIBUTIONAL PATTERNS OF KANSAS CHIRONOMIDAE TABLE 1. Chironomidae occurring in Kansas with known distributions primarily west of the Rocky Mountains or southwestern United States. Taxon Known distribution Kansas Records Thienemannimyia barberi Arizona, California, Cheyenne Co., Barton Co., Nevada, Utah, Washington Hamilton Co. Tanypus grodhausi California Cheyenne Co. Cryptotendipes ariel California Cheyenne Co. Stictochironomus naevus California, New Mexico Comanche Co. Paralauterborniella Nevada, California, New Wallace Co. subcinctum Mexico, Arizona Polypedilum isocerus Nevada, California Scott Co. TABLE 2. Chironomidae occurring in Kansas with known distributions primarily of a northern or northern Rocky Mountain nature. Taxon Known distribution Kansas Records Ablabesmyia pulchripennis Greenland, Manitoba, Alberta, Hamilton Co. Quebec, South Dakota, Colorado, Washington Diamesa heteropus Alaska, Washington, California, Cheyenne Co. British Columbia, Colorado, Idaho, Minnesota, Montana, Nebraska, Nevada, New Mexico, Utah, Wyoming. Hansen and Cook (1976) indicate that D. heteropus is the Diamesinae species most commonly encountered in the western United States. It has been my experience that D. heteropus is quite ubiquitous in upper alpine streams of the Beartooth Plateau region of Wyoming. The known distribu- tional range of this species extends from Alaska south to New Mexico along the Rockies, and east across the northern plains through Nebraska to Min- nesota. The Kansas record is from the South Fork of the Republican River near St. Francis in Cheyenne Co. Southern, Southeastern or Ozarkian Elements — Three species are assigned to this category, see table 3. Coelotanypus atus Roback is quite common and widespread in Kansas. Previous records for this species in- dicate that it occurs along the Texas-Louisiana coastal regions and south in- to Puerto Rico. However, a recent record by Parkin and Stahl (1981) of two specimens collected in Illinois, combined with Kansas records, suggests that this species may be much more widespread in the southern and central plains states. L. C. FERRINGTON, JR. 107 TABLE 3. Chironomidae occurring in Kansas with known distributions primarily of a southern, southeastern or possibly Ozarkian nature. Taxon Known distribution Kansas Records Coelotanypus atus Louisiana, Texas, Puerto Rico Widespread Tanypus neopunctipennis Bahamas, Mexico, Arizona, Widespread Florida, Illinois, lowa, Louisiana, Missouri, Nebraska, Oklahoma, Tennessee, Texas Sympotthastia sp. not known Cherokee Co. Roback (1971) provides a distributional map for Janypus neopunctipen- nis Sublette. This species is very common in the southcentral and south- eastern United States, with marginal populations extending west to New Mexico and Colorado and north to South Dakota. It is commonly collected by light trapping at the larger reservoirs within Kansas. In contrast to C. atus and T. neopunctipennis, which are known to be more widely distributed to the south and/or southeast of Kansas, the occur- rence of a species of Sympotthastia in Shoal Creek, Cherokee County is probably related to an Ozarkian Plateau influence. Shoal Creek originates in southwestern Missouri, flows northwest into Kansas and confluences with the Spring River south of Riverton, Kansas, less than six river miles west of the Kansas-Missouri border. The occurrence of Sympotthastia in Kansas was rather surprising. The genus has not been recorded with any regularity in the North American literature, and immatures are poorly known (Simpson and Bode, 1980). Limited ecological data suggests that species of this genus occur in swift cur- rent areas of pristine lotic habitats. Beck (1977) lists it as rheobiontic and saprophobic. All this evidence leads me to conclude that this species may be restricted in the central states region to the area of the Ozarkian Plateau in Missouri and Arkansas and surrounding streams in which appropriate habitat areas occur. Thus, it would not be expected to occur in Louisiana, Mississippi or other states to the southeast or south of Kansas. I initially became aware of the presence of this species in Kansas through the collection of a single pupa by Mr. Donald Huggins in January of 1981. Subsequent collections in February and March of 1982 by myself have in- dicated that the population density of this species is quite high during winter months in a section of Shoal Creek flowing through Schermerhorn Park south of Galena, Kansas. Several specimens have been reared and descrip- tions will be forthcoming. This is the southwestern most verifiable record for the genus. MEM. AMER. ENT. SOC., 34 108 DISTRIBUTIONAL PATTERNS OF KANSAS CHIRONOMIDAE Eastern Elements Or Widespread Species With Dense Populations In Central Plains Region — Table 4 lists species that show one of two types of geographic distributional patterns. These patterns are: (1) species common to the eastern United States and having their westernmost distributional limits in the Great Plains, or (2) species that are widespread in overall occur- rence but which seem to have more dense populations in the Great Plains. Species which are common in the eastern United States but have their westernmost distributional limits in the Great Plains are Conchapelopia americana (Fittkau) and Diamesa nivoriunda (Fitch). Records of rearings for C. americana by Roback (1981) indicate that this species occurs from New Hampshire to Florida along the east coast. In Kansas this species has been collected from two spring fed Flint Hills streams. Though not collected to date, it probably occurs in similar spring fed streams in the Osage Cuestas and Glaciated ecophysiographic regions of eastern Kansas. It is unlikely, however, that appropriate habitat exists for this species further west; the Flint Hills records thus probably represent its western distribu- tional limit in Kansas. Hansen and Cook (1976) state that D. nivoriunda is the most common northeastern species of Diamesa. This species is sometimes collected with Diamesa cheimatophila Hansen and Cook in the northeast and with Diamesa mendotae Muttkowski in midwestern states. The collection of D. nivoriunda in Bourbon County represents the southwestern most record for the species. It has been collected in Missouri, however, and the Kansas record thus represents only a small range extension. It is unlikely that this species occurs west of the Flint Hills. The occurrence of D. nivoriunda in Kansas, however, does suggest that D. mendotae could occur within the state. In view of the record of Diamesa chiobates Hansen and Cook in Osage County it would now seem not unlikely that there could potentially be four species of the genus that occur in Kansas. The remaining species in table 4 are species that are known to be widespread in occurrence but are not reported with sufficient regularity to suggest that they comprise substantive populations. Within Kansas all of these species, with the exception of D. chiobates, are widespread and com- monly encountered. This type of evidence thus tends to infer that the species are representative of a distinctive central plains element within the North American chironomid fauna. The apparent rarity of D. chiobates in Kansas may simply be an artifact of collection efforts to date. As is the case with most Diamesa, this species was collected in early spring as an adult and is thus probably a late winter/ early spring emerging species. Additional early spring collections will be re- quired to determine the commonness of this species in Kansas. L. C. FERRINGTON, JR. 109 TABLE 4. Chironomidae occurring in Kansas that are common to the eastern United States, or species that are known to occur in only one or a few widely disjunct areas but appear to have dense populations within the central plains region. Taxon Known distribution Kansas Records Conchapelopia americana Common in Eastern states in- Small prairie streams cluding New Jersey, South in Flint Hills Carolina, Pennsylvania, West Virginia, New Hampshire, Florida Diamesa nivoriunda Alabama, Indiana, Massachu- Bourbon Co. setts, Maryland, Michigan, Minnesota, Missouri, New- foundland, New York, Ontario, Quebec, Virginia, Ohio, Pennsylvania, Wisconsin Tanypus nubifer Manitoba, California, Kansas, Widespread Nebraska, Utah Tanypus concavus Michigan, Iowa, New York, Common Texas, Virginia Paramerina smithae California, South Dakota, Common Washington, Utah, Wyoming, Mexico Diamesa chiobates Minnesota, Wisconsin Osage Co. Telopelopia okoboji Iowa, Minnesota, New Mexico Common Lenziella cruscula South Dakota Common Tanypus nubifer Coquillett, Tanypus concavus Roback, Paramerina smithea (Sublette), Telopelopia okoboji (Walley) and Lenziella cruscula Saether have been collected throughout various summer months. 7. nubifer and P. smithae are more common in aquatic habitats in the high plains than elsewhere within the state. Their distribution outside the state of Kansas suggests that they may more appropriately be considered as part of a distinctive high plains fauna rather than central plains; however, additional distributional and abundance data will be needed to distinguish between these two characterizations. T. concavus has been collected with some regularity from ponds, reser- voirs and slow-moving streams in the Osage Cuestas ecophysiographic region of Kansas. Roback (personal communication) has indicated that this species is not common in any of the habitats from which it has been col- lected in the eastern United States. The remaining two species, 7. okoboji and L. cruscula, are common in streams and larger rivers with sand and sand/silt substrates throughout the state. They commonly are collected along with species of “‘Harnischia Com- MEM. AMER. ENT. SOC., 34 110 DISTRIBUTIONAL PATTERNS OF KANSAS CHIRONOMIDAE plex’? genera such as Robackia, Chernovskiia, Cryptochironomus and Cyphomella. Sublette (personal communication) has indicated that T. okoboji is a dominant Pentaneurini component of sandy bottomed streams in New Mexico. The predominance of these two species in sandy bottomed streams suggests that they may have their largest populations within streams of the central plains states. DISCUSSION As indicated in the introduction, it was previously my feeling that the Kansas chironomid fauna would be composed of (1) cosmopolitan species, (2) species that are widespread across the eastern United States and having their westernmost distributional limits in the central plains, and (3) species that are adapted to living in shifting sand or sand/silt habitats, a common river substrate type within the central plains region. Initial investigation of. the chironomid fauna of Kansas has shown this view to be an oversimpli- fication. While it is true that many species which occur in Kansas can be readily assigned to one of the above three categories, the species discussed in this paper indicate that additional geographic distributional patterns assist in determining the faunal associations that occur within the state. From the perspective of large scale distributional patterns it would now seem that the following distributional patterns are implicated as contributing to the species richness of Kansas Chironomidae: (1) Species with cosmopolitan distributions. (2) Eastern species — Species widely distributed across the eastern United States whose western distributional limits are in the central plains states. In Kansas the Flint Hills generally mark the western edge of the distributional boundary. (3) Western or southwestern species — Species recorded primarily from the west side of the Rocky Mountains, or from the southwest United States. These species appear to extend north and east into the high plains ecophysiographic region of western Kansas. (4) Northern species — Species distributed throughout Canada and having a southern distributional extension along the Rocky Mountains and into the high plains. These species appear to be restricted in Kansas to the high plains ecophysiographic region in the western part of the state. (5) Southeastern species — Species with dense populations in the southeast United States, but with marginal populations extending north and west into or throughout much of the central plains region. (6) Central plains species — Species that are specifically adapted to microhabitat conditions such as shifting sand or sand/silt substrates. L. C. FERRINGTON, JR. 111 Because of the predominance of these habitat conditions in the central plains, these species have their principal population densities in this region. Marginal populations of such species may occur in areas other than the central plains when appropriate localized microhabitat condi- tions exist; however, these populations would not be considered domi- nant species components of major ecophysiogiaphic regions. In addition to the large scale distributional influences, several subor- dinate influences associated with specific ecophysiographic regions in Kan- sas may contribute to the species richness of the state. As indicated in the results, the occurrence of Sympotthastia sp. in Cherokee County is most certainly associated with the influence of the Ozarkian Plateau spillover in- to Kansas. Stictochironomus naevus, which has only been collected to date within the Red Hills ecophysiographic region, may be an example of another very regionalized subordinate influence. While no data exists to date, it is likely, based upon their statewide uniqueness in terms of stream types and riparian vegetation, that the Chautauqua Hills and the Cherokee Lowlands ecophysiographic regions may yield additional Chironomidae that are restricted to these areas within the state. ACKNOWLEDGEMENTS I would like to acknowledge P.M. Liechti and D.G. Huggins for helpful discussions of geographic distributional patterns evidenced by Kansas Ephemeroptera and Odonata, respectively, and for assistance in collection of Chironomidae throughout the state. B.G. Coler read and provided helpful comments on the typescript; S.S. Roback provided verification of tanypod material; R.M. McGregor provided support for the research through the activities of the State Biological Survey of Kansas; S. Brown typed and edited the typescript. LITERATURE CITED CoLer, B.G. AND A. SLATER. 1982. Additions and corrections to the list of Dytiscidae of Kansas. Tech. Publ. State Biol. Surv. Kansas 12:43-48. Beck, W.M., Jr. 1977. Environmental requirements and pollution tolerance of common freshwater Chironomidae. U.S. Envir. Prot. Agen., Cincinnati, OH. EPA-600/4-77- 024. 260 pp. FEERINGTON, L.C., JR. 1981. Kansas Chironomidae, Part I. Tech Publ. State Biol. Surv. Kansas 10:45-51. 1982. Kansas Chironomidae, Part II: The Tanypodinae. Tech. Publ. State Biol. Surv. Kansas 12:49-60. GELHAuS, J.K. 1982. New records and distributional notes for Kansas crane flies, excluding Limonia (Diptera: Tipulidae). Tech. Publ. State. Biol. Surv. Kansas 12:70-88. MEM. AMER. ENT. SOC., 34 112 DISTRIBUTIONAL PATTERNS OF KANSAS CHIRONOMIDAE GILBERT, C. 1979. New records of Kansas Hydrophilidae (Insecta: Coleoptera). Tech. Publ. State Biol. Surv. Kansas 8:23-30. 1980. New records of Kansas Hydrophilidae: Epimetopus, Helochares, Hydrochus, Laccobius, and Tropisternus (Insecta: Coleoptera). Tech. Publ. State Biol. Surv. Kansas 9:54-59. HAMILTON, S.W. AND G.A. SHUSTER. 1979. Records of Trichoptera from Kansas, II: The families Glossosomatidae, Helicopsychidae, Hydropsychidae and Rhyacophilidae. Tech. Publ. State Biol. Surv. Kansas 8:15-22. 1980. Records of Trichoptera from Kansas, III: The families Limnephi- lidae, Phryganeidae, Polycentropodidae, and Sericostomatidae. Tech. Publ. State Biol. Surv. Kansas 9:20-29. HAnseEN, D.C. AND E.F. Cook. 1976. The Systematics and morphology of the Nearctic species of Diamesa Meigen, 1835 (Diptera: Chironomidae). Mem. Amer. Entomol. Soc. 30:1-203. Hucoins, D.G. 1978. Additional Records of Kansas Odonata. Tech. Publ. State. Biol. Surv. Kansas 6:1-35. 1981. New state and distributional records for Kansas Plecoptera. Tech. Publ. State Biol. Surv. Kansas 10:65-70. LiecuT!, P.M. 1978. Stenonema mayfly records for Kansas. Tech. Publ. State Biol. Surv. Kansas 6:53-58. 1979. Caenis mayfly records for Kansas. Tech. Publ. State Biol. Surv. Kansas 8:12-14. 1980. Baetis mayfly records for Kansas. Tech. Publ. State Biol. Surv. Kansas 9:15-19. 1981. Kansas mayfly records for the genera Potamanthus, Pentagenia, Ephemera, Ephoron, and Tortopus. Tech. Publ. State Biol. Surv. Kansas 10:52-56. 1982. Five additional Ephemeroptera genera from Kansas. Tech. Publ. State Biol. Surv. Kansas 12:13-16. AND D.G. Hucocins. 1977. Records of Megaloptera in Kansas. Tech. Publ. State Biol. Surv. Kansas 4:45-50. Mattocu, J.R. 1915. The Chironomidae, or midges of Ill. with particular reference to the species occurring in the Illinois River. Bull. Ill. State Lab. Nat. Hist. 10:275-543. May, E.M. 1982a. Records of Limonia crane flies in Kansas (Diptera: Tipulidae). Tech. Publ. State Biol. Surv. Kansas 12:89-93. 1982b. New records of Kansas Hydrophilidae (Insecta: Coleoptera). Tech. Publ. State Biol. Surv. Kansas 12:94-104. OLiIveR, D.R. 1981. Chironomidae, pp. 423-458 in McAlpine, J.F., et al., coordinators. Manual of Nearctic Diptera. Agr. Can. Monogr. 27. 674 pp. PARKIN, R.B. AND J.B. STAHL. 1981. Chironomidae (Diptera) of Baldwin Lake, Illinois, a cooling reservoir. Hydrobiologia 79:119-128. RosBack, S.S. 1971. The Subfamily Tanypodinae in North America. Monogr. Acad. Nat. Sci. Phila. 17:1-410. —_______. 1981. The Immature Chironomids of the Eastern United States. V. Pen- taneurini-Thienemannimyia group. Proc. Acad. Nat. Sci. Phila. 133:73-128. SAETHER, O.A. 1971. Four new and unusual Chironomidae (Diptera). Can. Entomol. 103:1799-1827. SHUSTER, G.A. AND S.W. HaMILTON. 1978. Records of the Trichopteran families Hydrop- tilidae, Philopotamidae and Psychomyiidae from Kansas. Tech. Publ. State Biol. Surv. Kansas 6:36-47. L. C. FERRINGTON, JR. 113 Smmpson, K.W. AND R.W. Bope. 1980. Common larvae of Chironomidae (Diptera) from New York State streams and rivers with particular reference to the fauna of artificial _ substrates. N.Y. St. Mus. Bull. 439. 105 pp. SLATER, A. 1979. Additions to and corrections to the list of aquatic and semiaquatic Heteroptera of Kansas. Tech. Publ. State Biol. Surv. Kansas 8:31-37. 1980. New Records of Dryopidae and Elmidae from Kansas. Tech. Publ. State Biol. Surv. Kansas 9:60-66. 1981. Aquatic and semiaquatic Heteroptera in the collection of the State Biological Survey of Kansas. Tech. Publ. State Biol. Surv. Kansas 10:71-88. 1982. Aquatic and semiaquatic Heteroptera in the collection of the State Biological Survey of Kansas: Addendum. Tech. Publ. State Biol. Surv. Kansas 12:39-42. STEWART, K.W. AND D.G. Huccins. 1977. Kansas Plecoptera (Stoneflies). Tech. Publ. State Biol. Surv. Kansas 4:31-40. SUBLETTE, J.E. 1960. Chironomid midges of California. I Chironominae, exclusive of Tanytarsini (= Calopsectrini). Proc. U.S. Nat. Mus. 112:197-226. Townes, H.K., Jr. 1945. The Nearctic species of Tendipedini (Diptera: Tendipedidae (=Chironomidae) ). Amer. Midl. Natur. 34(1):1-206. MEM. AMER. ENT. SOC., 34 som a : < \- a : . F : ¢ ie eS Ok : Lo ol 284.) RET 4 a. = y or wy j i Lay ye Eee Mf The Distribution of Chironomids in Geothermal Waters in New Zealand D.J. FORSYTH Department of Scientific and Industrial Research P.O. Box 415 Taupo, New Zealand ABSTRACT. — The fauna of geothermal waters of North Island, New Zealand, has a low number of species most of which are insects. Chironomid larvae usually dominate these com- munities and are pre-adapted to geothermal waters with temperatures up to 36°C and acidity as low as pH 1.8. All species belonged to the Tanypodinae and Chironominae. There were no obligate acidophiles and only one possible thermophile. Often only one chironomid species was found at each site suggesting that each species occupies a narrow niche. Geothermal waters generally support a limited biota which, because of its low species diversity and the small variations in flow, temperature and chemical composition of these waters, may be considered simpler to study than most other ecosystems. General investigations of the fauna of geother- mal waters are few, mostly confined to Algeria (Mason, 1939); Iceland (Tuxen, 1944); USA (Brues, 1927, 1932); New Zealand (Winterbourn, 1968, 1969, 1970, 1973; Winterbourn and Brown, 1967; McColl and Forsyth, 1973; Forsyth & McColl, 1974; Forsyth, 1977; Forsyth and MacKenzie, 1981). The fauna is mainly confined to insects, which are possibly favoured because of the protection from the harsh conditions afforded by their chitinous exoskeleton, and most are forms which breathe atmospheric oxy- gen. Exceptions within the latter group are the chironomid larvae which often dominate the fauna and which, in the absence of fish, may be at the top of the food chain directly above the primary producers. The temperature boundary separating a diverse fauna at ambient temperatures from a restricted fauna at warm temperatures in New Zealand is between 25° and 30°C. The upper temperature limits for most thermally adapted invertebrates fall between 32°C and 45°C. Chironomid larvae, especially the Chironominae, can tolerate extremes of environmental condi- tions and some are pre-adapted to warm or cold acid waters or warm neutral waters of the Taupo volcanic zone of North Island, New Zealand, where they may reach unusually high population densities in the absence of competition and their normal predators. 115 MEM. AMER. ENT. SOC., 34 116 CHIRONOMIDS OF NEW ZEALAND GEOTHERMAL WATERS Zz Fic. 1. Taupo volcanic zone of North Island, New Zealand, showing sites of standing geothermal waters (@) and volcanic cones (CY). In this paper the species distribution of chironomid larvae is examined in relation to pH and temperature and the abundance of algae, expressed as chlorophyll a, in the water. The Taupo volcanic zone (Fig. 1) is part of a belt of volcanism extending from the central volcanic plateau of North Island, New Zealand, to the Ker- madec Trench. In this zone are a variety of lakes, springs, streams and seepages under various degrees of geothermal influence. Ten sites were sampled at least once in winter, when water temperatures of neighbouring cold, monomictic lakes were between 8° and 10°C. These sites fell more or less into two groups depending on their pH. Standing waters usually were more acid than flowing waters. If hot springs enter a lake after only limited exposure to the atmosphere, the H.S is oxidised to H.SO,, and the pH is D.J. FORSYTH 117 lowered. However, if the spring water is more fully exposed to the air as in flowing waters, the H.S is rapidly lost to the atmosphere and the pH is little changed. RESULTS Standing waters (Fig. 1, Table 1) 1. Lake Rotowhero: This lake is both acid and thermal. It has an area of 2.5 ha and a maximum depth of 14 m, pH 3.1, and a minimum water temperature of 29.5°C. Geothermal water enters the lake through several springs at the lake margin. The mean concentration of chlorophyll a was 48.0 mg.m™’. The algae comprised two species of Chlorella. The fauna was limited to insects (apart from a rare copepod) with larvae of Chironomus zealandicus dominant at a maximum concentration of 31 000 m? in winter. The larvae disappeared in spring when the water temperature exceeded 34°C to reappear in autumn when the temperature fell below 34°C. The chironomid larvae were eaten by a Notonectid, Anisops wakefieldi and a damsel-fly Ischnura aurora. Considerable adult mortality was inflicted by insectivorous birds. Other insects present were a thermal mosquito Culex rotoruae, a Hydrophilid and two Dytiscid beetles and two Hemipterans. 2. Lake Rotokawa: This lake has an area of 62 ha and a mean depth of 5 m. There is a small geothermal input, but it is a cold lake with winter temperatures similar to those of other cold lakes in the area. There are only larvae of Chironomus zealandicus in Lake Rotokawa and these are con- fined to depths up to 1 m. The chlorophyll @ concentration in the lake in mid summer was 115.0 mg.m® derived from a large population of Euglena and less abundant Chlamydomonas. The only other benthic invertebrate was a leech, Helobdella sp. 3. Echo Lake has an area of about 3 ha and is 34 m deep. The water was turbid with sulphur particles and the concentration of algal pigment was at the limits of detection. On the other hand there was a large concentration of bacteria. The concentration of dissolved oxygen was never more than 0.5 g.m.°* and pH was 2.4. The water temperature through the water column was a uniform 19.5°C. No benthic invertebrates were present, but abundant larvae of the mosquito Culex rotoruae were near the lake margin. This species is confined to thermal waters of the Taupo volcanic zone. 4. Opal Lake: This has an area of 0.5 ha and a maximum depth of 5 m. There was a large population of flagellate algae and the concentration of chlorophyll a was 42 mg.m7?. The waters were cold and the pH was 4.3. There was no obvious inflow or outlet. The acid character of the water was probably due to H.S from a nearby fumarole. The chironomid fauna was MEM. AMER. ENT. SOC., 34 118 CHIRONOMIDS OF NEW ZEALAND GEOTHERMAL WATERS cl 09€ cs JoTINO J9}e1ID + + ap + + + 4 + Lit Oo TY WeeNg NSULUITE AA + 70 Out 79 sdulidg ndejore A + + + = ove 09 ssulidg 1yeI9}9yf S1JDM BUIMO]T + + + + 80 58 Se axe] yInosg + qr 70 Orr 8st axe] YON ne + + + + Gy G8 ev axe jedO + + 70 SG ie axe] OUT + 4b + SII O01 "% BMEYO}OY IALT + + + + + + SG OTOYMOIOY SALT SiajDM suIpUudDIs ig @ ge 42 do 9 So Ge ee Se pe Be & 2o 82 OG 2 be BS ae ss ae 8 ao Se Bo 2h = Fh GS Bg ae BS oo o. = < a 2 © a an » 2 (A. fe 26 $3 ho} Oe is a z eg = eo m7 @F SS ot a eee eee ee eos oe 3 G em 8 *(p10de1 ou = —) *pueleoZ MON “pues UVION jo ouoZ STUBSTOA odney, oY} JO SioyeM jeu y} ul (,-W' 3) p \jAydoso[yo Jo uoNeIUs.UOD pue ‘, oinjeiod wa} ‘Fd 0} uONRIaI JO syoasuT JayIO pue oBAIe] prwoUOIYS Jo uONNqUIsIq “T ATA L D.J. FORSYTH 119 more diverse here than in the other locations examined (Table 1) perhaps because of a less extreme geothermal influence. Opal Lake is the type locali- ty for Kiefferulus opalensis the larvae of which were abundant in crevices of the bark of submerged branches at the lake edge. Unlike the other species present Kiefferulus was rare in the sediments. Chironomus zealandicus was always dominant at a mean concentration of 447¢ larvae.m”’. Larvae of Macropelopia umbrosa were 1150 m”’, but were not recorded from elsewhere in the region. The parthenogenetic Paratanytarsus agameta ap- peared only in autumn. Other insects belonged to Odonata, Trichoptera, Hemiptera and Ceratopogonidae. Oligochaetes were found in this lake but not in the others mentioned in this paper. 5. Rainbow Mountain Lakes: Rainbow Mountain crater contains two small lakes of about 0.2 ha each. North Lake is 8 m deep with a pH of 1.8 and a water temperature of 14°C in winter. Hot water enters as condensate from a fumarole 2 m from the lake edge. The concentration of chlorophyll a was 0.4 mg.m™®. Clumps of senescent filamentous alga Cladophora sp. were at the surface and on the lake bed. A few larvae of Chironomus zealandicus were observed amongst the floating algae but there were none in the sedi- ment. The water was turbid with sulphur particles and other mineral material. This is the lowest pH at which chironomids have been recorded in New Zealand. South Lake about 8 m away, is 3 m deep with a pH of 2.5. The water is greenish and more optically clear than that of North Lake. The chlorophyll a concentration of the water was 0.8 mg.m™ and the water temperature was 8.5°C indicating no input of hot water. Larvae of Chironomus zealandicus were abundant in the sediment, while floating mats of filamentous alga Microspora pachyderma supported numerous larvae of Tanytarsus funebris. Adults of both species were emerging abundantly in winter when the water temperature was 8.5°C. Elsewhere chironomid emergence was not obvious at this temperature. Flowing waters (Fig. 2, Table 1) 1. Ketetahi Springs: Ketetahi Springs on the slopes of Mt Tongariro at an altitude of 1385 m above sea level is an area of numerous hot springs and fumaroles of different physical and chemical characteristics. Most of the springs were near boiling, turbid, and coloured by metal sulphides, but some were clear. Algal mats of Phormidium sp. and a coccoid blue-green alga Synechoccus sp. grew on the margins of one stream at a temperature of 53°C and pH 6.0. Downstream the spring became a thin film of water cross- ing a terrace of deposited iron hydroxide (limonite). Here larvae of Paratanytarsus agamenta built their tubes at a density of about 20 000.m°. In the mud and silt of small, shallow pools along its course were a few lar- MEM. AMER. ENT. SOC., 34 120 CHIRONOMIDS OF NEW ZEALAND GEOTHERMAL WATERS ; y Mt Tarawera Waimangu-——_—_(e F : Rainbow Mountain (e) Waiotapu Mt Tongariro Ketetahi Springs Fic. 2. Taupo volcanic zone of North Island, New Zealand, showing sites of flowing geothermal waters (@) and volcanic cones (€y)- vae of Chironomus zealandicus. The upper temperature limit of both species was 32°C. 2. Waiotapu: At Waiotapu thermal area the Champagne Pool overflows at 73°C on to the Artist’s Palette, a flat area of siliceous sinter overlain by fine silt and mud. Here the water temperature was 18°C and the pH 6.4. Abundant larvae of Chironomus zealandicus were exposed when the mud was disturbed. No other invertebrates were present. 3. Waimangu Valley: Waimangu Valley is a thermal area formed during the eruption of nearby Mt Tarawera in 1886. Frying Pan Lake, pH 3.8 and water temperature 56°C drains into the Waimangu Stream. Numerous hot springs along the stream margins supported large populations of larvae of D.J. FORSYTH 121 TABLE 2. Lower limits of pH and upper limits of water temperature for chironomids in thermal waters of the Taupo volcanic zone, North Island, New Zealand. pH it °C Macropelopia umbrosa 4.3 25.0 Chironomus zealandicus 1.8 34.0 Kiefferulus opalensis A, 25.0 Paratanytarsus agamenta 4.3 32.0 Tanytarsus funebris Ded 25.0 Tanytarsus sp. 4.2 36.0 Tanytarsus sp. which built their tubes in shallow pools at water temperatures up to 36°C and at pH 4.2. Adults were emerging from the slow flowing pools and swarming at the stream margin during the winter in- spection. Larvae of Culex rotoruae were found with them. Despite an exten- sive search, only one larva of Chironomus zealandicus was found here. Filamentous algae Phormidium sp. and Mastigocladus sp. with unicellular blue-green Synechoccus sp. were common particularly at higher tempera- tures. The same biota was present in a small stream draining the Inferno Crater at Waimangu where the pH was 5.5 (Table 1). Concentrations of chlorophyll @ in both streams were less than 2 mg.m™*. DISCUSSION All chironomids found in New Zealand geothermal waters belong to the Chironominae and Tanypodinae in contrast to some northern hemisphere hot springs where Orthocladinae have also been reported. Although six species of chironomid were found in geothermal waters, on- ly Opal Lake (4 species), Ketetahi Springs, South Lake on Rainbow Moun- tain and Waimangu Stream (2 species) had more than one species. The greater diversity in Opal Lake was probably due to the higher pH, low water temperature and relatively high concentration of chlorophyll a. Never- theless Macropelopia umbrosa and Keifferulus opalensis, although wide- spread, have a local distribution. In South Lake and Ketetahi Springs the species exploit different niches. Paratanytarsus agameta at Ketetahi Springs was probably living close to its upper temperature limit. It is not characteristic of thermal areas, but common in small, shallow, sheltered ponds where the summer water temperatures may briefly approach the winter temperatures measured at Ketetahi Springs. However, it may be unable to tolerate summer temperatures there. Tanytarsus funebris is recorded for the first time from MEM. AMER. ENT. SOC., 34 122 CHIRONOMIDS OF NEW ZEALAND GEOTHERMAL WATERS acid waters. Its usual habitat is in ponds, sewage waters and the littoral zone of lakes. The presence of Chironomus zealandicus at most of these sites il- lustrates its adaptability to extreme conditions. C. zealandicus and Tanytar- sus sp. are the only species that appear to be truly heat tolerant, and C. zealandicus and Tanytarsus funebris are the only species tolerant of extreme acid conditions. The mildly acid tolerant species, Kiefferulus opalensis and Macropelopia umbrosa have been found in neutral waters. Tanytarsus sp. may be the only obligate thermophile for it has not been found outside the Waimangu thermal valley. No chironomids are obligate acidophiles (Table 2). Likewise none of the other common invertebrates are obligate acidophiles, but Culex rotoruae and the Ephydrid flies are confined to thermal water. The Ephydrid flies were at all sites where there was hot flowing water. Microvelia macgregori was found only in standing waters (Table 1). It was rare for geothermal waters at temperatures below 36°C not to support chironomids. An exception was Echo Lake where chlorophyll a was virtually absent. At the other sites where the concentration of chlorophyll a was low, water was flowing exposing the larvae to more food. Algae appear to be essential for the support of chironomids in geothermal waters. Chironomid larvae are a characteristic element in this environment which supports a restricted insect community. A notable feature of these geothermal waters is that despite the spatial differences of temperature, water flow, and food availability at any one site, often only one species of chironomid is found there. Each species, therefore, appears to occupy a narrow niche in these extreme conditions. ACKNOWLEDGEMENTS To J. Davies, M. Gibbs, C. Howard-Williams, M. James and J. Priscu. REFERENCES Bruges, C.T. 1927. Animal life in hot springs. Quarterly Review of Biology 2:181-203. 1932. Further studies on the fauna of North American hot springs. Pro- ceedings of the American Academy of Arts and Sciences 67:185-303. ForsyTH, D.J. 1977. Limnology of Lake Rotokawa and its outlet stream. New Zealand Journal of Marine and Freshwater Research 11:525-539. ForsyTH, D.J. AND R.H.S. McCoii. 1974. The limnology of a thermal lake: Lake Roto- whero, New Zealand. II. General biology with emphasis on the benthic fauna of chirono- mids. Hydrobiologia 44:91-104. ForsyTH, D.J. AND A.L. MAcKENziE. 1981. Limnology of Opal Lake. New Zealand Jour- nal of Marine and Freshwater Research 15:279-283. D.J. FORSYTH 123 McCo1, R.H.S. AND D.J. ForsyTH. 1973. The limnology of a thermal lake: Lake Roto- whero, New Zealand. I. General description and water chemistry. Hydrobiologia 43: 313-332. Mason, I.L. 1939. Studies on the fauna of an Algerian hot spring. Journal of Experi- mental Biology 16:487-498. TuxEN, S.L. 1944. The Hot Springs of Iceland. Einer Munksgaardd, Copenhagen. 216 p. WINTERBOURN, M.J. 1968. The faunas of thermal waters uf New Zealand. Tuatara 16: 111-122. 1969. The distribution of algae and insects in hot spring thermal gradients at Waimangu, New Zealand. New Zealand Journal of Marine and Freshwater Research 3:459-465. 1970. The Hydrophilidae (Coleoptera) of New Zealand’s thermal waters. Transactions of the Royal Society of N.Z., Biological Science 12:21-28. 1973. Ecology of the Copland River warm springs, South Island, New Zealand. Proceedings of the N.Z. Ecological Society 20:72-78. WINTERBOURN, M.J. AND T.M. Brown. 1967. Observations on the faunas of two warm streams in the Taupo thermal region. New Zealand Journal of Marine and Fresh- water Research 1:38-50. MEM. AMER. ENT. SOC., 34 Effect of Metabolites on the Pyruvate Kinase of Chironomus plumosus Larvae CHRISTIAN FRANK Institut f. angewandte Zoologie Freie Universitat Berlin Haderslebenerstr.9, D 1000 Berlin 42 ABSTRACT. — Pyruvate kinase from Chironomus plumosus larvae was tested for regulatory characteristics of the last one-directional step in glycolysis. Due to the addition of KHCO; to the assay, the pH rose and the enzyme activity decreased. From the metabolites tested only the addition of ATP results in a decrease of activity. Alteration of the pH shows the greatest in- fluence on the activity of the PK. The pyruvate kinase (PK) catalyses the last one-directional step in the anaerobic degradation of glycogen. Therefore, it was assumed to be a con- trol enzyme of glycolysis (Newsholm & Start 1977). During anerobic metabolism in chironomid larvae, this reaction leads to the formation of lactate, ethanol and alanine (Frank 1977, 1983). The in- fluence of ATP, fructose-1, 6-biphosphate, KHCO; and of endproducts on the PK from 4th instar larvae of Chironomus plumosus should be shown. MATERIAL AND METHODS For all experiments, dredged 4th instar larvae of Chironomus plumosus and Glyptotendipes paripes from Tegeler Lake and Lake Heiligensee were used. The larvae were kept in plastic jars with tap water and no food the last two days before use. Enzyme assay. — The homogenisation medium contained 0,5 ml/25 mg wet weight of larvae, 20 mM Tris-HCl buffer, pH 7,2,1 mM MgCl. The clear supernatant (10” x 1000 g) was used for the assay at 25°C, 334 nm. The reaction was started by adding the supernatant fraction. Maximum reaction velocities were obtained under the following conditions, 1 ml volume: ATP: pyruvate-phosphotransferase E.C. 2.7.1.40 after Hoffman et al. (1979): 50 mM Tris-HCl buffer pH 7,2; 1 mM ADP; 0,1 mM fructose-1,6-biphosphate; 10 mM MgSO,; 75 mM KCI; 0,14 mM NADH; 12,0 mM phosphoenolpyruvate; 7 U/ml LDH; 50 ul supernatant. 125 MEM. AMER. ENT. SOC., 34 126 PYRUVATE KINASE OF CHIRONOMUS PLUMOSUS LARVAE uN E/min 6,0 6,5 7,0 7,0 8,0 85 pH Fic. 1. pH dependencies of pyruvate kinase activity, AE/min, o Tris buffer, x Tea buffer. RESULTS AND DISCUSSION The pH-dependency of the enzyme activity shows a maximum between pH 6,8-7,2. Different buffers show the same course but varying activities (Figure 1). KHCO; was added to the test mixture to simulate the carbon dioxide formed during anoxia. The addition of 250 mM KHCO, results in a decrease of activity by 50%, both in Chironomus plumosus and Glyptoten- dipes paripes larvae. Measurements of the pH of the assay mixture showed a rise in the pH. The decrease of activity is more a result of the alteration of the pH than of the addition of KHCO, (Figure 2). The addition of 5 mM (Tris buffer) or 10 mM ATP (TEA buffer) results in a decrease of activity by 82-92%. The addition of 1 mM fructose-1,6- biphosphate results in a reactivation of the PK (Fig. 3). The addition of endproducts of the anaerobic metabolism to the assay mixture (up to 5 mM) showed no significant differences to the control values. Only the influence of 5 mM alanine results in a light increase in ac- tivity. C. FRANK 127 A E/min 012 pH 010 85 0,08 8,0 0,06 75 0,04 7,0 0 100 200 300 500 750 1000 mM KHCO., Fic. 2. Effect of KHCO; on the activity of pyruvate kinase from Chironomus plumosus (A) and Glyptotendipes paripes (x) larvae. — pH alteration by addition of KHCO; to the assay. The maximum metabolic concentration of endproducts of Chironomus plumosus larvae measured during anoxia are: 0,12 mM/g wet weight lactate 0,3 mM/g wet weight ethanol 0.01 mM /g wet weight alanine Only unphysiologically high concentrations of metabolites (=>5 mM) show an influence on the activity of the PK, but they show no regulatory ef- fect. The greatest influence on the activity of the PK results from the alteration of the pH as also shown by Hoffman ef al. (1979). Holwerda et al. (1981) pointed out that in Mytilus, PK is the regulate in the last step of anaerobic glycolysis. In this species, the addition of 1 mM alanine to the assay results in a inhibition of activity by more than 75%. The activity of PK from MEM. AMER. ENT. SOC., 34 128 PYRUVATE KINASE OF CHIRONOMUS PLUMOSUS LARVAE 0,03 0,02 0,01 O 10 2o 5,0 10,0 mM ATP Fic. 3. Effect of ATP on the activity of pyruvate kinase from Chironomus plumosus l\ar- vae, o Tris buffer, x Tea buffer. Chironomus during anaerobiosis will increasingly be inhibited through the effects of alteration of pH and not by alanine. The activation of Chironomus PK by fructose-1,6-biphosphate shows patterns similar to those described by Munday ef a/. (1980). The activation of the PK could be interpreted as a feed-forward mechanism at the begin- ning of the anaerobic metabolism to ensure that the activity of PK is in- creased in relation to the activity of phosphofructokinase. In Chironomus the PK shows no allosteric characteristics. The regulation of the glycolysis in Chironomus might occur at the phosphofructokinase step as pointed out by Castellini and Somero (1981). ACKNOWLEDGMENTS This paper was supported by DFG grant FR.570/1. C. FRANK 129 REFERENCES CASTELLINI, M.A. AND G.N. Somero. 1981. Buffering capacity of vertebrate muscle: Corre- — lations with potentials for anaerobic funktion. — J. comp. Physiol. 143, 191-198. FRANK, C. 1977. Okologie und anaerober Stoffwechsel der Larven von Chironomus plu- mosus L. (Diptera, Nematocera). Diss. Universitat Tiibingen. 1983. Ecology, production an anaerobic metabolism of Chironomus plumosus L. larvae in a shallow lake. II Anaerobic metabolism. — Arch. Hydrobiol. [96:354-363]. HOFFMANN, K.H., T. MUSTAFA AND J.B. JORGENSEN. 1979. Role of pyruvate kinase, phos- phoenolpyruvate carboxykinase, malic enzyme and lactate dehydrogenase in anaerobic energy metabolism of Tubifex sp. — J. comp. Physiol. 130, 337-345. HoLtwerpA, D.A., P.R. VEENHOF AND H.A.A. VAN HEUGTEN. 1981. Regulation of mussel pyruvate kinase during anaerobiosis. — Posterabstract, 3d Congr. Europ. Soc. Comp. Physiol. Biochem. Noordwijkerhout. Munpbay, K.A., G.I. GiLEs AND P.C. Poat. 1980. Review of the comparative biochemistry of pyruvate kinase. — Comp. Biochem. Physiol. 67 B, 403-411. NEWSHOLME, E. A. AND C. Start. 1977. Regulation des Stoffwechsels, Verlag Chemie, Weinheim. 305 pp. MEM. AMER. ENT. SOC., 34 Ecology and Importance of the Chironomidae in the Trophic Structure and Biocenosis of Zoobenthos in the Lakes of the National Park of the Lithuanian SSR By ANTANAS GRIGELIS Institute of Zoology and Parasitology Academy of Sciences Vilnius, Lithuanian SSR ABSTRACT. — Twenty-five lakes in the National Park of the Lithuanian SSR were studied from 1976 to 1980 to determine the distribution of chironomid larvae according to sediment type, depth, and other environmental factors. Eighteen of these lakes are ranked according to an index of density and an index of biomass. The problem of sediment composition, as a significant factor in the distribution and development of chironomid larvae, was examined. INTRODUCTION Twenty-five lakes (Fig. 1) in the National Park of the Lithuanian SSR were investigated from 1976 to 1980. These lakes are located in the hills of eastern Lithuania (Grigelis et al., 1982) and differ in various topographical conditions. The total area of these lakes was 3680.7 ha. Some lakes, such as Gavys, Gavaitis, Vajuonis, Kretuonas, Usiai and others are artificially ex- cluded from the territory of the National Park as these lakes are more inten- sively used for recreational purposes. The lakes differ in surface area, depth, form of bed, length of orientation of their principal axis, width, and in their relationship to the direction of the prevailing westerly winds (Grigelis et al., 1975). RESULTS The investigation of the benthic fauna of 25 lakes in the National Park of the Lithuanian SSR from 1976 to 1980 showed that in the sublittoral and profundal zones of some lakes the dominant group of zoobenthos in terms of species composition, distribution, density and biomass was the Chironomidae. Chironomus plumosus, C. anthracinus, Sergentia longiven- tris, and Procladius spp. were widely distributed in these fresh water basins. On the basis of an index of density and biomass (Fig. 2, 3), the chironomids in the investigated lakes could be subdivided into four groups: 131 MEM. AMER. ENT. SOC., 34 132 LAKE CHIRONOMIDS OF LITHUANIAN SSR Fic. 1. Map of the larger lakes of the National Park of the Lithuanian SSR: 1. Utenas, 2. Utenykstis, 3. Baluosas, 4. Almajas, 5. Asekas, 6. Tauragnas, 7. Pakasas, 8. Ukojas, 9. Linkmenas, 10. Asalnai, 11. Dringis, 12. LuSiai, 13. Sakarvai, 14. Zeimenys, 15. Gavys, 16. Vajuonis, 17. Kretuonas, 18. Usiai. 1). Lakes in which the Chironomidae were the most important group constituting from 50.0 to 83.3 percent of the abundance and from 56.8 to 94.3 percent of the biomass. The lakes of this group differed in depth and substrate. Lakes LuSiai, Asalnai, and Tauragnas are deep with Sergentia longiventris as the dominant chironomid. Lakes GruodiSkis, Taramas, and Dringykstis are shallow with Chironomus plumosus as the dominant chironomid. A. GRIGELIS 133 90 80 70 60 50 40 30 Index Density of Chironomidae (%) 20 10 STASFponap rersn stuonfe, yeuresy Tersny seugeiney seunerty stasyA3utag sKarg stasyAsnq seuauyutTT SPUSH TY stqsyAupesy TRAIEYeS skuautaz seseyeg se foxy seuaqy seCeuty stqasyAueaqn sesonyeg seueiey stTeq1e@ stqreae9 STITEUSHTY Fic. 2. Distribution of the lakes of the National Park of the Lithuanian SSR according to the index of density for chironomids. 2). Lakes in which the Chironomidae were the second most important group of organisms constituting from 31.5 to 50.0 percent of the abundance and from 23.7 to 45.7 percent of the biomass. These lakes are both deep and shallow lakes as in group 1. 3). Lakes in which the Chironomidae were the second or third most im- portant group of organisms constituting from 12.0 to 19.0 percent of the abundance and from 14.3 to 21.4 percent of the biomass. 4). Lakes in which the Chironomidae were the third most important group of organisms or were absent, constituting from 0.0 to 6.8 percent of the abundance and from 0.0 to 10.0 percent of the biomass. These are lakes with hydrotroilite. The distribution, species composition, and development of the Chironomidae depends upon such factors as water temperature, chemical and physical properties (Grigelis et al., 1981; Zukaite, 1980), sediments, MEM. AMER. ENT. SOC., 34 134 LAKE CHIRONOMIDS OF LITHUANIAN SSR 100 90 80 70 60 50 40 30 Index Biomass of Chironomidae (%) 20 10 tersn TeETSNy seueqy Ted SPUSHTY skae9y selon sT4sTponay sKuowta7 seunety stqsyAsnq stuonfe, seuseiney yeutesy seusuyUuTT stqsyAueqn sTasyAusy Ty TeAIeHeS sTTeq Ted seseyed seleuty seuerey tegon stasyA3utig sTarear) STIPeusATY Fic. 3. Distribution of the lakes of the National Park of the Lithuanian SSR according to the index of biomass for chironomids. microflora of the substrate, vegetation cover, and other environmental fac- tors. In the profundal zone of the deep lakes the cold stenotherm larvae of Sergentia longiventris were the dominant organism in Lakes Sakarvai, LuSiai, Gavys, and Tauragnas, with Chironomus anthracinus the dominant organism in the profundal zone of Lake Dringis. The thickness of the A. GRIGELIS 135 hypolimnion in these lakes varied from 50 to 20m and occupied a dominant position in the total volume of the lake. Lakes Taramas, Baltelis and Utenykstis, which had hydrotroilite in the bottom were poor in chironomids. Lakes Alksnaitis and Gavaitis had hydrotroilite in the bottom and had no chironomids. These latter two lakes were chaoborid lakes. Our data confirms the distribution dependence of chironomid larvae on water temperature in the hypolimnion and bottom sediments. LITERATURE CITED GRIGELIS, A., G. BALKUVIENE, J. BANIONIENE, B. GAIVENIS, O. NAINAITE, AND E. ZuKAITE. 1975. Hydrobiological reserches in the Lithuanian lakes. In Lithuanian with English and Russian summaries. GRIGELIs, A., G. BALKUVIENE, J. BANIONIENE, E. ZUKAITE, J. KAVALIAUSKIENE, R. LENKAITIS, O. NAINAITE, A. ORLOVA, AND V. SUKACKAS. 1982. Biological productivi- ty of lakes of the National Park of the Lithuanian SSR. I. Hydrobiological characteristics. Lietuvos TSR Mokslu Akademijos darbai, C serija 3(79):54-63. In Rus- sian with Lithuanian summary. GRIGELIs, A., R. LENKAITIS, O. NAINAITE, AND E. ZuKAITE. 1981. Peculiarities of dis- tribution of cold-stenotherm hydrobionts in lakes of the National Park of the Lithuanian SSR. Verh. Internat. Verein. Limnol. 21:501-503. ZUKAITE, E. 1980. Thermal regime of LuSiai and Sakarvai lakes during period of summer thermal stagnation in 1978. Lietuvos TSR Mokslu Akademijos darbai, B serija 6(121): 83-91. In Russian with Lithuanian summary. MEM. AMER. ENT. SOC., 34 Appearance and Behavior of Nigerian and Californian Chironomid Larvae during Recovery from Desiccation GaIL GRODHAUS Vector Biology and Control Branch California Department of Health Services 215] Berkeley Way, Berkeley, California 94704 ERNEST C. BAY Western Washington Research and Extension Center Washington State University Puyallup, Washington 98371 LLOYD KNUTSON Insect Identification and Beneficial Insect Introduction Institute Beltsville Agricultural Research Center-W, USDA Beltsville, Maryland 20705 ABSTRACT. — Samples of dry substrate material containing Nigerian Polypedilum sp. near vanderplanki Hinton and Californian Tribelos atrum (Townes), Phaenopsectra pilicellata Grodhaus, and Hydrobaenus sp. were examined after immersion in water. Samples contained instars II, II-III, or IIl-IV, depending on the species. Each species demonstrated a characteristic folding or curling of the dormant larvae, which, except for Polypedilum, were enclosed in co- coons. After wetting, recovery took place most rapidly in Polypedilum and most slowly in Phaenopsectra. In Polypedilum it took 11-120 min for the first muscle contractions and 45-135 min for a return to normal appearance. In Phaenopsectra it took 11 hr-S days for the first mus- cle contractions and 12-48 hr for a return to near normal appearance. Polypedilum is suited to life in rock hollows, in which both inundation and drying are apt to occur suddenly. The Californian species are adapted to the vernal pool environment, in which hydrodynamic pro- cesses proceed at a slower rate; their life histories are probably regulated by diapause. Certain aquatic chironomid larvae are able to suspend activity for long periods of time while their habitat is without water. Chironomidae with this adaptation have been reported from Africa (Hinton 1951, Miller 1969, McLachlan and Cantrell, 1980), Australia (Edward 1968, Jones 1974), and North America (Grodhaus 1976, Grodhaus and Rotramel 1980, Wiggins et al. 1980). The present paper compares species from Nigeria and California with regard to the gross changes that occur while they recover from desicca- tion. The Nigerian species that we studied appears to be related to both Polypedilum yvanderplanki Hinton and P. dewulfi Goetghebuer but resembles P. vanderplanki more closely, on the basis of the single female 137 MEM. AMER. ENT. SOC., 34 138 RECOVERY FROM DESICCATION OF CHIRONOMID LARVAE reared from our sample. As were the larvae of P. vanderplanki studied by Hinton (1951, 1960, 1968), our African specimens were obtained from rock cavities. Our Californian larvae have been identified as Tribelos atrum (Townes), Phaenopsectra pilicellata Grodhaus, and an undescribed species of Hydrobaenus, each of which inhabits vernal pools (Grodhaus 1980). Ow- ing to the lack of specimens at the proper stage of recovery, the drought- resistant Polypedilum reported by Grodhaus and Rotramel (1980) was not included in the study. MATERIALS AND METHODS African material was collected from ‘‘grinding’”’ hollows on ‘‘inselberg’’ rocks near Zaria, northern Nigeria on 30 March 1973 by L. Knutson. Hollows as small as 10 cm long, 5 cm wide, and 5 cm deep yielded viable lar- vae. The larvae from California were extracted from soil from vernal pool sites listed by Grodhaus (1980). Dry substrate materials (i.e. debris from grinding hollows or soil from vernal pool sites) were stored in polyethylene bags prior to examination. All samples were wetted before examination to minimize mechanical damage to the larvae. Polypedilum larvae were abundant in the sample and were relatively easy to find. The other species required concentration by sieving and washing (Grodhaus 1980). Specimens were placed in dishes of tap water and observed at 10-60x. Additionally, specimens in cocoons had to be released mechanically with forceps to be seen. Specimens were stored and examined at room temperature (ca 21°C) unless noted otherwise. Results pertaining to different instars of a given species were pooled. Observations of material stored for similar lengths of time but from different localities were also combined. RESULTS Dormant larvae remain essentially unchanged for a few minutes after be- ing wet. When in that condition, specimens were considered to be in their “‘drought phase’’, which was used as a reference point for describing changes occurring later, during which time specimens were referred to as be- ing in their ‘‘recovery phase’’. Drought Phase Larva: All recently wetted larvae appeared completely lifeless. No cast skins or fecal pellets were associated with them to indicate activity during storage. Water content of the specimens has not been deter- mined, but soil samples matched with samples yielding viable Tribelos after 23 months of storage contained 4-5% water. G. GRODHAUS, E. C. BAY, L. KNUTSON 139 Drought phase larvae of each species showed visible evidence of desicca- tion. The antennae, anterior prolegs, anal papillae, and preanal setae were appressed against the body, and the oral cavity often contained an air vacuole. In Polypedilum and Tribelos, the body segments were shrunken, i.e. compressed, contracted, or both. The condition of Phaenopsectra was drastically different; the larva was inflated to near normal proportions, but a large vacuole occupied almost all of the interior of each body segment. This vacuolization caused larvae to float. Hydrobaenus larvae have not been studied critically enough to describe their condition in detail. The color of the recently wetted larvae was brownish red to yellowish red in Polypedilum and Phaenopsectra, yellowish red in Tribelos, and faint yellow in Hydrobaenus. Table 1 shows which instars were represented in the dry substrates. Viability was demonstrated by each stage except Polypedilum instar II, which was scarce in the sample. Proportions of instars were rather constant in samples from California, with instar II predominating in all instances. Table 1 also indicates the type of covering associated with the drought phase larva. Definite cocoons are constructed by 7. atrum, P. pilicellata, and Hydrobaenus sp. (See Grodhaus 1980 for descriptions.) Some larvae of Polypedilum sp. were seen to be within a cell lined with a silken membrane (Fig. 1). The membrane surrounding the illustrated specimen appears to be open at the anterior end. Whether all Polypedilum larvae were similarly en- cased was not ascertained. The membrane of Polypedilum was far less durable than that forming the wall of the Californian type cocoon, which could survive rough treatment and could be recognized long after the larva had emerged from it. All larvae whose recovery was observed were initially in a species-specific resting position (Table 1). The three species that assumed body folds were assigned positional indices according to the system of Danks (1979). The larva of Hydrobaenus sp. was not folded at any point but was curled inside its cocoon in a position similar to that figured for Ewkiefferiella claripennis (Lundbeck) by Madder et al. (1977). Recovery Phase Larva. Table 2 shows the interval between the exposure of specimens to water and the first muscular movements observed in three species. Prolonged storage seemed to retard the recovery of Tribelos and Phaenopsectra. The kinds of contractions and their usual sequences were: Polypedilum—rhythmic pulsation of pharynx; rhythmic pulsation of dorsal vessel; contraction of body, head, and mandibles; crawling. Tribelos—con- traction of prolegs and body; irregular contraction of dorsal vessel; contrac- tion of head and mandibles; crawling. Phaenopsectra—occasional contrac- tion of body and head and irregular contractions of dorsal vessel; gradually MEM. AMER. ENT. SOC., 34 140 RECOVERY FROM DESICCATION OF CHIRONOMID LARVAE TABLE 1. Condition of desiccated larval Chironomidae. Description of Resting surrounding Species Instar position* material Polypedilum sp II Delicate silk membrane, probably (nr. vanderplanki) Ill doesn’t enclose larva completely IV 6.5.1 Tribelos II Cocoon, encloses larva completely atrum Ill 11.1 Phaenopsectra II Cocoon, encloses larva completely pilicellata Il 4.4.4 Hydrobaenus sp. II curled Cocoon, encloses larva completely * According to Danks (1971), i.e., numerals indicate numbers of body segments and periods in- dicate where folds occur. TABLE 2. Recovery of desiccated larval Chironomidae. Estimated time elapsed from Period of immersion to: No. of dry storage lst muscle cessation of Species specimens (months) contraction water uptake Polypedilum sp. (nr. vanderplanki) 7 54-56 11-120 min 45-135 min Tribelos atrum 1 1 30 min 50 min 2 4 9-17 10-150 min 10-150 min “ 7 23-30 235-240 min 300 min i 4 84* 16 hr 200 min Phaenopsectra pilicellata 9 5-12 11-34 hr 12-30 hr ig 2 18 20-40 hr 20 hr 4 3 30-36 20 hr-S days 20 hr i 2 60 30-40 hr 40-48 hr *Kept at ca 4°C; others kept at room temperature. stronger, more complex movements leading to crawling (after termination of diapause, as discussed below). Viable Hydrobaenus sp. larvae were mak- ing simple contractions after several days, similarly to Phaenopsectra. The contractions of the pharynx and dorsal vessel that occurred in nearly | all of our Polypedilum larvae were similar to the movements of P. | G. GRODHAUS, E. C. BAY, L. KNUTSON 141 vanderplanki observed by Hinton (1951, 1960). These contractions ap- peared before any other movements in Polypedilum. In Tribelos and Phaenopsectra, pulsations of the dorsal vessel seemed to occur only after major body movements had begun, and these genera did not exhibit pharyngeal pulsations. After being wet for 48 hours, most larvae had made movements in one of the above categories or were clearly dead. An exceptional specimen of Phaenopsectra, however, did not show even simple contractions until it had been wet for 5 days. Most of the other Phaenopsectra specimens listed in Table 2 continued making only simple contractions for many days. This semiactive condition is presumed to be associated with diapause in which full mobility is delayed until after the larvae have been chilled for several weeks (Grodhaus 1980). Polypedilum and Tribelos larvae regularly began to crawl within the first 20 hours. During the early stages of recovery, specimens take up water until they either regain their normal appearance or reach a certain stable appearance. It is assumed that rehydration is complete in the first case and nearly com- plete in the second. The point where one or the other form of stability is reached is referred to in Table 2 as the time of ‘‘cessation of water uptake’’. In Polypedilum and Tribelos, rehydration is complete within a short period, but in Phaenopsectra the larva slowly enters a stage where it is still somewhat flaccid, though all of its vacuoles have been replaced by haemolymph. Full rehydration in Phaenopsectra probably takes place when diapause is completed and the larva regains full mobility. DISCUSSION Cocoon construction is associated with numerous species of chironomids that are aquatic during their entire larval life (Danks 1971). For these species, the cocoon-making behavior is an adaptation to cold winter condi- tions. As pointed out by Wiggins et al. (1980), a change in timing is all that would be needed to convert the behavior pattern of cold-tolerant species to one suited to temporary habitats. The timing of the building of cocoons by certain drought-resistant species is probably regulated by their state of development (Grodhaus 1980), not by receding water. The habit of larvae going into folded or curled positions in the substrate might be thought of as protection against too rapid movement of water through the tissues. Our results, however, indicate that water uptake is faster in the twice-folded Polypedilum than it is in the once-folded Tribelos, so it appears that the rate is not governed by the amount of folding. Polypedilum sp. near vanderplanki is anomalous in that its larva engages in MEM. AMER. ENT. SOC., 34 142 RECOVERY FROM DESICCATION OF CHIRONOMID LARVAE Fic. 1. Instar III] Polypedilum sp. near vanderplanki in a piece of substrate material ap- proximately 15 min. after wetting. folding without making a cocoon. Danks and Jones (1978) have suggested that a drawing of Hinton’s (Fig. 7, 1968) shows a cocoon of P. vanderplanki. This illustration could also be interpreted as a folded larva without a definite cocoon, comparable to the specimen, in Fig. 1 of the pre- sent paper. It is indeed possible that neither folding nor cocoon formation is obligatory. With regard to several Californian species, however, the present study and a previous one (Grodhaus 1980) indicate that viable larvae are in- variably folded and encased in cocoons. It is tempting to think that a cocoon might reduce the amount of water lost by a larva during drought, but we do not believe this to be the case. Be- coming dehydrated to an extreme degree may actually be advantageous. It has been suggested that dehydration permits P. vanderplanki to withstand higher temperatures than would otherwise be possible (Hinton 1960). Our visual observations, indicating severe dehydration in both Nigerian and Californian species, should be substantiated by moisture content measurements. A study by Jones (1975) indicates than another chironomid, Paraborniella tonnoiri Freeman, is only slightly dehydrated in its dormant state, during which this species can withstand drying of its rock-hollow habitat in Australia. G. GRODHAUS, E. C. BAY, L. KNUTSON 143 The protective value of the cocoon may simply be concealment for the Californian vernal pool chironomids. Larvae probably build their cocoons well in advance of the drying of the pools. In an exposed state they would become increasingly subject to attack as predaceous insects invade the pools. The habitat of the Polypedilum species of the presert study is similar to that described by Hinton (1950) for P. vanderplanki. Hinton noted that northern Nigerian rock cavities contain water intermittently during March and April and consistently from May until December. Especially during the first rains, these cavities are capable of rapid filling and drying. Larvae in- habiting such an environment would have to be quick to respond to hydrodynamic changes and would probably not be benefitted by having a diapause. As would be expected, Polypedilum is quick to regain mobility and to become fully hydrated in response to wetting alone. The Californian species inhabit large depressions on the ground, which require considerable precipitation before surface water appears. In a typical year, the earth becomes damp in October or November, after which an aquatic environment exists from December until March or April. Both the filling and drying processes take place slowly. Revival of the Californian species is slow, and the presence of water may only start the recovery pro- cess. The life histories of these species are probably regulated by diapause, in which larvae do not complete their metamorphosis until after they have been exposed to low temperatures (Grodhaus 1980), a pattern that fits the climate in northern California, in which cool weather and rain coincide. ACKNOWLEDGMENT We thank W.L. Murphy for offering many helpful suggestions during preparation of this manuscript. REFERENCES Danks, H.V. 1971. Overwintering of some north temperate and arctic Chironomidae. II. Chironomid biology. Can. Entomol. 103:1875-1910. Danks, H.V., AND J.W. Jones. 1978. Further observations on winter cocoons in Chirono- midae (Diptera). Can. Entomol. 110:667-669. Epwarp, D.H. 1968. Chironomidae in temporary freshwaters. Aust. Soc. Limnol. Newsl. 6:3-5. Gropuaus, G. 1976. Two species of Phaenopsectra with drought-resistant larvae (Diptera: Chironomidae). J. Kans. Entomol. Soc. 49:405-418. 1980. Aestivating chironomid larvae associated with vernal pools. P. 315- 322 in: Chironomidae. Ecology, systematics, cytology and physiology. (D.A. Murray, editor). Pergamon Press, Oxford. MEM. AMER. ENT. SOC., 34 144 RECOVERY FROM DESICCATION OF CHIRONOMID LARVAE GRODHAUS, G., AND G.L. ROTRAMEL. 1980. Immature stages of Polypedilum pedatum ex- celsius (Diptera, Chironomidae) from seasonally flooded tree-holes. Acta Univ. Carol. Biol. 1978:69-76. Hinton, H.E. 1951. A newchironomid from Africa, the larva of which can be dehydrated without injury. Proc. Zool. Soc. Lond. 121:371-380. 1960. Cryptobiosis in the larvae of Polypedilum vanderplanki Hint. (Chiro- nomidae). J. Ins. Physiol. 5:286-300. 1968. Reversible suspension of metabolism and the origin of life. Proc. R. Soc. B171:43-57. Jones, R.E. 1974. The effects of size-selective predation and environmental variation on the distribution and abundance of a chironomid, Paraborniella tonnoiri Freeman. Aust. J. Zool. 22:71-89. 1975. Dehydration in an Australian rockpool chironomid larva (Para- borniella tonnoiri). Proc. R. Entomol. Soc. Lond. A49:111-119. Mapper, M.C., D.M. ROSENBERG, AND A.P. WIENS. 1977. Larval cocoons in Eukief- feriella claripennis (Diptera: Chironomidae). Can. Entomol. 109:891-892. McLacutian, A.J., AND M.A. CANTRELL. 1980. Survival strategies in tropical rain pools. Oecologia 47:344-351. Mr.er, P.L. 1970. On the occurrence and some characteristics of Cyrtopus fastuosus Bigot (Dipt., Stratiomyidae) and Polypedilum sp. (Dipt., Chironomidae) from temporary habitats in western Nigeria. Entomol. Mon. Mag. 105:233-238. Wiceins, G.B., R.J. MAcKay, AND I.M. SmitH. 1980. Evolutionary and ecological stra- tegies of animals in annual temporary pools. Arch. Hydrobiol. Suppl. 58:97-206. Lotic Chironomids of the North Carolina Mountains Davip R. LENAT AND DANA REES FOLLEY Biological Monitoring Group North Carolina Division of Environmental Management Archdale Building Raleigh, North Carolina 27611 ABSTRACT. — Five years of chironomid data from streams and rivers in the North Carolina mountains have been combined to generate both a taxa list and some seasonal information. Approximately 172 species were recorded. The seasonal patterns observed for the more abun- dant taxa were often consistent with limitation by temperature and/or photoperiod, with 8 winter taxa, 21 summer taxa and 12 spring-fall taxa. However, 16 taxa had multiple peaks spread throughout the year. Temporal separation of some species may explain niche division within a diverse chironomid community; examples are given for 2 genera. The total chironomid community also showed seasonal variation with peaks in December-January, May- June and September. It is important to consider this normal seasonal variation when using in- dicator species concepts. INTRODUCTION The benthic macroinvertebrate community is frequently used as an indi- cator of water quality. Chironomidae comprise a large proportion of this community, but they are often treated as a single taxon in many routine surveys. This loss of information originates from both taxonomic uncer- tainties and difficulty in evaluating chironomid data. The latter problem can be addressed through the compilation of good ‘‘baseline’’ information. This investigation utilizes existing information collected by the North Carolina Division of Environmental Management to compile a list of larval Chironomidae collected from streams and rivers in the mountain regions of North Carolina. DATA SOURCES Quantitative information comes from 934 ‘‘kick’’ samples (Frost ef al. 1971) collected from 1978 to 1982. This particular collection technique can be utilized in a wide variety of lotic habitats and gives very consistent results (Pollard and Kinney 1979). Qualitative information is available from about 50 other sites. Chironomidae collected by Gurtz (1982) from streams in the Cowetta National Forest have also been utilized in this survey. 145 MEM. AMER. ENT. SOC., 34 146 NORTH CAROLINA LOTIC CHIRONOMIDS The quantitative information (kick samples) has been sorted by month to compile frequency (F) data. Monthly frequencies have been calculated as number of collections of a taxon divided by the total number of collections in any given month. The total number of collections in each month varied from 26 (November) to 170 (August). Average monthly abundance/sample (N) was also calcuated for the most abundant taxa. Both F & N can be used to generate the expected seasonal distribution of mountain chironomids. RESULTS AND DISCUSSION Table 1 lists larval chironomid taxa collected from North Carolina moun- tain streams and rivers. This list includes at least 172 species. Where several species (or species groups) are known to exist, but have not been routinely separated, the estimated number is given in parentheses. Many of the genera in Table 1 could not be easily dealt with at the species level. In particular, the Cricotopus/Orthocladius (C/O) group presents a special problem. We initially listed taxa in this group as C/O sp. 1, C/O sp. 2, etc. This ‘‘in-house’’ number has been retained in Table 1 to maintain consistency with prior publications. There are many species in the C/O group. Single samples often have 4-6 C/O group species. Coffman (1973) listed 23 Cricotopus/Orthocladius species for a Pennsylvania stream. We have recorded at least 26 species in the North Carolina mountains. Pollu- tion tolerance also varies widely within the C/O group, therefore, accurate identification is needed to use this group in environmental assessment. Table 2 lists the estimated chironomid taxa richness, by group, for our mountain samples. An estimate is also given for all DEM sampling, in- cluding Piedmont and Coastal Plain areas. The estimate for the entire mountain area (168 species) is not much greater than species richness recorded by Coffman (1973) for a single 3rd order stream (143 species). If the species richness values are converted to a percentage within each sub- family, the results are remarkably similar to other investigations of lotic Chironomidae. Lindegaard-Peterson (1972) and Boerger (1981) cite many studies with results comparable to North Carolina data. Seasonal trends for stream chironomids have usually been addressed through studies of adult emergence. Emergence patterns have been related to changes in temperature and photoperiod (Aagaard 1978). The seasonal patterns observed in Table 1 are also consistent with regulation by temperature and/or photoperiod. Several different types of seasonal patterns are illustrated in Figs. 1-4. Approximately 8 low temperature, (winter) species were found; examples are shown in Fig. 1. Orthocladius (Euorthocladius) sp. 3 is a eurythemal (broad temperature range) species with 3 maxima October to April, but it is D. R. LENAT AND D. R. FOLLEY 147 largely absent in July and August. Diplocladius cultriger is an example of a stenothermal (narrow temperature range) winter species, having a single maximum in November-December. This is in rough agreement with emergence data given by Coffman (1973) for this species, showing a single emergence period in February-April. Approximately 21 warm temperature (summer) species were found; examples are presented in Fig. 2. Cricoptopus bicinctus is a eurythermal species with maxima occurring from April to Oc- tober. Polypedilum illinoense and Robackia demeijerei have a more restricted seasonal distributions, with a pronounced maximum in July. Ap- proximately 12 species had spring-fall maxima; examples are given in Fig. 3. Another common pattern is seen in species having three or more maxima (approximately 16 taxa, see Fig. 4). This pattern may reflect either multiple generations or the combination of several species which are taxonomically inseparable. The seasonal curve shown for Diamesa could represent 1 sum- mer species and 1 spring-fall species. Likewise, the seasonal curve for Micropsectra could be attributed to a combination of a winter species, a spring-fall species and a summer species. The distinct seasonality of most chironomid species suggest that ‘‘in- dicator species’’ concepts must take into account normal seasonal variation. For example, some of the Cricotopus species which are associated with toxic conditions (C. infuscatus gr.) will normally be absent, or present in very low numbers, during winter months. The various seasonal patterns may be related to the concept of ‘‘niche division.’’ The chironomid community of a stream usually comprises a very large proportion (up to 50%) of the benthic macroinvertebrate species (Coffman 1973). How do so many species coexist without competitive ex- clusion? Investigators studying the zooplankton and phytoplankton com- munities in freshwater lakes have addressed a similar question. Hutchinson (1961) posed the ‘‘Paradox of the plankton,”’ contrasting the apparently uniform nature of the plankton environment with a highly diverse planktonic community. Subsequent studies have shown many different ways in which zooplankton species divide up the planktonic habitat. For ex- ample, Makarewicz and Likens (1975) showed that the major zooplankton taxa in Mirror Lake could be differentiated in relation to seasonality, depth preference and food size. Information is accumulating to suggest that chironomid species avoid competitive exclusion by similar mechanisms (Ramcharan & Patterson 1978). Several studies have shown spatial separation of chironomid species by means of substrate preferences (Jankovic 1979, Lindegaard-Petersen 1972)..Chironomids also may show different preferences along the stream contanuum. Penrose ef al. (1982) have demonstrated that certain MEM. AMER. ENT. SOC., 34 NORTH CAROLINA LOTIC CHIRONOMIDS 148 = =) I 7 Qh Hl Te ee eG Sas Mee (%)A €S 43 “ds snurouosryoordArotweq =e = = 6 Se a a I (%) A rd € ‘ds sadipuajoidAip = crane RC (ie lls ts ta a age aera aan Ged we (%) A BL 9€ "18 snaqny “OD = = = I = = (%}) 4 I I BULIe[G snWOUOITYIO}dAID xX — — ‘ds ewijodopelo —t ST ( 4 iS € C (4 (6 I a € (%)A $9 (G6 “dds snurouony5 (8€) ININONOUIHO = = = = i I = (%) 4 I I ‘dds eiAwiyo1aez, = = = (4 = a ple aa (%) A (4 (4 foyetqns “d x = — snjjoq snipepo1g xX — _ ‘ds sndAurjojIN xX — — ‘ds visivjen x = -- Byjasoyid “iu -7 x aad = R]Jasojidoau eiuipuniqey se LO Ot LO LY We Sr FE OL CO Wi Vl N Camm Cm SCO me SC, Ce wy OCmenhC. IGT MRCOs wkeSO (%)X -LZST OLZ (+p) “18 eidojadeyouoD x _— — ‘ds eyjoruipunig x = — el[aie] “VY x — — ejuefeied “vy = =, = @ = I = Sa) aa ee = (%)A £ € Tyoo][eut “Ww = = = = Zz ao = (%})A € € ByeUIO BIAWSaqeIqy (91) AVNIGOdANV.L Bed 7 II (0) 6 8 L 9 iS v € (4 I Pe92{10D (suon COX J310 #1BIOL — -9][09 #) Aduenbe1y yuo; Aq eieq “SIOATY 2 SUILIIIG UleJUNOW] PUI[OILD YON WOIJ exe] pluouolYyD “| I1av], 149 FOLLEY LENAT AND D. R. D. R. We el = = I yp §€ 5 7 — Zz I ae 4 A eS ne a a co a = = 6 im § S I = Ff ee ir a ee eee Z = @ Ss € LT oe = € G 10 80 pr €9 8€ I9 6FI SE — — 60 a = 6 S € yp 9 Se 6a € CE =—— iy oS aL = = — 2 Z - =— I i Of & po Pt 10 $0 10 $0 10 10 80 ef 9 61 Le 8 im z a 6 ” go el &@ ies a igs eee OR ye a SS Se ee ee eee i _— ae | = — cas PcOe 96 Pil ‘ds snowouoiysopnasg uwInyNd1aqn} * UINUSE]PDS * WINjoRy * asuaoull|l ° gyesaqyey * XBITRS A, Ay, Oy Ay Ay Ay AY (Z) winqotauos/ dastae * uinjn3ue win{IpsdAjog sadtaeyy “dq (z) ‘dds ensasdouseyug ‘dds sodipuajeied 2]e19]/BYO1sIU R][etuIOgJoneyeleg (uosyoer) | ‘ds -g aulpun ewyodopeperied SNWIOIYIOUOU SNOWIOUOIIYORIeg SIsua[epeAl “IU “JW snjoped ‘iu ‘dds sodipusjo11yy XNP sn[Niojjony ‘ds sadipuajoidA[H B[JeHjoeg “JU snueH SUBSIISIU SNWOUOIIYSOpUgq snjsopowloou *q SNSOAIOU *q ‘dds sadipua}o1s1q MEM. AMER. ENT. SOC., 34 NORTH CAROLINA LOTIC CHIRONOMIDS 150 ee L =. 9 S @ I = = (%)A oF 8Z TUBWYJOS (B9}]9]qNS) “L CO = 10 +0 $0 90 80 80 720 $0 70 10 N SSE 86 (+¢€) dds snsiejhuey L = ral €l 0% 1 SI 6 tl Ol 7 (%)A €s SO LE O1€ 16 EIl 6h FO SI Ht 6S 8 OF N @ZIL I€€ (+2) ‘dds snsieKurjoayy 5 nS TOES {Ts CUNY 20 ©) GI (0 CoG CARES) ES: COO (%)A Z = Z — — —- — — (%)4 L € ‘dds snsiejAuejeieg 60 v0 — 60 — €0 61 10 90 ST — 09 N 7801 911 Ge I ice Ma Guaen ah ieee Ce, SACS Ls (%)a (+p) ‘dds exjasdoso1yy] = = = — — _ — — — I — — (%)4 va I ‘ds eurjjoduiajsuoDd = = ¢ ‘ds‘9 = = 7 ds ‘9 = = ¢ = S € 9 S S @ @ = (%) A 16 97 “dds snsievjAuejope[D (81) INISUVLANV.L = = = (Pe = F = |] FF =| F (%) Ll BE ‘dds sojaquy = & I Z I => = = = = = gg (%)A SI ZI “ds snowouostysoj9Ng = = = = € € I ‘ = i == I (%)A I Ol ‘dds snowouoiyoouays - -—- -— => tr =! = fr = = Wis Ip ZI snjAy BLayIaeS i =— = ¢ 8 Oo € m ¢ = = = (%)A 6S pe foraftowiap “yf = = JOBIAR[D BIYDeGOY ZI Il Ol 6 8 L 9 S € Z I paqoa[10D (suon uOoXe | # [BIOL -J9|[09 #) Aouonbea14 yuo; Aq eieq (Quod) | alav 151 D. R. LENAT AND D.R. FOLLEY OT VT GG 14 (6 14 ££ 80 SI £0 SI (Zé a 60 cl €7 c0 eI eI 60 Of 90 8I Of 6€ N aA NN i>) = Oo ™ Ol 0 Cl cl s0 €¢ £0 OI (% A OLIT L8L SI ILI 6E1 TE Ol “ds (Snipe[s090}sON) sndojoo1lD ss “ds O/D (O/O=) snipepoyjiO/sndojoo1D (+7) ‘dds einauouAio9 “ds snipepojoeyoD ‘dds snipepoipie9 ‘ds snipejsouseydodig SUOIJIARLJ “G (9181) BJsapoul “g (91e1) eAIed “g “dds eiyug (16) AVNIIGV IOOHLUO Ja1aez eseyoduwiAs BOORAT[O BSOUILIPOIG— snueuIsuo] “qd Ipoed eljseyyo0d ‘dds ensesedg BAIN} BsaUIOJUOPO (+¢€) ‘dds esowviq (6) AVNISAWVIG ‘dds eyjourjjodwiais OSPATJUOW “JO BUTT[IdUID}S MEM. AMER. ENT. SOC., 34 NORTH CAROLINA LOTIC CHIRONOMIDS 152 = pa = C £ = C C 9 alate I (%)A L9 at pyoeqor “J9 (‘O) “O aI “ds O/D Te Le tO 1O 10 = €0 Gl Se ce CO OS N 100€ 871 “18 snqyeiquin4o (‘O) ‘oO i? ol © € py § 6 8 Se 81 OO + £(%)a ol ds O/9 L = P I = = | &% G&G. &¢ G GT Gx 86S OL "13 snualop ((O) snipefsoyiO L-dsO/9 = SS) a oe oS oS (%) A “9€ S SIQSOATAS “1U (snipejsos]) “5 6 ‘ds O/D = = = == = = = {I v = (%)A S 7 I ds (Dd) “Oo ze ds O/D eee SS eS ee SS (% a 02 € Blosestdy ("D) “O : 0s ‘ds O'D = = = = » OH = & SS (%)a 8b ay “13 SISUaIIAIA ("D) “OD 9p “ds O/D — — Z I ie _— —_- — - —- = — (%)4 tr L snede1IpulfAd “IU (‘D) *O pI ‘ds O/D = @0 Fi Cl GO Fe st @ SO SX = = N LEL LOI “13 saduea (9) ‘OD = Gi tt 8 We Mm im &§ iblob oo - (%)a 9 -ds Q/D 10 @EL 2% 80 61 LI It 8 60 10 — £¥0 N €Ssz IST “13 snqeosnjut ("D) *O L S&S Ww mW ie ( GS Gi 2 y (%)A sds oso Ol = SG €e6é CE Si FOC Se 6 — = N £092 661 snjoutog (*D) sndojoor9, (0) ee i iW 8 Ge ie GS We ff & v (%)A 1 ‘ds O/O pa EN RN aT Beqd Zl Il Ol 6 Chea 9 s nen CURE I payayop = (suo UOXe | ro # [B10 —-99]]09 4) AdJuonbe1y yuo; Aq ejeq a a ee ee ee (7,u05) | a1av I5)5) FOLLEY LENAT AND D. R. D. R. N tT -— = = = =| = |= =| FF L — — — —— — = — — = v — 2 = = = = ¢ = I I Z cs = = I I = > & = = = = = 0 CG SH 80 SE 30 = = = € z 8 a bl Gl 7 Ol iT — I co 2 GO TO TO SO ED GE OF Ih #9 86 % 6 CH. xc Sl g & OO we & ee € TT o£ & | z 7 - —- | = ¢€ = I = BY 9 Gin CReEy I p I 5 Z = =, = = 9 — ae <= te SS A ee eae | Se Re See ES ae ee — 9 eS ez ie 6 Sg Gl OO tO 10) — @o GF =] = Yo, a € € feo See Zt) ae ee eS p (%)A (%) (%)A (%)A (%) (%)I (%)A (%)a (%)A (% A IZ. (%)H Le 67 ST Ocs C6LI vOE 101 CCE TT Lt l€ 1X6 87 8p L ‘ds (a) 70) 19 ds O/D 9 ds (q) Oo Is “ds O/9 ¢ ds (gq) Oo 6z “ds O/D yds (ga) Oo 0z ds O/D € ds (q) Oo €1 ds O/O € ds O/O I -ds(¢q@-‘o z ds O/D € ‘ds q] edky (-g) ‘Oo vle “ds O/D z ds jj adky (a) oO z9 ‘ds O/D “ds [Jy odAy (snipejsoyjiong) “Oo Le ds O/D € ds(O) ‘Oo 9¢ “ds ‘O/D Z ‘ds ¢O) ‘Oo se ds O/D I -ds(O) Oo €€ “ds O/D (Z) “43 snqtstu (°O) ‘oO Ly ds O/D “18 fayreP9 (°O) “O rs ds O/D MEM. AMER. ENT. SOC., 34 NORTH CAROLINA LOTIC CHIRONOMIDS 154 N Ol lon = = = ¢ = = & SO 9 Ol Ol Tea = €0 0 = ft TL = 50 0 = 6 9 €0 SI 80 € NIG — 9 —_— L 9 ¢ yUoW Aq eed + a0) 8L7 LEE p9E P9199[[0D # [210.1 ont Icl vL 98 0 (suo -d9|[[09 #) Aduanboif (+ ¢) ‘dds sokydounry ‘ds eijqjiwisousry ‘dds snuseqoipAyH ‘ds snipe[oossii}019}0}4 “ds eyjorusjayH q snipepoury “iu snueh VY snipepouenN “iu snuep RIJolUs[oH “JU snuaD ‘ds snipepooyqi0aH (1) ‘13 eueyuoWOpnesd “gq (Z) 13 enseyjjod -q (Z) “43 eoTUOAap “gq (1) “43 puryerq “gq (p) ‘418 stuusdueyo “gq (Z) “48 reopeotaaiq Bl etosjorng” ‘ds snipejoosrodq JOsII}[NO snipe[sojdiq uoxe (),u0d) | 41aVv 155 FOLLEY LENAT AND D. R. D.R. vl co we ie We sh OF AM li ee > = € z = =| |= = |= fF = - - =—- — | Sah eee ae ee ot ie Sy ee { I I = I = —~ I (4 L € Sle ile peel ae E = = ¢ GxERE SS -G-= E> = = = = = = = = & é I _ = = = = I € I I — =" ¢ LOM = Se gh Gi Ee 60 #0 Of OS CE HO OWL ie hp te © Ww Ge @ & I —- F — | = i - — tl VI @O 10 = = 10 £0 60 I © 6 OO it € ® ie. & EL = = = = = = = i € 6 & © § I a Site AC ae IC eallgeh B Sali alte 2s SN a 61 ‘ds snipe[s0yj1OUAS “ds eqiwis B[Neoljap “1U eNWsosyY pds -y € ds -y Z ds uy 1ydeqos sndojoo909y4y SUIR[NWIS “JO “q ‘dds snipepooijsasq eijanbiy “iu “g | ‘ds e]jauesjonjeieg € ‘ds qd Z ds ‘d (g1499qQpun] =) | “ds SNUOUDOLNOWIeIe /snipejouseydeieg ‘ds sndojoouseied | ‘ds snipepojeeyorieg ISOUMOP “NJ snonyeq “IU "N (¢) ‘dds snipejoourn sadiosnj “Iu “JA ds snwiousonayy (,SeupioD,, =) ‘ds snipe[osodoT MEM. AMER. ENT. SOC., 34 nN Q eee a ee ees, ee ee SSS SS SS ee g ee iL 2 @ £ oe Se (%) A 0@ LI siqejnoe odAy Z, aeulIpe|ooyO o ee ee, ee er le 7 € xed sndoioj4x —_ —<——<$— — a 90 O77 Off Sh OT 80 fF SO 80° 90 tO SI N L697 607 ({) “18 sedisojoosip “L O i GOCE CO | ple = eee = 5 08 lh SS? Vl SO Tl tr Sl 60 VO SO I N 88h 08Z (1) “48 eorseaegq : hm OS @ @ oO C oO BS OC Ww Ol C& (%)A BIUdI9A.L Z E™@ TO OF Ci £0 CO YO LO BO — = BU N 106 OPI (+2) ‘dds ejjatuewoucryL | SS SS nn eee ee ee S wed ZI Il Ol 6 Sy ge 9 ¢ tee sae iC I paso = (suon UOXe | a 1410 # [RIO ——-99][09 #) ) Aduonboly{ se] (20 te eee S) Z (4,u09) | a31aV 156 D. R. LENAT AND D. R. FOLLEY 157 WINTER MAXIMA SP 3 > = ee D. CULTRIGER Fic. 1. Some chironomid species with winter maxima. TABLE 2. Approximate number of larval chironomid taxa collected in North Carolina freshwater lotic systems; DEM collections 1978-1982. Mountains All N.C. Collections Group # % # % Tanypodinae 16 10 28 12 Chironomini 38 23 56 24 Tarytarsini 14 8 21 9 Diamesinae 9 5 11 5 Orthocladiinae 91 54 117 50 Total 168 100 233 100 chironomid species prefer either streams (2nd order) or rivers (4th-5th order) within a single North Carolina watershed. Chironomids may also avoid competition by means of temporal separa- tion. Many examples may be drawn from Table 1. Fig. 5 illustrates the MEM. AMER. ENT. SOC., 34 158 NORTH CAROLINA LOTIC CHIRONOMIDS SUMMER MAXIMA <~ONZSMCOMATN 3s JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC C. BICINCTUS ------ P. ILLINOENSE vote cce ese ee R. DEMEIJEREI Fic. 2. Some chironomid species with summer maxima. TABLE 3. Number of maxima (frequency data) classified by sub- family and season. Season Subfamily Winter Spring-Fall Summer Tanypodinae 1 1 1 Chironominae 6 8 16 Diamesinae 2 2 2) Orthocladiinae 18 23 13 TOTAL 26 33 31 seasonal distribution of the 5 most abundant Polypedilum species (from Table 1). Polypedilum convictum gr. (includes P. aviceps) is a warm eurythermal taxon that is common from April to October. The other less common species show a distinct separation in the timing of the first spring D. R. LENAT AND D. R. FOLLEY 159 SPRING-FALL MAXIMA 58 48 32 10 000 WAS = 07 Fic. 6. Distribution of Fleuria lacustris in Hjarbak Fjord during 1968-81. Density in nos. m~ C. LINDEGAARD AND E. JONSSON 181 Polypedilum bicrenatum 1971 0-10 11-100 ==| 101-1000 1001-10 000 >10000 1981 2 km Fic. 7. Distribution of Polypedilum bicrenatum in Hjarbek Fjord during 1968-81. Density in nos. m”?. MEM. AMER. ENT. SOC., 34 182 CHIRONOMID SUCCESSION IN HJARBAEK FJORD species of oligochaets, chironomids and molluscs appears, which later develops to (4) the final stage where an ‘“‘equilibrium’’ of the different species is reached, depending on trophic status and morphometry of the im- poundment. The present fauna community in Hjarbaek Fjord, 15 years after the dam construction, has not yet reached an ‘““equilibrium”’ stage. A diverse invertebrate fauna was found in the lake a few years after the establishment of the dam, but later it was replaced by an oligochaet- chironomid dominated community, whose density is still increasing. Nuisance problems by midges. — Nuisance problems caused by midges have been reported from a number of sites (see Grodhaus 1975 for references). Outbreaks of chironomids have occurred in brackish water (e.g., the lagoons of the German Baltic Sea Coast, Thienemann 1954), or in brackish water turning into freshwater (e.g., IJsselmeer, Kruseman 1935, v.d. Torren 1939). Recently, Beattie (1981) described an incident of a chironomid plague at Wolderwijd, a border lake to IJsselmeer. There the invertebrate community was assumed to be in ‘‘equilibrium’’, with the chironomid larval density on average 1,000 ind. m~ (mainly Chironomus plumosus). Two reasons for such low larval densities causing midge nuisance were stated. Firstly, the mean depth of the lake is 2m, hence most of the 18 km? lake bottom is a suitable habitat for chironomid larvae, so the total numbers of adult chironomids emerging from the lake is high. Sec- ondly, the adult midges accumulate at specific localities along the shore, at- tracted either by light or by swarming points. In Hjarbeek Fjord, the density of larvae is much higher than in Wolder- wijd, and suitable larval habitats are found over 90% of the lake. Prevailing westerly winds concentrate the adult midges in certain areas. The dominating chironomid species have two or three generations per year (Lindegaard and Jonsson, in prep.), and consequently large numbers of adults are present from late May to September. Two plague types can be recognized at Hjarbaek Fjord: (1) All but one of the chironomid species have normal swarming behavior, forming under suitable weather conditions numerous swarms near the lake. (2) Fleuria lacustris shows a very different behaviour (Schlee 1980a,b). Newly hatched adults fly to the shore and cover the substrate in vast masses. A special social contact, in which the individuals touch each other, ensures that the formations do not break up. These formations may be disturbed either by man or by wind, but the midges will immediately try to fly back to reestablish their contacts. At certain times, F. /acustris may cover the shore surroundings over large parts of Hjarbaek Fjord. Newly hatched adults may remain for some time at their pupal exuviae, and from there they may C. LINDEGAARD AND E. JONSSON 183 Cladotanytarsus spp. 1971 y 0-10 E= | -100 =| 101-1000 = 1001-10000 > 10000 1981 2 km Fic. 8. Distribution of Cladotanytarsus spp. in Hjarbaek Fjord during 1968-81. Density in nos. m™. MEM. AMER. ENT. SOC., 34 184 CHIRONOMID SUCCESSION IN HJARBAEK FJORD directly invade boats in immense numbers, as they perceive the boats as parts of the shoreline. We consider several factors contribute to the serious chironomid plague at Hjarbaek Fjord. These are: (1) the average density of 32,000 larvae m” is higher than normally found in lakes. (2) No severe oxygen depletion occurs in the lake for long periods, and the sandy substrate making up the lake bot- tom remains undisturbed in stormy weather, thus the larval tubebuilding and food uptake is not disturbed as in shallow mud bottom lakes (Jonasson and Lindegaard 1979). Therefore, more than 90% of the bottom area in Hjarbaek Fjord (i.e., 22 to 23 km’) is suitable habitat for chironomid larvae. (3) Disappearance of other invertebrates and, to some extent, also of fish populations has resulted in low predation and may have reduced in- terspecific competition. The emergence of adults is, therefore, relatively high compared to other lakes. (4) Along with normal swarms, the unusual appearance and the special behaviour of F. /acustris result in a serious nuisance of crawling midges. ACKNOWLEDGMENTS The authors wish to thank Palle Uhd Jepsen, The Game Reserve Office, and Sigurd Christensen, School of Ecological Agriculture, for placing at our disposal their samples from 1968 to 1973 and from 1976 to 1977. Without this help, the description of the faunal succession would have been impossible. Knud Rasmussen, Water Board of Viborg Amt, is thanked for the 1981 samples. Finally, we wish to thank David Gordon, University of Western Australia, for help with linguistic corrections. REFERENCES ALEKSEVNINA, M.S. 1974. Rost i produktsiya massovykh vidov khironomid (Diptera, Tendipedidae) avandel’ty Volgi. (Growth and production of mass species of chironomids (Diptera, Tendipedidae) in the Volga avandelta). Zool. Zh. 53:720-727. ANDERSEN, F.S. 1949. On the subgenus Chironomus. Studies on the systematics and biol- ogy of Chironomidae II]. Vidensk. Medd. dansk naturh. Foren. 111:1-66. BEATTIE, D.M. 1981. Investigations into the occurrence of midge plagues in the vicinity of Harderwijk (Netherlands). Hydrobiologia 80:147-159. CHRISTENSEN, S. 1979. WHjarbaek Fjord 1966-78. Thesis, Freshwater Biological Labora- tory, University of Copenhagen. FitTtKAU, E.J. AND F. Reiss. 1978. Chironomidae. Jn: Illies, J. (ed.): Limnofauna Europaea, 2nd edition: G. Fischer, Stuttgart, 404-440. GropHaus, G. 1975. Bibliography of chironomid midge nuisance and control. Califor- nia Vector Veiws 22:71-81. C. LINDEGAARD AND E. JONSSON 185 Gr@NTVED, J. 1960. On the productivity of microbenthos and phytoplankton in some Danish fjords. Medd. Danm. Fiskeri- og Havunders. N. S. 3:55-92. HavinGA, B. 1941. De veranderingen in den hydrographischen toestand en in de macro- fauna von het IJsselmeer gedurende de jaren 1936-1940. Meded. Zuidersee — Comm. ned. dierk. Vereen 5:1-23. JepsEN, P.U. 1976. Feeding ecology of Goldeneye (Bucephala clangula) during the wing- feather moult in Denmark. Dan. Rev. Game Biol. 10, 4:1-22. 1978: Vildtreservatet Hjarbaek Fjord (The game reserve Hjarbaek Fjord). Danske Vildtunders¢gelser 30:1-68. JOnasson, P.M. AND C. LINDEGAARD. 1979. Zoobenthos and its contribution to the metab- olism of shallow lakes. Arch. Hydrobiol. Beih. Ergebn. Limnol. 13:162-180. KRUSEMAN, G. 1935. 8 mededeeling over Tendipedidae. Tijdschr. Entomolog. 78: LX- LXII. LENZ, F. 1933. Untersuchungen zur Limnologie von Strandseen. Verh. Internat. Verein Limnol. 6:166-177. 1954-62: Die Metamorphose der Tendipedinae (Chironomidae). Jn: Lindner, E. (ed.): Die Fliegen der palaearktischen Region 13c:139-260. LINDEGAARD, C. AND E. JONSSON. (in prep.) Abundance, population dynamics and produc- tion of Chironomidae (Diptera) in Hjarbaek Fjord, Denmark, during a period with mass developments. McLacuian, A.J. 1974. The development of chironomid communities in a new temperate impoundment. Ent. Tidskr. 95 suppl.: 162-171. MorDUCHAI-BOoLTovskol, F.D. 1961. Die Entwicklung der Bodenfauna in den Stauseen der Wolga. Verh. Internat. Verein. Limnol. 14:647-651. Muus, B.J. 1967. The fauna of Danish estuaries and lagoons. Medd. Danm. Fiskeri- og Havunders., N. S. 5:1-316. NurSALL, J.R. 1952. The early development of a bottom fauna in a new power reservoir in the Rocky Mountains of Alberta. Can. J. Zool. 30:387-409. PARMA, S. AND B.P.M. Kress. 1977. The distribution of chironomid larvae in relation to chloride concentration in a brackish water region of The Netherlands. Hydrobiologia. 52:117-126. PATERSON, C.G. AND C.H. FERNANDO. 1970. Benthic fauna colonization of a new reservoir with particular reference to the Chironomidae. J. Fish. Res. Bd. Canada. 27:213-232. SCHLEE, D. 1980a. Ungewohnlicke Varianten des Sozialverhaltens bei Zuckmiicken (Dip- tera: Chironomidae). Stuttgarter Beitr. Naturk. Ser. A 336:1-12. 1980b. Besonderheiten der Biologie und Morphologie von Fleuria lacustris (Diptera: Chironomidae). Stuttgarter Beitr. Naturk. Ser. A 340:1-23. Suitova, A.I. 1973. Novye i maloizvextnye Chironomidae (Diptera) fauny SSR. II. (New and little known Chironomidae (Diptera) of the USSR fauna. II). Inf. Byull. Inst. Biol. vnutr. Vod. 17:48-59. THIENEMANN, A. 1954. Chironomus. Leben, Verbreitung und wirtschaftliche Bedeutung der Chironomiden. Die Binnengewdsser 20:1-834. ToRREN, G. VANDER. 1939. Voorlopig verslag van de commissie inzake de muggenplagen om het IJsselmeer. Den Haag, Algemeine Staatsdrukkerij. MEM. AMER. ENT. SOC., 34 L wor, ; ies z ie * Effects of the Squaw Rapids Hydroelectric Development on Saskatchewan River Chironomidae (Diptera) P. G. MASON AND D. M. LEHMKUHL Department of Biology University of Saskatchewan Saskatoon, Saskatchewan Canada S7N 0WO0 AssTRACT. — Analysis of the Saskatchewan River chironomid fauna in the vicinity of the Squaw Rapids Hydroelectric Development showed that this family represented more than 2/3 of the total number of species of aquatic insects. Comparison of the upstream and downstream fauna showed distinct differences in the chironomid community. Downstream, some upstream species were eliminated and replaced by others, some species became less abundant, others became more abundant and still others remained unchanged. A number of environmental parameters appear to influence the chironomid community. An altered thermal pattern caused delay in the onset of spring emergence of most species examined. INTRODUCTION The use of benthic community structure to detect effects of impound- ments on riverine ecosystems has been discussed by various authors (Gore, 1980; Fraley, 1979; Williams and Winget, 1979; Brooker and Hemsworth, 1978; Armitage, 1978, 1977; Armitage ef. a/l., 1978; Merkley, 1978; Ward and Short, 1978; Young, ef. al., 1976; Goodno, 1975; Olson and Tarter, 1974; Ward, 1976a, 1976b, 1974; Lackey, 1973; Lehmkuhl, 1972; Spence and Hynes, 1971; Hilsenhoff, 1971). Many of these studies, however, have not identified all groups of the community to the species level. The Chironomidae in particular have been neglected, yet they comprise a signifi- cant part of the benthic community. Lehmkuhl (unpublished) studied the benthos of the Saskatchewan River in the vicinity of the Squaw Rapids hydroelectric development. He found approximately 62 species of aquatic insects, excluding the Chironomidae (Table I). The study reported herein of the Chironomidae in this part of the river was undertaken to complement Lehmkuhl’s information. The objectives were to (1) discover which species of chironomids occur upstream, within and downstream of the hydroelectric development and (2) define the effect of the development on the chironomid community. 187 MEM. AMER. ENT. SOC., 34 188 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE TABLE 1. Insects other than Chironomidae known to occur in the Saskatchewan River associated with Tobin Lake. Ephemeroptera Ametropodidae Ametropus albrighti Traver Baetidae Baetis tricaudatus Dodds Baetis sp. 2 Baetis sp. 3 Centroptilum bifurcatum McDunnough Centroptilum rivulare Traver Pseudocloeon sp. 1 Pseudocloeon sp. 2 Caenidae Brachycersus prudens (McDunnough) Caenis tardata McDunnough Ephemerellidae Ephemerella inermis Eaton Ephemeridae Ephoron album Say Heptageniidae Anepeorus rusticus (McDunnough) Heptagenia elegentula Eaton Heptagenia flavescens (Walsh) Heptagenia pulla Clemens Heptagenia solitaria McDunnough Macdunnoa nipawinia Lehmkuhl Pseudiron centralis McDunnough Rhithrgena jejuna Eaton Stenacron interpunctatum (Say) Stenonema terminatum Walsh Leptophlebiidae Leptophlebia cupida (Say) Traverella albertana (McDunnough) Metretopodidae Metretopus borealis Eaton Siphloplecton interlineatum (Walsh) Oligoneuriidae Lachlania saskatchewanensis Ide Tricorythidae Tricorythodes corpulentus Kilgore and Allen Tricorythodes minutus Traver Plecoptera Chloroperlidae Hastaperla brevis (Banks) P.G. MASON AND D.M. LEHMKUHL TABLE 1. (cont’d) Perlidae Acroneuria abnormis (Newman) Perlodidae Tsogenoides colubrinus (Hagen) Isoperla bilineata (Say) Isoperla longiseta Banks Pteronarcidae Pteronarcys dorsata (Say) Odonata Gomphidae Gomphus intricatus Hagen Gomphus notatus Bambur Ophiogomphus severus Hagen Hemiptera Corixidae Trichoptera Brachycentridae Brachycentrus occidentalis Banks Glossosomatidae Protoptila tenebrosa (Walker) Hydropsychidae Cheumatopsyche spp. Hydropsyche gutatta Pictet Hydropsyche occidentalis Banks Hydropsyche placoda Ross HAydropsyche recurvata Banks Hydroptilidae Hydroptila sp. Mayatrichia ayama Mosely Leptoceridae Athripsodes arielles Denning Ceraclyia sp. Nectopsyche diarina (Ross) Polycentropidae Neureclipsis bimaculata (Linnaeus) Psychomyiidae Psychomyia flavida Hagen Coleoptera Dytiscidae Diptera Simuliidae Simulium arcticum Malloch Simulium luggeri Nicholson and Mickel MEM. AMER. ENT. SOC., 34 189 190 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE TABLE 1. (cont’d) Simulium meridionale Riley Simulium rugglesi Nicholson and Mickel Simulium venustum Say Tabanidae Tipulidae StuDY AREA The Saskatchewan River system originates in the Rocky Mountains in western Canada. The upper reaches consists of two branches, the North and South Saskatchewan Rivers which flow more than 1200 kilometers eastward before converging to form the main Saskatchewan River which flows east for another 544 kilometers to Lake Winnipeg. The study focused on the Saskatchewan River in the vicinity of Tobin Lake (Figure 1), a hydroelectric development reservoir in central eastern Saskatchewan (N. lat. 53 30’; W. long. 103 30’). The lake, completed in 1962, is 74 kilometers long and covers an area of 30,000 hectares. The maximum water depth of 25.5 meters occurs at the dam. Water is taken from the lake through conduits submerged 4.8 meters below the surface, leading to the power canal. The water travels 4.8 kilometers in the power canal to the generating station where it passes through turbines and back into the Saskatchewan River. METHODS Five sampling stations, located about 85 kilometers upstream from the dam, within the impoundment and 6 and 23 kilometers downstream from the dam were chosen (Fig. 1). Collections of chironomids were made at each station during 1979, 1980 and 1981 to discover the species in the system. The pupal exuvial method was used during 1980 and 1981 to characterize com- munity composition and emergence patterns in the River (Rossaro and Fer- rarese, 1980; Laville, 1979; Wilson, 1979, 1977; Rossaro, 1978; Wilson and McGill, 1977; Wartinbee, 1976; Coffman, 1973; Wilson and Bright, 1973; and Brundin, 1966). Samples were taken from June to August in 1980 and April to November in 1981. Specimens were obtained in a plankton net with a mesh size of 80 pm, in two ways: (1) the net was held in flowing water for 10 minutes with 2 of the diameter below the surface and; (2) the net was held in a similar manner but dragged for 10 meters. Each sample was emp- tied into a pint jar of 95% ethanol and taken to the laboratory for analysis. Temperature, oxygen and current velocity were measured following each collection at each site. P.G. MASON AND D.M. LEHMKUHL 191 ii Hs Rh ; it ie Pe ati ais th Tbh Loke depth in metres a Z < ) a Hs, ws > ow : : a =< < . iz S Zz NS ie Ip = . O fF = < “ (7p) < (7p) Saskatchewan River and Tobin Lake. Fic. 1. 192 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE Samples were sorted in the laboratory, the number of each species counted and representatives placed on microscope slides for identification. Large samples were subsampled using the method described by Wartinbee (1979). Environmental data for each site was obtained from Saskatchewan En- vironment for the period (1979-1981). Additional information was taken from Durban (1981), Sawchyn (1974) and Royer (1969). Based on these data, and the surface temperature, oxygen and current velocity measurements taken during the study, a matrix was derived for principle components analysis (PCA) to determine differences between sampling sites and the parameters responsible for these differences. The PCA was done us- ing the SPSS statistical package program (Nie ef. a/., 1975). G88 Tanypodinae C4 Orthocladiinae [5 Tanytarsini De Diamesinae HB Chironomini 1980 June 5-Aug 10 13% 39% —De1s% ITN te 001 % - 17% 221%— DS K 619 % 127% ~88 2% UPSTREAM 6km DOWNSTREAM 23 km DOWNSTREAM 1981 April9-Nov5 -——21% 16% 14% 09% r—De03% //r— De 01% 206% 63% 98% 21%— f \ 23 4%— +912 % UPSTRE AM 6km DOWNSTREAM 23 km DOWNSTREAM Fic. 2. Composition of the chironomid population at each station during 1980 and 1981 (total number for each site is approximately 30,000). P.G. MASON AND D.M. LEHMKUHL 193 RESULTS AND DISCUSSION Chironomid Community — One hundred and thirty-one chironomid species (Table II) were collected from the Saskatchewan River and Tobin Lake. Nine of these occurred only in the lake and are typical lentic species (Table III). Thus, 122 species of chironomids were collected from the Saskatchewan River proper. Based on the species of aquatic insects col- lected from the study area, the Chironomidae constitute more than 2/3 of the fauna. Figure 2 shows the relative sizes of the populations of each major chironomid group for the entire period collected during 1980 and 1981. The difference in abundance of the groups, particularly the Chironomini, bet- ween years is due to the shorter sampling period in 1980 (June 5 to August 9) compared to that in 1981 (April 9 to November 5, the entire ice-free period). The Orthocladiinae tended to dominate in the early and late parts of the field season in 1981, while the Chironomini dominated during the middle period. As it is thus important to take samples during the entire ice- free season only the 1981 data will be discussed in detail. The fauna in the Saskatchewan River changes from a Chironominae- Orthocladiinae community upstream to one dominated by the Orthocla- diinae downstream (Fig. 2). The number of individuals of Tanytarsini is reduced downstream but at the same time there are more species downstream (Table IV). The number of individuals and species of Chironomini are fewer downstream compared to upstream while the Or- thocladiinae are rather constant in number of species at both locations (Table IV). Examination of weekly community structure gives additional informa- tion on the changes occurring. There is a clear delay of about 2 weeks in the onset of spring emergence downstream (Fig. 3), with 13 species emerging in the 2nd week of May upstream but only 2 downstream. However, from the beginning of June until the end of the field season the numbers of species emerging at upstream and downstream sites show no apparent difference. Depending on the week selected there may be more species emerging upstream from the reservoir than 23 kilometers downstream and vice versa. The number of species of major groups during the course of the 1981 field season is shown in Figures 4-6. Upstream, emerging Orthocladiinae dominate during the cooler parts of the field season (April, May, September, October and November) while emerging Chironominae are most abundant during the warmer months. Downstream, however, the Or- thocladiinae dominate throughout the season. The Chironomini are abun- dant during the summer but are not as abundant as at the upstream site. MEM. AMER. ENT. SOC., 34 194 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE TABLE 2. Chironomidae in Tobin Lake and adjacent stretches of the Saskatchewan River. Tanypodinae Pentaneurini Ablabesmyia (s.s.) sp. 1 Ablabesmyia (s.s.) sp. 2 Ablabesmyia (s.s.) sp. 3 Conchapelopia (s.s.) telema Roback Rheopelopia sp. Thienemannimyia senata (Walley) Procladiini Procladius (s.s.) denticulatus Sublette Procladius (s.s.) freemani Sublette Procladius (Psilotanypus) bellus (Loew) Diamesinae Diamesini Diamesa cf. cinerella Meigen Potthastia longimana Kieffer Chironominae Chironomini Chernovskiia amphitrite (Townes) Chironomus (s.s.) anthracinus Zetterstedt Chironomus (s.s.) decorus Johannsen Chironomus (s.s.) plumosus (Linnaeus) Chironomus (s.s.) sp. 1 Chironomus (s.s.) sp. 2 Chironomus (s.s.) sp. 3 Cladopelma sp. Cryptochironomus digitatus Malloch Cryptochironomus scimitarus (Townes) Cryptochironomus stylifera Johannsen Cryptochironomus sp. 1 Cryptochironomus sp. 2 Cryptochironomus sp. 3 Cryptochironomus sp. 4 Cryptotendipes darbyi Sublette Cyphomella gibbera Saether Demicryptochironomus sp. Dicrotendipes nervosus Staeger Endochironomus nigricans Johannsen Glyptotendipes (Phytotendipes) lobiferus (Say) Glyptotendipes (Phytotendipes) paripes (Edwards) Harnischia curtilamellata (Malloch) Microtendipes caducus Townes Microtendipes pedellus (De Geer) Nilothauma babiyi (Rempel) Parachironomus abortivus (Malloch) TABLE 2. (cont’d) P.G. MASON AND D.M. LEHMKUHL 195 Parachironomus frequens (Johannsen) Paracladopelma nereis (Townes) Paracladopelma winnelli Jackson Paracladopelma sp. 1 Paracladopelma sp. 2 Paracladopelma sp. 3 Paralauterborniella nigrohalterale (Malloch) Paratendipes albimanus (Meigen) Phaenopsectra obediens (Johannsen) Polypedilum (s.s.) sp. nr. aviceps Townes Polypedilum (s.s.) convictum (Walker) Polypedilum (s.s.) fallax (Johannsen) Polypedilum (s.s.) illinoense (Malloch) Polypedilum (s.s.) laetum (Meigen) Polypedilum (s.s.) obtusum Townes Polypedilum (Tripodura) digitifer Townes Polypedilum (Tripodura) scalaenum (Schrank) Polypedilum (Tripodura) sp. 1 Polypedilum (Tripodura) sp. 2 Polypedilum (Tripodura) sp. 3 Robackia claviger (Townes) Robackia demeijerei (Kruseman) Saetheria tylus Jackson Stenochironomus hilaris (Walker) Stictochironomus sp. 1 Stictochironomus sp. 2 Stictochironomus sp. 3 Xenochironomus (Anceus) scopula Townes Chironomini Genus 1 sp. Chironomini Genus 2 sp. Tanytarsini Cladotanytarsus sp. nr. viridiventris (Malloch) Cladotanytarsus sp. Constempellina sp. Micropsectra nigripila (Johannsen) Micropsectra dives (Johannsen) Micropsectra polita (Malloch) Micropsectra sp. Paratanytarsus confusus Palmen Paratanytarsus sp. nr. dimorphis Reiss Paratanytarsus laccophilus (Edwards) Paratanytarsus sp. nr. natvigi (Goetghebuer) Paratanytarsus sp. 1 Paratanytarsus sp. 2 Rheotanytarsus exiguus (Johannsen) Rheotanytarsus sp. | Rheotanytarsus sp. 2 MEM. AMER. ENT. SOC., 34 196 TABLE 2. HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE (cont’d) Stempellinella sp. Tanytarsus glabrescens Edwards Tanytarsus guerlus Roback Tanytarsus sp. 1 Tanytarsus sp. 2 Tanytarsus sp. 3 Chironominae Genus 3 sp. Orthocladiinae Acricotopus sp. Cardiocladius sp. Cricotopus (s.s.) bicintus (Meigen) Cricotopus (s.s.) curtus Hirvenoja Cricotopus (s.s.) politus (Coquillett) Cricotopus (s.s.) slossonae Malloch Cricotopus (s.s.) triannulatus (Macquart) Cricotopus (s.s.) cf. tremulus (Linnaeus) Cricotopus (s.s.) sp. 1 nr. tremulus (Linnaeus) Cricotopus (s.s.) sp. 2 nr. tremulus (Linnaeus) Cricotopus (s.s.) trifascia Edwards Cricotopus (s.s.) sp. Cricotopus (Isocladius) intersectus (Staeger) Cricotopus (Isocladius) sylvestris (Fabricius) Eukiefferiella sp. Nanocladius (s.s.) anderseni Saether Nanocladius (s.s.) crassicornis Saether Nanocladius (s.s.) spiniplenus Saether Orthocladius (Euorthocladius) rivicola Kieffer Orthocladius (Euorthocladius) ?rivicola Kieffer Orthocladius (s.s.) carlatus (Roback) Orthocladius (s.s.) mallochi Kieffer Orthocladius (s.s.) nigritus Malloch Orthocladius (s.s.) obumbratus Johannsen Orthocladius (s.s.) robacki Soponis Orthocladius (s.s.) sp. Parakiefferiella (s.s.) torulata Saether Psectrocladius (Allopsectrocladius) flavus Johannsen Psectrocladius (s.s.) simulans (Johannsen) Psectrocladius (s.s.) sp. 1 Psectrocladius (s.s.) sp. 2 Psectrocladius ss. sp. 3 Pseudosmittia sp. Rheosmittia sp. 1 Rheosmittia sp. 2 Synorthocladius semivirens (Kieffer) Thienemanniella cf. xena Roback Tvetenia vitracies (Saether) Orthocladiinae Genus 1 sp. Orthocladiinae Genus 2 sp. P.G. MASON AND D.M. LEHMKUHL 197 Hof Species Present Each Week GB Upstream 40 [3 6 km downstream (J 23 km downstream 30 20 | i all in i APR [Akay MAY MAYIMAY|MAYJUNUJUNTUNDUNIJUL [JUL |JUL[JUL JUL AUGAUGIAUG|AUG|SEP [SEP SEPJOCTOCTINOY Time | | #of Species Fic. 3. Total number of chironomid species collected each week from Saskatchewan River upstream, 6 kilometers downstream and 23 kilometers downstream from Squaw Rapids Dam. This reinforces the earlier observation that it is necessary to sample during the entire ice-free season to obtain a true picture of the river fauna. The thirty most abundant species were examined in detail and the follow- ing observations were made. While the upstream species Cyphomella gib- bera and Stenochironomus hilaris appear to be eliminated downstream from the reservoir, another, Polypedilum obtusum, which is not present upstream, appears in the community. Rheotanytarsus exiguus and Nanocladius anderseni, the dominant species upstream, show reduced numbers downstream, Cricotopus bicinctus, Orthocladius carlatus and Synorthocladius semivirens, a small part of the upstream fauna, increase in abundance in the downstream community. Orthocladius rivicola and Tvetenia vitracies constitute about the same percent of the populations upstream and downstream. The Orthocladiinae species, Synorthocladius semivirens and Cricotopus spp. are most abundant during midsummer downstream from the Dam while the numerous Chironomini species are most important above the reservoir. It appears as though some Orthocladiinae species replaced Chironomini species as the most abundant group during the summer. Mackey (1977b) noted that Cricotopus bicinctus and Cricotopus sylvestris dominated the summer fauna in the Thames River and the River Kent in Britain. MEM. AMER. ENT. SOC., 34 198 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE TABLE 3. Species of Chironomidae found only in Tobin Lake. Chironomus (s.s.) anthracinus Zetterstedt Chironomus (s.s.) plumosus (Linnaeus) Cladopelma sp. Cryptochironomus stylifera Johannsen Cryptochironomus sp. 1 Cryptochironomus sp. 4 Paratanytarsus sp. 1 Psectrocladius (s.s.) sp. 2 Ablabesmyia (s.s.) sp. 3 Upstream 198I WB Tanypodinae Tanytarsini FE Chironomini [2 Orthocladiinae Diamesinae 30 NO (eo) = fof Species Ss) = tif nil Fg Alli R MAY JUN JUL TIME Fic. 4. Number of species of each major chironomid group collected weekly upstream from Squaw Rapids Dam. Be OCT NOV | Bc Eee AUG SEP AP Environmental Factors — In addition to the information presented, studies by Lehmkuhl (1972), Brooker and Hemsworth (1979), Fraley (1979), Williams and Winget (1979), Merkely (1978), Ward (1976a, 1976b), Young et. al. (1976), Ward (1974), and Hilsenhoff (1971) have shown altered downstream community structure, presumably because of downstream en- vironmental changes resulting from the presence of reservoirs. Ward and Stanford (1979) state that temperature, flow and substrate are the major factors affecting macroinvertebrate distribution below reservoirs. P.G. MASON AND D.M. LEHMKUHL 199 TABLE 4. Number of species of each major group of the Chironomidae at upstream and downstream sites on the Saskatchewan River. Y mile 11 miles upstream downstream downstream Chironomini 39 20 34 Tanytarsini 9 10 17 Orthocladiinae 25 22, 28 Tanypodinae 5 3 3 Diamensinae 1 1 2 TOTALS 719 56 84 6km Downstream MB Tanypodinae Tanytarsini E} Chironomini C) Orthocladiinae 30 [) Diamesinae #of Species nN s) fo) TIME Fic. 5. Number of species of each major chironomid group collected weekly 6 kilometers downstream from Squaw Rapids Dam. Lehmkuhl (1979) concluded that temperature was the single most important environmental parameter affecting the fauna downstream from a reservoir. In an attempt to discover how the environment influences the communi- ty, environmental parameters were measured at the study sites and com- parison made. MEM. AMER. ENT. SOC., 34 200 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE [art i 23 km Downstream 198| | | Tanypodinae E24 Tanytarsini JUL AUG s TIME ED Chironomini E4 Orthocladiinae Fic. 6. Number of species of each major chironomid group collected weekly 23 kilometers downstream from Squaw Rapids Dam. [) Diamesinae 30 ES NO (oe) #of Species S&S EP ‘OCT 'NOV (i) PCA analysis of seasonal averages.—Average seasonal values of 27 en- vironmental parameters, for example oxygen, temperature, substrate organic carbon, turbidity and months of ice cover, were used in a Principal Components Analysis of the sample sites. Figure 7 is a plot of the principal factor scores of the five sites (2 lake, 3 river) studied. Those scores furthest from the origin, in either direction, on a given axis account for more of the variation (differ from the others more) than those scores closest to the origin. The horizontal Axis (1) accounts for 49.5% of the variation and separates the upstream site from the downstream sites. The lake sites are separated from the river sites on the vertical Axis (2) which accounts for 43.9% of the variation. Thus, the three major categories of sites differ from one another. The scattergram of the environmental parameters (Figure 8) shows high factor loadings for most of the parameters suggesting that all are important in separating sites. Based on average seasonal values, PCA analysis shows that upstream, lake and downstream sites differ from each other but that no single en- vironmental parameter stands above the others. Therefore, we turned to data from weekly measurements. P.G. MASON AND D.M. LEHMKUHL 201 Factor 2 Principle Factor Scores for Sites based on 27 Environmental Parameters Horizontal Factor 1 49.5%of Variation Vertical Factor 2 43.9% of Variation @ Sask River Upstream @Tobin Lake at Corral @@ @ Tobin Lake at Dam 1 @Sask River 6 km Downstream @© Sask River 23 km Downstream Factor 1 Fic. 7. Principle factor scores for sites based on 27 environmental! parameters (Factors 1 and 2). (ii) Data from weekly measurements.—Our data showed that en- vironmental parameters such as current velocity, substrate and organic car- bon have some limited influence on the distribution of individual species in the chironomid community and a detailed discussion is beyond the scope of this paper. Details are discussed in another paper (Mason and Lehmkuhl, in prep’n). Danks (1978) concluded that temperature and photoperiod are the two most important environmental factors affecting emergence time. In our analysis emergence patterns for each site are compared for the same time, and thus photoperiod is the same, and cannot account for observed dif- ferences. Nordlie and Arthur (1981) noted from the literature that oxygen also affects emergence. In our study, oxygen was always at or near satura- tion, and therefore it probably had little or no effect on chironomid emergence. Thus, the effect of temperature remains and will be examined especially as it relates to chironomid emergence. MEM. AMER. ENT. SOC., 34 202 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE Factor 2 Factor 1 °15 PCA Scattergram of Environmental Parameters Horizontal Factor] 425% of Variation Vertical Factor 2 43.9% of Variation °20 1s CURRVAR Reem 3:OXYGEN a feTROL ice $= INORGC PDISNO2 101TH 5 19 13:DISCL lator 15: ian 21:SO4 3: BeMONICEG 24: SUB) 27-SUB4 COSTES e3 e2 o14 °22 Fic. 8. PCA scattergram of environmental parameters (Factors 1 and 2). It is clear that reservoirs modify the temperature regime of a river depen- ding on release depth and thermal stratification (Ward and Stanford, 1979). Tobin Lake is an epilimnion release reservoir and the downstream thermal pattern is modified directly by lake surface temperatures. Temperature (Fig. 9) showed very definite seasonal differences between the upstream and downstream sites. This pattern was also observed in 1974, 1979 and 1980 and can be considered typical of the system. Observation of the emergence patterns of eight representative species (Figs. 10-17) showed that a definite delay in the onset of spring emergence occurred downstream. Microtendipes caducus and Orthocladius rivicola are the only two of these species which appear at the upstream and downstream sites simultaneously. M. caducus, however, is more abundant initially upstream than it is downstream. Polypedilum laetum, Robackia demeijerei, Rheotanytarsus exiguus, Potthastia longimana, Cricotopus trifascia and (C) TEMPERATURE 25 P.G. MASON AND D.M. LEHMKUHL 203 Upstream ----eee 6 km downstream oe — 23 km downstream Time Fic. 9. Temperature (degrees centigrade) in the Saskatchewan River during the 1981 Field Season. Tventenia vitracies all appear downstream almost a month after they do upstream. This appears to be the situation for many other species as well. Lesage and Harrison (1980) concluded that the rising spring water temperatures induced the spring emergence of Cricotopus spp. Most studies involving the effects of temperature on chironomid emergence have considered elevated temperatures only. Kochn and Frank (1979), Ferguson and Fox (1978), Langeford (1975) and Langeford and Duffern (1975) concluded that increased water temperatures had no effect on the emergence of aquatic insects. However, Nordlie and Arthur (1981) state that Brittain (1976) and Newel and Minshall (1978) concluded that alteration of normal thermal cycles can cause different developmental rates in aquatic insect life cycles and therefore produce variations in emergence times. Lesage and Harrison (1980) determined that emergence of Cricotopus trifascia occurred between 16 and 21 degrees Celsius, C. sy/vestris emerged at 18-19 degrees Celsius, and C. s/ossonae and C. politus emerged when the water was 20-21 or 22-23 degrees Celsius. A delay in reaching the threshold ambient temperature would therefore result in delay in the onset of MEM. AMER. ENT. SOC., 34 204 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE 10,0005 Microtendipes caducus 1981 ‘it [—) UPSTREAM Gl 6 km DOWNSTREAM [= 23km DOWNSTREAM F of SPECIMENS 1 [APRA MaYMay|May|MayJun|JuNsunsun{sut [ue [yuu [suLfaudaudaudaud|serse dseroctloct|Nov TIME Fic. 10. Emergence of Microtendipes caducus in the Saskatchewan River during 1981. 10,000 Polypedilum laetum 1981 GB Upstream 100 C9 6 km downstream (7) 23 km downstream 100 # of Specimens te) 5 L AUGAUGA SEP |SEP|SEP|OCTIOCTNOW Fic. 11. Emergence of Polypedilum laetum in the Saskatchewan River during 1981. P.G. MASON AND D.M. LEHMKUHL 205 10,000 Robackia demeijerei 1981 WB Upstream GE 6 km downstream (==) 23’ km downstream 100 #of Specimens fo} S) Fic. 12. Emergence of Robackia demeijerei in the Saskatchewan River during 1981. 10,000 Rheotanytarsus exiguus 1981 1000 + [——) UPSTREAM MM 6 km DOWNSTREAM g [= 23km DOWNSTREAM uJ = 100 4 O WwW oO Y) — ° 4 10 4 i 0 [arr ae Mayiar|war]war wun] sunjun[unsut sue [ui sue [sutfaudaudaudavelserse dserloc toc nov TIME Fic. 13. Emergence of Rheotanytarsus exiguus in the Saskatchewan River during 1981. MEM. AMER. ENT. SOC., 34 206 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE 10.000 Potthastia longimana 1981 GB Upstream 100 EE) 6 km downstream [= 23 km downstream 2 100- o £ .) ® a 7) ro} * 10: 0 Fic. 14. Emergence of Potthastia longimana in the Saskatchewan River during 1981. 10,000 Cricotopus trifascia 1981 GHB Upstream I [5 6 km downstream (=) 23 km downstream H#of Specimens Fic. 15. Emergence of Cricotopus trifascia in the Saskatchewan River during 1981. P.G. MASON AND D.M. LEHMKUHL 207 10,000 Orthocladius rivicola 1981 [J UPSTREAM MM 6 km DOWNSTREAM [= 23km DOWNSTREAM 1000 4 100 4 : | TIME FE of SPECIMENS Fic. 16. Emergence of Orthocladius rivicola in the Saskatchewan River during 1981. emergence of a species. This appears to be the situation in the Saskatchewan River in the vicinity of the Squaw Rapids hydroelectric development. Chironomids made up more than 2/3 of the known aquatic insect species in a regulated portion of the Saskatchewan River. Comparison of the chironomid fauna upstream and downstream from the Squaw Rapids hydroelectric develoment showed distinct differences — some species were eliminated downstream — others decreased in relative abundance, others increased in relative abundance and — still others were apparently not greatly affected by the reservoir. A delayed rise in spring temperature, while not apparently restricting chironomid distribution, caused delays of up to one month in the onset of spring emergence of most species examined. Other environmental parameters, particularly current velocity, substrate and organic carbon appear to affect distribution and growth and review of the available data is being done to determine the precise nature of their in- fluence on the chironomid fauna. ACKNOWLEDGEMENTS This research was supported by the junior author’s NSERC grant and by field research funds awarded to the senior author by the Institute for MEM. AMER. ENT. SOC., 34 208 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE 10,000 Tvetenia vitracies 1981 1] Upstream 1000 G6 km downstream (—) 23 km downstream 100: | n | | HL al # of Specimens Fic. 17. Emergence of 7vetenia vitracies in the Saskatchewan River during 1981. Northern Studies. Thanks is also given to Dr. L. Burgess for comments of the manuscript. REFERENCES ARMITAGE, P.D. 1977. Invertebrate drift in the regulated river Tees and an unregulated tri- butary Maize Bec. below Cow Green dam. Freshwat. Biol. 7:167-183. 1978. The impact of Cow Green Resevoir on invertebrate populations in the River Tees. Fresh. Biol. Assoc. Ann. Report 46:47-57. ARMITAGE, B.J., T.D. FORSYTHE, E.B. RODGERS AND W.B. WRENN. 1978. Brown Ferry biothermal research series I. Colonization by periphyton, zooplankton and macroinvertebrates. U.S. Env. Pro. Ag. Report #EPA-600/3-78-020:45 pp. BRITTAIN, J.E. 1976. The temperature of two Welsh lakes and its effects on the distribution of two freshwater insects. Hydrobiologia 48:37-49. BROOKER, M.P. AND R.J. HEMsworTH. 1979. The effect of the release of an artificial dis- charge of water on invertebrate drift in the River Wye, Wales. Hydrobiologia 59:155-163. BRUNDIN, L. 1966. Transantarctic relationships and their significance as evidenced by chironomid midges. Kungl. Svenska. Vetenskupacad. Handl. 11:1-472. COFFMAN, W.P. 1973. Energy flow in a woodland stream ecosystem II. The taxonomic composition and phenology of the Chironomidae as determined by the collection of pupal exuviae. Arch. Hydrobiol. 71:281-322. Danks, H.V. 1978. Some effects of photoperiod, temperature and food on emergence of three species of Chironomidae (Diptera). Can. Ent. 110:289-300. P.G. MASON AND D.M. LEHMKUHL 209 Durgan, J.K. 1981. Tobin Lake Fishery 1979, 17 years after impoundment. Sask. Fish. Lab. Tech. Report 81-5:147 pp. FERGUSON, V.M. AND R.C. Fox. 1978. A comparison of aquatic insects in natural inlets with those in heated effluent from the OCONEE nuclear station — Littoral zone. J. Ga. Ent. Soc. 13:202-213. FRALEY, J.J. 1979. Effects of elevated stream temperatures below a shallow reservoir on a cold water macroinvertebrate fauna. p. 257-272. In: Ecology of Regulated Streams. J.V. Ward and J.A. Stanford (editors). Plenum Press, New York and London. 398 pp. Goopno, E.J. 1975. Post-impoundment limnology of the East Lynn Lake tailwater, West Virginia. Proc. West Vir. Acad. Sci. 47:170-176. Gore, J.A. 1980. Ordinational analysis of benthic communities upstream and downstream of a prairie storage reservoir. Hydrobiologia 69:33-44. HILsENHOFF, W.L. 1971. Changes in the downstream insect and amphipod fauna caused by an impoundment with a hypolimnion drain. Ann. Ent. Soc. Amer. 64:743-746. Lackey, R.T. 1973. Bottom fauna changes during artificial reservoir destratification. Wat. Res. 7:1349-1366. LANGEFORD, T.E. 1975. The emergence of insects from a British River, warmed by power station cooling-water Part II. The emergence patterns of some species of Ephemeroptera, Trichoptera and Megaloptera in relation to water temperature and river flow, upstream and downstream of the cooling-water outfalls. Hydrobiologia 47:91-133. LANGEFORD, T,.E. AND J.R. DUFFERN. 1975. The emergence of insects from a British River, warmed by power station cooling-water Part I. The use and performance of insect emergence traps in a large spate-river and the effects of various factors on total catches, upstream and downstream of the cooling-water outfalls. Hydrobiologia 46:71-114. LAviLE, H. 1979. A study of the drift of pupal exuviae of chironomids at the confluence of the Truyere and the River Lot. Ann. Limnol. 15:155-180. LEHMKUHL, D.M. 1972. Change in thermal regime as a cause of reduction of benthic fauna downstream of a reservoir. J. Fish. Res. Board Can. 29:1329-1332. 1979. Environmental disturbances and life histories: Principles and ex- amples. J. Fish Res. Board. Can. 36:329-334. LESAGE, L. AND A.D. HARRISON. 1980. The biology of Cricotopus (Chironomidae: Ortho- cladiinae) in an algal-enriched stream: Part I. Normal biology. Arch. Hydrobiol. Suppl. 57:375-418. Mackey, A..P. 1977. Species diversity of some larval Chironomidae (Diptera) communi- ties in two rivers. Oikos 29:150-155. MERKLEY, W.B. 1978. Impact of Red Rock Reservoir on the Des Moines River. p. 62-76. In: Current Perspectives on River-Reservoir ecosystems. E. Benfield Jr. (editor). N.A. Benthol. Soc., Springfield, Illinois. NEWELL, R.L. AND G.W. MINSHALL. 1978. Life history of a multivoltiine mayfly, Tricorythodes minutus: an example of the effect of temperature on the life cycle. Ann. Ent. Soc. Amer. 71:876-881. Nie, N.H., C.H. Hutt, J.G. Jenkins, K. STEINBRENNER AND D.H. Bent. 1975. SPSS (Statistical Package for the Social Sciences), 2nd edition. McGraw-Hill, New York, New York. 675 pp. NorpuiE, K.J. AND J.W. ArTHUR. 1981. Effect of elevated water temperature on insect emergence in outdoor experimental channels. Env. Poll. (Series A) 25:53-65. Otson, D.E. AND D.C. TarTER. 1974. A pre-impoundment investigations of Beech Fork Basin Twelvepole Creek, Wayne and Cabell counties West Virginia, U.S.A. Proc. West Vir. Acad. Sci. 46:231-236. MEM. AMER. ENT. SOC., 34 210 HYDROELECTRIC DEVELOPMENT ON CHIRONOMIDAE Rossaro, B. 1978. The taxonomic composition and phenology of the Orthocladiinae, Dip- tera, Chironomidae, in Po River near Caorso (Piacenza) as determined by analysis of pupal exuviae. Riv. Hydrobiol. 17:287-290. RossARO, B. AND U. FERRARESE. 1980. A contribution to the knowledge of chironomids in Italy, including cluster analysis of presence-absence data and factor analysis with per cent composition of species in a stream. p. 203-208. In: Chironomidae, Ecology, Systematics, Cytology and Physiology. D.A. Murray (editor). Pergamon Press, Oxford and New York. 354 pp. Royer, L.M. 1969. A limnological and Fisheries survey of Tobin Lake, an impoundment on the Saskatchewan River, 1966-67. Sask. Fish. Lab. Report 1969:85 pp. SAwcHyn, W.W. 1974. Tobin Lake. A study of water quality, sediments and benthos. Sask. Res. Coun. Report 1975:79 pp. SPENCE, J.A. AND H.B.N. Hynes. 1971. Differences in benthos upstream and down- stream of an impoundment. J. Fish. Res. Board. Can. 28:35-43. Warp, J.V. 1974. A temperature-stressed ecosystem below a hypolimnial release mountain reservoir. Arch. Hydrobiol. 74:247-275. 1976a. Effects of flow patterns below large dams on stream benthos: A re- view. In: Instream Flow Needs Symposium, Vol. II. J.F. Osborn and C.H. Allman (editors). Amer. Fish. Soc. 1976:235-253. 1976b. Effects of thermal constancy and seasonal temperature displacement On community structure of stream macroinvertebrates. pp. 302-307. In: Thermal Ecology II, G.W. Esch and R.W. McFarlane (editors). ERDA Symposium Series (CONF-750425). WarRD, J.V. AND R.A. SHORT. 1978. Macroinvertebrate community structure of four special lotic habitats in Colorado, U.S.A. Verh. Intern. Ver. Limnol. 20:1382-1387. WARD, J.V. AND J.A. STANFORD. 1979. Ecological factors controlling stream zoobenthos with emphasis on thermal codification of regulated streams. p. 35-56. In: The Ecology of Regulated Streams. J.V. Ward and J.A. Stanford (editors). Plenum Publishing Corpora- tion. 378 pp. WARTINBEE, D.C. 1976. Quantitative determination of chironomid emergence from en- closed channels in a small lotic ecosystem. Amer. Midl. Nat. 95:479-485. 1979. Diel emergence patterns of lotic Chironomidae. Freshwat. Biol. 9:147-156. WILLIAMS, R.D. AND R.N. WINGET. 1979. Macroinvertebrate response to a flow manipula- tion in the Strawberry River, Utah (U.S.A.). p. 365-376. In: Ecology of Regulated Streams. J.V. Ward and J.A. Stanford (editors). Plenum Press, New York and London. 398 pp. Witson, R.S. 1979. Classifying rivers using chironomid pupal exuviae. p 209-216. In: Chironomidae, Ecology, Systematics, Cytology and Physiology. D.A. Murray (editor). Pergamon Press, Oxford. England. 354 pp. 1977. Chironomid pupal exuviae in the River Chew. Freshwat. Biol. 7:9-17. WILSON, R.S. AND P.L. BRIGHT. 1973. The use of chironomid pupal exuviae for character- izing streams. Freshwat. Biol. 3:283-302. WILson, R.S. AND J.D. McGmt. 1977. A new method of monitoring water quality in a stream receiving sewage effluent, using chironomid pupal exuviae. Wat. Res. 11:959-962. Younc, W.C., D.H. KENT AND B.G. WHITESIDE. 1976. The influence of a deep storage reservoir on the species diversity of benthic macroinvertebrate communities of the Guadalupe River, Texas, U.S.A. Tex. J. Sci. 27:213-224. Cytogenetic Studies on Chironomus plumosus L. (Diptera, Chironomidae) from different Populations and their Experimental Hybrids PARASKEVA MICHAILOVA Institute of Zoology Bulgarian Academy of Sciences I Ruski bul., Sofia — 1000, Bulgaria JURG FISCHER Zoologisches Institut, Universitat Bern Baltzerstrasse 3, CH — 3012 Bern, Switzerland ABSTRACT. — Chironomus plumosus egg masses were collected from populations in Lake Neuchatel (Switzerland), Lake Wohlen (Switzerland) and Deer Lake (Vancouver, Canada). The larvae reared in the laboratory, the crossing of adult midges (Neuchatel o o X Wohlen © 9; Neuchatel 9 9 XK Wohlen oO; Vancouver 0&0 X Wohlen 9 9 ) were initiated ar- tificially. The analysis of the polytene chromosomes of 4th instar larvae leads to the following conclu- sions: The individuals from Neuchatel and Wohlen differed by several homozygous inversions, making them clearly distinguishable. Such inversions have been found in the chromosome arms A, B, C, D and G. Four inversions (in arm: A, B, C and D) are forming polymorphic systems in both populations. In hybrid larvae the pairing of the homologue chromosomes is not disturbed in sections with similar band sequences. The populations from Neuchatel and Wohlen are closely related. They can be considered as subspecies which diverged on the basis of inversions. The difference between the populations of Vancouver and Wohlen are much more pro- nounced. We have found homozygous inversions in the arms B, D, E and G. In the hybrids, asynapsis of the homologue chromosomes is almost complete due to genic changes which are not reflected in the banding patterns. While the chromosomes of individuals from the Neuchatel and Wohlen populations have striking centromere regions, those of the Vancouver population have indistinct centromere regions. By means of the C-banding technique it could be established that in the European populations the heterochromatin is located mainly in the centromere regions, while in Vancouver population there is a considerable amount of heterochromatin in different parts of the chromosomes. The differing display of C-bands in the initial populations and the hybrids (Vancouver X Wohlen) reveals a different gene regulation which was first reported for chironomus and polytene chromosomes in general. The Vancouver population is not only geographically dis- tant from the European populations, the observed differences would even provide basis for the separation of the latter into an independent species. 211 MEM. AMER. ENT. SOC., 34 212 CYTOGENETIC STUDIES ON CHIRONOMUS PLUMOSUS L. INTRODUCTION The taxonomy of Chironomus plumosus has puzzled chironomid workers for many years. Some authors (Lenz, 1924; Tchernovski, 1948) considered it as a polytypic species: they distinguished several larval forms based on the length of the processes on the 8th larval segment. Keyl and Keyl (1959), Palmén and Aho (1969), Krieger and Wiilker (1971) regarded Ch. plumosus as a species-complex incorporating several species with distinct ecological and cytological peculiarities. The comparative karyological studies of Michailova and Maximova (1980) on distant populations in USSR and Bulgaria showed that the species is in the process of divergence. The aim of this study is to obtain new information on the taxonomic status of different populations of ‘‘the species’’ Chironomus plumosus. For this purpose we have analysed different populations cytotaxonomically: ad- jacent European (Neuchatel and Wohlen, Switzerland; Plovdiv and Durankulak, Bulgaria) as well as geographically distant populations (Switzerland /Bulgaria/Canada). We also analysed hybrids between dif- ferent populations. This will contribute to our knowledge on interpopula- tional relationships, the taxonomic status and the steps of evolution within the particular populations. As the species divergence causes not only karyotype differences but also differences in number, size and localization of the constitutive heterochromatin blockes (Gatti, Pimpinelli, Sautini, 1976), the distribution and localization of the constitutive heterochromatin in individuals from the natural populations and in hybrids were determined by C banding technique. For each population a cytological map was made in which the chromosomes are indexed by letters, and the sections within the chromosomes by digits. In this paper we will discuss only the results concerning the populations from Neuchatel, Wohlen and Vancouver. The Bulgarian populations will be discussed later. MATERIAL AND METHODS Adult midges in the process of laying egg masses were collected on the shores of Lake Neuchatel (Switzerland), Lake Wohlen (Switzerland) and Deer Lake (Vancouver / Canada). The lake Neuchatel is a large mesotrophic lake with a sandy sediment. Lake Wohlen is a reservoir formed by damming the river Aare; due to still considerable water flow it does not stratify. The sediment consists of a thick layer of silt. The Deer Lake is a small, highly eutrophic lake without current; the. sediment is similar to that of Lake Wohlen. P. MICHAILOVA AND J. FISCHER 213 The larvae hatching from the collected egg masses were reared in laboratory in containers of 1-101, using dechlorinated tap water. The substrate consisted of cellulose fibers and sterilized sediment from Lake Wohlen. The larvae were fed with minced Tetra Phyll flakes (Tetra-Werke, Melle, Western Germany). The crossings of adult midges were made by a method described earlier (Fischer, 1969). The staining methods (Aceto-Orceine and C-banding) also have been described earlier (Valkanov and Michailova, 1974; Michailova and Max- imova, 1980). RESULTS AND DISCUSSION The adults and larvae from the natural populations and all hybrid forms could not be distinguished morphologically. The analysis of the polytene chromosomes of the 4th instar larvae leads to the following conclusions: in | } SS =| Bid eporaie tk ab oly als sk lk ae ey NE %, fod tee 10\9 Fic. la. Chromosome I of initial populations (Neuchatel, Wohlen and Vancouver). Ne- Neuchatel; W-Wohlen, V-Vancouver. MEM. AMER. ENT. SOC., 34 214 CYTOGENETIC STUDIES ON CHIRONOMUS PLUMOSUS L. ex V Fic. 1b. Chromosome II of initial populations (Neuchatel, Wohlen and Vancouver). Ne- Neuchatel; W-Wohlen, V-Vancouver. all investigated populations, the Ist, 2nd and 3rd chromosomes are meta- centric, the 4th is acrocentric. Each of the natural populations has its characteristic pattern of inver- sions and other rearrangements (Fig. la, b, c, d), making them clearly distinguishable. Comparing individuals from Neuchatel and Wohlen, inver- sions have been found in the chromosome arms A, B, C, D and G. Four in- versions (in arms A, B, C and D) are forming polymorphic systems in both populations. Comparing the populations of Vancouver and Wohlen, we have found inversions in the arms B, D, E, and G. In the Vancouver population, polymorphic systems do not seem to exist. The polytene chromosomes of hybrids can differ considerably from those of their parents: while in the parents the homologue chromosomes conjugate along their entire length with very rare incomplete pairing, the pairing in the hybrids can be disturbed. In this repect, the two investigated types of hybrids differed fundamentally. P. MICHAILOVA AND J. FISCHER 215 ; , fet A ut saad WwW LOM Unrate UU es 7 S| - el a\" Fic. lc. Chromosome III of initial populations (Neuchatel, Wohlen and Vancouver). Ne- Neuchatel, W-Wohlen, V-Vancouver. In the hybrids Neuchatel X Wohlen, the homologue chromosomes are paired in sections with similar band sequences. In some sections inversions cause the formation of inversion loops (Fig. 3). The 3rd chromosomes of the two populations have identical banding patterns; here, in most cases, we have found complete synapsis in the hybrids (Fig. 4). The rare occurrence of asynapsis can be explained by the theory of heterocyclicity of father’s and mother’s chromosomes (Prokofyeva-Belgovskaya, 1946). The different stages of these chromosomes are evident from the staining intensity of the homologous bands, their thickness and alterations in the band sequence of the homologue. The most interesting feature of the combination Vancouver o X Wohlen 9 is that, even in case of apparently identical banding pat- tern, the homologues of all the chromosomes are strongly asynaptic. For ex- ample: in the Ist chromosome, synapsis occurs usually only between section 4/5-5/6 and several bands after 5/6 (Fig. 2). In the 2nd chromosome, synapsis can be observed usuallly in the first part of the chromosome (Fig. MEM. AMER. ENT. SOC., 34 216 CYTOGENETIC STUDIES ON CHIRONOMUS PLUMOSUS L. 10/UMA mt cage P. MICHAILOVA AND J. FISCHER 217 5). The homologues of the 3rd chromosome usually are completely separated (Fig. 6). The homologues of the 4th chromosome also are com- pletely separated in hybrids, but the same can be observed also in indi- viduals from natural populations. It appears that various factors are involved in the phenomenon of chromosomal pairing. The cases of asynapsis can be divided into three groups: 1. The homologue chromosomes have identical banding patterns. In this case the absence of pairing must be caused by differences on the molecular (at least submicroscopic) level. 2. The banding patterns of the two homologue chromosomes are not iden- tical: each homologon shows the parental arrangement. 3. One homologon has a clear banding pattern and the other shows some grain structure. This kind of asynapsis can be explained by the theory of heterocyclity of the father’s and mother’s chromosomes (Prokofyeva- Belgovskaya, 1946). To date, it is not clear what causes some tissues in certain organisms to grow by endomitosis and form polytene chromosomes. It also is not known which chromosome structures are responsible for the synapsis. Such struc- tures (‘‘synaptomeres’’) must exist (Nagl, 1972). We have not much infor- mation about them, but we can assume that they differ fundamentally from normal genes when considering ecological-genetical aspects of populations. As opposed to other genes, the synaptomeres cannot be affected by en- vironmental selection factors. Therefore one might consider the synap- tomeres to evolve at a higher rate than genes which are responsible for morphological or physiological features. Consequently, the lack of homologue pairing in hybrids should be one of the first visible differences between independently evolving populations. The different behaviour of the homologue chromosomes in the two types of hybrids (Neuchatel X Wohlen; Wohlen X Vancouver respectively) in- dicates that the two geographically adjacent populations are also biologi- cally closely related, while the geographically distant populations are distant biologically. The application of C-banding technique shows that the studied populations diverge also on the basis of amount and distribution of con- stitutive heterochromatin, which indicates some functional differences. In the Neuchatel and Wohlen individuals the heterochromatin is located in the centromere regions of the Ist, 2nd and 3rd chromosomes (in the Wohlen Fic. ld. Chromosome IV of initial populations (Neuchatel, Wohlen and Vancouver). Ne- Neuchatel, W-Wohlen, V-Vancouver. MEM. AMER. ENT. SOC., 34 218 CYTOGENETIC STUDIES ON CHIRONOMUS PLUMOSUS L. P. MICHAILOVA AND J. FISCHER 219 population also a few in the bands 1/2, 8/9 of the 2nd chromosome), while in the Canadian population there is a considerable amount of heterochro- matin distributed over the whole chromosomes (Fig. 7). In the hybrid (Neuchatel X Wohlen) only a few bands and the centromere regions are heterochromatic (Fig. 8). The reverse is the case with the hybrids Vancouver X Wohlen which show numerous C-bands, i.e., numerous genes are repressed, while in the parental forms there are only a few heterochromatic regions. C-bands were found in the hybrids Vancouver X Wohlen as follows: Ist chromosome, father’s homologon (Vancouver): sections 1, 4/3, 3/2, 2/1, 5/6, 6/7, 8/9, 10/11. Ist chromosome, mother’s homologon (Wohlen): section 1, 4/3, 3/2, 9/10. 2nd chromosome, father’s homologon: entire arm C and sections 7/8, 5/6, 6/7. 2nd chromosome, mother’s homologon: entire arm C and sections 7/8, 5/6, 8/9. 3rd chromosome, father’s homologon: sections 1, 1/2, 3/4, 4/5, 6/7. 3rd chromosome, mother’s homologon: centromere region and section Die 4th chromosome: both homologons: sections 2/3, 3/4, 4/5. When considering the relationship of populations, it would be interesting to compare also the mating behaviour. Unfortunately it has not been possi- ble till now to induce swarming and normal mating behaviour of our popu- lations of Ch. plumosus in the laboratory. In our work, the crossing was in- itiated artificially. Therefore, any possible ethological isolation factors were bypassed. Another interesting question is the behaviour of hybrid chromosomes in meiosis. For this purpose, we intend to analyse larvae derived from back-crossing. CONCLUSIONS Hybridization tests and cytogenetic analysis prove experimentally that “‘the species’? Chironomus plumosus is in the process of divergence to a Fics. 2-8. 2. Chromosome I of a hybrid (Vancouver o K Wohlen 9 ) with asynapsis anda complex heterozygous inversion; 3. Chromosome I of a hybrid (Neuchatel o X Wohlen 9); 4. Chromosome III of a hybrid (Neuchatel o XK Wohlen 9); 5. Chromosome II of a hybrid (Vancouver o& X Wohlen 9); 6. Chromosome III of a hybrid (Vancouver & X Wohlen 9); 7. C-bands of chromosome I of Canadian Population; 8. C-bands of chromosomes I, II and III of a hybrid Neuchatel o X Wohlen 9. MEM. AMER. ENT. SOC., 34 220 CYTOGENETIC STUDIES ON CHIRONOMUS PLUMOSUS L. varying extent in different regions of its distribution area. The constant high level of synapsis between hybrid homologues in the combination Neuchatel X Wohlen indicates that the two Swiss populations are closely related and diverged only on the basis of chromosome aberrations. The Canadian population is more distant phylogenetically. Asynapsis of homologue chromosomes is almost complete in hybrids Vancouver X Wohlen due to genic changes which are not reflected in the banding pat- terns. The genic alterations have accumulated during the evolution to such an extent that normal synapsis in the hybrids is no longer possible. These genic differences, together with the dissimilar arrangement of heterochromatin and the different homozygous inversions would provide basis for the separation of the Canadian population into an independent species. The different arrangement of C-bands in the initial populations and the hybrids (Vancouver X Wohlen) indicates a different gene regulation, a fact new for Chironomus and polytene chromosomes in general. ACKNOWLEDGEMENT This work was supported by the Canada Council, Ottawa. REFERENCES FISCHER, J. 1969. Zur Fortpflanzugsbiologie von Chironomus nuditarsis. Rev. suisse Tool. 76:23-55. Gattl1, N., S. PIMPINELLI AND G. SAauTINI. 1976. Characterization of Drosophila Hetero- chromatin. 1. Staining and Decondensation with Hoest 332558 and Quinacrine. Chromo- soma. 57:351-371. KeyL, H.G. AND I. Keyzt. 1959. Die cytologische Diagnostik der Chironomiden. I. Besti- mungstabelle fiir die Gattung Chironomus auf Grund der Speicheldriisen Chromosomen. Arch. f. Hydrobiol., 56:43-57. KITZMILLER, J., G. Frizzi AND W. BAKER. 1967. Evolution and Speciation within the maculipennis Complex of the Genus Anopheles. Genetics of Insect Vectors of Disease. 151-208. KRIEGER-WOLF, E. AND W. WULKER. 1971. Chrionomiden (Diptera) aus der Umgebung von Freiburg i. Br./mit besonderer Beriicksichtigung der Gattung Chironomus/ Beitr. natural. Forsch. Sud. d. W. DTL. 30:133-145. LENZ, F. 1924. Die Chironomiden der Wolga. Arb. Biol. Wolga-Station. 3:97-126. MicHatLova, P. AND F. Maximova. 1980. Karyological Variability of Chironomus plu- mosus L. (Diptera: Chironomidae) and its Importance for the Species Divergence. Acta Zoologica Bulgarica. 14:5-18. Naci, W. 1972. Chromosomes. Wilhelm Goldmann Verlag Miinchen. PALMEN, E. ANDL. AHO. 1966. Studies on the Ecology and Phenology of the Chironomidae of the Northern Baltic. 2. Camptochironomus Kieff. and Chironomus Meig. Annali Zool. Fenn. 3:217-244. P. MICHAILOVA AND J. FISCHER 221 PROKOFYEVA-BELGOVSKAYA, A.A. 1946. MHeterocyclicity of the Parental Chromosomes set. Comp. Rendus de |’ Academy Sciences de 1’ USSR. 2:18-25. TcHERNOVsKiI, A.A. 1949. Key of the Larvae of the Fam. Tendipedidae. USSR. VALKANOV, A. AND P. MicHAILOvA. 1974. Untersuchungen tiber den Karyotypus und Chro- mosomenpolymorphism bei Thalassomyia frauenfeldi Schiner (Diptera, Chrionomidae) von der bulgarischen Schwarzmeerkiiste. Bull. Inst. Zool. Mus. 40:5-16. MEM. AMER. ENT. SOC., 34 jie 4 % + 1 + oe An Inventory of the Irish Chironomidae (Diptera) D.A. MURRAY AND P. ASHE Department of Zoology University College Dublin Belfield, Dublin 4, Ireland ABSTRACT. — The resident Irish biota is poor in species in comparison with that of Britain and mainland Europe. Only two thirds of the native British flora occurs naturally in Ireland and the native Irish fauna is similarily poor in species. Studies on the Irish Chironomidae in recent years have shown that approximately 347 species are now known to occur. An inventory of these species is given in this paper. To date 69% and 29% respectively of the western Palaearc- tic genera and species and 70% of the species known from Britain have been recorded from Ireland. INTRODUCTION The family Chironomidae is represented by approximately 1400 species in 186 genera in the 25 European zones of the western Palaearctic region recognised by Illies in Limnofauna Europaea (Illies, 1978: Fittkau and Reiss 1978). Since the publication of the list of Chironomidae in Ireland (Murray 1972) additional records have been given by Fahy and Murray (1972); Mur- ray (1976a,b); Pinder and Cranston (1976); Douglas and Murray (1980); Dowling et al (1981); Dowling and Murray (1981); Murray and O’Connor (1982) and Murray and Ashe (1982). Lists of species are a “‘neglected mine of valuable information whose ex- ploitation should do much to clarify the zoogeographical and ecological components of animal distribution’’ (Blackith and Blackith 1975). Com- pilation of such lists forms a basic and important component of ecological research and there still exists a need for continued effort in this field of chironomid research in many countries, including Ireland. The need is all the more pressing with the realisation that many habitats are being steadily eroded and damaged not only by natural forces but also through human neglect, exploitation, pollution and other aspects of 20th century life. An inventory of the Chironomidae currently known to occur in Ireland, zone 17 of Limnofauna Europea (Illies l.c.), is given in Table 1. This list incor- porates the records of the last decade and includes corrections of some species names from the previous list (Murray 1972) which have been necessitated as a result of recent taxonomic revisions. Previously unpub- 223 MEM. AMER. ENT. SOC., 34 224 INVENTORY OF IRISH CHIRONOMIDAE lished records are denoted by an asterisk‘*‘*’’. Distributional data of these latter species in Ireland is given in Ashe, Hayes and Murray (in prep.). Included in the list are some species records which require comments. Anatopynia plumipes (Fries) This monospecific genus is known only from middle and northern Europe. Larvae live, according to Brundin (1949), in the littoral region of small lakes and ponds. Fittkau (1962) regards it as a rare species. The record of this species from Ireland, based on a mature male pupa, is in general agreement with previous ecological information but its emergence, in March, is three/four weeks earlier than previously reported. Telmatopelopia nemorum (Goetgh.) This doubtful record is based on a single female taken by Fahy (Fahy and Murray |.c.). The material has since been mislaid and there have been no ad- ditional records in the interim period. Diamesa permacer (Walk.) The adult male of this rare palaearctic species, collected by C.F. Hum- phries in 1950 near Dublin, is to be found in the British Museum of Natural History, London. The material had been sent to Dr. P. Freeman for iden- tification and the specimen in question had obviously been reared from the larva since both larval and pupal exuviae have recently been located in the slide collection made by Humphries and now on deposit in the Zoology Department, University College, Dublin. Bryophaenocladius virgo Thien. This record is based on the identification of a single pupal exuviae (leg. Ashe) from the River Flesk, Killarney. Morphological characters of the specimen agree entirely with the descriptions given by Strenzke (1950) but there remain a number of species in this genus whose pupa is, as yet, unknown. Georthocladius luteicornis (Goetgh.) Larvae of this monospecific genus have previously only been found in moist or wet turf or bog biotopes. The single pupal exuviae recorded from the River Flesk (leg. Ashe) in all probability originated from moist soil above the river bank. The species is not known from Britain. Orthocladius (Euorthocladius) ? rusticus Goetgh. For some time now a characteristic Euorthocladius pupal exuviae, differ- ing from descriptions of the existing species, has been known to the authors. Associated adults resemble O. (E.) thienemanni Kieff. The pupal ab- dominal chagrin on the posterior edge of tergites IV-VIII and the three dor- D.A. MURRAY AND P. ASHE 225 socentral setae on the thorax are much more robust in comparison with those on the other members of the genus. This species has been collected in numerous rivers in Ireland (leg. Murray, Ashe) and is also known from two locations in Norway and one in France (leg. Murray). Recent examinations of slide material prepared by C.F. Humphries has revealed that reared material had been sent by her to Goetghebuer for identification. Some slides in Humphries’ collection are labelled ‘‘O. rusticus’?. The possibility therefore exists that Goetghebuer identified the associated male and com- municated this information to Humphries. It is not possible to state at this time whether the specimens belong to O. rusticus or to a new species. Orthocladius (O.) dentifer Brundin and O. (Pogonocladius) consobrinus (Holm.) The record of O. dentifer is based on a re-examination of some material of O. (Pogonocladius) consobrinus made available by Murray to Dr. P. Cranston (Pinder and Cranston, 1976). To date this is the sole record of O. dentifer from Ireland or Britain. O. (P.) consobrinus has been recorded from four Irish locations. It is predominantly a late winter /spring emerging species. Pupal exuviae have been collected from lake surfaces during January/February. Rheosmittia spinicornis Brund. Pupal exuviae of this species have been obtained from the River Slaney in S.E. Ireland (leg. Hayes). Only two species are known in the genus neither of which occur in Britain. The pupa is readily identifiable on the basis of the characteristic palmate abdominal setae. Harnischia falcata Kieff. This species was recorded in Limnofauna Europaea (Fittkau and Reiss, l.c.). The original source of the record is unknown to the present authors. Microchironomus deribae (Freem.). Damaged adults of this species were obtained from the fuel filtering system of a helicopter based at the Baldonell military airport near Dublin. The validity of including the record as part of the Irish fauna is questionable as the helicopter had been refueled in France some time prior to the discovery of the insects in the filter (Murray and O’Connor 1982). Corynocera ambigua Zett. Adult males and females of this brachypterous species have recently been taken from L. Corrib (leg. Connolly) a large limestone lake in western Ireland. C. ambigua has a predominantly boreal distribution and is not yet recorded from Britain. MEM. AMER. ENT. SOC., 34 226 INVENTORY OF IRISH CHIRONOMIDAE Our current knowlege of the Irish Chironomidae shows that 347 species in 128 genera are known to occur (Table 2). This compares with 460 species and 133 Genera in Britain and approximately 1200 species and 186 genera in Europe. Approximately 38% of the European chironomid species has been recorded from Britain while only 29% has so far been recorded from Ireland. Three hundred and twenty seven species on the Irish list are known from Britain. It is noteworthy, however, that some species on the Irish list, including those belonging to the genera Anatopynia, Georthocladius, Rheosmittia and Corynocera, are not yet recorded from Britain. Conversely there are fourteen genera and some 158 species found in Britain which have not been recorded from Ireland. Approximately 70% of the species known from Britain are also recorded in Ireland. This trend compares favourably with records for two other insect groups with aquatic life stages since only 73% and 62% of the Trichoptera and Ephemeroptera respectively, recorded in Britain are known from Ireland. (Table 3). DISCUSSION Absence of a taxon from the faunal list of a particular area is, according to Blackith and Blackith (1975), attributed to two clear propositions: that the members of the taxon never reached the area in question or, if they did, were unable to maintain a permanent population. It must be recognised, however, that the content and length of national or regional faunal lists also reflects the intensity of sampling and variety of habitats sampled. Thus rare species may remain undetected for years. The flora and fauna of the island of Ireland, lying to the west of mainland Europe and separated from its nearest neighbour Britain by the Trish sea, is characterised by a lower number of species than that of Britain or Europe. Only two thirds of the native British Flora occurs naturally in Ireland and likewise Ireland contains only one third of the reptiles and half of the land mammals found in Britain (Mitchell 1976). It has been postulated that events during and after the last glaciation have led to this situation. Many authors believe that the plants and animals now present in Ireland are post-glacial immigrants as it is suggested that few could have survived at maximum glaciation (Healy 1979). The route for the floral and faunal recolonisation of Ireland from Europe lay through Britain but as sea levels rose on the retreat of ice the land bridge between Britain and Ireland was severed long before Britain was isolated from the continent (Mitchell 1976) thus establishing a barrier to dispersal. Whereas this scenario might validly apply to the terrestrial fauna using land bridges as dispersal routes it need not necessarily apply to insects with the ability to fly. It would seem D.A. MURRAY AND P. ASHE 227 that there has been ample time in the 10,000 years since glaciation for most European invertebrate species to have reached Ireland but the fact remains that the Irish fauna does appear impoverished in relation to Britain’s. The lack of ecological variety has been cited as a possible factor to explain the low number of species in Ireland’s fauna (Healy l.c.). Ireland lacks the extensive areas of dry heath or chalkland and is deficient in deciduous woods in comparison with Britain but in relation to the aquatic insects this need not necessarily present a serious obstacle. Nevertheless all the available data indicates that the Irish chironomid fauna exhibits a trend similar to other freshwater groups and undoubtedly appears impoverished in comparison with Britain. Keeping in mind the fact that the length of faunal lists is partly a reflec- tion of the intensity of work done on a particular group it is anticipated that the number of species on the Irish Chironomid list will increase in the years ahead. For example little work has been done to date on the semi-aquatic chironomids in Ireland. Likewise it is highly probable that many of the species recorded in Ireland and not yet reported from Britain will eventually be found there. It is suggested that a combination of geographical isolation and the prevailing westerly /south-westerly Atlantic winds have constituted effective barriers preventing largescale immigration. The presence of some species with a boreal distribution may additionally support a theory that local refugia existed to allow some species survive maximum glaciation. REFERENCES BLAcKITH, R.E. AND R.M. BLackiTH. 1975. Zoogeographical and ecological determinants of Collembolan distribution. Proc. R. Ir. Acad. 75B 18:345-368. BRUNDIN, L. 1949. Chironomiden und andere Bodentiere der sudschwedischen Urge- birgssen. Rep. Inst. Freshwat. Res. Drottingholm. 30:1-914. Douctas, D.J. AND D.A. Murray. 1980. A Checklist of the Chironomidae (Diptera) of the Killarney Valley Catchment, pp 123-130. In MURRAY, D.A. (Ed.): Chironomidae- Ecology, Systematics, Cytology and Physiology. Pergamon Press, Oxford, 354 pp. Dowluinc, C. AND D.A. Murray. 1981. The distribution of the Chironomidae (Diptera) in two Irish blanket bogs. Proc. R. Ir. Acad. 81B:53-61. DowuInc, C., J.P. O’CoNNoR AND M.F. O’GrRApy. 1981. A baseline survey of the Caragh, an unpolluted river in southwest Ireland: Observations on the macro- invertebarates. J. Life Sci. R. Dubl. Soc. 2:147-159. Fauy, E. AND D.A. Murray. 1972. The Chironomidae from a small stream system in western Ireland with a discussion of species composition of the group (Diptera). En- tomol. Ts. 93:148-155. Firrkau, E.J. 1962. Die Tanypodinae (Diptera, Chironomidae). Abh. Larvalsyst. In- sekten, 6:1-453. Firrkau, E.J. AND F. ReIss. 1978.. Chironomidae, pp 404-444. In J. Illies (Ed.) Limno- fauna Europaea, 2nd. edn. Gustav Fischer Verlag, Stuttgart, 534 pp. MEM. AMER. ENT. SOC., 34 228 INVENTORY OF IRISH CHIRONOMIDAE HEALY, B. 1978. Records of Enchytraeidae (Oligochaeta) in Ireland. J. Life Sci. R. Dubl. Soc. 1:39-70. ItuEs, J. 1978. Limnofauna Europaea. 2nd. edn. Gustav Fischer Verlag, Stuttgart, 532 pp. MITCHELL, F. 1976. The Irish Landscape. Collins, London. Murray, D.A. 1972. A list of the Chironomidae (Diptera) known to occur in Ireland with notes on their distribution. Proc. R. Ir. Acad. 72B:275-293. 1976a. Buchonomyia thienemanni Fitt. a rare and unusual species (Diptera, Chironomidae) recorded from the Killarney area, Ireland. Ent. Gaz. 27; 1976b. Thienemannimyia pseudocarnea n.sp. a palaearctic species of the Tanypodinae (Diptera, Chironomidae). Ent. Scand. 7:191-194. Murray, D.A. AND P. AsHE. 1982. First Records of the subfamilies Podonominae and Telmatogetoninae (Diptera, Chironomidae) from Ireland. Ir. Nat. J. 22:546-547. Murray, D.A. AND J.P. O’CONNOR. 1982. Microchironomus deribae (Freem.) (Diptera, Chironomidae) a fuel contaminant in an Irish helicopter. Ent. Mo. Mag. 118:44. PINDER, L.C.V. AND P.S. CRANSTON. 1976. Morphology of the male imagines of Ortho- cladius (Pogonocladius) consobrinus and O. Glabripennis with observations on the tax- onomic status of O. glabripennis. Ent. Scand. 7:19-23. STRENZKE, K. 1950. Systematik, Morphologie, und Okologie der terrestrischen Chirono- miden. Arch. Hydrobiol. (Suppl) 18:207-414. TABLE 1. Chironomidae recorded from Ireland. Previously unpublished records denoted by* TANYPODINAE Ablabesmyia longistyla Fitt. A. monilis (L.) A. phatta (Egg.) *Anatopynia plumipes (Fries) *Apsectrotanypus trifascipennis (Zett.) Arctopelopia barbitarsis (Zett.) A. griseipennis (Wulp) Clinotanypus nervosus (Meig.) Conchapelopia melanops (Wied.) C. pallidula (Meig.) C. viator (Kieff.) Guttipelopia guttipennis (Wulp) Krenopelopia binotata (Wied.) K. nigropunctata (Staeg.) Larsia atrocincta (Goetgh.) L. curticalcar (Kieff.) Macropelopia goetghebueri Kieff. M. nebulosa (Meig.) M. notata (Meig.) Monopelopia tenuicalcar (Kieff.) Natarsia nugax (Walk.) N. punctata (Fabr.) Nilotanypus dubius (Meig.) Paramerina cingulata (Walk.) P. divisa (Walk.) Procladius (Procladius) barbatus Brund. . (P.) choreus (Meig.) . (P.) crassinervis (Zett.) . (P.) culiciformis (Linn.) . (P.) sagittalis Kieff. . (P.) signatus (Zett.) . (P.) simplicistilus Freem. . (P.) sp. cf. crassinervis (Zett.) Procladius (Psilotanypus) albinervis (Kieff.) P. (Ps.) flavifrons Edw. P. (Ps.) lugens Kieff. P. (Ps.) ruffovittatus (Wulp) Psectrotanypus varius (Fabr.) Rheopelopia eximia (Edw.) R. maculipennis (Zett.) *R. ornata (Meig.) Tanypus punctipennis (Meig.) Telmatopelopia nemorum (Goetgh.) Thienemannimyia carnea (Fabr.) T. laeta (Meig.) T. lentiginosa (Fries) T. northumbrica (Edw.) T. pseudocarnea Murray Trissopelopia longimana (Staeg.) Xenopelopia falcigera (Kieff.) na} ‘ae} qe) 'x9) '9e) '9) a9) D.A. MURRAY AND P. ASHE 229 X. nigricans Fitt. *Zavrelimyia barbatipes (Kieff.) Z. hirtimana (Kieff.) Z. melanura (Meig.) Z. nubila (Meig.) PODONOMINAE Parochlus kiefferi (Garrett) BUCHONOMYIINAE Buchonomyia thienemanni Fitt. PRODIAMESINAE Monodiamesa bathyphila Kieff. Prodiamesa olivacea (Meig.) P. ruffovittata (Goetgh.) DIAMESINAE Diamesa bohemani Goetgh. D. chiron (Hal.) D. cinerella (Meig.) D. incallida (Walk.) D. insignipes Kieff. *D. permacer (Walk.) D. thienemanni Kieff. Potthastia gaedii (Meig.) P. longimana Kieff. P. montia (Edw.) Protanypus morio (Zett.) Pseudodiamesa branickii (Now.) P. nivosa (Goetgh.) TELMATOGETONINAE Thalassomya frauenfeldi Schin. ORTHOCLADIINAE Acricotopus lucens (Zett.) Brilla longifurca Kieff. B. modesta (Meig.) Bryophaenocladius furcatus (Kieff.) B. nitidicollis (Goetgh.) *B. subvernalis (Edw.) *B. virgo Thien. Camptocladius stercorcarius (deGeer) Cardiocladius fuscus Kieff. Chaetocladius dentiforceps (Edw.) *C. perennis (Meig.) Clunio marinus Hal. Corynoneura carriana (Edw.) MEM. AMER. ENT. SOC., 34 . celeripes (Winn.) . celtica Edw. . coronata (Edw.) . edwardsi Brund. . lacustris Edw. lobata Edw. . scutellata Winn. ricotopus (C.) albiforceps (Kieff.) . (C.) annulator Goetgh. . (C.) bicinctus (Meig.) (C.) curtus Hirv. . (C.) epipphium (Zett.) (C.) festivellus (Kieff.) (C.) fuscus (Kieff.) (C.) pulchripes Verr. . (C.) similis Goetgh. (C.) tremulus (L.) . (C.) triannulatus (Macq.) (C.) trifascia Edw. . (C.) tristis Hirv. . (Isocladius) laricomalis Edw. (I.) ornatus (Meig.) (I.) reversus Hirv. . (1.) sylvestris (Fabr.) . (I.) tricinctus (Meig.) . (1.) trifasciatus (Meig.) . (Nostococladius) lygropis Edw. *Diplocladius cultiger Kieff. Epoicocladius flavens (Mall.) Eukiefferiella brevicalcar (Kieff.) . claripennis (Lund.) . clypeata Kieff. . coerulescens Kieff. . devonica (Edw.) . dittmari Lehm. . gracei (Edw.) . ilkleyensis (Edw.) . minor (Edw.) Eurycnemus crassipes (Panz.) *Georthocladius luteicornis (Goetgh.) Gymnometriocnemus brumalis Edw. *Halocladius (Halocladius) fucicola (Edw.) H. (H.) variabilis Staeg. H. (H.) varians (Staeg.) H. (Psammocladius) braunsi (Goetgh.) Heleniella ornaticollis (Edw.) Heterotanytarsus apicalis (Kieff.) Heterotrissocladius grimshawi (Edw.) H. marcidus (Walk.) * APDQOGHAGAGAGAZOHHGOOOGAAGHOHOO eolesi coil esi col colmcsimes| 230 INVENTORY OF IRISH CHIRONOMIDAE *Krenosmittia camptophleps (Edw.) Limnophyes exiguus (Goetgh.) L. gurgicola Edw. L. minimus (Meig.) L. prolongatus (Kieff.) L. truncorum Goetgh. Metriocnemus atriclavus (Kieff.) M. fuscipes (Meig.) M. hirticollis (Staeg.) M. hygropetricus Kieff. M. picipes (Meig.) *Nanocladius (Nanocladius) balticus (Palm.) N. (N.) bicolor (Zett.) *N. (N.) rectinervis (Kieff.) Orthocladius (Eudactylocladius) femineus (Edw.) . (Evorthocladius) rivicola (Kieff.) . (E.) rivulorum (Kieff.) . (E.) thienemanni Kieff. . (E.) rusticus ? Goetgh. (Orthocladius) dentifer Brund. . (O.) frigidus (Zett.) . (O.) oblidens (Walk.) . (O.) rubicundus (Meig.) . (O.) rhyacobius (Kieff.) . (O.) saxicola (Kieff.) . (Pogonocladius) consobrinus (Holm.) Orthosmittia albipennis (Goetgh.) Paracladius conversus (Walk.) Parakiefferiella bathophila (Kieff.) P. coronata (Edw.) Paralimnophyes hydrophilus (Goetgh.) Parametriocnemus stylatus (Kieff.) Paraphaenocladius impensus (Walk.) P. irritus (Walk.) P. pseudirritus Str. Paratrichocladius rufiventris (Meig.) P. skirwithensis (Edw.) Paratrissocladius excerptus (Walk.) Psectrocladius (Allopsectrocladius) obvius (Walk.) P. (A.) platypus Edw. P. (Mesopsectrocladius) barbatipes (Kieff.) P. (Monopsectrocladius) calcaratus (Edw.) . (Psectrocladius) barbimanus Edw. . (P.) edwardsi Brund . (P.) fennicus Stora SOOOGOGOOOOO uu 'U P. (P.) psilopterus Kieff. P. (P.) sordidellus (Zett.) “Psectrocladius’’ turfaceus Kieff *Pseudorthocladius curtistylus (Goetgh.) P. filiformis (Kieff.) Pseudosmittia trilobata (Edw.) Rheocricotopus chalybeatus (Edw.) *R. fuscipes (Kieff.) *R. gouini (Goetgh.) *Rheosmittia spinicornis Brund. Smittia aterrima (Meig.) S. contingens (Walk.) S. edwardsi Goetgh. *S. leucopogon (Meig.) S. pratorum Goetgh. Synorthocladius semivirens (Kieff.) *Thalassosmittia thalassophila (Beq. and Goetgh.) Thienimaniella acuticornis Kieff. T. clavicornis Kieff T. flavescens Edw. T. majuscula Edw. T. vittata Edw. Tvetnia bavarica (Goetgh.) T. clavescens (Edw.) T. discoloripes (Goetgh.) T. verralli (Edw.) Zalutschia humphriesiae Dow. and Murr. CHIRONOMINAE Chironomini Chironomus (Camptochironomus) tentans Fabr. Chironomus (Chironomus) annularius Auct. . (C.) anthracinus Zett. . (C.) aprilinus Meig. . (C.) cingulatus Meig. . (C.) dorsalis (Meig.) Auct. . (C.) longistylus Goetgh. (C.) lugubris Zett. . (C.) pilicornis Fabr. (C.) plumosus (L.) . (C.) prasinus Meig. . (C.) pseudothummi Str. . (C.) riparius Kieff. . (C.) salinarius Kieff. . (C.) sp. A. (Pinder) Cladopelma laccophila Kieff. ADADAGGAAGDADGOe D.A. MURRAY AND P. ASHE 231 C. psittacinus (Meig.) C. viridula (Fabr.) Cryptochironomus lateralis (Goetgh.) C. supplicans (Meig.) Cryptotendipes pseudotener Goetgh. Demeijerea rufipes (L.) Demicryptochironomus vulneratus (Zett.) Dicrotendipes lobiger (Kieff.) D. nervosus (Staeg.) D. notatus (Meig.) D. pulsus (Walk.) D. tritomus Kieff. *Einfeldia longipes (Staeg.) *E. pagana (Meig.) Endochironomus albipennis (Meig.) E. dispar (Meig.) E. impar (Walk.) E. tendens (Fabr.) Glyptotendipes barbipes (Staeg.) G. gripkoveni Kieff. G. pallens (Meig.) G. paripes Edw. G. viridis Macq. Graceus ambiguus Goetgh. Harnischia falcata Kieff. (?) *Kiefferulus tendipediformis Goetgh. Lauterborniella agrayloides Kieff. Microchironomus deribae Freem. (in helicopter fuel) Microtendipes caledonicus Edw. M. chloris Kieff. *M. confinis (Meig.) M. pedellus (deGeer) M. rydalensis Edw. M. tarsalis (Walk.) Nilothauma brayi (Goetgh.) Pagastiella orophila (Edw.) Parachironomus arcuatus Goetgh. . frequens (Joh.) . monochromus (Wulp) . parilis (Walk.) . subalpinus Goetgh. . Swammerdami Kreus. . tenuicaudatus (Mall.) Paracladopelma laminata Kieff. Paralauterborniella nigrohalteralis (Mall.) Paratendipes albimanus (Meig.) P. nudisquama (Edw.) P. plebejus (Meig.) vuvyUU'U MEM. AMER. ENT. SOC., 34 Phaenopsectra (Phaenopsectra) flavipes (Meig.) P. (P.) punctipes (Wied.) P. (Sergentia) coracina (Zett.) Polypedilum (Pentapedilum) nubens Edw. . (Pe.) sordens (Wulp) . (Pe.) tritum (Walk.) . (Polypedilum) acutum Kieff. . (P.) albicorne (Meig.) . (P.) arundinetum Goetgh. . (P.) convictum (Walk.) . (P.) cultellatum Goetgh. . (P.) laetum (Meig.) . (P.) nubeculosum (Meig.) . (P.) pedestre (Meig.) . (Tripodura) bicrenatum Kieff . (T.) pullum (Zett.) . (T.) quadriguttatum Kieff. . (T.) scalaenum (Schr.) Stenochironomus gibbus (Fabr.) S. hibernicus Edw. Stictochironomus histrio (Fabr.) S. maculipennis (Meig.) S. pictulus (Meig.) S. rosenschoeldii (Zett.) Xenochironomus xenolabis Kieff. Pseudochironomini Pseudochironomus prasinatus (Staeg.) * ne} 9) Ne) le} le) Me} Me) Me} te) le) Ie} Me) 'of 'e) Tanytarsini Cladotanytarsus atridorsum Kieff. C. difficilis Brund. C. mancus (Walk.) *C. nigrovittatus (Goetgh.) C. vanderwulpi Edw. *Corynocera ambigua Zett. *Microspectra apposita (Walk.) . atrofasciata (Kieff.) . bidentata Goetgh. . contracta Reiss . fusca (Meig.) . groenlandica And. . junci (Meig.) . lindrothi Goetgh. . notescens (Walk.) . recurvata Goetgh. *Neozavrelia luteola Goetgh. Parapsectra nana (Meig.) *Paratanytarsus austriacus Kieff. ee ee a es I 232 INVENTORY OF IRISH CHIRONOMIDAE *P. bituberculatus (Edw.) eile P. confusus Palmen T *P. dimorphis Reiss. T P. inopertus (Walk.) se P. intricatus Goetgh. T P. laccophilus Edw. T P. tenuis (Meig.) 1, *Rheotanytarsus distinctissimus (Brund.) T R. pentapoda Kieff. T Stempellina bausi (Kieff.) T Stempellinella brevis Edw. AIT Tanytarsus bathophilus (Kieff.) ily *T. brundini LInd. T T. buchonius Reiss and Fittkau debilis (Meig.) . eminulus Walk. . glabrescens Edw. . gracilentus Holmer. . gregarius (Kieff.) . lestagei (aggregate) medius Reiss and Fittkau . pallidicornis Walk. . quadridentatus Brund. . Signatus Wulp . Striatulus Lindeberg. . sylvaticus Wulp . usmaensis Pag. Virgatanytarsus arduennensis (Goetgh.) *T. curticornis Kieff. V. triangularis (Goetgh.) TABLE 2. The occurrence of chironomid genera in Ireland, Britain and mainland Europe. Numbers of species recorded to date in Ireland given in brackets ( ). IRELAND BRITAIN EUROPE Tanypodinae 24 (55) 24 Dy Podonominae 1 (1) 2 5 Telmatogetoninae 1 (1) 2 3 Buchonomyiinae 1 (1) 1 1 Diamesinae 4 (13) 6 11 Prodiamesinae 2 (3) 3 3 Orthocladiinae 52 (134) $2 70 Chironominae 43 (139) 43 66 TOTAL 128 (347) 133 186 D.A. MURRAY AND P. ASHE 233 TaBLE 3. The known occurrence of genera and species of the Chironomidae and species of the Ephemeroptera and Trichoptera in Ireland, Britain and mainland Europe. IRELAND BRITAIN EUROPE CHIRONOMIDAE Genera 128 133 186 %Europe 69% 71% 100% % Britain 92% 100% — Species 347 460 1200 %Rurope 29% 38% 100% % Britain 70% 100% — EPHEMEROPTERA Species 29 47 217 %Europe 13% 21% 100% % Britain 62% 100% — TRICHOPTERA Species 144 197 895 % Europe 16% 22% 100% % Britain 73% 100% — MEM. AMER. ENT. SOC., 34 & eR me oz’ i ‘ ‘ P wPEt ™ go" oe iw ~s iat Male Dimorphism in an Arctic Chironomid (Diptera: Chironomidae) D.R. OLIVER Biosystematics Research Institute Agriculture Canada Ottawa, Ontario, Canada KIA OC6 ABSTRACT. — Oliveridia tricornis (Oliver) has two male morphs differing primarily in antennal structure. The long-antennal morph has 13 flagellomeres and the short-antennal morph has female-like antennae with 5-7 flagellomeres. No differences in depth or substrate preference of the larvae, nor seasonal emergence or behaviour of adults were found. The morphs occur in a ratio of 3 short-antennal morphs: 1 long-antennal morph. INTRODUCTION A number of male arctic chironomids are known to have structural modifications such as reduced antennae and wings, strengthened legs, and enlarged hypopygia (Downes 1962; Oliver 1976; Danks 1981). Usually all males of a species share these structural modifications with little or no variation between individuals. But male dimorphism apparently exists in at least two species, Oliveridia tricornis (Oliver 1976) and Hydrobaenus fusistylus Saether (1976). In O. tricornis the two morphs differ primarily in antennal structure. Although Oliver (1976) regarded the two morphs as belonging to the same species, the possibility that they belonged to two sym- patric, sibling species was conceded. This paper is a contribution towards resolving the problem of whether or not male dimorphism exists in Olive- ridia tricornis. Information on the taxonomy, ecology and life history of O. tricornis can be found in Welch (1973, 1976) and Oliver (1976). No comparable dimor- phism has been found in females or immature stages. The larvae inhabit small to large oligotrophic to ultra-oligotrophic lakes. They prefer silt substrates, either in areas of pure silt or in rocky areas with the spaces be- tween the rocks filled with silt. Larval development takes two years. Emergence begins soon after ice-off or in late seasons during the period of ice-off. Adults rarely swarm but congregate for mating on lake shores or on the ice. The males exhibit some of the structural modifications generally associated with loss of the habit of mating in swarms (Brundin 1966; Downes 1969; Hansen and Cook 1976; Oliver 1981). The antennal plume is 235 MEM. AMER. ENT. SOC., 34 236 MALE DIMORPHISM IN AN ARCTIC CHIRONOMID Yo (3 W/ Mi TN nN @ —_ Fics. 1-3. Oliveridia tricornis (Oliver). Antennae: 1. Male, long-antennal morph; 2. Male, short-antennal morph; 3. Female. reduced in both morphs, the legs are somewhat elongated and strengthened, and the hypopygium is enlarged and elaborated. MATERIALS AND METHODS In 1971 and 1972, Dr. H.E. Welch carried out an intensive trapping pro- gram of chironomids emerging from Char Lake (74°12’N, 95°53’ W), Cor- nwallis Island, N.W.T., Canada (Welch 1973). In 1971, submerged emergence traps were laid out along transects radiating from shallow to deep areas of the lake (see Fig. 1 in Welch 1973). Along transects A and C the traps were set over 1, 2, 3, 4, 6, 8, 10, 16, 19, 22, and 23m depths. They were emptied every three days after solar noon. The emergence trap specimens from 1971 were analyzed to determine if differences in depth distribution or pattern or annual emergence existed be- tween the two morphs. There was no essential difference in results from the various transects. Therefore only the results from transect C are presented herein. ANTENNAL ANATOMY The antenna of the long-antennal morph (Fig. 1) has 13 flagellomeres, sensilla chaetica on flagellomeres 1-3 and 13, and no subapical seta. The D.R. OLIVER 237 Fic. 4. Oliveridia tricornis (Oliver). Head with both long and short antennae, frontal view. plume is partly reduced and the extent of reduction varies between in- dividuals. That of the short-antennal morph (Fig. 2) has 5-7, usually 6, flagellomeres, sensilla chaetica on each flagellomere, and a strong subapical seta. It is female-like (cf. Fig. 3) including the setal arrangement and absence of an antennal groove. Several thousand specimens from Char Lake and elsewhere in the North American arctic have been examined and no intermediates between the antennae of the two morphs have been found. However, among the emer- gence trap specimens from Char Lake, three specimens had both types of antennae (Fig. 4). The long antenna was present on the left side of the head in two of the specimens and on the right side of the third. DEPTH DISTRIBUTION The total number of each morph emerging from various depths along transect C is presented in Fig. 5. Emergence patterns and numbers of specimens were similar between transects A and C (compare Fig. 5 with Fig. 6 of Welch 1973). Numbers of short-antennal morphs emerging outnumbered those of the long-antennal morphs by about three to one. Both morphs emerged from all depths sampled, and in the same pattern. Emergence peaked at 2-4m (rock- silt zone) and decreased at 4-8m (moss zone). It peaked again at 10m (silt zone) and steadily decreased from 10-25, the maximum depth sampled. ANNUAL EMERGENCE The annual emergence of each morph is presented in Fig. 6. The short- antennal morph outnumbered the long-antennal morph by about 3:1 but there was no difference in the pattern of annual emergence. There was also MEM. AMER. ENT. SOC., 34 238 MALE DIMORPHISM IN AN ARCTIC CHIRONOMID 120 110 100 90 80 70 60 50 40 30 20 10 EMERGENCE PER DEPTH ZONE Tp pe i 0) 2 4 6 8 10 12 14 16 18 20 22 24 DEPTH (m) Fic. 5. Comparison of total emergence of long-antennal morphs (LA) and short-antennal morphs (SA) against depth. 100 = 90 5 80 70> NUMBER EMERGING PER TRAP PERIOD T I I I I T T I Ql 2s 2) 30) 2 5 8 vl 4 We 2) 23 Aad 2 i 4 7 24 JULY AUGUST SEPTEMBER Fic. 6. Comparison of annual emergence of long-antennal morphs (LA) and short- antennal morphs (SA). no difference in the total duration of emergence between the two morphs as the onset and termination of emergence is the same. Cumulative emergence curves (Fig. 7) emphasize the similarity of annual emergence patterns. As in D.R. OLIVER 239 CUMULATIVE PERCENTAGE OF ANNUAL EMERGENCE 7 27 Ss 2 5 8 WW 14 47 2 23 2 2 JULY AUGUST Fic. 7. Comparison of cumulative’ emergence of long-antennal morphs (LA) and short- antennal morphs (SA). other arctic chironomids, emergence is highly synchronized (Danks and Oliver 1972; Welch 1973), with emergence of both morphs attaining 50% of the annual total at the same time. DISCUSSION The long-antennal and short-antennal morphs of O. tricornis emerged simultaneously from various depths throughout the emergence period. No depth or substrate preference was exhibited between the larvae of the two morphs. The adults congregated in the same areas and mated on substrate. MEM. AMER. ENT. SOC., 34 240 MALE DIMORPHISM IN AN ARCTIC CHIRONOMID No difference in mating behaviour was observed (Oliver 1976; pers. obs.). The two morphs participated in the rarely occurring swarms (Oliver 1976). Thus, there is no evidence that reproductive isolation exists between the two morphs. To the contrary, all ecological and life history data strongly in- dicated that there is ample opportunity for exchange of genetic material be- tween the two morphs and it is concluded that they belong to the same species. The 3:1 ratio of short-antennal morphs is not restricted to Char Lake in the Nearctic. The two morphs consistently were collected together in a 3:1 ratio in other areas of the Nearctic (Oliver 1976; pers. obs.). The short- antennal morph is dominant in the Nearctic but it has not been collected in Spitzbergen (Oliver 1976; Saether pers. comm.). This suggests that the short-antennal morph, undoubtedly the more derived form, has originated in the Nearctic or at least in Beringia. ACKNOWLEDGEMENTS I am indebted to Dr. H.E. Welch, Freshwater Institute, Winnipeg, for allowing me to analyze the emergence trap specimens from Char Lake. I also wish to thank Mr. J.A. Downes, Dr. D. Rosenberg, and Dr. H.E. Welch for critical comments on the manuscript. REFERENCES BRUNDIN, L. 1966. Transantarctic relationships and their significance as evidenced by chir- onomid midges. With a monograph of the subfamilies Podonominae and Aphroteniinae and the austral Heptagyiae. K. svenska Vetenska-Acad. Handl. 11:1-472. Danks, H.V. 1981. Arctic Arthopods. A review of systematics and ecology with particular reference to the North American fauna. Ent. Soc. Can., Ottawa. 608 pp. Danks, H.V. AND D.R. OLIVER. 1972. Seasonal emergence of some high arctic Chirono- midae (Diptera). Can. Ent. 104:661-686. Downes, J.-A. 1962. What is an arctic insect? Can. Ent. 94:143-162. 1969. The swarming and mating behaviour of Diptera. A Rev. Ent. 14:271- 298. HANSEN, D.C. AND E.F. Cook. 1976. The systematics and morphology of the Nearctic spe- cies of Diamesa Meigen, 1835 (Diptera: Chironomidae). Mem. Amer. Ent. Soc. 30:1-203. OLtver, D.R. 1976. Chironomidae (Diptera) of Char Lake, Cornwallis Island. N.W.T., with descriptions of two new species. Can. Ent. 108:1053-1064. 1981. Redescription and systematic placement of Oreadomyia albertae Kevan and Cutten-Ali-Khan (Diptera: Chironomidae). Quaest. Entomol. 17:121-128. SAETHER, O.A. 1976. Revision of Hydrobaenus, Trissocladius, Zalutschia, Paratrissoclad- ius, and some related genera (Diptera: Chironomidae). Bull. Fish. Res. Bd. Can. 195:1-287. WEtcH, H.E. 1973. Emergence of Chironomidae (Diptera) from Char Lake, Resolute, Northwest Territories. Can. J. Zool. 51:1113-1123. 1976. Ecology of Chironomidae (Diptera) in a Polar Lake. J. Fish. Res. Bd. Can. 33:227-247. A Field Method to Quantitatively Sample Sand Invertebrates JANICE G. PETERS ANNELLE R. SOPONIS Entomology & Structural Pest Control Florida A&M University Tallahassee, Florida 32307 ABSTRACT. — The use of CO, flotation was found to be 89-95.6% effective in separating in- vertebrates from sand samples. It can be used under field conditions to obtain samples of living material and to make quantitative estimates of sand populations. For quantitative work, two fresh CO, rinses are necessary. The paper also gives preliminary estimates of populations of Chironomidae in the clean sand of the Blackwater River, Florida. Many streams in northern Florida, particularly the acid streams, have a unique benthos associated with sand substrates. One of these is the Blackwater River, an undisturbed, protected, sand-bottom river in the Blackwater River State Forest, Okaloosa and Santa Rosa Counties, Florida. Many studies on sand substrates have been restricted by a lack of precise methods for separating the fauna from the sand, but there are two studies that provide quantitative estimates of the Blackwater River sand fauna (Bass & Hitt 1977, Tsui & Hubbard 1979). Separating the fauna from the sand either with sieves or by hand has presented problems. Invertebrates, particularly small ones, are probably missed in both processes, and hand-sorting is too tedious and time- consuming for general use. Other separation methods include flotation (soap, oil, sugar, CO,), elutriation, and centrifugation (Edmondson & Winberg 1971). Centrifugation requires that samples be sorted with a cen- trifuge in the laboratory, limiting its use in the field. We tried the other methods at least once under field conditions in an effort to find a method that separated live invertebrates from sand with relative ease and accuracy. Flotation with CO, offered the greatest promise and this paper gives results of experiments designed to find the most efficient way of using CO; to float live invertebrates from the clean sand. METHODS Studies were made in the clean sand of the Blackwater River, Okaloosa Co., Florida, which was described by Beck (1973), Peters and Jones (1973), 241 MEM. AMER. ENT. SOC., 34 242 FIELD QUANTITATIVE SAMPLING OF SAND INVERTEBRATES Bass and Hitt (1977), and Peters and Peters (1977). Sand analysis showed that 75-90% of sand by weight was in the size categories from 0.25-1.00 mm diameter (1,2) with less than 0.5% very fine sand and silt (<0.125 mm diameter). Changes in the relative proportions of © 1 and @2 occurred with changes in current. Experiments were conducted on 26-IV, 2-V, and 9-V-1981 in water 36 to 68 cm deep with a current of 45 to 60 cm/sec and temperatures of 21.2 to 23.5°C. Circular samplers of three standard lengths (S cm, 10 cm, 15 cm) and various widths retrieved approximately 100 cm? of sand per sample. A corer of PVC pipe was used to sample depths greater than 15 cm. Since compact- ness of the sand varied and some of each sample was lost during retrieval, actual sand volume was measured for each sample so that the number of animals per unit volume of sand could be determined with precision. Each sample was placed in a 0.35 litre jar. The jar was filled with water saturated with CO., corked, agitated 10-30 sec, and the elutriate decanted over filter paper. At first, we used bottles of club soda as a source of CO,, but later changed to a soda water bottle with CO, gas canisters. For ex- periments, the filter paper, the sample, and the sand were stained with a solution of 1% rose bengal and preserved in 10% formalin. Formalin was used to keep oligochaetes intact until counted and this was fairly successful. We then counted all specimens floated by the CO, and all specimens left in the sand under a microscope. Each taxon in a sample was treated as an in- dividual trial. Only trials containing more than 10 specimens were included in Table 3. RESULTS Experiments with a single CO, rinse collected only 59.8 + 20.8% of in- vertebrates present (13 trials, range 26.7-94.0%), a result too low and too variable for quantitative work. Results for two samples with four rinses, recharging the CO, before the first and third rinse, are given in Table 1 (data pooled for all groups except Chironomidae). Most animals were col- lected only after the freshly charged first and third rinses, suggesting that a fresh CO, charge was necessary. Table 2 gives total data for a single core us- ing three fresh CO, rinses. Although Chironomidae were under- represented, it was these data that influenced the choice of two CO, rinses since a third rinse only contributed about 5% to the total numbers. Subse- quent samples used two fresh CO, rinses. Two CO, rinses, both freshly charged, will give overall estimates of better than 90% of populations pre- sent (Table 3). In the Chironomidae, an average 95.6% of the population was estimated (range 89.3-100%). J. G. PETERS AND A. R. SOPONIS 243 Number of rinses N 1 2 3 4 Chironomidae 67 Oho CEO Sdo5 «S760 Di TOs V%o8 — S359 GA Other benthos 230 56.5 65.2 90.9 91.3 TABLE 1. Percentage of invertebrates collected after four consecutive rinses with CO, recharged before first and third rinse. Depth Vol- Number/ into ume nee WN Tn sand (cm?) Taxon 1 2 SrsanGaallotall On0— 60 CH 2 2 0 0 4 LA EN 9 6 1 1 17 cm NE 3 2 @) 0) 5 (exe) @) 1 (0) ) 1 4.7- 70 CH ] 3 6) 0) 4 10.1 EN 34 17 3 4 58 cm NE 6 1 1 0) 8 CO Ti OQ 2B 1 14 10.1- Va EN 15 20 2 1 38 157 NE 1 1 @) 0 2 cm CO 4 10 ] 6) 15 15.7- 64 EN 30 4 1 0) S5 20.7 NE 4 0 0) (8) 4 cm CO 78 1 4 0) 83 266 198 68 15 i. 288 TABLE 2. Numbers of invertebrates collected in a single core (2V-1981, water depth of 55 cm) treated with three rinses. CO, was recharged before each rinse. Abbreviations: CH, Chironomidae; EN, Enchytraeidae; NE, Nematoda; CO, Copepoda. MEM. AMER. ENT. SOC., 34 244 FIELD QUANTITATIVE SAMPLING OF SAND INVERTEBRATES N col- N pre- Average Taxon Trials lected sent % /trial Chironomidae 6 424 436 yay es 20) Enchytraeidae 10 910 1016 90.57 VEsSri6 Nematoda 6 363 374 O52 oS Copepoda 3 104 WA 89.0 + 9.1 aS 1801 1938 O2.7 eS TABLE 3. Percentages of invertebrates collected after two rinses with CO, recharged before each rinse. Few of the invertebrates living in the sand can be identified with any degree of certainty. Some of the Copepoda belong to the harpacticoid genus Parastenocaris (based on Pennak 1978). Although they have been sent to specialists, the Nematoda are not yet identified. A single species of En- chytraeidae, Barbidrilus paucisetus Loden & Locy, represents the oligochaetes in clean sand. The Chironomidae, identified from larvae or pupae, belong to the Orthocladiinae (Lopescladius Oliveira, Rheosmittia Brundin, Corynoneura Winnertz, Thienemanniella Kieffer) and the Har- nischia group of Chironomini [Robackia claviger (Townes), ‘‘Cryp- tochironomus’’ sp. Pagast, Demicryptochironomus Lenz]. Additional genera and species of Chironomidae are likely to be found in sand samples from other dates and habitats. DISCUSSION The CO, flotation method was efficient in clean sand substrates and ap- parently anesthetized and bubbled out invertebrates present in the sand; in- dividuals returned directly to fresh water remained alive. It was also effec- tive in fine gravel/sand (unpublished data). The method was unsuccessful in silt habitats because silt was bubbled off with the fauna. We sampled Chironomidae in different sand habitats of the Blackwater and our results using CO, flotation can be compared with those of two pub- lished reports. For comparison, we have converted all results to numbers/m? to a depth of 10 cm into the sand. Bass and Hitt (1977) J. G. PETERS AND A. R. SOPONIS 245 1,2-111-1981* 9,12-V-1981** Body length N % N % =6mm 3 4 5 1S <6mm >1 mm 321 3708) 201 12.7 =1mm 532 Goll (S72 87.0 * 28 samples totaling 1998 cm® sand. ** 26 samples totaling 2315 cm? sand. TABLE 4. Size groupings of Chironomidae collected from sand habitats in the Blackwater River, Florida. surveyed the fauna of the Blackwater using an Ekman dredge and sieves. Their samples taken ‘‘near-shore”’ and “‘mid-river’’ averaged 31 (2-II-1977) and 113 (9-V-1977) chironomids/m?. Tsui and Hubbard (1979) hand-sorted litre samples of sand which had been taken from random depths and preserved in formalin stained with rose bengal. These samples averaged 6,500-9,000 chironomids/m? (II, III, [V-1974). Our averages were 53,744 (1,2-III-1981) and 75,723 (9,12-V-1981). These averages were calculated from totals for three 10 cm samples each at depths of 38 to 45 cm (current 44 to 45 cm/sec) and 60 to 72 cm (current 60 to 62 cm/sec). Mean numbers/cm? for the two habitats were respectively: March, .846 + .102 and .153 + .058; May, 1.044 + .118 and .469 + .211. The differences between our data and those of Bass and Hitt (1977) demonstrate that dredge and sieve methods underestimate populations of sand chironomids. Sieves cannot adequately separate sand particles from midges when both are of similar size, and Flannagan (1970) showed that the Ekman dredge is not reliable in sand. Only when all chironomids less than 6 mm in body length (Table 4) are excluded from our data do the CO, flota- tion estimates compare with the dredge and sieve estimates. Table 4 gives totals for all Chironomidae collected from samples 2 cm to 27 cm into the sand. These data give a rough estimate of the size distribution of chironomids: most were larvae of 1 mm body length or less (Table 4). Quantitative estimates based only on the percentage of larvae greater than 1 mm body length would approximate the figures given by Tsui and Hubbard (1979). However, the samples of Tsui and Hubbard were taken in a dif- ferent year when the river was 15 to 58 cm deeper, so differences in estimates of chironomid populations may represent annual population fluc- tuations or a broadened dispersal. The increased substrate available with a MEM. AMER. ENT. SOC., 34 246 FIELD QUANTITATIVE SAMPLING OF SAND INVERTEBRATES wider stream bed compensates for an apparent reduction in species density, as was shown for mayflies by Lehmkuhl and Anderson (1972). We do suspect that hand-sorting large samples of sand overlooks many chironomids less than 1 mm in body length. In using hand-sorting to evaluate invertebrates left in the sand (Tables 1-3), we used small samples sorted with care under a microscope, and it seems that the small midges were the first to be floated off by CO,.; however, some may have been missed since every sand grain was not individually examined. If so, overall efficiency may be somewhat less than calculated in Table 3. Whether or not hand-sorting is as accurate as CO, flotation, the use of CO, has other ad- vantages: sorting time is reduced; small larvae are easily retrieved; and live material is collected. ACKNOWLEDGMENTS We thank Jarl K. Hiltunen, Great Lakes Fishery Laboratory, Ann Arbor, for identification of oligochaetes, and William L. Peters, A. Elizabeth Gor- don, and Michael D. Hubbard, Florida A&M University, for critical review of the manuscript. We also thank the Florida Game and Freshwater Fish Commission, the Florida Division of Forestry, and Robert Carr, State Fish Hatchery, for help and support. This research was supported by a research program (FLAX 79009) of CSRS, USDA. LITERATURE CITED Bass, D.G., JR. AND V.G. Hitt. 1977. Ecology of the Blackwater River system, Florida. Northwest Streams Res. Proj., Florida Game & Freshw. Fish Comm. Beck, W.M., Jr. 1973. Chemical and physical aspects of the Blackwater River in North- western Florida. Jn: Peters, W.L. & J.G. Peters (eds.), Proc. Ist Int. Conf. Ephemeroptera. E.J. Brill, Leiden. p. 231-239. EDMONDSON, W.T. AND G.G. WINBERG (eds). 1971. A Manual on Methods for the Assess- ment of Secondary Productivity in Fresh Waters (IBP Handbook No. 17). Blackwell Sci. Publ., Oxford. FLANNAGAN, J.F. 1970. Efficiencies of various grabs and corers in sampling freshwater benthos. J. Fish. Res. Board Can. 27:1691-1700. LEHMKUHL, D.M. AND N.H. ANDERSON. 1972. Méicrodistribution and density as factors affecting the downstream drift of mayflies. Ecology 53:661-667. PENNAK, R.W. 1978. Freshwater Invertebrates of the United States. 2nd ed. Wiley Inter- science, New York. PETERS, W.L. AND J. JoNEs. 1973. Historical and biological aspects of the Blackwater River in Northwestern Florida. In: Peters, W.L. & J.G. Peters (eds.), Proc. Ist Int. Conf. Ephemeroptera. E.J. Brill, Leiden. p. 242-251. J. G. PETERS AND A. R. SOPONIS 247 AND J.G. PETERS. 1977. Adult life and emergence of Dolania americana in Northwestern Florida (Ephemeroptera: Behningiidae). Int. Rev. Gesamten Hydrobiol. 62:409-438. Tsul, P.T.P. AND M.D. HupBarp. 1979. Feeding habits of the predaceous nymphs of Dolania americana in Northwestern Florida (Ephemeroptera: Behningiidae). Hydro- biologia 67:119-123. MEM. AMER. ENT. SOC., 34 »y* Observations on the Life-cycles of some Chironomidae in Southern England L.C.V. PINDER Freshwater Biological Association East Stoke, Wareham Dorset, BH20 6BB, England ABSTRACT. — The life-cycles of 10 species of chironomid inhabiting the bottom sediments of a stream in southern England are described. The number of generations per year ranged from 1 to 5, but the pattern for several species differed between the 2 years under discussion, thus illustrating the difficulty of interpreting insect life-cycles from observations made at a single site over a restricted period of time, as is usually the case. INTRODUCTION Despite the wealth of literature dealing with various aspects of the ecol- ogy of the Chironomidae (Fittkau, Reiss & Hoffrichter, 1976; Hoffrichter & Reiss, 1981), the life-cycles of very few species have been elucidated and this is especially the case in relation to riverine species. The observations on which this paper is based were made on a 100 m reach of the Tadnoll Brook, a small chalk-stream in southern England, pre- viously described by Pinder (1974). Chalk streams derive the bulk of their flow from springs emanating from the chalk aquifer. As a result, they are characterized by having relatively stable flow and temperature regimes. In the south of England, chalk spring- water has a more or less constant temperature of 10°C-11°C which has the effect of making the rivers it feeds relatively warm in winter and cool in summer. January is typically the coldest month with a mean river-water temperature of about 6°C and July is the warmest with a mean of about 15°C. Discharge follows a seasonal pattern with peak flows in late autumn and winter, decreasing to a minimum in late summer. Violent floods are rare. The physical, chemical and biological characteristics of southern English chalk-streams are described by Ladle and Casey (1979). THE Stupy SITE Although more or less a typical chalk-stream in terms of its chemistry, the study reach of the Tadnoll Brook differs from most chalk-streams in that its 249 MEM. AMER. ENT. SOC., 34 250 LIFE CYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE E oO 2 oO f= (5) & Qa 1970 1971 Fic. 1. Discharge recorded a short distance upstream of the sampling site. (Asterisk in- dicates that stream was over its banks.) bed is mainly composed of sand and organic detritus rather than gravel. The detritus deposits, which are mostly to be found in slack marginal areas and in the lee of weedbeds, develop during summer and early autum when dis- charge is low and are dispersed by the first floods which usually occur in autumn. Discharge data are available for most of the period under consideration (Fig. 1), and show that flow-rate was high during most of the autumn and winter of 1970, but remained low for the whole of 1971. Temperature data are only available for 1971 (Table 1) but data from other nearby rivers suggest that water temperatures in 1970 were little dif- ferent. SAMPLING METHODS Samples were obtained at roughly 2 week intervals between March 1970 and the end of 1971. On each sampling occasion, 30 cores, each 55 mm in diameter, were taken at random from within a 100 m reach of stream, which at this point is 4 m to 5 m wide with a depth in summer of between 0.5 m TABLE 1. Monthly mean water temperatures at the study site during 1971. JAN FEB MAR APR MAY JUN JUL AUG SEPT OCT NOV DEC 5.6 507) 1) 9.0 NOS 17233 S57 aE 3} IO a 7/ 6.6 L.C.V. PINDER 251 and 1.5 m. The depth to which the corer was inserted into the substratum varied according to the nature of the sediment, but was at least 200 mm. Preliminary sampling had indicated that very few larvae occurred below 100 mm. Samples were washed through a sieve of 125 wm aperture, sorted under 10x magnification and mounted in dimethyl hydantoin formaldehyde (DMHF) resin for subsequent identification and measurement. RESULTS Despite using a relatively fine sieve (125 pm), very few first instar larvae were found. Undoubtedly first, and to a lesser extent second, instar larvae are capable of passing through such a sieve with ease. In presenting the data for individual species estimates of total population density are shown (Fig. 2-11) together with the proportion of fourth instar larvae. Logarithmic, 95% confidence intervals were calculated from the data and in general fell within about 30% of the mean, at least when population size was reasonably large. However, in the interpretation of life-cycles, it is the trends in population size and changes in the proportions of age classes which are of interest, rather than statistically significant differences be- tween particular samples. SUBFAMILY TANYPODINAE Apsectrotanypus trifascipennis (Zett.) Fig. 2 This was the most abundant species of the subfamily with a peak popula- tion density approaching 2000 m~ in November 1971. However, the major- ity of larvae occurred in association with deposits of organic detritus (Pinder 1980), especially coarse material of allochthonous origin so that local densities were much greater. The decline in the number of larvae following autumn floods is well shown by the 1970 data, whereas in 1971 when discharge remained low, no such dramatic decline occurred. The small number of larvae present in the early part of the year makes in- terpretation of this part of the life-cycle very difficult. In both years, the main recruitment occurred from August onwards with evidence of 2 in- fluxes of young larvae in August/September and October/November. Lindegaard-Petersen (1972) found pupae of this species in Linding A dur- ing May and June and from August to September. Ringe (1974) recorded a small number of adults emerging from the Breitenbach in May and a main emergence period extending from July to September. In the present study, a MEM. AMER. ENT. SOC., 34 252 LIFE CYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE 6 of 47 ers No. A. trifascipennis larvae m~-2(x 10-2) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1970 18 14 trifascipennis larvae m~2(x10-9 Q oy O Ov one, O a i 0-0-0,,0—-0 5 ot a Leer RSG PETES POP rae JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1971 No. A. Fic. 2. Population density of A. trifascipennis. Solid line = all larvae, broken line = fourth instar only. few pupae were taken in May 1970 and pupae occasionally occurred in samples during July, August and September of both years. Thus, the life-cycle appears to be similar in each of these 3 rivers with an overwintering population emerging in late spring. In the Tadnoll Brook, Overwintering occurred as third and fourth instar larvae. A small summer generation apparently emerged during July and 2 overlapping generations during late summer resulted in a rapid increase in population density. The L.C.V. PINDER 253 220 140 100 60 No. P. choreus larvaem LS) (o} JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1970 600 520 9 - pS [o) (4) Q (eo) O O 200 Cae O 2 No. P. choreus larvae m re) C) {o} O 120 A 40 p ey \ aN hing e) CaS x 0O-0=—0-O>G Ly woul [| li JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1971 Fic. 3. Population density of P. choreus. Solid line = all larvae, broken line = fourth in- star only. second of these produced the overwintering population, the success of which was largely dependent on the extent to which beds of organic detritus were disrupted by autumn and winter spates. Procladius choreus (Meigen) Fig. 3 Apart from A. trifascipennis, this was the only tanypodine to occur in reasonably large numbers in samples. Trends in population density were MEM. AMER. ENT. SOC., 34 254 LIFE CYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE similar to those of A. trifascipennis and the 2 species occupied apparently similar niches. In 1970, no larvae of P. choreus were found until the end of June. Thereafter, 3 well-defined generations were evident in July/August, August/September and October/November. During the last of these generations, almost all larvae in samples were second instars and the population declined to zero by the end of November, well in advance of the first autumn floods. Larvae did not reappear in samples until May of 1971. Thereafter there was some evidence of a small generation emerging early in July and 2 major periods of recruitment in August/September and October/November. As in the previous year, the great majority of larvae from October onwards were second instars and although numbers declined from early November, a reasonable population persisted to the end of the year. Although there was no sign of an overwintering population at this site in either of the preceding years, it is probable that the very small number of larvae (all 4th instar) which were found in May 1971 represented the residue of an overwintered population which entered the study reach by immigra- tion from upstream. If this interpretation is correct, the life-cycle consists of an overwintering generation emerging in April/May and 2 summer generations. Ford (1957) came to a similar conclusion regarding the timing of the overwintering generation, but detected only 1 protracted summer generation, probably because he sampled at monthly intervals. SUBFAMILY DIAMESINAE Potthastia gaedii (Meigen) Fig. 4 In 1970, numbers of this species were low, with evidence of 2 distinct generations, in spring and late summer respectively. In 1971, when numbers were much greater, the pattern of generations was much more difficult to discern. No indication of an overwintering population was found until the begin- ning of March 1971 when low numbers of third and fourth instar larvae began to appear in samples and persisted until the beginning of May. The timing of this agrees with that of the first apparent generation of 1970 when numbers were much greater. A clearly defined generation occurred in early summer, emerging in the second half of June. After a short break, larvae reappeared and thereafter were present to the end of the year. From July onwards, the data indicate at least 3 periods of recruitment: in July, late August to early September and October/November, with small numbers of L.C.V. PINDER 255 Dy, Oc NOV DEC No. P. gaedii larvae m-2(x10-4) nN P. gaedii m-2 (x1074 No. JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Fic. 4. Population density of P. gaedii. Solid line = all larvae, broken line = fourth in- star only. second and third instar larvae persisting to the end of the sampling period. P. gaedii thus had 4 generations at the study site during 1971. SUBFAMILY PRODIAMESINAE Prodiamesa olivacea (Meigen) Fig. 5 Third and fourth instar larvae of this species first occurred in samples in late May 1970 and by mid-July virtually all had developed to fourth instar. Numbers declined from July onwards and by the latter half of September were absent, reappearing in October and overwintering, largely as fourth in- MEM. AMER. ENT. SOC., 34 256 LIFE CYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE =? No. P. olivacea larvae m JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1970 = No. P. olivacea larvae m JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1971 Fic. 5. Population density of P. olivacea. Solid line = all larvae, broken line = fourth in- star only. star larvae from which adults emerged in early spring. Subsequent trends in 1971 were similar to those of the previous year except that the overwintering population, to the end of December, consisted principally of second and third instar larvae. P. olivacea thus has 2 annual generations in the Tadnoll Brook, with summer development being slow by comparison with the other species con- sidered. Pinder (1974) found adults on sticky-traps during July, August and September, indicating a protracted emergence period. Ford (1957) also noted a similar timing for the overwintering generation but failed to demonstrate adequately the occurrence of a summer genera- tion. L.C.V. PINDER 257 2 || x On A? '~o we di Fr if Yr JAN FEB MAR APR MAY JUN JU o fo) No. O. fulva larvae m-2 OCT NOV DEC 1971 Fic. 6. Population density of O. fulva. Solid line = all larvae, broken line = fourth instar only. Odontomesa fulva (Kieffer) Fig. 6 Except for very occasional second instar larvae towards the end of the year, this species was absent from samples in 1970. Evidence of a small overwintering population was found from the beginning of 1971 with all larvae developing to the fourth instar by early March and emerging as adults by mid-April. Larvae reappeared in samples during mid-May with evidence of a period of emergence during June. Subsequent cohorts are more difficult to define but there are indications of 3 periods of recruitment of young larvae, in July, August and September. No fourth instar larvae were found after the beginning of November, but a small population of se- cond instar larvae persisted to the end of the year. O. fu/va thus appeared to have 5 generations: an overwintering generation which persisted from Oc- tober to April, a well-defined spring generation and 3 overlapping sum- mer/autumn generations. SUBFAMILY CHIRONOMINAE Paratendipes albimanus (Meigen) Fig. 7 This was the only species to have a single generation, the timing of which was identical in the 2 years, although the number of larvae was much smaller in 1971. MEM. AMER. ENT. SOC., 34 258 LIFE CYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE 179 1971 Proportion in 3 Freee WIGTAR 2, 3 & 4 instar Uo, Baa (1970) uly VS July 12 = a i= © Al - 14 x< a ey 20 ; = 12h © = uy 27 | ey 27 = 10- & 2 ar aij aoensS el avoust 2 E a Or —f August 10 a accuse a @ : Ar a oe bz, ae oo 16 fo) z ar | August 24 = | August 23 JUN JUL AUG Fic. 7. Population density of P. albimanus, with proportion in instars 2-4 on successive sampling dates. Solid line = 1970 data, broken line = 1971. Larvae first appeared in significant numbers in July, reached a peak early in August and had virtually disappeared by the end of August. Otherwise only very occasional first and second instar larvae occurred in samples. This life-cycle is very similar to that described for the same species living in a Michigan headwater-stream by Ward and Cummins (1978) and also corresponds with Lehmann’s (1971) description of a flight period during July and August on the Fulda. Ward and Cummins (1978) found that in their situation the population overwintered as first and second instar larvae. The few larvae which were found in the Tadnoll Brook in months other than June, July and August were all early instar which suggests that a similar situation exists here. Paracladopelma camptolabis Kieffer Fig. 8 No trace of an overwintered population was found in early 1970. A clearly defined generation was evident from late May, emerging in July and 2 periods of recruitment of young larvae were apparent in late summer and a further period of recruitment in October gave rise to the overwintering population. This pattern was repeated almost precisely in 1971 except that the overwintering population persisted and was considerably augmented in late February/March, presumably by immigration from upstream. Thus, the life-cycle consists of an overwintering generation, emerging in April, a well defined early summer generation and 2 strongly overlapping generations in late summer, early autumn. L.C.V. PINDER 259 = No. P. camptolabis larvae m JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1970 360 320 SSS No. P. camptolabis larvae m2 i // //// Mii : Lf if MULL z Feaiibeere JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1974 IS Fic. 8. Population density of P. camptolabis. Solid line = all larvae, broken line = fourth instar only. Polypedilum convictum (Walker) Fig. 9 Trends in population density of this species were very different between the 2 years. In 1970, larvae first appeared in mid-June with 2 overlapping but reasonably well-defined generations between then and mid-September; whereafter no further larvae were found until March of the following year. Second and third instar larvae were found during March 1971 with fourth instars appearing early in April. By the end of April all larvae were in the MEM. AMER. ENT. SOC., 34 260 LIFECYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE =2 No. P. convictum larvae m JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC No. P. convictum larvae m ~2 (o) fo} T 20 L 40 l [ O oO JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 1971 Fic. 9. Population density of P. convictum. Solid line = all larvae, broken line = fourth instar only. fourth instar and emergence was completed by late May. Two periods of recruitment of young larvae were apparent during the summer, the first in June/July giving rise to an emergence in August, and the other in August and September, corresponding respectively with the 2 generations observed in 1970. An autumn population first appeared in October and persisted as second instar larvae until the end of the year. Evidently this represented the over- wintering generation. L.C.V. PINDER 261 340 300 260 220 180 140 100 60 No. P. cultellatum larvae m~2 20 FEB MAR No. P. cultellatum larvae m~2 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Fic. 10. Population density of P. cultellatum. Solidline = all larvae, brokenline = fourth instar only. On occasions large fluctuations in the density of larvae occurred which were not obviously related to emergence or to recruitment through oviposi- tion. In late July there appears to have been a mass emigration of third in- star larvae, compensated for by immigration of fourth instar larvae shortly afterwards. A similar explanation probably applies to the temporary drop in numbers in late August, though on this occasion the differences were not MEM. AMER. ENT. SOC., 34 262 \LIFECYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE statistically significant. Presumably such movements are initiated by local changes in environment. It is not possible to say what these changes were but discharge can be eliminated as a possibility since it was more or less con- stant over the period in question. Polypedilum cultellatum Goetghebuer Fig. 10 A well defined peak in numbers of this species occurred in late April 1970, most of which were fourth instars and had disappeared by early May. These larvae must have entered the study reach in the drift and probably represented the tail-end of an overwintered generation. Subsequently, there was evidence of 2 overlapping generations between May and July. No larvae were found during August or September 1970, but a population of third and fourth instar larvae appeared in October and persisted over winter, emerg- ing in late April or early May. Subsequent trends were similar to those of the previous year, except that larvae of an additional, distinct generation were present during August and September. Cladotanytarsus vanderwulpi Edwards Fig. 11 Small numbers of this species occurred in April in both years and gave rise to an emergence in late April/early May. Subsequent clearly defined, separate generations occurred in June/July and August/September. Whereas there was no sign of a generation overwintering from 1970, a large number of second instar larvae persisted to the end of 1971. It is probable that the relatively small number of larvae, all fourth instar, which were found in March and April of 1970 and 1971 represented immigrants from a population which overwintered upstream of the sampling reach. The species thus has a protracted overwintering generation extending from October to April followed by 2 well defined summer generations. DISCUSSION A range of life-cycles is discernible amongst the species considered, from the single annual generation of P. albimanus through to the probable 5 of Odontomesa fulva. In only 3 species, P. albimanus, P. olivacea and C. vanderwulpi with 1, 2 and 3 generations respectively, were all generations distinct. In all other species they overlapped more or less strongly for at least part of the year. Usually this was during late summer and early autumn, but in the case of P. cultellatum it was the 2 early summer genera- tions which overlapped, the later one being distinct. Both this species and P. convictum occur in relatively large numbers in the Tadnoll Brook and they apparently occupy similar niches. In this context, it is interesting to note that the life-cycle of P. cultellatum enabled it to have 2 generations in early summer in advance of the summer generation of P. convictum. L.C.V. PINDER 263 - 3) — T ice 1 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC No. C. vanderwulpi larvae m 2 (x10 No. C. vanderwulpi larvae m-2 (x1079) JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Fic. 11. Population density of C. vanderwulpi. Solid line = all larvae, broken line = fourth instar only. Development times for each generation may be estimated with reasonable accuracy for most species using the available data. The overwintering generation generally occupied about 30 weeks (28 to 33), the only outstan- ding exception being O. fulva which required only about 24 weeks to com- plete this part of its life-cycle. For those species with more than 2 annual MEM. AMER. ENT. SOC., 34 264 LIFE CYCLES OF SOUTHERN ENGLAND CHIRONOMIDAE generations, development time varied between 6 and 13 weeks in spring and early summer and 5 to 10 weeks later in the summer and early autumn. In many cases, recruitment of larvae was evident in late winter or early spring indicating that immigration had occurred from upstream. Similarly in 1970, the autumn generation of P. cultellatum was lacking at the study- site, but an overwintering generation was present from mid-October on- wards. In this instance, the overwintering population must either have been derived from larvae entering the reach in the drift or from oviposition by immigrant females. The general point arising from this variability is that a particular site may not be consistently suitable for colonization by a particular species so that observations made at one site, especially over a restricted period of time, are liable to produce an incomplete picture and lead to erroneous conclusions regarding life-cycles. Ladle et a/. (1977) also noted variation in the number of generations of Simulium spp. apparent at different sampling sites. In this present study, several species of chironomid showed no indication of a population overwintering from 1969 or 1970. In contrast, all species maintained relatively high population densities to the end of 1971. An unusual feature of the latter months of 1971 was the lack of any significant increase in discharge. The study reach was an almost straight channel with rather uniform cross-section of steep banks and flat bottom. A spate is therefore likely to influence the entire reach in a uniform manner. Other sections of stream, incorporating meanders and deeper pools would be more variably affected and be more likely to retain suitable habitats. The 2 tanypodine species are particularly prone to be washed out by spates since their preferred substratum of organic detritus can only persist in regions of low water velocity. In 1970 the number of P. choreus however, declined well before the onset of autumn floods. P. choreus larvae were shown to overwinter mainly as second instars whereas A. trifascipennis overwintered in the third and fourth instar. The smaller size of overwinter- ing P. choreus larvae made them suitable prey for A. trifascipennis and their remains were frequently found in the guts of the latter species. The apparently consistent life-cycle of Paratendipes albimanus in rivers in Canada and Europe is particularly interesting in view of the work of Ward and Cummins (1979) which showed that the life-cycle can be modified by altering the quality of food given to the larvae and their observation that in lakes this same species apparently has 2 annual generations. ACKNOWLEDGEMENTS Thanks are due to Mrs. Angela Matthews who assisted at all stages of this study and to Mrs. Valerie Palmer who typed the manuscript. L.C.V. PINDER 265 REFERENCES FITTKAU, E.J., F. REISS AND O. HOFFRICHTER. 1976. A bibliography of the Chironomidae. K. norske Vidensk. Selsk. Mus. Gunneria 26:1-177. Forp, J.B. 1957. A study of the biology and distribution of mud-dwelling chironomid larvae in a chalk stream. Ph.D. Thesis. (University of Southampton.) 225 pp. HOFFRICHTER, O. AND F. REISS. 1981. Supplement 1 to ‘‘A bibliography of the Chironomi- dae.’’ K. norske Vidensk. Selsk. Mus. Gunneria 37:1-68. LADLE, M., J.A.B. Bass, F.R. PHILPOTT AND A. JEFFERY. 1977. Observations on the ecol- ogy of Simuliidae from the River Frome, Dorset. Ecological Entomology 2, 197-204. LADLE, M. AnD H. Casey. 1979. The ecology of southern English chalk streams. Salm. Trout Mag. 217:52-57. LINDEGAARD-PETERSON, C. 1972. An ecological investigation of the Chironomidae (Dip- tera) from a Danish lowland stream (Linding A). Arch. Hydrobiol. 69:465-507. PINDER, L.C.V. 1974. The Chironomidae of a small chalk stream in southern England. Ent. Tidskr. Suppl. 95:195-202. 1980. Spatial distribution of Chironomidae in an English chalk stream. In D.A. Murray (ed.), Chironomidae, Ecology, Systematics, Cytology and Physiology. Pergamon Press, Oxford. 153-161. RincE, F. 1974. Chironomiden-Emergenz 1970 in Brietenbach und Rohrwiesenbach. Schlitzer Produktionsbiologische Studien (10). Arch. Hydrobiol. Suppl. 45:212-304. Warp, C.M. AND K.W. Cummins. 1978. Life-history and growth pattern of Paratendipes albimanus in a Michigan headwater stream. Ann. ent. Soc. Am. 71:272-284. _.. 1979. Effects of food quality on growth of a stream detritivore, Paratendipes albimanus (Meigen) (Diptera: Chironomidae). Ecology 60:57-64. MEM. AMER. ENT. SOC., 34 7K Chironomid Longitudinal Distribution and Macroinvertebrate Diversity along the Llobregat River (NE Spain) NARCIS PRAT, MARIA-ANGELS PUIG, GLORIA GONZALEZ AND XAVIER MILLET Departament d’Ecologia Facultat de Biologia Universitat de Barcelona Avda. Diagonal, 645, Barcelona-28, Spain ABSTRACT. — Qualitative samples were taken at 18 riffles along the Llobregat river (NE Spain) during August and September 1979 and April 1980. For each sample, the populations and com- position of the macroinvertebrate fauna present were considered. Composition of the chironomid fauna, its distribution along the river, and temporal and spatial changes are discussed. The differences between the species-rich communities of the upper and medium rif- fles, and the simplified communities of the lower river reaches, are a consequence of human disturbance such as flow regulation and pollution. A high correlation exists between the extent of disturbance and changes in species richness and diversity patterns along the river, when these are based on the macroinvertebrate collections and on the relative abundance of each species in the sampled communities. The patterns are similar to those predicted by the ‘‘In- termediate Disturbance Hypothesis’’ (Ward & Stanford 1983). The importance of completing the classification of Chironomidae to the same taxonomic level as the other macroinvertebrates, before any general speculation can be made regarding species richness and diversity patterns along the rivers, is discussed. INTRODUCTION Chironomidae are an important component of the macroinvertebrate fauna in rivers, streams and brooks, and considerable work has been devoted to the identification of species living on stones in riffle areas. Therefore, the community structure is well known for some European and American rivers (Thienemann, 1954; Lindegaard, 1972; Lehmann, 1971; Lesage & Harrison, 1980; Laville, 1981). Some of these studies were based primarily on the collection of pupal exuviae or on the use of adult emergence traps, due to the difficulty of larval species identification, especially those of the Cricotopus-Orthocladius group. (Hirvenoja, 1973; Soponis, 1977; Kownacki & Zosidze, 1980). This is the first study of the chironomids of a Spanish river although some previous information is available regarding riverine species caught near different Catalonian rivers (Prat, 1976) and Spanish reservoirs (Prat, 1979, 1980). In this work we present the species composition and longitudinal distribution of chironomid larvae in the river Llobregat (NE 267 MEM. AMER. ENT. SOC., 34 268 | CHIRONOMID DISTRIBUTION ALONG LLOBREGAT RIVER Ni PAPER-MILL a MOUNTAIN RIVER LA BAELLS__, SS RESERVOIR se INTERMEDIATE ZONE LOWER ZONE RUBI RIVER STRONG POLLUTED a =6TRIBUTARY ANOIA RIVER POLLUTED TRIBUTARY STRONG POLLUTED ZONE 68 | +—SALT INPUT SALT | | 0 20 Km — | Fic. 1. Location of the river Llobregat and sampling points along the river. The most strik- ing features affecting macroinvertebrates are indicated. Spain). We shali also discuss the importance of having a correct generic or specific identifications of the larvae, if one wishes to recognize the diversity patterns of the invertebrate fauna along the rivers. Stupy AREA The river Llobregat is located in Catalonia (NE Spain). It is a typical mediterranean river 145 km. long, with a mean annual discharge of 700 Hm? and characteristic seasonal changes in flow with maxima in spring and fall and reduced flow in summer (Prat ef al., 1982). The river flow is regulated from up stream by the ‘‘La Baells’’ dam with a storage capacity of 125 Hm’. Many weirs, built in the XIX century to provide electrical power N. PRAT, M.-A. PUIG, G. GONZALEZ AND X. MILLET 269 to textile mills, are present along the river course. Water is alkaline, and domestic and industrial effluents enter the river especially in its lower stretches. Detailed information concerning the morphological, physical and chemical characteristics of the river is included in Puig ef a/. (1981), and Prat et al. [1982, and 1983 (in press)]. METHODS Eighteen sampling stations were established along the river (Fig. 1). At each station 23 physico-chemical measurements were taken, together with macroinvertebrate samples. Each station was visited in September and December of 1979 and in April of 1980. An extensive effort was undertaken in order to collect most of the species present on stones. Only rocky fast-current zones were studied, using stratified samples (Resh, 1980). Material was preserved and later sorted in the laboratory. In the sorting process, we routinely counted the animals present and made the assumption that the relative proportions of the dif- ferent species in the sample reflected their relative abundance on stones. The data were used to attempt a quantification of diversity. In any case, a minimum of 200 individuals were sorted, counted and identified, and at least 100 chironomid larvae were identified from each sample. RESULTS Chironomid components of the fauna and longitudinal distribution. — Larvae of the Chironomidae were diverse and abundant in all 18 stations along the river. Pupae, occasionally found and some of these, were males with fully developed genitalia; therefore, the most common larval types could be identified to species. The associations were found after the ex- amination of all pupae present in the samples. In many cases, the larval ex- uvia was attached to a male pupa permitting clear larval-pupal-adult association. In other cases, however, it has been impossible to identify the larvae to the specific level, especially in some scarce larvae of the Eukief- feriella and Cricotopus-Orthocladius group (Fig. 2). The pattern of species distribution in three different seasons emphasizes the importance of seasonal change (Fig. 2). Some species are rare in summer or winter (as Eukiefferiella sp. 1, Eukiefferiella claripennis or Nanocladius rectinervis) and common in spring (April 1980). Other species seem to be found only in certain parts of the river. Diamesa spp. with Paracricotopus niger and Parorthocladius nudipennis are com- mon only upstream from the La Baells reservoir (sampling point 60). On the MEM. AMER. ENT. SOC., 34 270 Thienemannimyia group Ablabesmyia sp. Rheopelopia ornata Diamesa spp. Potthastia gaedi Sympotthastia zavreli Thienemanniella sp. Paracricotopus niger Parorthocladius nudipennis Tvetenia calvescens Eukiefferiella claripennis Eukiefferiella sp. 1 Eukiefferiella sp. 2 Cardiocaldius fuscus Synorthocladius semivirens Rheocricotopus chalybeatus Nanocladius rectinervis Brillia_ sp. Parametriocnemus stylatus Cricotopus gr. arnulator Cricotopus bicinctus Cricotopus sylvestris Orthocladius (E) gr. rivicola Orthocladius (E) rivulorum Orthocladius (O) sp. 1 Orthocladius (O) sp. 2 Orthocladius (O) sp. 3 Orthocladius (O) spp. Polypedilum gr. convictum Polypedilum gr. nubeculosum Pentapedilum sp. Cryptochironomus sp. Chironomus sp. Limnochironomus sp. Micropsectra sp. Neozavrelia sp. Tanytarsus sp. Fic. 2. 56 57 a CHIRONOMID DISTRIBUTION ALONG LLOBREGAT RIVER 54 53 60 52 67 68 103 102 101 97 95 94 91 — a {_—_} ——S |__|} {—__] — ——| ——— ——} E _——. =e) ——" 90 89 ———SS SS —— _———} ae) __ OOO a ———} aaa eee a [—— — = —=]) SS SS i——s= £ 3 i= =| =a — : = a —} Oe — — e333 a5 a ———— ——— upper half black-rectangles. April 1980; Lower all black rectangles. Longitudinal distribution of chironomids along the Llobregat at three different | seasons of the year. September 1979; upper-white lined rectangles. December 1979; Middle- N. PRAT, M.-A. PUIG, G. GONZALEZ AND X. MILLET 271 other hand, Synorthocladius semivirens, Nanocladius rectinervis, and Tvetenia calvescens are more frequent in the intermediate reaches, between the sampling points 60 and 67 (Fig. 2). Isocladius sylvestris, Chironomus sp. and Cricotopus bicinctus are common in the lower parts of the river. Data about the relative abundance of the different species will be presented in Prat et al. (1983, in press). Species richness and diversity. — Species richness and diversity indices have been used in many invertebrate surveys on rivers (Bournaud & Keck, 1980) as a measure of the community structure. One of the most widely used indexes is the Shannon’s (Statzner, 1981) which some authors (Cook, 1976; Godfrey, 1978) state to be inversely correlated with the water pollution. One of the problems encountered in the use of such an index to macroinvertebrate collections is the taxonomic level of identification. Nor- mally, im papers concerning species richness or diversity of the macroinvertebrates in rivers, the identification level is not the same for all the organisms sampled, especially in the Chironomidae, which, despite their abundance, are considered, in most cases, only to the family level. For these reasons, absolute values in number of species or diversity can be very dif- ferent as we found in the Llobregat. In this river, the richness of the invertebrate fauna (sensu Ward & Stan- ford, 1981) at three distinct times of the year (Figs. 3 and 4), presents a quite different set of values depending on whether we consider all the Chironomidae as a single taxon (Fig. 3) or whether each midge species is in- dividually identified (Fig. 4). However, the pattern of the total species number along the river may be very similar in both cases, with the max- imum richness in the upper and middle reaches (Figs. 3 and 4, stations 54 to 67). If the chironomids are identified as separate species, richness in station 60 (Fig. 4) is very close to the value found by Ward (1974) below deep- release reservoirs. This station is 4 km. downstream La Baells reservoir and the water temperature is influenced by the hypolimmical water released from the dam (Prat ef a/., 1982, Fig. 16 Puig ef a/., 1981). The relative number of Chironomidae is also very important in all the sta- tions along the river. The diversity indices of all the macroinvertebrate groups including the Chironomidae, as a single taxon (Fig. 5), are obviously lower than the indices computed with due consideration to the relative pro- portion of the different midges listed in Fig. 2 (Fig. 6). All the species names of the macroinvertebrates and their relative abundance in the samples on September 1979, can be found in Prat ef a/., (1983, in press). With regard to the diversity pattern along the river, no differences appear to exist in it whether the Chironomidae are included as a unique taxon or counted by species. In both cases, this pattern parallels the total species richness (Figs. 3 to 6). On the other hand, the absolute values of the diver- MEM. AMER. ENT. SOC., 34 40 30 272 CHIRONOMID DISTRIBUTION ALONG LLOBREGAT RIVER WU BS S77 Sh 53} 60 D2 O/ be 103 102 10197 95 94 «(ON 90 89 Fic.3. Number of taxonomic units of macroinvertebrates along the Llobregat if the Chironomidae are included as a single taxon. Solid black line—September 1979; dashes— December 1979; dot and dash line—April 1980. TU. 5) SY S45 53 60 o/ oe 103. 102 101 97 95 94 9) 90 89 Fic. 4. Number of taxonomic units along the Llobregat. Chironomids are entered as the numbers of species listed in Fig. 2. Linear symbols as in Fig. 3. sity vary in each case. If we consider the Chironomidae as a single taxon, the highest diversity score is 3 bits and only a few sampling points have diversities over 2 bits (Fig. 5). If we take into account the relative presence of the different chironomid species, the highest diversity is close to 4 bits and diversities under 2 bits are only frequent in the lowest and very polluted part of the river (Fig. 6). N. PRAT, M.-A. PUIG, G. GONZALEZ AND X. MILLET 273 Thus, a general increase in the diversity index is produced if we determine Chironomidae to the same taxonomic level as the other groups. In this sense, if we are to attempt to give an accurate picture of real species richness and its diversity pattern along the rivers, it is necessary to identify the chironomids to species. DISCUSSION The river zonation versus the river continuum concept. — Although many attempts have been made to provide a detailed picture of the rivers (hydrologic, morphometric and even the water quality) only broad, im- precise zones have been established. One of these attempts is the zonation of the rivers using the macroinvertebrates (the Illies & Botosaneanu system). This is based on the presence-absence data of some species in the different sampling points. As a result ‘‘rithron’’ and ‘“‘potamon’’ zones can be de- fined. In the Llobregat, despite the seasonal fluctuations, the longitudinal distribution of the macroinvertebrates, including chironomids, allows us to divide the river into different zones (Prat ef a/., in press, and Fig. 1). The higher, intermediate and lower zones can be easily recognized in the river, either by the water quality criteria or by the longitudinal distribution of the species. The reservoir of La Baells is the limit of the typical mountain zone of the Llobregat (Fig. 1 St. 60). The input of salt after station 68 represents a turn- ing point in the ecology of the river, and after station 91 the strong pollution inputs limit the fauna to just a few resistant species. These zones can be defined by presence of characteristic species in each section but also by the relative abundance of the species. Diverse authors have pointed out the difficulties in applying Illies & Botaseneanu (1963) methodology to some rivers (Hawkes, 1975). This oc- curs because of the lack of any discontinuity, so that the river can be de- fined as a continuum, where the species are replaced successively along its course and no definite zones can be established. The so-called ‘‘river con- tinuum concept’”’ (Vannotte ef a/., 1981; Hawkins & Sedell, 1981) has been applied successfully to some American rivers but as Ward & Stanford (1983) have pointed out, the applicability of the ‘‘river continuum concept’’ to the majority of the streams and rivers is very difficult, because in most cases these are altered by the human activities as dams, weirs or pollution. The changes can be so striking that all the community structure is altered, even in short distances, although different ‘‘zones’’ can be distinguished in the pattern of the species distribution looking to the fluvial ecosystem as a MEM. AMER. ENT. SOC., 34 274 CHIRONOMID DISTRIBUTION ALONG LLOBREGAT RIVER 3) S/ Sh 53 60 52 Of 103 102101 97 95 94 9 90 89 Fic. 5. Shannon diversity of the macroinvertebrate communities along the Llobregat calculated on the basis of the relative presence of the different taxonomic units, considering the chironomids as a single taxonomic group. Linear symbols as in Fig. 3. 56 57 54 53 60 2 O/ (3 103 102 101 97 95 % 89 90 89 Fic. 6. Shannon diversity of macroinvertebrate communities along the Llobregat, when the chironomids are entered as the taxonomic units of the Fig. 2. Linear symbols as in Fig. 3. N. PRAT, M.-A. PUIG, G. GONZALEZ AND X. MILLET 275 whole. In the case of river Llobregat the ‘‘river continuum concept’’ cannot be applied because of the great human impact. On the other hand, the idea of ‘‘Intermediate Disturbance Hypothesis’’ seems to be part of our pattern of species richness and diversity. The disturbance in the Llobregat, and the patterns of species richness and diversity along the river. — In the Llobregat the disturbance by man is diverse and frequent (Prat et al., 1982). It changes the patterns of species distribution, species richness and the relative presence of the species. However, criteria still exist that allow us to recognize different zones (Prat et al. 1982). These zones were recognized, as we have noted previously, by using the Illies and Botosaneanu (1963) criteria. As we have seen before, the species richness runs parallel to the pattern of diversity (Figs. 3-6) and both seem to be related to the disturbance level. The already defined zones along the Llobregat, can be delimitated by the exter- nal influences, and associated with the sharp changes in species richness and diversity (see for instance the differences between the sampling points 68 and 103 before and after the salt inputs). Therefore, the patterns of the species richness and diversity are well correlated to the level of disturbance in each sampling point or zone. We can then consider the faunal richness fluctuations within the frame of disturbance level. At its origin (station 56) the Llobregat is undisturbed and the low water temperature is almost constant (Prat ef a/., 1982, Fig. 16). Species richness is not high, and as the relative abundance of each species is very low, the diversity can reach high values (Figs. 4 and 6). At station 57, the Llobregat is still a mountain river, but the water quality has been strongly affected by a paper mill and by sewage effluents. As a result, species richness decreases (nearly all the species present are Chironomidae as can be seen by comparing Figs. 3 and 4) along with diver- sity. Higher species richness and diversity are regained by the river between stations 53 and 54, where no human influence is present. The dam (station 60) increases the species richness and diversity in the river (Figs. 4 and 6), probably as a result of the regulation of the flow and damping of temperature fluctuations. A sharp decrease in species richness and diversity takes place between sta- tions 67 and 103 (Figs. 4 and 6). Here the river receives salt from mining ac- tivity and organic pollution increases. The disturbance is followed by a decrease of the richness and diversity of some very abundant species, especially Hydropsyche exocellata (Trichoptera). A great change in the quality of water occurs when the Llobregat receives some tributaries in its lower part, and the input from sewage and industrial MEM. AMER. ENT. SOC., 34 276 CHIRONOMID DISTRIBUTION ALONG LLOBREGAT RIVER effluents in the area around Barcelona (sampling points 89 and 90). Few species occur in these zones, mainly chironomids and tubificids. Despite the lack of a more accurate quantitative data and the broad periods between samples, species richness and diversity patterns seem to agree with the basic idea of the ‘“Intermediate disturbance hypothesis’’ given by Ward & Stanford (1983) as applied to rivers. Higher diversity is found in areas with an intermediate disturbance (st. 53 to 67), more than in very constant conditions (station 56) or when the disturbance is very great (stations 57, 103, 94) or even extreme (station 91 to the mouth). The Llobregat seems to answer to human impact-increasing or keeping a higher species richness and diversity when the extent of such actions is moderate (according to the flow conditions), and impairing the community structure when the disturbance is greater. The application of some ideas like the ‘Intermediate disturbance hypothesis’? can be a helpful instrument for generalization in the very disturbed rivers, as are those of the Mediterranean basin, and could be an alternative to ‘‘river zonation”’ or “‘river continuum’’ concepts. In any case, one should not attach excessive importance to species richness and diversity until an equivalent degree of taxonomic level is reached in the different macroinvertebrate groups considered in macroinvertebrate surveys. In some cases, as in the Llobregat, we cannot dispense with the most frequent and abundant group, the Chironomidae, despite the systematic difficulties involved. ACKNOWLEDGEMENTS Thanks are expressed to Prof. Dr. Margalef for helpful comments and criticisms. Financial support has been received from the Universitat de Barcelona and Diputacio Provincial de Barcelona. REFERENCES BOURNAUD, M. AND G. KECK. 1980. Diversite specifique et structure des peuplements de macro-invertebres benthiques au long d’un cours d’eau: le Furans (Ain). Acta Oecologica Oecol. Generalis. 1:2:131-150. Cook, S.E. 1976. Quest for an index of community structure sensitive to water pollution. Environ. Pollut. 11:269-288. GODFREY, P.J. 1978. Diversity as a measure of benthic macroinvertebrate community re- sponse to water pollution. Hydrobiologia 57:2:111-122. Hawkes, H.A. 1975. River zonation and classification. In: B.A. Whitton (ed.), River ecology, Blackwell Scientific Publications, Oxford. 725 pp., pp. 313-374. HAWKINS, C.P. AND J.R. SEDELL. 1981. Longitudinal and seasonal changes in functional N. PRAT, M.-A. PUIG, G. GONZALEZ AND X. MILLET 7 organization of macroinvertebrate communities in four Oregon streams. Ecology 62:387- 397. Hirvenoya, M. 1973. Revision der gattung Circotopus van der Wulp und ihrer Verwandten (Diptera, Chironomidae). Ann. zool. fenn. 10:1-363. ILLIES, J. AND L. BOTOSANEANU. 1963. Problemes et methodes de la classification et de la zonation ecologique des eaux courantes, considerees surtout du pont de vue faunistique. Mitt int. verein. Limnol. 12:57 pp. Kownack!i, A. AND R.S. ZosmpzeE. 1980. Taxocens of Chironomidae (Diptera) in some rivers and streams of the Adzhar ASSR (Litte Caucasus Mnts). Acta Hydrobiol. 22:1:67-87. LAVILLE, H. 1981. Récolte d’exuvies nymphales de chironomides (Diptera) dans le haut-Lot de la source (1295 m), au confluent de la Truyére (223 m). Annls. Limnol. 17(3):225-289. LEHMANN, J. 1971. Die Chironomiden der Fulda. Systematische, Okologische und faunis- tische Untersuchungen. Arch. Hydrobiol. Suppl. 37:466-555. LESAGE, L. AND A.D. HARRISON. 1980. The biology of Cricotopus in an algal enriched stream: Part. 1. Normal biology. Arch. Hydrobiol. Suppl. 57:375-418. LINDEGAARD-PETERSEN, C. 1972. An ecological investigation of the Chironomidae (Dip- tera) from a Danish lowland stream (Linding A). Arch. Hydrobiol. 69:465-507. MarGALEF, R. 1981. La Biosfera, entre la termodinamica y el juego. Omega. Barcelona. 236 pp. Prat, N. 1977. Quironémidos de Catalunya Graellsia XXXI: 157-185. 1979. Quirondmidos de los embalses espafioles (1) Graellsia XX XIII:37-96. 1980. Quironédmidos de los embalses espafoles (II) Graellsia XXXIV:59-119. Prat, N., M.A. Puic, G. GONZALEZ AND M.J. Tort. 1982. Prediccid i control de la qualitat de les aigiies dels rius Besds i Llobregat. I: Els factors fisics i quimics del medi Monografies 6:206 pp. Servei del medi ambient de la Diputacio de Barcelona. Prat, N., M.A. Puic, G. GONZALEZ, J.M. TorT AND M. EstrRADa. in press. The Llogrebat: A mediterranean river fed by the Pyrenées. Jn: B.A. Whitton (ed.), The ecology of Euro- pean rivers. Studies in Ecology. Blakwell. London. Puic, M.A., I. BAutista, M.J. Tort AND N. Prat. 1981. Les larves de trichoptéres de la riviére Llobregat (Catalogne, Espagne). Distribution longitudinale et relation avec la qualité de l’eau. Proc. of the 3rd Int. Symp. on Trichoptera. G.P. Moretti (ed.). Series entomologica, Vol. 20, Dr. W. Junk Publishers, 303-309 pp. ResH, V.H. 1979. Sampling variability and life history features: basic considerations in the design of aquatic insect estudies. J. Fish. Res. Board. Can. 36:290-341. Soponis, A. 1977. A revision of the nearctic species of Orthocladius (Orthocladius) van der Wulp (Diptera: Chironomidae) Mem. Entomol. Soc. Canada. 102:187 pp. STATZNER, B. 1981. Shannon-Weawer diversity of the macrobenthos in the Schierensee- brooks (North Germany) and problems of its use for the interpretation of the community structure. Verh. Internat. Verein. Limnol. 21:782-786. THIENEMANN, A. 1954. Chironomus. Leben, Verbreitung und wirtschaftliche Bedeutung der Chironomiden. Binnengewdsser 20:834 pp. VANNOTE, R.L., G.W. MINSHALL, K.W. Cummins, J.R. SEDELL AND C.E. CUSHING. 1980. The river continuum concept. Can. J. Fish. Aquat. Sci. 37:130-137. Warp, J.V. 1974. A temperature stressed stream ecosystem below a hypolimnial release mountain reservoir. Arch. Hydrobiol. 74:247-75. Warp, J.V. AND R.G. DuFForD. 1979. Longitudinal and seasonal distribution of macro- MEM. AMER. ENT. SOC., 34 278 | CHIRONOMID DISTRIBUTION ALONG LLOBREGAT RIVER invertebrates and epilitic algae in a Colorado springbrook-pond system. Arch. Hydrobiol. 86:3:284-321. WarbD, J.V AND J.A. STANFORD. 1983. The intermediate disturbance hypothesis: An ex- planation for biotic diversity patterns in lotic ecosystems. Jn: T.D. Fontaine and S.M. Bartell (eds.), Dynamics of lotic ecosystems. Ann Arbor Science Publishers, Ann Arbor, Mich. 347-356. Three New Species of Lopescladius Oliveira, 1967 (syn. ‘‘Cordites’’ Brundin, 1966, n. syn.), with a Phylogeny of the Parakiefferiella group OLE A. SAETHER Freshwater Institute Winnipeg, Manitoba, Canada and Zoologisk Museum, Univ. of Bergen N-5000 Bergen/Univ. Norway ABSTRACT. — LOPESCLADIUS Oliveira, 1967 is shown to be the valid name for Cordites Brun- din, 1966, illustrated from pupa. Diagnoses for all stages are given. Males and pupae of two new species, Lopescladius fittkaui, Lopescladius verruculosus, the male of one new species Lopescladius inermis, and the pupae of six more species are described. The mainly Neotropical (, but also Nearctic) Lopescladius is shown to be a member of the Parakiefferiella group (Lopescladius, Gynnidocladius Subl. et Wirth, Rheosmittia Brund., Parakiefferiella Thien., Stilocladius Rossaro, Saetheriella Halv., Krenosmittia Thien., Epoicocladius Zavi.) with Gyn- nidocladius from the subantarctic Campbell Island as its sister genus. Rheosmittia apparently form the sister group of these two genera combined. Male imagines of Lopescladius Oliveira associated with numerous pupal exuviae and collected by D. E.J. Fittkau, Zoologisches Staatssammlung, Munich, from Brazil and Mexico were sent me by Dr. F. Reiss, also of Zoologisches Staatssammlung, for description of the pupal stage. A closer examination revealed that the pupae were identical to “‘Cordites’’ Brundin as described by Brundin (1966) on pupae only and thus an invalid name. The specimens from near Manaus in Brazil, and from near Tocuman in Mexico, belonged to two closely related species both apparently differing from the only previously described species, Lopescladius minutissimus Oliveira. Later I received pupal exuvia collected by Dr. Fittkau from Peru and Brazil and belonging to yet another six species. Dr. W.P. Coffman, University of Pittsburgh, in collaboration with Dr. S.S. Roback, The Academy of National Sciences, Pittsburgh, is describing all stages of a new species of ‘“‘Cordites.’’ I received mature pupae from Dr. Coffman for com- parison. Although the male genitalia of their species is quite dissimilar, the remaining male characters as well as the pupae clearly show that also this species is congeneric with Lopescladius, but possibly deserve a separate subgenus. After presenting this paper I received the male of an additional new species from Dr. Ferrington, University of Kansas. 279 MEM. AMER. ENT. SOC., 34 280 NEW SPECIES OF LOPESCLADIUS The general terminology in the following descriptions follows Sather (1980). The measurements are given as ranges followed by a mean when three or more specimens are measured; n=number measured. The types are deposited in the Zoologisches Staatssammlung, Munich (ZSM) and Museum of Zoology, Bergen (ZMBN). Lopescladius Oliveira, 1967:417 Cordites Brundin, 1966; 428, invalid name I.C.Z.N. 13b, n. syn. Unknown gen. & sp. near Corynoneura; Roback 1953:113. Type species: Lopescladius minutissimus Oliveira, 1967:417, by original designation. Other included species: Lopescladius fittkaui n. sp., Lopescladius inermis n. sp., Lopescladius verruculosus n. sp., Lopescladius sp. A (pupa), Lopescladius sp. B (pupa), Lopescladius sp. C (pupa), Lopescladius sp. D (pupa), Lopescladius sp. E (pupa), Lopescladius sp. F (pupa), Lopescladius n. subgen., n. sp. (Coffman and Roback in prep.). Diagnostic characters. — The combination of strongly protruding, but small, strongly pubescent to short haired eyes; reduced antepronotum; wing broad, without costal extension, with R>.3 fused with R,.>, high VR, Cu, straight, and squama bare; cordiform ta, and small pulvilli; gonostylus bent medially with apical megaseta absent; inferior volsella minute and spine-like (Lopescladius s. str.) or broadly digitiform (Lopescladius n. subgen.); and gonocoxite (in Lopescladius s. str.) with caudal elongation; easily separates the males from all other orthoclads. The pupae are characterized by the lack of frontal setae and thoracic horn; the unique leg sheath arrangement with all leg sheaths directed straight backwards with apices joined along sutures; the presence of caudal spines on tergites (I) II-VIII and sternites II-VII (9) or II-VIII (@); the absence of hooklets on tergite II and of pedes spurii A and B; and the anal lobe without fringe, with long, apical, dorsally curved, digitiform, moveable projections each carrying 3 equally long macrosetae. The larvae are characterized by a long antenna longer than head capsule with a long whip-like ultimate segment; S I bifid; premandible with 1 apical tooth and fine brush; mandible with short apical tooth and 6 inner teeth, setae interna present; mentum with broad median tooth and 5 pairs of lateral teeth; separate, claw-bearing parapods; and procercus with 1 strong anal seta. Male. — Eyes small, strongly protruding, with microtrichia as long as the height of an om- matid (hence between pubescent and hairy). Antenna with 11-13 flagellomeres; antennal groove reaching flagellomere 4-5; flagellomeres 2, 3, 4 and ultimate with sensilla chaetica; setal plume weak, with one stronger developed and one weakly developed whorl of setae on each flagellomere; AR between 0.4 and 1.3. Temporals apparently absent. Palp 5-segmented, O. A. SAETHER 281 segments progressively longer, third segment with 1 weak sensillum clavatum. Antepronotum more or less reduced, lobes narrowed medially, separated or in narrow contact, without or with 1 lateral seta. Dorsocentrals few, anterior ones stronger than posterior ones; acrostichals ap- parently absent; about 2 prealars. Scutellars few, in transverse row. Wing broad, membrane with fine punctation of microtrichia barely visible at 350 X. Anal lobe indicated. Costa not or barely extended, ending between M;,, and Cu,; R23 fused with R,,;, partially indicated as a line; vannal fold ends distad, An basad to FCu; Cu, straight; brachiolum with 1 setae, other veins and squama all bare. Sensilla campaniformia about 6 at base of brachiolum, 3 below seta, and about 6 at apex of brachiolum; | at base of subcosta, 1 on FR, and | at base of R,. Pulvilli very small, but distinct. Comb and hind tibial spurs normal. Pseudospurs and sensilla chaetica absent. BV and SV high, ta, strongly cordiform. Setae of abdomen very few, often on- ly 1 very long seta on each tergite. Anal point absent, tergum IX at most with a few weak setae. Phallapodeme well developed, with triangular or rounded aedeagal lobe. Transverse ster- napodeme broadly curved to nearly straight with or without weak oral projections. (The preparations of two of the described males are not very good with the apodemes indistinct.) Gonocoxite without inferior volsella or, with a spine-like inferior volsella (Lopescladius s. str.), or with a broadly digitiform, bare inferior volsella (Lopescladius undescribed subgenus); with a long Protanypus-like apical extension (Lopescladius s. str.) or without such an extension (Lopescladius n. subgen.) Gonostylus with a more or less distinct median bend, without crista dorsalis and apical megaseta. Pupa — Minute pupae (1.3-3.3 mm long). Frontal setae absent. Frontal apotome without anterior triangular projection, evenly and weakly rounded to reniform shape, smooth to strongly rugulose. Antennal sheath above pedical without pearls to covered with rugulosity. Ocular field with 1 weak postorbital only. Thoracic horn absent. Dorsocentrals 4, 2 anterior and 2 posterior grouped. Thorax rugulose. Wing sheath smooth or rugulose and reticulate. Leg sheaths all directed straight backwards with apices joined along sutures, front leg sheath end- ing at conjunctives I/II, mid leg in the middle of tergite II and slightly basad of apex of wing sheath, hind leg ending at conjunctive II/III. Abdomen arched. Tergite I and sternites I and IX without shagreen; tergites II-I[X and sternites II-VIII with transverse rows of fine group shagreen to very strong spinules in anterior third to half, with reticulation of polygones often with smaller pattern of more faint polygone reticulation inside. Tergites I (II)-VIII and ster- nites II-VII (Q ) or II-VIII (oc) with caudal rows of erect spines. Tergite II without hooklets. Conjunctives of tergites and sternites with small distinct to indistinct greyish polygones. Pedes spurii A and B absent. Segment I with 5 pairs of D setae, 3 pairs of V setae and 3 pairs of L setae. Segments IJ-VII each with 4 hair-like L setae, 1-2 of them very small; segment VIII with 2 hair-like L setae. Setae D., D;, V3 and V; broad, lamelliform; tergite VIII with 3 pairs of D setae. O setae and apophyses absent. Anal lobe without fringe, with long, apical dorsally curved, digitiform, moveable, smooth, wrinkled or rugulose projections each carrying 3 equal- ly long macrosetae. Genital sac also curved dorsad. Larva — A description will be given by Coffman and Roback (in prep.) and Cranston, Oliver and Szether (1982). SYSTEMATICS The male imago of Lopescladius s. str. look strikingly different from other orthoclads. The extremely protruding eyes; the broad wings; the distinctly cordiform, Cardiocladius-like ta,; the absence of a megaseta; and the Protanypus-like elongation of the gonocoxite; appears to indicate rela- MEM. AMER. ENT. SOC., 34 282 NEW SPECIES OF LOPESCLADIUS tionships with plesiomorphic genera as well as with highly derived genera. Also the pupa is unique, particularly with respect to the leg sheath arrange- ment and the projections of the anal lobe. The larva is peculiar, but do, however, show some similarities with other genera such as Rheosmittia Brundin (Cranston & Saether in ms). In the key to male Orthocladiinae by Brundin (1956) Lopescladius will key to Camptocladius v.d. Wulp from which it, however, differs in a number of significant details. The Cu, is straight which contradicts a place- ment in either of the Smittia, Parakiefferiella or Pseudosmittia-groups. However, as shown by Szether (1981) for some species of Smittia Holmgr. and Pseudosmittia Goetgh., the straight Cu, almost certainly is a result of FCu being moved far distad and is not an original plesiomorphous condi- tion. The medially bent gonostylus of Lopescladius is very similar to and found only in some members of the Parakiefferiella-group namely Rheosmittia Brund., Gynnidocladius Subl. & Wirth, Parakiefferiella Thien., and Stilocladius Rossaro (Brundin 1956, Sublette & Wirth 1980, Seether 1982, Cranston & Sather in ms). The same genera and Lopescladius also have some members with reduced numbers of flagellomeres; all except Stilocladius and some Parakiefferiella, but with the addition of Krenosmit- tia Thien, have R2,3 fused with R4,;; all except Parakiefferiella, but with the addition of Saetheriella Halvorsen (1982), have pubescent to hariy eyes; and they probably all (the larva of Gynnidocladius is not known) have a whip- like ultimate larval antennal segment. The pupal anal lobe have an apical elongation of differing shape in all known pupae of the Parakiefferiella- group; the pulvilli are present, but small or vestigial probably in all genera; and the ante-pronotum is somewhat reduced except in Epoicocladius Zavt. and in some Parakiefferiella. All in all, Lopescladius clearly appear to be an aberrant and apomorphic member of the Parakiefferiella-group. Whether this group perhaps should include some other genera, particularly of the Pseudosmittia group, which share a few of the same apomorphous features cannot be decided without a better knowledge of the immature stages and of the females. Although the same restrictions applies to the Parakiefferiella group as here outlined a tentative shceme of argumentation delineating the cladogen- sis of the genera can be attempted (Fig. 1). Trends showing the same direc- tion are grouped. The following trends are used (a=apomorphous, p = plesiomorphous): TRENDS 1 — Eyes reduced, strongly protruding (a); less protruding (p). — Costal extension reduced (a); long (p). — Cu, secondarily straightened (see above) (a); curved (p). — Distinctly cordiform ta, (a); slightly cordiform to cylindrical ta, (see Trends 3) (p). O. A. SAETHER 283 — Anal point absent (a); present (p). — Megaseta of gonostylus lost (a); present (p). — Leg sheath arrangement of pupa unique (see above) (a); normal (p). — Projections of anal lobe very long and moveable (a); short to long, not moveable (p). (Several other trends could be added) TREND 2 — Male antenna reduced to 5 nonplumose flagellomeres (a); 11-13 plumose flagello- meres (p). TRENDS 3 — Slightly to distinctly cordiform ta, (a); cylindrical ta, (p). — Anal point reduced or absent (a); well developed (p). — Acrostichals and scutal tubercle absent (a); acrostichals and/or scutal tubercle present (The acrostichals may be secondarily reduced also in Rheosmittia (see trends 5). Whether acrostichals are present or absent in Krenosmittia has not been examined. Other genera of the Parakiefferiella group all have acrostichals, or scutal tubercle or tuft.) — Cnu, straight or at most sharply downcurved at apex (a); Cu, sinuate (p). (The sinuate Cu, is a synapomorphy for the Smittia - Parakiefferiella - Pseudosmittia groups, and symplesiomorphous within these groups.) TRENDs 4 — Scutum extending hunch-like forward (a); normal (p). — Anal macrosetae of pupa absent (a); well developed (p). (Although the pupa of Gynnidocladius is not known it will in all likelihood not possess the autapomorphies of Rheosmittia.) — D, of pupal tergites II or III to IV flattened and finely split (a); normal (p). TRENDS 5 — VR higher than 1.4 (a); VR lower than 1.4 (p). (Secondarily high also in Epoicocladius) — Acrostichals and scutal tubercles absent in all or some species of each genus, i.e. tendency to reduction present (a); present (p). — Ros fused with R,,.; (a); R23 not fused (or secondarily fused in some Parakiefferiella and in Krenosmittia) (p). TREND 6 — Eyes naked without trace of microtrichia between central ommatids (a); eyes pubescent or hairy (p). (See Trends 11). There is some doubt whether Parakiefferiella really is monophyletic as male P. coronata (Edw.) differ distinctly from the other members. However, the immatures appear quite similar to other members of the genus. TRENDS 7 — R>,; fused with R,,; in all or at least some species of each genus; i.e. tendency to fusion present (a); R23; separate and distinct (secondarily fused in Krenosmittia) (p). TRENDs 8 — Anal point narrowed (a); broad-based (p). — Hind tibial comb very oblique (a); normal (p). (The same tendency to an oblique comb is found also in Saetheriella.) — Anal lobe projection reduced (a); present (p). (This trend is regarded as a secondary reduc- tion since Stilocladius possesses the synapomorphies of trend 9.) TRENDs 9 — Gonostylus with a characteristic median bend (a); normal (p). — All genera with some members with reduced numbers of flagellomeres (a); 13 flagellomeres (p). (Epoicocladius gynocera (Edw.) have completely female antenna. Howver, as the immatures are unknown the placement is uncertain. Furthermore, the complete femini- zation of the antenna probably cannot be regarded as the same trend as reduction of a few flagellomeres.) MEM. AMER. ENT. SOC., 34 284 NEW SPECIES OF LOPESCLADIUS — Ultimate segment of larval antenna long, tapering, thread-like (a); normal (p). (Although the larva of Gynnidocladius is not known, this trend probably belongs here.) TRENDs 10 — Costal extension short (a); long or (in Lopescladius) secondarily reduced (p). — Antenna with straight apical seta (a); without (p). (This trend occurs also in the Pseudor- thocladius and Smittia-groups (Szther & Sublette 1983, Brundin 1956).) TREND 11 — Eyes pubescent or hairy (secondarily bare in Parakiefferiella) (a); eyes bare (p). (This trend could be interpreted as going in the opposite direction and probably do so in more plesiomorphic groups.) TRENDs 12 — R»,; fused with R4,.; (a); Roz separate or secondarily fused (see trends 5 and 7). — Larval antenna 4-segmented (a); antenna 5-6 segmented (p). (Parallelly 4-segmented in Epoicocladius, see trend 14.) — Larval maxillary palp unique, Tanytarsiinae-like (a); normal (p). — Procercus with 1 elongate anal seta, half the length of the body (a); anal setae equally long (secondarily one long anal seta in Lopescladius) (p). — Pupal thoracic horn beset with scale-like plates ending in spinules (a); thoracic horn when present without plates (p). (Several other autapomorphics could be added here.) TREND 13 — Antepronotum reduced, with lobes narrowed medially (a); antepronotum not reduced (p). TREND 14 — Larval antenna 4-segmented (a); antenna 5-6 segmented (secondarily 4-seg- mented in Krenosmittia) (p). — Abdomen of larva and pupa with a thick covering of setae (a); normal amount of setae (p). A number of other trends based on the aberrant immatures could be added. TREND 15 — Anal lobe of pupa with an apical more or less well developed elongation (a); with- Out (p). A number of trends could be added here. These, however, depend on whether the com- bined Heterotrissocladius and Cardiocladius groups forms the sister group of the Parakief- feriella group as will be suggested by combining the data in Sather (1977) and Saether and Halvorsen (1981) or if another group is a more likely sister group. A number of trends in the above analysis are partly contradictory. Ob- viously both parallel selections and underlying synapomorphies (Saether 1979) obscure the synapomorphies. If the weighting were placed more on characters such as hairy eyes and fusion on R2.3 with R4,.; instead of on the shape of the gonostylus, the thin ultimate larval segment etc. the placement of the genera from Krenosmittia to Parakiefferiella could well be different. The placement of Lopescladius, Gynnidocladius, Rheosmittia and of Epoicocladius, however, seem relatively well founded. It probably is premature to speculate about the zoogeographical impor- tance of the findings since the actual distribution of many of the genera not is well known. However, the sister group relationship between the mainly neotropical Lopescladius and Gynnidocladius from New Zealand’s subant- arctic Campbell Island as well as the fact that Saetheriella and Stilocladius up to now are known exclusively from the southeastern states and, for the 285 VM MMMM? YM UMMM UMMM MMMM GMM WM Gl _ MLL WMV MA VM MUM MUL UMMM SAETHER CMM MMM (Eat) Gil YIM GMM UMMM ares) { VM EY 8 VL: = [Eee = v VE € (eral = rd =a] O. A. MMM Ol snippjosedo7 SMIPDID —- DIFFISOAYY —DIa!JaJaI4 — SMIPOIDOIIS§-—-DJJaMSYJSDS DIWWUUSOUBy — SNIPD|SOD}od3 -opiuuks =DUDd Scheme of argumentation delineating the cladogenesis of the Parakiefferiella group Fic. 1. by means of trends 1-15. MEM. AMER. ENT. SOC., 34 286 NEW SPECIES OF LOPESCLADIUS last genus, the part of Europe which was part of Gondwanaland , indicate a Gondwanian origin of most of the group. —_ KEY TO KNOWN MALES OF LOPESCLADIUS OLIVEIRA Gonocoxite without caudal extension; inferior volsella broadly digitiform .............. Rea Oe Ore oO Lopescladius n. subgen. n. sp. (Coffman & Roback in prep.) Gonocoxite with caudal extension; inferior volsella spiniform or absent ................ AO AE RO Te Lopescladius'ss Str: sc. fs sac cie oi 00 SOO Antenna with 11 flagellomeres, AR about 1.1; inferior volsella tapering to sharp point, bare (Fig. 2D); extension of gonocoxite without broad apical seta ................. Peer eat eT te Ae oe tte eS Ces ci REROED OSB. Gre OME ocaro Lopescladius fittkaui n. sp. Antenna with 13 flagellomeres, AR higher than 1.2 or lower than 0.8; inferior volsella with a parallelsided base, median seta, and a tapering apex (Fig. 2E) or absent; extension of gonocoxite with or without a broad apical seta........................-.--- 3 AR about 0.75; inferior volsella and broad seta of gonocoxite extension present (Fig. 2E) Ae RE i eG CO Ci OC NAT SUDO SES B's Lopescladius verruculosus n.sp. AR higher than 1.2 or lower than 0.6; inferior volsella and broad seta of gonocoxite exten- sion absent (Fig. 5; Oliveira 1967 fig. 3))-..2.2.-------esese eres sess sense eee 4 ARiaboutilis i5 arccernccsvorenee Lopescladius minutissimus Oliveira (Oliveira 1967 fig. 3) AR AD OU OSS cts saeiece Stccntas archaeon eae tesel epee S ene oe erate neers Lopescladius inermis n. sp. KEY TO KNOWN PUPAE OF LOPESCLADIUS OLIVERIA . Spinules of anterior shagreen bands on tergites and sternites strong, spine-like; anal lobe projection rugulose, about equally wide in the middle as at base, about 2.3 times as long as wide; total length about 2.4mm ............... 2... eee eee eee eee ene Heber s Oe alo home cad c Lopescladius n. gen. n. sp. (Coffman & Roback in prep.) Spinules of anterior shagreen bands not conspicuously strong; anal lobe projection smooth or when rugulose pupa more than 3.0 mm long, projection widest at base, about 2.6 times as long as wide; total length 1.3-2.1 mm or3.0-3.3mm....................-. sMetaealed ute ap Oa cain tape aan eademntcaepe Lee EOpescladiusiS Strive one te 3) 1 ee Exuvia dark brown; total length 3.0-3.3 mm; anal lobe projection very distinctly rugulose (BI YGF) eta tiha einai kano eotans egenetes Seas eg eee eee eA ae eee eee Lopescladius sp. F Exuvia pale yellowish to greyish brown; total length 1.3-2.1 mm; anal lobe projection smooth: to:slightly rugulose .... 3. 4 lec ees sews & os ser nee Ot es Sa a eee 3 Extensive rugulosity extends onto frontal apotome, antennal and wing sheaths (Fig. 4A, B) fsgottuinaas a) apie sds ash avavcuedeveiehueeee sree Means eae crise tele ace Ane ease thee Ene Ene eee 4 Rugulosity well developed, but frontal apotome, antennal and wing sheaths smooth or at most slightly wrinkled or rugulose (Fig. 3A)...............--eeeeee cree eeees 5 Lateral rugulosity of abdominal segments extending for most of the length of each seg- ment (Fig. 4D), digitiform projections of anal lobe 90-113 um long and about 2.8 times as long as basally wide ..................... Lopescladius verruculosus n.sp. Lateral rugulosity of abdominal segments restricted to medial half (Fig. 4E); digitiform projections of anal lobe 75-83 um long and about 3.4-3.7 times as long as basally (0 (RE A a te A eee ny iPods Gilats ols. 4 6n'0.6 bya So G.c6 Lopescladius sp. E Tergite I with more than 2 posterior spines (Fig. 6A, B)............-...---+eeeee eee 6 Tergite I usually without, at most with 2 weak, posterior spines (Fig. 3D)............. 8 O. A. SAETHER 287 6. Anal lobe projection slightly wrinkled (Fig. 6E) to distinctly rugulose medially (Fig. 6D) or near apex, 109-135 um long and 1.2-1.4 times as long as anal macrosetae; exuvia CARES ALCON 0.0.0.6 Od aSlow b uaa.g DOL cIaIolG lceeae old OSLO DICICES ORCI CRI RRR Ione Nee meena 7 Anal lobe projection nearly smooth, 86-101 »m long and 0.9-1.1 times as long as anal macrosetae; exuvium very pale yellowish...................... Lopescladius sp. A 7. Anal lobe projection distinctly rugulose medially (Fig. 6D) 120-135 um long, about 3.6-4.1 times as long as basally wide.............. 0c cee cece eee ueeaee Lopescladius sp. C Anal lobe projection slightly wrinkled but not rugulose (Fig. 6E) 109-116 um long, about 2.9-3.1 times as long as basally wide.....................-.-5. Lopescladius sp. D 8. Exuvium very pale greyish yellow including cephalothorax; anterior spinules in shagrena- tion of segments very sparse, present mostly only anteriolaterally.................. Bong a COOOL O CIS co ONO ern O TOR ra a CeIn ont Cre natn RMI te eee eee Lopescladius sp. B Exuvium with brown cephalothorax and pale greyish brown abdomen; anterior spinule bands weak but nearly complete (Fig. 3D)....................... L. fittkauin. sp. Lopescladius fittkaui n. sp. Type locality: Brazil, Amazonas, near Manaus, Reserva Duke. Type Material: Holotype, male, light trap, Upper Rio Marauia, Estado Amazonas, Missionstation Sao Antonio, Brazil, 9/1/1963, E.J. Fitt- kau, in the collection of Zoologisches Staatssammlung, Munich, West Germany (ZSM, number on slide A 473). Paratypes, male, as holotype, drift, undisturbed forest, Igarapé Barro Branco, Reserva Duke, near Manaus, Estado Amazonas, Brazil, mature male pupa, 11 pupal exuvia 8-10/5/1961, E.J. Fittkau, A 174-1; 1 exuvium, as above except 2/2/1961, A 116-1 (ZSM, ZMBN). Diagnostic characters: See key on p. 286. Etymology: Named in honour of Dr. E.J. Fittkau, Zoologisches Staats- sammlung, Munich, the collector of two of the new species described here. Male imago (n=2 except when otherwise stated) Total length about 1.3 mm (1). Wing length 0.58-0.60 mm. Total length/ wing length about 2.2 (1). Wing length/length of profemur 3.33-3.44. Coloration yellowish brown. Head (Fig. 2A). Antenna with 11 flagellomeres, ultimate flagellomere 131 pm (1) long, AR 1.09 (1). Clypeus with 2 setae. Cibarial pump, tentorium and stipes as in Fig. 2A. Tentorium 56 pm (1) long, 7 um (1) wide. Palp lengths/(micrometers, n=1): 8, 9, 17, 32, 45. Thorax (Fig. 2B). Antepronotum with 1 lateral seta. Dorsocentrals apparently 3-4, prealars 2. Scutellum with 2 setae. Wing (Fig. 2C). VR 1.51-1.56. Legs. Spur of front tibia 19 ym long, spurs of middle tibia 15m and absent, of hind tibia 26m and 8-9 ym. Width at apex of front tibia 17-18 ym, of middle tibia 15 ym, of hind tibia 20 pm. Comb with 7-8 setae, 11-17 wm long. Lengths (micrometers) and proportions of legs: MEM. AMER. ENT. SOC., 34 288 NEW SPECIES OF LOPESCLADIUS Fic. 2. Lopescladius spp., male. — A-D. L. fittkauin. sp. — A. Eye, cibarial pump, ten- torium and stipes. — B. Thorax. — C. Wing. — D. Hypoygium — E. L. verruculosus, hypopygium. O. A. SAETHER 289 fe ti ta, ta, ta; ta, ta; Pp: 173 188-203 75-79 41 30 11 23 P2 225-236 236-249 83-84 34 23 11 19-21 D3 176-184 227-240 69-73 32-34 23-26 9-11 19-21 LR BV SV BR Pi 0.39-0.40 4.14-4.32 4.76-4.80 1.7-1.8 P2 0.34-0.35 6.17-6.61 5.50-5.76 2.3 Pp; 0.31-0.34 5.52-5.60 5.79-5.80 1.9-2.0 Hypopygium (Fig. 2D). Tergum IX with 4-6, 5 (3) setae; laterosternite [IX apparently bare. Apodemes not measurable. Gonocoxite 143um (3) long, distance from base to base of gonostylus 79-83ym; inferior volsella tapering, bare, 10-13ym long, 4-5ym wide at base, 1.5m side one third from base. Gonostylus 38-41, 39um (3) long. HR 3.45-3.80, 3.68 (3); HV about 3.2. Pupa (n=10 except when otherwise stated) Total length 1.51-1.86, 1.65 mm. Thorax of exuvia brown, abdomen pale greyish brown. Cephalothorax (Fig. 3A-C). Frontal apotome (Fig. 3A), antennal sheath and wing sheath smooth. Postorbital 19-23, 22um (3) long. Median antepronotal setae 41-49, 43um (8) and 30-49, 37um (8) long; lateral antepronotals 38-41, 38um (9) and 4-11, 6um (8) long. Posterior precorneal seta (Fig. 3A) 38-45, 40um (8) long; median and anterior both 26-38, 32um long. Anterior two dorsocentrals (Dc,, Dc.) both 26-45, 391m (9) long; Dc; 8-26, 15um (8) long; Dc, 15-38, 26um (8) long. Distance between Dc, and De, 11-21, 15um; between Dc, and Dc; 88-116, 100um; between De; and Dc, 9-34, 22um. Abdomen (Fig. 3D-G). Shagrenation and chaetotaxy as in Figs. 3D-G and generic diagnosis. Number of caudal spines on tergites I-VIII as: 0-2, 0; 9-15, 12; 10-14, 11; 9-13, 11; 9-12, 11; 8-13, 10; 8-14, 10; 6-10, 8. Number of caudal spines on sternites II- VIII as: 4-7, 5; 8-11, 9; 7-11, 9; 7-11, 9; 7-10, 9; 6-10, 8; 0 (6 9 9), 5-9, 8 (4 o@). Digitiform projection of anal lobe smooth, about 2.8 times as long as wide, 86-101, 94um long. Anal macrosetae 83-101, 92m long. Larva. Unknown. Lopescladius verruculosus n. so. Type locality: Mexico, Prov. Michoacan, river mouth about 100 km south of Tocuman. Type material: Holotype, male, river mouth about 100 km south of Tocuman, prov. Michocan, Mexico, April 1981, E.J. Fittkau, in the collection of Zoologisches Staatssammlung, Munich, West Germany (ZSM). Paratypes, 60 pupal exuvia, as holotype. Other material: pupa, Mississippi River, near Cordoba, Illinois, 13/9/72, D.L. Andersen (ZSM, ZMBN). Diagnostic characters: See key on p. 286. MEM. AMER. ENT. SOC., 34 290 NEW SPECIES OF LOPESCLADIUS ae O. A. SAETHER 291 Etymology: From Latin verruculosus, full of small warts, referring to the rugulosity which extends onto the frontal apotome and the antennal and wing sheaths of the pupa. Male imago (n=1) Total length about 1.4 mm. Wing length about 0.6 mm. Total length/wing length about 2.2 mm. Wing length/length of profemur about 3.3 mm. Coloration yellow brown. Head. Antenna with 13 flagellomeres, ultimate flagellomere 128m long, AR 0.74. Palpal segments 4 and 5 respectively 364m and 60um long. Other details of head not measurable. Thorax. Antepronotum apparently bare. Dorsocentrals apparently 2-3, prealars 2. Scutellum with 2 setae. Wing. VR not measurable. Legs. Spur of front tibia 191m long, spurs of middle tibia 15um and 7um long, of hind tibia 21pm and perhaps absent. Width at apex of front and middle tibia 194m, of hind leg 23m. Comb with 9 setae, 11-17~m long. Lengths (micrometers) and proportions of legs: fe ti ta, ta, ta; ta, tas LR BV SV BR i 191 221 86 49 39 15 26 0.39 3.86 4.78 2.0 oF 248 244 90 43 34 15 24 0.37 5.00 5.46 25) Ps 210 244 71 34 26 15 24 0.29 5.28 6.37 2.0 Hypopygium (Fig. 2E). Tergum IX with 6 setae; laterosternite IX apparently bare. Apodemes not measurable. Gonocoxite 169m long, distance from base to base of gonostylus 79um; in- ferior volsella with median seta, 13um long, 2.5um wide and parallel-sided in basal half; 1.54m wide about 0.6 from base and tapering to point, projection of gonocoxite with conspicuous spine-like apical seta. Gonostylus 60um long, HR 2.81, HV about 2.4. Pupa (n= 10) Total length 1.56-1.94, 1.77 mm. Exuvia pale yellowish brown. Cephalothorax Frontal apotome (Fig. 4A), antennal sheath and wing sheath covered with rugulosity. Postorbital 11-21, 17pm long. Median antepronotal setae 26-38, 324m and 23-34, 26um (8) long; lateral antepronotals 19-38, 294m and 4-8, 6um long. Anterior precorneal seta 15-34, 234m long; median and posterior both 11-19, 14um long. Anterior two dorsocentrals (Dc,, De2) and posterior dorsocentral (Dc,) all 15-16, 20um long, Dc; 11-19, 144m long. Distance between De, and Dec, 11-23, 16m; between Dc, and De; 90-124, 106um; between Dc; and De, 6-38, 14um. Abdomen (Fig. 4C-D). Shagrenation and chaetotaxy as in Fig. 4C-D and generic diagnosis. Number of caudal spines on tergites II-VIII as: 8-16, 12; 9-16, 12; 9-12, 11; 10-13, 11; 9-11, 10; 6-10, 8; 2-9, 7. Number of caudal spines on sternites II-VIII as: 4-9, 6; 8-13, 10; 8-12, 10: 7-12, 9; 6-10, 8; 6-8, 7;0(9 9), 4-7, 5(o @). Digitiform projection of anal lobe smooth, about 2.8 times as long as basally wide, 90-113, 102m long. Anal macrosetae 90-116, 103m long. Larva. Not known with certainty. Fic. 3. Lopescladius fittkaui n. sp., pupa. — A. Frontal apotome. — B. Chaetotaxy of cephalothorax. — C. Leg sheath arrangement. — D. Tergites. — E. Sternites. — F. Reticula- tion and spinules of tergites. — G. Polygones of conjunctives. MEM. AMER. ENT. SOC., 34 292 NEW SPECIES OF LOPESCLADIUS —— O. A. SAETHER 293 Lopescladius inermis n. sp. Type locality: U.S.A., Kansas, Meade Co., near Meade Co. State Lake. Type material: Holotype, male, artesian spring at Meade Co. State Lake, Meade Co., Kansas, U.S.A., 16/7/80, P. Liechti & M. Orbois, in the collection of Museum of Zoology, University of Bergen (ZMBN No. 73). Diagnostic characters: See key on p. 286. Etymology: From Latin inermis meaning unarmed referring to the lack of a spine-like inferior volsella. Male imago Total length about 1.59 mm. Wing length 0.73 mm. Total length/ wing length 2.18. Wing length/length of profemur 3.24. Coloration yellowish brown with darker vittae, postnotum and lower parts of preepisternum. Head. Antenna with 13 flagellomeres, ultimate flagellomere 137u4m long, AR 0.51. Tem- poral setae absent except for 1 weak postorbital. Clypeus with 2 setae. Cibarial pump, ten- torium and stipes as in Fig. SA. Tentorium 71um long, 7pm wide. Stipes 774m long. Palp lengths (micrometers): 11, 15, 21, 30, 58. Thorax (Fig. 5B). Antepronotum with 1 lateral seta. Dorsocentrals 4, prealars 2. Scutellum with 2 setae. Wing (Fig. 5C). VR 1.42. Brachiolum with 1 seta, other veins bare. Legs. Spurs of front tibia 211m long, spurs of middle tibia 17um and absent, of hind tibia 41pm and 15um. Width at apex of front tibia 21ym, of middle tibia 194m, of hind tibia 26um. Comb with 11 setae, 9-17um long. Lengths (micrometers) and proportions of legs: fe ti ta, ta, ta; ta, ta; LR BV SV BR Pi ANS) 255 90 51 41 13 28 0.35 4.28 5.33 Dp) P2 244 313 90 41 32 11 26 0.29 5.85 6.19 3.0 Ps 233 302 86 38 30 11 28 0.29 5.81 6.20 2.5 Hypopygium (Fig. 5C). Tergum IX with 7 setae, laterosternite IX apparently bare. Phalla- podeme strongly developed, 994m long, with large aedeagal lobe. Transverse sternapodeme broadly rounded without oral projections. Gonocoxite 180um long, distance from base to base of gonostylus 116um; inferior volsella absent, perhaps represented by a weak seta. Gonostylus 57ym long. HR 3.16, HV 2.79. Remarks. The species appear intermediate between Lopescladius minutissimus and L. fittkaui. The hypopygium is lacking an inferior volsella as apparently also L. minutissimus does, while the hypopygium otherwise is very similar to that of L. fittkaui. Also while the antenna has 13 Fic. 4. Lopescladius spp., pupae. — A. Lopescladius verruculosus, frontal apotome and antennal sheath base. — B. Lopescladius sp. E, frontal apotome, C-D. Lopescladius ver- ruculosus. — C. Abdomen. — D. Lateral view of segment IV. — E-F. Lopescladius sp. E. — E. Lateral view of segment IV. — F. Tergites VIII-LX and anal lobe. MEM. AMER. ENT. SOC., 34 NEW SPECIES OF LOPESCLADIUS 294 , cibarial pump, tentorium and stipes. Lopescladius inermis n. sp., male. — A. Eye Fic. 5. — B. Thorax. — C. Wing. — D. Hypopygium. O. A. SAETHER 295 flagellomeres the AR is lower than in L. minutissimus and L. verruculosus. Unfortunately L. minutissimus could not be reexamined. However, judging on the figures the hypopygium appears similar to L. verruculosus and these two species combined could well form the sister group of L. inermis and L. fittkaui combined. Lopescladius sp. A (Fig. 6A) These exuvia have a pale yellowish coloration, 3-6 spines on tergite I (Fig. SA) and a size of 1.7-2.0 mm. All other details appear identical to those of L. fittkaui n. sp. Material examined: 29 pupal exuvia, left tributary to River Huallago at Huanoco, 1 900 m.a.s.l., Peru, 1963, E.J. Fittkau; 18 pupal exuvia, tributary Rio Chaohamayo at San Ramon, 1 700 m.a.s.l., Peru, 15/5/63, E.J. Fittkau. Lopescladius sp. B (Fig. 6B, C) These exuvia have a very pale greyish yellowish coloration including the cephalothorax, no caudal spines on tergite I, and a length of 1.31-1.56 mm. The anterior shagreen of the abdomen consists mostly of a few anteriolateral spinules only. Except for details connected with the smaller size the exuvia otherwise are very similar to these of L. fittkaui n. sp. and could con- ceivably be a form of that species. Material examined: 94 exuvia, drift, Rio Negro, 2 km below Tapunquara, Brazil, 6/2/63, E.J. Fittkau. Lopescladius sp. C (Fig. 6D) These exuvia have the digitiform anal lobe projection rugulose particularly medially. The frontal apotome also is weakly rugulose medially, while there is some slight rugulosity at base of the antennal sheaths. The size is about 2.0 mm, the digitiform anal lobe projections are 120-135um, about 3.6-4.1 times as long as basally wide and about 1.3-1.4 times as long as the anal macrosetae. There are several caudal spines present on tergite I and the anterior transverse rows of spinules in the group shagreen are nearly complete. Material examined: 4 exuvia, drift, Rio Marauia, Brazil, 28/1/63, E.J. Fittkau. Lopescladius sp. D (Fig. 6E) These exuvia have the digitiform anal lobe projection slightly wrinkled, but not rugulose. The projection is 109-116um long, about 2.9-3.1 times as long as basally wide, and about 1.2-1.3 times as long as the anal macrosetae. The length of the exuvia are 1.81-1.95 mm and the coloration greyish brown. The frontal apotome, antennal and wing sheath are nearly smooth. Tergite I has about 6 caudal spines and the spinules in the anterior group shagreen are placed in a nearly continuous line. The species appear close to Lopescladius sp. C. MEM. AMER. ENT. SOC., 34 296 NEW SPECIES OF LOPESCLADIUS O. A. SAETHER 297 Material examined: 17 exuvia, Panguana, Peru, 3° 37'S, 74° 56’ W, ca. 260 m a.s.l. 3/4/82, E.J. Fittkau. Lopescladius sp. E (Fig. 4B, E-F) These exuvia have a pale yellowish brown coloration, a length of 1.43-1.52 mm; the digitiform projections are smooth, about 75-83ym long, and about 3.4-3.7 times as long as basally wide; and the anal macrosetae are 71-81m long. They resemble L. verruculosus n. sp. in having the frontal apotome, antennal sheath and wing sheath covered with rugulosity. The rugulosity particularly of the frontal apotome (Fig. 4B) is weaker and while the tergites in L. verruculosus have a strong lateral rugulosity extending for most of the length of each segment (Fig. 4D), the lateral rugulosity of Lopescladius sp. E. is restricted to the median half (Fig. 4B). The species could be a form of L. verruculosus. Material examined: 23 exuvia, drift, Rio Tocantus near Baiao, Brazil, 28/10/1960, E.J. Fittkau. Lopescladius sp. F (Fig. 6F) These exuvia have a dark brown coloration, a size of 3.0-3.4 mm, the digitiform projections are distinctly rugulose (Fig. SF) and about 150-160um long, tergite I have a few caudal spines, the frontal apotome is weakly and partly rugulose, the antennal sheaths are smooth, the wing sheath have some rugulosity near apex, and the posterior bands of anterior spinules on the tergites are placed on a continuous transverse line. Material examined: 5 exuvia, drift, middle part of Rio Prete da Eva, Estado Amazonas, Brazil, 14/10/1965, E.J. Fittkau. ACKNOWLEDGEMENTS I am indebted to Drs. F. Reiss and E.J. Fittkau, Zoologisches Staats- sammlung, Munich, Germany, and Dr. L.C. Ferrington, University of Kan- sas, Lawrence, Ka. for material, to Dr. W.P. Coffman, University of Pitts- burgh, Pittsburgh, Pa., for allowing me to examine material of the new subgenus and species, and to my wife Mrs. U. Saether for making the draw- ings and typing the manuscript. Fic. 6. Lopescladius spp., pupae. — A. Lopescladius sp. A, tergites I-II. — B. Lopescladius sp. B, tergites I-IIl. — C-F. Tergites VIII-IX and anal lobe. — C. Lopescladius sp. B. — D. Lopescladius sp. C. — E. Lopescladius sp. D. — F. Lopescladius sp. F. MEM. AMER. ENT. SOC., 34 298 NEW SPECIES OF LOPESCLADIUS REFERENCES BRUNDIN, L. 1956. Zur Systematik der Orthocladiinae (Dipt., Chironomidae). — Rep. Inst. Freshwat. Res. Drottningholm 37:5-185. 1966. Transantarctic relationships and their significance, as evidenced by chironomid midges. With a monograph of the subfamilies Podonominae and Aphrote- niinae and austral Heptagyiae. — K. Sevenska VetenskAkad. Handl. 11:1-472. COFFMAN, W.P. AND S.S. ROBACK. [in preparation] CRANSTON, P.S., D.R. OLIVER AND O.A. SAETHER. 1982. Orthocladiinae. In Wiederholm, T. (ed.): Chironomidae of the Holarctic region. Part 1. Larvae. — Ent. scand. Suppl. CRANSTON P.S. AND O.A. SAETHER. A revision of European Rheosmittia Brundin [in MS]. HALvorsENn, G.A. 1982. Saetheriella amplicristata gen. n., sp. n., a new Orthocladiinae (Diptera: Chironomidae) from Tennessee. Aquatic Insects. 4:131-136. Ropack, S.S. 1953. Savannah River tendipedid larva (Diptera: Tendipedidae — Chiro- nomidae). — Proc. Acad. Nat. Sci. Philad. 105:91-132. SAETHER, O.A. 1979. Underlying synapomorphies and anagenetic analysis. — Zool. Ser. 8:305-312. 1980. Glossary of chironomid morphology terminology (Diptera: Chironomidae). — Ent. scand. Suppl. 14:1-51. 1981. Orthocladinae (Chironomidae: Diptera) from British West Indies, with descriptions of Antillocladius n. gen., Lipurometriocnemus n. gen., Compterosmit- tian. gen., and Diplosmittia n. gen. — Ent. scand. Suppl. 16:1-46. 1982. Orthocladinae (Diptera: Chironomidae) from SE U.S.A., with descriptions of Plhudsonia, Unniella, Platysmittia n. genera and Atelopodella n. subgen. — Ent. scand. 13:465-510. SAETHER, O.A. AND G.A. HALVORSEN. 1981. Diagnoses of Tvetenia Kieff., emend., Drat- nalia n. gen., and Eukiefferiella Thien., emend., with a phylogeny of the Cardiocladius group (Diptera: Chironomidae). — Ent. scand. Suppl. 15:269-285. SAETHER O.A. AND J.E. SUBLETTE. 1983. A review of the genera Doithrix n. gen. Geor- thocladius Strenzke, Parachaetocladius Wiilker, and Pseudorthocladius Goetghebuer (Diptera: Chironomidae, Orthocladiinae). — Ent. scand. Suppl. 20 [in press]. SUBLETTE, J.E. AND W.W. WirtTH. 1980. The Chironomidae and Ceratopogenidae (Diptera) of New Zealand’s subantarctic islands. — N.Z.Y. Zool. 7:299-378. ee eee Wing Character Variation in the Nearctic species of Orthocladius (Orthocladius) van der Wulp (Diptera: Chironomidae): a Principal Components Analysis H.M. SAVAGE AND A.R. SOPONIS Department of Entomology Florida A & M University Tallahassee, Florida 32307, USA ABSTRACT. — Variation in eight characters, wing length, wing width, distance from arculus to crossvein rm, distance from arculus to fork of the cubitus, the numbers of radial and squamal setae, thorax length, and abdomen length, was investigated in males of 22 species of Or- thocladius (Orthocladius) using principal components analysis. The first 3 principal com- ponents (PC) accounted for 96-98 percent of the observed variation: PC1, a general size factor, accounted for 77-83 percent; PC2, which largely corresponds to variation in radial setae number, accounted for 10-15 percent; and PC3, which contrasts squamal setae number with wing size, accounted for 3-5 percent. Within males of O. (Orthocladius), the number of radial setae and thorax length display negative allometry, while the number of squamal setae displays positive allometry with respect to size. This analysis suggests that wing length may be used as a reliable index of size, and that the numbers of radial and squamal setae may prove useful for separating species or species groups within genera of the Orthocladiinae. INTRODUCTION The Nearctic species of the subgenus Orthocladius (Orthocladius) van der Wulp were recently revised by Soponis (1977). Species recognition within Orthocladius and many other chironomid genera is based primarily on the genitalic structure of adult males, e.g., Townes 1945, Brundin 1956. However, descriptions of chironomids are among the most quantitative of all zoological descriptions and generally include detailed data on various lengths and counts, e.g., Oliver 1977, Saether 1977. Even though selected quantitative characters have been used to separate species (Saether 1976), much of the available data has gone unanalyzed. The objective of this study was to evaluate wing character variation in males of O. (Orthocladius) through the use of principal components analysis (PCA). Thorax and abdomen lengths were also included in the analysis to facilitate the evaluation of size effects. Twenty-two of the 29 Nearctic species recognized by Soponis (1977) were represented by sample sizes sufficiently large to allow analysis. 299 MEM. AMER. ENT. SOC., 34 300 WING CHARACTER VARIATION IN NEARCTIC ORTHOCLADIUS Fic. 1. Wing (a), thorax (b), and abdomen (c) of male of Orthocladius showing measurements and setae with character abbreviations: WL - wing length; ARM - distance from arculus to crossvein rm; AFCU - distance from arculus to fork of the cubitus; WW - wing width; R - number of setae on vein R; SQUM - number of setae on squama; TH - thorax length; ABD - abdomen length. H.M. SAVAGE AND A.R. SOPONIS 301 METHODS Definitions and abbreviations for characters treated herein are provided in the legend of Fig. 1. Species studied are numbered and alphabetically listed in the legend for Fig. 2: the corresponding numbers from Fig. 2 are employed in lieu of names in all ordinations (Figs. 2-4). One to 3 sets of character values were selected for each species. For each species, 1 set consisted of the mean values for all specimens studied by Soponis (1977). Additionally, for each species the largest and smallest specimens for which reliable data on all 8 characters were available were in- cluded. The mean value and sample size for each species are provided by Soponis (1977), and data on the largest and smallest specimens are available from the authors. Complete data sets for large specimens of O. dentifer Brundin and O. robacki Soponis, and small and large specimens of O. trigonolabis Edwards were unavailable. PCA on the correlation matrix of standardized characters (Table 1) and variance-covariance matrix of the log transformed data (Table 2), followed by ordination on the first 3 principal components from the correlation matrix were employed to reduce the dimensionality of the data and to ex- amine relationships among selected body length and wing characters. For n characters, PCA produces n linear combinations of the original characters, or n principal components (PC), such that the first PC is the longest axis of the concentration ellipsoid of the multivariate normal density function in the n-dimensional character space for the taxa being investigated. Thus, PC1 is the linear combination of characters that explains the largest portion of the variation in the data. The second PC is defined to be that linear com- bination of the original characters that explains the second largest amount of variation, subject to the restraint that PC2 be uncorrelated, or or- thogonal to PC1. Subsequent PCs are defined in a similar fashion. Each PC is characterized by 1 eigenvalue and n coefficients, or 1 coefficient for each character (Pimentel 1979). The relative size of an eigenvalue corresponds to the percent of the total variance explained by that PC (Tables 1-2). Each coefficient represents the direction of variance for 1 character on the associated PC. Interpretation of PCs is facilitated by the use of a correla- tion coefficient between each character and PC (Tables 1-2). The use of 3 PC scores for each species allows for assessment of species variation and the effects of size within a species (Figs. 2-4). However, the size and position of the triangle for each species is highly dependent upon the selection of the large and small specimens, and provides only a rough approximation of the size and position of the ellipse of species variation. In addition, triangles for species represented by small sample sizes, particular- ly O. knuthi Soponis (11, Figs. 2-4) and O. wiensi Saether (22, Figs. 2-4), provide poor estimates of the true species ellipse. MEM. AMER. ENT. SOC., 34 302 WING CHARACTER VARIATION IN NEARCTIC ORTHOCLADIUS Fic. 2. Males of 22 species of O. (Orthocladius). Ordination by principal components analysis of standardized character correlations, component 1 vs. 2. Dots connect values for 4 large species (4, 15, 16, 17); dashes connect values for 3 small species (1, 3, 10). Species are listed alphabetically and numbered as follows: 1 — annectens Saether; 2 - appersoni Soponis; 3 - carlatus (Roback); 4 - charensis Soponis; 5 - clarkei Soponis; 6 - decoratus (Holmgren); 7 -dentifer Brundin; 8 - dorenus (Roback); 9 - hazenensis Soponis; 10 - hellenthali Soponis; 11 -knuthi Soponis; 12 - lapponicus Goetghebuer; 13 - mallochi Kieffer; 14 - manitobensis Saether; 15 - nigritus Malloch; 16 - obumbratus Johannsen; 17 - oliveri Soponis; 18 - robacki Soponis; 19 - subletti Soponis; 20 - trigonolabis Edwards; 21 - tryoni Soponis; 22 - wiensi Saether. lle 3° 10 13, 14s 1J .22616 Z 2 SS + + + + + + + = + => + = ies *. + + + + + RCL > a 5.0 -4.0 7 -2.0 -1.0 ° $2182 2.0 . 3.0 4.0 5.0 1s S35 ea! e277 s° fe 8 SS 5 S65 SSS 2S get 19 6™2 2e=— Se 72 BB 1 +=====-_ = 1 Fic. 3. Males of 22 species of O. (Orthocladius). Ordination by principal components analysis of standardized character correlations, component 1 vs. 2. Dots connect values for a species with large PC2 scores (21); dashes connect values of 4 species with small PC2 scores (1, 9, 15, 19). Species are numbered as in Fig. 2. H.M. SAVAGE AND A.R. SOPONIS 303 17° Fic. 4. Males of 22 species of O. (Orthocladius). Ordination by principal components analysis of standardized character correlations, component 1 vs. 3. Dots connect values for a species with large PC3 scores (5); dashes connect values for a species with small PC3 scores (6). Species are numbered as in Fig. 2. Conclusions drawn from Anderson’s (1963) test of multivariate isometry and Jolicoeur’s (1963a, 1963b) paired comparison method were verified by regression analyses employing Hotelling’s T* and Bonferroni confidence in- tervals, and by separate PCA on selected data subsets. For example, separate PCA were conducted on the 6 length characters, and on the 4 wing size and vein length characters. Recently, several biostatisticians (Mosimann 1970, Sprent 1972, Hum- phries et al. 1981) have criticized the labelling of PC1, or the first general factor, as a size factor and the labelling of subsequent bipolor components as shape factors. These authors point out that size and shape are not com- pletely dissociated by PCA and that each PC will contain both size and shape components (Humphries et al. 1981). Although this argument is valid, see discussion of PC2 and PC3, labelling PC1 as a general size factor may be beneficial if the vast majority of the variation explained by the PC is directly related to size. Within males of O. (Orthocladius), PC1 and each of the 6 length variables are very highly correlated (r > .94). Thus, labelling PCI as a general size factor facilitates dimension reduction and interpreta- tion of the data; even though, PC1 may contain a small shape component, and subsequent PCs may contain size components. The argument against labelling subsequent bipolor components as shape variables is more ap- propriate as these components are generally more homogeneous mixtures of size and shape. Herein, subsequent components are treated as contrast be- tween particular groups of characters. MEM. AMER. ENT. SOC., 34 304 WING CHARACTER VARIATION IN NEARCTIC ORTHOCLADIUS RESULTS AND DISCUSSION PCA on the correlation matrix of the standardized data (Table 1) and on the variance-covariance matrix of the log transformed data (Table 2) pro- duced similar results. Differences in the percentage of variation explained by each PC for the 2 procedures (Tables 1-2) largely result from the larger variances of the radial and squamal setae characters. However, the correla- tion coefficients between each character and PC (Tables 1-2) are generally of the same magnitude and sign and similar interpretations may be drawn from each procedure. The first 3 PCs account for 96-98 percent of the variation in the 8 characters surveyed. The first PC accounts for the vast majority of the variation in the data, between 77.36 and 82.67 percent (Tables 1-2), and ap- pears to be most parsimoniously interpreted as a general size factor. All characters are highly correlated with PC1 (Tables 1-2) except for radial setae number, which is only weakly correlated with the other length characters and hence size. The 6 length characters, WL, ARM, AFCU, WW, TH, ABD (Fig. 1), are very highly and approximately equally cor- related with PC1 (r > .94), or size. Allometric relationships for each character with respect to size may be assessed by inspection of the coefficients for each character on PCl, the general size component, from the log transformed data (Table 2, top). Generalization of the allometry equation to the multivariate case (Anderson 1963, Jolicoeur 1963a, 1963b) allows for a test of the null hypothesis of multivariate isometry, or that all characters increase at constant rates with respect to size. For males of 0. (Orthocladius) the multivariate isometry hypothesis may be rejected (X? = 101, n = 62, P <_ .005). It is apparent from the coefficients in Table 2 that all characters do not increase at a cons- tant rate with increasing size. Characters with coefficients for PC1 roughly equal to .3535, or 1/./n, display isometry, or increase at a constant rate with respect to size. Characters such as SQUM with coefficients larger than .3535 display positive allometry, or increase at increasing rates with increasing size. Characters such as R and TH with coefficients less than .3535 display negative allometry, or increase at decreasing rates with increasing size. Allometric relationships between character pairs may also be investigated by use of their PC1 coefficients (Jolicoeur 1963), although no significance test is available (Pimental 1979). Thus, SQUM displays positive allometry with respect to each of the other characters as its coefficient on PC1 is greater than that for each of the other characters, while R and TH display negative allometry with respect to the remaining wing characters and ab- domen length. Regression analyses employing Hotelling’s T? and Bonfer- roni confidence intervals, and separate PCA on various data subsets in- 305 SOPONIS SAVAGE AND A.R. H.M. OE - Le0"- pse + L7O + COT 1 COC - 67 - €Od 9P0°- 0L0°- vrl- 6S8° + 6eI - €OI- te}e= 980°- cod €96- €v6- St6 - 60S "- 1S6- 96°- C96 - 196°- TOd ddv HL WnOs ud MA NOdAV Wav IM €Od PUR ZOd ‘1Od PUL JO}OBIBYO YORa UsaMJaq S}UDIOIJJooo UOTe[IIIOD 90'S pst - 1v0'- 908° + SO + vIT- €87- £67 - Sie - £Od I@ SI 8£0°- Sv0'- 061 - p96 + SOT - 6L0°- $60°- 890°- tod SELL ore - Guo 6ES - €Sc- Ove - 9TE- 8Se- SEE - lod paureldxe dav HL WNOS u NM NOdAV Wav IM OUBLIVA €Od PUR ZOd ‘1Od Pur JOJOBIVYO YORA UdaMIJEq SJUDIOIJJaOD ‘(snippjr0Yyj4{O) *O JO sayeul 10J eJep pawtIOJsueI] BO] JO XIYeUI DOULTIBAOD-JdURTIVA UO WOd “T ITAVL Of € $00°- 960° + trp + 970°- 880°- 14 Lel- 6€1- tOd COL c10- 620°- veO- S88" + col - LLO’- 140) is 890°- cod L978 196°- 976 - 688°- A) a 796 °- 786 - 7L6- 086°- IOd poureldxe ddv HL WnOs u NA NOdAV Wav TM DOULTIVA Of €Od PUL ZOd ‘TOd pure Jo}OBIvYO YORE UdaMJaq SJUSIOIJJIOO UOTRIaIIOD *(SnIppj20YI4IOC) °C JO So[eUl 10} XIIeUI UONLIAIIOD pezipiepuejs 94} UO WOd “| ATV L MEM. AMER. ENT. SOC., 34 306 WING CHARACTER VARIATION IN NEARCTIC ORTHOCLADIUS dicate that ARM displays positive allometry with respect to AFCU, while all other paired comparisons among WL, ARM, AFCU, WW and ABD appear isometric. For taxonomic purposes it is often convenient to select 1 character to represent size for species within homogeneous genera. The representative size character, such as wing length in O. (Orthocladius), should be: 1) iso- metric with respect to PC1, or the general size component; 2) highly cor- related with PC1 and other length characters; 3) reliably measured; and 4) commonly used and reported in the literature. The negative coefficients of PC1 (Tables 1-2) result in large component scores for small species (Figs. 2-4) such as O. annectens Saether (1), O. carlatus (Roback) (3), and O. hellenthali Soponis (10), and small (negative) scores for large species, such as O. charensis Soponis (4), O. nigritus Malloch (15), O. obumbratus Johannsen (16), and O. oliveri Soponis (17). While the 3 small species and 4 large species may be separated based solely upon size (Fig. 2), values for large and medium-sized species widely overlap due in part to the larger variances associated with increased size. The 3 small species, particularly O. annectens (1), appear to be distinctly smaller than many medium-sized species; however, as additional specimens of both small and medium-sized species become available these gaps will likely close. PC2, which accounts for 10.22 to 15.21 percent of the variation (Tables 1-2), contrasts the number of radial setae with all remaining characters. Correlation coefficients between PC2 and characters other than R are small (Table 1) so that PC2 may be loosely referred to as the radial setae compo- nent. Species with large numbers of radial setae, such as O. tryoni Soponis (21, Fig. 3), will have large PC2 component scores, while species with small numbers of radial setae for their size, such as O. annectens (1), O. hazenen- sis Soponis (9), O. nigritus (15), and O. subletti Soponis (19), will have small component scores (Fig. 3). However, size plays a significant role in determining PC2 scores. In species with relatively constant radial setae counts, or small variances for R, such as O. clarkei Soponis (5), O. obum- bratus (16), and O. subletti (19), large specimens will display slightly decreased PC2 scores due to the subtraction of increased values for the re- maining characters, even though these characters are represented by very small coefficients (Table 1, Fig. 3). PC2 separates O. tryoni (21) from several species with small component scores (Fig. 3) and suggests that O. tryoni could be separated from O. an- nectens (1), 0. hazenensis (9), 0. nigritus (15), and O. subletti (19) by R counts or a reduced PC2 variable such as 9(R) — ARM — WW, where the distance from the arculus to crossvein rm (ARM) and wing width (WW) are expressed in micrometers. The reduced PC2 variable correctly assigns all H.M. SAVAGE AND A.R. SOPONIS 307 specimens to species for the O. tryoni versus O. nigritus (n = 29) and O. tryoni versus O. hazenensis (n = 30) comparisons, while correctly separating 87.5 percent of the O. tryoni and O. subletti specimens (n = 24). However, the reduced PC2 variable could correctly separate only 45 percent of the specimens (n = 49) for the O. fryoni versus the small species O. an- nectens comparison. The reduced PC2 variable performs better than any single character such as R counts for large and medium-sized species, but not for comparisons between O. tryoni and smaller species such as O. an- nectens. Ninety-two percent of the specimens (n = 49) for an O. tryoni ver- sus O. annectens comparison may be correctly separated based only upon R counts. Poor separation between O. ¢ryoni and smaller species results from the subtraction of greater ARM and WW values in O. tryoni, which nullifies the differences in R counts between O. fryoni and smaller species. Despite size dependency, reduced PC2 variables may be potentially useful tools for separating particular species, and the use of radial setae counts merits further study. PC3 accounts for 3.3 to 5.1 percent of the variation in the data, and con- trasts the wing size and wing vein length characters, WL, WW, ARM, and AFCU (Fig. 1) with the number of squamal setae (Tables 1-2). Species with large component scores for PC3, such as O. clarkei (5, Fig. 4) have relative- ly small wings and large numbers of squamal setae, while species with small values for PC3, such as O. decoratus (Holmgren), (6, Fig. 4), have relatively large wings and small numbers of squamal setae. Size plays an important role in determining PC3 scores via the 4 wing characters. In species that do not display strong positive allometry for SQUM, such as O. annectens (1), O. carlatus (3), O. clarkei (5), O. decoratus (6), O. dorenus (8), O. hazenen- sis (9), O. obumbratus (16), and O. oliveri (17), larger specimens will have slightly smaller PC3 scores (Fig. 4). Differences in PC3 scores (Fig. 4) between O. clarkei (5) and O. decoratus (6) suggest employment of a reduced PC3 variable to separate these species. The variable SQUM — WL can correctly separate 10 of 11 specimens for which data are available; however, a critical evaluation of this variable’s ability to separate these species must await collection of additional specimens. ACKNOWLEDGMENTS We thank J.H. Epler for comments on an earlier draft of this manuscript. This study was supported by a research grant (FLAX 79009) of CSRS, USDA, to Florida A & M University. MEM. AMER. ENT. SOC., 34 308 WING CHARACTER VARIATION IN NEARCTIC ORTHOCLADIUS LITERATURE CITED ANDERSON, T.W. 1963. Asymptotic theory for principal component analysis. Ann. Math. Stat. 34:122-148. BRUNDIN, L. 1956. Zur Systematik der Orthocladiinae (Dipt. Chironomidae). Rep. Inst. Freshwat. Res. Drottningholm 37:5-185. Humpnries, J.M., F.L. BOOKSTEIN, B. CHERNOFF, G.R. SMITH, R.L. ELDER, AND S.G. Poss. 1981. Multivariate discrimination by shape in relation to size. Syst. Zool. 30:291-308. JoLticoeuR, P. 1963a. The multivariate generalization of the allometry equation. Bio- metrics 19:497-499. 1963b. The degree of generality of robustness in Martes americana. Growth 27:1-27. MosIMANN, J.E. 1970. Size allometry: size and shape variables with characterization of the lognormal and generalized gamma distributions. J. Amer. Stat. Assoc. 65:930-945. Outver, D.R. 1977. Bicinctus-Group of the genus Cricotopus Van der Wulp (Diptera: Chironomidae) in the Nearctic with a description of a new species. J. Fish. Res. Bd. Can. 34:98-104. PIMENTEL, R.A. 1979. Morphometrics the multivariate analysis of biological data. Ken- dall/Hunt, Dubuque. 276 pp. SAETHER, O.A. 1976. Revision of Hydrobaenus, Trissocladius, Zalutschia, Paratrisso- cladius and some related genera (Diptera: Chironomidae). Bull. Fish. Res. Bd. Can. 195:1-287. 1977. Taxonomic studies on Chironomidae: Nanocladius, Pseudochirono- mus, and the Harnischia complex. Bull. Fish. Res. Bd. Can. 196:1-143. Soponis, A.R. 1977. A revision of the Nearctic species of Orthocladius (Orthocladius) van der Wulp (Diptera: Chironomidae). Mem. Entomol. Soc. Can. 102:1-187. SPRENT, P. 1972. The mathematics of size and shape. Biometrics 28:23-27. Townes, H.K., JR. 1945. The Nearctic species of Tendipendini (Diptera, Tendipedidae (= Chironomidae)). Amer. Midl. Natur. 34:1-206. Emergence of Polypedilum (Chironomidae) in a Sand-Bottomed Stream of Northern Florida ANNELLE R. SOPONIS Laboratory of Aquatic Entomology Florida A and M University Tallahassee, Florida 32307 U.S.A. ABSTRACT. — Emergence data are given for 5 species of Polypedilum: aviceps Townes, convic- tum (Walker), fallax (Johannsen), i/linoense (Malloch), and scalaenum (Schrank). Overlapp- ing emergence occurs among all the species in Turkey Creek. During the summer P. aviceps may emerge downstream in Rocky Comfort Creek, a 4th order stream. Turkey Creek is pro- bably a more favorable habitat for P. aviceps than for P. convictum, based on numbers and patterns of emergence. All species have longer emergence periods in Turkey Creek, Florida, than in L’Achigan River, Quebec, which is expected if latitude restricts seasonal emergence. INTRODUCTION Only a few North American studies have dealt with seasonal emergence of chironomids in Jotic habitats at the species level, and all of these (Coff- man 1973, Cloutier and Harper 1978, Harper and Cloutier 1979, Boerger 1981) have been conducted in streams and rivers above 40°N latitude. This paper is concerned with the seasonal emergence of 5 species of Polypedilum in Turkey Creek, Florida (30° N latitude). The emergence patterns and emergence periods of these species are compared with those of a higher latitude in North America. Corbet’s (1964) theory of latitude restriction of seasonal emergence in North America holds for all 5 species: all species have longer emergence periods in Florida. Similarities and differences in emergence patterns are probably due to factors other than latitude, such as favorable or unfavorable habitat. Stupy SITE AND METHODS Turkey Creek is a 3rd order (Leopold et al. 1964), sand-bottomed stream located in Gadsden County, Florida (30° 29’N, 84° 35’ W). At the sampling site the average depth was 15-30 cm, the average width, 4 m. Pupal exuviae were collected in 2 drift nets, one placed 2 m and the other 1 m from the stream bank. These nets (735~m opening) were placed in the stream for 24 hrs biweekly from February 1979 to January 1980. Data from 309 MEM. AMER. ENT. SOC., 34 310 EMERGENCE OF POLYPEDILUM both nets were pooled. Species identifications were based on pupal associa- tions of reared material, using the keys of Maschwitz (1976) for the larva, pupa, and adult. The study site and methods are described in more detail in Soponis (1980). This study is based on 280 specimens. RESULTS AND DISCUSSION Emergence data for the 5 species of Polypedilum are presented in Fig. 1. P. aviceps and P. fallax emerge during the winter months with no emergence from May to September/October. P. i//inoense emerges during a similar period, and, in addition, a single exuvium was collected in August. P. convictum emerges throughout the year except for the winter months of December, January, and February. P. scalaenum emerges in 3 waves throughout the year. Harper and Cloutier (1979) reported on the emergence for the same 5 species in the higher latitude stream, L’Achigan River, Quebec (46°N, 74°W). The emergence of these species is longer in Turkey creek (6 to 9 months) than in L’Achigan River (3 to 4 months). P. convictum emerges from mid-May to mid-August in Quebec, and March to November in Florida. P. scalaenum emerges in June in Quebec, but in June and 7 other months in Florida. P. aviceps and P. fallax emerge May to September in Quebec, and October/November to May in Florida. P. i/linoense emerges from June to August in Quebec, and from November to June, and August in Florida. Shorter periods of emergence of aquatic insects are typical of higher latitudes in North America, probably as a result of temperature (Corbet 1964). Patterns of emergence are strikingly different for P. convictum in the 2 streams. In Turkey Creek P. convictum has an amodal emergence whereas in L’Achigan River it has bimodal to trimodal emergence. The differences in patterns may be due to the numbers of specimens collected which, aside from collecting techniques, may reflect favorable or unfavorable conditions for the species. The low numbers (n= 16) and amodal emergence suggest that Turkey Creek is not a favorable habitat for P. convictum, and that L’Achigan River is. By contrast, the large numbers (n= 186) and distinct pattern of P. aviceps suggest that Turkey Creek and L’Achigan River are favorable habitats for this species. However, this needs to be confirmed with more study. I agree with Harper and Cloutier (1979) that variations in emergence patterns of the same species in different sites and in different years reflect the favorableness of the habitat for the species at that place and time. A. R. SOPONIS 311 Eee eA Medd ASS OXON (Dd aviceps convictum fallax illinoense sScalaenum Fic. 1. Emergence of 5 species of Polypedilum from February 1979 to January 1980. Numbers of pupal exuviae on vertical axis, months on horizontal axis. P. aviceps emerges in Turkey Creek in months (November to May) that are almost complementary to those (May to September) in L’Achigan River. This might suggest that P. aviceps, a lotic species often found in nor- thern waters (Maschwitz 1976, Harper and Cloutier 1979, Boerger 1981) is cold-adapted and has switched its life cycle in Florida to take advantage of the cooler months. However, life history data on Polypedilum in Turkey Creek (Russell and Soponis, unpub.) suggest that P. aviceps is also emerg- ing in the summer because large numbers of 3rd and 4th larval instars are drifting in late May and June. If this is true and adults are emerging, but MEM. AMER. ENT. SOC., 34 312 EMERGENCE OF POLYPEDILUM pupal skins are not recovered in Turkey Creek, then it raises the possibility that P. aviceps is emerging downstream. There is some indirect evidence to support downstream emergence of P. aviceps. lovino and Miner (1970) collected adults of P. aviceps in emergence traps in an Arkansas reservoir in June, July, and September. However, they were unable to find the larvae in spite of sampling the benthos for one year. In the same study they found low level, transient populations of certain species in the reservoir, mostly orthoclads, which they attributed to an in- flux from spring-fed streams. The possibility exists that larvae of P. aviceps drifted from the streams into the reservoir and emerged there. It seems likely that P. aviceps from Turkey Creek could emerge in the summer in Rocky Comfort Creek. In Turkey Creek, larvae of P. aviceps in- habit leaf packs in the current. Turkey Creek joins a 4th order stream, Rocky Comfort Creek, 0.36 km downstream. Rocky Comfort Creek emp- ties into Lake Talquin, which in 5.7 km downstream from the collecting site on Turkey Creek. It’s unlikely that such lotic larvae would survive in the lentic conditions of Lake Talquin. Dendy (1944) showed that many stream dwelling chironomid larvae that drifted into lakes died there. Although Ward and Cummins (1978) found larvae of Paratendipes albimanus (Mg.) in a lake as well as Ist to 3rd order streams, P. albimanus is primarily a len- tic species. Further study is needed to confirm the downstream emergence of P. aviceps. Overlapping emergence occurs in all 5 species in Turkey Creek. In all months except September at least 2 species are emerging at the same time. In March and May all congeners are on the wing. This seems to be true of Polypedilum in 2 other streams. In Linesville Creek, Pennsylvania, Coff- man (1973) recorded the emergence of 5 species of Polypedilum in June, and 4 of these were on the wing from June to August. In L’Achigan River, 16 species of Polypedilum emerged from May to September. However, in Bigoray River, Alberta, Boerger (1981) found no overlap between Polypedilum scalaenum and Polypedilum braseniae (Leathers). Boerger did not report the emergence for 4 less common species of Polypedilum, and I suspect that emergence of some of the species overlap. Since lotic habitats support many species of Polypedilum, complex resource partitioning by the larvae can be expected. Harper and Cloutier (1979) found evidence that P. aviceps dominates the lower part of riffles, and P. convictum dominates the upper part. In Turkey Creek, at least 3 species of Polypedilum partition the leaf pack habitat (Russell and Soponis, unpub.). In summary, species of Polypedilum are found emerging from many lotic habitats in North America, and the same species are found in streams of dif- ferent latitudes. Species do tend to have longer emergence periods at lower A. R. SOPONIS 313 latitudes as Corbet predicts. Overlapping emergence appears to be common within the genus although some species do not overlap in the same habitat. Differences in species patterns of emergence may indicate the favorableness of the species habitat. ACKNOWLEDGMENTS I would like to thank A.E. Gordon, C.L. Russell, and J.H. Epler for their helpful comments. This research was supported by NSF (RIM78-17403) and SEA/CR (FLAX 79009). LITERATURE CITED BoERGER, H. 1981. Species composition, abundance and emergence phenology of midges (Deiptera: Chironomidae) in a brown-water stream of West-Central Alberta, Canada. Hydrobiol. 80: 7-30. CLouTIER, L. AND P.P. HARPER. 1978. Phénologie de Tanypodinae de ruisseaux des Laurentides (Diptera; Chironomidae). Can. J. Zool. 56:1129-1139. COFFMAN, W.P. 1973. Energy flow in a woodland stream ecosystem: II. The taxonomic composition and phenology of the Chironomidae as determined by the collection of pupal exuviae. Arch. Hydrobiol. 71:281-322. CorBET, P.S. 1964. Temporal patterns of emergence in aquatic insects. Can. Entomol. 96:264-279. Denpy, J.S. 1944. The fate of animals in stream drift when carried into lakes. Ecol. Mongr. 14:333-357. Harper, P.P. AND L. CLouTmR. 1979. Chironomini and Pseudochironomini of a Québec highland stream (Diptera: Chironomidae). Entomol. Scand. Suppl. 10:81-94. Iovino, A.J. AND F.D. MINER. 1970. Seasonal abundance and emergence of Chironomidae of Beaver Reservoir, Arkansas (Insecta: Diptera). J. Kansas Entomol. Soc. 43:197-216. LEOPOLD L.B., M.G. WoLMAN, AND J.P. MirieR. 1964. Fluvial processes in geomor- phology. W.H. Freeman and Co.: San Francisco, 522 pp. MaAscuwitz, D.E. 1976. Revision of the nearctic species of the subgenus Polypedilum (Chironomidae: Diptera). Ph.D. thesis, Univ. of Minnesota, 325 pp. Soponis, A.R. 1980. Taxonomic composition of Chironomidae (Diptera) in a sand- bottomed stream of northern Florida, pp. 163-169. IN: Murray, D.A. (Ed.), Chironomidae Ecology, Systematics, Cytology, and Physiology. Pergamon Press: New York. Warp, G.M. AND K.W. Cummins. 1978. Life history and growth patterns of Paratendipes albimanus in a Michigan headwater stream. Ann. Amer. Entomol. Soc. 71:272-284. MEM. AMER. ENT. SOC., 34 Communities of Chironomidae (Diptera) from an acid-stressed headwater stream in the Adirondack Mountains, New York KARL W. SIMPSON Center for Laboratories and Research New York State Department of Health Albany, NY 12201 ABSTRACT. — During the summer of 1980 the benthic invertebrates of Silver Run, a small headwater stream in the Adirondack Mountains of upstate New York, and its tributaries were investigated to document differences between tributaries with different pH. Communities of Chironomidae in strongly acidic first-order streams (pH 4.48-5.20) contained 40-70% fewer taxa than the community in a physically comparable, less acidic stream (pH 5.72-6.71). The acidobiontic (strongly acidic) communities were strongly dominated by one or more of the following taxa: Conchapelopia americana/flavifrons, Eukiefferiella claripennis gr., Cricotopus tremulus gr., and Cricotopus vierriensis. These results suggest that communities of Chironomidae respond to decreased pH in much the same way as do total macroinvertebrate communities: many species are replaced with a few dominant ones. INTRODUCTION The acidification of precipitation and surface waters in eastern North America is well documented (see summary by Haines 1981; also Cogbill 1976, Cogbill and Likens 1974, Likens and Bormann 1974, Likens ef al. 1979). Surface waters at high altitudes, such as the Adirondack and White Mountain regions, are particularly susceptible to acidification. Because their geologic substrates are highly resistant to chemical weathering, there is little buffering of the acid before precipitation enters the surface water systems (Hall ef al. 1980). The present study was part of a project to document differences in the biota of a small stream in the Adirondack Mountains whose tributaries ex- hibit varying degrees of acidity. Information concerning the midge fauna will be presented here. Data concerning the total macroinvertebrate fauna will be presented elsewhere (Simpson and Bode, in preparation). STuDY AREA Silver Run is a small, short headwater stream located approximately 20 km south-southwest of the village of Blue Mountain in Hamilton County, 315 MEM. AMER. ENT. SOC., 34 316 CHIRONOMID COMMUNITIES IN ADIRONDACK STREAM New York (latitude 43°42’, longitude 74°35’) (Fig. 1). It originates at about 850 m elevation, flows west and then southwest, and discharges to the South Branch of the Moose River, which eventually flows via the Black River into Lake Ontario. Silver Run falls 150 m over its course of 6.2 km and receives two main tributaries, Cellar Brook and Bradley Brook, 2.4 and 1.3 km from its mouth, respectively. This system is attractive for study because of the different pH in the physically similar streams. Between April and September 1980, Silver Run upstream of Cellar Brook ranged from pH 5.72 to 6.71. Based on defini- tions established by the US Environmental Protection Agency (Harris and Lawrence 1978, Hubbard and Peters 1978, Surdick and Gaufin 1978), the fauna at this site is acidophilic (occurring at pH 5.5-7.0). During the same period, Cellar Brook and Bradley Brook ranged from pH 4.48 to 5.20; their faunas are classified as acidobiontic (occurring at pH less than 5.5). The different pH regimes in these streams may be related to stream gra- dient. Cellar and Bradley brooks flow through high-gradient mountain valleys with little accumulated soil to buffer the water. Silver Run originates in a lower-gradient area. As would be expected, pH values in Silver Run downstream of the tributaries reflected the combined flow of the three source waters, ranging from 4.82 to 6.22. Samples were collected at four sites: (1) Silver Run upstream of both tributaries, (2) Cellar Brook near its mouth, (3) Bradley Brook near its mouth, and (4) Silver Run downstream of both tributaries (Fig. 1). Stations 1-3 were all on first-order streams and were very similar with regard to flow (Table 1), width (5-6 m), and current speed (10-20 cm/sec). Although sta- tion 4 was considerably wider (18 m) and had higher flows, samples could be collected from areas with similar depth and current speed. The substrate at stations 1 and 2 consisted of rubble 3-20 cm in diameter with sand and gravel underneath. The substrate at stations 3 and 4 was similar but also contained numerous larger rocks, some several meters in diameter. METHODS Macroinvertebrates were sampled quantitatively with a Surber square- foot sampler (net mesh size = 1.05 mm). All rocks within the square frame were removed and thoroughly cleaned, and the underlying sand and gravel were agitated to a depth of about 10 cm. Three samples were taken at each site on 2 July 1980 and three on 9 September 1980. The samples were K. W. SIMPSON 317 Bradley © Mtn NEW YORK Little Moose Lake Fic. 1. Map of Silver Run and tributaries, showing locations of sampling sites. preserved in 70% ethanol with rose bengal stain (Mason and Yevich 1967) and later picked under a dissecting microscope. Midges were cleared in warm KOH (10%), rinsed successively in water and ethanol, and mounted in Euparal. Communities of midges were characterized by taxonomic richness and the number of individuals per sample. Diversity indices were not used due to the small sample size (Weber 1973). Even when all three samples were pool- ed, fewer than 100 individuals were obtained at any site. Pooled samples were used to assess changes in taxonomic richness; however, means of the values for individual samples showed the same trends. To evaluate trends in the occurrence of individual taxa, only those occur- ring in two or more samples at any one site were used. These are termed common taxa. RESULTS The taxa found at each sampling site are listed in Table 2. At station 1, the Orthocladiinae were dominant in terms of both the number of taxa and the number of individuals contributed to the total midge fauna (Fig. 2). The Tanytarsini were the second most abundant, followed by the Tanypodinae and Chironomini. MEM. AMER. ENT. SOC., 34 CHIRONOMID COMMUNITIES IN ADIRONDACK STREAM 318 ce L@ (63 06'T £70 68°0 81 9 II C9 (6:4 crs uny JeAflg eMOT—P cv (G3 che vL'0 80°0 07 0 LI G 6 C61 sry so'V yooig Ae[peig—f€ (G4 L@ ce 890 90°0 Os a0) 8I € 6 OGES OS 09'P yooig Ie[[s9>—Z (43 LZ oe 1¢'0 90°0 LYV0 cl 6 8 IL9 go's st9 uny Joaqig 1eddq—t Xe, CUI Ss«LRIPOTN Xe CUTJNSsCUR IPO XE CUI —sCURIPOIN xe COUYSCs«CURIPO uoe007T / UONFIS (s7) (S/W) (90) Hd AWATIONPUOD MOT ainjeloduis |, “(0861 Jequieides 67 pue flidy € UesMjoq Use} s]USUIOINSeeU 6 UO peseq) seliejnqi pue unYy IOATIS Jo siojoutesed Jeommoys-oo1skyd aulog “] ATaVL K. W. SIMPSON 319 The midge fauna at station 1 was not strongly dominated by any par- ticular taxon. The two most abundant taxa collectively comprised 27% of the total individuals in July and 40% in September. Numerous rare organisms were also found at this site: 11 taxa were represented by only one individual in the six pooled Surber samples (July plus September). Eleven taxa occurred in two or more samples at station 1 (Table 3) and appeared to be important components of the community. In July, three of these common taxa were Tanytarsini, two were Orthocladiinae, and one was Tanypodinae. In September all six common taxa were Orthocladiinae. Only Rheocricotopus sp. was commonly found during both sampling periods. At stations 2 and 3, the structure of the midge community was markedly different than at station 1. Many fewer taxa (40-70%) were collected at each site, and the two most abundant taxa contributed a much greater propor- tion to the total number of individuals (as much as 90%). Although the number of orthoclad taxa was 50% lower, this group still contributed the most taxa and individuals to the total community. The contribution of Tanytarsini declined from 30% of the total individuals at station 1 to an average of 3% at stations 2 and 3. Only five taxa were common in the acidobiontic communities at stations 2 and 3 (Table 4): one Tanypodinae and four Orthocladiinae. Con- chapelopia americana/flavifrons' was common at one or both sites during both months and was dominant at station 3 in September. Eukiefferiella claripennis group was common at both sites in July and comprised 75% of the individuals at station 3. Cricotopus tremulus group and Cricotopus vier- riensis Goetgh. occurred during both months but were common and more abundant in September. Station 4 yielded the same number of taxa as station 3 but substantially fewer individuals than any of the other sampling sites. An average of only 7 individuals per Surber sample was obtained (= 75 individuals/m? of streambed). Eukiefferiella claripennis group was the only common taxon, and it was common during both sampling periods. No Tanytarsini were col- lected at this site. A comparison of the fauna found at station 1 versus those found at sta- tions 2-4 demonstrates the paucity of taxa in the more strongly acidified ‘Conchapelopia americana Fittkau and C. flavifrons (Joh.) occurred concurrently in our samples. Species identifications were made on mature (4th instar) larvae using Roback’s (1981) key characters of size and were verified with pupal characters visible in prepupal specimens. Most samples contained numerous early instars that could not be identified with certainty, hence the designation Conchapelopia americana/flavifrons. MEM. AMER. ENT. SOC., 34 CHIRONOMID COMMUNITIES IN ADIRONDACK STREAM 320 “18 DIUDADG Diuajad puax “IU DjjaluuDWaualY SuasAluUas *IU SniIppjzoyjoUds ‘ds sndojoo1io0ayy (YOL) 1y2eqpunN] snueuds0laUDIDg 7 dS ppjaiaffarynivg ‘ds sndojoo1o0saypw ‘ds pjjiusouaLy Ioyjees (xadpji1y snippjo0ssisjosajapzy ‘13 po1UOAap Dijalaffalynyz “Id SIuUadLIDjD Djjalsaffalyng "18 1mYyaig djjalsaffaryng ‘ds snippja0yjiQ /sndojooup 4 ‘ds (snippja0sy) sndojooupD *Y3}90DH sisuaiiiaia sndojoIuUgD "13 snjniuad sndojoouD snjouiz0anjf “iu sndojoouga (uaSIEJ\) SNjoUIIIq sndojoIuUD deULIPe[OOUWO “ds IAW asAvZ 5 0 O e e e e I (oye MA) Snioluquif sndduvnjojin. *«8E (6 e +8 *6 «9 (Yor) suodfianyf /NexWY VUDIIaWUD DIdojadpyou0D e e e e e e e S (AaTTe MA) 1y90)J0U viduseqniqy seulpoddAuey, ee- e — * * * eoeereeencece3oee#ee#eeeee eeee-reeees35§qoese354«e04ene e * eoen-r-eeeEegenreemeee#eeeee e er-wrtrrmee0euemreee#ee#eeeentwee e + * * ennwoeemnmewoee#esg#geoeus05o#54eernrn e@ — om eon emee ee ome nne-n eemeeeermneeeee e e _ e e — N e mn v € C I v € C I uole1S Jaquiaidas 6 Ane Z (0861) 21ed ‘(e]dures | ul pormnd90 sioyjo je ‘saydures ¢ ul pormM000 = x, ‘sojdures Z Ul palimo00 = ,) solieingiy pue uny JoATIg ‘sojdwies Jaqing 2014} Ul sepIWoUoIIYO Jo saouepuNnqe poajoog ‘7 aTaV I 321 SIMPSON K. W. eoewee eeeoeeee eneee —™e@e@e¢.e@ eenree eee-r-eee Or 8€ —e@e@e@ e@ @ (ZG 6S Ce siejsul AjIeay# exe} JO JOquUINN S[ENPIAIPUI JO Joquinyy poulwMisjapuy # “18 Snjions snsivj]dudy, *pury iuipuniqg snsipjcuv ‘ds pjjauijjadwuajs “13 SNNBIXA SNSADJAUDJOBYY (WOOT) ¢vIj0d vijzasdoss1py TUISIeAUP | (YooTTeW) asuaouny: winjipaddjod (OAR) WnjoIAUuod winjipaddjog SOUMO YL, SdadIAD Wnjipaddjog ‘ds sadipuajojdaj5 (YOO]JeW) Snjsapowuoau sadipuajoiiqg TUIWIOUOIIYD sPUTWIOUOIIYD 322 CHIRONOMID COMMUNITIES IN ADIRONDACK STREAM streams. The six Surber samples from station 1 yielded 30 midge taxa, in- cluding all those listed in Table 2 except Zavrelimyia sp., Hetero- trissocladius hirtapex? Saether, and Glyptotendipes sp. In contrast, the 18 samples from stations 2-4 collectively contained only 16 taxa. If mere presence is used to determine a taxon’s environmental requirements, these 16 taxa could be classified as acidobiontic. DISCUSSION The number of midge taxa (33) found during the course of this study is relatively low, considering that intensive studies of other small streams have yielded over 100 species (Coffman 1973, Ringe 1974, Boerger 1981). A com- plete species list for the system would doubtless be considerably longer, especially if drift and emergence samplers were utilized. The number of in- dividuals collected is also very meager, considering that midge densities in small streams (first- to third-order) often range from several hundred to several thousand individuals per m? (e.g., Newbold ef al. 1980, Pinder 1980). These sparce results may reflect ineffective sampling to some degree, but they are also due in part to the oligotrophic nature of high-altitude mountain streams (Pennak 1977). Members of the subfamily Orthocladiinae are well known for dominating the midge fauna in small streams (Brennan ef a/. 1981, Coffman 1973, Drake 1982, Pinder 1980, Thienemann 1954). Thus, their predominance in the Silver Run system is not surprising. However, the absence of Diamesinae was unexpected. The physical traits of the stream seem well suited for them, and some species are reported to be acid-tolerant (Curry 1965, Scullion and Edwards 1980). Some resident species may have been missed because there was no winter sampling, when many diamesin species are most active. Midges are important in the macroinvertebrate communities of acidified streams, although they often are identified only to the family level (Arnold et al. 1981, Sutcliffe and Carrick 1973, Pratt and Hall 1981). More precise taxonomic work could enhance the results of such studies. For example, reduced species richness has been cited repeatedly as a major biological ef- fect of increased acidity (Haines 1981, Hall et a/. 1980). Since midge richness also declines substantially, accurate identification of midges would amplify the observed reduction in taxonomic richness of the total macroinvertebrate community. One approach toward understanding the effects of acidification is to compile a list of acidobiontic taxa. The occurrence of relatively few taxa in severely stressed waters facilitates the development of a species assemblage K. W. SIMPSON 323 ap) Chironomini a ME Tonytarsini > Tanypodinae S 80 (] Orthocladiinae = (am) Le © a 40 Y, ca Y ud Yj => =) za O =) Zz STATION | 2 3 4 Fic. 2. Number of individuals and number of taxa of midges collected in Silver Run and tributaries, July (J) and September (S), 1980. Results are composites of 3 Surber samples for each collection site and time. indicative of such conditions. Conchapelopia contains some of the most successful inhabitants of strongly acidified streams. In addition to our observations regarding Conchapelopia americana and C. flavifrons, Roback (1981) reported these same species as occurring in water with pH 4.1 to 7.0. Along with ‘‘C. flavifrons var.’’ these are the only members of the Thienemannimyia group found at pH 4.1 to 5.0. Scullion and Edwards MEM. AMER. ENT. SOC., 34 324 CHIRONOMID COMMUNITIES IN ADIRONDACK STREAM TABLE 3. Common acidophilic Chironomidae in the Silver Run system (those occurring in 2 or more samples from station 1). Taxon July September Conchapelopia americana/flavifrons xX Cricotopus vierriensis Cricotopus/Orthocladius sp. Mesocricotopus sp. x Parametriocnemus lundbecki Rheocricotopus sp. x Thienemanniella nr. xena Tvetenia bavarica gr. Micropsectra polita? Rheotanytarsus exiguus gr. Tanytarsus brundini? ~*~ x x KK xx TABLE 4. Common acidobiontic Chironomidae in the Silver Run system (those occurring in 2 or more samples from Cellar Brook and/or Bradley Brook). Taxon July September Conchapelopia americana/flavifrons x Cricotopus tremulus gr. Cricotopus vierriensis Eukiefferiella claripennis gr. x Heterotrissocladius hirtapex? xx KK MK (1980) reported Conchapelopia pallidula (Mg.) as one of only two insects found in a small stream with pH less than 3.5. Eukiefferiella claripennis group is the most tolerant group in the genus Eukiefferielia to a number of environmental perturbations, including low pH (Bode, 1983). Larvae of E. claripennis (Lundb.) have been found in alkaline chalk streams with pH above 8.0 (Pinder, personal communica- tion). Roback (1974) reported ‘‘Eukiefferiella sp.’’ as occurring at pH greater than 8.5. Further taxonomic work must be done to determine if species in this group have different pH tolerances or if a single species can tolerate a wide range of pH. Although Cricotopus is one of the most common and widely distributed lotic midge genera, I could find little information regarding its occurrence in acidified waters. Data presented by Beck (1977), Curry (1965) and Roback (1974) suggest that most species prefer neutral or slightly alkaline waters. Many midges are adversely affected by decreased pH. For example, Pratt and Hall (1981) tentatively recommended midges (and mayflies) as in- K. W. SIMPSON 325 dicators of acid precipitation in small mountain streams, based on increased drift rates in response to experimental acidification. Their study documented the sensitivity of midges as a group to decreased pH, but it did not distinguish between sensitive and tolerant taxa because the midges were identified only to the subfamily level. Except for Conchapelopia americana/flavifrons and Cricotopus vierrien- sis, all of the common acidophilic taxa in our study (Table 3) were much less abundant or absent in the acidobiontic communities. Although some of these taxa probably are limited directly or indirectly by low pH, con- siderable additional sampling would be necessary to confirm that the observed distributions were not due to sampling variability. Moreover the responses of the same species may be different in other streams due to dif- ferences in stream chemistry. Haines (1981) discussed discrepancies between reported responses of congeneric organisms to decreased pH and warned of the difficulty of determining cause and effect relationships in complex ecosystems. Some of the taxa which appear acid-sensitive in the present study are in genera reported by Scullion and Edwards (1980) to contain acid-tolerant representatives (Micropsectra, Rheocricotopus, Rheotanytar- sus, Tanytarsus). Additional work must be done to produce a list of acid- sensitive midge species. It is beyond the scope of this work to discuss in detail the many ways in which low pH can affect biological communities. Haines (1981) and Sutcliffe and Carrick (1973) discuss several potential mechanisms. In high- elevation areas of the Northeast mobilization of aluminum at low pH is thought to be a particularly important consequence of acid precipitation, with severe biological and ecological ramifications (Cronan and Schofield 1979, Hall et a/. 1980). In Silver Run, low pH is evidently not solely respon- sible for the observed differences in the midge fauna. If it were, station 4 should support a fauna intermediate between that found at station 1 and those found at stations 2 and 3. ACKNOWLEDGMENTS Mr. James Colqhoun (New York State Department of Environmental Conservation) suggested the study area and provided the pH data. Larry Abele, Robert Bode, Ricky Graham, Dave Ouderkirk, and Robert Peck assisted in the collection, sorting and identification of samples. New York State’s Biological Stream Monitoring Program is jointly sponsored by the US Environmental Protection Agency under Section 106 of the Pure Waters Act and the New York State Departments of Health and Environmental Conservation. Additional financial support for this study was provided by a Section 208 grant from the Bureau of Water Resources, New York State MEM. AMER. ENT. SOC., 34 326 CHIRONOMID COMMUNITIES IN ADIRONDACK STREAM Department of Environmental Conservation. Drs. Peter Cranston and Clive Pinder determined the proper placement of Tanytarsus brundini? (= Micropsectra curvicornis sensu Chernovskii). Mr. Bode and Drs. C.O. Berg and G.W. Fuhs critically reviewed the manuscript. LITERATURE CITED ARNOLD, D.E., P.M. BENDER, A.B. HALE AND R.W. LicHT. 1981. Studies on infertile, acidic Pennsylvania streams and their benthic communities. pp. 15-33. In R.L. Singer, (ed.), Effects of acid precipitation on benthos. North Amer. Benthol. Soc. ISBN 0-941908-00-3. 154 pp. Beck, W.M., JR. 1977. Environmental requirements and pollution tolerance of common freshwater Chironomidae. EPA-600/4-77-024. 261 pp. BopE, R.W. 1983. Larvae of the North American Eukiefferiella and Tvetenia (Diptera: Chironomidae). N.Y.S. Museum Bull. No. 452. 40 pp. BoerGER, H. 1981. Species composition, abundance and emergence phenology of midges (Diptera: Chironomidae) in a brownwater stream of West-Central Alberta, Canada. Hydrobiol. 80:7-30. BRENNAN, A., A.T. WALENTOWICZ AND A.J. MCLACHLAN. 1981. Midges (Diptera: Chiron- omidae) from the upper reaches of a spate river. Hydrobiol. 78:147-151. COFFMAN, W.P. 1973. Energy flow in a woodland stream ecosystem: II. The taxonomic composition and phenology of the Chironomidae as determined by the collection of pupal exuviae. Arch. Hydrobiol. 71(3):281-322. CocsBiLL, C. 1976. The history and character of acid precipitation in eastern North America. U.S. Forest Serv. Gen. Tech. Rep. NE-23:363-370. CoGBILL, C. AND G. LikENS. 1974. Acid precipitation in the northeastern United States. Water Resources Res. 10:1133-1137. CRONAN, C.S. AND C.L. SCHOFIELD. 1979. Aluminum leaching to acid precipitation: effects on high elevation watersheds in the Northeast. Science 204:304-306. Curry, L.L. 1965. Survey of environmental requirements for the midge (Diptera: Tendi- pedidae). pp. 127-141. In Biological problems in water pollution, 3rd seminar, 1962. U.S. Public Health Serv. Publ. No. 99-WP-25. 424 pp. DrakE, C.M. 1982. Seasonal dynamics of Chironomidae (Diptera) on the Bulrush Schoe- noplectus lacustris in a chalk stream. Freshwater Biol. 12:225-240. Haines, T.A. 1981. Acidic precipitation and its consequences for aquatic ecosystems: a re- view. Trans. Am. Fish. Soc. 110:669-707. Hatt, R.J., G.E. LIkens, S.B. FIANCE AND G.R. HENDREY. 1980. Experimental acidifica- tion of a stream in the Hubbard Brook Experimental Forest, New Hampshire. Ecology 61(4):976-989. Harris, T.L. AND T.M. LAWRENCE. 1978. Environmental requirements and pollution tolerance of Trichoptera. EPA-600/4-78-063. 310 pp. HUuBBARD, M.D. AND W.L. PETERS. 1978. Environmental requirements and pollution toler- ance of Ephemeroptera. EPA-600/4-78-061. 461 pp. LIKENS, G. AND F. BORMANN. 1974. Acid rain: a serious regional environmental problem. Science 184:1176-1179. LIKENS, G., R. WRIGHT, J. GALLOWAY AND T. BUTLER. 1979. Acid rain. Sci. Amer. 241(4):43-51. Mason, W.T. JR. AND P.P. YEvIcH. 1967. The use of phloxine B and rose bengal stains to facilitate sorting benthic samples. Trans. Am. Microsc. Soc. 86(2):221-223. K. W. SIMPSON 327 NEWBOLD, J.D., D.C. ERMAN AND K.B. Rosy. 1980. Effects of logging on macroinverte- brates in streams with and without buffer strips. Can. J. Fish. Aquat. Sci. 37:1076-1085. PENNAK, R.W. 1977. Trophic variables in Rocky Mountain trout streams. Arch. Hydrobiol. 80(3):253-285. PINDER, L.C.V. 1980. Spatial distribution of Chironomidae in an English chalk stream. In D.A. Murray, (ed.), Chironomidae—ecology, systematics, cytology and physiology. Pergamon Press, New York, pp. 153-161. Pratt, J.M. AND R.J. Hatt. 1981. Acute effects of stream acidification on the diversity of macroinvertebrate drift. pp. 77-95. In R.L. Singer (ed.), Effects of acid precipitation on benthos. North Amer. Benthol. Soc. ISBN 0-941908-00-3. 154 pp. RinGE, F. 1974. Chironomiden Emergenz 1970 in Breiten und Rohr wiesenbach. Arch. Hydrobiol. Suppl. 45:212- Rospack, S.S. 1974. Insects (Arthropoda: Insecta). Chapter 10. In C.W. Hart, Jr. and S.L.H. Fuller, (eds.), Pollution ecology of freshwater invertebrates. Academic Press, New York, 389 pp. 1981. The immature chironomids of the eastern United States. V. Pentaneurini- Thienemannimyia group. Proc. Acad. Natur. Sci. Philadelphia 133:73-127. SCULLION, J. AND R.W. Epwarps. 1980. The effects of coal industry pollutants on the macroinvertebrate fauna of a small river in the South Wales coal field. Freshwater Biol. 10:141-162. Simpson, K.W. AND R.W. BopeE. In preparation. The macroinvertebrate fauna of Silver Run, a mountain stream system with variable acidity. SuRDICK, R.F. AND A.R. GAUFIN. 1978. Environmental requirements and pollution tolerance of Plecoptera. EPA-600/4-78-062. 417 pp. SUTCLIFFE, D.W. AND T.R. CARRICK. 1973. Studies on mountain streams in the English Lake District. I. pH, calcium and the distribution of invertebrates in the River Duddon. Freshwater Biol. 3:437-462. THIENEMANN, A. 1954. Chironomus. Leben, Verbreitung und wirtschaftliche Bedeutung der Chironomiden. Binnengewasser 20:1-834. WEBER, C.I.(ED.). 1973. Biological field and laboratory methods for measuring the quality of surface waters and effluents. EPA-670/4-73-001. MEM. AMER. ENT. SOC., 34 ci ‘ Gr $3, Paracricotopus mozleyi n. sp. from Georgia, U.S.A. (Diptera: Chironomidae) JOHN W. STEINER United States Geological Survey, Water Resources Division National Water Quality Laboratory 6481-H Peachtree Industrial Blvd. Doraville, Georgia 30340, U.S.A. ABSTRACT. — The adult male, pupa and larva of Paracricotopus mozleyin. sp. were collected from a vertical rock seep in Lumpkin County, GA. All stages are described and illustrated and characters are given to distinguish the life stages of this species from the other species of the genus. This is the second species of Paracricotopus to be described from North America. Saether (1980) described Paracricotopus glaber and revised the genus which contained two other species, the palearctic P. niger Kieffer and P. uliginosis Brundin. He placed Paracricotopus Thienemann and Harnisch into a group with Nanocladius Kieffer, Mesocricotopus Brundin, Psec- trocladius Kieffer, and MRheocricotopus Thienemann and Harnisch. Saether’s new species, P. glaber was described from three associated specimens collected from a seepage area on a mountainside outcrop in Oconee County, South Carolina. Recently, I collected specimens of a new, species, P. mozleyi, from a similar rocky seepage area in Lumpkin County, Georgia. The adult male, pupa and the larva of P. mozleyi are described here following the morphological terminology of Saether (1980a). The Measurements are in microns and are expressed as means or ranges. Paracricotopus mozleyi n. sp. Type locality: Vertical rock seep 4.2 km northeast of Stonepile Gap crossroads on north side of State Route 60, Lumpkin County, Georgia. Elevation: 781 m. Type Material: Holotype: pharate male and pupal skin (recovered from salamander gut). Paratypes: reared pharate male with cast pupal and lar- val skins, 1 pupal skin, 1 partial pupal skin, 2 4th instar larvae, 1 4th instar larval head (recovered from salamander gut), 1 3rd instar larva. All specimens collected by John W. Steiner, 26 X. 81. All in coll. U.S. Nat. Mus. Diagnosis: The life stages of P. mozleyin. sp. can be distinguished from 329 MEM. AMER. ENT. SOC., 34 330 PARACRICOTOPUS MOZLEYI N. SP. those of P. glaber and P. niger by these combinations of characters: Adult male: anal point proper without setae but with 3-4 setae at base; squama with 3-4 setae; inferior volsella apparently weak; last three palpal segments as 62:80:100. Pupa: tergites VII-IX with anterior shagreen; largest precorneal seta longer than thoracic horn. Larva: last three antennal segments subequal; Lauterborn organs longer than segment III. Fics. 1-5. LE \ — 5 Paracricotopus mozleyi n. sp. Male — 1. Antenna. — 2. Hypopygium (right, ventral view; left, dorsal view). — 3. Anal point. — 4. Palpus. — 5. Scutellum. J. W. STEINER 331 Etymology: This species is named in honor of Dr. Sam Mozley of North Carolina State University. MALE (n=2: both pharate) Dark olive-brown body with red eyes. Total length about 3200. Head: eyes hairy, diameter of largest facet 10. Palpal segments (Fig. 4) in the ratio: 27:25:60:80:100. Outer vertical setae 2. Antenna (Fig. 1.) with 13 flagellomeres in the ratio: 33:23:20:20:27:27:30:30:28:27:26:26:268; AR: 0:84. Thorax: scutellum (Fig. 5) with six strong setae. Dorsocentral setae 6-9; prealars 3. Haltere dark. Wing: Squama with 3-4 setae. Hypopygium: anal point proper (Fig. 3) very small and without setae, length 8-11, base with 3-4 setae. Gonocoxite length 175, gonostylus (Fig. 2) length 83. Laterosternite IX with 2-3 setae. HR: 1.92-2.14. PUPA (n=4) Light Brown with darker spines. Total length 3070. Cephalothorax: \ength 1270. Thoracic horn (Fig. 7) entirely smooth or with two shallow lateral notches, length 95. Anterior precorneal seta length 125; median precorneal seta length 105. Median anteprontal setae 2, length of each about 90. Dorsum of cephalothorax weakly reticulate. Dorsocentral setae 4, distance between Ist and 2nd 50. Wing sheaths smooth. Abdomen: length 1800. Anterior shagreen on tergites VII-IX (Fig. 8). Median or posterior shagreen on sternites II-VIII. Numbers of large (8+) spines in median transverse rows on T III-VII as 8-20, 20-24, 20-29, 18-26, 0-3; numbers of small (8—) spines in same rows as 6-10, 9-13, 5-11, 7-8, 0-2. Numbers of large caudal spines on T II-VIII as 18-26, 30-34, 28-34, 34-37, 46-53, 45-52, 44-50. Numbers of small caudal spines on T II-VIII as 18-26, 22-26, 27-41, 34-38, 31-40, 35-38, 18-20. Sternites V-VII (Fig. 6) each with an irregular row of short caudal spines. Pedes spurii A only on S VI, perhaps present on S VII as 1-2 minute spinules. Conjunctives II/III through V/VI each with 3-5 rows of thin spinules. Anal lobe without fringe. Apical spines absent. Anal macrosetae subequal, length 85-95. LARVA — 4th INSTAR (n=4) Tan body with golden brown head capsule. Total length about 4050. Head: length about 400, width about 280. Mouthparts and occipital margin dark brown. Mentum with 11 teeth (Figs. 13, 17). Median tooth pointed with 1st laterals adpressed; last lateral teeth reduced, mentum width: 85. Ventromental plate crescent shaped, thin, curving to base of mentum. Mandible (Fig. 12) sharply curved with three inner teeth; seta subdentalis short, truncate; seta interna with six or seven thin, regular, apically serrate filaments; mandible length 111. Epipharynx with SI (Fig. 14) apically bifid, SII simple; pecten epipharyngis with three blunt teeth, basal sclerite about as long as ungula. Premandible (Fig. 15) simple with mesal lobe; premandibular brush produced into a single large serrate seta; premandible length: 67. Maxilla (Fig. 11) with large palp and 6-7 smooth lacinial chaetae. Antenna (Fig. 10) 5-segmented with ring organ at base of segment I; blade with sclerotized base, length: 32; length of segments: 51:19:6:5:4; Lauterborn organs paired at of apex of II, length: 7; Segment I with two long basal lateral setae and mounted on short, sclerotized, spurred tubercle; AR: 1.48. MEM. AMER. ENT. SOC., 34 332 PARACRICOTOPUS MOZLEYI N. SP. Body: length about 3600. Anal tubules (Fig. 16) much longer than posterior parapods. Pro- cerci sclerotized and dark with large mesal spurs, each with five long (500) dark anal setae and two small lateral setae. Supraanal setae reduced. Posterior parapods each with 16 strong yellow claws. Long claws of anterior parapods serrate. Se ee eS Seams, Te ye! i \ s 4 ] ee emgeco | ¥ s Gass Ses ara eens ented e: \ j 18,0 a tJ = 7/50 td = SCG Opa ae a NO, /3m2 0600 MEM. AMER. ENT. SOC., 34 = w 40 30 Tanytarsus flavellus 20 10 1100 1600 2100 0200 0700 TIME (HOURS) 359 360 DIEL PERIODICITY IN CHIRONOMID EMERGENCE NIELSEN, E. T. 1962a. A note of the control of Glyptotendipes paripes. Mosq. News 22:114-115. 1962b. Contributions to the ethology of Glyptotendipes (Phytotendipes) paripes Edwards. Oikos 13:48-75. OuiveR, D. R. 1968. Adaptation of Arctic Chironomidae. Ann. Zool. Fenn. 5:111-118. 1971. Life history of the Chironomidae. Ann. Rev. Entomol. 16:211-230. PALMEN, E. 1955. Diel periodicity of pupal emergence in natural populations of some chironomids. Ann. Zool. Soc. Vanamo 17:1-30. 1956a. Periodic emergence in some chironomids — an adaptation to noc- turnalism. K. G. Wingstrand (ed.): Bertil Hanstrom, Zoological Papers in honour of his sixty-fifth birthday. pp. 148-256. 1956a [1958]. Diel periodicity of pupal emergence in some North European chironomids. Proc. Int. Cong. Entomol., 10th. 2:219-224. 1958. Periodic emergence in some North European chironomids. Verh. Int. Ver. Limnol. 41:817-822. PHILUIPP, P. 1938. LExperimentelle Studien zur Okologie von Chironomus thummi Kieffer. Zool. Anz. 122:237-245. REMMERT, H. 1955a. Untersuchungen uber das tageszeitlich gebundene Schlupen von Pseu- dosmittia arenaria (Dipt., Chiron.). Z. vergl. Physiol. 37:338-354. 1955b. Tageszeitlich gebundenes Schlupfen bein Pseudsmittia arenaria (Dipt. Chironomidae). Naturwiss. 42:261. 1965. Uber den Tagesrhythmus arktischer Tiere. Zschr. Morphol. Okol. Tier. 55:142-160. Scott, W., AND D. F.OppykE. 1941. The emergence of insects from Winona Lake. Invest. Ind. Lakes 2:3-14. TokunaGA, M. 1932. Morphological and biological studies on a new marine chironomid fly, Pontomyia pacifica, from Japan. I. Morphology and Taxonomy. Mem. Coll. Agric. Kyoto Univ. 19:1-5S6. 1934. Periodical emergence of a marine midge, Pontomyia pacifica Toku- naga. Proc. Kansai Entomol. Soc. 7. Woot, D. AND J. KuGLER. 1969. Circadian rhythm in chironomid species (Diptera) from the Hula nature preserve, Israel. Ann. Zool. Fenn. 6:94-97. A Reconnaissance of the River Rhine using Chironomidae Pupal Exuviae (Insecta: Diptera) RONALD S. WILSON AND SUSAN E. WILSON Department of Zoology University of Bristol, England ABSTRACT. — Samples of Chironomidae (Insecta: Diptera) pupal exuviae were taken from 62 stations along the River Rhine, its tributaries and associated lakes during a 3-week period in July/August 1981. Good samples proved easy to obtain by either hand-net or surface drift-net from all sections of the river, including both the alpine headwaters and the lowland reaches in Holland, as well as from the Bodensee and the Grand Canal d’ Alsace. The pattern of water quality was derived from consideration of both the sample diversity and the pollutional tolerance of taxa, and the results match well with the saprobien system classification (1969/74), and follow the main trends of chemical water quality. The technique of exuvial sampling is shown to have a high potential as a tool for water quali- ty assessment in large rivers of international importance. INTRODUCTION During the period of 23.7.81 to 10.8.81 the authors travelled the length of the Rhine from the Alps to the North Sea, taking a set of over seventy samples of chironomid pupal exuviae. Samples were taken from many of the more important tributaries and from the Bodensee as well as from the main river itself. Both surface drift-netting and bankside flotsam collecting techniques were employed, depending on the river conditions, although flotsam collections were preferred as they are less subject to diurnal varia- tion (McGill, 1981). There were two principal aims to this work. Firstly, to demonstrate that the technique of exuvial collecting could be carried out on a major interna- tional river, and secondly, to show whether the samples could be analysed to reveal the water qualities of the different sections of the river, as they have been successful in doing for rivers in Great Britain. Much of the British work has been carried out under contract with the Department of the Environment. ge In the event, very little difficulty was experienced in obtaining large samples of exuviae from all the places sampled. In the Vorderrhein, Hinter- rhein and Alpenrhein, the fast flow and turbulent nature of the river made it necessary to use the drift-nets extensively. The exuvial types have been identified to genus, using keys prepared by the authors and Dr. P.H. Langton, and some species identifications have 361 MEM. AMER. ENT. SOC., 34 362 RECONNAISSANCE OF RHINE RIVER been checked by Dr. F. Reiss and Dr. P.H. Langton. A full species list will be published later. The River Rhine — The Rhine is a major international river, flowing through Switzerland, Lichtenstein, Austria, France, West Germany and Holland. It has become famous not only for its navigational importance, but also for its pollutional state. Effluents from major industrial centres (such as K6In and the Ruhr Valley) as well as city sewage are carried in its waters, and are the cause of much concern to the downstream nations. Even the oligotrophic Bodensee (Lake Constance) is under threat from effluents originating in Switzerland. The river (Figure 1) is 1320km long, and can conveniently be divided into sections for study. The VORDERRHEIN rises in Lake Toma on Piz Badus at 2344km above mean sea level, and links at Reichenau with the HINTER- RHIEN which flows from the Rheinwaldhorn glacier, to form the APLENRHEIN which then flows into the BODENSEE at 395m. Below the lake the river can be split into consecutive sections: the HOCHRHEIN from the Bodensee to just below Basel, the SUDLICHER and the NORDLICHER OBERRHEIN, from Basel to Karlsruhe and Karlsruhe to Mainz respectively, the MITTELRHEIN from below Mainz to just above Bonn, the NIEDERRHEIN from Bonn to Emmerich on the Dutch border, and finally the WAAL which flows from Emmerich to the North Sea (Friedrich & Miiller, 1983). One of the exits has been dammed to form the HARINGVLIET, a freshwater lake, which has also been sampled in this study. In general terms, serious pollution begins at Basel, which is the beginning of commercial navigation. From this point to just south of Karlsruhe the river forms the boundary between France and Germany and receives the saline effluents from French mines at and below Strasbourg. The river then flows past Mannheim where it is joined by the polluted river Neckar, past Worms and on to Mainz, where it receives another seriously polluted tributary, the Main. Throughout this Oberrhein section, the river has been subject to extensive restructuring to aid navigation, and cuts through se- quences of old meanders which are more or less static. They form an impor- tant faunal refuge from which the main river can be continuously repopulated. Mention must also be made of the canal system near Strasbourg, and in particular the Grande Canal d’ Alsace linking Basel and Strasbourg, and which carries most of the commercial shipping. It is regulated to maintain its flow at the expense of the main river, and during dry weather it may carry 90% of the total river flow (Rat von Sachverstan- digen fiir Umweltfragen, 1976). 363 R. S. WILSON AND S. E. WILSON SP (ROTTERDAM | 44 a ,.— a2 s y Z2™ 2) S SS , Oey, ¢ eo Waal NIOMEGEN eS ey en Maas y ‘ es — Emscher =~ DUISBURG tf) Ruhr DUSSELDORF Wupper 4, A, Hin @ gS KOLN “o & S> oe eS S STRASBOURG ex S$ OBERRHE Ty (NORDL ICHER) of & °@ 11] Grand Canal Seat cate ! KONSTANZ d'Alsace HOCHRHEIN BODENSEE “ sper BASEL Aare & k= $ x LY > CHUR SS Saas ani Sketch map of the River Rhine showing some of the principal cities and tributaries, Fic. 1. and the regional divisions. MEM. AMER. ENT. SOC., 34 364 RECONNAISSANCE OF RHINE RIVER At Mainz the waters of the Main flow along the north bank, and are kept largely separate from the Rhine water by the presence of midstream islands until they mix completely at Bingen where the river enters the Mittelrhein gorge. The water here is very deep and fast flowing, with steep rocky banks. It is joined by the Mosel at Koblenz, and then flows on to Bonn, where it emerges from the gorge into a more open countryside. From here it may be termed the Niederrhein, and is subject to the most severe pollution of its whole course, from sewage and industrial effluents from Bonn, Leverkusen, Koln, Dusseldorf and the great industrial complex of the Ruhr. The highly polluted tributaries of the Wupper, the Ruhr, the Emscher and the Lippe which enter in this section, were selected for special study, as it was considered that they would show the characteristics of extreme pollu- tion. Below Emmerich the Rhine enters Holland, and divides at Nijmegen into two rivers, the Lek and the Waal. Of these two, the Waal is the more impor- tant, but there are many connecting channels and canals both between them and with the Maas. This network of waterways links Rotterdam and the Hook of Holland, as well as cities further afield. The Lek passes through Rotterdam, and together with the Waal and Maas discharge to the North Sea at the Hook of Holland. Many of the features of the Rhine have been manipulated by man in order to maintain navigation. The banks in many areas of the lower river have been stabilised with groins set out at right-angles to the bank. These form pockets of relatively still water, but even so the passage of the great barges cause considerable wave action on the shore. Further up the river, especially in the Mittelrhein, the channel is divided by a series of islands. In order to maintain an adequate channel for shipping under dry weather con- ditions, walls associated with the islands have also been constructed in the river, for instance at Mainz. These are only visible above water at low water, but provide lentic areas where sediment may be deposited, in the cen- tre of the fast-flowing main stream. In general, however, the river may be considered as a turbulent and rapid river throughout most of its course, with a bed principally of stones and gravel. Large accumulations of sand and silt can only occur at the downstream end, especially in the delta and estuary regions. METHODS Sampling and sampling stations — Table 1 gives a list of the stations from which samples were taken during this survey. They were selected to give a complete coverage of the river, and included samples from the lower end of R. S. WILSON AND S. E. WILSON 365 TABLE |. List of sampling sites. VORDERRHEIN VRI1 Oberalppass VR2_ Disentis Bridge VR3 Tavanasa Bridge HINTERRHEIN HR1 Hinterrhein HR2 Clugin Bridge HR3 Rodels Bridge ALPENRHEIN SR1 Untervaz Bridge SR2_ Sevelin Bridge SR3 Diepoldsau Bridge SR4_ Altenrhein Zoll BODENSEE Bl Aeschach B2 Nonnenhorn B3 Horn B4 Berlingen (Untersee) HOCHRHEIN Rl Rheinklingen R2 Eglisau R3a Murg (Laufenburg) R3b Murg R4 Rheinweiler OBERRHEIN (SUDLICHER) RS Weisweil R6 Ichenheim R7 Sollingen OBERRHEIN (NORDLICHER) R8 Leopoldshafen R9 Germersheim R10 Philippsburg R11 Rheinhausen R12 Speyer R13 ~~Bruhl R14 Rheindurkheim R15 Mainz, Mombacher (S. bank) R16 Mainz, Biebrich (N. bank) R17 Assmannshausen MITTELRHEIN R18 Loreley R19 Irlich R20 Erpel (Remagen Bridge) MEM. AMER. ENT. SOC., 34 NIEDERRHEIN R21 Niederkassel R22 Leverkusen R23 Hitdorf R24 Kaiserwerth R25 Gotterswickershamm East R26 Gotterswickershamm R27 Rees R28 Emmerich WAAL W1 = Dodewaard W2 Rossum W3 _~—sCBrakel HARINGVLIET Hl Hellevoetsluis Tributaries AARE AA1 Dottingen GRAND CANAL D’ALSACE CA1 Chalampé CA2 Geiswasser NECKAR NE1 Neckarhausen MAIN MAI Russelsheim (10 samples) MOSEL MOl1 Lay (10 samples) WUPPER WUI1 Vohwinkel WU2 Mungsteiner Brucke WU3 Landwehr WU4 Burrig (Leverkusen) RUHR RH1 Duisburg Hafen EMSCHER RE! Stapp LIPPE LI1 Oberlippedorf 366 RECONNAISSANCE OF RHINE RIVER some of the major tributaries. Help in the selection was given by Professor Kinzelbach of Mainz University, and Dr. Caspers of Bonn University, and is here gratefully acknowledged. At each station, shoreline flotsam collections were made with a fine- meshed hand net on a triple extendable handle, or with a series of 3 surface drift-nets, 2 with openings of 10 x 30 cm and one of 10.5 x 61 cm. The net mesh was 250 pm aperture in each case. Each sample was briefly examined in a white bowl to confirm that an adequate number of exuviae had been taken, before moving on to the next station. All samples were sieved at the water’s edge through a 250 pm sieve, placed in polythene bags, labelled, and fixed with formalin. They were then placed in a sealed, insulated container for transportation. Measurements were made at each station of the water temperature, con- ductivity, pH and Dissolved Oxygen, using HACH portable meters, Models 17250, 19000 and 16046. The values must be treated with caution, however, since they were necessarily taken from the shoreline, and may bear little resemblance to measurements taken in the main stream. The turbidity was also recorded by passing 20 ml of water through a 13 mm diameter glass fibre filter paper, which was then mounted on the record card. It could be seen that the wash from a passing barge significantly increased the turbidi- ty, but that the silt then settled rapidly again. Laboratory procedure and sample analysis — Each sample when opened was rinsed through a 250 mu sieve to remove the formalin, and suspended in a bowl of water. Subsamples were taken by the ‘‘scoop’’ method (Wilson & McGill, 1977, 1979) and approximately 200 exuviae were mounted on slides, 20 to a slide, using dimethyl-hydantoin-formaldehyde mountant. These exuviae were identified, and counted, and the results stored on disc using a Commodore 3032 Series PET Computer. The samples from each river section were added to give ‘‘amalgamated samples,’’ which reflect the general characteristics of the section in question. These, together with the original samples were then analysed to reveal the percentage composition of species and subfamilies, and the Menhinick and Shannon-Weaver Diver- sities (See Hellawell, 1978). The assessment of water quality followed the technique developed under contract with the Department of the Environment for the River Trent and other British rivers (Wilson, unpub. reports to the DOE, 1976-82) which en- tailed the allocation of each species or taxon of chironomid identified to four categories based on an empirical knowledge of their pollution “tolerance’’ (Wilson and McGill, 1982). The tolerance considered was prin- cipally with respect to low oxygen and organic pollution. Category A was R. S. WILSON AND S. E. WILSON 367 the least tolerant, B and C intermediate, and D the most tolerant. Each of these was calculated both for species presence/absence data and for relative abundance data. RESULTS In this paper the samples taken down the main river have been amal- gamated for each region (see Figure 1 and Table 1). Samples taken on tributaries have been given individually: in most cases only a single sample was taken near the point of discharge of the tributary to the main river, but two stations were sampled on the Grand Canal d’Alsace, and four on the Wupper. Table 2 gives some of the more important data derived from the samples, including the percentage of taxa and exuviae in the tolerance categories A, B, C and D, the Menhinick and the Shannon- Weaver diversities, as well as the percentage of sediment-dwellers. It appears from current work on the River Trent in Britain, that the sediment-dwellers may be more exposed to the effects of industrial and heavy metal pollution, and may be selectively eliminated where this occurs. Certainly the levels of heavy metals in the Rhine sediments are very high, especially at the lower end of the river (Forstner and Wittmann, 1979; Miller, 1971; van der Veen and Huizenga, 1980). Taxa have been allocated to tolerance categories and as sediment-dwellers according to the authors’ best estimates, after a close study of the literature, coupled to personal observations and subsequently modified by discussion with colleagues. As more detailed information becomes available on the ecology of the different species, this categorization will undoubtedly have to be modified. The categories chosen are as follows: A. INTOLERANT: found only in clean waters. B. FACULTATIVE/INTOLERANT: common in clean waters, but occasionally found in situations of mild pollution. C. FACULTATIVE/TOLERANT: often found in abundance in mild- ly polluted waters, but also present in clean waters. D. TOLERANT: capable of survival under conditions of severe pollu- tion, but may also be found in low numbers in clean waters. Of the two diversity indices used, the Menhinick diversity depends on the numbers of taxa relative to the numbers of exuviae sampled, whereas the MEM. AMER. ENT. SOC., 34 RECONNAISSANCE OF RHINE RIVER 368 9°08 € 8S Ol Me iw Or BS A A 3 OL TI 180 09 9 ET 6p J€ 61 -T Ww 9 CE OT (46 vl 0 L0 S81 Vb oy ie Wo Wp 66° Ge SI £9'T 790 VT VIC 6 8€ th OL 9€ IC Ce I2 LOC 801 SG O'9€ Cf th CC vl ve OC 8¢ B8I LOC 80°1 ve 9°87 Se a Wy tf ite @¢ SE il LOT ev SE VAY Lt! OC €€ O€ LO 81 OF SZ OEE S61 tp y°89 Sp I ve O1 LE 6I OF PI I@E€ €6'1 LE te vl Il cl €f 9 8f 8 Le Lt 08°C 90°C Vs ULI 8 Lt! cl €9 II O@ 62 OF S8C byl Te £8 C 8¢ cl 8S bl 61 Cd SYP 68°C Sv STANK BX} @Q 7 OF a ov GeO) te Vi H d (%) —pingy PU | Key PARAM (YOIUTY uouueys = - Ua SI9][9Mp (%) Sat1osajeD Ailend AjISIOAIG JUSUIPaS exe] sojdures 0¢~ a 899 61 L8LI LT L99 87 CHIT Os $6 ce Ovol €9 9L8 LS L9S 6V 68S SE 079 9€ deIANXa jo jo “ON “ON *SOLILINGII} P9j9Jas PUL SUIYY IOATY OY) JO suorse1 jedrioutid 9y) 10} eyep sjdures oiseg nn OM OM i jo “ON LAITADNIAVH IVVM NIBFHYYACHIN NIGHYTALLIN (N) NICHYAAIO (S) NIGHYAAIO NIHHYHOOH aaSNACOd NIHHAUNAd TV NIZHYYaLNIH NIFHAYAaCYOA TAEGCKAD 369 WILSON WILSON AND S. E. R. S. eC 0°66 g9¢ L'86 6°86 OTT es VL Eee Is 8 I cl OEE L Ol eee S34 0'S¢ ce€ c 8I 001 TI@ L9@ v 67 OST sat L St v1 86 19 66 66 8€ oF ce 08 6S vl cI 91 18 sé SI 8t 9E 67 owtroo (6S 0s cf CC €€ 97 os L9 Lé 0¢ 97 LC 1X6 Sc Te vl 0@ VC O€ 61 v7 oo sé IT 61 6¢ Is T 90°0 Tec 800 L0'0 10'¢ 9EC €L I 69'T S8I OL T 60° 69°C (oan 170 19°T 970 C70 9ST cel ec l €0'1 Stl vel am | 6£ 7 LIZ v0? LAUY4 6€7 601 002 87 Ove LAK OIZ CCC VOT VIC 8I €7 61 SI LI 02 oT ce -— mt JaopoddtyisqoO dddIT ddeis aYaHOSWA sinqsing aqdHnNa sing Iyompue Ty Ioula}ssun|y JOXUIMYOA ddddnM Key THSOW UIIOUJEssny NIVI uosneyleyooN UVAOAN JOSSEMSIOD odwejeyD AOVSTV.d IVNVO GNVaYD uosul}j0q qavv SOLIDINGILT, MEM. AMER. ENT. SOC., 34 370 RECONNAISSANCE OF RHINE RIVER PERCENTAGE ABUNDANCE Vorderrhein Hinterrhein Alpenrhein Bodensee Hochrhein Oberrhein (S) Oberrhein (N) Mittelrhein Niederrhein Waal Haringvliet ALISYSATC AIINTHNAW R. S. WILSON AND S. E. WILSON 371 Shannon-Weaver diversity is calculated from the relative abundance of the different taxa. The formulae are as follows: — Menhinick Diversity = S_ Ne> (where S = total taxa, and N = total exuviae) Shannon-Weaver Diversity = - Sigma Pi.log, Pi (where Pi is the proportion of individuals in the ith taxon). Bar-charts giving the relative percentages of subfamilies and tribes are shown in Figure 2 for the main Rhine regions and in Figure 3 for the tributaries. Table 3 shows how the numbers of species in the principal sub- families and tribes found in the Rhine alter between the regions. The numbers of species given in the table are estimates of those that would be expected to occur in samples of 500 exuviae in each case. The actual sample size for the different regions varies from 220 to 2142 exuviae, depending on the number of samples amalgamated, and the estimated species count was calculated by assuming that the Menhinick diversity remains constant for different sized samples (a condition that is fulfilled within a 10% to 15% error, see Wilson, unpub. report to the DOE, 1980), and then working out the number of species that would give the same diversity in a sample of 500 exuviae. This procedure therefore does not give the maximum number of species which could be found, but facilitates com- parisons between the regions. The Vorderrhein, Hinterrhein and Alpenrhein are dominated by Ortho- cladiinae (with estimates of 24, 24 and 38 spp. respectively), each section has only a single Chironomini and 4 or 5 Tanytarsini; Bodensee is dominated by Chironomini (17 spp.) and Tanytarsini (11 spp.), and has a reduced number of Orthocladiinae (8 spp.); the Hochrhein again shows Orthocladiinae as dominant (19 spp.), but with a significant number of Chironomini (12 spp.) and Tanytarsini (6 spp.); while the remainder of the river shows a falling-off in numbers of species reaching the lowest point in the Niederrhein with only 7 Orthocladiinae, 4 Chironomini and 2 Tanytar- sini. It is interesting to contrast Bodensee with the Haringvliet, which have totals of 43 and 18 species respectively. In Figures 4 and 5 the proportions of taxa and exuviae are plotted as pro- portions of the appropriate diversity index. Taxa presence/absence data are Fic. 2. Bar-charts of percentage relative abundance of taxa, and the Menhinick Diversity, for amalgamated data for the different region of the River Rhine. (Key: Tanypodinae (ex. Pen- taneurini), black; Pentaneurini, cross-hatched; Diamesinae, stippled; Prodiamesinae, squares; Orthocladiinae, white; Chironomini, vertical hatching; Tanytarsini, dots). MEM. AMER. ENT. SOC., 34 372 RECONNAISSANCE OF RHINE RIVER PERCENTAGE ABUNDANCE F > Dottingen = ™ SB oO ~ > Chalampe = Zz Zo Geiswasser 5 = i SS = =a ™ Neckarhausen R 2) 2) ra n (2p) oO) = © = NIVW S Lay D m Vohwinkel Mungsteiner = ag Landwehr m — 2) Duisburghafen = Mm = (Zp) Stapp 2 m | 7 . | Ober lippedorf |: 3 | ™m | ALISYSAIG AIINTHNAW R. S. WILSON AND S. E. WILSON 373 associated with the Menhinick diversity, and the relative abundance data with the Shannon-Weaver diversity. The diagrams in Figures 4 and 5 give a visual indication of water quality as they combine the diversity with the pollutional tolerance categories (A to D). The authors feel that it is very important to consider both diversity and tolerance in arriving at an assessment of water quality. High diversity coupled to a high proportion of intolerant taxa indicates a river of excellent quality, if they are both low, then the water is strongly polluted. If the diversity is low, but the proportion of intolerants high, then the water is of good quality, but the fauna is ‘‘restricted’’ through natural physical or chemical conditions. This condition is typical of high mountain areas with low productivity, but no pollution. If the diversity is high and coupled to a high proportion of tolerant forms, then the water is organically ““enriched’’, but not severely polluted. Other categories may be devised, depending on the setting of limiting values for the diversity and proportion of intolerant /tolerant taxa, and the limiting values used in this study are given in Table 4. Lakes normally have a high proportion of sediment-dwellers, and as these tend to be more tolerant, they may be assigned to a lower category than is warranted; care therefore must be taken in the interpretation of the data. The joining of the points in these diagrams by straight lines is not intended to imply a gradation of intermediate values, but only to facilitate the eye in comprehending the overall patterns. The term ‘‘sediment-dwellers’’ as used in this study, is defined as those larvae which normally live in sand, silt or mud; but excludes those species whose larvae are principally associated with only sparsely silted rocks. It is important to distinguish between the data based on the presence/absence of taxa, and that based on the relative abundance of ex- uviae, and the two data sets give different information on the river condi- tions. Presence/absence of species gives equal weighting to each species pre- sent, however low their numbers, and can therefore be significantly biased by the occurrence of rare species. On the other hand, each species of chironomid has its own period (s) of maximum emergence, and so the actual numbers of exuviae collected may vary significantly between different seasons of the year. Thus single samples taken at different seasons will not give the same relative abundances, and should not strictly be compared. Fic. 3. Bar-charts of percentage relative abundance of taxa, and the Menhinick Diversity, for data from selected tributaries of the River Rhine. (Key: Tanypodinae (ex. Pentaneurini), black; Pentaneurini, cross-hatched; Diamesinae, stippled; Prodiamesinae, squares; Orthocla- diinae, white; Chironomini, vertical hatching; Tanytarsini, dots). MEM. AMER. ENT. SOC., 34 RECONNAISSANCE OF RHINE RIVER 374 So A ee ee eS ae 180 vL 0 790 80'T 801 ev I col £61 90°@ pri Saat AUISIOATP YOTUTYUSIAL (G6 899 L8LI L99 (GAYS S6S Ovol 9L8 LOS 68S 079 poullexe ovlANXe [210] 8I oT vl VC VC GE VAY 17 OV (G3 CE STIVLOL IS G 4 £ t3 9 9 II S ¢ v TursejAue L 9 L v 8 6 Il Cl LI if if if TuIWOUOIIY 9 9 [L Gil Ol el 61 8 8e VC VC sVUTIpe[SOyUO 0 0 0 0 0 0 if 0 if 0 0 seuIsoUeIpOld 0 0 0 0 0 0 if i if 6 € svulsouleld 0 if G 6 if if (6 if 0 0 if TulMoue ued 6 i 0 0 6 G 4 S if if 0 (tulmeuRjued Xe) aseuipodsuey PION pns JoqyA uloyd uleys uleys uloys uldys 20s ulous uleys uleys soqiL sulley jeem JOpoIN JOIN 10qO 10qO yooH uopog uod]y IQUI,R JeploA ® sarprweszqns “UdAIB OS]B oI UOIIOIS YORE JO} SPIANXS PUL saisads [e101 “(1x9 29S sprejep JouIny 10J) AVISIOATP YOIUTYLOIA] 24} 0} BUIpsIOIOe Zuruoniodoid hq aseo yore Ul sBIANXS QOS JO gidures & 0} a]qeJajol aie sloquinu sy], “OUlYyY Jeary 24} JO SUOTIIES jediourid oy} ul punoy soqiz} pur serpiureyqns ULEUI SY} UI Satseds pruoUOITYS JO IOQUINN “£ alav | R. S. WILSON AND S. E. WILSON 375 Single samples comprising a set taken all within a short priod of time, as were those from the Rhine, may have their relative abundances compared within the set, but should not be compared with data from other seasons or places. Relative abundance data becomes more valid when a series of samples has been taken from each station at different seasons, for instance monthly through the year. The presence/absence data is not so subject to these restrictions, as most of the river species have extended emergence periods during which they can be taken in a sample, but perhaps only at a low level of abundance. Ex- perience has shown that presence/absence data give more consistent results between different seasons than do relative abundance data (Wilson, unpub. reports to the DOE, 1976-82). More weight is therefore placed on presence / absence data in this present study. In addition, different interpretations must be applied to the two data sets. The presence of a species implies that it has the ability to survive in the habitat, and that its niche is supported by the prevailing conditions. The community of species therefore outlines the range of niches available and hence the range of conditions within the river. By considering the communi- ty as a whole, one may to a large extent avoid the problems of using in- dicator species which may be present or absent for a number of unrelated reasons. It also avoids undue emphasis being placed on any single exuviae which may be present by chance. All the species in the community are linked together by the actual water quality in which they all live, and which sets the limits of the extremes of tolerance within which the community exists. Within these limits, various other factors, such as substrate, food, preda- tion etc. all play their parts in shaping the community, but they will often exert more control on the numbers of each species present than on its ab- solute survival. Relative abundance data can give a measurement of the suc- cess of different species, and therefore indicate the relative extent of each niche within the ecosystem. It is an oversimplification to state that the numbers of a species indicate the amount of substratum available to it and therefore that a high proportion of mud-dwellers indicates a high propor- tion of mud in the substratum, but in general this must be true, and can be usefully examined in assessing river conditions from exuvial samples. To categorize the samples with respect to water quality, both the diversity and the proportions of categories A, B, C and D must be taken into ac- count. Table 5 gives pollutional categories allocated according to the scheme shown in Table 4, which can be compared with the saprobic data for 1969-74 (Rat von Sachverstaéndigen fiir Umweltgragen, 1976). Only presence/absence data has been used to calculate the values for Table 5. MEM. AMER. ENT. SOC., 34 RECONNAISSANCE OF RHINE RIVER 376 0 x AOD (a+ Vv) AjISIOAIG yorUTy Us|] MOT wpe] ysIH MOT [nyaqnop (peyote) [njiqnop MOT poo3 ysty (pe}d11]se1) poos JUd][29xXO WI9}SAS uoiqoides s]UsWIUIOD 0, Xx Yo 001 (+) (d+0+4+V) MOT v MOT uINIpe|\| € wInIps|] ysIH Z wINIpsT] ysIH al MOT ystH VI A}ISIOAIG Sse[D yoruryusyy Ayyend UONBOIJISse[D Aypend IajeA “UIa]SAS usiqoides ay} 0} diysuorjejel May] pue ‘elep [eIANXe WOIJ UONBOdIJISse[O AjeNnb 19jeM dy} BUIALIOP JO pOYjJoU oy] BUIMOYS BIQUI_ ‘“p ATAV], R. S. WILSON AND S. E. WILSON 377 RELATIVE ABUNDANCE DATA SHANNON DIVERSITY oO PRESENCE/ABSENCE DATA MENHINICK DIVERSITY Vorderrhein Hinterrhein Alpenrhein Bodensee Hochrhein Oberrhein (S) Oberrhein (N) Mittelrhein Niederrhein Waal Haringvliet Fic. 4. Water quality diagrams for the principal regions of the River Rhine, with the tolerance categories A, B, C & D plotted as proportions of the diversity. Upper diagram: relative abundance data and Shannon-Weaver Diversity; lower diagram: taxa presence/absence-data and Menhinick Diversity. (Key: A, white; B, vertical hatching; C, stip- pled; D, black). MEM. AMER. ENT. SOC., 34 378 RECONNAISSANCE OF RHINE RIVER DISCUSSION Of the two major objectives, that of testing the sampling procedure on a major international river, has been satisfactorily accomplished. Good samples of exuviae were readily taken at each sampling station. The wide applicability of the technique has therefore been clearly demonstrated. No other single sampling technique can obtain comparable samples from rapid, stony, mountain rivers, from large, deep, lowland rivers, as well as from canals and lakes. The second objective, of analysing the samples to reveal the water quali- ty, has also been satisfactorily achieved using techniques developed for British rivers. It is difficult to prove that the results are in accord with the currently known physics and chemistry of the Rhine, since the information available to the authors is mostly sketchy or out of date. For the purpose of this preliminary reconnaissance, however, it is sufficient to confirm that the exuvial samples and analyses reflect the major trends in the quality of the Rhine. Table 5 shows that the saprobity assessments of 1969-74 match well with the quality classifications derived from the exuvial data, with if anything, some recent improvements being demonstrated. There is evidence that levels of organic and heavy metal pollution (but not salt or organo- chlorides) have decreased between 1971 and 1980, especially in the Nieder- rhein (Arbeitsgemeinschaft Rhein-Wasserwerke, 1981; Miiller, 1979; Son- theimer, Gimbel and Weindel, 1979). Figure 4 illustrates the good quality of the Vorderrhein and Hinterrhein which have an excellent, if a little restricted, chironomid fauna. The Alpenrhein has a highly diverse fauna again of good quality, but demonstrating enrichment in this fast-flowing section of the river. The Bodensee has a rich chironomid fauna (Reiss, 1968), and the current samples show that there is a high proportion of more tolerant forms. There is also a high proportion of sediment-dwellers — a feature to be expected in a lake. It therefore appears that the water quality is not first-class, being assessed at Class 1B. However, care must be exercised in the interpretation of lake samples as no detailed work has yet been done to show the relation- ships between the shore line exuvial collections and the trophic status. It is likely that lakes will have to be assessed against a modified set of quality criteria. The Hochrhein has a good fauna, and is tentatively classed as IB along with its major tributary the Aare. The Aare has an exceptionally high diver- sity, which probably indicates slight enrichment. Work by Bloesch (1977), Caspers (1980) and Kinzelbach (1978) demonstrate the good diversity of the general invertebrate fauna in this region. R. S. WILSON AND S. E. WILSON 379 TABLE 5. Comparison between saprobic classification and exuvial classification for the principal regions of the Rhine and selected tributaries. Quality classification Saprobic (1969-74) Exuvial (1981) River Rhine Vorderrhein Hinterrhein Alpenrhein Bodensee Hochrhein Oberrhein (S) Oberrhein (N) Mittelrhein Niederrhein Waal Haringvliet Tributaries (in rank order) Aare (Dottingen) Gd. Canal d’ Alsace (Geiswasser) Wupper (Vohwinkel) Gd. Canal d’ Alsace (Chalampé) Mosel (Lay) Wupper (Mungsteiner Br.) Ruhr (Duisburghafen) Neckar (Neckarhausen) Main (Russelheim) Lippe (Oberlippedorf) Wupper (Landwehr) Wupper (Burrig) Emscher (Stapp) } 2 (enriched) 2 (enriched) 3 3 3 MEM. AMER. ENT. SOC., 34 380 RECONNAISSANCE OF RHINE RIVER The Oberrhein shows a progressive fall-off in quality downstream. Although the Stidlicher Oberrhein is still classed as IB, it suffers a reduction in diversity and in % Intolerant, compared with the Hochrhein (Figure 6). The percentage of sediment-dwellers also falls off very rapidly at this point. The Siidlicher Oberrhein receives a large quantity of effluent from industry and sewage below Basel and Strasbourg. Samples were also taken from the Grand Canal d’Alsace which runs parallel to this section of the Rhine. It is a very large concrete channel which carries most of the commercial traffic between two cities. Exuviae were found in abundance along the concrete canal edge, and the analyses show that the fauna was Class 2 at Chalampé but improved downstream to Class IB at Geiswasser. The Nordlicher Oberrhein shows a further reduction in diversity. It receives the waters of the Neckar and the Main, both of which show a polluted chironomid fauna exemplified by a severe reduction of the in- tolerant categories A and B. The Neckar fauna is more restricted than that of the Main, and may be subject to more severe industrial, as opposed to organic, pollution than the Main. It was established by taking samples from both sides of the river at Mainz and looking at the species composition, that the Main affects only the north bank of the Rhine downstream of its entry. The waters of the Main and the Rhine mix completely only at Bingen, where the river enters the narrow Mit- telrhien gorge (Kinzelbach, pers. comm.). In the Mittelrhein the river is steep-sided and rocky, and is therefore dif- ferent in character from the sections above and below. The quality index does not change, however, nor does the taxa composition alter to any great extent, but an improvement can be seen in an increase of the relative abun- dance of category A and B exuviae. There is also a reduction in the % sediment-dwelling taxa. The Mosel is the principal tributary entering in this section, but has a higher exuvial quality index (Class 2) than the Neckar or Main (both Class 3). The Neiderrhein flows through the industrial heart of Germany and receives the drainage from some of the most highly populated and industrial areas in the country. It is not surprising therefore that there is both a reduc- tion of diversity and of intolerant taxa and exuviae in this section, although on the quality assessment it is still just in Class 2. Of the tributaries sampled, the lower stations of the Wupper and the Emscher reveal a most seriously polluted condition. The set of samples taken down the Wupper show how this river alters under the impact of heavy sewage pollution. The top station at Vohwinkel lies above the major sewage works; Mungsteiner Brucke is characterised by a great increase of R. S. WILSON AND S. E. WILSON 381 2 GRAND CANAL D’ ALSACE NECKAR MOSEL RUHR (PRE MAIN WUPPER EMSCHER =) RELATIVE ABUNDANCE DATA SHANNON DIVERSITY Oo — PRESENCE/ABSENCE DATA os |Pa | MENHINICK DIVERSITY — qe (= _ ® (— ) i o e cB) 1S} (= ® = = os Cc ®o ® oD Oo oO o Om & D Q D an iS < MaiDn aS c [oY Cc Ee Oo cS = ee 2 oO Oo SF = — oOo = oO (cb) — wn = = OA a. — 7) 4 Wn Nx n = oO Dt NG a i v o-_ rs) ) p> &£ & & Fae & ® fo) = wo 5) =| oO =) S OO 35.55 s= a =) Se) = az =) SS Ss saa 1A ie) =) Fic. 5. Water quality diagrams for selected tributaries of the River Rhine, with the tolerance categories A, B, C & D plotted as proportions of the diversity. Upper diagram: relative abundance data and Shannon-Weaver Diversity; lower diagram: taxa presence/ absence data and Menhinick Diversity. (Key: A, white; B, vertical hatching; C, stippled; D, black). MEM. AMER. ENT. SOC., 34 382 RECONNAISSANCE OF RHINE RIVER tolerant taxa and reduction of intolerant taxa; while at Landwehr the sam- ple contained tolerant taxa only, coupled to an exceptionally low diversity, indicating very severe pollution. This condition persists downstream to Bur- rig, just above the confluence with the Rhine. The Emscher appeared to be a fast-flowing canalised river with clear water of very high conductivity (4000 puS), which foamed excessively over a weir at its point of discharge to the Rhine. At each of these very highly polluted stations the chironomid population was reduced almost completely to a single species, Chironomus riparius Meigen. In contrast, the Ruhr was revealed as an organically enriched river of Class 2, with a diverse, but tolerant, fauna. The Lippe was assigned to Class 3 and appeared to be rather more polluted. There is no significant change in the condition of the Rhine as it passes in- to Holland. Figure 4 shows that there is only a marginal increase in the taxa diversity in the Waal as compared with the Niederrhein. The relative abun- dance data, however, show an increase in the proportion of tolerant ex- uviae, perhaps due to enrichment with organic matter. In the Waal the percentage of sediment-dwelling species (as previously defined) drops to the lowest value for the whole river below Bodensee, although their relative numbers increase slightly. It is possible that the sedi- ment, in which the heavy metals tend to accumulate (Miiller, 1979), is too toxic to maintain a good chironomid fauna. In the single Haringvliet sample distinct changes in the fauna can be ob- served, probably consequent on its being in effect a freshwater lake. These changes include an increase in taxa diversity, and in category D taxa. This is balanced by an increase in the proportion of intolerant exuviae. In no way, however, does it approach the good quality of the Bodensee, and has been allocated to Class 3 using the existing quality criteria; but it must be pointed out again that this assessment may be too low a value for lentic situations. Of the actual genera and species identified, a few require detailed men- tion. Chironomus riparius is a well-known indicator of organic enrichment, and made up over 95% of the exuvial samples from the Emscher and the lower Wupper. This species was also found in the Rhine itself at Gotters- wickershamm below the Emscher inflow; and elsewhere was present low numbers in the Waal, the Main and the Hochrhein. There were two species of Rheotanytarsus exuviae present throughout most of the system, R. photophilus (Goetghebuer) and R. rhenanus Klink. The latter has been recently identified by Klink (in press). They dominated the fauna of the Mittelrhein and Niederrhein where the larval tubes could be seen very commonly on rocks along the shoreline. High numbers of Rheotanytarsus have previously been reported near Bonn by Caspers R. S. WILSON AND S. E. WILSON 383 100f % INTOLERANT DIVERSITY e @ 100pf % SEDIMENT DWELLERS Niederrhein Haringvliet Oberrhein (N) Mittelrhein — (22) ~— i= — {o)) = (= = cd) 2 So Vorderrhein Hinterrhein Alpenrhein Bodensee Hochrhein Waal Fic. 6. Graphs of % Intolerant (open circles, taxa; closed circles, relative abundance), Diversity (open circles, Menhinick; closed circles, Shannon-Weaver) and % Sediment-dwellers (open circles, taxa; closed circles, relative abundance) for the principal sections of the River Rhine. MEM. AMER. ENT. SOC., 34 384 RECONNAISSANCE OF RHINE RIVER (1980b). They were also present commonly in the Siidlicher Oberrhein above Strasbourg, and in the Grand Canal d’Alsace. Below Strasbourg in the Nordlicher Oberrhein, their numbers were remarkably low, as they were also in the Neckar and Main. High numbers were found in the Mosel, the Lippe and on the south bank of the river at Mainz, opposite to the point of inflow of the Main. The larvae of Rheotanytarsus are filterfeeders, and therefore depend on the suspended organic matter including phytoplankton which is commonly present (Backhaus & Kemball, 1978) and their distribution must be related to the balance between the availability of particulate food and the avoidance of areas of heavy siltation. Field observations showed that the suspended matter in the river increased markedly downstream, building up after the entry of the Main to a maximum in the Niederrhein and Waal. This distribution coincides with the dominance of Rheotanytarsus, and it may be that the fast flow or the frequent wave action on the shore due to passing barges is sufficient to keep the larvae free from excess silt deposition. An interesting difference in the distribution of the two species was noted, with R. photophilus dominating the more highly polluted regions and tributaries in the lower part of the river, and R. rhenanus more common in the Hochrhein. More detailed descriptions of the distributions of particular species or genera could be made, some of which appear to be linked to water quality, but are omitted because much would be speculative. It is clear, however, that sampling chironomid pupal exuviae has great potential as a simple and cheap technique in the biological assessment of water quality. It is ap- plicable to both lotic and lentic waters, although it may find particular ap- plication under conditions where other forms of sampling are difficult or dangerous. The authors feel that it may also prove of use in detailed monitoring of rivers of international importance. ACKNOWLEDGEMENTS The authors would like to thank Dr. N. Caspers, Dr. E.J. Fittkau, Pro- fessor R. Kinzelbach and Dr. F. Reiss, for their kind help in arranging the sampling trip and for discussion of the work. Dr. L.J. Huizenga very kindly supplied recent information on the chemistry of the Rhine. They are also much indebted to Dr. F. Reiss and Dr. P.H. Langton for taxonomic help, and to Miss E. Wells for sorting and mounting the samples. R. S. WILSON AND S. E. WILSON 385 REFERENCES BackKHAus, D. AND A. KEMBALL. 1978. Gewdssergtitverhaltnisse und Phytoplanktonent- wicklung im Hochrhein, Obberhein und Neckar. Arch. Hydrobiol. 82:166-206. BLoetscH, J. 1977. Bodenfaunistische Untersuchungen in Aare und Rhein. Schweiz. Z. Hydrol. 39:46-67. Caspers, N. 1980a. Die Makrozoobenthos-Gesellschaften des Hochrheins bei Bad Sack- ingen. Beitr.naturk.Forsch.StidwDtl. 39:115-142. 1980b. Die Makrozoobenthos-Gesellschaften des Rheins bei Bonn. Deche- niana (Bonn) 133:93-106. FRIEDRICH, G. AND D. MuLLER. 1983. Rhine. In: WHITTON B.A. (Ed.) Ecology of European Rivers. Blackwell, Oxford. HELLAWELL, J.M. 1978. Biological surveillance of rivers. N.E.R.C. Handbook, Water Research Centre, Stevenage. KINZELBACH, R. 1978. Veradnderungen der Fauna des Oberrheins. Beih. Veroff. Natur- schutz Landschaftspflege Bad. Wiirtt. 11:291-301. Kink, A. in press. Rheotanytarsus rhenanus n.sp., a common midge of the lithorheo- philic fauna in large lowland rivers (Diptera, Chironomidae). Ent. Bei., Amsterdam. McGu1, J.D. 1981. Distribution of Chironomidae (Diptera) throughout the River Chew drainage system, Avon, England. Unpublished PhD thesis, University of Bristol. MULLER, G. 1979. Schwermetalle in den Sedimenten des Rheins — Veradnderungen seit 1971. Umschau 79 24:778-783. RAT VON SACHVERSTANDIGEN FUR UMWELTFRAGEN. 1976. Umweltprobleme des Rheins. 3. Sondergutachten vom Marz 1976. Stuttgart, Mainz (W. Kohlhammer). Reiss, F. 1968. Okologische und systematische Untersuchungen an Chironomiden des Bodensees. Ein Beitrag zur lakustrischen Chironomidenfauna des nérdlichen Alpenvor- landes. Arch.Hydrobiol. 64:176-323. SONTHEIMER, H., R. GIMBEL AND W. WEIDEL. 1979. Die Rheinwasserqualitat. In: 7. Arbeitstagung der Internationalen Arbeitsgemeinschaft der Wasserwerke im Rheinein- Zugsgebiet, Basel 29-31 Mai, 1979. SONTHEIMER, H., A. GLOECKLER AND W. WEINDEL. 1981. Der Rhein in dem Jahren 1980 und 1981. Arbeitsgemeinschaft Rhein-Wasserserke e.V. Jahresbericht 1981. Karlsruhe. VAN DER VEEN, C. AND J. HUIZENGA. 1980. Combatting river pollution taking the Rhine as an example. Prog. Wat. Tech., 12:1035-1059. Witson, R.S. AND J.D. McGmt. 1979. The use of chironomid pupal exuviae for bio- logical surveillance of water quality. Department of the Environment, Water Data Unit Technical Memorandum No. 18. 1977. A new method of monitoring water quality in a stream receiving sewage effluent, using chironomid pupal exuviae. Water Research 11:959-962. 1982. A practical key to the genera of pupal exuviae of the British Chirono- midae. University of Bristol Printing Office. MEM. AMER. ENT. SOC., 34 x Se ci ey oat - q MeEmorrs OF THE AMERICAN ENTOMOLOGICAL SOCIETY The Cresson Types of Hymenoptera. Ezra T. Cresson. 1916. $42.00. A Venational Study of the Suborder Zygoptera (Odonata), with Keys for the Identifica- tion of Genera. Philip A. Munz. 1919. $10.00, The Blattidae of Panama. Morgan Hebard. 1920. $8.00. The Types of Hymenoptera in the Academy of Natural Sciences of Philadelphia other than those of Ezra T, Cresson, Ezra T, Cresson, 1928, $8.00. The Eumastacinae of Southern Mexico and Central America, James A.G, Rehn and John W.H. Rehn. 1934, $8.00, The Generic Names of the Sphecoid Wasps and Their Type Species. W.S.L. Pate. 1937, $8.00. A Revision of the North American Species Belonging to the Genus Pegomyia, H.C. Huckett, 1941, $6.00, Catalogue and Reclassification of the Nearctic Ichneumonidae, Henry K, Townes, Jr. 1944. $20.00, 13. Elachistidae of North America (Microlepidoptera), Annette F. Braun. 1948. $10.00. Classification of the Blattaria as Indicated by their Wings (Orthoptera), John W.H. Rehn. 1951. $10.00, — A Taxonomic Study of the North American Licinini with Notes on the Old World Species of the Genus Diplocheila Brulle (Coleoptera). George E. Ball. 1959, $15.00. A Taxonomic Study of the Milliped Family Spirobolidae (Diplopoda: Spirobolida). William T. Keeton, 1960, $1£.00, . The Genus Bucculatrix in America North of Mexico (Microlepidoptera). Annette F. Braun. 1964, $15.00. The Butterflies of Libera, Richard M. Fox, Arthur W. Lindsey, Jr., Harry K. Clench and Lee D, Miller, 1965. $12.50. 20. A Revision of the Mexican and Central American Spider Wasps of the Subfamily Pom- pilinae (Hymenoptera; Pompilidae), Howard E. Evans, 1966, $12.50. A Taxonomic and Zoogeographic Survey of the Scarabaeinae of the Antilles (Coleoptera: Scarabacidac). Eric G. Matthews. 1966. $5.00 A Monograph of the Ithomiidae (Lepidoptera) Part Ill. The Tribe Mechanitini Fox. Richard M. Fox. 1967. $9.00. A List of New North American Spiders, 1940-1966. Beatrice R. Vogel. 1967. $9.00. The Higher Classification, Phylogeny and Zoogeography of the Satyridae (Lepidoptera), Lee D, Miller. 1968, $7.00, The Schizopteridae (Hemiptera: Heteroptera) With the Description of New Species from Trinidad. Michael G. Emsley. 1969. $6.50. A Taxonomic Revision of the Aquatic Beetle Genus Laccophilus (Dytiscidae) of North America. James R. Zimmerman. 1970, $12.00. A Revision of the Nearctic Species of the Tribe Parydrini (Diptera: Ephydridae). Philip J, Clausen and Edwin F, Cook. 1971, $7.00. Tischeriidae of America North of Mexico (Microlepidoptera). Annette F. Braun. 1972. $7.00. The Shield-backed Katydids of the Genus Idiostatus. David C. Rentz. 1973. $9.50. The Systematic and Morphology of the Nearctic Species of Diamesa Meigen, 1835 (Diptera: Chironomidae). Dean Cyrus Hansen and Edwin F, Cook. 1976. $10.00. The Stoneflies (Plecoptera) of the Rocky Mountains. Richard W. Baumann, Arden R, Gaufin and Rebecca F, Surdick. 1977. $10.50. The Genus Isoperla (Plecoptera) of Western North America; Holomorphology and Systematics, and a New Stonefly Genus Cascadoperla. Stanley W, Szezytko and Kenneth W. Stewart, 1979, $7.50. ’ Revision of the Milliped Genus Sigmoria (Polydesmida: Xystodesmidae). Rowland M. Shelley, 1981. $11.00 Proceedings of the 8th iniernaiional Symposium on Chironomidae, Jacksonville, Florida, July 25-28, 1982, 1983, $25.00, In making inquiries relative to publications, address: THE AMERICAN ENTOMOLOGICAL SOCIETY 1900 Race Street Philadelphia, PA 19103 eetAwhiwhM VATE OF ISSUE, PLEASE STAMP DATE OF RECEIe 7 AND able IN YOUR LISRARY. MEMOIRS OF THE AMERICAN ENTOMOLOGICAL SOCIETY NUMBER 35 RECONSIDERATION OF THE MILLIPED 3NUS SIGMORIA, WITH A REVISION OF LTOTARIA AND AN ANALYSIS OF THE SENERA IN THE TRIBE APHELORIINI (POLYDESMIDA: XYSTODESMIDABE) By ROWLAND M., SHELLEY DONALD R. WHITEHEAD PUBLISHED BY THE AMERICAN ENTOMOLOGICAL SOCIETY AT THE ACADEMY OF NATURAL SCIENCES PHILADELPHIA ok MAY 1.54 198 \ a “ 1986 ~ LIBRARIES __* MEMOIRS OF THE AMERICAN ENTOMOLOGICAL SOCIETY NUMBER 35 A RECONSIDERATION OF THE MILLIPED GENUS SIGMORIA, WITH A REVISION OF DELTOTARIA AND AN ANALYSIS OF THE GENERA IN THE TRIBE APHELORIINI (POLYDESMIDA: XYSTODESMIDAEB) By ROWLAND M. SHELLEY DONALD R. WHITEHEAD MUMIA Ay, Wow Wy An! My > S Gt OMO Te PUBLISHED BY THE AMERICAN ENTOMOLOGICAL SOCIETY AT THE ACADEMY OF NATURAL SCIENCES PHILADELPHIA 1986 SELWYN S. ROBACK EDITOR (Issued April 4, 1986) PRINTED IN THE UNITED STATES OF AMERICA TABLE OF CONTENTS CC TT PART I. REVISION OF SIGMORIA, by Rowland M. Shelley ....... MaxOnomicihanacters cy ac ase cause och GG Coie Cte snes ces w ae ls Genusps7o71onig Chambenlimrrrr. sense de des ase os ca seses a. INGVaOISUDE EM Chamammneren tat oars er oes eae ceed esac os IXGS7 HO SJOSSIES 5.0.0. 6% wre bes pS Bore idig BOTS GP oNMLIe Rie a aS iene eae SubgenusyRu@iHOrialCauseyae sss soo c hs oe es cone sete ees SubgenussDixionia' Chamberlin. .0 055. en. cone see oe a. Subgenus) Sie7oria Chamberlinvan ane ceases sss soso sees Subgenus;@rogrania Shelley anaes. soe oe ewe eno e cr oe ae Subgenusi@/enroriaiGhamberlinuyneee se) ose odes ese ane Subsenus@neiopus OOMISeraanes doe se ae sone eee ee ea. Subpenusis/e171a ChambeniMernnn see ce ec daa ee ac Subgenuss-avoriaiilotimaneecae acess jose c sree nas ee ac PART II. REVISION OF DELTOTARIA, by Rowland M. Shelley ... axonomici@hanactensrrna. sae core oes see ce aoe ere low aa ees Genus Velfotariai\Causey nec cee oe ee oe heise oh we wid seen PART III. ECOLOGY, by Rowland M. Shelley.................... PART IV. DISTRIBUTION, by Rowland M. Shelley............... PART V. COLORS AND COLOR PATTERNS, by Rowland M. Shelley and Donald R. Whitehead ......................... PART VI. RELATIONSHIPS WITHIN SIGMORIA, by Donald R. Whitehead and Rowland M. Shelley .............. Analy ticalvenoplemSiericwitcrcuct aie oscars che eesce Siore sana, esansted | aoa WEISS CM Chale ere a bes ute be cues eau teawetnoued ahnsteain esis sneraiereae IRB OVE. 8s 6,68 igo BO k 6 OBS 0 Po Oe OLIN EC eee rar LIBRO s 6-6 3 ANS Bd S58 OGIO BOO OOO Re PRL CECE eee arora SIMONE 5 806 66 0. FA AALS TOE OOOO CO ee en eee Ree (CHOGITUEOS 5 Sono Bola HOOT COMO OO ao OTe (UADIOHD 6-6. 015.0 bio PES DS Ee SO ee ne re (QUT ODUS ooo 0 0:6. UH Hone Bro Oo ErOte Oso RCE Oe OER eae a ee ners SUBTLE S' SS 6 BLK CED DS OTE ON Ee OE a er ore FONORD 5 652608 82 G08 & S'O-G O60 6 WG OO OTE E SORE ECOM RTC ee renee Subgenenicdrelationshipsmersarciricee erieiea siecle Sak sete sc teucl emer Problems with the Preliminary Subgeneric Hypothesis .......... MEM. AMER. ENT. SOC., 35 Resolution of the Problems.....................e cece cece wees 197 Relationships Among the Subgenera ......................... 199 DIsCUSSION a..c5d.ndi) dele se Go ee 201 PART VII. THE GENERA OF THE TRIBE APHELORIINI, by Rowland M. Shelley .........2..4.. ae eee 205 Tribe Aphelorimty 0.2.5... icns0ntesc vio sarees ata ek Se eee 205 Genus Alpheloria Chamberlinis. 4506 40 oe ee eee 207 Genus) Srachoria Chamberline.-) epee 210 Genus) 2ynoriag Chambering =.-5-- ee eee eee 210 GenusiEurcillaria’ Shelley nasser eee Eee 211 PART VIII. RELATIONSHIPS IN THE TRIBE APHELORIINI, by Rowland M. Shelley and Donald R. Whitehead............ 212 The Generic Hypothesis:. 2222.55.03 ose eee eee 212 The Tribal Hypothesis: «25224: 3 so sear aoe eee 214 DISCUSSION 6 oescc a2, cnegacena 19,005,' Seem cece EE Re ORR Cane ene 216 LITERATURE CITED |. 3.o2.0.0406 5 6 S06,c00 60 Suchen Gee eee 218 INDEX TO TAXA . 06 305 acs os, a be oly Oe ee 222 MEMOIRS OF THE AMERICAN ENTOMOLOGICAL SOCIETY NUMBER 35 A RECONSIDERATION OF THE MILLIPED GENUS SIGMORIA, WITH A REVISION OF DELTOTARIA AND AN ANALYSIS OF THE GENERA IN THE TRIBE APHELORIINI (POLYDESMIDA: XYSTODESMIDAE) BY ROWLAND M. SHELLEY’ AND DONALD R. WHITEHEAD? INTRODUCTION This paper reviews the genera in the xystodesmid milliped tribe Apheloriini (Xystodesminae) and provides diagnostic statements for those deemed valid. For Sigmoria, which is considered valid since the concepts of the older genus-group names, Apheloria and Brachoria, have not been critically analyzed, we present an expanded diagnosis, descriptions of 10 new species, redescriptions of five species, supplemental accounts of 15 species transferred into the genus, and additional information on distribu- tion, ecology, and relationships. Sigmoria now consists of 65 species, with three divided into a total of nine subspecies. For Deltotaria we present a diagnosis, accounts of the gonopods of the two species in ‘‘sigmoid”’ ter- ‘North Carolina State Museum of Natural History, P.O. Box 27647, Raleigh, North Carolina 27611. 2Systematic Entomology Laboratory, IIBIII, Agricultural Research Service, USDA, c/o U.S. National Museum, NHB 168, Washington, DC 20560. 1 MEM. AMER. ENT. SOC., 35 2 XYSTODESMID MILLIPEDS minology’, diagnoses of the two subspecies of the type species, and distribu- tional and ecological information. The paper concludes with analyses of the Apheloriini and the four principal xystodesmid tribes in eastern North America. We use five taxonomic categories throughout this work in reference to the Apheloriini: genera, subgenera, species groups, species, and subspecies. In various respects their usages are not traditional to xystodesmid classifica- tions, and explanations are in order. Genera are taxa with apparent geographic and lineage independence. Among the Apheloriini, Sigmoria is a cohesive geographic entity and forms a ‘‘mosaic’’ of largely allopatric or parapatric parts (subgenera, species groups, species, and subspecies). The other genera form their own mosaic patterns, which overlie each other and Sigmoria to varying degrees. Accept- ance of this minimal generic concept allows us to remain within traditional limits of xystodesmid classification, because no clear autapomorphies have yet been found among gonopodal or somatic characters for many genera in- cluding Sigmoria, and there may be none. Consequently, it is geographic cohesiveness that gives these genera their apparent individuality or reality, which in turn justifies generic distinction. Deltotaria, however, is defined by an autapomorphy, the coxal apophysis. Thus, Deltotaria and Sigmoria are distinct, sympatric geographic entities, but only the former is defined cladistically. Subgenera are not customarily employed in milliped classifications, and we do not advocate their use. However, our analysis of Sigmoria resulted in the recognition of eight major components, all of which have published genus-group names available. It is convenient to recognize these major en- tities, amd we conserve the senior name at the subgeneric level. The subgenera may have one or more species groups, tend to be allopatric or parapatric, do not show clear patterns of lineage independence, and are dif- ficult to relate in standard cladistic fashion. Lack of lineage independence is reflected by similarities (S3ymplesiomorphies) between proximal elements of adjacent subgenera, which we think reflect ancestral continuities. This pat- tern is a prime reason why the major components are not separate genera; they are sections of a single geographic entity. We are convinced that seven subgenera are monophyletic; Sigiria, however, may be paraphyletic. Begin- >The term ‘‘sigmoid’’ has no taxonomic value and refers collectively to the various apheloriine taxa whose gonopodal acropodites curve in a vaguely sigmoidal fashion. To stand- ardize taxonomic treatments and to facilitate comparisons between species, Shelley (1981a) developed new terminology for these acropodites and divided them into sections or zones. This method of characterizing apheloriine acropodites has been used in subsequent papers and for simplicity is referred to as ‘‘sigmoid’”’ terminology. R.M. SHELLEY & D.R. WHITEHEAD 3 ning with the first subgeneric account, that of Rudiloria, subgeneric assign- ment is indicated throughout the text either before or after the specific name, as for example ‘‘ainsliei (Falloria).’’ When two or more species in se- quence belong to the same subgenus, the designation follows the last — thus “‘nela, brooksi, and coronata (Dixioria).’’ Species groups combine closely related species, some of which might unite as superspecies or even prove to be conspecific. They are geographically coherent and represent our estimates of specific relation- ships. In most cases these groups differ considerably from those proposed by Shelley (1981a). Species are taxa which we hypothesize to be reproductive isolates, ap- parently segregated from geographically proximal taxa and not connected by intergrades. Evidence for separation comes in various forms. Ranges may be so proximal that discontinuities are abrupt, as between pela (Chamberlin) and wrighti Hoffman, and translineata Shelley and lyrea Shelley. Phenotypically different taxa such as brooksi Hoffman and cor- onata (Hoffman) that appear to be syntopic in part of their ranges also are considered reproductively isolated. Additionally, there may be a clear range disjunction between taxa that otherwise could be considered subspecies, the chief example being that of rubromarginata (Bollman) and austrimontis Shelley. Despite their residual intergrades, we consider both full species because genetic interchange is no longer possible; hence, the latter is elevated to specific status. The gap between J/eucostriata Shelley and xerophylla Shelley is occupied by several other taxa, and we consider this further evidence for reproductive isolation. Subspecies are taxa which are reasonably homogeneous throughout their ranges but which connect with other such taxa through intergrade or in- termediate forms. We reserve this category for /atior (Brolemann), trimaculata (Wood), and nigrimontis (Chamberlin), geographically varied but clinally continuous species with uninterrupted ranges where gene flow is still possible. The systematic treatment of Sigmoria in Part I proceeds in the clockwise geographic sequence of north to south to west, both among the subgenera and their component species groups. Species accounts vary. A few new records are presented for those previously discussed (Shelley 1981a), but otherwise they are merely listed. The new species are described; houstoni Chamberlin and mimetica (Chamberlin) are redescribed; and the treatments of the transferred species vary according to the depths of previous accounts, some being redescribed, some having gonopodal descriptions in ‘‘sigmoid”’ terminology, and others merely summarized. A new name is proposed for Sigmoria simplex Shelley, 198la, to avert homonymy with Croatania MEM. AMER. ENT. SOC., 35 4 XYSTODESMID MILLIPEDS simplex Shelley, 1977. Full synonymies are provided for the newly incor- porated species; synonyms of previously included taxa are found in Shelley (1981a). To facilitate identifications we provide anatomical keys to subgenera and species of Sigmoria and a tabular presentation (Table 3) of geographical distributions for species of Sigmoria and Deltotaria. The keys refer to illustrations in this paper plus Shelley (1980a, 198la and b, 1982, 1983a, 1984a); the table includes a color pattern code. Holotypes of all pertinent nominal species were seen except those of Polydesmus trimaculatus (Wood), which is lost; Fontaria lutzi Jacot, for which clear illustrations and reasonably proximal material are available; and Hubroria picapa Keeton, Fontaria pela Chamberlin, and Dixioria dac- tylifera Hoffman, for which paratypes are available. Two requests to the NMNH failed to produce the holotype of H. picapa, which Keeton claimed to have deposited there both in his 1960 paper and in a 1976 letter to the senior author. He kindly loaned paratypes from his collection, and after his death in 1980, these were deposited in the FMNH and the personal collec- tion of R.L. Hoffman. Type specimen notations are always of the senior synonym. Unless otherwise stated in the locality listings, collections were made by the senior author and assistants. The acronym GSMNP denotes the Great Smoky Mountains National Park; acronyms of sources of preserved study material cited in the text are as follows: AMNH — American Museum of Natural History, New York, NY. ANSP — Academy of Natural Sciences, Philadelphia, PA AU — Department of Zoology and Entomology, Auburn University, Auburn, AL FMNH — Field Museum of Natural History, Chicago, IL FSCA — Florida State Collection of Arthropods, Gainesville, FL MCZ — Museum of Comparative Zoology, Harvard University, Cam- bridge, MA NMNH — National Museum of Natural History, Smithsonian In- stitution, Washington, DC NCSM — North Carolina State Museum of Natural History, Raleigh, NC. The invertebrate catalog numbers of material from this col- lection are indicated in parentheses. PMNH — Peabody Museum of Natural History, Yale University, New Haven, CT RLH — Private collection of Richard L. Hoffman, Radford, VA ROM — Royal Ontario Museum, Toronto, Ontario, Canada RVC — Private collection of the late Ralph V. Chamberlin, now being accessioned by the NMNH. —————_— "Ss a sesrtrttSSt”s~—=<; OO!!! R.M. SHELLEY & D.R. WHITEHEAD 5 UMN — Department of Entomology, Fisheries, and Wildlife, University of Minnesota, St. Paul, MN WAS — Private collection of William A. Shear, Hampden-Sydney, VA LITERATURE REVIEW The papers pertaining to the species of Sigmoria in previous works (Shelley 1981a, 1983a) were discussed at those times; here major references on the additional taxa are summarized. The history of Deltotaria was reviewed by Hoffman (1961), the only additional citation being his (1979) inclusion of it in the tribe Apheloriini, where he said it consisted of five species in the southern Appalachian region. The oldest species discussed herein, trimaculata Wood (1864, 1865), is one of four originally placed in Fontaria Gray (1832)*. One other, F. rileyi Bollman (1888), was proposed in the 19th century, and Chamberlin (1918a,b) added F. mimetica and F. pela. In 1921 Chamberlin erected Apheloria for the new species ainsliei and F. montana Bollman, which was designated type species. Both were included under Apheloria by Chamberlin and Hoffman (1958), but Hoffman (1978a) observed that the genus was heterogeneous in its original proposal. Conse- quently, he removed ainsliei from Apheloria but left it unassigned. We agree with Hoffman’s assessment and herewith place the species in Sigmoria. In 1938 Jacot described F. /utzi from Keene, New Hampshire. A year later the number of pertinent nominal taxa increased dramatically, when Chamberlin (1939) proposed A. keuka, here placed in synonymy under the nominate subspecies of trimaculata, and the genera Sigmoria, Sigiria, and Cleptoria, with seven, one, and two species, respectively. Shelley (1981a) dealt with Sigiria and the species originally assigned to it and Sigmoria, and we can now report that C. macra Chamberlin is also properly assigned to the latter because its characters grade into those possessed by more “‘typical’’ forms. Cleptoria therefore becomes a subgenus of Sigmoria. Its second original species was rileyi, which Chamberlin (1939) transferred from Fontaria. The decade of the 1940’s brought more nominal species and the initial ef- fort at reduction through synonymies. Chamberlin (1943) described Sigmoria houstoni from, supposedly, Houston, Texas, but it actually oc- curs in the Cumberland Plateau of southern Tennessee. Loomis (1943) sug- 4 Because of the extreme confusion surrounding this name, we agree with Hoffman (1979) that Fontaria should be placed on the list of officially rejected names in zoology. MEM. AMER. ENT. SOC., 35 6 XYSTODESMID MILLIPEDS gested that C. macra was a synonym of rileyi and expressed doubt about the validity of Cleptoria, a most perceptive insight that we now agree with. Loomis’ insight into relationships showed again in 1944, when he described C. shelfordi. He realized 40 years ago that this form was congeneric with macra, something that neither Hoffman (1967) nor Shelley (1980a, 1981b) perceived. Chamberlin (1947) proposed Dixioria for the new species, den- tifer, which Hoffman (1956a) correctly synonymized with pe/a, in turn transferred from Fontaria. In 1949 Chamberlin described Apheloria tortua; and Hoffman added A. antrostomicola, kleinpeteri, and picta, four Virginia locality records for trimaculata, which Attems (1938) had trans- ferred into Apheloria, and Deltotaria coronata. The decade of the 1950’s featured the Checklist, three papers by Hoff- man, and one by Causey. Hoffman, (1950) summarized Sigmoria, which in- cluded the transfer of mimetica from Fontaria, and remarked that the absence of gonopodal drawings hampered determination of its generic posi- tion. However, he did not provide any, and mimetica still has not been il- lustrated. Hoffman (1951) identified four races of Apheloria trimaculata based on the shape and size of the middorsal spot. Fontaria lutzi was synonymized with the nominate subspecies, and A. antrostomicola and A. tortua were reduced to subspecific status. The fourth subspecies, A. ¢. in- carnata, occurring in Ontario, Canada, was newly diagnosed. Causey’s sole contribution to Sigmoria came in 1955, when she established the genus Rudiloria for the new species mohicana. Hoffman (1956a) transferred Deltotaria coronata into Dixioria and recognized two species: dactylifera, and pela, the latter with six subspecies. Chamberlin and Hoffman’s checklist of the North American fauna (1958) listed without change the species from all previous studies except the last. Keeton (1960) began the ensuing decade with the erection of Hubroria for the new species, picapa. The only other pertinent study of this period was the revision of C/eptoria by Hoffman (1967). He recognized five species — marca; rileyi, with two subspecies; abbotti and bipraesidens, both newly diagnosed; and divergens (Chamberlin) — and transferred shelfordi out of the genus but left it unassigned. Recent contributions to the taxa under study include the diagnosis of Apheloria guyandotta, the transferral of mohicana into that genus, and the suppression of Rudiloria by Shear (1972). Hoffman (1979) assigned Apheloria, Sigmoria, Cleptoria, Dixioria, and Hubroria, to the new tribe Apheloriini along with Croatania, which Shelley (1977) established for four species in the Carolinas. Rudiloria was considered a synonym of Apheloria by Hoffman (1979), the most recent action on that taxon, although Hoff- man (1978a) had revived it for 10 nominal species of Apheloria. The order R.M. SHELLEY & D.R. WHITEHEAD 7 of appearance of these two studies was reversed by publication delays. More recently Shelley (1980a) erected Brevigonus for shelfordi and (1981b) added a second species, arcuatus. Shelley (198la) published a monograph on Sigmoria transferring houstoni and mimetica out of the genus but leaving them unassigned. In the final relevant work, Shelley (1983a) returned divergens to Sigmoria. ACKNOWLEDGMENTS In the nearly ten years that the senior author has been collecting and studying apheloriine millipeds in the southeastern United States, he has been fortunate in receiving assistance from many people. Foremost among them are the following colleagues at major museums, who loaned the in- dicated holotype specimens and other material from the collections under their care: Norman I. Platnick, AMNH, Rudiloria mohicana; Selwyn S. Roback, ANSP, Dixioria dentifer, Deltotaria brimleii, D. brimleardia, and D. tela; Herbert W. Levi, MCZ, Apheloria ainsliei and A. guyandotta; and Jonathan Coddington, NMNH, Apheloria antrostomicola, A. picta, A. kleinpeteri, A. trimaculata incarnata, Cleptoria abbotti, C. bipraesidens, C. rileyi alabama, Deltotaria coronata, Dixioria pela acuminata, D. p. brooksi, D. p. fowleri, D. p. wrighti, Fontaria rileyi, and Phanoria philia. Access to the types of Fontaria mimetica, Cleptoria macra, Sigmoria houstoni, Apheloria tortua and A. keuka in the RVC collection was courtesy of Richard L. Hoffman, who at that time was curating this collec- tion and who also loaned his extensive and invaluable collection of apheloriine millipeds. William A. Shear donated his personal collection, and the following curators kindly provided material from institutional holdings: Michael L. Williams (AU), Larry E. Watrous (FMNH), Howard V. Weems, Jr., (FSCA), Charles L. Remington (PMNH), David Barr (ROM), and Philip J. Clausen (UMN). We are indebted to the National Park Service, United States Department of the Interior, for cooperating with this research by allowing the senior author to collect in the Great Smoky Mountains National Park for seven years. Most of the material from the Park was discussed in Shelley (1981a), but four more species are reported here. The National Audubon Society granted permission to sample in Four Holes Swamp Sanctuary, which resulted in the easternmost records of simplex. The following offices issued permits that allowed collection in the indicated state parks: Division of State Parks, North Carolina Depart- ment of Natural and Economic Resources (Mt. Jefferson State Park); Divi- sion of State Parks, South Carolina Department of Parks, Recreation, and MEM. AMER. ENT. SOC., 35 8 XYSTODESMID MILLIPEDS Tourism (Santee State Park); Tennessee Department of Conservation (Cedars of Lebanon, Edgar Evins, Fall Creek Falls, Henry Horton, and Lake Radnor State Parks); the Parks and Historic Sites Division, Georgia Department of Natural Resources (Augusta Canal, Crooked River, Hard Labor Creek, Kolomoki Mounds, and Mistletoe State Parks); the State Parks Division, Alabama Department of Conservation and Natural Resources (Chattahoochee and Chewacla State Parks); and the Division of Recreation and Parks, Florida Department of Natural Resources (Torreya State Park). Cathy Wood typed and retyped the manuscript, and Renaldo G. Kuhler, NCSM scientific illustrator, prepared Figures 5, 14, 19, 25, 28, 32, 38, 42, 46, 61, 64, 70, 74, 78, 85, 89, 93, 98, 100, 106, 110, 116, 122, 128, 138, and 142. Henrik Enghoff (Zoologisk Museum, Copenhagen) and Peter Mundel (NMNH) provided valuable criticism throughout the manuscript; Mary Mikevich and Michael Schauff (Systematic Entomology Laboratory) added comments on the chapters on relationships. Lastly, the senior author wishes to express his deep gratitude to certain colleagues at the NCSM, who have both supported and encouraged his basic taxonomic research on millipeds throughout his employment. Financial support for this research was rendered in part by the National Science Foundation Grants Nos. DEB 7702596 and 8200556, and was used to fund necessary travel and sub- sistence, employment of field assistants, and publication costs. PART I. REVISION OF SIGMORIA by Rowland M. Shelley Taxonomic Characters The taxonomically important characters of Sigmoria are as listed previously (198la), but with the inclusion of more species, they require elaboration. Most of these characters are found on the male gonopods, and in much of the eastern United States collection of males is mandatory for positive generic determinations. One exception is the northeast (New England, New York, Canada, and northern Pennsylvania and New Jersey), where ¢rimaculata and Apheloria corrugata (Wood) are the only two apheloriine species. They can be distinguished by color pattern, the former having yellow middorsal spots and the latter displaying yellow metatergal stripes. Coloration.— Color and color pattern are useful clues to the identity of an individual of either sex, and I (1981a) presented a table that now must be enlarged (Table 1). As before, all forms display a black metatergal base color and colored paranota, but a middorsal spot pattern is added to the R.M. SHELLEY & D.R. WHITEHEAD TABLE 1. Pattern Paranotal spots only Paranotal and mid-dorsal spots Paranotal spots and metatergal stripes Color red, orange, or yellow red only yellow only both yellow, interior of latter sometimes tinted with red both yellow to orange both lemon yellow both orange both red both violet or purple both white or light yellow paranota red, stripes blue paranota red, stripes white *taxa that are polymorphic for color Colors and color patterns of species and subspecies of Sigmoria Taxa stenoloba, latior latior, some Iatior intergrades*, simplex* australis*, macra, shelfordi, arcuata, robusta, abbotti, rileyi, saluda catawba, simplex*, pela, dactylifera, brooksi, acuminata, coronata, watauga, wrighti guyandotta, trimaculata inornata*, areolata, latior munda*, some latior inter- grades*, simplex* whiteheadi planca inornata*, truncata, sigiri- oides, quadrata, lati- curvosa, stibarophalla, latior munda*, latior hoff- mani, some latior inter- grades*, rubromarginata, triangulata, nigrimontis, australis*, divergens, austrimontis, persica, serrata, bipraesidens, haerens, divaricata, thrinax stenogon, disjuncta leucostriata fumimontis, bidens, trans- lineata, lyrea, tuberosa, xerophylla, ainsliei, aphe- lorioides, prolata, mime- tica, crassicurvosa, pendu- lata, picapa, forficata, houstoni, abbreviata nantahalae MEM. AMER. ENT. SOC., 35 10 XYSTODESMID MILLIPEDS several stripe and paranotal spot patterns of before. The last category is also expanded, since some species display only red or yellow paranota in addi- tion to the highly variable colors of /atior latior and stenoloba. In addition to the color variation sometimes seen along the inner margin of the paranotal spots and metatergal stripes, the size of the middorsal spots in guyandotta and trimaculata varies from small discrete dots, as found in northern populations of the latter, to large semilunar blotches that may even spread laterad into diffuse, indistinct stripes connecting with the paranota markings. To avoid confusion, these species are listed only in the ‘‘paranotal and middorsal spots’’ category, but. this variation should be noted because it demonstrates how spot and stripe patterns relate to each other. Table 1 summarizes the most common colors and color patterns for each species and subspecies of Sigmoria, omitting occasional variations ex- hibited by such taxa as bidens, translineata, divergens, ainsliei, mimetica, and pendulata. Taxa that are commonly polymorphic for color and/or pat- tern are indicated by an asterisk. A clear paranotal spot and metatergal stripe pattern is evident on preserved material of mohicana and rigida, but since the colors are unknown, they are omitted. I have not personally col- lected them, and there is no indication of living colors in either the literature or on labels with the specimens. The various colors and patterns tend to cluster throughout the range of Sigmoria and also within that of the Apheloriini (Figs. 148-150, 159). These phenomena are discussed in Part V. Sterna.— Except for the more densely pilose condition in tuberosa, the only aspects of the sterna that aid in determinations are the configuration and degree of development of the process of the 4th sternum and, to a lesser extent, those of the 5th. Variation of the processes is as described previously (Shelley 1981a), and Table 2 presents updated information on that of the 4th sternum. The newly included species add to each of the three columns to bring the totals to 37 species in which the process is shorter than the widths of the adjacent sterna, 19 in which they are subequal, and 9 in which the projection is longer. Gonopodal Characters.— Most of the specific characters in Sigmoria obtain in aspects of the male gonopods. The newly included species increase the scope of variation and require supplemental accounts for each zone of the acropodite plus ones on thickness, the in situ arrangements, and the prefemoral process. 1. In situ arrangements.— The in situ arrangements of the newly in- cluded species fit, for the most part, the overlapping, crossing, and parallel patterns described previously, although a telopodite will occasionally be dislodged giving an atypical arrangement. The only new configurations oc- a TABLE 2. Shorter (than widths of adjacent coxae) latior stenoloba areolata stibarophalla inornata truncata sigirioides stenogon nantahalae leucostriata xerophylla rubromarginata austrimontis triangulata whiteheadi nigrimontis disjuncta prolata mimetica crassicurvosa pendulata pDicapa abbreviata houstoni australis simplex divergens dactylifera mohicana rigida guyandotta planca persica agrestis serrata divaricata thrinax MEM. AMER. ENT. SOC., 35 R.M. SHELLEY & D.R. WHITEHEAD Subequal quadrata laticurvosa translineata lyrea fumimontis ainsliei aphelorioides bidens macra rileyi shelfordi pela brooksi acuminata coronata wrighti watauga haerens trimaculata Length of process of 4th sternum in males in species of Sigmoria Longer tuberosa catawba saluda yemassee forficata abbotti arcuata robusta bipraesidens 11 12 XYSTODESMID MILLIPEDS cur in ainsliei, aphelorioides, and the subgenera Rudiloria and Dixioria. In ainsliei the acropodites intertwine (Fig. 89), which is really a manifestation of the overlapping pattern caused by the circular configuration. Conversely, in most males of aphelorioides the acropodites lie above and below each other (Fig. 85) and rarely overlap. In the subgenus Rudiloria the telopodites are oriented differently on the coxa and face mediad. Here and in Dixioria they are directed transversely across the body and tend to insert in the op- posite side of the aperture instead of extending over the anterior margin and lying between the 7th legs. This arrangement is shown in Figures 5 and 14, in Shear (1972, Fig. 1) for guyandotta, and in Hoffman (1978, Fig. 2) for t. trimaculata. 2. Prefemoral process.— With more species and more diversity, the prefemoral process loses all effectiveness as a specific criterion. A basally divided process occurs in four species rather than two (ainsliei, forficata, translineata, and lyrea), and the configuration in bidens is approached in proximal populations of prolata. Thus, even in these cases, determinations based solely on the process are inadvisable. Moreover, species are now known without a process or with only a vestige — pendulata, bipraesidens, rileyi, guyandotta and mohicana. 3. Characters of the acropodite. a) Thickness.— The newly added species included those with the heaviest, most massive gonopods in the genus (rileyi, abbotti, bipraesidens, robusta, crassicurvosa, and pendulata). In a few, such as mohicana, trimaculata, pela, and wrighti, the acropodites are rather thin and fragile, and similar to those of stenogon and Ssigirioides. The others fall between these extremes. b) General curvature. — It is in general curvature that the newly added species depart the most from those previously included in Sigmoria, and the generic concept must now be broadened to encompass forms whose acropodites circumscribe a complete or nearly complete loop. Three species are involved: ainsliei, aphelorioides, and trimaculata. In these the anterior bend and apical curve are more or less continuous through the peak, so that the acropodal regions are obscure. The overall configuration is of a smoothly continuous circle, with the tip overlapping or nearly overlapping the basal zone in medial view. Futhermore, many of the newly added species do not possess coplanar basal and distal zones. The latter region angles generally laterad from the peak, and is coplanar with the peak in houstoni and picapa. Heretofore, only stibarophalla, disjuncta, and divergens had laterally directed distal zones, but this character is also displayed by australis, prolata, mohicana, the subgenus Dixioria, and certain species of the translineata group of the subgenus Falloria. R.M. SHELLEY & D.R. WHITEHEAD 13 c) Basal zone. — The basal zone has more taxonomic significance than formerly thought. Previously, the tubercles of tuberosa were the only modifications of the outer surface of this region, but with the inclusion of the southeastern lowland forms, more specializations are known. The medial edge of the basal zone is deeply emarginate and notched in catawba and saluda and sinuous to shallowly notched in simplex and yemassee, and the basalmost projection is enlarged in the first three. Sigmoria arcuata, rileyi, abbotti, robusta, and shelfordi have a single basal projection that ranges from a large spur to a strong, distinct spine, which may be homologous with the basalmost projection in catawba, saluda, and simplex. In the trimaculata group the basal zone is directed mediad rather than anteriomediad. d) Anterior bend. — The anterior bend varies from well defined, in species like brooksi, rigida, and australis, to broad and poorly defined in those with circular acropodites. e) Peak. — The peak is tilted laterad in the mimetica group of Falloria, exposing the undersurface in medial view. It is also much thicker in those species with heavy, massive acropodites. Jf) Apical curve and distal zone. — As stated earlier, more species are now known with laterally directed distal zones that are not coplanar with basal zones. In houstoni and picapa the distal zone is actually coplanar with the peak. Extremely short distal zones, coplanar with the basal zones and directed perpendicularly from the peak, are found in five species of the subgenus Cleptoria. The opposite extreme is found in ainsliei and trimaculata, where the region completes the loop of the acropodite and overlaps the basal zone. Sigmoria shelfordi joins truncata in lacking this section, the acropodite terminating at the distal extremity of the peak. Ad- ditional lobes and flanges are found on the distal zones of pendulata and crassicurvosa. g) Tip. — Except for variants of such species as aphelorioides and ar- cuata, which have reflexed tips, all the newly added taxa have simple, blunt to acute, terminations. h) Medial flange. — Considerable additional variation is known for the medial flange. It can be located much more proximally and arise well onto the basal zone. In the forms of Cleptoria with heavy acropodites it loses some of the laminate character, becoming correspondingly thick and heavy. The ultimate condition obtains in bipraesidens, where the flange is not a lamina but a thickening or swelling of the entire medial surface of the peak with the general contours of the flange in ri/eyi when viewed ventrally. In the mimetica group the flange is narrow and the peak is tilted laterad, so that in medial view the undersurface is visible and the flange is obscure. MEM. AMER. ENT. SOC., 35 14 XYSTODESMID MILLIPEDS Likewise, the lamella is obscured in medial view in the subgenus Rudiloria, because the basal zone which carries it is directed mediad. Sigmoria ainsliei, aphelorioides, and nantahalae and the subgenus Dixioria lack the flange. i) Tooth. — Additional species with a tooth on the medial margin of the acropodite include rigida and the seven species of the subgenus Dixioria. Others, like rileyi, robusta, catawba, and saluda have broadly rounded or triangular distal lobes on the flange, which may represent the tooth. A small, sharp accessory tooth arising from the under surface of the acropodite at or near the location of the marginal tooth is also present in all species of Dixioria except dactylifera. The tooth or teeth are usually located on the peak but may also be on the distal zone. J) Lateral flange. — The lateral flange is absent from most newly added species, and in most of the others it is not laminate. In the southeastern species with heavy acropodites (rileyi, robusta, abbotti, catawba and saluda) it is more lobe-like, more ventrad than laterad, and poorly demar- cated from the acropodite stem in the first three, where it forms the highest point of the acropodite arch. k) Prostatic groove. — The prostatic groove crosses from the medial to the lateral sides of the acropodite at various positions depending on how the telopodite is situated on the coxa and whether or not the peak is tilted. !) Solenomerite. — In most species of Sigmoria the prostatic groove runs along the main stem of the acropodite, avoiding flanges, lobes, and other adornments, and opens terminally at the tip of the distal zone. In the subgenus Cheiropus, however, the acropodite is either apically divided or there is a small subterminal structure on which the groove opens. This struc- ture, or the terminal division carrying the groove, is called the solenomerite, and its position and configuration are taxonomically important. Genus SIGMORIA Chamberlin Previous gonopodal descriptions and that of the 6th sternum (Shelley 198la, 1983a) are rewritten to accommodate the newly incorporated species. Other somatic features are unchanged. Diagnosis. — Coxa without apophysis; acropodite curving in variable sigmoid to coiled or circular configurations, with or without variable flanges, lobes, or teeth on different zones, but always without a transverse groove or cingulum on outer surface. Description. — Sternum of segment 6 with or without variable convex recession or excava- tion between 7th legs to accommodate apical curvatures of acropodites. Gonopods in situ with variable configurations; usually with each acropodite extending mediad beyond midline of aperture and over- or under-lapping other acropodite proximal to midlength in midline, curving slightly anteriolaterad over opposite side of aperture then back R.M. SHELLEY & D.R. WHITEHEAD 15 over midline with tips curving dorsad and crossing, apical curvatures extending to varying lengths over anterior margin of aperture and sternum of segment 6; some species with acropodites crossing only once in peak or apical curve regions in midline of aperture, apical curves extending anteriad only slightly beyond anterior margin of aperture; others with acropodites not overlapping but medial edges touching or nearly touching in midline and ex- tending anteriad in subparallel arrangement over anterior margin of aperture and sternum of segment 6, apical portions of curvatures in these forms projecting either dorsad or dor- solaterad; still others with acropodites curving dorsad over opposite side of aperture and either touching or approaching opposite coxa and lying in front of and behind each other or curving laterad toward midline and overlapping opposite member. Coxa moderate to large, without apophysis, connected by membrane only, no sternal remnant. Telopodite set terminally on coxa, latter not projecting distad beyond prefemoral region. Prefemur moderate in size, with or without variable prefemoral process, latter ranging from nubbinlike vestige to long, massive structure extending well beyond tip of acropodite, linear to bisinuate, simple or divided basally into two long, nearly equal components, or divided apically and bifurcate. Acropodite thin and fragile to thick and massive, with torsion, curved through one or more vertical planes in vague- ly sigmoidal to loosely coiled or circular configurations as seen in situ and in medial and lateral views, configurations as follows: basal zone extending subventrad from prefemur; bent sharply or broadly anteriad (anterior bend) from 1/4 to 1/2 of total acropodal length; usually curved sharply or broadly dorsad or laterad apically (apical curve) near 2/3-3/4 length forming arc with variable diameter; portion between anterior bend and apical curve (peak of arch) coplanar with basal and distal zones or bowed mediad and not coplanar with latter regions, flattened or high and gently rounded in medial view, extending to or beyond level of prefemoral process; region forming apical curve (distal zone) varying in length, either projecting nearly directly dorsad, curved inward into acropodite arch and essentially coplanar with basal zone, or bent abruptly laterad, curving through more than one vertical plane, and not coplanar with basal zone, occasionally coplanar or nearly coplanar with peak, either of subequal width throughout most of length (except apically) or tapering smoothly and continuously to acuminate tip. Basal zone variable in length, inner surface usually directed anteriad, occasionally mediad, with or without variably laminate flange on medial surface and adornments on outer surface, latter in- cluding densely packed tubercles, a row of dentate spurs, or single basal spines; medial edge usually entire, occasionally variably expanded and finely scalloped or deeply and irregularly notched; lateral edge occasionally expanded and lamellate. Peak of arch variable in length and width, usually with inner surface directed toward arch, occasionally tilted laterad and facing mediad, with or without flange of variable length, width, and configuration on medial edge, arising from proximal part of basal zone to proximal part of peak, usually terminating from near midlength of peak to proximal part of distal zone, with or without variable laminate to spiniform tooth arising from medial edge or outer surface distal to flange, separated to varying lengths from flange, occasionally fused to latter; occasionally with small, sharp accessory tooth on undersurface at level of marginal tooth. Distal zone present or absent; when present with or without variable flanges or lobes proximally on both medial and lateral sides, one or both often arising on distal extremity of peak and distally on medial side; also with or without long, narrow, laminate flange on lateral side, arising near beginning of apical curve, terminating proximal to tip, flange occasionally modified into subtriangular lobe, long narrow subrectangular projection, or greatly enlarged and swollen, forming outermost point on acropodite arch; usually without but occasionally with teeth; when absent sometimes replaced by variably positioned solenomerite, best viewed laterally. Termination variable; either blunt or acuminate unmodified end of peak or distal zone; or small, reflexed lamina on medial edge directed at sharp angle (often perpendicularly) from distal zone; termination occasionally angled from distal zone but not separated into distinct lamina. Prostatic groove arising in pit MEM. AMER. ENT. SOC., 35 16 XYSTODESMID MILLIPEDS on prefemur, running along stem of acropodite and crossing from medial to lateral sides at various locations, opening terminally on tip of distal zone or on reflexed lamina when present. Females agreeing essentially with males in somatic features, except paranota usually more strongly depressed, creating appearance of more highly arched or vaulted body. Distribution. — As currently understood (Fig. 145), Sigmoria occupies a large, contiguous, irregular area and five smaller, disjunct ones. The former extends from Ontario and northern New England to northeastern Florida, and longitudinally from the outer coastal Plain of the Carolinas and Georgia, the Blue Ridge Province of Virginia, and the New England Proy- ince, to piedmont Georgia, the Ridge and Valley Province of Tennessee, and Central Ohio. One disjunct area encompasses most of the Cumberland Plateau,’ eastern Highland Rim, and Nashville Basin of Tennessee.° Another is located mostly in the Coastal Plain of south-central Alabama, ranging northward slightly into the Fall Zone. The third spans the Chat- tahoochee and upper Appalachicola Rivers in the Coastal Plains of south- western Georgia, southeastern Alabama, and the adjacent part of Florida; and the fourth is located in extreme southern Georgia and the northern part of peninsular Florida. The fifth disjunct area, in the Cumberland Plateau of north-central Alabama and shown by the dot, is represented by a single male of rileyi, which became available as the paper was being drafted. The lacunae in Tennessee and those between the more southern disjunct areas are slightly over 40 miles wide and may be sampling artifacts that will even- tually be joined. However, I have not found representatives there in nearly 10 years of field work and show them to accurately depict present knowledge. The blank area in south-central Georgia is around 90 miles in diameter and though nearly surrounded by Sigmoria, seems likely to survive additional field work. Sigmoria thus inhabits seven physiographic provinces and 17 states plus Ontario. The total north-south length is over 1,000 miles, and that from the Atlantic Ocean to central Tennessee is around 550 miles. The Alabama River seems a likely western distributional limit in Alabama. Species. — Sixty-five, with three consisting of nine geographic races. KEY TO SUBGENERA OF SIGMORIA 1. Color of metatergal stripes contrasting with that of paranota or both either white or pale VSM OW: sca: Sisvdccrs, Soe Sede RGA GRU IOC oe es Falloria Hoffman Metaterga either uniformly black, variably spotted middorsally, or with stripes con- colorous with paranota but not white or pale yellow...................-.---- 2 * The local name, Cumberland Plateau, is used in this paper to refer to the southern part of the Appalachian Plateaus Physiographic Province. ° The Nashville Basin and the surrounding Highland Rim are the southernmost sections of the Interior Low Plateaus Physiographic Province. R.M. SHELLEY & D.R. WHITEHEAD 7) 2. Gonopods in situ with acropodites extending mostly mediad and projecting over opposite side of aperture, occasionally with sides tilted and leaning over anterior OEE HD 5G:0 Sit 0.t.0'S 0 0 OU a OIG OES Otel CTO CEES IO EERSTE PRE PRR oan gee 3 Gonopods in situ with acropodites extending mostly anteriad, projecting well beyond anterior margin of aperture and inserting between legs of segment 6........... 4 3. Caudolateral corners of paranota rounded through midbody segments; acropodite usually with accessory tooth arising from undersurface at or distal to level of tooth; metaterga uniformly black, never with spots or stripes........ Dixioria Chamberlin Caudolateral corners of paranota rounded only on anteriormost segments; acropodite without accessory tooth; color variable, metaterga usually with spots and stripes ... 0:0:6:6 8:d:0°0: & eeehidhaic eB balls tol eC ESS CRC ERS cr RE nee on Rudiloria Causey 4. Acropodites either with distal zones curving strongly laterad and obscured in medial view by stems or medial flanges, or with variably positioned solenomerites replacing Gistalizonesfeneme ayy enter een iret siciors oc ivare ainehsuriha rae eee dantsts coaeenae doen 0D B.U:8.0-0.0 1B DRO IG OOO SENG One OOe OCR ERS CIR NCE on ee Neen ee Cheiropus Loomis Acropodites without solenomerites; distal zones visible in medial views............. 5 5. Acropodites usually massive and heavily sclerotized; basal zones usually with variable spines or teeth on outer or medial surface .............. 0... c cece cece ee eens 6 Acropodites thin and fragile to moderate, never massive; basal zones without modifica- (HKOING: 5.9 o.oa'o Helo U 0-0 Sido -ErBAETeCO a ole ath lator care tos hel OEM DnE RCN CRANE ep ERG CT cic ee RE ner 7 6. Basal zones with variable teeth or variably notched expansions on proximomedial edges; prefemoral processes large to enormous, usually extending beyond tips of ACKOPO Gites eer ee Seren awe acess aoa an a cookie ween bands Seema ae Croatania Shelley Basal zones with at most a single spine on outer surfaces; prefemoral processes small to AOS om. ans bora t's Om eed CG O CORIO eon trcitib cee! oectaer ere en earn Cleptoria Chamberlin 7. Acropodites either with reflexed tips or with this general configuration ................ ee eerie ore ctv tele. aesuadate Weedon uaceaeenlegs Sigmoria Chamberlin Tips of acropodites simple, blunt to acuminate.................... Sigiria Chamberlin KEY TO SPECIES OF SIGMORIA (based primarily on adult males) The difficulty of devising a key to the variable species of Sigmoria is greatly magnified by the addition of more congeners than were in the original revision (Shélley 1981a). Even less can now be said about a given species that does not also apply to another, and it is harder to compose couplets so as not to exclude important variants. Qualifying modifiers such as ‘‘usually’’ are therefore found throughout the key, because of the high frequency of exceptional variants. Also necessary are relative comparisons and subjective criteria like thickness, whose utility depends upon one’s ex- perience with ‘‘sigmoid’’ xystodesmids. For these reasons and to facilitate usage, most couplets contain more than one comparative criterion. In the following key the three subdivided species are treated as species, so all subspecies and intergrades key out at the same couplet. Except for lateral flanges, most characters of dissected gonopods are best seen in medial view, MEM. AMER. ENT. SOC., 35 18 XYSTODESMID MILLIPEDS particularly those involving comparisons such as ‘‘...extending beyond level of distal zone.’’ The few other instances when the lateral perspective is desirable are so stated. Figure numbers and generalized range descriptions are given to enhance the key. To avoid repetition and to distinguish them from drawings in this paper, illustrations in my previous revision (Shelley 1981a) are indicated by an asterisk after the number. References are cited for figures in other papers. As all species of Sigmoria display vivid pigmentation, and the colors and patterns vary considerably, a checklist (Table 3) is presented after the key in which the distribution by state and color pattern is tabulated. Thus, where locality and color are known, determinations may be easier with this table than with the anatomical key. The two complement each other nicely, and the forms of De/totaria are also included in the table for the sake of comple- tion. 1. Gonopods with variably positioned solenomerites, best visible in lateral perspective ee eT een ne eee ee RAS OO Ah aT OM OH oo 0000000006 59 Without thisicharacter sc... 2 Ake Rae Seg ed a ae ea ea 2) 2. Gonopods with variable medial flanges, occasionally vestigial.................... 12 Without this: character gs...oc:24:< macs sees aoe Gears 5) oes oe emo oon te 3 3. Acropodite configuration circular, distal zone curving broadly into arch in medial view, extending beyond prefemoral process and overlapping or nearly overlapping basal zone; toothiabsent: s..8.ei ae Sa ee Oe eee 4 Acropodite configuration otherwise, either resembling number 7, an inverted L, or curving broadly and continuously to but not beyond level of prefemoral process; {R010} 81 6 0) R~) || Men Rene een ene ee eee ee Pere oI ata cn og co edd 6.0 000-6 5 4. Prefemoral process divided basally into two long, unequal components; distal zone coplanar with basal zone (Figs. 90-91); Knox, Sevier, and Blount cos, TN ......... eee a tr eens earn cece re ceaaiciao: Gok our a acronis oto Ao b.0° G0 ainsliei (Chamberlin) Prefemoral process simple, upright, and acuminate; distal zone curving sublaterad from peak, not coplanar with other sections; (Figs. 86-87); Swain Co. NC, to McMinn Oi TIN ee io ccc Ra NE (CPE SOE TOT SES PIC o SPE aphelorioides, new species 5. Acropodite with minute, sharply pointed accessory tooth arising from under surface of acropodite stem or main tooth at or distal to latter......................-4-- 6 Without thisicharacter ..2:. s1<.h. escsad ura paste ara Ses san een Ore Oe eee 11 6. Outer margin of distal extremity of peak greatly expanded into broadly rounded lobe; accessory tooth arising from base of main tooth; prostatic groove crossing to lateral surface at anterior bend; latter sharp, well defined (Figs. 39-40); Washington Co., VA; and’Johnson' Coss RNase sete con oe eee brooksi (Hoffman) Outer margin of distal extremity of peak at most only faintly expanded, not lobe-like; accessory tooth arising from acropodite stem; anterior bend broad, poorly defined; prostatic groove crossing to lateral side on prefemur (Figs. 20-22, 26, 29-30, 33-36) ee ee ee ee ea anna oro tia ria ecremes aie oso ORO .0.0.0.0.0000 7 7. Acropodite configuration a flattened, open arch, overhanging and extending slightly be- yond level of prefemoral process; peak linear; distal zone short, terminating well above level of prefemoral process (Figs. 26, 29-30)...............-0eeeeeeeee 8 ee a es 10. 11. 12. 13. 14. 15. 16. 17. R.M. SHELLEY & D.R. WHITEHEAD 19 Acropodite configuration a smooth, continuous curve, arch overhanging and extending slightly beyond level of prefemoral process; peak gently curved and rounded, apex at midlength; distal zone relatively long, extending to or near level of prefemoral PROCESS (HISS820222933-36))am.ce cic dia cloves od ehcic) easyer Seacacl Siaiei Gveveveneva,ntinvernn ene de 9 Tooth present (Figs. 29-30); Tazewell Co., VA, to Watauga Co., NC.................. 5°610-0°0-995.0 6 Dedaphalo: Oty OS SLE ORMONO G-OueT CEnLcoes CROONER TCP NE met ee coronata (Hoffman) Tooth absent (Figs. 26-27); Johnson Co., TN................... acuminata (Hoffman) Accessory tooth arising distal to tooth, fully visible in medial view (Figs. 35-36); Watauga ANGUWATKESTCOSE IN Carey eae woh eh vores arsue a eteeat awed Guekcusveetco ae watauga, new species Accessory tooth arising at same level as tooth (when present), partly obscured in medial VIC WACRISSHe2 02226 35 -SA)I ay water wats (enck-espccey ota decascteeeyapeyalsieue tous Sincere create a sisi & 10 Tooth always present, quadriform, apically dentate (Figs. 33-34); Johnson Co., TN, to Galdwell(ComING cia sel-wrcets tse hacrckcna trast eaves ae ensues wrighti (Hoffman) Tooth present or absent, triangular to spiniform (Figs. 20-24); Carter Co., TN, to Mc- OW el NC OMIIN Cra errant See eoroo caine Momeni wieccbaunuctaas vonsnduonsilics pela (Chamberlin) Tooth large, extending in medial view well below level of distal zone and nearly to that of prefemoral process, inner apical corner produced, extending into arch; lateral flange absent (Figs. 43-44); corners of paranota rounded beyond midlength; paranota yellow, metaterga black, without stripes; Ashe Co., NC ...............--000000: hD'O 10:0 8 Oia 9 a ERGO A 0:0 OE OT Ger Te ORE RER ROCCE ECan dactylifera (Hoffman) Tooth subequal in length to that of distal zone, but short, barely projecting below peak, terminating well above level of prefemoral process; lateral flange present (Figs. 63-64*); corners of paranota rounded only in anterior half of body; paranota red, metatergal stripes white; Swain Co., NC, to Towns and Unioncos.,GA .......... B20. BOO Abita Cen at CLE aL0 DEO Ory GEOR PRC ORDR SET CHELERT Fac? mre RRC ee nantahalae Hoffman Prefemoral process usually present; occasionally vestigial....................... 18 Prefemoral process usually absent ............. 0. cece cee cece eet eee eters 13 Acropodite of normal or moderate thickness or thin and fragile, latter poorly sclerotized eT eR ere ee ee earthly a ouanv ay seaueprrdacians anatiolavaacarents 14 Acropodite thick and massive, heavily sclerotized ..................0eceeeeeeees 16 Basal zone with basal spine on outer surface (Shelley 1981b, Figs. 3-5); Pickens to Abbe- WU EOS.5 NOws 5 oes ae OO oe aoe Rene INES Sota Aigner ome eeceree cares arcuata (Shelley) Writhoutithisichanactenerty-sasics.teschsereie ajo o aren oetevete ote ai Sleeves iris atau a soayerscaliey sue 15 Distal zone flared, lamellate (Fig. 9); Wyoming Co., WV to Menifee Co., KY .......... 50D 45100 OG ATOLD DAO AS OATS BOTS Ot OO OOO Oe Eee ee guyandotta (Shear) Distal zone of nearly equal width throughout except for tip, without lamellae (Fig. 11); NShlandi€oriOHer es oncconsniiee krone on anced anemonmyaw s mohicana (Causey) Peak flat and relatively long, about 1/3 of acropodite length; distal zone relatively short, terminating well above level of prefemur, directed perpendicularly from peak, form- ing rectangular acropodal arch (Figs. 50-51, 57-58) ...........-.--2+---0005- 17 Peak short, high, and rounded; distal zone relatively long, terminating at level of pre- femoral process, arch an inverted U (Fig. 123); Putnam and Dekalb cos., TN ...... D-AiaI0.0 ARS Otol Ob. A BOLO Ot Ona EOS Gl DEAS OIE TOTNES RC EES ee mene ere pendulata, new species Distal zone relatively broad, sides curving and converging to blunt tip; medial flange thick but distinctly laminate (Figs. 50-51); Clarke Co., GA, to Lee and Jefferson COS Ae peg oe acne ae rest age ate a eam oA vans SiO Ghemrone cena eraal etn gs aysiese le’ rileyi (Bollman) Distal zone relatively narrow, inner edge straight, outer curving inward to acuminate tip; medial flange represented by thickening of entire medial surface of peak (Figs. S58) aACKSOM(COn GAs yo cicosess- epeve sue eo eveuate ovaierevetnouels bipraesidens (Hoffman) MEM. AMER. ENT. SOC., 35 20 18. 19. 20. 21. Web 23. 24. MSc 26. Ail 28. XYSTODESMID MILLIPEDS Distal zone and apical curve absent; acropodite terminating at distal extremity of peak Ba ee ee ern orien am steers DE iD AoE Oc BNO GOW 6.0-9°0'0.d 0-0 0 19 Distal zone and apical curve present, length and configuration variable............ 20 Medial flange arising on basal zone and terminating on distal extremity of peak; acropo- dite relatively thick and heavy, with or without proximal spine on outer edge of basal zone, with or without short, acute spur on medial or outer surfaces of peak (Shelley 1980a, Figs. 7-12); Oconee to McCormick cos.,SC ..................... FR es rare ene yar ee acest cae ear es oarcotoros ruc Gersarentomaaare dict, clacornicctr a 6 shelfordi (Loomis) Medial flange arising on proximal part of peak and terminating just proximal to tip; acropodite relatively thin and blade-like, without spines or spurs (Figs. 37-38*); Yancey and Mitchell cos., NC ........... cece cece eee teen ees truncata Shelley Basal zone with teeth, spines, and/or tubercles on medial or outer surfaces (Figs. 47-48, 53-54, 62-63, 92-93*; Shelley 1977, Figs. 1, 3-4,6)...............c.0eeseeee 21 Without these features, surfaces of basal zone smooth..................-...-.:- Di Proximomedial edge of basal zone expanded and irregular, with small denticulations to sharplysacuteiteethiandispinesi sneer een ener rcenicnnicneicnerneat 22 Medial edge of basal zone smooth and entire, not expanded ..................... DS) Prefemoral process terminating below level of distal zone (Figs. 47-48); Chester to Berkeley cosiviS@s nines acetone eat eee ee simplex (Shelley) Prefemoral process large, extending to or beyond level of distal zone (Shelley, 1977, Figs. | NS) ee ee ear ere are ene Ant nunnery teh aed din'c10 d.0-0-0.0 0 ¢ 23 Prefemoral process apically bifurcate (Shelley 1977, Fig. 3); Laurens to Aiken cos., SC Pe mera a rina oi hokcanecanteh ao Ader OM eRaain cide chao DIOATT OIG SOW od saluda (Shelley) Prefemoral process simple, not bifurcate .............. 0c ccc cence eee ene teen ae 24 Prefemoral process of approximately uniform width except near tip, bisinuately curved; proximomedial expansion of basal zone small and inconspicuous, with only small denticulations (Shelley 1977, Fig. 6); Beaufort and Jasper cos.,SC ............... Pe ar earn aR ee ene China narrate tah aici midis ti aiaio, G yemassee (Shelley) Prefemoral process subglobose basally, distal 1/3 narrower and bent into arch; proximo- medial expansion broad, irregularly notched (Shelley 1977, Fig. 1); Lincoln Co., NG sto'Chester'CoipS@nd i ee ee ee ee catawba (Shelley) Outer margin of basal zone with dense cluster of minute tubercles (Figs. 92-93*); Swain Cos NGee eae i Se ee eee tuberosa (Shelley) Outer margin of basal zone with basal, acute spine................00cceeeneeeue 26 Distal zone with outer margin continuous with lateral flange (lobe), without indentation, sides broadly rounded and converging to blunt tip; medial flange conspicuous, ter- minating in rounded lobe (Figs. 62-63); Oconee and Anderson cos.,SC ........... SDS ER a SRS Fe ae Sec Se robusta, new species Distal zone sharply demarcated from lateral flange (lobe) by deep indentation, sides gently curved and converging to acuminate tip; medial flange inconspicuous, without distal lobe (Figs. 53-54); Hart to Burkecos.,GA....... abbotti (Hoffman) Acropodite with peak tilted laterad, exposing under-surface in medial view; course of prostatic groove visible in medial view up to distal extremity of peak or proximal part of distal zone (Figs. 111,117, 120))........................00--seeees 28 Orientation of peak normal or tilted mediad ................. 0.0.2 e eee e eee eeee 29 Acropodite massive, heavily sclerotized; distal zone with broadly rounded lobe at mid- length on medial surface or with terminal medial lamina; tip a short, acuminate, and laminate projection from apex of distal zone (Figs. 117-120); Smith and Dekalb COSH TING xc ck eae ae ie ene aoe crassicurvosa, new species a 29. 30. 31. 32. 33. 34. 35. 36. Sr R.M. SHELLEY & D.R. WHITEHEAD 21 Acropodite moderately thick and heavy; distal zone without modifications; inner corner produced to form tip (Figs. 111-112); Robertson to Maurycos.,TN.............. 0.01050 £:0.0/0-0°8:0,0.5 OOOO PAG. Can DRORECECrOND Die CrOn ene Cnet ment mimetica (Chamberlin) Prefemoral process divided from base to midlength into two long, prominent, nearly equal components (Figs. 94-96, 99-100, 68-69*, 72-73*)................-000- 30 Prefemoral process with or without subterminal spurs or at most only shallowly or apical- ly bifurcate, components small and usually unequal........................ 33 Distal zone long, curving broadly well below level of uppermost component and nearly to level of prefemoral division; tip pseudoreflexed (Figs. 72-73*); Swain Co., NC, and Blount Cor aeINeseraen eres err tercmeciewe oe seesed apa ketal wna aia nane lyrea Shelley Distal zone short, terminating above prefemoral components; tip blunt or acuminate 9. 0-6'0 1'6.3:9:0.9.6-010.0 7 UDO OOOO URE OORT EIS eer RC CRT Tene rien Caen ane eT 31 Acropodite with rounded lobe on lateral edge at apical curve, peak tilted mediad exposing lateral margin in medial view; distal zone coplanar with basal zone; tip blunt (Figs. 68-69*); Swain Co., NC, and Blount Co., TN................ 0... e esse ee ee eee 5-0. 6 Beg. C0 0:9 GUE G. BO EOIO A PCEETCEG eC R RSEe EO RORCRT ERR RE eN ar translineata Shelley Acropodite without lobe on lateral margin; orientation of peak normal; distal zone curv- ing generally laterad from peak, not coplanar with basal zone; tip acuminate... 32 Prefemoral process divided near midlength, components relatively short, straight, and nearly equal; medial flange short and narrow, arising near midlength of peak, margin gently rounded; distal zone short, barely projecting into arch (Figs. 99-100); Bledsoe and Van Burencos., TN.................2-.0005 abbreviata, new species Prefemoral process divided basally, components moderate to long, lateral one broadly or sharply curved ventrad near midlength; medial flange variable, arising from prox- imal part of basal zone to peak, broadly rounded or triangular on distal extremity of latter; distal zone moderately long, curving well into arch (Figs. 94-96); Scott to EXAMItONYCOS wl Nie ram ioarnieie un eave iiaiessuretavacee: svete forficata, new species Basal zone with inner surface directed mediad; medial flange narrow located entirely on basal zone, inconspicuous in medial view (Figs. 2,6, 15)..................-- 34 Basal zone with inner surface directed generally anteriad; medial flange with configura- tion and position variable, not located entirely on basal zone, conspicuous or in- COW OOMONS 5 oc ctw cred ooo DOOD SEDO OD Sco tLe arenas Comintern cee ee eee 35 Acropodite with distinct spiniform or triangular tooth near midlength of peak (Figs. 15-16); metaterga with stripes along caudal margins; Kanawha, Clay, and Nicholas COB WAY oid Siete ole BU 8 ono OURO EIR CRTOLOEEA CIREREE chr en Cia Caene Piro aeter G rigida, new species Without this character; metaterga usually with variable middorsal spots, occasionally with stripes; Ontario and northern New England to Washington Co., VA Eee Mest Boat eee MAG acest tara eecure cask sae aca aca Wes Salata nus Ga MEGANE Marianas trimaculata (Wood) Prefemoral process long, extending beyond level of distal zone (Figs. 79-83)........ 36 Prefemoral process terminating well below level of distal zone................... 37 Distal zone curving dorsad (downward in medial view) from peak, coplanar with basal zone; tooth present on proximal part of distal zone, widely separated from medial flange; prefemoral process greatly enlarged to globose basally (Figs. 80-81*); Sevier (S065 LUNs ero deakata 0 0 cee D OM TG ERC MEN re Coie eer PREC Reece tr mean eee bidens (Causey) Distal zone directed laterad from peak, not coplanar with basal zone; tooth absent; pre- femoral process variable but at most only moderately swollen basally (Figs. 79-83); Semele (CoS, DUN ence oro tain ob oi ceo Ra See Le eee OMe ORO RC oeeetoe prolata, new species Distal zone curving generally laterad from peak, not coplanar with basal zone...... 38 MEM. AMER. ENT. SOC., 35 22 38. 39. 40. 41. 42. 43. 44. 45. XYSTODESMID MILLIPEDS Distal zone with variable length, directed generally dorsad from peak, essentially co- planar with basal'zone oc000. dfs Sone ts ohn ste ae en ee 44 Medial flange arising on basal zone or proximal part of peak, variably broad, obscuring at least narrow part of acropodite stem in medial view (Figs. 71, 75, 25*, 27*; Shelley 1983: Figs2) :3i.sheseu a. eee ace se wees eee ne ee eee 39 Medial flange located entirely on distal zone, peak and basal zone entirely visible in medial view (Fig. 129*); Oconee Co., SC, to Dawson Co., GA................... ee ec ae ar rere ce Recents heneen aie nictcacgnid arG.o.ohehasOnr oO -ax0.) disjuncta Shelley Medial flange long and narrow, arising on basal zone, terminating in broadly rounded lobe at midlength of distal zone; ventral surface of prefemoral process convex distad (Figs. 103-104); Franklin, Grundy, and Marioncos.,TN..... houstoni Chamberlin Medial flange terminating proximal to distal zone, configuration variable; prefemoral processiwithoutiexcavationS- nee aeraee ee ce eee ceo eee 40 Prefemoral process relatively long and bisinuately curved, extending beyond midlength of basal zone and terminating near level of distal zone (Figs. 107-108); Morgan Co., TRIN ee sae recs sek cae eee See Dicapa (Keeton) Prefemoral process relatively short, not bisinuate, much shorter than half the length of basal zone, terminating well below level of distal zone...................... 41 Medial flange thick and heavily sclerotized, not especially laminate; prostatic grove crossing from medial to lateral sides on basal zone (Shelley 1983a, Figs. 2-3); Spar- tanburg and Pickens cos., SC, to Polk and Transylvania cos., NC...............- Ee ee rae Morir Roots cee erences c-AB Se. 00 divergens Chamberlin Medial flange variably broad but thin and laminate, weakly sclerotized; prostatic groove crossing to lateral side at anterior bend.................---. + eee eee e eens 42 Acropodites thick and massive, heavily sclerotized, too large to overlap in aperture, one thus directed mediad and other projecting anteriad over most of 6th sternum; prefemoral process directed anteriad and parallel to peak, not toward distal part of acropodite (Figs. 25-28*); Buncombe, McDowell, and Rutherford cos., NC ....... Fed eS aUa RE Se WES cee RS See eee stibarophalla Shelley Acropodite moderately thick and heavy, overlapping normally over aperture and extend- ing beyond anterior margin; prefemoral process directed toward distal part of ACTOPOGILE: is csics see sae Shae RA eae Ak te See 43 Medial flange moderately broad, distally triangular; distal zone short, in lateral view not extending beyond level of broad, triangular part of medial flange (Figs. 71-72); Hampton Co., SC, to Wilcox Co., AL.................0.5- australis, new species Medial flange narrow, only slightly wider distad, not triangular; distal zone moderately long, in lateral view curving into arch and well beyond level of medial flange (Figs. 75=16)? RatrickiGon WA wd ee cede eee ee ee whiteheadi, new species Medial flange arising on basal zone, terminating on peak; peak with or without spur on outer surface near anterior bend; lateral flange not laminate, in form of broadly rounded lobe projecting ventrad and forming highest point of arch, poorly demar- cated from distal zone (Figs. 65-67); Greenville to Newberry cos.,SC ............. ote ee A GO MATT Neo tion aoa bo macra (Chamberlin) Medial flange absent from basal zone, arising on peak or distal zone; peak without spurs; lateral flange present or absent, laminate when present, not forming highest point of anche Ae oes ne Sa ee ee er eee 45 Medial flange on proximal portion of peak (Figs.3*, 31*, 45*, 57*, 76*, 84*, 88*) ee een Rr er aes TRARY PIN IR Rian wats Aen ee aecea, Chciata & HOI. hOrd.0.0.0'0'0.0 0.0.0 0.0 46 46. 47. 48. 49. 50. 51. S28 53. 54. SSE R.M. SHELLEY & D.R. WHITEHEAD 23 Medial flange small, not obscuring stem of acropodite in medial view; distal zone with opposing proximal lobes on medial and lateral margins, lateral lobe larger (Figs. Gye ORF) oo a) tras chose Jeu a-nlcher bs cierce ice Guchute imu rhd ime trict et EERE eRe eRe a ne 47 Medial flange larger, obscuring at least small section of stem of acropodite in medial view; distal zone either without lobes or with lobe only on lateral side......... 48 Portion of distal zone distal to lobes relatively long, bent sharply into arch; lateral flange present (Figs. 84-85*); paranota and metatergal stripes white to light yellow; Cocke ANCES E VIET COS sa IN Me icgs sees eee worstensta west tet yi te asesvevoueaoe aim eeareanans leucostriata Shelley Portion of distal zone distal to lobes relatively short, only slightly curved into arch; lateral flange absent (Figs. 88-89*); paranota red, metatergal stripes blue; McMinn ComaNitoiGilmernCorvGAy.. sac sae ces ee cru hens ie aes xerophylla Shelley Wateralitlangevabsentesrcret voit cs tenccste wieie seks ere ols Sos Woe anew wlensin be wena eovoeuniamiaaue 49 eateralitiany eypresen twarwrctucrsaeracersthausisners cuenotnn, ahapcaysley oe ttn Gkuaetenere col & 51 Arch of acropodite subtending a square; peak flattened, length nearly equal to distal zone (Figs. 45*, 47*); Lexington, Saluda, and Edgefield cos.,SC................. tea lasa te oe Ob GO Geer eho EeCee a ROL CN RO Cron not RICE a eR IC een eee tae quadrata Shelley Arch of acropodite otherwise; peak gently curved or rising to apex at beginning of apical curve, shorter or longer than distal zone................0 0.0 cc secs eeeeeees 50 Peak of arch gently curved, shorter than distal zone; medial flange relatively short, Margin acuminate; distal zone relatively long, projecting dorsad from peak (Fig. Ale) nYian ceyiC On NG waynes stabs: soir raed coteee ore Sis: Sa yanscaliaeceeduaiacy Say < sigirioides Shelley Peak of arch with apex distad at beginning of apical curve, much longer than distal zone; medial flange relatively long and narrow, margin variable; distal zone relatively short, projecting dorsad from peak or curved into arch (Figs. 31*, 33*); Mitchell, Yancey and McDowell cos., NC.............0 00 c cee eeeeeee inornata, new name Medial flange short, obscuring at most only short section of stem of acropodite; peak of arch rounded, anterior bend and apical curve more or less continuous through peak, poorly defined; distal zone curved but not projecting significantly into arch (Figs. SHlet) Sis) PUNT esol Croerens| Gigi stasntk oto ron corn Ga ck Eee Diao Cc cena laticurvosa Shelley Medial flange longer, obscuring long section of acropodite stem; peak not rounded, anterior bend and apical curve sharp and well defined, not continuous through peak; distal zone curved or straight, projecting wellinto arch................ Sy Apical curve bisinuate, with two inward bends into arch; lateral flange variable, absent, triangular, or rectangular (Figs. 57*, 59-60*); Transylvania and Henderson cos., NC BU OUI OREO O's 6 OOO DSC ECLA OLO HIGHEST CRCSERCeERCR RRCa Sn eea stenogon Chamberlin Apical curve smoothly rounded and continuous, forming arc with variable diameter; lateral flange present, usually long and narrow, occasionally with margin projecting and irregular, never triangular or rectangular.....................-2.-.08: 53 Acropodite thick and heavy; peak tilted mediad, exposing lateral margin in medial view; tooth absent (Fig. 76*); Blount Co., TN...............-.---- fumimontis Shelley Acropodite moderately thick and heavy; peak oriented normally, lateral edge obscured inhmedialviewAtoothipresentianr mene ere cies cielo nrsiteneisi: 54 Peak flexed ventrad at midlength; tooth attached to lamina of medial flange but well separated from lobe of latter; tip simple (Fig. 21*); Buncombe Co., NC........... 3 0:60 6 eS ONGC OLD GEERT 'S LENO. TOROS: DRC ERS PREC es em areolata Shelley Peak relatively flat; tooth present or absent, not attached to lamina of medial flange; ip Me flex el Ree Py aemepew veneer aes ra aera ae ses pera eer aircat lentils canederaiate 55 Medial flange narrow, poorly demarcated from stem of acropodite, depth much less than that of apical curvature; when present, tooth usually well separated from MEM. AMER. ENT. SOC., 35 24 56. 57. 58. Sy), 60. 61. 62. XYSTODESMID MILLIPEDS medial flange, often greatly reduced and rounded (Figs. 12-15*); Wilkes and Catawhbacos-3NC xiwens shh haao eee Ge iret eine eee stenoloba Shelley Medial flange broad to large, well demarcated from stem of acropodite; depth equal to or greater than that of apical curvature; when present, tooth located at distal ex- tremity of medial flange, usually subacuminate (Figs. 3*, 8*, 9*); southern West Virginia to southern South Carolina......................... latior (Brolemann) Distal zone tapering to acuminate tip; medial and lateral flanges large and conspicuous; peak of arch gently rounded (Figs. 97*, 105*).............. 000 cee ceeueee 57 Distal zone of nearly equal width throughout except for tip; latter blunt; flanges reduced, medial often vestigial; peak either curved or flattened (Figs. 108-112*, 114-121*, 123-124*, 126-127*); Greene and Unicoi cos., TN, to Buncombe and McDowell COSI QING fceelnie herd Se aise Dike Cote ee eee nigrimontis (Chamberlin) Lateral flange triangular, located entirely on peak; distal zone bisinuately curved distad, directed toward prefemoral process (Fig. 105*); Cocke Co., TN, and Madison CONG vas wankers i ee ace Se eed nee triangulata Shelley Lateral flange with margin broadly rounded, occurring from distal extremity of peak to midlength of distal zone; latter bent sharply inward into arch at termination point of flanges\(Figs::97*, 100*; 102*)).c Actas eee atcha cede a ae one ae ee hee eee 58 Acropodites in medial view showing profile of arch and distal zone; edge of medial flange visible in medial view, face directed anteriad; prefemoral process relatively long, recurved at midlength and directed toward distal zone, simple or bifurcate with one component usually much longer (Fig. 97*); Sevier and Cocke cos., TN, to Tran- sylvania'CosING ie en tee TR eee erane rubromarginata (Bollman) Distal zone twisted mediad revealing part of face of medial flange or all of faces of medial and lateral flanges in medial view; proximal portion of arch visible, distal obscured by flanges; bend of distal zone partly or completely obscured by flanges in medial view; prefemoral process variably short and straight, usually bifurcate with sube- qual components (Figs. 100-103*); Buncombe to Burke cos., NC................. sin bl RS euch Pate tec MPa Seco ERS SWC OT ET rT austrimontis Shelley Acropodite with row of distinct spurs on basal zone (Shelley 1982, Figs. 4,8,12).... 60 Without thischaracter: 4.2 sins Gas ee ee oe SIO Eee 62 Acropodite apically trifurcate, solenomerite directed away from gonopod along axis of acropodite; latter a broad, continuous curve, anterior bend poorly defined (Shelley 1982, Figs. 11-12); Polk and Hendersoncos., NC................ thrinax (Shelley) Acropodite apically blunt to subacute, with at most two terminal projections; soleno- merite located beneath distal extremity of peak, directed perpendicularly and laterad to peak; acropodite bent sharply at 1/3 to 1/2 length, anterior bend well defined ee ane ee Renae ce ie aera ir en Rh Sa cucu, 2.4.0-0:0,0 000.0 61 Solenomerite fused to undersurface of peak; spurs moderate in size, located entirely on basal zone (Shelley 1982, Figs. 3-4); McDowell, Buncombe, and Henderson cos., I Oa eer Po eee CRM NRA eet teen mia orOto. 0.0 0:0:5,0.0.0 haerens (Shelley) Solenomerite a separate bisinuate process, narrowly segregated from peak; spurs moderate to large in size, extending around anterior bend to base of lateral flange on peak (Shelley 1982, Figs. 7-8); Rutherford, Henderson, and Polk cos., NC Rr ier er teria ie cies n rimi ch ta cero mms osc oe teeG Oa Ob divaricata (Shelley) Acropodite broadened distad into cupulate expansion or cap; solenomerite either beneath and partly or completely obscured by cap, or located laterad at its base........ 63 Acropodite relatively narrow; widest basally, tapering smoothly to acuminate tip; solenomerite about half the length of acropodite, arising near base of, and running R.M. SHELLEY & D.R. WHITEHEAD 25 subparallel to, acropodite (Shelley 1984a, Figs. 10-11); Fall Zone of central Georgia MCDM a eer Ren ee ey Ree cies re bevieye ces stave, chaberyaricari sar dl miea chris levee eis vaciev aie persica (Hoffman) 63. Solenomerite distinctly visible on lateral side of acropodite at base of expansion; except for eastern populations, margin of expansion not strongly serrated (Shelley 1984a, Figs. 3-7); Thomas Co., GA, to Hernando Co., FL.............. planca (Loomis) Solenomerite located beneath cap and partly or completely obscured; margin of ex- Pansionistronelyssenratedaar-watsrer ee rset evel mesic ieee coors cities eisai cae 64 64. Acropodite in form of continuous, broadly curved arc, extending in medial view beyond level of prefemur; margin of expansion with a single row of up to 12 teeth; solenomerite a thickened, sclerotized boss (Shelley 1984a, Figs. 13-15); Charleston Co., SC, to Clarke and Burke cos.,GA....................0.. agrestis (Loomis) Acropodite in form of inverted L, stem relatively straight, bent sharply anteriad at level of cap, overhanging prefemur in medial view; margin of expansion highly serrate, with two or more rows of variable teeth; solenomerite a thin, falcate projection tucked under teeth, distal to boss (Shelley 1984, Figs. 19-22); Liberty Co., GA, to DUNC Lo) 5-5 18H Lerch cate cc: ono kia an ain Ton Ree een eee ee eae serrata (Shelley) SIGMORIA (RUDILORIA) Causey, new status Rudiloria Causey, 1955:28. Chamberlin and Hoffman, 1958:47. Jeekel, 1971:285. Hoffman 1978:6 Type species. — Rudiloria mohicana Causey, 1955, by original designa- tion. Diagnosis. — Metatergal color pattern variable, with spots or stripes of varying widths and diameters, at least partly concolorous with paranotal markings; gonopods in situ with acropodites extending mostly over op- posite side of aperture, not projecting anteriad between legs of segment 6; acropodites relatively thin and fragile, oriented on coxa with inner surface directed mediad, with narrow medial flange present on basal zone, largely undetectable in medial view because of acropodal orientation; tip simple. Remarks. — Occurring from Virginia to Canada and northern New England, the subgenus Rudiloria occupies the largest and most northerly range in Sigmoria s. lat. At first glance most of the species do not appear congeneric with Jatior (Sigmoria) or the species formerly described (Shelley 198la) as having acropodites shaped like the number 7, but the most ex- treme forms with circular acropodites connect with the latter through a con- tinuum of intermediate forms and there are important similarities with known species of Sigmoria s. lat. in North Carolina and Tennessee. Over the years these intermediate forms have received names which I retain, there being no compelling reason to do otherwise. More collecting is needed in eastern Kentucky, northern West Virginia, southern and eastern Ohio, and western Pennsylvania before their true statuses can be determined. In the only case where ample material is available, trimaculata from the counties MEM. AMER. ENT. SOC., 35 XYSTODESMID MILLIPEDS 26 tats eects I HYsiUM “TT + if psnvjomM “OT ap ap + if DIDUOLOD “6 + if DIDUIUNID “8 + + if pjad *L (01401x1q) als Ol ppisid “9 + OI DUDIIYOU °*¢ + + iC pyjopuvkns “fy + Z “Sp3JUl DIDjNIVUIA] ~€ + G Lajaduiajy “1 “7 + + ap + + + + EZ DIDINIDULAY “1 “| (0140]1pNY) DILOWBIC ouewO puejsuq AN Wd GN HO AM WA AX NI ON OS VD Td IV apo) uoxe |, MON ‘umouyun uloyed pue sIO[OD “OT “uaoys Sulpyieds pue ‘sedis yesioye}0ur “sjods jejouvied a3uvio 10 MOTO UIA 6 “sadiiys ovyM YqIM B310]e]9Ul ‘sjods jeyouvied pol UUM °8 ‘sadijs onjq YIM ed10}e}9Ul ‘sjods yejouvsed pol UUM “L ‘sodiijs jediojyejou pue sjods jejouered Moyak ajed 10 o114yM YIM “9 -sodiis e3iojejyou pue sjods Jejouesed ajdind 10 Je[OIA GUM “Ss -sodis [e8ioyejou! pue sjods Jejouvied pol UM “sodiijs jedioyejow pue siods jeyouvied a3uei0 10 MOTJOA UM “€ -sjods [esIOppIw YIM e3I9}e}0Ul ‘sjods yejouered pol 0} MOT[as YAN “7 -yorlq Ajorque e310} e19Ul ‘sjods yejouesed pol 0} MOTTOs UUM T susoyed IO[OD 10} uso}jed 10J09 pue o1e}s Aq viuDJOJjaq pue DILOWBIG JO Saioadsqns pue saoads Jo uolnqisiq “€ Fav =) 0) Dil R.M. SHELLEY & D.R. WHITEHEAD +++ 4+ 4 +++ 4 +++ 4+ + + XDUIIY] “LE DIDIIADAIP “9€ suasavy “SE SUABIBAID “PE Djjpydosvqus “¢¢ S1DAISND “TE (sndosiay)D) DIIDU “TE Ipsofjays “O€ DIONIAD *6Z DISNGOd *3Z sSuapisavidiq “LZ 1710QqD “97 1dafid “SZ (p140}d9/D) aassDUlak “PZ xajduiis *€Z DPNIDS “ZZ DQMDJDI “7 (o1unjDOID) DSOAANIIJD] “OT pjpaponb *6| pqojouajs *g “sp3}UI 4010) “LT juouffoy *] “OI ppunu 7] “ST 4010] *] “yl (D1LOUBIG) piafijajovp *§ | 1syOO1G “ZT MEM. AMER. ENT. SOC., 35 XYSTODESMID MILLIPEDS 28 + ++ 44+ 44+4+4++4+ + ++ 4+ + + olejuQ purlsuq AN MON Vd GN HO AM VWA AX NL ON OS WO Td TV 2p) > oor nm yrrrnrsyryrsyrvsrsyrtrmrnnn rwrowd Sap1os0jaydv °6S DyJxydosax °3¢ DIDIAJSOINA] “LS ADIDYDIUDU “9G (0140]]0.J) pjountsip °¢¢ uOsOUaIS “HS DIDIOAID *€G DIDULOUI “7G DIDIUNL] “TS SaplOlisis “QC 1091UN “U “6p DsojnsuDd “u “8p DIPAUajUl “U “Ly SIJUOUMAZIU “U “Op DDjNsuvl4] “Sp IPDOYAHlyYM “py SIJUOUNAISND *€p DIDUIBIDWOAGNA “TH (0141816) pa1siad “|p pounjd ‘Or DIDIIIS “GE SIJSALBD “QE uoxe | 29 R.M. SHELLEY & D.R. WHITEHEAD + ++ +4444 +++ +44 St Se oS ol oS oS ol ol lS el ol Da] “sp8qUr Majiuisg pijiyd -q Majuisq *q D1IDIOJJAq DSOAANIISSDAD “EL, Diojnpuad *¢L DINAUIU “TL pdpaid "OL DIDIAAIGGD °69 1uojsnoy °g9 pjpof4of *L9 pso1aqn} “99 pjpjoid °S9 suapiq “9 1a1jSUID “€9 Dald] °79 syjuowmiunf “19 Dypaulsuvn14 *09 MEM. AMER. ENT. SOC., 35 30 XYSTODESMID MILLIPEDS of western Virginia, it is possible to combine nominal species to form subspecies and synonyms. The acropodites of the forms of Rudiloria are relatively thin and fragile and curve more mediad than anteriad, thus projecting more over the op- posite side of the aperture than over the anterior margin and the 6th ster- num. The inner surface of the basal zone is directed mediad, and since the thin, narrow medial flange is located chiefly in this region, it it barely detec- table in medial view. The most variable aspect of the acropodites is the distal zone, which can be moderate to very long, overlapping the basal zone and forming a loop or circular acropodite. This condition is similar to that in ainsliei, aphelorioides, (Falloria) and species of Apheloria. The prefemoral processes range from absent to short and triangular to moderately long and uncinate. The subgenus is comprised of five known forms in a single species group, the trimaculata group. Two forms are con- sidered races of a single species. Components. — trimaculata (Wood) [t. trimaculata, t. kleinpeteri (Hoffman)], mohicana (Causey), guyandotta (Shear), rigida new species. Sigmoria (Rudiloria) trimaculata (Wood), new combination Diagnosis. — A moderate-size species of Sigmoria with medial flange on distal part of basal zone and with variable color pattern, paranota yellow, metaterga with yellow, red, or orange markings on caudal edges ranging from small, discrete middorsal spots to variable semilunar blotches diffus- ing into paranotal markings and becoming stripes; gonopods with following diagnostic characters: prefemoral process moderately long, uncinate; acropodite thin and fragile, forming completely or nearly complete, broadly symmetrical, loop overhanging and extending well beyond level of prefemoral process; basal zone with inner surface directed mediad; anterior bend and apical curve poorly defined; peak gently curved to flattened, con- tinuous through anterior bend and apical curve with basal and distal zones; latter either coplanar with basal zone or curving through several planes and not coplanar with basal zone, extending downward from peak in medial view then curving into arch and terminating in acute, falcate tip, with or without broadly rounded lobes at midlength; medial flange inconspicuous in medial view. Remarks. — This species has the largest range in the genus, extending from Canada north of Lake Ontario to southwestern Virginia. There is marked stability in the configuration of the gonopods from central Virginia northward, but in Giles County, the acropodite begins to widen at midlength of the distal zone (base of the upwards curve). South of Giles —— R.M. SHELLEY & D.R. WHITEHEAD 31 County there is a broad lobe at this position whose proximal part is oriented perpendicularly to the medial field of view and whose distal part lies in this plane. Samples from Mountain Lake Biological Station and northern Giles County display intermediate conditions ranging from a slightly wider acropodite at this point to a small rounded lobe. I therefore recognize two subspecies of this assemblage, the oldest available name being trimaculata (Wood, 1864), assigned to a population of the northern form from Sus- quehannah County, Pennsylvania. The oldest available name from the range of the southern race is kleinpeteri (Hoffman 1949). Two names were proposed for intergrade forms, tortua (Chamberlin 1949) and picta (Hoff- man 1949), and for the sake of convenience, I place them in synonymy under the nominate subspecies. Sigmoria (Rudiloria) trimaculata trimaculata (Wood) Figs. 1-3 Polydesmus (Fontaria) trimaculata Wood, 1864:6; 1865:223-224, Figs. 53-54. Fontaria lutzi Jacot, 1938:571-572, Fig. 1. Apheloria trimaculata: Attems, 1938:170. Loomis, 1944:173 (in part). Hoffman, 1949:378. Apheloria keuka Chamberlin, 1939:10-11, pl. 4, Fig. 32. Chamberlin and Hoffman, 1958:19 NEW SYNONYMY. Apheloria coriacea (nec Koch): Chamberlin, 1947:25, Fig. 5. Apheloria tortua Chamberlin, 1949:101, Fig. 23. NEW SYNONYMY. Apheloria antrostomicola Hoffman, 1949:372-374, pl. 26, Figs. 1-2. NEW SYNONYMY. Apheloria picta Hoffman, 1949:376-378, pl. 26, Figs. 5-6. NEW SYNONYMY. Apheloria trimaculata trimaculata: Hoffman, 1951:2-3, Fig. 1b. Chamberlin and Hoffman, 1958:20. Apheloria trimaculata antrostomicola: Hoffman, 1951:3-4, Fig. 1c. Chamberlin and Hoff- man, 1958:20-21. Apheloria trimaculata incarnata Hoffman, 1951:4-5, Fig. la. Chamberlin and Hoffman, 1958:21. NEW SYNONYMY. Apheloria trimaculata tortua: Hoffman, 1951:5-6, Fig. 1d. Chamberlin and Hoffman, 1958: 21. Rudiloria trimaculata: Hoffman, 1978, Figs. 2, 5. Apheloria (Rudiloria) trimaculata trimaculata: Kevan, 1983:2968. Type specimens. — Not known to exist; the type locality is in Susque- hannah Co., PA (Wood 1864, 1865). Diagnosis. — Peak relatively flat; distal zone at most only slightly wider at midlength (base of upwards curve), without lobes. Color in Life. — Variable. Most forms with yellow paranota and concolorous, yellow mid- dorsal spots on caudal edges of otherwise black metaterga, the spots varying in size from small, discrete circles to semilunar blotches spreading laterally along segment margins; some forms with spots connecting by thin line with paranotal markings, others with broad yellow stripes MEM. AMER. ENT. SOC., 35 32 XYSTODESMID MILLIPEDS along caudal edges showing little or no evidence of middorsal spots (see also Hoffman 1949, 1951). Male from Tucker Co., WV. — Length 33.9 mm, maximum width 7.7 mm, W/L ratio 22.7%, depth/width ratio 59.7%. Segmental widths as follows: collums 6.4mm _ 15th 7.4 2nd-3rd 7.1 16th 7.1 4th 7.4 17th 6.1 Sth-14th 7.7 18th 4.9 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.1 mm, interantennal isthmus 1.4 mm. Antennae reaching back to middle of 4th paranota, relative lengths of antennomeres 2>3>5=6>4>1>7. Genae without impressions. Facial setae as follows: epicranial 1-1, interantennal absent, frontal 1-1, genal 2-2, clypeal about 10-10, labral about 18-18, merging with clypeal series and continuing for short distance along genal border, about 4 setae per side. Dorsum smooth, polished, moderately coriaceous on paranota. Collum broad, ends not pro- duced beyond those of following tergite. Paranota moderately depressed, continuing slope of 3 Fics, 1-3. Sigmoria (Rudiloria) trimaculata trimaculata. 1, process of 4th sternum of male from Tucker Co., WV, caudal view. 2, telopodite of left gonopod of the same, medial view. 3, the same, lateral view. Scale line = 1.00 mm for all figures. R.M. SHELLEY & D.R. WHITEHEAD 33 dorsum, caudolateral corners rounded through segment 10, becoming blunt and progressively more acute caudally. Peritremata thin but distinct, moderately elevated above paranotal sur- face, ozopores located caudal to midlength, opening dorsolaterad. Process of 4th sternum moderately long, apically divided, length equal to widths of adjacent coxae (Fig. 1); 5th sternum with two low, paramedian knobs between anterior legs and elevated flattened areas between posterior legs; 6th sternum flattened, with very slight convex impres- sion on caudal surface. Postgonopodal sterna flat and plate-like, with bicruciform impressions on segments 8-9 and variably broad, shallow, central impressions on remaining segments. Cox- ae with low, blunt tubercles beginning on segment 9, becoming longer and sharper caudally; prefemoral spines arising on segment 5, becoming progressively more acute posteriorly. Gonopodal aperture ovoid, 3.4 mm wide and 1.8 mm long at midpoint, indented anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Hoffman, 1978, Fig. 2) with acropodites curving mediad across midline of aperture, curving dorsad over op- posite side then laterad toward midline, overlapping opposite member. Gonopod structure as follows (Figs. 2-3): Prefemoral process moderately long, acute, and falcate, directed mediad, with low, basal spur. Acropodite thin and fragile, in form of flattened loop nearly forming complete circle, overhanging and extending well beyond level of prefemoral process; basal zone relatively long, inner surface directed mediad, sides tapering distad, without modifica- tions; anterior bend moderately broad; peak moderately long, flattened, not continuous with basal zone; apical curve broad, poorly defined; distal zone long, curving broadly downward from peak, linearly into arch, then upwards in a short, linear stretch terminating in acute, falcate tip, sides diverging slightly proximad and broadest at base of upwards curve, tapering smoothly and continuously thereafter; tip directed toward peak in medial view. Medial flange narrow and inconspicuous in medial view, arising on basal zone and blending with medial edge at anterior bend. Lateral flange absent. Prostatic groove crossing to lateral surface at anterior bend, continuing to terminal opening. Female from Tucker Co. — Length 40.7 mm, maximum width 7.4 mm, W/L ratio 18.2%, depth/width ratio 78.4%. Cyphopods in situ with receptacles visible in aperture, valves directed dorsad. Receptacle large, hood-like, enveloping valves, surface rugulose. Valves small, equal, surfaces finely granulate. Variation. — The gonopods of trimaculata are constant throughout the range and do not change with changes in color pattern. Ecology. — The material I collected in West Virginia and New York was discovered under moist hardwood litter in sheltered sites in deciduous forests. Distribution. — Northern New England and southern Ontario to west- central Virginia, mostly in upland provinces. This species and Apheloria corrugata (Wood), also found in Ontario, are the only two xystodesmids known from eastern Canada, although Pleuroloma flavipes Rafinesque may occur in southern Ontario (Shelley 1980b, Kevan 1983). The range of trimaculata, around 550 miles long, is the longest documented to date in the Apheloriini. It spans many large rivers, including the St. Lawrence, Con- necticut, Hudson, Delaware, Susquehannah, and Potomac. Material was examined as follows: MEM. AMER. ENT. SOC., 35 34 XYSTODESMID MILLIPEDS CANADA. — ONTARIO: Durham Co., Kendal, M, 30 May 1966, and M, 19 July 1967, I.M. Smith (ROM). Leeds Co., Charreys Locks, M, 29, May 1973, J.C.E. Riotte (ROM). Halton Co., Speyside, F, 10 May 1964 (ROM). Frontenac Co., Gull Lake nr. Arden, M, 4 September 1948, R.E. Crabill (NMNH). UNITED STATES. — MAINE: Oxford Co., Norway, F, date unknown, O. Harger (PMNH). NEW HAMPSHIRE. — Sullivan Co., Corbin Park, M, date unknown, B.L. Brooks (MCZ). Rutland Co., Tweed River, M, 10 September 1934, collector unknown (UMN). Ben- nington Co., Beartown, M, 9 June 1954, F.P. Rindge (AMNH). MASSACHUSETTS. — Berkshire Co., Mt. Greylock, M, 15 June 1936, C.H. Blake (MCZ); and Williamstown, M, 14 June 1936, R. Dow (MCZ). NEW YORK. — Steuben Co., Lake Keuka, M, September 1905, collector unknown (RVC). Essex Co., Keene Valley, 2M, 22 June and 21 July 1917, H. Notman (AMNH). Greene Co., near Lanesville on NY hwy. 214, 6M, 7F, 18 August 1978 (NCSM A2402). Ulster Co., Phoenicia, M, 25 July 1909, L.J. Barnum (AMNBH). PENNSYLVANIA. — Potter Co., locality unknown, M, F, 3 July 1906, H.W. Fowler (ANSP). Schuykill Co., Tuscarora St. Pk., 5M, 3F, 2 August 1981, H.W. Levi (MCZ). WEST VIRGINIA. — Tucker Co., along WV hwy. 32, 5.5 mi. E. Lanesville, Monogahela Nat. For., 7M, F, 23 August 1978 (NCSM A2412). VIRGINIA. — Highland CO., 2 mi. N Williamsville, Bull Pasture R. Gorge, 3M, F, 4 June 1972, R.L. Hoffman (RLH). Bath Co., locality unknown, M, 2F, date and collector unknown (MCZ). Rockbridge Co., Vesovius, F, 11 August 1971, R.H. Perry (RLH). Alleghany Co., 2 mi. NW Clifton Forge, M, 3F, 14 June 1947, R.L. Hoffman (RLH); W side Potts Mtn., 2F, 18 July 1982 R.L. Hoffman (RLH); 8 mi. SW Lowmoor, 16 June 1947, R.L. Hoffman (NMNBH). Remarks. — Because color and color pattern are variable in many eastern xystodesmids, I do not think subspecies should be based solely on these characters. They should be accompanied by corresponding changes in the gonopods as in J/atior (Sigmoria). This was my philosophy with Pleuroloma (Shelley 1980b) and races were not recognized in P. flavipes Rafinesque. As shown in Table 1, several species of Sigmoria s. lat. are polymorphic for color and most have at least an occasional variant, so dif- ferences in color or color pattern are not a good basis for taxonomic deci- sions. The patterns in ¢. trimaculata are also more diverse than previously thought, as a yellow transverse stripe variant occurs in West Virginia. Additional literature records of trimaculata are as follows: Susquehan- nah Co., PA (Wood 1864, 1865); Keene, Cheshire Co., NH (Jacot 1938); Ithaca, Tompkins Co., NY (Chamberlin, 1939, Loomis 1944); Potter Co., PA, and Garrett Co., MD (Chamberlin 1947); near Staunton and Lynd- hurst, Augusta Co., VA (Hoffman 1949); and an unspecified site in Quebec, Canada (Kevan, 1983). Sigmoria (Rudiloria) trimaculata kleinpeteri (Hoffman), new combination, new status Figs. 4-7 Apheloria trimaculata (nec Wood): Loomis, 1944:173 (in part). Apheloria kleinpeteri Hoffman, 1949:375-376, Pl. 26, Figs. 3-4. Chamberlin and Hoffman, 1958:19. R.M. SHELLEY & D.R. WHITEHEAD 35 Type specimens. — Male holotype (NMNH) and two male paratypes (RLH) collected by R.L. Hoffman and H.I. Kleinpeter, 30 June 1947, from Burkes Garden, Tazewell Co., Va. Female allotype (NMNH) collected by J.E. Graf, 5 June 1940, at same locality. Diagnosis. — Peak gently curved; distal zone with broadly rounded lobes at midlength (base of upwards curve), the proximal one perpendicular to field of vision in medial view, the lateral one visible from this respective. Variation. — The gonopods of kleinpeteri are uniform and do not vary appreciably from the condition in the holotype. Ecology. — Unknown. Distribution. — Western Virginia and adjacent southern West Virginia. Material was examined as follows: WEST VIRGINIA. — Mercer Co., Bluefield, East River Mts., M, F, 25 June 1961, R.L. Hoffman (RLH); and along US hwy. 52 above Bluefield, 2M, 20 June 1968, R.L. Hoffman (RLH). VIRGINIA. — Giles Co., W of Thessalia, near head of Sugar Run, 2M, 3 August 1980, R.L. Hoffman (RLH). Bland Co., 2 mi. S. Long Spur, W side of Little Walker Mtn., F, 16 Oc- tober 1966, R.L. Hoffman (RLH). Tazewell Co., Burkes Garden, F, 5 June 1940, J.E. Graf (NMNB), M, 30 June 1947, R.L. Hoffman and H.I. Kleinpeter (NMNH), and M, July 1970, collector unknown (RLH). TYPE LOCALITY. Wythe Co., 3 mi. ESE Wytheville, E slope Sand Mtn., F, 1 June 1981, R.L. Hoffman (RLH). Washington Co., along VA hwy. 80 at Hayters Gap, 2M, 8 June 1974, D.W. Ogle (RLH); bluffs on N. Fork Holston R., 4M, 2F, 11 May 1981, D.W. Ogle (RLH); and Brumley Gap, 4M, 24 May 1978, R.L. Hoffman (RLH). Sigmoria (Rudiloria) trimaculata intergrades The samples of trimaculata from Mountain Lake Biological Station, in northern Giles County, Virginia, display conditions intermediate between those of the two subspecies. They vary from a slightly wider distal zone at midlength to a small rounded lobe at this point, the latter being illustrated by Hoffman (1949, Figs. 5-6). The names Apheloria tortua and A. picta have been assigned to these forms, and are synonymized with the nominate race. Specimens were examined as follows: VIRGINIA. — Giles Co., Mountain Lake Biological Station, M, July 1942, A.C. Cole (RVC) and M, F, June 1947, H.H. Hobbs et al. (NMNH). Sigmoria (Rudiloria) guyandotta (Shear), new combination Figs. 8-9 Apheloria guyandotta Shear, 1972:494-496, Figs. 1-2. Type specimens. — Male holotype (MCZ) and six male and two female paratypes (WAS) collected by P. Vogel and M. McGraw, 28 May-1 June 1968, from Shaft Hollow near McGraws, Wyoming Co., WV. MEM. AMER. ENT. SOC., 35 36 XYSTODESMID MILLIPEDS Diagnosis. — A moderate-size species of Sigmoria with medial flange ex- tending from distal portion of basal zone to midlength of peak and with yellow paranota and red or red-orange middorsal spots, latter tending to 6 7 Fics. 4-7. Sigmoria (Rudiloria) trimaculata kleinpeteri. 4, process of 4th sternum of male from Washington Co., VA, caudal view. 5, gonopods in situ, ventral view of the same. 6, telopodite of left gonopod of the same, medial view. 7, the same, lateral view. Scale line for fig. 5 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 37 form diffuse or distinct metatergal stripes; gonopods with following diagnostic characters: prefemoral process absent; acropodite moderately thick, arch leaning anteriomediad and extending well beyond level of prefemur; basal zone short, inner surface directed mediad; anterior bend poorly defined; peak relatively long, sides narrowing with highest point distad; apical curve well defined; distal zone moderately long, coplanar with basal zone, directed nearly ventrad from peak, flared for entire length with lamellae on both sides, narrowing distad to acuminate tip; medial flange narrow and inconspicuous in medial view; lateral flange narrow and in- distinct. Color in Life. — Variable within species and individuals. According to Shear (1972) the paranota are yellow and the metaterga are black with red or red-orange median spots that may or may not form distinct or diffuse stripes along the caudal edges connecting with the paranotal spots. The collum of the holotype has an anteriomedian red spot with a diffuse yellow stripe along the caudal margin. Holotype. — The following descriptions of the pregonopodal sterna and the gonopods in “‘sigmoid’’ terminology are supplemental to the more complete description by Shear (1972). 4th sternum with small, divided lobe, much shorter than widths of adjacent coxae (Fig. 8); 5th sternum with minute, barely detectable elevations between anterior legs, flat and un- modified between posterior legs; 6th sternum flattened, without depression on caudal edge. Gonopods in situ (see Shear, 1972, Fig. 1) with acropodites extending mediad across midline, lying one in front of other, curving dorsad over opposite side and inserting on opposite prefemur. Gonopod structure as follows (Fig. 9): Prefemoral process absent. Acropodite moderately thick, arch leaning anteriomediad, overhanging and extending well beyond level of prefemur; basal zone relatively short, with inner surface directed mediad; anterior bend moderately broad, poorly defined; peak relatively long, rising continuously to apex at distal ex- tremity, sides narrowing distad; apical curve broad but well defined; distal zone moderately long, coplanar with basal zone, directed nearly ventrad from peak, flared for entire length with lamellae on both sides, narrowing and curving distad to subacuminate tip. Medial flange long and narrow, indistinct, terminating and blending into inner margin of peak distal to midlength. Lateral flange short, narrow, and inconspicuous, located on distal zone opposite ventral bend. Prostatic groove crossing to lateral side at anterior bend, continuing to terminal opening. Male Paratypes. — The male paratypes agree with the holotype in all particulars. Female Paratype. — Shear (1972) briefly described somatic features. Cyphopods in situ with receptacles visible in aperture, valves directed dorsad. Receptable large, hood-like, enveloping valves, surface rugulose. Valves small, equal, surfaces finely granulate. Variation. — The male from Menifee County, Kentucky, has a small, acute, nubbinlike prefemoral process. Ecology. — Unknown. Distribution. — Known from two small, allopatric populations in southern West Virginia and east-central Kentucky. Comparatively little sampling has taken place in eastern Kentucky, however, and guyandotta MEM. AMER. ENT. SOC., 35 38 XYSTODESMID MILLIPEDS should occur in the intervening area. Material was examined from the following new localities: KENTUCKY. — Rowan Co., Moreland, M, 14 July 1961, A Razor (FSCA). Menifee Co., 4.5 mi. NE Slade, Dunkan Hollow, M, 3F, 31 August 1957, L. Hubricht (RLH). Remarks. — Except for being flared, the distal zone of guyandotta has the same basic configuration as that of mohicana, and the rounded lobe on the outer margin at the apical curve and the beginning of the flare appears homologous to the lobe in that position in mohicana. Though more pro- nounced, the lobes on the distal zone in ¢. kleinpeteri, when examined from the ventral perspective, have a similar configuration to the flange of guyan- dotta. However, the distal zone of guyandotta is shorter than that of f. kleinpeteri, the flanges are smaller, and the portion distal to the flanges is absent. Thus, the acropodite of guyandotta is intermediate between those of t. kleinpeteri and mohicana, and the forms may be clinally continuous and conspecific. A final decision on their statuses is left to future investigators with access to more material from uncollected areas in Kentucky, West Virginia, and Ohio. Sigmoria (Rudiloria) mohicana (Causey), new combination Figs. 10-12 Fontaria trimaculata (nec Wood): Williams and Hefner, 1928:108-109, Fig. 10c. Rudiloria mohincana Causey, 1955:28-29, Fig. 6. Rudiloria mohicana: Chamberlin and Hoffman, 1958:47. Hoffman, 1978, Fig. 4. Apheloria mohicana: Shear, 1972:496-497, Fig. 3. Type specimens. — Male holotype (AMNH) collected by L. Gray, August 1951, from Mohican State Park, Ashland Co., OH. There are no paratypes. Diagnosis. — A moderate-size species of Sigmoria with medial flange ex- tending from distal extremity of basal zone to midlength of peak and with metatergal stripe color pattern; gonopods with following diagnostic characters: prefemoral process absent; acropodite thin and fragile, arch high and rounded, extending beyond level of prefemur; basal zone with in- ner surface directed mediad; anterior bend poorly defined; peak gently rounded; apical curve with two broad bends; distal zone moderately long, directed laterad from peak, not coplanar with other sections, widest basally with slight proximal lobe on outer margin, tapering to acuminate tip; medial flange narrow and inconspicuous; lateral flange absent. Color in Life. — Unknown. Causey (1955) described the color pattern after two years in alcohol, and although the pigments had disappeared, the striped pattern was evident. It still is R.M. SHELLEY & D.R. WHITEHEAD 39 12 9 Fics. 8-12. 8-9, Sigmoria (Rudiloria) guyandotta. 8, process of 4th sternum of holotype, caudal view. 9, telopodite of left gonopod of the same, medial view. 10-12, Sigmoria (Rudiloria) mohicana. 10, process of 4th sternum of holotype, caudal view. 11, telopodite of left gonopod of holotype, medial view. 12, the same, lateral view. Scale line = 1.00 mm for 10-12; 1.14 mm for 8-9. MEM. AMER. ENT. SOC., 35 40 XYSTODESMID MILLIPEDS today after 30 years of preservation, and there are broad stripes along the caudal margins of the metaterga and both edges of the collum. Holotype. — The following account of the pregonopodal sterna and the gonopods in “‘Sigmoid”’ terminology supplements previous descriptions by Causey (1955) and Shear (1972). 4th sternum with small, undivided lobe, much shorter than widths of adjacent coxae (Fig. 10); 5th sternum with slight elevations between 4th legs and smaller ones between Sth legs; 6th ster- num flat, without depression in caudal edge. In situ arrangement of gonopods unknown. Gonopod structure as follows (Figs. 11-12): prefemoral process absent. Acropodite thin and fragile, forming high arch, gently and un- evenly rounded, overhanging and extending beyond level of prefemur; basal zone relatively long, inner surface directed mediad, without modifications; anterior bend broad, poorly de- fined; peak gently rounded, highest proximal to midlength; apical curve with two broad bends; distal zone moderately long, curving laterad from peak, not coplanar with basal zone, with two bends, widest basally with slight proximal lobe on outer margin, sides narrowing and tapering smoothly to subacuminate tip. Medial flange narrow and inconspicuous, arising near anterior bend, terminating by blending into inner margin of peak near midline, without lobes. Lateral flange absent, possibly represented by slight lobe on outer margin at proximal corner of distal zone. Prostatic groove crossing to lateral surface on basal zone, continuing to terminal open- ing. Distribution. — Known from the type locality and probably also Mariet- ta, Washington County, and Athens, Athens County, Ohio (Williams and Hefner 1928). Remarks. — The acropodite of mohicana is convergent with forms of nigrimontis intermedia (Sigiria) in its thin, fragile structure, the double bend in the distal zone, and the apical configuration (see Shelley 1981la, Figs. 114-121, p. 114). Also noteworthy is the laterally directed distal zone, which is not coplanar with the basal zone. This feature is less pronounced than that in species of Falloria in the Cumberland Plateau of Tennessee, which occur nearly due south of mohicana. Thirty years after its discovery, the type specimen is still the only authen- tic individual of mohicana. Thus, nothing can be said about variation, female anatomy, and ecological preferences since the habitat was not in- dicated in the original description. This one male is also the only authentic specimen of Sigmoria s. lat. from the state of Ohio, which can be expected to have a wealth of forms because the type locality is in the north-central part, about 60 miles south of Lake Erie. I agree with Shear (1972) that mohicana probably occurs throughout the hill country of southern Ohio, and one of the remaining needs in the study of Sigmoria s. lat. is more material from this state and adjacent parts of West Virginia and Kentucky. I include under mohicana the records of Fontaria trimaculata from Athens and Marion, Ohio, published by Williams and Hefner (1928). Their species was probably not corrugata, for which they reported a different col- or pattern and which they claimed was abundant in this section of Ohio. R.M. SHELLEY & D.R. WHITEHEAD 41 The gonopod drawings are adjacent, and although from different angles, appear to be of different forms. Hoffman (1951) also reported seeing a new genus and species from Ohio that had been sent to him by Hefner’ and sug- gested that it was the same as Williams and Hefner’s form. Their drawing of “‘trimaculata’’ shows a moderately slender, symmetrically curved acropodite with a slightly greater apical curve directed inward into the arch. Sigmoria (R.) mohicana appears like this from certain perspectives, and there is a notable similarity between the figure of Williams and Hefner and that of mohicana published by Shear (1972, Fig. 3). The absence of a prefemoral process from Williams and Hefner’s form also corresponds to the condition in mohicana. Sigmoria (Rudiloria) rigida Shelley, new species Figs. 13-16 Type specimens. — Male holotype (RLH) and one male and three female paratypes collected by L. Hubricht, 16 June 1956, on Elk River bluffs, 1-1.5 mi. S Clay, Clay Co., WV. Oné male and three female paratypes (RLH) taken by same collector at same locality, 28 May 1952. Diagnosis. — A moderate-size species of Sigmoria with medial flange located on distal portion of basal zone and with metatergal stripe color pat- tern; gonopods with following diagnostic characters: prefemoral process varying from small nub to small, triangular structure; acropodite thin and fragile, arch flattened and extending beyond level of prefemoral process; basal zone relatively long, inner surface directed mediad; anterior bend well defined; peak long, flattened, edges generally parallel; apical curve with two bends; distal zone moderately long, coplanar with basal zone, linear basal- ly, bending into arch at 2/3 length, widening slightly then narrowing to acuminate tip; medial flange narrow and indistinct in medial view; tooth present at midlength of peak, rigid and spiniform; lateral flange absent. Color in Life. — Unknown. Preserved material shows a clear pattern of stripes along the caudal margins of the metaterga connecting the paranotal markings and along both the anterior and posterior edges of the collum. Holotype. — Length 34.5 mm, maximum width 7.9 mm, W/L ratio 22.9%, depth/ width ratio 63.3%. Segmental widths as follows: collum 6.4 mm 16th 7.3 2nd 7.0 17th 6.7 3rd 7.7 18th 5.2 4th-15th 7.9 7 This sample, from Marietta, Ohio, is present in Dr. Hoffman’s collection, but the gonopods are lost and its assignment to mohicana needs confirmation. MEM. AMER. ENT. SOC., 35 42 XYSTODESMID MILLIPEDS Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.3 mm, interantennal isthmus 1.5 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>4=5=6>1>7. Genae without impressions. Facial setae as follows: epicranial, interantennal, and genal absent, fron- tal 1-1, clypeal about 10-10, labral TOD, 14 16 | 15 | Fics. 13-16. Sigmoria (Rudiloria) rigida. 13, process of 4th sternum of holotype, caudal view. 14, gonopods in situ, ventral view of paratype. 15, telopodite of left gonopod of holotype, medial view. 16, the same, lateral view. Scale line for fig. 14 = 1.00 mm; line for other figs. = 1.00 mm for each. a a R.M. SHELLEY & D.R. WHITEHEAD 43 Dorsum smooth, polished, becoming moderately coriaceous on anterior half of paranota. Collum moderately broad, ends broadly rounded and extending slightly beyond those of following tergite. Paranota moderately depressed, continuing slope of dorsum, caudolateral corners rounded through segment 8, blunt on 9-14, becoming progressively more acute posteriorly. Peritremata thin but distinct, moderately elevated above paranotal surface; ozopores located caudal to midlength, opening dorsolaterad. Process of 4th sternum apically divided, shorter than widths of adjacent coxae (Fig. 13); 5th sternum with two paramedian knobs between anterior legs, shorter than widths of adjacent coxae, and broad elevated area between caudal legs; 6th sternum flat, without convex reces- sions in caudal edge. Postgonopodal sterna flat and plate-like, with slight bicruciform impres- sions on 8 and 9. Coxae with small tubercles beginning on segment 12, becoming progressively larger posteriorly; prefemoral spines arising on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture ovoid, 2.7 mm wide and 1.5 mm long at midpoint, without indenta- tions, sides flush with metazonal surface. Gonopods in situ (Fig. 14, of paratype) with acropodites extending mediad across midline, lying one in front of other, curving dorsad over opposite side and touching opposite coxa. Gonopod structure as follows (Figs. 15-16): Prefemoral process small, triangular, directed toward tip of acropodite; prefemur with small, acute spur on medial surface at juncture with basal zone, spur obscured in medial view. Acropodite relatively thin and fragile, forming high, flattened arch, overhanging and extend- ing beyond level of prefemoral process; basal zone relatively long, inner surface directed mediad, without modifications; anterior bend sharp, well defined; peak moderately long, flat- tened, edges parallel with slight concavity on outer margin; apical curve in two parts, a broad, proximal curve followed by a linear section then a sharp distal curve; distal zone moderately long, coplanar with basal zone, with a linear basal portion bent abruptly into arch at 2/3 length, widening slightly apically then narrowing rapidly to subacuminate tip; latter directed toward prefemur. Medial flange narrow and indistinct, located entirely on basal zone, barely detectable in medial view. Tooth present, a short, rigid spiniform projection at midlength of peak. Lateral flange absent. Prostatic groove crossing to lateral side proximal to anterior bend, continuing to terminal opening. Male Paratypes. — The male paratypes agree with the holotype in all particulars. Female Paratype. — Badly fragmented and unmeasurable. Cyphopods in situ with openings of valves visible in aperture. Receptacle moderate, located anteriomediad to valves, surface rugulose. Valves moderate, equal, surfaces finely granulate. Variation. — Males from Kanawha and Nicholas counties have only a small nubbinlike prefemoral process, and the tooth is slightly shorter and more rounded apically. Ecology. — Unknown. Distribution. — Known only from a small triangular area between US highways 19 and 119 in the Elk and Gauley River drainages in central West Virginia. Material was examined as follows: WEST VIRGINIA. — Kanawha Co., 0.5 mi. N Falling Rock, Elk R. bluffs, M, 28 May 1952, L. Hubricht (RLH). Clay Co., 1-1.4 mi. S Clay, Elk R. bluffs, M, 3F, 28 May 1952, and 2M, 3F, 16 June 1956, L. Hubricht (RLH) and M, F, 7-9 September 1958, MacMillan and Richmond (RLH). TYPE LOCALITY. Nicholas Co., 4 mi NE Swiss, M, 2 September 1965, R.L. Hoffman (RLH). MEM. AMER. ENT. SOC., 35 44 XYSTODESMID MILLIPEDS Remarks. — Similarities between rigida and stenogon/nantahalae (Sigiria/Falloria), some 240 miles south in southwestern North Carolina are worth noting. The tooth arises suddenly at midlength of the peak, just as in nantahalae, and the double bend in the distal zone is similar to that in stenogon, although the latter is somewhat bisinuate. The only differences between rigida and mohicana are the tooth (present in rigida, absent in mohicana), the prefemoral process (present in rigida, ab- sent in mohicana), and a slightly higher and more rounded arch in ees Ge NS \ Par , q CES Nene Ror RNY ae 7 EE ee} TCP Fic. 17. Distribution of species of Sigmoria from Virginia northward, excepting the subgenus Dixioria. Ovals, /. latior; dots, t. trimaculata; half shaded dots, t. kleinpeteri; X, trimaculata intergrades; squares, guyandotta; triangles, rigida; diamonds, mohicana; asterisk, whiteheadi. Open symbols indicate literature records believed to be reliable. R.M. SHELLEY & D.R. WHITEHEAD 45 mohicana. The two are thus closely related and may eventually be shown to be subspecies. The hypothesized connection between ¢. kleinpeteri, guyan- dotta, and mohicana may therefore continue eastward and include rigida. If so, Rudiloria will contain only one species, and the forms will connect in a spiral configuration that curls inward and terminates centrally. SIGMORIA (DIXIORIA) Chamberlin, new status Dixioria Chamberlin, 1947:28. Hoffman, 1956a:6-7; 1979:158. Chamberlin and Hoffman, 1958:31. Jeekel, 1971:259. Type species. — Dixioria dentifer Chamberlin, 1947, by original designa- tion. Diagnosis. — Color pattern constant, paranota yellow, metaterga uniformly black, collum with concolorous yellow stripe along anterior margin; caudolateral corners of paranota rounded through midbody segments; gonopods in situ with acropodites extending mostly over opposite side of aperture and inserting beside opposite coxa, perhaps with sides lean- ing Over anterior margin but apices not projecting over latter nor between legs of segment 6; acropodites relatively thin and fragile, without flanges; tooth usually present, located on peak or distal zone; smaller accessory tooth usually present, arising from undersurface of acropodite at or distal to level of tooth; tip simple. Remarks. — The subgenus Dixioria occupies a roughly triangular area in the contiguous corners of North Carolina, Tennessee, and Virginia. The northernmost known locality is in Tazewell County, Virginia, and the Nolichucky River in North Carolina and Tennessee forms a sharp southern distributional limit. The cove dwelling species of the other subgenera do not occur north of this watercourse and are replaced by Dixioria. The only sym- patric congeners are /atior (Sigmoria) and t. kleinpeteri (Rudiloria), the former being an ecological generalist and thus able to occur outside coves. The subgenus Dixioria is distinguished by both somatic and gonopodal features. The caudolateral corners of the paranota are distinctly rounded over the anterior 2/3 of the body, and even the caudalmost are apically blunt rather than acute. In all other species groups, the caudolateral corners are rounded only through about segments 5-7, after which they become blunt and progressively more acute caudally to around segment 15. Thus, females as well as males can be accurately placed in Dixioria on the basis of this diagnostic trait. As in the subgenus Rudiloria the gonopods in situ ex- tend more mediad than anteriad, projecting over the opposite side of the aperture and often curving dorsad so that the distal zone of one acropodite MEM. AMER. ENT. SOC., 35 46 XYSTODESMID MILLIPEDS touches the coxa of the other. The gonopods are characterized by large prefemoral processes and relatively thin, fragile acropodites. A variable tooth is usually present on the distal extremity of the peak or proximal half of the distal zone, and the forms lacking it have a slightly expanded, rounded medial margin there. Medial and lateral flanges are absent. Six of the seven species display a smaller and sharply acute accessory tooth, which arises from the undersurface of the acropodite at or distal to the level of the main, marginal, tooth. This accessory tooth also is diagnostic for Dixioria. Hoffman (1956a) revised this assemblage, recognizing two species, one with six subspecies. Here I synonymize one subspecies and elevate the others to specific rank. Although the true relationship of acuminata to both pela and coronata in Tennessee remains to be settled, enough collecting has taken place in North Carolina to show that pela, wrighti, and coronata are not linked by intergrades as in /atior (Sigmoria). All have sharp distribu- tional boundaries and are therefore considered reproductively isolated, and a new species exists between wrighti and coronata. As brief descriptions were previously provided by Hoffman (1949, 1956a), I describe only pela, brooksi, dactylifera, and the new species. Supplemental accounts, describ- ing the process of the 4th sternum and the gonopods in ‘‘sigmoid’’ ter- minology, are provided for the others. Dixioria consists of a single species group, the pela group. Components. — pela (Chamberlin), acuminata (Hoffman), coronata (Hoffman), wrighti (Hoffman), watauga new species, brooksi (Hoffman), dactylifera (Hoffman). Sigmoria (Dixioria) pela (Chamberlin), new combination Figs. 18-24 Fontaria pela Chamberlin, 1918b:122-123. Attems, 1938:167. Dixioria dentifer Chamberlin, 1947:28-29, Fig. 13. Dixioria pela pela: Hoffman, 1956a:8-11, Fig. la. Chamberlin and Hoffman, 1958:31. Apheloria pela: Wray, 1950:44; 1967:44. Type specimens. — Male and female paratypes (RLH, MCZ) collected by R. Thaxter in July of unknown year at Burbank, Carter Co., TN. The holotype is missing from the latter repository where both Hoffman (1956a) Fics. 18-24. Sigmoira (Dixioria) pela. 18, process of 4th sternum of paratype caudal view. 19, gonopods in situ, ventral view of male from Mitchell Co., NC. 20, telopodite of left gonopod of paratype, medial view. 21, the same, lateral view. 22, telopodite of left gonopod of another paratype, medial view. 23, distal zone of male from 3.8 mi. SE Poplar, Mitchell Co., NC, medial view. 24, the same of male from 3.8 mi. SE Spruce Pine, Mitchell Co. Scale line for fig. 49 = 1.00 mm; line for other figs. = 1.00 mm for 23-24, 1.25 mm for 22, 1.43 mm for 18 and 21, and 1.66 mm for 20. R.M. SHELLEY & D.R. WHITEHEAD MEM. AMER. ENT. SOC., 35 47 48 XYSTODESMID MILLIPEDS and Chamberlin and Hoffman (1958) report its deposition. It may be among the Chamberlin material now being accessioned by the NMNH. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process long, thick, and Variably straight, curved, or bisinuate; acropodite thin and fragile, con- figuration a broad, smooth, continuous curve forming nearly complete circle with prefemoral process; anterior bend and apical curve broad, poorly defined; peak gently curved, apex near midlength; distal zone long, curving broadly dorsad and bending abruptly inward into arch distal to midlength; tooth present or absent, distomedial margin of distal zone slight- ly expanded and gently rounded when present, not expanded when absent, tooth located proximal to midlength of distal zone, length and configura- tion variable; accessory tooth present, arising from inner surface of distal zone at level of tooth, variably triangular to spiniform; lateral flange ab- sent. Color in Life. — Paranota yellow; metaterga black without stripes or spots; collum with broad yellow stripe along anterior margin connecting paranotal spots. Male Paratype. — Length 35.0 mm, maximum width 7.2 mm, W/L ratio 20.6% depth/ width ratio 68.0%. Segmental widths as follows: collum 5.5 mm 14th-15th 7.0 2nd 6.7 16th 6.1 3rd-12th 7.2 17th 5.2 13th 6.8 18th 4.8 Somatic features as in /. /atior, with following exceptions: Width across genal apices 4.1 mm, interantennal isthmus 1.6 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>6>4=5>1>7. Genae with distinct central impressions. Facial setae as follows: epicranial and interantennal absent, frontal 1-1, genal 2-2, clypeal about 8-8, labral about 12-12. Dorsum finely granulate, becoming smooth and polished on paranota, with very little wrinkling. Collum broad, ends not extending beyond those of following tergite. Paranota moderately depressed, continuing slope of dorsum, caudolateral corners rounded through seg- ment 11, blunt on 12-14, becoming progressively more acute posteriorly. Peritremata flat and indistinct, slightly elevated above paranotal surface; ozopores located caudal to midlength, opening dorsolaterad. Sternum of segment 4 with moderate process between 3rd legs, length equal to widths of ad- jacent coxae (Fig. 18); that of segment 5 with two low, medially coalesced knobs between 4th legs, much shorter than widths of adjacent coxae, and rounded, elevated areas between Sth legs; that of segment 6 with convex recession between 7th legs to accommodate apical cur- vatures of acropodites, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with bicruciform impressions on segments 8-10 and variably broad, shallow, central impressions on remaining segments. Coxae with low, blunt tubercles arising on segment 9, becoming pro- gressively longer and sharper caudally; prefemoral spines arising on segment 5, becoming pro- gressively longer and more acute posteriorly. Gonopodal aperture ovoid, 3.5 mm wide and 1.8 mm long at midpoint, without indenta- tions, sides elevated above metazonal surface. Gonopods in situ (Fig. 19, not this specimen) R.M. SHELLEY & D.R. WHITEHEAD 49 with acropodites projecting mediad and crossing each other in midline, curving broadly over opposite side of aperture then back toward midline, extending slightly beyond anterior margin. Gonopod structure as follows (Figs. 20-21): prefemoral process a moderately long, upright peg-like projection, directed toward peak and extending beyond level of tip of acropodite, of nearly equal width throughout, narrowing apically, with distal spur on lateral side. Acropodite relatively thin and fragile, configuration a smoothly curved arch overhanging and extending well beyond level of prefemoral process; basal zone moderately long and gently curved, without modifications; anterior bend broad, poorly defined, continuous with apical curve through peak; peak gently curved, apex at midlength; apical curve broad, poorly defined; distal zone long, projecting laterad from peak and not coplanar with basal zone, bending abruptly into arch distal to midlength and tapering to blunt tip; latter directed toward anterior bend. Medial and lateral flanges absent. Tooth moderately long, triangular, and apically subacuminate, located distal to midlength of distal zone; accessory tooth shorter, triangular, and apically acute, located at same position and obscured in medial view. Prostatic groove crossing to lateral side on prefemur and continuing to terminal opening. Other Male Paratypes. — The type males vary more than those in any other sample. The prefemoral process varies from bisinuate to gently curved, and the arch of the acropodite ex- tends to the level of the prefemoral process in some individuals and well beyond the latter in others. The tooth varies from distinct and sharply triangular to a small, rounded vestige, and the distal zone is slightly wider and has a gently curved medial edge distal to the teeth in the lat- ter condition (Fig. 22). Female Paratype. — Length 37.6 mm, maximum width 8.7 mm, W/L ratio 23.1%, depth/width ratio 65.5%. Agreeing closely with males in somatic features except coxal tubercles more sharply acute. Cyphopods in situ with sides of receptacles visible in aperture, valves directed laterad. Receptacle moderately large, cupped around medial ends of valves, surface rugulose. Valves moderate, equal, surface finely granulate. Variation. — Variation in the prefemoral process is as described for paratypes, but most individuals resemble the bisinuate condition. On the acropodites the arch is broader in some males, and the great majority lack the tooth. In these the accessory tooth is visible in medial view (figs. 23-24). About half of the available males have an expanded, rounded distomedial margin of the distal zone (Fig. 23), and about half have a narrow, linear edge (Fig. 24). There is no detectable geographic pattern to the distributions of these two forms. Ecology. — Sigmoria (D.) pela is a cove dwelling species. As stated previously by Shelley (198la), the mountains in northwestern North Carolina north of the Nolichucky-Toe Rivers are more segregated from each other and have fewer ridges and clusters of peaks than those south of this boundary. Consequently, cove environments are rarer, and most forms of Dixioria are found along streams under thin layers of leaves on relatively hard substrates. Distribution. — Sigmoria (D.) pela spans the Blue Ridge Province im- mediately north of the Nolichucky River, from near the escarpment in McDowell-Burke counties, North Carolina, to western Carter County, Ten- nessee, near the boundary with the Ridge and Valley Province. I have col- MEM. AMER. ENT. SOC., 35 50 XYSTODESMID MILLIPEDS lected extensively south of the Nolichucky and can confirm Hoffman’s prediction (1956a) that it forms the southern distributional limit. The range is only two counties wide in each state, and the northern limit does not cor- respond to any particular physiographic feature. Material was examined as follows: TENNESSEE. — Carter Co., 1 mi. NW Hampton, Doe River bluff, M, F, 3 May 1951, L. Hubricht (RLH); and Burbank, MM, FF, complete date unknown, R. Thaxter (MCZ, RLH) TYPE LOCALITY. Unicoi Co., 3 mi. E Erwin, co. rd. 2457 at Rock Creek Rec. Area, Cherokee Nat. For., 2M, F, 20 May 1978 (NCSM A1973). NORTH CAROLINA. — Avery Co., 4.4 mi. N Elk Park, along co. rd. 1305, 1.7 mi. N jet. co. rd. 1307, 2M, 8 September 1973 (NCSM 1999); 2.8 mi. N Elk Park, Elk Falls on co. rd. 1305, M, 29 October 1977 (NCSM A1774); Cranberry, M, 2F, June 1896, A. Wenzell (ANSP); 4 mi. N Newland, along US hwy, 19E, 0.8 mi. S. jct. co. rd. 1168, M, F, 20 June 1980 (NCSM A3323); 6.5 mi. W Newland, along co. rd. 1130, 0.8 mi. W. jct. US hwy. 19E, 3M, 2F, 20 June 1980 (NCSM A3320); and 4 mi. S Newland, along co. rd. 1143, 1.7 mi. NW jet. US hwy. 221, 3M, 20 June 1980 (NCSM A3324). Mitchell Co., 12 mi. NW Bakersville, along co. rd. 1320, 0.7 mi. N jet. co. rd. 1321, 0.8 mi. N jct. co. rd. 1320, 5M, F, 24 July 1975 (NCSM A409); 1.5 mi. N Poplar, along co. rd. 1323, 2.1 mi. N jct. co. rd. 1321, M, 21 May 1978 (NCSM A1976); 11 mi. N Spruce Pine, along NC hwy 261, 0.1 mi. W jct. co. rd. 1159, M, 21 June 1980 (NCSM A3325); 3 mi. NW Spruce Pine, along co. rd. 1164, 0.5 mi. W jet. co. rd. 1162, 2M, 2F, 21 June 1980 (NCSM A3331); and 3.8 mi. SE Spruce Pine, along co. rd. 1129, 0.4 mi. N jet. co. rd. 1128, 2M, 24 June 1975 (NCSM A408). McDowell-Burke cos. — county line on US hwy. 221, 3F, 22 June 1980 (NCSM A3332). Remarks. — The female only sample from the last locality is tentatively assigned to pela, the only proximal species. The holotype of D. dentifer, said by Hoffman (1956a) and Chamberlin and Hoffman (1958) to be at the AMNH, is actually at the ANSP. Sigmoria (Dixioria) acuminata (Hoffman), new combination, new status Figs. 25-27 Dixioria pela acuminata Hoffman, 1956a:11-12, Fig. 1b. Type specimen. — Male holotype (NMNH) collected by J.A. Fowler and R.L. Hoffman, 19 June 1950, from Johnson Co., TN, at the summit of Holston Mountain, 2 mi. W of Shady Valley. There are no paratypes. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process long, thick, and Fics. 25-27. Sigmoria (Dixioria) acuminata. 25, gonopods in situ, ventral view of holo- type. 26, telopodite of left gonopod of holotype, medial view. 27, distal zone of the same, lateral view. Scale line for fig. 25 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 26 MEM. AMER. ENT. SOC., 35 27 SI 52 XYSTODESMID MILLIPEDS curved; acropodite thin and fragile, configuration a flattened, open arch; anterior bend broad, poorly defined; peak flattened and linear with distal margins slightly expanded and gently rounded; apical curve sharp, well defined; distal zone short, bending sharply into arch, lying parallel to and beneath distal extremity of peak; tooth absent; accessory tooth present, located distad on peak; lateral flange absent. Color in Life. — Paranota yellow; metaterga black without stripes or spots; collum with broad yellow stripe along anterior margin connecting paranotal spots. Holotype. — Process of 4th sternum moderately long, equal in length to widths of adjacent coxae. Gonopods in situ (Fig. 25) with acropodites projecting mediad and crossing each other in midline, curving broadly over opposite side of aperture and extending slightly beyond anterior margin. Gonopod structure as follows (Figs. 26-27): prefemoral process moderately long, upright, indented and curving slightly near midlength, tapering distad to subacuminate tip, directed toward peak. Acropodite relatively thin and fragile, configuration a flattened, open arch extending just beyond level of prefemoral process; basal zone moderately long, gently curved, without modifications; anterior bend broad, poorly defined; peak flattened, linear, distal margins slightly expanded and rounded; apical curve sharp, well defined; distal zone short, not coplanar with basal zone, curving sharply into arch and lying parallel to and beneath distal extremity of peak; tip acuminate. Tooth and medial and lateral flanges absent. Ac- cessory tooth present, arising from inner surface of peak at level of expanded medial margin. Prostatic groove crossing to lateral side on prefemur and continuing to terminal opening. Variation. — The male from the second locality agrees closely with the holotype. Distribution. — Known only from Holston Mountain in western Johnson County, Tennessee. The following specimen was examined in addi- tion to the holotype: TENNESSEE. — Johnson Co., 3 mi. S Shady Valley, along TN hwy. 91, M, 17 May 1964, R.L. Hoffman (RLH). Remarks. — This is the westernmost form of Dixioria. Its true status awaits more sampling in southern Johnson and northern Carter counties, Tennessee, and it might be a subspecies of pela, forms of which also lack the tooth and have a slightly expanded distomedial margin of the peak. As Holston Mountain is near the western periphery of the Blue Ridge Province, future investigators should also look for acuminata in the adjacent Ridge and Valley Province. Just as coronata occurs in this region farther north, acuminata may inhabit it on the western side of the subgeneric range. Sigmoria (Dixioria) coronata (Hoffman), new combination Figs. 28-31 Deltotaria coronata Hoffman, 1949:380-381, Pl. 26, Figs. 7-8. Dixioria pela coronata: Hoffman, 1956a:12-13, Fig. 1c. Chamberlin and Hoffman, 1958:31. Dixioria pela fowleri Hoffman, 1956a:13-14, Fig. le. NEW SYNONYMY. R.M. SHELLEY & D.R. WHITEHEAD 53 Type specimens. — Male holotype, female allotype, and male paratype (NMNH) collected by H.I. Kleinpeter and R.L. Hoffman, 1 July 1947, from Mt. Rogers, Grayson Co., VA. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process long, variably curved to bisinuate; acropodite moderately thick and heavy, configuration a flat- tened, open arch; anterior bend broad, poorly defined; peak linear with margins essentially straight; apical curve sharp, well defined; distal zone short, curving gently into arch and nearly overlapping tooth; latter located on distal extremity of peak, apically rounded and moderately long, nearly equal in length to distal zone; accessory tooth present, arising from inner surface of peak beside base of tooth; lateral flange absent. Color in Life. — Paranota yellow; metaterga black without stripes or spots; collum with broad yellow stripe along anterior margin connecting paranotal spots. Holotype. — Process of 4th sternum moderately long, equal in length to widths of adjacent coxae. Gonopods in situ (Fig. 28, ventral view of paratype) with acropodites projecting mediad and crossing each other near midlengths in midline, curving broadly over opposite side of aperture and extending slightly beyond anterior margin. Gonopod structure as follows (Figs. 29-30): prefemoral process relatively long, upright, curving slightly into arch near midlength, apically blunt, directed toward tooth. Acropodite moderately thick and heavy, configuration a flat- tened, open arch extending to level of prefemoral process; basal zone moderately long, gently curved, without modifications; anterior bend broad, poorly defined; peak flattened and linear, margins straight; apical curve sharp, well defined; distal zone short, curving gently into arch and nearly overlapping tooth, not coplanar with basal zone; tip subacuminate. Medial and lateral flanges absent. Tooth moderately long, apically rounded, located distad on peak; ac- cessory tooth shorter, triangular, and apically acute, located at same position and obscured in medial view. Prostatic groove crossing to lateral side on prefemur, continuing to terminal opening. Variation. — The distal zone does not curve inward into the arch in northern males (Fig. 31). Otherwise, all gonopods agree closely with those of the holotype. Ecology. — Sigmoria (D.) coronata is a cove dwelling species. Distribution. — The Watauga River appears to form the southern boundary for this form, which ranges northward into the Ridge and Valley Province to the northern part of southwestern Virginia. It is the only species of Dixioria occurring in all three states. Material was examined as follows: VIRGINIA. — Tazewell Co., Burkes Garden, spring on NE slope of Beartown Min., M, 28 August 1977, R.L. Hoffman (RLH). Bland Co., 10 mi. SW Bland, along US hwy. 52 near summit of Big Walker Mtn., 2M, F, 24 June 1950, R.L. Hoffman and J.A. Fowler (NMNH). Wythe Co., Big Walker Mtn., M, 5 May 1955, collector unknown (RLH). Grayson Co., Mt. Rogers, 2M, F, 1 July 1947, R.L. Hoffman and H.I. Kleinpeter (NMNH) and 2M, 24 August MEM. AMER. ENT. SOC., 35 54 XYSTODESMID MILLIPEDS 1978, D.W. Ogle (RLH) TYPE LOCALITY; and Comer’s Rock Rec. area, 2M, F, 15 June 1950, L. Hubricht (RLH). Washington Co., 4 mi. SW Konnarock, base of Laurel Mtn., 2M, F, 28 April 1951, L. Hubricht (RLH). TENNESSEE. — Johnson Co., Backbone Rock Rec. Area., Cherokee Nat. For., approx. 4 mi. S Damascus, VA, 2M, F, 11 July 1962, and M, 2 June 1974, R.L. Hoffman (RLH); and along US hwy. 421, 1 mi. W NC state line, several MM, FF, 1 May 1965, collector unknown (RLH). NORTH CAROLINA. — Ashe Co., 14 mi. NW Jefferson, along co. rd. 1359, 0.5 mi. S. jet. co. rd. 1360, 4M, 3F, 17 June 1980 (NCSM A3295); 12 mi. NW Jefferson, along co. rd. 1356, 0.8 mi. SE jet. co. rd. 1358, 4M, 17 June 1980 (NCSM A3297); 10 mi. NW Jefferson, along co. rd. 1319, 1.7 mi. S jet. co. rd. 1320, M, 21 July 1972 (NCSM 1180); 8.5 mi. SW Jef- Fics. 28-31. Sigmoria (Dixioria) coronata. 28, gonopods in situ, ventral view of paratype. 29, telopodite of left gonopod of holotype, medial view. 30, the same, lateral view. 31, peak and distal zone of male from Bland Co., VA, medial view. Scale line for fig. 28 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 55 ferson, along co. rd. 1100 at Long Hope Cr., 3M, 2F, 17 June 1980 (NCSM A3299); Bina, several MM, FF, 17 May 1964, R.L. Hoffman (RLH); and between Idlewood and Gap Cr., 2M, F, 1 May 1965, collector unknown (RLH). Watauga Co., 3 mi. SE Zionville, headwaters of Meat Camp Cr., M, 3F, 5 September 1971, R.L. Hoffman and L.S. Knight (RLH); 8 mi. NW Boone, along co. rd. 1213, 1.4 mi. SE jct. co. rd. 1201, 4M, 3F, 19 June 1980 (NCSM A3308); and 4 mi. N Vilas, along US hwy. 421, M, 11 June 1962, R.L. Hoffman (RLH). Remarks. — As shown in figures 29 and 31, the types of pela fowleri closely resemble those of coronata. The slightly wider gap between the tooth and distal zone could result from a slightly different orientation of the structure for drawing. This is clearly insignificant and p. fowleri is placed in synonymy under coronata, making it the northernmost representative of Dixioria. It occurs sympatrically and perhaps syntopically with brooksi at Backbone Rock Recreation Area, Johnson County, Tennessee. Sigmoria (Dixioria) wrighti (Hoffman), new combination, new status Figs. 32-34 Dixioria pela wrighti Hoffman, 1956a:15, Fig. 1f. Type specimens. — Male holotype (NMNH) collected by R.L. Hoffman, 3 August 1949, from Avery Co., NC, approximately 5 mi NE Linville, on the east side of Grandfather Mountain. The female allotype and male paratype, which Hoffman (1956a) reported to be in the NMNH, were not examined. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process long, curved caudad apically; acropodite thin and fragile, configuration a variably continuous curve directed downward to or below level of prefemoral process, forming complete or nearly complete circle; anterior bend broad, poorly defined; peak variably rounded, apex near midlength; apical curve broad, poorly defined, distal zone long, not coplanar with basal zone, curving broadly dorsad and more abruptly into arch distal to midlength; tooth present near midlength of distal zone, relatively long and nearly overlapping level of tip of acropodite, configuration variably quadriform with sides and distal margin irregular to serrate; accessory tooth arising at level of tooth, triangular to spiniform; lateral flange absent. Color in Life. — Paranota yellow; metaterga black without stripes or spots; collum with broad yellow stripe along anterior margin connecting paranotal spots. Holotype. — Process of 4th sternum moderately long, equal in length to widths of adjacent coxae. MEM. AMER. ENT. SOC., 35 56 XYSTODESMID MILLIPEDS Gonopods in situ (Fig. 32, not this specimen) with acropodites projecting mediad and cross- ing each other near midlengths in midline, curving broadly over opposite sides of aperture then back toward midline with tips overlapping, curvatures located entirely within aperture. Gonopod structure as follows (Fig. 33-34): prefemoral process relatively long, moderately thick, bent caudad apically, tip subacuminate, directed toward anterior bend. Acropodite relatively thin and fragile, configuration a smoothly continuous curve overhanging and extend- ing well beyond level of prefemoral process, curving downward to or below level of latter forming nearly complete circle; basal zone relatively short, broadly curved; anterior bend broad, poorly defined; peak gently rounded, apex in midlength, sides not expanded; apical curve broad, poorly defined; distal zone moderately long, curving broadly into arch and ex- tending beyond level of tooth, not coplanar with basal zone; tip subacuminate, directed toward anterior bend. Medial and lateral flanges absent. Tooth located near midlength of distal zone, relatively long, sides parallel, with four apical dentations; accessory tooth arising from inner surface of distal zone at same location, shorter and apically acute; partly obscured in medial view. Prostatic groove crossing to lateral side on prefemur, continuing to terminal opening. Variation. — The degree of jaggedness or serration of the tooth varies on each individual, and there may be from three to five marginal dentations. The acropodite arch is also flattened more in some males, but the distal zone always extends to near the level of the prefemoral process. Ecology. — Sigmoria (D.) wrighti is a cove dwelling species. Distribution. — A narrow area in the Blue Ridge Province between the Linville and Watauga rivers, extending onto the escarpment in Caldwell County, North Carolina. Specimens were examined as follows: TENNESSEE. — Johnson Co., along US hwy. 321, 1 mi. W NC state line, 2M, 3F, 1 May 1965, collector unknown (RLH). NORTH CAROLINA. — Watauga Co., 2 mi. W Sugar Grove, along US hwy. 321, M, 1 May 1965, collector unknown (RLH); 9 mi. W Boone, along co. rd. 1157, 0.1 mi. S jct. co. rd. 1123, 4M, F, 19 June 1980 (NCSM A3314); 7 mi. SW Boone, along NC hwy. 105, 0.8 mi. SW jet. co. rd. 1559, 2M, 19 June 1980 (NCSM A3315); and along US hwy 221 near Caldwell Co. line, M, 31 July 1931, A.W. Petrunkevitch (PMNH). Avery Co., 4.1 mi. S Elk Park, along NC hwy. 194, 3M, 3F, 20 May 1956, R.L. Hoffman (RLH); between Linville and Banner Elk, M, 15 June 1953, R.L. Hoffman (RLH); 13.5 mi. N Newland, jct. US hwy. 321 and co. rd. 1316, M, 19 June 1980 (NCSM A3310); 7 mi. E Newland, along co. rd. 1514, 1 mi. E jet. US hwy. 221, 2M, 20 June 1980 (NCSM A3316); 8 mi. SW Newland, along co. rd. 1530, 0.4 mi. N jet. co. rd. 1511, M, F, 20 June 1980 (NCSM A3318); and 5 mi. NE Linville, along US hwy. 221 on E side Grandfather Mtn., M, 3 August 1949, R.L. Hoffman (NMNH); TYPE LOCALITY. Caldwell Co., Edgemont, 2M, F, 22 June 1980 (NCSM A3333). Fics. 32-36. 32-34, Sigmoria (Dixioria) wrighti. 32, gonopods in situ, ventral view of male from Watauga Co., NC. 33, telopodite of left gonopod of holotype, medial view. 34, the same, lateral view. 35-36, Sigmoria (Dixioria) watauga, holotype. 35, telopodite of left gonopod, medial view. 36, the same, lateral view. Scale line for fig. 32 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD S7/ 33 MEM. AMER. ENT. SOC., 35 58 XYSTODESMID MILLIPEDS Sigmoria (Dixioria) watauga Shelley, new species Figs 35-36 Type specimens. — Male holotype (NCSM 2008) and five male and seven female paratypes collected by R.M. Shelley, 8 September 1973, from yard of residence on Goforth Rd., 0.5 mi. N US highway 321, Blowing Rock, Watauga Co., NC. Male and female paratypes deposited in FSCA. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process relatively long, bisinuately curved; acropodite moderately thick and heavy, configuration a smooth continuous curve extending slightly beyond level of prefemoral process and directed downward toward latter; anterior bend and apical curve broad, poorly defined; peak gently curved, apex at midlength; distal zone moderately long, not coplanar with basal zone, directed downward toward prefemoral process; tooth present near midlength of distal zone, relatively long, apically subacuminate and irregular; accessory tooth short and triangular, arising distal to latter and fully visible in medial view; lateral flange absent. Color in Life. — Paranota yellow; metaterga black without stripes or spots, collum with broad yellow stripe along anterior margin connecting paranotal spots. Holotype. — Length 38.2 mm, maximum width 8.0 mm, W/L ratio 20.9%, depth/width ratio 67.5%. Segmental widths as follows: collum 6.8 mm 15th 7.6 2nd 7.2 16th 6.8 3rd 7.8 17th 6.0 4th-14th 8.0 18th 5.0 Somatic features as in /. /atior, with following exceptions: Width across genal apices 3.9 mm, interantennal isthmus 1.6 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>4>5=6>1>7. Genae with faint central impressions. Facial setae as follows: epicranial and interantennal absent, frontal 1-1, genal 2-2, clypeal about 10-10, labral about 16-16. Dorsum smooth, polished, becoming slightly coriaceous on anterior halves of paranota. Collum broad, ends extending well below those of following tergite. Paranota moderately depressed, continuing slope of dorsum, caudolateral corners rounded through segment 12, blunt on segments 13-16. Peritremata distinct, strongly elevated above paranotal surface, ozopores located caudal to midlength, opening dorsolaterad. Sternum of segment 4 with moderate process between 3rd legs, length equal to widths of ad- jacent coxae; 5th sternum with two low, rounded, medially coalesced knobs between 4th legs and more separated, equivalent projections between Sth legs; 6th sternum convexly recessed between 7th legs to accommodate apical curves of acropodites, 7th legs set slightly farther apart than 6th. Postgonopodal sterna generally flat, with variably broad, shallow, central im- pressions. Coxae with low, rounded tubercles beginning on segment 16; prefemoral spines beginning on segment 6, becoming progressively longer and sharper caudally. Gonopodal aperture elliptical, 3.9 mm wide and 1.4 mm long at midpoint, without indenta- tions, sides elevated above metazonal surface. Gonopods in situ with acropodites projecting mediad and crossing in midline, curving broadly over opposite side of aperture and extending R.M. SHELLEY & D.R. WHITEHEAD 59 dorsad, lying entirely over aperture. Gonopod structure as follows (Figs. 35-36): prefemoral process relatively long, bisinuate, apically subacuminate. Acropodite moderately thick and heavy, configuraton a smooth, continuous curve extending slightly beyond level of prefemoral process and curving downward in medial view (dorsad); basal zone moderately long, gently curved; anterior bend broad, poorly defined, continuous through peak with apical curve; peak relatively short, gently curved, apex at midlength; apical curve broad, poorly defined; distal zone moderately long, curving slightly laterad from peak and not coplanar with basal zone, projecting downward in medial and lateral views and extending to level of prefemoral process; tip subacuminate, directed toward coxa. Medial and lateral flanges absent. Tooth moderately long, apically subacuminate and irregular, located near midlength of distal zone; accessory tooth short and acute, arising distal to tooth and fully visible in medial view. Prostatic groove crossing to lateral side on prefemur and continuing to terminal opening. Male Paratypes. — The male paratypes agree with the holotype in all particulars. Female Paratype. — Length 39.4 mm, maximum width 8.4 mm, W/L ratio 21.3%, depth/ width ratio 69.0%. Cyphopods in situ with openings of valves and corners of receptacles visible in aperture. Receptacle moderate, cupped around lateral sides of valves, surface rugulose. Valves small, equal, surfaces finely granulate. Variation. — The gonopods are all quite similar. The tooth is slightly longer in western populations, and the midlength bend of the prefemoral process is considerably broader in that from Wilkes County. Ecology. — Sigmoria (D.) watauga is primarily a cove species, although it can occur away from water sources. The type specimens were found under ivy, in moist litter, in corners of walkways, and in the basement of the residence at Blowing Rock. Distribution. — A triangular area in the Blue Ridge Province of North Carolina, mostly in Watauga County. The species is common in the com- munities of Boone and Blowing Rock. Specimens were examined as follows: NORTH CAROLINA. — Watauga Co., 12 mi, WNW Boone, along US hwy. 321, 1.5 mi. N Avery Co. line, 2M, 129 June 1980 (NCSM A3309); Boone, Appalachian State Univ. cam- pus, 2M, 3F, 26 April 1974 (NCSM 2212); 1.9 mi. S Boone, along co. rd. 1549, 0.7 mi. jet. co. rd. 1550, 4M, 26 April 1984 (NCSM 2216); 6.2 mi. E Boone, along co. rd. 1508, 1.2 mi. S jet. Blue Ridge Pkwy., 4M, 3F, 18 June 1980 (NCSM A3303); 2.2 mi. N Blowing Rock, along co. rd. 1541, 1.5 mi. N ject. co. rd. 1552, 2M, 26 April 1974 (NCSM 2221); and Blowing Rock, residence on Goforth Rd., 0.5 mi. N jct. US hwy. 321, 6M, 7F, 8 September 1973 (NCSM 2008) TYPE LOCALITY. Wilkes Co., 18.1 mi. WNW Wilkesboro, Jeffress Pk. Recreation area on Blue Ridge Pkwy., 4M, 2F, 18 June 1980 (NCSM A3302). Remarks. — The male from Boone that Hoffman (1956a) considered an intergrade between pela wrighti and p. coronata is referable to watauga. Sigmoria (Dixioria) brooksi (Hoffman), new combination Figs. 37-40 Dixioria pela brooksi Hoffman, 1956a:12, Fig. 1d. Dixioria brooksi Hoffman, 1969:227. MEM. AMER. ENT. SOC., 35 60 XYSTODESMID MILLIPEDS Type specimens. — Male holotype and male paratype (NMNH) collected by Dr. and Mrs. S.T. Brooks, 14 August 1941, on Holston Mountain near Damascus, Washington Co., VA. Hoffman (1956a) stated that male and female topoparatypes were deposited in the Carnegie Museum, Pittsburgh, PA, but I did not find them in a visit there in 1978. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process long, thick, and linear, tapering distad to acuminate tip; acropodite moderately thick and heavy, configuration an upright, open arch with apex overhanging level of pre- femoral process; anterior bend sharp, well defined; peak moderately long and straight, outer margin expanded distad into broadly rounded lobe forming apex of arch; distal zone short, terminating well above prefemoral process, directed laterad from peak and not coplanar with basal zone, sides tapering to acuminate tip directed toward basal zone; medial and lateral flanges absent; tooth present on distal extremity of peak, short and truncate, extending to same level as tip of distal zone, distal margin linear, angling to lower inner corner; accessory tooth present, short and acute, arising from base of tooth, obscured by latter in medial view. Color in Life. — Unknown. Hoffman (1956a) did not report this, and there is no fresh material. However, faded markings on the preserved specimens indicate the coloration of other members of Dixioria. Holotype. — Length 35.2 mm, maximum width 7.6 mm, W/L ratio 21.6%, depth/width ratio 60.5%. Segmental widths as follows: collum 5.4 mm 13th-15th 7.0 2nd 6.9 16th 6.8 3rd 7.2 17th 6.1 4th-12th 7.6 18th 4.9 Somatic features similar to those of /. /atior, with following exceptions: Width across genal apices 3.9 mm, interantennal isthmus 1.4 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>4=5=6>1>7. Genae with faint central impressions. Facial setae as follows: epicranial and interantennal absent, frontal 1-1, genal 2-2, clypeal about 12-12, labral about 18-18. Dorsum smooth, polished, slightly coriaceous on anterior halves of paranota. Collum broad, ends extending slightly beyond those of following tergite. Paranota moderately de- pressed, continuing slope of dorsum, caudolateral corners rounded through segment 13, blunt on 13-15, acute on remaining segments. Peritremata moderately distinct, slightly elevated above paranotal surface; ozopores located caudal to midlength, opening dorsolaterad. Sternum of segment 4 with moderate process between 3rd legs, length equal to widths of ad- jacent coxae (Figs. 37); that of segment 5 with slight medial process between 4th legs, much shorter than widths of adjacent coxae, and low, rounded, elevated areas between Sth legs; that of segment 6 with moderate, convex recession between 7th legs to accommodate apical cur- vatures of acropodites, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with bicruciform impressions on segments 8-10, flattened and plate-like thereafter. Coxae with low, R.M. SHELLEY & D.R. WHITEHEAD 61 blunt tubercles arising on segment 13; prefemoral spines arising on segment 8, becoming pro- gressively longer and sharper caudally. Gonopodal aperture ovoid, 3.2 mm wide and 1.9 mm long at midpoint, indented slightly anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 38, not this specimen) with acropodites angling mediad and apices lying over and under each other in midline, extending forward slightly beyond anterior margin. Gonopod structure as follows (Figs. 39-40): prefemoral process relatively long, linear, tapering distad to acuminate tip, directed toward distal zone. Acropodite moderately thick and heavy, well sclerotized, arch upright, bent abruptly at 2/3 length and extending to but not beyond level of prefemoral Fics. 37-40. Sigmoria (Dixioria) brooksi. 37, process of 4th sternum of holotype, caudal view. 38, gonopods in situ, ventral view of male from Johnson Co., TN. 39, telopodite of left gonopod of holotype, medial view. 40, the same, lateral view. Scale line for fig. 38 = 1.00 mm; line for other figs. = 1.00 mm for 37, 1.23 mm for 39-40. MEM. AMER. ENT. SOC., 35 62 XYSTODESMID MILLIPEDS process; basal zone relatively long, nearly straight; anterior bend sharp, well defined; peak moderately long and linear, outer margin expanded distad into rounded lobe forming highest point on acropodite arch; apical curve sharp, well defined; distal zone short, angling slightly laterad from peak, not coplanar with basal zone, directed downward toward prefemur then bending into arch, nearly overlapping tooth; tip subacuminate, directed toward basal zone. Medial and lateral flanges absent. Tooth short, apically broad and truncate, outer corner slightly produced, located at distal extremity of peak; accessory tooth equal in length to, and arising from base of, tooth, obscured in medial view, apically acuminate. Prostatic groove crossing to lateral side at anterior bend, continuing to terminal opening. Male Paratype. — The male paratype agrees with the holotype in all particulars. Female Paratype. — As the specimen supposedly at the Carnegie Museum could not be located, the female characters of brooksi are unknown. Variation. — The male from Tennessee agrees closely with the holotype. Distribution. — Known from the type locality and the following site, which is only about four miles southeast. Remarks. — Although brooksi possesses an accessory tooth and resembles coronata in the general form of the acropodite and the short distal zone that curves inward, nearly overlapping the tooth, it is not con- specific. The two occur sympatrically at Backbone Rock Recreation Area and hence are reproductively isolated. The male of brooksi was even discovered in a vial with one of coronata, suggesting that they are syntopic. The course of the prostatic groove differs, as it crosses to the lateral surface at the anterior bend in brooksi and on the prefemur in coronata. Likewise, they are distinguished by the configuration of the peak, which is expanded into a distolateral lobe in brooksi and is nearly linear in coronata. The lobe in brooksi culminates the tendency toward enlargement at this position found in dactylifera and acuminata. It is similar in appearance to, and ap- parently convergent with, that of translineata (Falloria) (compare Figs. 39-40 with Figs. 68-69 in Shelley (1981a)). TENNESSEE. — Johnson Co., Backbone Rock Recreation Area, Cherokee National Forest, along TN hwy. 41 about 4 mi. S of Damascus, VA and the state line, M, 2 June 1974, R.L. Hoffman and L.S. Knight (RLH). Remarks. — Although brooksi possesses an accessory tooth and resembles coronata in the general form of the acropodite and the short distal zone that curves inward, nearly overlapping the tooth, it is not con- specific. The two occur sympatrically at Backbone Rock Recreation Area and hence are reproductively isolated. The male of brooksi was even discovered in a vial with one of coronata, suggesting that they are syntopic. The course of the prostatic groove differs, as it crosses to the lateral surface at the anterior bend in brooksi and on the prefemur in coronata. Likewise, they are distinguished by the configuration of the peak, which is expanded into a distolateral lobe in brooksi and is nearly linear in coronata. The lobe aaa. -—-—t—C SS R.M. SHELLEY & D.R. WHITEHEAD 63 in brooksi culminates the tendency toward enlargement at this position found in dactylifera and acuminata. It is similar in appearance to, and ap- parently convergent with, that of translineata (Falloria) (compare Figs. 39-40 with Figs. 68-69 in Shelley (1981a)). Sigmoria (Dixioria) dactylifera (Hoffman), newcombination _ Figs. 41-44 Dixioria dactylifera Hoffman, 1956a:15-16,Fig. 2. Type specimens. — Male and female paratypes (AMNH, NMNBH) col- lected by C.M. and R.D. Breeder, 22 August 1910, at Mill Hill (precise loca- tion unknown), Ashe Co., NC. The male holotype said by Hoffman (1956a) to be deposited in the AMNH is not there and apparently is lost. Diagnosis. — A moderate-size species of Sigmoria without medial flange and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process moderately long and gently curved ventrad, apically subacuminate; acropodite moderately thick and heavy, configuration an open arch leaning over and extending slightly beyond level of prefemoral process; anterior bend broad, poorly defined; peak relatively long, margins essentially straight; apical curve sharp, well defined; distal zone short, terminating well above level of prefemoral process, directed laterad from peak and not coplanar with basal zone, sides tapering to acuminate tip directed parallel to prefemoral process; medial and lateral flanges absent; tooth present on distal extremity of peak, relatively long, extending below level of distal zone, sides parallel, distal margin linear, inner apical corner produced to form subacuminate tip pro- jecting inward into arch and directed toward basal zone; accessory tooth absent. Color in Life. — Paranota bright lemon yellow; metaterga dark glossy black without stripes or spots; collum with narrow yellow stripe along anterior margin connecting paranotal markings. Male Paratype. — Length 56.9 mm, maximum width 7.4 mm, W/L ratio 20.1%, depth/width ratio 60.8%. Segmental widths as follows: collum 5.9 mm 15th 7.2 2nd 6.5 16th 6.8 3rd 7.0 17th 5.8 4th-14th 7.4 18th 4.2 Somatic features similar to those of /. /atior, with following exceptions: Width across genal apices 3.9 mm, interantennal isthmus 1.7 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>5>6>4>1>7. Genae with faint central impressions. Facial setae as follows: epicranial and interantennal absent, frontal 1-1, genal 2-2, clypeal about 9-9, labral about 12-12, merging with clypeal series and continuing for short distance along genal borders, about 3 setae per side. MEM. AMER. ENT. SOC., 35 64 XYSTODESMID MILLIPEDS Dorsum smooth, polished, slightly coriaceous on anterior halves of paranota and extending mediad along margins of strictures. Collum broad, ends extending slightly beyond those of following tergite. Paranota moderately depressed, continuing slope of dorsum, caudolateral corners rounded through segment 13, blunt on 14-16, acute on remaining segments. Peritremata moderately thick and distinct, well elevated above paranotal surfaces, ozopores located caudal to midlength, opening dorsolaterad. Sternum of segment 4 with small process between 3rd legs, shorter than widths of adjacent coxae (Fig. 41); that of segment 5 with low projection between 4th legs, much shorter than 43 A \ 44 Fics. 41-44. Sigmoria (Dixioria) dactylifera. 41, process of 4th sternum of paratype, caudal view. 42, gonopods in situ, ventral view of male from Mt. Jefferson St. Pk., Ashe Co., NC. 43, telopodite of left gonopod of paratype, medial view. 44, the same, lateral view. Scale line for fig. 42 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 65 widths of adjacent coxae, and higher, flattened, elevated areas between Sth legs; that of seg- ment 6 with deep, convex depression between 7th legs to accommodate apical curvatures of acropodites, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with bicruciform impressions on segments 8-11, with variably broad, shallow, central depressions on remaining segments. Coxae with low blunt tubercles arising on caudal legs of segment 13; prefemoral spines arising on segment 6, becoming longer and sharper caudally. Fic. 45. Distribution of the subgenus Dixioria in North Carolina, Tennessee, and Virginia. Stars, pela; triangles, acuminata; dots, coronata; squares, wrighti; asterisks, watauga; diamonds, brooksi, stars in dots, dactylifera. Open symbols represent unconfirmed records from Hoffman (1956a) that are believed to be correct. The dashed lines show the approximate boundaries of the Blue Ridge Province. MEM. AMER. ENT. SOC., 35 66 XYSTODESMID MILLIPEDS Gonopodal aperture ovoid, 3.3 mm wide and 1.8 mm long at midpoint, indented slightly aneriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 42, not this specimen) with acropodites crossing in midline of aperture, curving slightly over opposite side then back toward midline with apices overlapping, extending slightly beyond anterior margin. Gonopod structure as follows (Figs. 43-44): Prefemoral process moderately long, thick, and heavy, curving broadly ventrad and directed toward tooth, tip subacuminate. Acropodite moderately thick and heavy, arch broadly curved, leaning over and extending slightly beyond level of prefemoral process; basal zone short, broadly curved; anterior bend broad, poorly defined; peak relatively long and straight, apex near midlength; apical curve sharp, well de- fined; distal zone short, curving strongly laterad from peak and not coplanar with basal zone, directed downward from peak in medial view with only very slight curve; tip acuminate, directed parallel to prefemoral process. Medial and lateral flanges absent. Tooth long, exten- ding below level of distal zone, directed linearly from distal extremity of peak with sides parallel, inner apical corner produced to form acuminate tip projecting inward into arch and directed toward basal zone; accessory tooth absent. Prostatic groove crossing to lateral side on prefemur, continuing on this surface to terminal opening. Male Paratypes. — The male paratypes agree with the holotype. Female Paratype. — Length 37.2 mm, maximum width 7.9 mm, W/L ratio 21.2%, depth/ width ratio 64.6%. Cyphopods in situ with receptacle visible in aperture, valves directed laterad. Receptacle large, hood-like, enveloping medial and ventral surfaces of valves, surface rugulose. Valves moderate, equal, surfaces finely granulate. Variation. — All males of dactylifera are nearly identical to each other. Ecology. — The male I collected in Mt. Jefferson State Park was dis- covered at the picnic area near the summit under moist litter in a deciduous forest. The site lacked rhododendron and a water source, and could not be classified as a cove. Distribution. — Known only from three localities in the northwestern corner of North Carolina. Material was examined as follows: NORTH CAROLINA. — Ashe Co., Mill Hill, exact location unknown, 6M, 5F, 22 August 1910, C.M. and R.D. Breeder (AMNH, NMNH) TYPE LOCALITY; 2 mi. SE Creston, Three- Top Mtn., 4M, 4F, 23 July 1963, R.L. Hoffman (RLH); and Mt. Jefferson St. Pk., M, 29 August 1976 (NCSM A1412). Remarks. — The principle diagnostic features of dactylifera are the distinctive configuration of the tooth and the absence of the accessory tooth. SIGMORIA (SIGMORIA) Chamberlin, new status Sigmoria Chamberlin, 1939:7. Hoffman, 1950:1-2; 1979:158. Chamberlin and Hoffman, 1958:49. Jeekel, 1971:287. Shelley, 1981a:16-18. Type species. — Sigmoria munda Chamberlin, 1939, by original designa- tion. SS R.M. SHELLEY & D.R. WHITEHEAD 67 Diagnosis. — Paranota yellow to red, metatergal color pattern variable, either uniformly black or with concolorous stripes connecting paranotal markings; gonopods in situ with acropodites extending well beyond anterior margin of aperture and inserting between 7th legs; acropodites moderately thick, oriented normally on coxa with inner surface directed anteriomediad; basal zone without modifications; with variable laminate medial flange located on peak; with or without variable lateral flange and variable tooth on peak distal to flange; tip either reflexed or with this configuration. Remarks. — The nominate subgenus occupies a broad area from southern West Virginia to the Savannah River and the Atlantic Ocean. In the north it overlaps all of Dixioria and the southern extremity of Rudiloria, and in the south it overlaps the South Carolina forms of Cleptoria and all species of Croatania except yemassee. Although less common in the moun- tains, it ranges westward into the central Blue Ridge Province and overlaps parts of Sigiria and Cheiropus. Most of its range is due to the widespread, ubiquitous /atior, a clinally continuous species composed of three geographic races connected through a broad area of intergrades (Shelley 1976, 1981a). In the western Piedmont of North Carolina occurs a second species, stenoloba, that apparently arose from /atior by ecological isolation and is comprised of two allopatric populations. Two additional, localized species occur in the eastern Piedmont and Fall Zone of southern South Carolina. They are sufficiently different from /atior and stenoloba that I divide Sigmoria s. str. into two species groups. The Latior Group In the /atior group, the acropodites are moderately thick and heavy, the anterior bend and apical curve are sharp and well defined, and the peak is flattened and constitutes about 1/3 of the total length. The medial flange is variably broad, located on the proximal part of the peak, and clearly demarcated from the acropodite stem. The distal zone terminates in a reflexed tip; has a long, narrow flange; and usually exhibits a tooth distal to the medial flange. The /atior group overlaps partly the quadrata group, and its range is essentially that of the subgenus. Components. — latior (Brolemann) [/. latior, 1. munda Chamberlin, /. hoffmani Shelley], stenoloba Shelley. Additional Records. — |. latior — VA, Patrick Co., Blue Ridge Parkway at mile 174.3, M, 23 June 1984, R.L., and C. Hoffman (RLH); /. munda — SC, Spartanburg Co., Spartanburg, M, 3 October 1943, R.L. Wenzel (FMNH). MEM. AMER. ENT. SOC., 35 68 XYSTODESMID MILLIPEDS The Quadrata Group The acropodites in this group are characterized by reduced medial flanges and simple tips that project inward into the arches at distinct angles from the distal zones. A lateral flange may or may not be present, but other acropodal specializations are absent. The prefemoral process is moderately long. The two species occupy small areas in the eastern Piedmont and Fall Zone of southern South Carolina. Components. — quadrata Shelley, /aticurvosa Shelley. Additional Record. — quadrata — SC, Lexington Co., Leesville, M, 7 October 1949, L. Brodie (FMNH). SIGMORIA (CROATANIA) Shelley, new status Croatania Shelley, 1977:305-306. Hoffman, 1979:159. Type species. — Croatania catawba Shelley, 1977, by original designa- tion. Diagnosis. — Paranota yellow or red, metaterga either uniformly black or with concolorous stripes connecting paranotal markings; gonopods in situ with acropodites extending well beyond anterior margin of aperture, in- serting between 7th legs; prefemoral process very large, usually extending beyond level of tip of distal zone; acropodites moderately thick and heavy to massive, well sclerotized, oriented normally on coxa with inner surface directed anteriomediad; basal zone with proximomedial edge expanded and irregularly notched, basalmost projection largest, spiniform; medial flange variable but laminate, located entirely on peak, expanded distad into lobe, latter sometimes detached from flange; lateral flange present, lobe-like, usually clearly demarcated from acropodite stem; tip blunt. Remarks. — The subgenus Croatania occurs to the east and north of Cleptoria in piedmont South Carolina and ranges into south-central North Carolina. The center of diversity is the Broad River Valley of north-central South Carolina where catawba, saluda, and simplex are parapatric. The fourth species, yemassee, is allopatric in the outer Coastal Plain. The gonopods, widely separated and parallel in situ (see Fig. 46 of simplex), ex- hibit a number of specializations, the most obvious being an irregularly notched, proximomedial expansion of the basal zone, which varies in length and degree of jaggedness and is small and comparatively inconspicuous in yemassee. Three species — catawba, saluda, and yemassee — also have enormous prefemoral processes extending to or beyond the levels of the distal zones. Females of this group can be distinguished by the large, con- voluted cyphopodal membrane, which protrudes through the medial corner of the aperture. R.M. SHELLEY & D.R. WHITEHEAD 69 Croatania is composed of four species which are united in a single species group, the catawba group. They were illustrated and described in detail by Shelley (1977), and the gonopods are here redescribed in ‘‘sigmoid’’ ter- minology. New diagnoses and color descriptions are also presented along with new locality records for each species except yemassee. A discussion of variation is given for simplex. Components. — catawba (Shelley), saluda (Shelley), simplex (Shelley), yemassee (Shelley). Sigmoria (Croatania) catawba (Shelley), new combination Croatania catawba Shelley, 1977:306-312, Figs. 1-2, 7, 11-12, 16. Filka and Shelley, 1980:28, Fig. 54. The three new records of catawba add Cherokee County, South Carolina, to its known range and conform to the previous account of variation. Diagnosis. — A moderate-size species of Sigmoria with medial flange on peak and with yellow paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process massive, globose basally, bent ventrad distal to midlength and narrowing to acuminate tip, extending beyond level of tip of acropodite; latter thick and heavy, arch high and rounded, extending to but not beyond level of prefemoral process; basal zone with ridge on inner surface, proximomedial edge expanded into deeply notched plate, basalmost projection in form of short spine; anterior bend poorly defined; peak moderately long, gently curved; apical curve well defined; distal zone short, angling into arch, tip blunt; medial flange arising at anterior bend, expanding distad into triangular lobe on distal extremity of peak; lateral flange in form of thick lobe on proximal part of distal zone, well demarcated from acropodite stem. Color in Life. — Paranota bright lemon yellow, occasionally orange; metaterga black, without stripes; collum with or without broad concolorous stripe along anterior margin. Holotype Gonopods. — Prefemur moderate with massive process arising on dorsal side, subglobose basally, widest proximal to midlength and blackened along inner edge, bent ven- trad at 2/3 length and tapering to acuminate tip; latter directed toward anterior bend, extend- ing well beyond level of distal zone. Acropodite thick and heavy, well sclerotized, arch upright, a smooth, continuous curve overhanging but not extending beyond level of prefemoral process; basal zone with ridge on inner surface, proximomedial edge expanded into deeply notched plate, basalmost projection enlarged into short spine; anterior bend broad, poorly de- fined; peak moderately long, gently curved, apex at midlength; apical curve sharp, well de- fined; distal zone short, coplanar with basal zone, angling into arch and directed toward latter; tip blunt. Medial flange arising proximad on peak, expanded distad into triangular lobe on distal extremity. Lateral flange in form of thick, heavily sclerotized lobe on proximal part of distal zone, well demarcated from latter. Prostatic groove running along lateral side of ridge on MEM. AMER. ENT. SOC., 35 70 XYSTODESMID MILLIPEDS basal zone, crossing to lateral side of acropodite at anterior bend and continuing to terminal opening. Distribution. — South-central piedmont North Carolina to north-central piedmont South Carolina (Shelley 1977). Specimens were examined from the following new localities: SOUTH CAROLINA. — Cherokee Co., 4.6 mi SE Blacksburg, along SC hwy. 5, 0.3 miS jet. SC hwy. 68, F, 10 May 1977 (NCSM A1481). Chester Co., 11.6 mi W Chester, Leeds Campground, Sumter Nat. For., 2M, F, 30 April 1977 (NCSM A1486); and 12.4 mi W Lowrys, along SC hwy. 9 at Broad R., 2M, 1 May 1977 (NCSM A1500). Union Co., 0.1 mi N Jonesville, along SC hwy. 18, 2M, 2F, 2 May 1977 (NCSM A1505); 2.3 mi W Carlisle, along SC hwy. 24 at Cane Cr., 2M, 3F, 2 May 1977 (NCSM A1513); and 5.0 mi N Carlisle, along SC hwy. 86 at Neals Cr., Sumter Nat. For., 6M, 2F, 2 May 1977 (NCSM A1514). Remarks. — The two males in sample A1500 from Chester County were found dead and lying on top of leaves. They had been recently decapitated by an unknown carnivore and were still pliable. One male lacked the head and collum, and the other had lost the head, collum, and segments 2 and 3. The bites were irregular and parts of the next segments had been destroyed, but there was no evidence of continuous chewing. The segments seem to have been removed by a single, clean bite, which stopped short of segment 5 and the anteriormost defensive glands. This situation is identical to that I reported for Dicellarius t. talapoosa (Chamberlin) from Cheaha Mountain, Alabama (Shelley 1984c). Sigmoria (Croatania) saluda (Shelley), new combination Croatania saluda Shelley, 1977:312-316, Figs. 3, 8, 13. As with catawba, the new material of saluda conforms to previously known variation, but the range is expanded to encompass parts of Laurens, Lexington, Saluda, and Aiken counties, South Carolina. Its north-south distribution from the Enoree to the Savannah Rivers, is unchanged, but the area is extended eastward into the Fall Zone. Diagnosis. — A moderate-size species of Sigmoria with medial flange on peak and with red paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process large, sides parallel, curving broadly ventrad, apically bifurcate with medial component longer than lateral and extending beyond level of distal zone; acropodite thick and heavy, leaning dorsad and extending beyond level of prefemoral process; basal zone with proximomedial edge expanded into shallowly notched plate, basalmost projection enlarged into spine; anterior bend poorly defined; peak moderately long, gently curved; apical curve well defined; distal zone R.M. SHELLEY & D.R. WHITEHEAD all short, angling into arch, tip blunt; medial flange arising at anterior bend, expanding distad into triangular lobe on distal extremity of peak and prox- imal part of distal zone; lateral flange in form of thick lobe on proximal part of distal zone, well demarcated from latter. Color in Life. — Paranota usually entirely red, color occasionally restricted to peritremata; metaterga black, without stripes; collum with red strip along anterior margin. Holotype Gonopods. — Prefemoral process not globose basally, of nearly equal width throughout, broadly curved ventrad, bifurcate apically, medial component longer than lateral component and extending beyond level of distal zone, directed toward anterior bend. Acropodite thick and heavy, well sclerotized, arch broad, a smooth continuous curve leaning dorsad and extending well beyond level of prefemoral process; basal zone without ridge, prox- imomedial edge expanded into shallowly notched plate, longer and more irregularly notched than in catawba, basalmost projection enlarged into short spine. Anterior bend broad, poorly defined; peak moderately long, gently curved, apex at midlength; apical curve sharp, well de- fined; distal zone short, coplanar with basal zone, angling into arch and directed toward prefemur; tip blunt. Medial flange arising at anterior bend, expanding distad into broadly rounded lobe on distal extremity of peak and proximal part of distal zone. Lateral flange in form of thick, heavily sclerotized lobe on proximal part of distal zone, well demarcated from latter. Prostatic groove crossing to lateral side at anterior bend, continuing to terminal open- ing. Distribution. — Central and south-central piedmont South Carolina be- tween the Enoree and Savannah Rivers, ranging east into the Fall Zone. Specimens were examined from the following new localities: SOUTH CAROLINA. — Laurens Co., 11.2 mi NE Clinton, along SC hwy. 72 at Duncan Cr., M, F, 9 May 1977 (NCSM A1564); 8.6 mi. S Clinton, along SC hwy. 38 at Little R., 3M, 9 May 1977 (NCSM A1565); and 2.0 mi. NE Cross Hill, along SC hwy. 560 at Campbell Cr., M, 9 May 1977 (NCSM A1566). Greenwood Co., 8.4 mi. SW Ninety Six, along SC hwy. 702, 0.2 mi. W Saluda Co. line, M, 4 May 1977 (NCSM A1524). Newberry Co., 16.0 mi. NE Newberry, along SC hwy. 45 at Enoree R., M, F, 3 May 1977 (NCSM A1517); and 6.3 mi. E Newberry, along US hwy. 176 at Mud Cr., 4M, 2F, 3 May 1977 (NCSM A1521); 11.5 mi. E Newberry, along US hwy. 176, 0.1 mi. S Pomaria, 3M, 2F, 14 July 1979 (NCSM A2804); 6.5 mi. SE Newberry, along SC hwy. 244, 0.5 mi. E ject. SC hwy. 315, M, F, July 1979 (NCSM A2806); and 9.0 mi. SE Newberry, along SC hwy. 403 at Camping Cr., M, F, 14 July 1979 (NCSM A2805). Richland Co., 20.0 mi. NW Columbia, along US hwy. 176 at Wateree Cr., F, 14 July 1979 (NCSM A2803); and 14.0 mi. NW Columbia, along SC hwy. 244 at Hollingshead Cr., 3M, F, 14 July 1979 (NCSM A2802). Lexington Co., West Columbia, M, 3F, 24 June 1958, collector unknown (FSCA); 13.0 mi. W Lexington, along SC hwy. 59 at Little Cr., 3M, 15 July 1979 (NCSM A2813); and 5.2 mi. NE Leesville, along SC hwy. 158, 0.3 mi. S jct. SC hwy. 54, 3M, 23 November 1977 (NCSM A1809). Saluda Co., 4.0 mi. WNW Batesburg, along US hwy. 178 at Clouds Cr., 2M, 23 November 1977 (NCSM A1807); 6.0 mi. S Saluda, along SC hwy. 193, 0.1 mi. W jct. SC hwy. 119, 5M, 3F, 15 July 1979 (NCSM A2811); and 6.5 mi. E Saluda, along US hwy. 178, 0.2 mi. W jct. SC hwy. 29, M, F, 15 July 1979 (NCSM A2811). Edgefield Co., 1.9 mi. ENE Trenton, along SC hwy. 75 at Tiger Cr., 3M, 2F, 23 November 1977 (NCSM A1804); and 7.5 mi. S Edgefield, along SC hwy. 34 at Double Branch Cr., M, 16 July 1979 (NCSM A2821). Aiken Co., Lexington Co. line on SC hwy. 109 at Chinquapin Cr., 2M, F, 15 MEM. AMER. ENT. SOC., 35 72 XYSTODESMID MILLIPEDS July 1979 (NCSM A2818); and North Augusta, along SC hwy. 230, 1.7 mi. E I-20, 2M, 2F, 17 September 1980 (NCSM A3509). Sigmoria (Croatania) simplex (Shelley), new combination Figs. 46-48 Croatania simplex Shelley, 1977:316-319, Figs. 4-5, 9, 14. Priority for the name simplex is with this species, proposed four years before the one herein renamed inornata (Sigiria). The original description of simplex was based only on the type series from Chester County, South Carolina, and the range could not be reported, nor could variation be discussed aside from that demonstrated by paratypes. Diagnosis. — A moderate-size species of Sigmoria with short medial flange located on distal extremity of peak or proximal part of distal zone and with yellow or red paranota, metaterga with or without narrow, con- colorous transverse stripes; gonopods with following diagnostic characters: prefemoral process moderately long, curving broadly ventrad, apically acuminate or bifurcate; acropodite moderately thick, arch flattened or a smooth, continuous curve, overhanging and extending beyond level of prefemoral process; basal zone with proximomedial edge expanded into variably notched plate, basalmost projection largest; anterior bend poorly defined; peak moderately long, gently curved or flattened; apical curve variably defined; distal zone moderately long, curving broadly into arch or extending ventrad from peak and bent abruptly into arch at midlength, apically acuminate; medial flange represented by distal lobe and short preceding lamella, either located on distal extremity of peak or arising there and terminating near midlength of distal zone; lateral flange either absent or represented by widening in proximal portion of distal zone. Color in Life. — Paranota bright lemon yellow, occasionally red; metaterga black, with or without concolorous stripes along caudal edges; collum with stripe along anterior margin. Holotype Gonopods. — Prefemoral process moderately long (Fig. 47), terminating well below level of distal zone, crescent shaped, widest at 1/3 length, then tapering smoothly and continuously to acuminate tip, latter directed toward anterior bend. Acropodite moderately thick and heavy, arch upright, a smooth, continuous curve overhanging and extending beyond level of prefemoral process; basal zone without ridge, basal proximomedial expansion short, shallowly notched, basalmost projection enlarged into short spine; anterior bend broad, poorly defined; peak relatively short and high, gently rounded, apex at midlength; apical curve broad, poorly defined; distal zone moderately long, curving broadly into arch, tapering smoothly and continuously to acuminate tip; latter directed toward anterior bend. Medial flange short, in- conspicuous, represented only by short, blunt tubercle and short attached lamella at level of apical curve, this homologous to lobe of catawba. Lateral flange indistinct, represented by broad rounded area at midlength of distal zone. Prostatic groove crossing to lateral side at anterior bend, continuing to terminal opening. R.M. SHELLEY & D.R. WHITEHEAD 13; 47 Fics. 46-48. Sigmoria (Croatania) simplex. 46, gonopods in situ, ventral view of male from Fairfield Co., SC. 47, telopodite of left gonopod of holotype, medial view. 48, telopo- dite of left gonopod of male from Richland Co., SC. Scale line for fig. 76 = 1.00 mm; line for other figs. = 1.00 mm for 77; 0.64 mm for 48. 74 XYSTODESMID MILLIPEDS Variation. — Some specimens have yellow stripes along the caudal margins of the metaterga, whereas others, like the types, lack these mark- ings. Both patterns were found 4.2 mi. N Columbia (samples A2800-2801), but there were no detectable anatomical differences. Material from Berkeley and Orangeburg counties had red paranota and stripes. The gonopods of eastern populations vary considerably from the types, and a representative example is shown in Figure 48. Aside from occasional bifurcate individuals, as found in the type series, the prefemoral process is generally of the same shape throughout the range, although it becomes longer and narrower in eastern populations. In the Kershaw County males it projects beyond the level of the distal zone. The proximomedial expansion is shorter and less jagged in some males, and this does not conform to any geographic pattern. The medial flange, however, is broad and laminate in eastern forms, and the distal zone likewise projects inward into the arch of the acropodite at a sharper angle, often at right angles. The configuration shown in Figure 12 is actually more common and representative of simplex than that of the holotype, which is restricted to the northwestern range periphery and may represent a terminal population effect. Thus, the situa- tion is similar to that in macra (Cleptoria), where the holotype is atypical for the species. Distribution. — Farifield, Richland, Kershaw, Orangeburg, and Berkeley counties can be added to the distribution of simplex, which is now known from both the Piedmont and Coastal Plain of South Carolina. The type locality is the northernmost, as I have not found the species in Union, York or Lancaster counties. The range abuts that of catawba in Chester County, and in Richland County, simplex is segregated from saluda by the Broad River. Here and in Fairfield County the waterway is the western limit, but northern populations in Chester County do not extend to the river, being replaced on the eastern side by catawba. Material was examined from the following localities: SOUTH CAROLINA. — Chester Co., 7.2 mi. SE Chester, along SC hwy. 44, 0.7 mi. W jct. SC hwy. 347, F, 1 May 1977 (NCSM A1492); and 7.2 mi. SW Chester, along SC hwy. 42 at Sandy R., 4M, F, 1 May 1977 (NCSM A1502). Farifield Co., 16.8 mi. NW Winnsboro, along SC hwy. 215, 0.5 mi. N jet. SC hwy. 302, 2M, 2F, 3 May 1977 (NCSM A1516); 12.0 mi NW Winnsboro, along SC hwy. 28 at Mobley Cr., 2M, 13 July 1979 (NCSM A2793); 7.0 mi. NW Winnsboro, along SC hwy. 303, 1.5 mi. N jct. SC hwy. 22, 2M, 2F, 13 July 1979 (NCSM A2792); 13.0 mi. SW Winnsboro, along SC hwy. 46 at Sawney’s Cr., M, 2F, 13 July 1979 (NCSM A2788); 10.5 mi. SW Winnsboro, along SC hwy. 213 at Little R., 3M, 2F, 13 July 1979 (NCSM A2794); 9.5 mi. S Winnsboro, along SC hwy. 64 at Little Cedar Cr., 5M, F, 13 July 1979 (NCSM A2796); and 9.0 mi. NE Winnsboro, along SC hwy. 234 at Wateree Cr., M, 13 July 1979 (NCSM A2791). Kershaw Co., 11.0 mi. NNW Camden, along SC hwy. 40 at Flat Rock Cr., 2M, F, 12 July 1979 (NCSM A2789). Richland Co., 15.0 mi. NNE Columbia, along LS ——— CU~S—~—~S~S—~— R.M. SHELLEY & D.R. WHITEHEAD US) SC hwy. 54 at Five Mile Cr., M, F, 13 July 1979 (NCSM A2798); 15.0 mi. NNW Columbia, along SC hwy. 967 at Horse Cr., 3M, F, 13 July 1979 (NCSM A2797); 12.0 mi. NNW Colum- bia, along SC hwy. 59, 0.1 mi. N ject. SC hwy. 215, M, 23 November 1977 (NCSM A1811); 7.0 mi. N Columbia, along SC hwy. 61 at North Branch Cr., 9M, 7F, 14 July 1979 (NCSM A2800-A2801); and 3.0 mi. N Columbia, along US hwy. 321 at Crane Cr., 2M, F, 14 July 1979 (NCSM A2799). Orangeburg Co., Santee State Park, M, 9 September 1980 (NCSM A3494). Berkeley Co., 20.0 mi. W Moncks Corner, along SC hwy. 27, 10.0 mi. N ject. I-26, bluff area of Four Holes Swamp Sanctuary, 4M, 2F, 9 September 1980 (NCSM A3493). Remarks. — The in situ gonopodal configuration in this and other species of Croatania is the parallel arrangement shown in Figure 46. Sigmoria (Croatania) yemassee (Shelley), new combination Croatania yemassee Shelley, 1977:319-321, Figs. 6, 10, 15. Sigmoria yemassee is still known only from the two collections cited in 1977. I have extensively surveyed the environments in southeastern South Carolina and visited the type locality twice and the other site once at the susie re ahs Fic. 49. Distribution of the subgenus Croatania in North and South Carolina. Triangles, catawba; stars, simplex; dots, saluda; squares, yemassee. MEM. AMER. ENT. SOC., 35 76 XYSTODESMID MILLIPEDS proper times of the year, without finding a single individual. The color in life is therefore unknown. Diagnosis. — A moderate-size species of Sigmoria with medial flange on distal extremity of basal zone and proximal part of distal zone; gonopods with following diagnostic characters: prefemoral process large, bisinuately curved, bent ventrad apically and narrowing to acuminate tip, extending to level of distal zone; acropodite thick and heavy, arch high and narrowly rounded, extending just beyond level of prefemoral process; basal zone slightly expanded, with small denticulations on proximomedial edge, not noticeably expanded; anterior bend poorly defined, peak rounded, short and high, apical curve well defined; distal zone very short, angling into arch, apically blunt; medial flange in two parts, a narrow, inconspicuous lamina on distal extremity of basal zone and narrow, rhomboidal projection on distal extremity of peak, representing distal lobe of flange in catawba; lateral flange a short quadrate lobe on distal zone, margin finely serrate. Color in Life. — Unknown. Holotype Gonopods. — Prefemoral process upright, bisinuately curved, not globose basal- ly, bent ventrad apically and narrowing abruptly to acuminate tip; latter directed toward mid- point of telopodite. Acropodite moderately thick and heavy, well sclerotized; arch high and narrowly rounded, upright, slightly overhanging prefemoral process; basal zone without ridge, proximomedial expansion short, with only small denticulations, none enlarged; anterior bend poorly defined; peak short, apex at midlength; apical curve sharp, well defined; distal zone short, projecting inward for short distance into arch, tip broadly rounded and directed toward basal zone. Medial flange located on distal extremity of basal zone and proximal part of distal zone; former a short, narrow, inconspicuous lamina; latter a linear rhomboidal projection. Lateral flange represented by quadrate lobe, margin finely serrate. Prostatic groove crossing from medial to lateral sides at midlength of basal zone, continuing to subterminal opening. SIGMORIA (CLEPTORIA) Chamberlin, new status Cleptoria Chamberlin, 1939:9. Chamberlin and Hoffman, 1958:28. Hoffman, 1967:5-7; 1979:158. Jeekel, 1971:254. Brevigonus Shelley, 1980a:32-34; 1981b:54-55. NEW SYNONYMY. Type species. — Of Cleptoria, C. macra Chamberlin, 1939, by original designation; of Brevigonus, Cleptoria shelfordi Loomis, 1944, by original designation. Diagnosis. — Paranota red, metaterga either uniformly black or with conclorous red stripes connecting paranotal markings; gonopods in situ with acropodites extending well beyond anterior margin of aperture, insert- ing between 7th legs; prefemoral process moderate to small, frequently ab- ———E—— i ”— ssi“ (COS”S”;”;”*”*;*é~‘izsS R.M. SHELLEY & D.R. WHITEHEAD 77 sent; acropodites moderately thick to massive, heavily sclerotized, oriented normally on coxa; basal zone with or without variable basal spine on outer surface; medial flange variable in shape and location, moderately laminate to a thickened swelling of medial surface of acropodite stem, located on basal zone and/or peak, occasionally with suggestion of distal lobe; lateral flange occasionally laminate but usually swollen, thickened, and lobe-like, poorly demarcated from acropodite stem, located more ventrad than laterad, forming highest point of acropodite arch; tip variable. Remarks. — Cleptoria occurs chiefly in piedmont South Carolina and Georgia, and its species have the largest bodies and the most massive, heavi- ly sclerotized acropodites in the genus. The medial flanges are usually thick, stiff projections instead of thin, flexible lamellae as in other subgenera. Likewise, only one species has a laminate lateral flange; otherwise the struc- ture is a swollen, broadly rounded, and poorly demarcated lobe on the proximal part of the distal zone. It typically projects ventrad more than laterad and forms the highest point of the acropodal arch. The distal zone is long and blade-like in one species and is absent from another. In forms with lobe-like lateral flanges, the region is short, broad, and directed perpen- dicularly from the peak, imparting an overall ‘‘bird’s head’’ appearance to the acropodite (Bollman 1888). Cleptoria presents the most difficult taxonomic decisions in Sigmoria s. lat., particularly with the highly variable South Carolina forms. Central and southern populations of macra possess sharply acute spurs on the outer sur- faces of the peaks as do adjacent populations of shelfordi, the parapatric southern form with a shortened acropodite (lacking the distal zone and apical curve). This condition could derive from that in macra by reduction of the lateral flange (lobe) and the short distal zone, but the picture is com- plicated by the random absence of the spur in shelfordi. Additionally, ran- dom males of shelfordi have spines on the basal zones, which are absent from macra. Thus, two principal variants of shelfordi exist (Shelley 1980a). Its range is unequally divided by arcuata, which has a long distal zone and basal acropodal spine. Sigmoria (Cleptoria) robusta, with a swollen, rounded lateral flange and a short distal zone, occurs west of arcuata. Its acropodite is heavier than that of macra and is similar to the Georgia species; it also lacks the spur but has the basal spine. Thus in South Carolina, forms with shortened acropodites (the variants of shelfordi) con- nect forms with ‘‘bird’s head’’ acropodites (robusta and macra ), which in turn are segregated elsewhere by a form with a long curved distal zone (ar- cuata) that also bisects the range of the first form (shelfordi)! Features like the basal spine and midlength spur traverse the sharp boundaries between the overall acropodal forms. The spine occurs in robusta, arcuata, and ran- MEM. AMER. ENT. SOC., 35 78 XYSTODESMID MILLIPEDS domly in shelfordi; the spur also occurs randomly in shelfordi and in all populations of macra except the northern ones. The picture is clearer in Georgia because there is only one acropodal form, the ‘‘bird’s head.’’ Hoffman (1967) recognized three species, and I concur although the status of bipraesidens, known only from the holotype, could change when more material is available from northwest of Athens. Unlike the parapatric rileyi, bipraesidens lacks the spine on the basal zone; has a longer, more acute, and more strongly indented distal zone; and the entire medial surface of the peak is thickened to represent the flange. Sigmoria (Cleptoria) abbotti, the parapatric species to the east of rileyi, is most common along the Savannah River between Hart and Burke counties. It differs from rileyi in having a prefemoral process; an expanded, laminate anterior margin of the basal zone; a thin, narrow, indistinct medial flange; and a stronger basal spine. Similarities in the basal zones and prefemora of abbotti and robusta suggest prior connection and that the ancestral range curved northward from Hart County. Following segregation by the Savan- nah River, the forms diverged as revealed by the outer margins of the distal zones, rounded and continuous in robusta and strongly indented in abbotti. Thus, the ‘‘bird’s head’”’ acropodal configuration and the spine on the basal zone traverse the Savannah River, whereas the spur, the short acropodite of shelfordi, and the long curved one of arcuata are restricted to the northern side. The last three conditions evolved subsequent to the “‘bird’s head’’ con- figuration, which antedates the present course of the Savannah River. Equally interesting are small, allopatric populations of ri/eyi in eastern and central Alabama. Reasonably thorough searches have failed to produce rileyi in western Georgia, and the main range is disjunct from that in eastern Alabama by about 100 miles. Thus, geographical isolation brought about by the Savannah River seems to have promoted divergence of robusta and abbotti, which are only about 20 miles apart, but isolation has not affected the Alabama populations of rileyi, which are also segregated by major waterways and five times as much distance! A complete description is presented for robusta, but only diagnoses and color statements are provided for shelfordi and arcuata, since their gonopods were characterized in ‘‘sigmoid’’ terminology by Shelley (1980a, 1981b). Hoffman (1967) published full descriptions for rileyi, abbotti, bipraesidens, and macra, so only diagnoses, color statements, and redescriptions of the gonopods are needed. The species of Cleptoria form a homogeneous unit, and thus there is only one, the ri/eyi, species group. Components. — rileyi (Bollman), macra (Chamberlin), shelfordi (Loomis), bipraesidens (Hoffman), abbotti (Hoffman), arcuata (Shelley), robusta new species. —————— SF R.M. SHELLEY & D.R. WHITEHEAD 719 Sigmoria (Cleptoria) rileyi (Bollman), new combination Figs. 50-52 Fontaria rileyi Bollman, 1888:345. Attems, 1938:167. Cleptoria rileyi: Chamberlin, 1939:10. Chamberlin and Hoffman, 1958:28. Cleptoria rileyi rileyi: Hoffman, 1967:12-15, Figs. 2, 7-8. Cleptoria rileyi alabama Hoffman, 1967:16-17, Figs. 9-10. NEW SYNONYMY. Hoffman (1967) provided a detailed description and illustrations of the holotype of ri/eyi along with a short account of the allopatric population in Lee County, Alabama. Toward the end of the present study a nearly iden- tical male was obtained from Jefferson County, about 100 miles northwest of Lee County. I find no meaningful differences between the Alabama forms and those in central Georgia and no justification for taxonomic recognition. The following supplemental account characterizes the sternal process of the holotype and its gonopods in ‘‘sigmoid’’ terminology, and addresses aspects of variation, ecology, and distribution. I also present ad- ditional drawings of the gonopods including one from ventral view, which shows the flange protruding on the medial side. Type specimens. — Male holotype (NMNH) collected by L.M. Under- wood, August 1887, at Macon, Bibb Co., GA. Diagnosis. — A large species of Sigmoria with medial flange on peak and with red paranota, metaterga without stripes; gonopods with following diagnostic characters: prefemoral process absent; acropodite massive, arch flattened and squared, extending well beyond level of prefemur; basal zone moderately long and broad, with small basal spine on caudal edge; anterior bend well defined; peak long, flat, constricted distad; apical curve well defined; distal zone short, directed perpendicularly from peak, outer margin moderately indented, sides narrowing apically to subacuminate tip; medial flange thick but relatively laminate, arising on proximal portion of peak, expanding into broadly rounded distal lobe near midlength of peak; lateral flange represented by thick, broadly rounded lobe on proximal part of distal zone, protruding ventrad, demarcated from acropodite stem by im- pression on lateral surface. Color in Life. — Paranota red; metaterga dark glossy black, without stripes along caudal edges; collum also without stripes along either margin. Holotype. — Process of 4th sternum moderately long, equal in length to widths of adjacent coxae (Hoffman, 1967, Fig. 2). Gonopods in situ with acropodites projecting anteriad from aperture, extending forward in parallel arrangement over anterior margin to between 7th legs. Gonopod structure as follows (Figs. 50-52): Prefemoral process absent. Acropodite massive, heavily sclerotized, forming rec- tangular arch overhanging and extending well beyond prefemur; basal zone moderately broad, anterior edge slightly thinner than caudal but not especially laminate or expanded, outer margin wider basally with short spiniform projection; anterior bend sharp, well defined, ap- proximately a right angle; peak relatively long, straight, and level, constricted on both margins MEM. AMER. ENT. SOC., 35 80 XYSTODESMID MILLIPEDS at distal extremity; apical curve sharp, well defined; distal zone short, with distolateral striae, outer margins moderately indented, margins gently curved and narrowing to blunt tip; latter directed toward coxa. Medial flange relatively thick, moderately laminate, arising proximally Fics. 50-55. 50-52, Sigmoria (Cleptoria) rileyi. 50, telopodite of left gonopod of holotype, medial view. 51, the same, lateral view. 52, the same, ventral view. 53-55, Sigmoria (Clep- toria) abbotti. 53, telopodite of left gonopod of male from Lincoln Co., GA, medial view. 54, the same, lateral view. 55, the same, ventral view. Scale line = 1.00 mm for all figures. ——— R.M. SHELLEY & D.R. WHITEHEAD 81 on peak, curving sharply mediad and forming broadly rounded lobe near midlength of peak, terminating abruptly distal to lobe. Lateral flange lobe-like, located on proximal part of distal zone, broadly rounded and produced ventrad forming highest point of acropodite arch, poorly demarcated from acropodite stem by only slight impression in distal zone. Prostatic groove crossing to lateral side at anterior bend, curving onto distal zone and opening terminally. Variation. — Gonopodal variation in rileyi is negligible. Males from cen- tral Georgia are nearly identical to each other, and the only difference in those from Alabama is a slight enlargement on the anterior surface of the prefemur in some individuals, which suggests a vestigial prefemoral process. Ecology. — Sigmoria (Cleptoria) rileyi occurs under thin layers of leaves on relatively hard substrates near water sources. Distribution. — Comprised of three allopatric populations, one in cen- tral Georgia, and one in Lee and one in Jefferson counties, Alabama (Fig. 68). These areas are in the Piedmont Plateau, the Fall Zone, and the Cumberland Plateau, respectively. The Georgia population occupies a linear area roughly 60 miles long between I-85 and I-75, and three trips to sections of Georgia west of that shown in figure 68 produced specimens of Dynoria medialis Chamberlin and Sigmoria (Cheiropus) persica, but none of rileyi. At best the species is uncommon here. Material was examined from the following new localities: GEORGIA. — Clarke Co., Athens, 3M collected in June, July, and October 1948-1972 (FSCA). Oconee Co., Bogart, 2M, 14 October 1973, R.L. Duffield (RLH). Morgan Co., Hard Labor Creek St. Pk., 4M, 3F, 15 June 1958, N.B. Causey (FSCA). Putnam Co., 7 mi. S Eaton- ton, Sinclair Lake Rec. Area, Oconee Nat. For., M, 20 November 1977 (NCSM A1786); and 7 mi SE Monticello, along GA hwy. 212 at Jasper co. line, 3M, 20 November 1977 (NCSM A1788). ALABAMA. — Lee Co., Chewacla St. Pk., 7M, 4F, 21 May 1980 (NCSM A3122); Auburn, 8M, 13 June 1959, N.B. Causey (FSCA) and Auburn Univ. campus, 2M, June-July 1973, F.M. Scale and R. Russell (AU). Jefferson Co., 10 mi. S Birmingham, along AL hwy. 150, 1.2 mi. W jct. US hwy. 31, M, Fall 1984, R.E. Ashton (NCSM A4228). Sigmoria (Cleptoria) abbotti (Hoffman), new combination Figs. 53-55 Cleptoria abbotti Hoffman, 1967:18-21, Figs. 13-16. This species was known from only two localities when it was described and illustrated by Hoffman (1967). Much more material is available now, providing insight into variation and distribution. The lateral flange, which Hoffman called the subapical lobe, seems overemphasized in his drawings, and I therefore present figures of a male from Lincoln County, Georgia, ap- proximately 55 miles NW of the type locality. MEM. AMER. ENT. SOC., 35 82 XYSTODESMID MILLIPEDS Type specimens. — Male holotype (NMNH) and one male and one female paratype collected by L. Hubricht, 22 May 1960, from 5 mi. SW Waynesboro, Burke Co., GA. Male and female paratypes deposited in RUere Diagnosis. — A large species of Sigmoria with medial flange extending from distal extremity of basal zone to midlength of peak and with red paranota, metaterga without markings; gonopods with following diagnostic characters: projection of prefemur present, either distinct process or fold of anterior lamina of basal zone; acropodite massive, leaning over and extend- ing beyond level of prefemoral process; basal zone very broad, inner surface convex with thin, laminate lateral margin and with short, acute, basal spine on caudal margin; anterior bend poorly defined; peak short; apical curve well defined; distal zone short, directed perpendicularly from peak, outer margin strongly indented, sides tapering to subacuminate tip; medial flange thick, narrow, and inconspicuous, not particularly laminate, terminating in slight lobe on distal part of peak; lateral flange represented by thick, broad- ly rounded lobe on proximal portion of distal zone, protruding ventrad, demarcated from acropodite stem by impression on lateral surface. Color in Life. — Paranota red; metaterga dark glossy black, without stripes along caudal edges; collum also without stripes. Holotype. — 4th sternal process long, apically divided, longer than widths of adjacent coxae (Hoffman 1967, Fig. 4). Gonopods in situ with acropodites projecting anteriomediad from aperture, overlapping in midline and extending forward over anterior margin to between 7th legs. Gonopod structure as follows (Figs. 53-55): Prefemur with rounded lobe on medial surface; prefemoral process short and conical, directed toward apical curve. Acropodite massive, heavily sclerotized, leaning over and extending well beyond level of prefemoral process; basal zone broad, with short, acute spine proximad on outer margin, medial surface convex with anterior edge expanded into thin lamina; anterior bend broad, poorly defined; peak short and thick; apical curve sharp, well defined; distal zone short, with distolateral striae, outer margins gently curved and nar- rowing to subacuminate tip; latter directed toward coxa. Medial flange thick, narrow, and in- conspicuous, arising on basal zone, terminating at midlength of peak, latter with slight thickening on medial surface distal to flange representing lobe of latter. Lateral flange lobe- like, located on proximal part of distal zone, broadly rounded and produced ventrad, leaning laterad and demarcated from distal zone by slight depression. Prostatic groove crossing to lateral side at anterior bend, curving onto distal zone and terminating apically. Variation. — The anterior edge of the basal zone is thin and laminate, and in many males, what is called the prefemoral process is really just the expanded, folded proximal end of this lamina, which appears as a separate projection because its profile is viewed in medial perspective. Few males in addition to the types have a truly separate subconical process (Hoffman 1967) that extends clearly beyond the margin of the lamina. The structure occurs randomly throughout the range, being found on males from Hart, —2_——$$ <_< R.M. SHELLEY & D.R. WHITEHEAD 83 Elbert, Lincoln, and Burke counties. The medial flange, inconspicuous in all specimens, also varies randomly. There is a slight thickening of the medial surface of the peak at the level of the apical curve, which is not really visible in medial view and is best seen from the ventral perspective. As in rileyi, I think this represents the distal lobe of the medial flange, and it is continuous with the medial flange in males from Columbia County. The size of the spine on the outer surface of the basal zone also varies, but the projection on the 4th sternum is longer than the widths of the adjacent coxae in all individuals. Ecology. — Sigmoria (Cleptoria) abbotti is typically found in pre- dominantly hardwood areas under thin layers of leaves on relatively hard substrates near water sources. I also encountered it in similar spots in predominantly pine woods, under the few deciduous trees in the area. Distribution. — The Piedmont Plateau and the inner edge of the Coastal Plain along the southern side of the Savannah River from Hart to Burke counties, Georgia. The area is about 85 miles long, but abbotti is most com- mon between US highways I-85 and I-20. It extends to 30 miles south of the river, there abutting the range of rileyi (Fig. 68). Material was examined from the following new localities: GEORGIA. — Oconee Co., Watkinsville, M, F, 26 March 1959, W. Tarpley (FSCA). Hart Co., 8 mi. ESE Hartwell, along co. rd. 1724 at Little Cedar Cr., 2M, 20 July 1979 (NCSM A2841). Elbert Co., 11 mi. NNE Elberton, along GA hwy. 368 at Pickens Cr., 2M, F, 20 July 1979 (NCSM A2840); Nancy Hart St. Pk., M, F, 19 July 1979 (NCSM A2938). Oglethorpe Co., Oconee Nat. For., 3M, 8 June 1973, R.L. Duffield (RLH). Greene Co., 10.8 mi. NW Greensboro, Oconee R. Rec. Area, Oconee Nat For., 2M, F, 22 November 1977 (NCSM A1799). Lincoln Co., 6 mi. N. Lincolnton, along GA hwy. 79 at Mills Cr., 4M, F, 19 July 1979 (NCSM A2835); 2.6 mi. NW Lincolnton, along GA hwy. 904 at Soap Cr., 2M, F, 19 July 1979 (NCSM A2832); and 6 mi. SE Lincolnton, along GA hwy. 220 at Cherokee Cr., M, 2F, 19 July 1979 (NCSM A2830). Columbia Co., Mistletoe St. Pk., 2M, F, 18 July 1979 (NCSM A2825); 10 mi. NE Appling, along US hwy. 221 at Clark Hill Dam, 2M, F, 19 July 1979 (NCSM A2828); 15.5 mi. N. Harlem, along GA hwy. 104, 5M, 19 July 1979 (NCSM A2829); 10 mi. NW Augusta, along GA hwy. 28, 0.1 mi. E Savannah R., 2M, F, 13 September 1980 (NCSM A3513); and Augusta Canal St. Pk., 3M, 13 September 1980 (NSCM A3511). Remarks. — The female from Lincoln County, assigned to rileyi by Hoffman (1967), is here referred to abbotti, the only species in this county. Although similar, I think that abbotti and rileyi are reproductively isolated because parapatric populations in Oconee, Morgan, Greene, and Oglethorpe counties maintain their respective identities and show no evidence of hybridization. Sigmoria (Cleptoria) abbotti has a prominent basal spine on the outer surface of the basal zone, whereas that in ri/eyi is barely detectable. The anterior margin of this section, expanded and laminate in abbotti is enlarged in some males to represent the prefemoral MEM. AMER. ENT. SOC., 35 84 XYSTODESMID MILLIPEDS process. Sigmoria (Cleptoria) rileyi, however, lacks both the lamina and a prefemoral process; its medial flange is also more distinct and laminate than that of abbotti. Sigmoria (Cleptoria) bipraesidens (Hoffman), new combination Figs. 56-59 Cleptoria bipraesidens Hoffman, 1967:16-17, Fig. 11 Hoffman (1967) provided a brief description of the holotype and only known specimen of this species, but he did not illustrate the process of the 4th sternum nor show a medial view of the gonopods. I therefore present these figures along with lateral and ventral views of the gonopods and selected anatomical details. No additional material is available to provide information on variation or distribution. Type specimen. — Male holotype (NMNH) collected by L. Hubricht, 4 April 1953, at Jefferson, Jackson Co., GA. Diagnosis. — A large species of Sigmoria with thickening on medial sur- face of peak representing medial flange and with red paranota and transverse metatergal stripes; gonopods with following diagnostic characters: prefemoral process absent; acropodite massive, arch flattened and extending well beyond level of prefemur; basal zone without trace of spine on outer surface; anterior bend broad, poorly defined; peak relatively long, flat; apical curve well defined; distal zone moderately long, outer margin strongly indented, directed perpendicularly from peak, inner edge linear, outer edge converging to acuminate tip; medial flange non-laminate, a thickening on peak, narrowest along inner surface and widest on outer; lateral flange represented by thick, broadly rounded lobe on proximal part of distal zone, protruding ventrad, demarcated from acropodite stem by im- pression on lateral surface. Color in Life (Hoffman 1967). — Paranota red; metaterga black with red transverse stripes along caudal edges; collum with red stripes along both anterior and posterior margins. Holotype. — Width across genal apices 4.9 mm; genae with distinct central impressions. Facial setae with epicranial, interantennal, genal, and frontal series absent; clypeal about 12-12, labral about 16-16. Terga smooth, polished, becoming moderately coriaceous on paranota. Latter moderately depressed, continuing slope of dorsum, caudolateral corners rounded through segment 8, blunt on 9-14, becoming progressively more acute posteriorly. Peritremata distinct, strongly elevated above paranotal surfaces; ozopores located near midlength, opening dorsolaterad. Sternum of segment 4 with large, apically divided process between 3rd legs, longer than widths of adjacent coxae (Fig. 56); that of segment 5 with two long, paramedial, basally coalesced knobs between 4th legs, length equal to widths of adjacent coxae, and low, flattened, elevated areas between Sth legs; that of segment 6 with shallow, convex, depression between 7th legs to accommodate curvatures of acropodites. Coxae with low, blunt tubercles arising on R.M. SHELLEY & D.R. WHITEHEAD 85 postgonopodal legs of segment 9, becoming sharper and more acute posteriorly; prefemoral spines arising on segment 5, becoming longer and sharper caudally. Gonopodal aperture ovoid, 4.4 mm wide and 2.7 mm long at midpoint, indented anteriolaterad, sides elevated above metazonal surface. In situ arrangement of gonopods 57 59 Ie 58 Fics. 56-59. Sigmoria (Cleptoria) bipraesidens. 56, process of 4th sternum of holotype, caudal view. 57, telopodite of left gonopod of the same, medial view. 58, the same, lateral view. 59, the same, ventral view. Scale line = 1.00 mm for 56 and 59; 1.60 mm for 57-58. MEM. AMER. ENT. SOC., 35 86 XYSTODESMID MILLIPEDS unknown. Gonopod structure as follows (Figs. 57-59): prefemoral process absent. Acropodite thick and massive, heavily sclerotized, forming rectangular arch, overhanging and extending well beyond prefemur; basal zone moderately broad, without spine on outer surface, with short blunt, basal lobe on medial side; anterior bend sharp, well defined, approximately a right angle; peak relatively long, straight, and level, constricted very slightly on both margins on distal extremity; apical curve sharp, well defined; distal zone moderately long, directed perpen- dicularly from peak, outer margin strongly indented, inner edge linear, outer straight prox- imally then curving beyond midlength and converging with inner at acuminate tip; latter directed toward coxa. Medial flange represented by thickened non-laminate area along entire width of peak, beginning proximally and terminating on distal extremity, narrowest on inner margin, widest on outer. Lateral flange lobe-like, located on proximal part of distal zone, broadly rounded and produced ventrad forming highest point of acropodite arch, leaning slightly laterad and demarcated from distal zone by slight depression. Prostatic groove crossing to lateral side at anterior bend, bending sharply (90°) onto distal zone and opening terminally. Remarks. — The striped color pattern of bipraesidens (Hoffman 1967) should be confirmed in fresh material, since this is the only reported in- dividual in Cleptoria with metatergal markings. The differences between bipraesidens and rileyi are even less distinct than those between the latter and abbotti. However, their close geographical proximity combined with the longer distal zone and non-laminate medial flange of bipraesidens suggest reproductive isolation. The status of bipraesidens should be reviewed when more material is available. Sigmoria (Cleptoria) robusta Shelley, new species Figs. 60-63 Type specimen. — Male holotype (NCSM A1553) and one male and one female paratypes collected by R.M. Shelley, 7 May 1977, in Oconee Co., SC, 10.8 mi. SW Seneca, along SC highway 168, 0.3 mi. S junction with SC highway 86. Male and female paratypes (NCSM A2061) collected by R.M. Shelley and W.B. Jones, 10 June 1978, in Oconee Co., 5.7 mi SE Oakway, along SC highway 66 at Beaverdam Cr. SC. Male paratype deposited in FSCA. Diagnosis. — A large species of Sigmoria with medial flange extending between distal extremity of basal zone and proximal portion of distal zone and with red paranota, metaterga without markings; gonopods with follow- ing diagnostic characters: prefemoral process short, blunt, and narrow; acropodite massive, arch flattened and extending beyond level of prefemoral process; basal zone very broad, inner surface convex and lateral margin thin and laminate, with sharply acute basal spine on caudal edge; anterior bend broad, poorly defined; peak flattened but rising to distal apex; apical curve sharp well defined; distal zone short, directed perpen- dicularly from peak, outer margin not indented, sides curving broadly to blunt tip; medial flange thick but relatively laminate, long, narrow, and in- R.M. SHELLEY & D.R. WHITEHEAD 87 conspicuous, with broadly rounded distal lobe; lateral flange represented by thick broadly rounded lobe on proximal part of distal zone, produced ven- trad, demarcated from acropodite stem by impression on lateral surface. Color in Life. — Paranota red; metaterga black, without stripes along caudal margins; col- lum without stripes along either margin. Holotype. — Length 48.3 mm, maximum width 11.4 mm, W/L ratio 23.6%, depth/ width ratio 58.8%. Segmental widths as follows: collum 7.7 mm 14th 11.2 2nd 9.1 15th 10.8 3rd 9.8 16th 10.2 4th 10.6 17th 8.9 5th-7th 10.9 18th 6.4 8th-13th 11.4 63 Fics. 60-63. Sigmoria (Cleptoria) robusta. 60, process of 4th sternum of holotype, caudal view. 61; gonopods in situ, ventral view of paratype. 62, telopodite of left gonopod of holotype, medial view. 63, the same, lateral view. Scale line for fig. 61 = 1.00 mm; line for other figs. = 1.00 mm for 62-63; 1.33 mm for 60. MEM. AMER. ENT. SOC., 35 88 XYSTODESMID MILLIPEDS Somatic features similar to /. /atior, with following exceptions: Width across genal apices 1.5 mm, interantennal isthmus 1.9 mm. Antennae reaching back to middle of 4th paranota, relative lengths of antennomeres 2>3>6>4=5>1>7. Genae with faint central impressions. Facial setae as follows: epicranial, interantennal, genal, and frontal absent, clypeal about 11-11, labral about 16-16. Terga smooth, polished, becoming coriaceous on paranota. Collum broad, ends extending well below those of adjacent tergite. Paranota moderately depressed, containing slope of dor- sum, caudolateral corners rounded through segment 5, blunt on segments 6-11, becoming pro- gressively more acute posteriorly. Peritremata sharp and distinct, strongly elevated above paranotal surface; ozopores located near midlength, opening dorsolaterad. Sternum of segment 4 with large, apically divided process between 3rd legs, much longer than widths of adjacent coxae (Fig. 60); that of segment 5 with two medially coalesced knobs between 4th legs and longer, more separate projections between Sth legs, latter equal in length to widths of adjacent coxae; sternum of segment 6 without depression on caudal edge, with two rounded, elevated areas between 7th legs, shorter than widths of adjacent coxae. Postgono- podal sterna with two blunt lobes on caudal edge of segment 8 arising from bicruciform im- pression; segment 9 also with bicruciform impression but flattened between caudal legs; re- maining sterna becoming progressively flatter and more plate-like posteriorly, with variably broad, shallow, central impressions. Coxae with low, blunt tubercles arising on segment 9, becoming larger posteriorly; prefemoral spines arising on segment 5, becoming progressively longer and sharper caudally. Gonopodol aperture elliptical, 4.9 mm wide and 2.4 mm long at midpoint, without indenta- tions, sides flush with metazonal surface. Gonopods in situ (Fig. 61, of paratype) with acropodites projecting anteriad from aperture, nearly parallel but angling towards each other and extending slightly beyond anterior margin, not overlapping or touching. Gonopod struc- ture as follows (Figs. 62-63): Prefemur with acutely triangular, ventromedial lobe; prefemoral process blunt and narrow, arising from lamina of basal zone, directed toward apical curve. Acropodite massive, heavily sclerotized, forming rectangular arch, overhanging and extending well beyond level of prefemoral process; basal zone broad, with sharply acute spine basally on outer margin, medial surface convex with anterior edge expanded into thin lamina; anterior bend sharp, well defined; peak moderately long and straight, angling slightly ventrad with apex distad; apical curve sharp, well defined; distal zone short and very broad, outer margin not in- dented, gently rounded and converging with inner to form blunt tip, with short striations on lateral surface; tip directed toward coxae. Medial flange thick, heavy, and narrow, relatively laminate, arising on distal extremity of basal zone, curving across anterior bend and narrowing on proximal part of peak, then expanding into broadly rounded distal lobe, terminating on proximal part of distal zone. Lateral flange lobe-like, located on proximal part of distal zone, produced ventrad, forming highest point of acropodite arch, poorly demarcated from acropodite stem by slight impression in distal zone, continuous with outer margin of distal zone. Prostatic groove crossing to lateral side at anterior bend, curving onto distal zone and opening terminally. Male Paratypes. — The male paratypes agree with the holotype in all particulars. Female Paratype. — Length 41.5 mm, maximum width 9.9 mm, W/L ratio 23.9%, depth/ width ratio 63.6%. Agreeing closely with males in somatic features, except ends of collum not produced beyond those of following tergite. Cyphopods in situ with corner of receptacle and valves visible in aperture, latter directed caudolaterad. Receptacle large, cupped around medial side of valves, with ridges and lobes, surface rugulose. Valves large, equal, surfaces finely granulate. R.M. SHELLEY & D.R. WHITEHEAD 89 Variation. — Except for slight changes in the length and shape of the prefemoral process and the spine on the basal zone, the gonopods in this species are essentially uniform. Likewise, all males possess elevated areas on the 6th sternum between the 7th legs and lack a depression at this point. Ecology. — Sigmoria (Cleptoria) robusta occurs under thin layers of leaves on relatively hard substrates near water sources. Distribution. — The Piedmont Plateau in the western corner of South Carolina between the Toxaway and Chatooga Rivers. The species extends to the base of the Blue Ridge Escarpment but apparently is absent from the mountains. Specimens were examined as follows: SOUTH CAROLINA. — Oconee Co., 7.1 mi. NE Walhalla, along SC hwy. 24 at Little R., M, 7 May 1977 (NCSM A1557); 4.8 mi. W Seneca, along SC hwy. 13 at Corneross Cr., 2F, 7 May 1977 (NCSM A1554); 5 mi. S Seneca, along SC hwy. 54 at Hartwell Res., 2M, 7 May 1977 (NCSM A1555); 10.8 mi. SW Seneca, along SC hwy. 168, 0.3 mi. S jct. SC hwy. 86, 2M, F, 7 May 1977 (NCSM A1553) TYPE LOCALITY; and 5.7 mi. SE Oakway, along SC hwy. 66 at Beaverdam Cr., M, F, 10 June 1978 (NCSM A2061). Anderson Co., 10.3 mi. SW Pendleton, along SC hwy. 192 at Beaverdam Cr., M, 4F, 7 May 1977 (NCSM A1552). Remarks. — Sigmoria (Cleptoria) robusta shares ancestry with abbotti and represents a population that became isolated north of the Savannah River. They are similar in the narrow medial flanges and the expanded, laminate anterior margins of the basal zones that lead into the prefemoral processes. However, the spine on the basal zone is larger in robusta, and there is a distinct distal lobe on its medial flange instead of a thickening or boss as in abbotti. Another difference obtains on the 6th sternum, which is moderately depressed between the 7th legs in abbotti and displays two elevated lobes in robusta. Sigmoria (Cleptoria) macra (Chamberlin), new combination Figs. 64-67 Cleptoria macra Chamberlin, 1939:9, pl. 4, Figs. 36-37. Chamberlin and Hoffman, 1958:28. Hoffman, 1967:7-11, Figs. 3, 5, 6, and 12. Cleptoria rileyi Loomis, 1943:402. Hoffman (1967) presented a detailed description of macra, and his medial gonopodal illustration shows the medial flange arising on the basal zone and terminating on the peak. The flange in macra is thus in the same posi- tion as those in shelfordi and arcuata (Shelley 1980a, 1981b) and a close relationship is confirmed by new material with a spur on the outer surface of the acropodite at the anterior bend (Figs. 65-67). This spur is homologous to that in shelfordi, leaving no doubt that the species are con- generic. Loomis understood this fauna better than any subsequent author, MEM. AMER. ENT. SOC., 35 90 XYSTODESMID MILLIPEDS as he (1944) assigned shelfordi to the same genus as macra, and (1943) ex- pressed doubts as to the validity of Cleptoria. Diagnosis. — A large species of Sigmoria with medial flange extending between midlength of basal zone and distal extremity of peak and with red paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral processes short and variably triangular; acropodite moderately thick and heavy, arch either flattened or gently rounded, ex- tending only slightly beyond level of prefemoral process; basal zone moderately long, without modifications; anterior bend broad, poorly de- fined; peak variable, flattened or gently curved, with or without sharply acute spur on outer surface; apical curve well defined; distal zone short, outer margins slightly indented, directed perpendicularly from peak, sides curving and narrowing to blunt tip; medial flange variable in length and configuration, arising on distal extremity of basal zone, terminating from midlength of peak to proximal portion of distal zone, usually with curved 66 Fics. 64-67. Sigmoria (Cleptoria) macra. 64, gonopods in situ, ventral view of male from Newberry Co., SC. 65, acropodite of left gonopod of the same, medial view. 66, distal part of acropodite of male from Greenville Co., SC, medial view. 67, the same, lateral view. Scale line for fig. 64 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 91 proximal margin, then indented and terminating in rounded or triangular distal lobe; lateral flange a broadly rounded lobe on proximal part of distal zone, protruding ventrad, demarcated from acropodite stem by slight im- pression on lateral surface. Color in Life. — Paranota red; metaterga black, without stripes along caudal edges; collum also without stripes along either margin. Holotype. — Process of 4th sternum moderately long, equal in length to widths of adjacent coxae (Hoffman 1967, Fig. 3). Gonopods in situ (Fig. 64, not this specimen) with acropodites crossing at midlength in midline of aperture, extending forward over anterior margin to between 7th legs. Gonopod structure as follows (Hoffman 1967, Figs. 65-67): Prefemoral process short, nearly straight, tapering to acuminate tip, directed toward distal zone. Acropodite thick and heavy, well sclerotized, forming broad, rectangular arch, overhanging and extending slightly beyond level of prefemoral process; basal zone moderately long, without modifications; anterior bend sharp, well defined; peak moderately long, broad, and flattened, rising continuously to apex at distal extremity, with thin curved lamella on outer margin of lateral surface; apical curve sharp, well defined, approximately a right angle; distal zone short, with outer edge slightly in- dented and angling sharply inward to form blunt tip; latter directed toward coxa. Medial flange relatively broad, thin, and laminate arising on distal extremity of basal zone, curving broadly across anterior bend and terminating on distal extremity of peak. Lateral flange lobe- like, located on proximal part of distal zone, broadly rounded and produced ventrad, poorly demarcated from acropodite stem by slight depression in lateral surface. Prostatic groove crossing to lateral side at anterior bend, running along outer margin of peak, curving sharply and extending through center of distal zone to terminal opening. Variation. — {1 have collected macra five times, all south of the type locality, and each male possesses a short acute spur on the outer margin at the anterior bend, or at midlength of the medial flange (Figs. 65-67). Hoff- man (1967) did not mention this feature in his account of variation, but my specimens, collected at nearly the same site in Newberry County possess it (Fig. 66). Northern populations evidently lack the spur and are atypical. Males from the center of the range in Greenville County have broadly rounded, nearly triangular lobes distally on the medial flange (Figs. 66-67). Distribution. — A linear area approximately 60 miles long in piedmont South Carolina, ranging from north of Greenville to the Saluda River in southern Newberry County. Hoffman (1967) predicted that macra would probably be found in North Carolina but I have sampled intensively north of the type locality without encountering it. Sigmoria (Cleptoria) divergens Chamberlin is common in this area and along the North Carolina-South Carolina state line in the Blue Ridge escarpment. As with simplex (Croatania), the type locality is near the northern range limit. Material was examined from the following new localities: SOUTH CAROLINA. — Greenville Co., 8.7 mi. SW Fountain Inn, along SC hwy. 68 at Reedy Cr., M, 3F, 11 June 1978 (NCSM A2069); and 12.5 mi. SW Fountain Inn, along SC MEM. AMER. ENT. SOC., 35 92 XYSTODESMID MILLIPEDS hwy. 51 at Mountain Cr., M, 11 June 1978 (NCSM A2068). Newberry Co., 5.2 mi. N Chap- pells, along SC hwy. 56, 0.2 mi. N jet. SC hwy. 347, 4M, 2F, 3 May 1977 (NCSM A1523) and M, F, 22 April 1982 (NCSM A3927). Sigmoria (Cleptoria) shelfordi (Loomis), new combination Cleptoria shelfordi Loomis, 1944:172-173, Fig. 4. Chamberlin and Hoffman, 1958:28. Brevigonus shelfordi: Shelley, 1980a:35-41, Figs. 1-13; 1981b:55-S6. Occurring along the northern side of the Savannah River in piedmont South Carolina (McCormick, Abbeville and Oconee counties), shelfordi was adequately described in two previous papers (Shelley 1980a, 1981b), the latter characterizing the gonopods in ‘‘sigmoid’’ terminology. Aspects of variation, ecology, and distribution were also discussed. No new records are available. In the first account (Shelley 1980a) two gonopodal variants were described. Variant A has a basal spine on the outer surface of the basal zone, a large medial flange that partly or completely obscures the acropodite stem in medial view, and a compact curvature with the tip directed toward the coxa. It usually lacks the spur, and the stem is apically entire. Variant B lacks the basal spine, has a narrower medial flange that reveals the prostatic groove to the crossover point, and has a more open cur- vature with the tip directed generally parallel to the coxa. This form, which includes the type specimens, possesses the spur and is usually apically in- dented. Occasionally, characters from one form appear in the other (i.e. the spur of B in A, and the apical configuration of A in B), proving that they are not reproductively isolated. Subspecific recognition is also not justified because they occur sympatrically and are interspersed throughout the range. Therefore, shelfordi is a polymorphic species. Although the two variants are intermixed in McCormick and Abbeville counties, form A tends to oc- cur more in the west proximal to the ranges of arcuata and robusta, and form B more in the east and north proximal to macra. In form B, absence of the basal spine is shared with macra, and presence of a spur is shared with southern populations of macra. Since shelfordi and macra are proximal and lack known intergrades, they are considered separate species even though geographic variation in these features crosses cladistic lines. I return to shelfordi the Oconee County record [Clemson vic., under dead pig, 2M, F, 18 July 1962, J.A. Payne (RLH)], which I (1980a) assigned to it and later (1981b) switched to arcuata. This record is troublesome because it is segregated by the range of arcuata. The form lacks a prefemoral process and has a stronger basal spine, a broader medial flange, and greater acropodal curvature than the other males of shelfordi. The Oconee County R.M. SHELLEY & D.R. WHITEHEAD 93 population, parapatric with robusta, agrees with arcuata in lacking a prefemoral process, but despite proximity, this is best considered con- vergence. Diagnosis. — A large species of Sigmoria with medial flange extending between proximal part of basal zone and distal extremity of peak and with red paranota, metaterga without stripes; gonopods with following diagnostic characters: prefemoral process present or absent, variable; acropodite heavily sclerotized, arch extending only slightly beyond level of prefemoral process; basal zone relatively long, about 2/3 of acropodite length, with or without variable basal spine on outer edge; anterior bend variable; peak about 1/3 of acropodite length, flattened or gently curved, with or without short, acute spur on outer or medial surfaces; apical curve and distal zone absent, acropodite terminating in blunt inner corner of peak, apical edge variable; medial flange thick but laminate, length and configuration variable; lateral flange absent. Color in Life. — Metaterga red; paranota black without stripes along caudal edges; collum with or without narrow stripe along anterior margin. aay, Co, Fic. 68. Distributions of australis and the subgenus Clepforia in the southeastern states. Dots, australis; triangles, rileyi; squares, abbotti; asterisk, bipraesidens; stars, robusta, diamonds, arcuata; ovals, shelfordi; half shaded dots, macra. MEM. AMER. ENT. SOC., 35 94 XYSTODESMID MILLIPEDS Sigmoria (Cleptoria) arcuata (Shelley), new combination Brevigonus arcuatus Shelley, 1981b:56-60, Figs. 1-5. Sigmoria (Cleptoria) arcuata occurs in the Savannah River Valley be- tween the Oconee County locality and all others of shelfordi. It ranges northward into Pickens County near the Blue Ridge Front. The previous paper (Shelley 1981b) described arcuata in detail and discussed variation, ecology, and distribution. Since no new records are available, I present here only a diagnosis and color statement. Diagnosis. — A large species of Sigmoria with medial flange usually ex- tending between proximal part of basal zone and midlength of peak and with red paranota, metaterga without markings; gonopods with following diagnostic characters: prefemoral process absent; acropodite moderately thick, arch a broad curve extending beyond level of prefemur; basal zone with prominent acute spine basally on caudal edge, length variable; anterior bend variable; peak gently curved or flattened; apical curve poorly defined; distal zone moderately long, curving broadly into arch, narrowing smoothly and continuously distally, tip simple or reflexed; medial flange with variable tooth on slightly wider proximal part, then narrowing and either ter- minating by blending into outer margin of peak or widening again and ter- minating abruptly at midlength of peak; lateral flange laminate, located on apical curve and proximal part of distal zone. Color in Life. — Paranota red; metaterga dark glossy black, without stripes along caudal edges; collum also without stripes along either margin. SIGMORIA (CHEIROPUS) Loomis, new status Cheiropus Loomis, 1944:170-171. Chamberlin and Hoffman, 1958:25. Jeekel, 1971:253. Hoffman, 1979:159. Shelley 1984a:265-267. Stelgipus Loomis, 1944:173. Jeekel, 1971:288. Hoffman, 1979:159. Fontaria: Chamberlin and Hoffman, 1958:33. Lyrranea Hoffman, 1963:114-115; 1979:159. Prionogonus Shelley, 1982:460-462. NEW SYNONYMY. Types species. — Of Cheiropus, C. plancus Loomis, 1944, by original designation; of Ste/gipus, S. agrestis Loomis, 1944, by original designation; of Lyrranea, L. persica Hoffman, 1963, by original designation; of Prionogonus, P. haerens Shelley, 1982, by original designation. Diagnosis. — Paranota usually red, occasionally orange, metaterga usually with concolorous stripes connecting paranotal markings, rarely uniformly black; gonopods in situ with acropodites extending well beyond anterior margin of aperture and inserting between 7th legs; prefemoral R.M. SHELLEY & D.R. WHITEHEAD 95 process present or absent, moderate when present; acropodites moderately thick and heavy, occasionally massive, well sclerotized, oriented normally on coxa with inner surface directed anteriomediad; basal zone with or without row of spurs on outer surface, occasionally extending onto peak, otherwise unmodified; medial flange present or absent, variable but usually broad and dilated, laminate to thickened, located on peak or distal part of acropodite in forms lacking distal zone; lateral flange present or absent, laminate; distal zone curving either strongly laterad and obscured in medial view by acropodite stem or medial flanges, or replaced by variably positioned solenomerite, latter best revealed in lateral view; tip variable. Remarks. — The subgenus Cheiropus, containing the southernmost apheloriine species, is united chiefly by the laterally directed distal zone or its solenomerite substitute. It lacks geographic cohesion and is divided into four species groups. The Australis group The australis group contains a relatively undifferentiated species in the Coastal Plains of South Carolina, Georgia, Alabama, and Florida. It oc- curs in three allopatric populations that have not diverged and are unques- tionably conspecific. Component. — australis, new species Sigmoria (Cheiropus) australis Shelley, new species Figs. 69-72 Type specimens. — Male holotype (NCSM A2874) and three male and two female paratypes collected by R.M. Shelley and P.T. Hertl, 15 September 1979, from Torreya State Park, Liberty Co., FL. The lushness of this park makes it a popular arthropod collecting site and australis has been taken there many times. Too numerous to be cited with complete data, additional paratypes of both sexes are available in the NCSM, FSCA, and WAS. Diagnosis. — A moderate to large species of Sigmoria with medial flange extending between distal extremities of basal zone and peak and with red paranota and variable metaterga, with or without red transverse stripes; gonopods with following diagnostic characters: acropodite moderately thick, arch in form of inverted L, extending slightly beyond level of prefemoral process; basal zone relatively long; anterior bend and apical curve well defined, latter forming arc with narrow diameter, peak short, MEM. AMER. ENT. SOC., 35 96 XYSTODESMID MILLIPEDS gently curved; distal zone short, curving laterad from peak, not coplanar with other sections, obscured in medial view by medial flange, sides narrow- ing to acuminate tip; medial flange relatively long, widening into triangular lamina on distal extremity of peak then terminating abruptly; lateral flange narrow and inconspicuous, located on distal extremity of peak. Color in Life. — Paranota red; metaterga black, with or without concolorous red stripes along caudal edges connecting paranotal markings; collum with or without red stripes along both margins. 2, Fics. 69-72. Sigmoria (Cheiropus) australis. 69, process of 4th sternum of holotype, caudal view. 70, gonopods in situ, ventral view of male from Camden Co., GA. 71, telopo- dite of left gonopod of holotype, medial view. 72, the same lateral view. Scale line for fig. 70 = 1.00 mm; line for other figs. = 1.00 mm for 71-72, 1.33 mm for 69. R.M. SHELLEY & D.R. WHITEHEAD 97 Holotype. — Length 47.3 mm, maximum width 12.8 mm, W/L ratio 27.1%, depth/ width ratio 56.3%. Segmental widths as follows: collum 7.9 mm 7th 12.6 2nd 8.5 8th-14th 12.8 3rd 9.5 15th 12.0 4th 11.0 16th 11.2 Sth 12.0 17th 9.7 6th 12.3 18th 6.8 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 5.1 mm, interantennal isthmus 1.8 mm. Antennae reaching back to middle of 4th paranota, relative lengths of antennomeres 2>3>4>5=6>1>7. Genae with distinct central impressions. Facial setae as follows: epicranial, interantennal, and genal ab- sent, frontal 1-1, clypeal about 8-8, labral 14-14. Dorsum appearing smooth and polished, but moderately coriaceous. Collum broad, ends ex- tending well below those of following tergite. Paranota relatively flat, interrupting slope of dorsum and subparallel to substrate, caudolateral corners rounded through segment 4, blunt on 5-8, becoming progressively more acute posteriorly. Peritremata distinct, sharply elevated above paranotal surface; ozopores located near middle of peritremata, opening dorsolaterad. Sternum of segment 4 with small process, barely elevated above sternal surface, much shorter than widths of adjacent coxae (Fig. 69); that of segment 5 with two low paramedial knobs between anterior legs and slightly elevated flattened areas between caudal legs; 6th ster- num with only slight depression between 7th legs, these set slightly farther apart than 6th legs. Postgonopodal sterna flattened and plate-like, with shallow transverse grooves between leg pairs on segments 8-12 and variably broad, shallow, central impressions on remaining segments. Coxae with low, rounded tubercles beginning on segment 10; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture elliptical, 4.1 mm wide and 1.3 mm long at midpoint, indented anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 70, not this specimen), with acropodites projecting ventrad from aperture, bending laterad and extending over opposite side in front of and behind each other, curving anteriad with one behind over- lapping other and one in front extending slightly beyond anterior margin. Gonopod structure as follows (Figs. 71-72): Prefemoral process short, blunt, directed toward distal zone. Acropodite moderately thick, forming inverted L shaped arch, overhanging and extending slightly beyond level of prefemoral process; basal zone relatively long, unmodified; anterior bend sharp, well defined; peak relatively short, gently curved; apical curve relatively sharp, well defined, forming arc with very narrow diameter; distal zone short, curving laterad from peak, not coplanar with other regions, obscured in medial view by medial flange; tip acuminate. Medial flange arising on distal extremity of basal zone, widening into triangular lamina on distal extremity of peak and terminating abruptly thereafter. Lateral flange a short, narrow, inconspicuous lamella on peak opposite trianglar part of medial flange. Prostatic groove crossing to lateral side at anterior bend, continuing to terminal opening. Male Paratypes. — Some male paratypes have more acuminate prefemoral processes, but otherwise all agree closely with the holotype. Female Paratype. — Length 42.9 mm, maximum width 11.8 mm, W/L ratio 27.5%, depth/ width ratio 62.7%. Cyphopods in situ with edge of receptacles and opening of valves visible in aperture. Receptacle moderate, situated on medial sides of valves, with ridge on this side, sur- face rugulose. Valves moderate, equal, surfaces finely granulate. MEM. AMER. ENT. SOC., 35 98 XYSTODESMID MILLIPEDS Variation. — Aside from minor changes in the length and bluntness of the prefemoral process and the breadth of the medial flange, the gonopods of australis are quite constant throughout the range. Size and color pattern vary, however, and as noted under ‘‘Color in Life,’’ individuals may or may not have red stripes along the caudal margins of the metaterga and collum. The types and the sample from Decatur County, Georgia, lack stripes, but as far as one can tell from faded preserved specimens, all others possess them. All the individuals I collected in Alabama, Georgia, and South Carolina were striped. Regarding size, the types are near maximum for the species. The opposite extreme is found along the Atlantic Coast, as shown by a male from Beaufort County, South Carolina, which measures 29.1 mm in length, 7.4 mm in width, W/L ratio 25.4%, depth/ width ratio 53.4%. Large males ap- pear extremely broad in dorsal view, but this is an illusion created by the unusually flat paranota. This effect is less noticeable in small individuals. Ecology. — Sigmoria (Cheiropus) australis occurs in a variety of mesic deciduous forests under thin layers of leaves on relatively hard substrates, preferably near water sources. In Crooked River State Park, Camden County, Georgia, I found australis on sandy-humus soil in a live oak- magnolia forest, but it was absent from the litter of these species occurring instead under other hardwoods. It occurs syntopically with serrata at this site, but they are dominant at different times of the year (Shelley 1984a). Sigmoria (Cheiropus) australis was more abundant in July 1977 but was ab- sent in October 1980, when serrata was present. The life histories therefore seem to be adjusted to minimize ecological competition. A comparable situation may exist in western Georgia at Kolomoki Mounds State Park, Early County, where australis and Dynoria medialis Chamberlin occur syn- topically in typical climax piedmont forest. I have visited this locality twice and found them to be equally abundant in November 1977, whereas D. medialis was more common in May 1983. Distribution. — The Coastal Plains of Alabama, Georgia, southern South Carolina, and northern Florida, ranging from near the Alabama River to the Atlantic Ocean and extending inland to the edge of the Fall Zone, just crossing the Alabama River northwest of Montgomery. The distribution spans several large rivers including the Chattahoochee, Altamaha, and Savannah, but australis has not been encountered south of the St. Marys River in Nassau or Duval counties, Florida. As shown in Fig. 68 the available material clusters into three disjunct areas — the Atlantic coast from Hampton County, South Carolina, to Camden County, Georgia; along both sides of the Chattahoochee River from the inner Coastal Plain of Russell County, Alabama, to Liberty County, Florida; and in central Alabama from Autauga to Conecuh counties. There are no R.M. SHELLEY & D.R. WHITEHEAD 99 anatomical differences between these groups, however, and all are unques- tionably conspecific. Additional collecting may eventually connect the two areas in Alabama, but I doubt if the Chattahoochee and Atlantic Coastal populations will ever be connected. Specimens were examined as follows: ALABAMA: Autauga Co., 2.5 mi. SE Prattville, M, F, 26 August 1960, L. Hubricht (RLH). Lowndes Co., 3 mi. E Braggs, 2F, 4 July 1960, L. Hubricht (RLH). Wilcox Co., along AL hwy. 41, 2.5 mi. S Dallas Co. line, 2M, F, 22 May 1980 (NCSM A3124); 3.5 mi. N Camden, 2M, 11 November 1962, L. Hubricht (RLH); and 2 mi. S Oak Hill, M, 10 May 1960, L. Hubricht (RLH). Butler Co., 3 mi. SE Searcy, 6M, 5F, 9 April 1960, L. Hubricht (RLH); 9.8 mi. NE Greenville, along US hwy. 31 at Pigeon Cr., F, 2 juvs., 21 April 1983 (NCSM A4039); and 3 mi. NW McKenzie, along US hwy. 31, 2M, 2F, 9 April 1960, L. Hubricht (RLH). Monroe Co., 12 mi. NNE Monroeville, along AL hwy. 21, 4.9 mi. N jct. AL hwy. 265, M, 22 April 1983 (NCSM A4046). Conecuh Co., 2.2 mi. SW Evergreen, 10M, 10F, 26 June 1961, L. Hubricht (RLH). Russell Co., 3 mi. N Uchee, 4M, 2F, 12 June 1960, L. Hubricht (RLH); and 14.9 mi. S. Phenix City, along AL hwy. 165, 0.3 mi. N jct. AL hwy. 38, 2F, 26 April 1983 (NCSM A4064). Barbour Co., 13 mi. E Clayton, M, F, 5 September 1959, L. Hubricht (RLH) and 2.7 mi. S Eufaula, along US hwy. 431, 0.5 mi. S jct. AL hwy. 30, F, 29 April 1983 (NCSM A4067). Henry Co., 5 mi. W Capps, M, 6 August 1960, L. Hubricht (RLH) and 2.8 mi. NE Abbeville, along AL hwy. 47, 2.1 mi. E jct. AL hwy. 95, 2F, 29 April 1983 (NCSM A4069). Dale Co., 5.6 mi. NE Ozark, along AL hwy. 105 at Judy Cr., M, 29 April 1983, (NCSM A4071). Houston Co., 4 mi. E Webb, 2F, 9 July 1967, D.R. Whitehead (RLH) and Chattahoochee St. Pk., 4M, 3F, 30 April 1983 (NCSM A4080). FLORIDA: Jackson Co, 5.7 mi. W Greenwood, M, 11 July 1973, R.M. Blaney (FSCA); Florida Caverns St. Pk., several males and females taken in April, May, June, and July from 1957-1975 by various collectors (FSCA, WAS); Marianna, males and females collected from 1961-1970 by various persons (FSCA); Three Rivers St. Pk, 8M, 5F, 19 July 1953, W.A. Shear (WAS); and 3.3 mi. E Sneads, along Appalachicola R., 3M, 2F, 7 September 1959, L. Hubricht (RLH). Liberty Co., Torreya St. Pk., 4M, 2F, 15 September 1979 (NCSM A2874) and several other males and females taken from March-October from 1968-1977 by various collectors (NCSM, FSCA, WAS) TYPE LOCALITY. GEORGIA: Early Co., Kolomoki Mounds St. Pk., 2M, 19 November 1977 (NCSM A1783) and M, 1 May 1983 (NCSM A4023). Decatur Co., 1.4 mi. W Climax, 2M, F, 7 September 1959, L. Hubricht (RLH). Camden Co., Crooked River St. Pk., 3M, 8F, 3 July 1977 (A1605) and M, 2 October 1980 (NCSM A3590). Chatham Co., 1.7 mi. E Silk Hope, 2M, 12 September 1959, L. Hubright (RLH). SOUTH CAROLINA: Hampton Co., 0.8 mi. SE Hampton, M, 19 September 1959, L. Hubricht (RLH). Beaufort Co., 2.5 mi. W. Bluffton, along SC hwy. 46, M, 12 September 1980 (NCSM A3507). Jasper Co., 7 mi. S Hardeeville, M, 19 September 1959, L. Hubricht (RLH); and Ridgeland, M, 28 March 1975, D. Brody (AMNH). Remarks. — Sigmoria (Cheiropus) australis has been reported twice before (Shelley 1979a, 1984a), first as representing an undiagnosed apherloriine genus and secondly as an undescribed species of Hubroria. I once thought australis related to species in the Cumberland Plateau of Ten- nessee for which the name Hubroria is available. Since the medial flange is broadest distad and the distal zone curves laterad from the peak, australis does conform in these traits to the Cumberland species. I therefore have MEM. AMER. ENT. SOC., 35 100 XYSTODESMID MILLIPEDS spent days searching for Sigmorias from Jackson and Marshall counties to Birmingham, and since these efforts were unproductive, I do not think that the Cumberland fauna occurs south of the Tennessee River. Except for the medial flanges there is little similarity between the gonopods of australis and those of any Cumberland species as there would be in the case of a direct relationship. Only after revising Cheiropus (Shelley 1984a) did the af- finity of australis for stibarophalla become evident. The latter, in the eastern Blue Ridge Mountains of North Carolina, is some 200 miles north- west of the closest known locality of australis in South Carolina. The region between stibarophalla and australis is occupied by more divergent forms of Cheiropus, Like stibarophalla, australis has a short prefemoral process, a sharp apical curve forming an arc with very narrow diameter, a short distal zone that curves laterad from the peak, and a medial flange that arises at the anterior bend and terminates at the apical curve, obscuring the distal zone in medial view. The similarities between their gonopods is evident by com- paring Figs. 71-72 with Figs. 25-26 in Shelley (1981a). Of nearly equal interest to the relationship with the allopatric stibarophalla are the allopatric populations of australis (Fig. 68). The two Alabama areas may eventually be joined by further collecting, but I doubt that the middle one will ever be connected with the eastern. There is no anatomical divergence and the forms in all three areas vary within similar limits. The Divergens Group The divergens group was proposed by Shelley (1983a) for a single species, but it is now enlarged to accommodate two species in the eastern Blue Ridge Mountains of the Carolinas. The distal zones curve strongly laterad from the peak but maintain their identities as distinct acropodal regions. Details are presented in Shelley (198la, 1983a). Components. — divergens Chamberlin, stibarophalla Shelley The Haerens Group In this and the following group, the distal zone is indistinguishable as a separate acropodal region and is represented instead by a solenomerite that is best viewed in lateral perspective. In the haerens group, the structure is located terminally on the distal extremity of the peak, perpendicular to the acropodal axis in two species and coaxial in one. The three included species, previously placed in the genus Prionogonus, were discussed in detail by R.M. SHELLEY & D.R. WHITEHEAD 101 Shelley (1982). They are united by a synapomorphy, the row of spurs along the basal zone. Components. — haerens (Shelley), divaricata (Shelley), thrinax (Shelley). The Planca Group The planca group contains the southernmost species in Sigmoria, which were formerly the sole components of the genus Cheiropus (Shelley 1984a). Here the solenomerite occurs at different positions on the acropodite stem, but always proximal to the tip. The basal zones are unmodified. The acropodite of agrestis displays a sigmoid curvature, which is absent in the curvilinear species planca and serrata. The peak is also greatly enlarged and thickened in these three species and often possesses marginal dentations. The fourth species, persica, is highly modified with the solenomerite arising basally from the acropodite. Further details are available in Shelley (1984a). Components. — planca (Loomis), agrestis (Loomis), persica (Hoffman), serrata (Shelley) Additional Record. — serrata — GA, Camden Co., Cumberland Island, M, 5 July 1984 (NCSM A4223). SIGMORIA (SIGIRIA) Chamberlin, new status Sigiria Chamberlin, 1939:9. Chamberlin and Hoffman, 1958:48. Jeekel, 1971:287. Type species. — Sigiria scorpio Chamberlin, 1939 [=Sigmoria rubromarginata (Bollman)], by original designation. Diagnosis. — Paranota yellow, red or violet/purple, metaterga with con- colorous stripes connecting paranotal markings; gonopods in situ with acropodites extending well beyond anterior margin of aperture and insert- ing between 7th legs; prefemoral process moderate; acropodites thin and fragile to moderately thick and heavy, oriented normally on coxa with inner surface directed anteriomediad; basal zone without modifications; medial flange present, laminate, location varying from proximal portion of peak to distal zone, lateral flange usually present, variably laminate; distal zone variable but usually more or less coplanar with basal zone. Remarks. — Sigiria is revived from synonymy under Sigmoria for a heterogeneous assemblage in the southern Appalachians with concolorous paranota and metatergal stripes. The acropodites are highly variable but tend to connect through a spectrum of intermediate forms. Two species groups are recognized partly on the basis of geography, and partly on color. MEM. AMER. ENT. SOC., 35 102 XYSTODESMID MILLIPEDS The Rubromarginata Group The rubromarginata group includes forms previously in this group, the nigrimontis and inornata [=simplex (Shelley 1981a)] groups, plus a new species on the Blue Ridge escarpment of Virginia. All except the last have red paranota and red metatergal stripes, this species being yellow. The acropodites are highly variable, but the extremes tend to join through forms of nigrimontis intermedia, which is appropriately named. The species-group name of simplex (Shelley 198la) is changed to avoid homonymy with simplex Shelley (1977), and the former eastern race of rubromarginata, which is continuous with intergrades on the Blue Ridge Front, is elevated to specific status. Its range is isolated from rubromarginata by around 30 miles, and anatomical divergence has occurred. Components. — rubromarginata (Bollman); austrimontis Shelley; whiteheadi, new species; nigrimontis n. nigrimontis (Chamberlin), n. in- termedia (Hoffman), n. angulosa Shelley, n. unicoi (Shelley); inornata, new name; truncata Shelley; sigirioides Shelley. Sigmoria (Sigiria) austrimontis Shelley, new status Sigmoria rubromarginata austrimontis Shelley, 1981a:102-103, Figs. 100-103. Under this binomial I include the homogeneous population in the South Mountains of North Carolina and the contiguous, heterogeneous assemblage of forms anatomically intermediate between it and rubromarginata occurring in the Blue Ridge escarpment and western pied- mont lowlands. Genetic interchange with rubromarginata is no longer possible, as the ranges are now disjunct, and as stated in the introduction, such populations are now recognized at the specific level. This situation is clearly one where a formerly continuous range, with localized gene pools in the peripheries and intergrade forms in an intermediate geographical posi- tion, underwent vicariance partitioning. Sigmoria (Sigiria) whiteheadi Shelley, new species Figs. 73-76 Type specimens. — Male holotype and two female paratypes (RLH) col- lected by R.L. Hoffman, 20 May 1983, from Patrick Co., VA, along Laurel Creek on Blue Ridge parkway at mile 174.3. Male paratype (RLH) taken by same collector at same locality, 23 June 1984. Diagnosis. — A small species of Sigmoria with medial flange on peak and with yellow paranota and yellow metatergal stripes; gonopods with following diagnostic characters: prefemoral process short, blunt; R.M. SHELLEY & D.R. WHITEHEAD 103 acropodite moderately thick and heavy, curvature forming narrow arc; peak short and gently curved; distal zone curving laterad from peak, not coplanar with basal zone, bent sharply inward into arch at midlength; medial flange thin and narrow, margin linear, poorly demarcated from acropodite stem; lateral flange located opposite medial, flared outward proximad. Color in Life. — Paranota bright lemon yellow; metaterga black with concolorous yellow stripes along caudal margins connecting paranotal markings; collum with yellow stripes along both anterior and posterior edges. Fics. 73-76. Sigmoria (Sigiria) whiteheadi. 73, process of 4th sternum of holotype, caudal view. 74, gonopods in situ, ventral view of holotype. 75, telopodite of left gonopod of the same, medial view. 76, the same, lateral view. Scale line for fig. 74 = 1.00 mm; line for other figs. = 1.00 mm for 73 and 76; 1.25 mm for 75. 104 XYSTODESMID MILLIPEDS Holotype. — Length 31.1 mm, maximum width 6.9 mm, W/L ratio 22.2%, depth/ width ratio 62.3%. Segmental widths as follows: collum 5.1 mm 15th 6.5 2nd 6.1 16th 6.2 3rd 6.5 17th 5.5 4th 6.7 18th 4.4 Sth-14th 6.9 Somatic features similar to those of /. /atior, with following exceptions: Width across genal apices 3.7 mm, interantennal isthmus 1.1 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>4=6>5>1>7. Genae with trace of central impressions. Facial setae as follows: epicranial 2-2, interantennal absent, frontal 1-1, genal 3-3, clypeal about 8-8, labral about 14-14. Terga smooth, polished, becoming moderately coriaceous on paranota. Collum broad, ends broadly rounded and extending slightly beyond those of following tergite. Paranota moderate- ly depressed, continuing slope of dorsum, caudolateral corners rounded through segment 9, blunt on 10-15, and more acute caudally. Peritremata relatively flat, faintly elevated above paranotal surface. Ozopores located caudal to midlength, opening dorsolaterad. Process of 4th sternum small, divided, much shorter than widths of adjacent coxae (Fig. 73); 5th sternum with low paramedian knobs between 4th legs and flattened, elevated areas between 5th, both much shorter than widths of adjacent coxae; 6th sternum with slight recession be- tween caudal legs to accommodate apical curvatures of acropodites. Postgonopodal sterna elevated above stricture, flattened, with bicruciform impressions on 8-9 and variably broad, shallow central depressions thereafter. Coxae without tubercles; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture ovoid, 2.6 mm wide and 1.4 mm long at midpoint, indented slightly anteriolaterad, sides elevated above metazonal surface. Gonopods jn situ (Fig. 74) with acropodites projecting anteriomediad from aperture, overlapping in midline and curving dor- sad over opposite side, extending only slightly beyond anterior margin. Gonopod structure as follows (Figs. 75-76): Prefemoral process short, blunt, directed toward peak. Acropodite moderately thick and heavy, well sclerotized, tightly curved forming narrow arch, overhanging and extending well beyond level of prefemoral process; basal zone relatively long, widest basal- ly; anterior bend sharp, well defined; peak short and gently curved; apical curve broad, poorly defined; distal zone moderately long, curving slightly laterad from peak and not coplanar with basal zone, bent abruptly (90°) inward into center of arch at midlength or termination point of flanges, relatively straight distal to bend with sides narrowing smoothly and continuously to acuminate tip; latter directed toward basal zone. Medial flange thin and narrow, arising im- perceptibly on proximal part of peak but terminating clearly at midlength of distal zone, margin curving distad. Lateral flange larger than and located opposite medial, arising on distal extremity of peak, terminating at midlength of distal zone, flared outward proximad, margin linear thereafter. Prostatic groove crossing to lateral surface at anterior bend, continuing to terminal opening. Female paratype. — Length 31.8 mm, maximum width 7.7 mm, W/L ratio 24.2%, depth/ width ratio 65.0%. Cyphopods in situ with corners of receptacles visible in apertures, valves directed dorsolaterad. Receptacle large, cupped over ventral corners of valves, surface rugulose. Valves moderate in size, subequal, surfaces finely granulate. Male paratype. — The male paratype agrees with the holotype in all details. Ecology. — Dr. Hoffman discovered the type specimens in rhododen- dron litter in a rhododendron/red maple woods. Since this spot is along a R.M. SHELLEY & D.R. WHITEHEAD 105 creek, whiteheadi seems to be a cove species. It is only the second one found in rhododendron leaves, the other being truncata (Shelley 1981a). Others occur near rhododendron, but always in litter of associated hardwood species, usually red maple or dogwood. A male of /. latior (Sigmoria) was collected only a foot or so from the male paratype. Distribution. — Known only from the type locality. Remarks. — The closest collection of the rubromarginata group to whiteheadi is Morganton, North Carolina, approximately 120 miles SSW, where austrimontis occurs. The acropodite of whiteheadi is thicker and heavier than that of austrimontis, but in both the distal zones are twisted approximately 90° mediad revealing the faces of the medial and lateral flanges in medial view. However, the apical curve is narrower than in austrimontis, forming an arc of shorter diameter, and the distal zone distal to the flanges curves more strongly into the acropodite arch. Thus, the distal half of the distal zone, which is longer than in austrimontis, is partly obscured in medial view by the flanges. The prefemoral process is also longer than in austrimontis and is not bifurcate. Sigmoria (Sigiria) inornata Shelley, new name Sigmoria simplex Shelley, 1981a:45-49, Figs. 29-34. Sigmoria (Sigiria) inornata is proposed as a new name for this species to avert homonymy resulting from the inclusion of Croatania in Sigmoria and the resultant transferral of C. simplex Shelley, 1977. Priority for the species-group name is with the transferred species. The Stenogon Group As presently conceived, the stenogon group is markedly different in com- position from that in its original proposal (Shelley 1981a). One species, nan- tahalae (Falloria), is now in another subgenus, and two species have been added that were previously in the /atior and nigrimontis groups. the gonopods of the three species of the stenogon group appear dissimilar in medial view, but considerable conformity is apparent in lateral perspective (compare Figs. 22, 60, and 130 in Shelley (1981a)). The color varies from yellow in the north through red in the central part of the group’s range to purple or violet in the south. Components. — stenogon Chamberlin, areolata Shelley, disjuncta Shelley. MEM. AMER. ENT. SOC., 35 106 XYSTODESMID MILLIPEDS SIGMORIA (FALLORIA) Hoffman, new status Falloria Hoffman, 1948:93-94. Chamberlin and Hoffman, 1958:33. Jeekel, 1971:264. Hoff- man, 1979:159. Hubroria Keeton, 1960:42. Hoffman, 1979:159. NEW SYNONYMY. Type species. — Of Falloria, Apheloria bidens Causey, 1942, by original designation; of Hubroria, H. picapa Keeton, 1960, by original designation. Diagnosis. — Paranota usually red, occasionally white or light yellow, metaterga with contrasting stripes, usually blue, connecting paranotal markings except when latter white or light yellow; gonopods in situ with acropodites extending well beyond anterior margin or aperture and insert- ing between 7th legs; prefemoral process highly variable, deeply or apically divided, long and basally globose or moderate or short; acropodites usually moderately thick and heavy, oriented normally on coxa with inner surface directed anteriomediad or with inner surface directed mediad; basal zone usually without modifications, occasionally with dense tubercles; medial flange usually present and laminate, location varying from basal to distal zones; lateral flange present or absent, variably laminate; distal zone variable, either coplanar with basal zone or directed strongly laterad and nearly coplanar with peak. Remarks. — Falloria is revived from synonymy under Sigmoria as the oldest available name for the heterogeneous assemblage with largely con- trasting paranotal and metatergal colors in the western part of the generic range. The acropodites are highly variable, and eight species groups are recognized. The Nantahalae Group The nantahalae group contains only the single species, which is unique to the subgenus in the red/white coloration, the absence of a medial flange, and the presence of a tooth. Further details are available in accounts by Hoffman (1958a) and Shelley (1981a). Component. — nantahalae Hoffman. The Leucostriata Group The /eucostriata group is characterized by a greatly reduced medial flange on the proximal part of the peak, small medial and lateral lobes opposing each other proximally on the distal zone, and a curved or bent distal zone R.M. SHELLEY & D.R. WHITEHEAD 107 which extends into the arch. It consists of two allopatric species in the western fringe of the Blue Ridge Province in Tennessee and north Georgia. One, /eucostriata, is the only member of the subgenus with concolorous paranota/metaterga; the other, xerophylla, exhibits the typical red/blue pattern. Components. — leucostriata Shelley, xerophylla Shelley. The Bidens Group Previously, the bidens group was monobasic, but I now add an undescribed species with a contiguous range to the north. It has been known for years, but thinking that it might be referrable to Hubroria, I withheld description until the validity of this name could be assessed. Whereas bidens has a long, narrow medial flange on the proximal part of the peak and a separate subconical tooth distal to the flange (Shelley 198la), the new species lacks a tooth and its short medial flange has a broad variable lobe, arises distally on the peak, and terminates on the proximal part of the distal zone. Its affinity for bidens is demonstrated by the prefemoral processes. That of bidens is basally globose, while that of the new species is variable and moderately globose in the most proximal population. The bidens group occurs in the Sevier County, Tennessee, portion of the GSMNP, from Elk- mont to Greenbrier. Components. — bidens (Causey), prolata, new species. Sigmoria (Falloria) prolata Shelley, new species Figs. 77-83 Type specimens. — Male holotype (NCSM A1939) and 4 male and 7 female paratypes collected by R.M. Shelley and W.B. Jones, 19 May 1978, in the Ramsey Cascade Parking Area, Greenbrier Section, GSMNP, Sevier Co., TN. Male and female paratypes deposited in FSCA. Diagnosis. — A large species of Sigmoria with red paranota and blue transverse metatergal stripes; gonopods with following diagnostic characters: prefemoral process long, configuration variable, widest basally, tapering distad, extending beyond level of tip of acropodite; latter moderately thick and heavy, arch broadly curved and overhanging prefemoral process; basal zone continuous with peak and distal zone through broad, poorly defined anterior bend and apical curve; distal zone variable but generally moderately long, directed laterad from peak, not coplanar with basal zone, bent abruptly inward into arch at midlength in medial view; tip elongate and acuminate or reflexed; medial and lateral MEM. AMER. ENT. SOC., 35 108 XYSTODESMID MILLIPEDS flanges short, opposing each other on distal zone and distal extremity of peak, with variable lobes. Color in Life. — Paranota red; metaterga black with wide, blue stripes along caudal edges connecting paranotal markings; collum with blue stripes along both anterior and posterior edges. Holotype. — Length 41.1 mm, maximum width 9.7 mm, W/L ratio 23.6%, depth/ width ratio 58.8%. Segmental widths as follows: collum 7.8 mm 14th 9.4 2nd 8.6 15th 9.0 3rd 9.1 16th 8.5 4th 9.4 17th 7.3 Sth-13th 9.7 18th 5.7 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.9 mm, interantennal isthmus 1.7 mm. Antennae extending back to caudal edges of 3rd paranota, relative lengths of antennomeres 2>3>6>5>4>157. Genae with distinct central impressions. Facial setae as follows: epicranial, interantennal, and genal absent, frontal 1-1, clypeal about 10-10, labral 16-16. Dorsum smooth, polished, with only faint wrinkling on anterior half of paranota. Collum broad, ends produced slightly beyond those of following tergite. Paranota moderately de- pressed, continuing slope of dorsum, caudolateral corners rounded through segment 6, blunt on 7-13, becoming progressively more acute posteriorly. Peritremata distinct, strongly elevated above paranotal surface, ozopores located caudal to midlength, opening dorsolaterad. Process of 4th sternum small, apically divided, much shorter than widths of adjacent coxae (Fig. 77); knobs between anterior legs of 5th sternum minute, lower than broad elevated areas between posterior legs; 6th sternum convexly recessed between 7th legs to accommodate apical curvature regions of acropodites. Postgonopodal sterna flat and plate-like, with bicruciform impressions on segments 8-9 and variably broad, shallow, central impressions on remaining segments. Coxae with low, blunt tubercles beginning on segment 8, becoming sharply acute on 10-15 and lowly rounded on remaining segments; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture elliptical, 4.0 mm, wide and 1.9 mm long at midpoint, indented slightly anteriolaterad, sides elevated above metazonal surface and thickened. Gonopods in situ (Fig. 78, of paratype) with acropodites overlapping near midlengths in midline of aperture, extend- ing forward beyond anterior margin with distal zones crossing. Gonopod structure as follows (Figs. 79-80): Prefemoral process long, extending beyond level of tip of acropodite, widest basally, tapering smoothly and continuously to acuminate tip, latter curved gently anteriad, directed toward apical curve. Acropodite moderately thick, a broad, continuous curve with in- distinguishable regions overhanging and extending well beyond level of prefemoral process; peak and basal zone continuous through anterior bend and continuous with distal zone through apical curve; distal zone moderately long, curving laterad and not coplanar with basal Fics. 77-83. Sigmoria (Falloria) prolata. 77, process of 4th sternum of holotype, caudal view. 78, gonopods in situ, ventral view of paratype. 79, telopodite of left gonopod of holotype, medial view. 80, the same, lateral view. 81, telopodite of left gonopod of male from Porter Creek parking area, medial view. 82, distal part of acropodite of the same, lateral view. 83, telopodite of left gonopod of male from Roaring Fork Nature Trail, medial view. Scale line for fig. 78 = 1.00 mm; line for other figs. = 1.00 mm for 79-83, 1.33 mm for 77. R.M. SHELLEY & D.R. WHITEHEAD MEM. AMER. ENT. SOC., 35 109 110 XYSTODESMID MILLIPEDS zone, bent abruptly inward into arch at midlength and tapering smoothly and continuously to acuminate tip; latter directed toward distal curve of prefemoral process; medial flange located opposite lateral flange, arising on distal extremity of peak, terminating in broad, triangular lobe at midlength of distal zone. Lateral flange arising at apical curve, terminating at midlength of distal zone, margin gently rounded. Prostatic groove crossing to lateral side on basal zone, continuing to terminal opening. Male Paratypes. — The male paratypes agree with the holotype in all particulars. Female Paratype. — Length 42.8 mm, maximum width 10.4 mm, W/L ratio 24.3%, depth/ width ratio 68.3%. Cyphopods in situ with corner of receptacles visible in apertures, valves directed caudolaterad. Receptable very large, completely enveloping valves, with ridges and lobes, surface rugulose. Valves small, subequal, surfaces finely granulate. Variation. — Considerable variation is evident in both the prefemoral process and acropodite of prolata. All samples from the Greenbrier section of the GSMNP differ from the types, which are from the northernmost point reachable by road in this area. Those from Porter Creek parking area (NCSM A3007), the southernmost point in Greenbrier, have a more linear prefemoral process, a shorter distal zone with a reflexed tip, and a short distal projection on the medial flange (Figs. 81-82). Males collected along Rhododendron Creek on the entrance road to Greenbrier (NCSM A1942) resemble those from Porter Creek except the medial flange is trapezoidal and lacks the projection. The male collected along Roaring Fork Nature Trail (NCSM A1895), about 4 miles south of Greenbrier, displays a broadly linear medial flange that is not rounded, a narrow and inconspicuous lateral flange, a long distal zone, a reflexed tip, and a prefemoral process that is moderately globose basally and similar to that in contiguous populations of bidens (Fig. 83). Thus specimens from the geographic extremes, Ramsey Cascade and Roaring Fork, exhibit long distal zones, but only those from the former have reflexed tips. There are also north-south geographic trends toward a narrower medial flange and a basally broader prefemoral process, which grades into the globose condition found in bidens (see Shelley 1981a, Figs. 80-81, p. 82). Ecology. — Sigmoria (F.) prolata is a cove inhabiting species. Distribution. — Known only from the Greenbrier and Roaring Fork sec- tions of the GSMNP near Gatlinburg in Sevier County, Tennessee. I have searched in vain for prolata along TN highway 73, which runs along the western edge of the Park in Tennessee and connects these two sections, and thus believe it to be restricted to these secluded areas of the Park. The species is abundant in Greenbrier and can be easily collected in May or June. It is comparatively rare in Roaring Fork, however, as two subsequent trips to the site of the first collection failed to produce more individuals. Specimens were examined as follows: TENNESSEE: Sevier Co., Greenbrier Section, GSMNP, 8.2 mi. ENE Gatlinburg, Ramsey Cascade pkg. area, 5M, 7F, 19 May 1978 (NCSM A1939) TYPE LOCALITY; 7.5 mi. ENE R.M. SHELLEY & D.R. WHITEHEAD 111 Gatlinburg, Porter Creek pkg. area, 9M, SF, 8 May 1980 (NCSM A3007); 4.0 mi. ENE Gatlin- burg, along Greenbrier entrance rd. 1.8 mi. E jct. TN hwy. 73, 2M, 3F, 19 May 1978 (NCSM A1942); and unknown sites in Greenbrier, M, 14 June 1939, D.H. Lowrie, and M, 15 June 1942, C.H. Seevers (both RLH). Roaring Fork Nature Trail between 3rd and 4th bridges, 2.9 mi. ESE Gatlinburg, M, F, 16 May 1978 (NCSM A1895). Remarks. — Sigmoria (F.) prolata links forms in the GSMNP, which are clearly referrable to Sigmoria s. lat., with those in the Cumberland Plateau with laterally directed distal zones that could be assigned to Hubroria. Hence, this species is a major reason why the latter name is placed in synonymy. The Tuberosa Group The single species of this group is characterized by a number of apomor- phic traits including hirsute postgonopodal sterna, a circular gonopodal aperture, tubercles on the outer surface of the basal zone of the acropodite, fusion of the tooth and medial flange, an additional flange on the inner sur- face of the distal zone, and a unique, complex tip. The species is known only from the eastern part of the GSMNP and adjacent areas in Swain County, North Carolina. Component. — tuberosa Shelley. The Aphelorioides Group A single species group is proposed for aphelorioides. The general con- figuration of the prefemoral process is shared with xerophylla and nan- tahalae, and the circular acropodite is similar to that of ainsliei; however, there are important differences in the two loops as discussed in the species accounts. The aphelorioides group occurs between xerophylla and the translineata group in the Great Smoky Mountains south of the National Park. As with the latter, it spans the state line, occurring in the western fringe of North Carolina. Components. — aphelorioides, new species. Sigmoria (Falloria) aphelorioides Shelley, new species Figs. 84-87 Type specimens. — Male holotype (NCSM A2468) and 3 male and 2 female paratypes collected by R.M. Shelley and W.B. Jones, 12 October 1978, from Monroe Co., TN, 17.3 mi. SE Madisonville, along TN highway 165, 0.5 mi W Bald River Falls, Cherokee National Forest. Three male MEM. AMER. ENT. SOC., 35 112 XYSTODESMID MILLIPEDS paratypes (A2466) taken by same collectors on same date, 14.3 mi SE Madisonville, along TN highway 165 at Tellico Ranger Station, Cherokee National Forest. Two male and one female paratypes (RLH) collected by L. Hubricht, 12 June 1953, from Monroe Co., 1.5 mi. E Tellico Plains, Tellico River Gorge. Male and female paratypes deposited in FSCA. Diagnosis. — A large species of Sigmoria without medial acropodal flange, with red paranota and blue transverse metatergal stripes, gonopods 87 Fics. 84-87. Sigmoria (Falloria) aphelorioides. 84, process of 4th sternum of holotype, caudal view. 85, gonopods in situ, ventral view of paratype. 86, telopodite of left gonopod of holotype, medial view. 87, the same, lateral view. Scale line for fig. 85 = 1.00 mm; line for other figs. = 1.00 mm for 84; 1.14 mm for 86-87. R.M. SHELLEY & D.R. WHITEHEAD 113 with following diagnostic characters: acropodites in situ usually lying over and under one another; prefemoral process relatively long, upright, acuminate; acropodite moderately thick and heavy, slightly overhanging prefemoral process, configuration circular, anterior bend and apical curve broad but well defined, distal zone not coplanar with basal zone, extending laterad from peak, curving down behind arch of acropodite and extending nearly to level of basal zone, expanded proximal to tip, latter acuminate, bent abruptly dorsad. Color in Life. — Paranota red; metaterga black with wide, blue transverse stripes along caudal edges connecting paranotal spots; collum with blue stripes along both anterior and caudal edges. Holotype. — Length 42.4 mm, maximum width 10.7 mm, W/L ratio 25.2%, depth/ width ratio 61.7%. Segmental widths as follows: collum 7.0 mm 14th 10.5 2nd 8.9 15th 10.2 3rd 9.6 16th 9.9 4th-8th 10.4 17th 8.7 9th-13th 10.7 18th 6.1 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.8 mm, interantennal isthmus 1.9 mm. Antennae reaching back to middle of 4th paranota, relative lengths of antennomeres 2>3>4=5=6>1>7. Genae with slight central impressions. Facial setae as follows: epicranial 1-1, interantennal and frontal ab- sent, genal 2-2, clypeal about 15-15, labral about 18-18, merging with clypeal series and con- tinuing for short distance along genal border, 2 setae per side. Terga relatively coriaceous in middorsum, more so on paranota. Collum broad, ends extend- ing well beyond those of following tergite. Paranota moderately depressed, angled ventrad and continuing slope of dorsum; caudolateral corners rounded on segments 1-4, becoming blunt and progressively more acute thereafter. Peritremata distinct, sharply elevated above metazonal surface; ozopores located in swelling just caudal to midlength, directed dor- solaterad. Process of 4th sternum (Fig. 84) deeply divided apically, length subequal to widths of adja- cent coxae; Sth sternum with small paramedial knobs between 4th legs and larger knobs between 5th legs; both shorter than widths of adjacent coxae; 6th sternum convexly recessed between 7th legs to accommodate curvature of acropodites. Postgonopodal sterna relatively flat and plate-like, with shallow central impressions, becoming deeper caudally. Coxal tubercles beginning on 10th legs, becoming longer and sharper posteriorly; prefemoral spines arising on segment 5, becoming progressively more acute caudally. Gonopodal aperture broadly ovoid, 4.4 mm long and 1.9 mm wide at midpoint, indented anteriolaterad, sides strongly elevated above metazonal surface. Gonopods in situ (Fig. 95, of paratype) with acropodites lying over and under each other in midline of aperture, not inter- twining, peaks extending forward just beyond anterior margin of aperture. Gonopod structure as follows (Figs. 86-87): Prefemur moderate in size, with relatively long, upright, acuminate prefemoral process arising on anterior side, bent slightly at midlength, with suggestion of tooth basally. Acropodite relatively thick and heavy, well sclerotized, circumscribing nearly complete circle of considerably more than one vertical plane, extending slightly beyond outer level of prefemoral process. Basal zone long, without modifications; anterior bend broad, well de- fined; peak gently curved and rounded, highest at midlength, leaning slightly mediad and MEM. AMER. ENT. SOC., 35 114 XYSTODESMID MILLIPEDS essentially coplanar with basal zone; apical curve relatively broad; distal zone long, extending laterad from peak and not coplanar with basal zone, expanding markedly near midlength then narrowing rapidly to acuminate tip; recurved and slightly bisinuate. Prostatic groove crossing to lateral side at anterior bend, running along inner surface of distal zone to opening at tip. Male paratypes. — The prefemoral process is uncinate on a few paratypes, and the expan- sion of the distal zone is much broader in the males from Tellico Gorge. Female paratype. — Length 42.7 mm, maximum width 10.7 mm, W/L ratio 25.1%, depth/ width ratio 81.3%. Cyphopods in situ with side of receptacle visible in aperture, valves directed dorsad. Receptacle large, cupped over ventral surfaces of valves, surface rugulose. Valves relatively small, subequal, surfaces finely granulate. Variation. — Neither size nor color pattern varies appreciably in aphelorioides. The most notable gonopodal variation involves the midlength expansion on the distal zone, which is broadest in the center of the range (Monroe Co.) and reduced in the eastern and western peripheries. The expansion is barely detectable in the male from North Carolina and resembles the condition in ainsliei. In the paratypes from Tellico Gorge and the male from McMinn County, the tip is broader and not as recurved as in other specimens (reflexed in the latter), and the distal part of the distal zone is “‘scoop-shaped.”’ Ecology. — Sigmoria (F.) aphelorioides inhabits moist rhododendron coves. Distribution. — A small area in the Blue Ridge Province south of the Lit- tle Tennessee and north of the Hiwassee Rivers, lying just north of the area of xerophylla. In Tennessee the species in known only from the Cherokee National Forest, and it occurs in the Tellico Wildlife Management area in eastern Monroe County. The range extends across the North Carolina state line into the western fringe of Swain County, and aphelorioides should be expected in the Joyce Kilmer-Slickrock Wilderness Area of Graham Coun- ty. Specimens were examined as follows: TENNESSEE. Monroe Co., 8.8 mi E Madisonville, along unnumbered rd. off co. rd. 2568, M, 3F, 13 October 1978 (NCSM A2475); 14.3 mi SSE Madisonville, along TN hwy. 165 at Tellico Ranger Station, 3M, 12 October 1978 (NCSM A2466); 17.3 mi SSE Madisonville, along TN hwy. 16, 0.5 mi W Bald River Falls, 4M, 2F, 12 October 1978 (NCSM A2468) TYPE LOCALITY; and 1.5 mi. E Tellico Plains, Tellico Gorge, 2M, F, 12 June 1953. L. Hubricht (RLH). McMinn Co., 11.2 mi. SE Athens, along co. rd. 4371, 0.5 km W ject co. rd. 4276, M, F, 13 October 1978 (NCSM A2479). NORTH CAROLINA: Swain Co., 0.3 mi N Tapoco (in Graham Co.), along trail off US hwy. 129 just inside Graham Co. line, M, 26 June 1974 (NCSM 2460). Remarks. — Though phenotypically similar to ainsliei in having a cir- cular acropodite, aphelorioides differs in that the loop passes through con- siderably more than one vertical plane. In aphelorioides the distal zone curves laterad from the peak, whereas it is not coplanar with the basal zone in ainsliei. These looped or circular configurations are convergent with the > Se eee R.M. SHELLEY & D.R. WHITEHEAD 115 condition in Apheloria, hence aphelorioides’ specific name; they are also convergent with that in trimaculata (Rudiloria). The distal zone is not broader at midlength in ainsliei, but most males of aphelorioides display a distinct expansion, particularly those from the center of the distribution. This expansion is in about the same position on the distal zone as the lobes in trimaculata kleinpeteri (Rudiloria). Thus, in the Great Smoky Mountains and from Virginia to Canada there are forms with circular acropodites that possess and lack an expansion on the distal zone. Their relative statuses, however, differ; those in the Smokies are reproductively isolated, whereas the others connect through intergrades and represent geographic races. In addition to the acropodal differences, the distinct prefemoral processes of ainsliei and aphelorioides also indicate reproductive isolation. In contrast, the projections of ¢. trimaculata and t. kleinpeteri (Rudiloria) are nearly identical. The Translineata Group Four species, two newly described, are added to the translineata group, bringing its composition to seven species, four in the Great Smoky Moun- tains area of Tennessee and North Carolina and three in the Cumberland Plateau of Tennessee. One species, ainsliei, occurs in the former and ex- tends west into the fringe of the adjacent Ridge and Valley Province. The species are united by a divided prefemoral process; translineata, lyrea, ainsliei, and forficata have large, deeply divided structures, but it is shorter, and the division more shallow, in fumimontis. The processes are large and cupped, and the division subapical in houstoni and abbreviata. In translineata, lyrea, and fumimontis the peak is tilted laterad, and the medial flanges are long and narrow, arising on the anterior bend and terminating near the beginning of the apical curve. The first two also display a rounded lobe on the lateral edge of the acropodite near the beginning of the apical curve. The acropodal configurations vary, with translineata and fumimon- tis having flattened peaks and short distal zones, while ains/iei has a circular acropodite that forms a complete loop. Sigmoria (F.) lyrea exhibits an in- termediate configuration but does not occupy an intermediate geographical position. In the Blue Ridge forms the distal and basal zones are coplanar, but the former curves strongly laterad in the Cumberland species and is nearly coplanar with the peak in houstoni. At first glance, ainsliei does not appear congeneric, as its circular acropodite is convergent with those in Apheloria (it was originally assigned to this genus by Chamberlin (1921)), and it also lacks acropodal adornments, most notably a medial flange. MEM. AMER. ENT. SOC., 35 116 XYSTODESMID MILLIPEDS Despite these differences, the divided prefemoral process of ainsliei is clear- ly indicative of shared ancestry with other members of the translineata group. Components. — translineata Shelley; lyrea Shelley; fumimontis Shelley; ainsliei (Chamberlin); forficata, new species; houstoni Chamberlin; ab- breviata, new species. Fics. 88-91. Sigmoria (Falloria) ainsliei. 88, process of 4th sternum of holotype, caudal view. 89, gonopods in situ, ventral view of male from Sevier Co., TN. 90, telopodite of left gonopod of holotype, medial view. 91, the same, lateral view. Scale line for fig. 89 = 1.00 mm; line for other figs. = 1.00 mm for 88 and 91; 1.33 mm for 90. R.M. SHELLEY & D.R. WHITEHEAD 117 Sigmoria (Falloria) ainsliei (Chamberlin), new combination Figs. 88-91 Apheloria ainsliei Chamberlin, 1921:232, Fig. 1. Attems, 1938:168, Fig. 184. Chamberlin and Hoffman, 1958:18. Type specimen. — Male holotype (MCZ) collected by George G. Ainslie on unknown date from unspecified locality in Knox Co., TN. Diagnosis. — A large species of Sigmoria without medial acropodal flange, usually with red paranota and blue transverse stripes along caudal margins of metaterga; gonopods with following diagnostic characters: acropodites interlocking and intertwined; prefemoral process large, divided basally into two unequal components, lateral component longer and directed laterad behind base of acropodite, medial component upright and directed ventrad; acropodite moderately thick and heavy, slightly overhang- ing prefemoral process, configuration circular, anterior bend and apical curve broad but well defined, distal zone coplanar with basal zone, extend- ing into arch and overlapping latter; tip narrowing on both sides to blunt, central termination. Color in Life. — Paranota usually red, occasionally blue, metaterga with blue stripes along caudal margins. Holotype. — Length 52.0 mm, maximum width 12.5 mm, W/L ratio 24.0%, depth/ width ratio 68.0%. Segmental widths as follows: collum 9.0 mm 13th 12.3 2nd 10.4 14th 12.0 3rd 10.9 15th 11.5 4th 11.6 16th 10.6 Sth-7th 12.0 17th 9.3 8th-12th 12.5 18th 6.4 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.9 mm, interantennal isthmus 2.0 mm. Antennae reaching back to middle of 3rd paranota, relative lengths of antennomeres 2>3>4=5>6>1>7. Genae with slight central impressions. Facial setae as follows: epicranial, interantennal, and genal absent, clypeal about 12-12, labral about 16-16. Terga smooth, polished, moderately coriaceous on paranota. Collum broad, ends extending slightly beyond those of following tergite. Paranota moderately depressed, angled ventrad and continuing slope of dorsum, caudolateral corners rounded through segment 5, blunt on 6-13, becoming progressively more acute posteriorly. Peritremata relatively flat, not sharply elevated above paranotal surface. Ozopores located in swellings near middle of paranota, opening dor- solaterad. Process of 4th sternum (Fig. 88) moderate, subequal to widths of adjacent coxae; Sth ster- num with large paramedial knobs between both leg pairs, shorter than widths of adjacent cox- ae; 6th sternum convexly recessed between both leg pairs to accommodate apical curvatures of acropodites. Postgonopodal sterna relatively flat and plate-like, with large, shallow central im- pressions on caudal segments and transverse grooves between leg pairs on all segments. Small coxal tubercles present on legs of segments 9-16; prefemoral spines beginning on legs of seg- ment 6, becoming progressively longer and sharper caudally. MEM. AMER. ENT. SOC., 35 118 XYSTODESMID MILLIPEDS Gonopodal aperture elliptical, 5.4 mm wide and 2.4 mm long at midpoint, indented anteriolaterad, sides raised above metazonal surface. Gonopods in situ (Fig. 89, not this specimen) with acropodites projecting ventrad from aperture, bending sharply mediad and overlapping and intertwining in midline, peaks extending forward beyond anterior edge of aperture and inserting in depression between 7th legs, apices directed ventrad on respective sides of aperture. Gonopod structure as follows (Figs. 90-91): Prefemur large, subglobose, with basally divided prefemoral process arising on anterior side, components of process diverg- ing by more than 90°; lateral component longer than medial, directed laterad and extending behind base of acropodite, tip blunt, obscured in medial view; medial component apically bifurcate and upright, directed ventrad. Acropodite moderately thick and heavy, circumscrib- ing complete circle of slightly more than one vertical plane, extending only to outer level of prefemoral process. Basal zone long, without modifications; anterior bend broad, well de- fined; peak gently curved and rounded, highest at midlength, leaning mediad and overhanging prefemur, not coplanar with basal zone; apical curve relatively broad but well defined; distal zone long, curving broadly into arch and overlapping basal zone, thus forming complete circle with peak and distal extremity of basal zone, essentially coplanar with peak, bent slightly at midlength and again just proximal to tip; latter blunt, directed subventrad, only slightly discontinuous with distal zone. Medial and lateral flanges absent. Prostatic groove crossing to lateral side of acropodite at anterior bend and continuing to opening in center of tip. Description of Female. — Based on specimen from 16.3 km E Maryville, Blount County, TN (NCSM A2460). Length 55.5 mm, maximum width 12.4 mm, W/L ratio 22.3%, depth/ width ratio 78.2%. Cyphopods in situ with edge of receptacle visible in aperture, valves directed dorsolaterad. Receptacle large, globose, located entirely on medial sides of valves, surfaces rugulose. Valves moderate and subequal in size, diverging dorsad, surfaces finely granulate. Variation. — Sigmoria (F.) ainsliei varies considerably in size, with some individuals dramatically larger than others. Dimensions of the male and a female from 10.2 miles east of Maryville (NCSM A2460) were 56.1 mm in length and 12.8 mm in width, and 59.8 mm in length and 13.3 mm in width, respectively. On the gonopods the prefemoral process is highly variable. Males from Knox County possess a simple, two-pronged structure, as described for the holotype, but most individuals from Blount and Sevier counties have one or two additional spurs on the ventral surface arising near the juncture of the two components. The configuration of the acropodite is reasonably constant, but the distal zone is slightly bisinuate in a few males and widens distally (proximal to the bend at the tip) in ones from the southern part of the range. Ecology. — In the Blue Ridge Province in and near the GSMNP., ainsliei inhabits the moist rhododendron coves typical of Sigmoria s. lat. However, these environments are rare in the more western Ridge and Valley Province, particularly in the valley between the Smokies and Knoxville. Here I have found ainsliei in several predominantly hardwood habitats under leaves near water, but sample A2460 was taken under thick layers of leaves on a steep bank. R.M. SHELLEY & D.R. WHITEHEAD 119 Distribution. — A small area in the adjacent fringes of the Ridge and Valley and Blue Ridge Provinces, ranging from just north of the French Broad River on the University of Tennessee campus in Knoxville to Ten- nessee Highway 73 (Little River Road) in the GSMNP. The species does not penetrate far into the GSMNP and has not been encountered east of Highway 73, for example along the road to Cades Cove. However, ainsliei is common in the Park near the Sinks, Metcalf Bottoms Picnic Area, and Little Greenbrier School, as well as in adjacent areas outside the park near Townsend and Wear Valley. It should be expected in western Sevier County and in the corner of Loudon County near Friendsville, although these areas contain so much cleared land that suitable milliped habitat is rare. Specimens were examined as follows: TENNESSEE: Knox Co., locality and date unspecified, M, G.G. Ainslie (MCZ) TYPE SPECIMEN; Knoxville, Cherokee Bluff, 2M, 17 May 1951, L. Hubricht (RLH) and University of Tennessee campus, M, 18 April 1972, W. Tolbert (RLH). Blount Co., 10.2 mi E Maryville, along unnumbered rd., 1.0 km N jet. co. rd. 2427, M, 2F, 11 October 1978 (NCSM A2460); 6.4 me SE Maryville, along unnumbered rd., 3.2 km N ject. Foothills Parkway, F, 11, October 1978 (NCSM A2462); 3.4 mi N Townsend along co. rd. 2422, M, 9 April 1981 (NCSM A3659); and GSMNP, along TN hwy. 73 at crossing of Little R., M, F, 17 May 1978 (NCSM A1932). Sevier Co., 1.4 mi. W Gatlinburg, along Norton Creek Rd. off US Hwy. 441, 3M, F, 8 May 1980 (NCSM A3015); Little Greenbrier School, GSMNP, 5M, 6F, 17 May 1978 (NCSM A1935); and GSMNP, along TN hwy 73, 0.6 mi N Metcalf Bottoms Picnic Area, 2M, F, 9 August 1981, R.M. Shelley and H. Enghoff (NCSM A3723). Remarks. — Sigmoria (F.) ainsliei lacks a medial flange, but a new medial edge arising at the anterior bend is suggestive of a flange. This is ac- tually caused by torsion. The medial surface of the basal zone shifts to the lateral side at the anterior bend, and a new medial edge arises at this point, extends the length of the peak, and terminates at the apical curve, whereupon a third medial edge arises. The acropodite is not broader through the peak, at it would be with a flange, and tapers into the distal zone. Sigmoria (Falloria) forficata Shelley, new species Figs. 92-96 Type specimens. — Male holotype (NCSM A2630) and two male and two female paratypes collected by R.M. Shelley and R.K. Tardell, 7 May 1979, 4.2 mi. SE Crab Orchard, along Fall Creek near Ozone Twp., Cumberland Co., TN. Male and female paratypes deposited in FSCA. Diagnosis. — A large species of Sigmoria with medial flange extending between proximal parts of peak and distal zone and with red paranota and blue metatergal stripes; gonopods with following diagnostic characters: prefemoral process massive, usually divided basally into two long com- MEM. AMER. ENT. SOC., 35 120 XYSTODESMID MILLIPEDS ponents, medial one shorter and usually more linear, lateral one longer and curved broadly anteriad; acropodite relatively thin, arch broadly curved Fics. 92-96. Sigmoria (Falloria) forficata. 92, process of 4th sternum of holotype, caudal view. 93, gonopods in situ, ventral view of paratype. 94, telopodite of left gonopod of holotype, medial view. 95, the same, lateral view. 96, telopodite of left gonopod of male from Scott Co., TN, medial view. Scale line for fig. 93 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 121 and extending over prefemoral process; basal zone gently curved, with or . without a basomedial projection; anterior bend and apical curve poorly defined; distal zone moderately long, directed laterad from peak and not coplanar with other sections, curving into arch distad and tapering to acuminate tip; medial flange variable but with broadly triangular lobe on distal extremity of peak; lateral flange inconspicuous, a small lobe on distal zone opposite distal extremity of medial flange. Color in Life. — Paranota red; metaterga black with wide blue stripes along caudal edges connecting paranotal markings; collum with blue stripes along both anterior and posterior margin. Holotype. — Length 48.8 mm, maximum width 11.3 mm, W/L ratio 23.2%, depth/ width ratio 58.4%. Segmental widths as follows: collum 6.8 mm 14th 10.8 2nd 8.5 15th 9.8 3rd 9.9 16th 9.5 4th 10.3 17th 8.4 Sth-6th 10.8 18th 6.3 7th-13th 11.3 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.9 mm, interantennal isthmus 1.7 mm. Antennae relatively long, reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>6>4=5>1>7. Genae with faint central impressions. Facial setae as follows: epicranial and interantennal absent, frontal 1-1, genal 2-2, clypeal about 12-12, labral about 16-16. Dorsum smooth, polished, and moderately coriaceous, especially on anterior halves of paranota. Collum broad, ends extending well below those of following tergite. Paranota strongly depressed, angling sharply downward and exceeding slope of dorsum, caudolateral corners rounded through segment 7, blunt on 8-13, and becoming progressively more acute posteriorly. Peritremata relatively flat and indistinguishable, ozopores located near middle of peritremata, opening dorsolaterad. Sternum of segment 4 with long, apically divided process, longer than widths of adjacent coxae (Fig. 92); that of segment 5 with long apically divided process between anterior legs and two medially coalesced knobs between Sth legs, projections nearly equal in length to each other and to widths of adjacent coxae; 6th sternum deeply and convexly recessed between 7th legs to accommodate apical curvatures of acropodites. Postgonopodal sterna as follows: those of segments 8-10 with flattened elevated areas and blunt caudally directed lobes subtending anterior and posterior coxae, respectively, these strongest on segment 8 and progressively less pronounced on 9-10; remaining sterna without elevations between anterior legs and with pro- gressively smaller ones between posterior legs, becoming progressively more plate-like posteriorly with variably broad central impressions. Coxae with blunt tubercles beginning on caudal legs of segment 9, sharply acute on 11-14, diminishing thereafter; prefemoral spines on segment 5, becoming progressively longer and more acute posteriorly. Gonopodal aperture ovoid, 4.6 mm wide and 2.6 mm long at midpoint, strongly indented anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 93, of paratype) with acropodites angling mediad and overlapping in midline of aperture, then curving laterad and extending slightly beyond anterior margin; prefemoral processes with components crisscrossing. Gonopod structure as follows (Figs. 94-95): Prefemoral process massive, divided basally into two components, a short, straight medial one, and a longer, broadly curved MEM. AMER. ENT. SOC., 35 122 XYSTODESMID MILLIPEDS anterior one, both directed toward basal zone and tapering smoothly and continuously to acuminate tips. Acropodite rather thin, forming broadly curved arch, extending nearly to level of distal extremity of prefemoral process; basal zone gently curved, continuous with peak through anterior bend; latter broad, poorly defined; peak gently curved, apex near midlength; apical curve broad, poorly defined; distal zone moderately long, curving laterad from peak and not coplanar with basal zone or peak, curving inward distally and tapering to acuminate tip; latter directed toward basal zone. Medial flange large and conspicuous, broadly triangular, arising on proximal part of peak, terminating on proximal part of distal zone. Lateral flange small and inconspicuous, a broadly rounded lobe opposite distal half of medial flange. Pro- static groove crossing to lateral side on peak and continuing to terminal opening. Male Paratypes. — The male paratypes agree with the holotype in all particulars. Female Paratype. — Length 47.0 mm, maximum width 11.3 mm, W/L ratio 24.0%, depth/ width ratio 69.9%. Cyphopods in situ with corners of receptacles partly visible in aper- ture, partially obscured by large, convoluted cyphopodal membranes, valves directed caudad. Receptable moderately large, completely enveloping valves, with lobes and ridges, surface rugulose. Valves small, subequal, surfaces finely granulate. Variation. — The sternal projections are similar to those of the holotype in all males except those from Hamilton County, where they are smaller. In this individual the process on segment 4 is equal in length to the widths of the adjacent coxae; the anterior projection on segment 5 is absent; and there are broad, separate, elevated areas between the posterior legs that are much shorter than the adjacent coxal widths. The postgonopodal sternal lobes on segments 8-10 are also smaller. On the gonopods, the males from Cumberland County agree closely with the holotype, although that from 8.5 miles E of Crab Orchard (NCSM A2766) has a small triangular projection from the medial edge of the basal zone near midlength. The male from Scott County, the northernmost locali- ty, has a more rounded medial flange that arises proximally on the basal zone and has a distinct spine from the proximomedial edge (Fig. 96). The latter closely resembles the condition in arcuata (Cleptoria) in piedmont South Carolina. The Morgan County male has a broad prefemoral process that is cupped on the ventral surface and not as deeply divided, so the medial projection appears more as a basal spur than a separate structure. The medial flange in the Morgan County male is less expanded and ter- minates more abruptly on the distal zone. South of Cumberland County, the Bledsoe County male is nearly identical to the holotype, but in those from Hamilton County, the medial component of the prefemoral process is longer and curved toward the other, which in turn is narrower and more sinuous than in the holotype. Ecology. — Sigmoria (F.) forficata is a cove inhabiting species. Distribution. — The Cumberland Plateau of Tennessee. Sigmoria (F.) forficata occurs sporadically across the width of Tennessee from near the Kentucky to near the Georgia/ Alabama borders. It may eventually be dis- en R.M. SHELLEY & D.R. WHITEHEAD 123 covered in McCreary County, Kentucky, but the Tennessee River marks its © southern range limit, as extensive searches south of the waterway in Marion and Hamilton counties, and adjacent parts of Georgia and Alabama, have been unsuccessful. Specimens were examined as follows: TENNESSEE. — Scott Co., 7.0 mi. S. Huntsville, along co. rd. 2342 near Brimstone Cr., M, F, 8 June 1979 (NCSM A2724). Morgan Co., 4.3 mi. SW Wartburg, along Emory R. on edge of Catoosa Wildlife Man. Area, M, 2F, 7 June 1979 (NCSM A2718). Cumberland Co., 8.5 mi. N Crossville, along Fox Cr. in Catoosa Wildlife Man. Area, M, 2F, 15 June 1979 (NCSM A2765); 4.2 mi. SE Crab Orchard, along Falls Cr. near Ozone, 3M, 2F, 7 May 1979 (NCSM A2630) TYPE LOCALITY; and 8.5 mi. E Crab Orchard, along co. rd. 4382, 2.2 mi. W jct. co. rd. 2590, M, 2F, 15 June 1979 (NCSM A2766). Bledsoe Co., 2.8 mi. SE Pikeville, along TN hwy. 30, M, F, 25 May 1983 (NCSM A4137). Hamilton Co., Signal Mountain, M, F, 18 August 1956, R.L. Hoffman (RLH); and Rainbow Lake Wilderness Area nr. Signal Moun- tain, M, 22 May 1983 (NCSM A4113). Remarks. — Some traits of forficata are convergent with ones found in congeners in piedmont South Carolina, many miles east across the Ap- palachians. The large, convoluted cyphopodal membrane is similar to that in catawba, saluda, and simplex (Croatania); and the long process on the 4th sternum closely resembles those in these species plus shelfordi and ar- cuata (Cleptoria). The distinct basal spine on the medial flange in the Scott County male also resembles those in these species as does the slightly distal triangular lobe in a Cumberland County male. Lastly, the prefemoral proc- ess is massive in forficata, catawba, and saluda, and the curvature of the lateral component is similar to that of the entire structure in catawba and saluda (compare Figs. 30 and 32 with Figs. 1, 3, 7, and 8 in Shelley 1977). Sigmoria (Falloria) abbreviata Shelley, new species Figs. 97-100 Type specimens. — Male holotype (A4136) and one male and three female paratypes collected by R.M. Shelley, 24 May 1983, from Fall Creek Falls State Park, Van Buren Co., TN. Additional paratypes collected in same locality as follows: two females by same collector, 17 June 1976; and one male and four females by R.M. Shelley and R.K. Tardell, 13 May 1979. Male and female paratypes deposited in FSCA. Diagnosis. — A moderate-size species of Sigmoria with medial flange on distal extremity of peak and with red paranota and blue metatergal stripes; gonopods with following diagnostic characters: prefemoral process large, ventral surface convex distad, divided at midlength into two subequal, diverging components; acropodite thin and fragile, arc broadly curved and extending over but not beyond level of prefemoral process; basal zone MEM. AMER. ENT. SOC., 35 124 XYSTODESMID MILLIPEDS broadly curved; anterior bend and apical curve poorly defined; peak and distal zone relatively short; latter curving sublaterad from former, not coplanar with other sections, bent into arch distad and tapering to acuminate tip; medial flange a small, rounded lobe; lateral flange narrow 99 100 | Fics. 97-100. Sigmoria (Falloria) abbreviata. 97, process of 4th sternum of holotype, caudal view. 98, gonopods in situ, ventral view of paratype. 99, telopodite of left gonopod of holotype, medial view. 100, the same, lateral view. Scale line for fig. 98 = 1.00 mm; line for other figs. = 1.00 mm for 99-100; 1.33 mm for 97. R.M. SHELLEY & D.R. WHITEHEAD 125 and inconspicuous, represented by a slightly wider acropodite stem opposite medial flange. Color in Life. — Paranota red; metaterga black with wide blue stripes along caudal margins connecting paranotal spots; collum with blue stripes along both anterior and posterior edges. Holotype. — Length 37.0 mm, maximum width 9.1 mm, W/L ratio 24.6%, depth/width ratio 60.4%. Segmental widths as follows: collum 6.9 mm 15th 8.9 2nd 7.2 16th 8.7 3rd 7.9 17th 7.5 4th 8.8 18th 5.4 5th-14th 9.1 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.4 mm, interantennal isthmus 1.4 mm. Antennae reaching back to middle of 3rd paranota, relative lengths of antennomeres 2>3>5>6=4>1>7. Genae with distinct central impressions. Facial setae as follows: epicranial, interantennal, and genal ab- sent, frontal 1-1, clypeal about 16-16, labral 22-22. Dorsum smooth, polished, moderately coriaceous. Collum broad, ends extending slightly beyond those of following tergite. Paranota moderately depressed, continuing slope of dor- sum, caudolateral corners rounded through segment 6, blunt on 7-13, becoming progressively more acute caudally. Peritremata thick and conspicuous, strongly elevated above paranotal surface, ozopores located caudal to midlength, opening dorsolaterad. Process of 4th sternum moderately long, equal to widths of adjacent coxae (Fig. 97); Sth sternum with two paramedian knobs between anterior legs, shorter than widths of adjacent coxae, and broad, flattened, elevated areas between posterior legs; 6th sternum convexly recessed between 7th legs to accommodate apical curvatures of acropodites, 7th legs set slightly farther apart than 6th. Postgonopodal sterna with bicruciform impressions on 8-9, becoming progressively flatter and more plate-like posteriorly with variably broad, shallow, central im- pressions. Coxae with blunt tubercles beginning on segment 10, becoming progressively more acute posteriorly; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture elliptical, 3.4 mm wide and 1.7 mm long at midpoint, indented anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 98, of paratype) with acropodites angling toward midline, either touching or overlapping near midlengths, then curving anteriolaterad and extending beyond anterior margin. Gonopod structure as follows (Figs. 99-100): Prefemoral process large, curving slightly ventrad at midlength and convex distally on ventral surface, divided into two diverging, subequal, acuminate components, directed toward midlength of acropodite. Acropodite moderately thin, forming broadly curved arc, overhanging prefemoral process; basal zone broadly curved, continuous with peak through anterior bend; latter broad, poorly defined; peak moderately long; apical curve broad, poorly defined; distal zone relatively short, curving sublaterad from peak and not coplanar with other sections, bending inward into arch distad and tapering to blunt tip. Medial flange small, margin smoothly rounded, located on distal extremity of peak. Lateral flange barely detectable, a slightly wider spot in stem of acropodite opposite medial flange. Prostatic groove crossing to lateral side on basal zone, continuing to terminal opening. Male Paratypes. — The only variation in the male paratypes involves the shape and relative lengths of the projections of the prefemoral process. Female Paratype. — Length 39.6 mm, maximum width 9.9 mm, W/L ratio 25.0% depth/width ratio 59.6%. Cyphopods in situ with corners of receptacles protruding through MEM. AMER. ENT. SOC., 35 126 XYSTODESMID MILLIPEDS folded, convoluted membranes in aperture, valves directed caudomediad. Receptacle large, located along anterior sides of valves, with ridges and lobes, surface rugulose. Valves moderate, subequal, surfaces finely granulate. Variation. — The Bledsoe County male agrees with the holotype. Ecology. — Sigmoria (F.) abbreviata is a cove inhabiting species. Distribution. — Known only from the central Cumberland Plateau. Specimens were examined as follows: TENNESSEE. — Van Buren Co: Fall Creek Falls St. Pk., 2F, 17 June 1976 (NCSM A888), M, 4F, 13 May 1979 (NCSM A2693), and 2M, 3F, 24 May 1983 (NCSM A4136) TYPE LOCALITY. Bledsoe Co., 22 mi. E Spencer, along TN hwy 30, 0.5 mi. E jct. TN hwy. 101, M, 2F, 14 May 1979 (NCSM A2697). Remarks. — IJ have collected abbreviata two places in Fall Creek Falls State Park. The sample with the holotype came from behind the park of- fice; the others were taken along the trails behind the nature center. Although it has a bifurcate prefemoral process, abbreviata shares more features with houstoni than with forficata. The bifurcation is more of an elaboration on the squared/cupped condition found in houstoni, as the cor- ners are elongated and the margin between them depressed, than it is true bifurcation, in which separate projections arise from a common base. The affinity between abbreviata and houstoni is also seen in the acropodites, as the medial flanges are in the same general locations, and the distal zones have the same general curvatures. The flange and distal zone are shorter (abbreviated) in abbreviata, hence the specific name. These is no indication of a flange on the basal zone in abbreviata as in some variants of forficata, and abbreviata also lacks the sternal modifications of the latter. Perhaps abbreviata and houstoni were once joined by intergrade forms that have disappeared leaving two isolated populations. Both warrant specific recognition because of these differences. Sigmoria (Falloria) houstoni Chamberlin Figs. 101-104 Sigmoria houstoni Chamberlin, 1943:144, Fig. 1. Hoffman, 1950:5. Chamberlin and Hoff- man, 1958:50. Type specimens. — Male holotype and female allotype (RVC); locality, date, and collector unknown. The vial label and the three prior references give Houston, Harris County, Texas, as the type locality, but this is clearly the result of a sampling mixup. As shown in Fig. 151, no member of the tribe Apheloriini occurs within 250 miles of Houston. The type locality should be somewhere in Franklin, Grundy, or Marion counties, Tennessee, the only places where houstoni has been authentically taken. The date of R.M. SHELLEY & D.R. WHITEHEAD 127 collection and the collector reported by these authors may also be wrong. Because of this confusion, the following description is of a male collected from Marion County, along TN highway 108, 5.8 mi S of the Grundy County line. Diagnosis. — A moderate-size species of Sigmoria with medial flange ex- tending between midlengths of basal and distal zones and with red paranota and blue metatergal stripes; gonopods with following diagnostic characters: prefemoral process moderately long, ventral surface convex distad, apical margin with corners produced, with or without additional spurs; acropodite 103 Fics. 101-104. Sigmoria (Falloria) houstoni. 101, process of 4th sternum of male from Marion Co., TN, caudal view. 102, gonopods in situ, ventral view of the same. 103, telopo- dite of left gonopod of the same, submedial view. 104, the same, lateral view. Scale line for fig. 102 = 1.00 mm; line for other figs. = 1.00 for 101 and 104; 1.14 mm for 103. MEM. AMER. ENT. SOC., 35 128 XYSTODESMID MILLIPEDS relatively thin, curving broadly over but not beyond distal extremity of prefemoral process; basal zone gently curved; anterior bend poorly defined; peak short, continuing overall curvature of acropodite; apical curve well defined; distal zone relatively long, directed laterad and coplanar with peak, tapering distally to acuminate tip; medial flange long and narrow, expanded into rounded terminal lobe on proximal part of distal zone; lateral flange absent. Color in Life. — Paranota red; metaterga black with wide blue stripes along caudal edges connecting paranotal markings; collum with blue stripes along both margins. Description. — Length 37.1 mm, maximum width 8.9 mm, W/L ratio 24.0%, depth/ width ratio 48.3%. Segmental widths as follows: collum 5.8 mm 14th 8.4 2nd 6.9 15th 8.1 3rd-4th 7.9 16th 7.5 Sth-7th 8.7 17th 6.5 8th-13th 8.9 18th 5.0 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.1 mm, interantennal isthmus 1.3 mm. Antennae relatively short, extending only to caudal edge of 2nd tergite, relative lengths of antennomeres 2>3>4>6>5>1>7. Genae with faint central impressions. Facial setae as follows: epicranial, interantennal, frontal, and genal absent, clypeal about 10-10, labral 12-12. Dorsum smooth, moderately coriaceous in midline. Collum broad, ends extending slightly beyond those of following tergite. Paranota moderately depressed, continuing slope of dor- sum, caudolateral corners rounded through segment 4, blunt in midbody region, becoming progressively more acute caudally. Peritremata distinct, strongly elevated above paranotal sur- face; ozopores located caudal to midlength, opening dorsolaterad. Sternum of segment 4 with small lobe between 3rd legs, much shorter than widths of adja- cent coxae (Fig. 101); that of segment 5 with low knobs between 4th legs, much shorter than widths of adjacent coxae, and low flattened areas between Sth legs; sternum of segment 6 con- vexly recessed between 7th legs to accommodate apical curvatures of acropodites. Postgonopodal sterna smooth and flat, bicruciately impressed on segments 8-10, with variably broad, shallow, central impressions on remaining segments. Coxae with blunt tubercles begin- ning on segment 5, sharp and distinct on all legs. Gonopodal aperture elliptical, 2.9 mm wide and 1.6 mm long at midpoint, indented slightly anteriolaterad, sides raised above metazonal surface. Gonopods in situ (Fig. 102) with acropodites angling mediad and overlapping in midline of aperture, curving laterad, over- lapping again, and projecting slightly beyond anterior margin. Gonopod structure as follows (Figs. 103-104): Prefemoral process moderately large, rectangular, narrowest basally, convex ventrally, distal corners curved ventrad and directed toward basal zone. Acropodite thin and laminate, forming broadly curved, open arc, extending over but not beyond level of distal ex- tremity of prefemoral process; basal zone gently curved, continuous through anterior bend with peak; anterior bend broad, poorly defined; peak continuing general curvature of basal zone and anterior bend; apical curve short, well defined; distal zone moderately long, curving directly laterad and coplanar with peak, tapering smoothly and continuously to acuminate tip, latter slightly bisinuous. Medial flange long, arising proximally on basal zone, continuing through peak as linear lamina, expanding into rounded lobe on proximal part of distal zone, terminating abruptly at distal extremity of lobe near midlength of distal zone. Lateral flange SS. R.M. SHELLEY & D.R. WHITEHEAD 129 absent. Prostatic groove crossing from medial to lateral sides around proximal part of peak, continuing to terminal opening. Description of Female from Grundy Co. — Length 42.8 mm, maximum width 10.6 mm, W/L ratio 24.8%, depth/width ratio 69.8%. Cyphopods in situ with corners of receptacles protruding through folded, convoluted membranes in aperture, valves directed caudomediad. Receptacle large, located along anterior sides of valves, with ridges and lobes, surface rugulose. Valves moderate, equal, surfaces finely granulate. Variation. — The gonopods of houstoni are rather uniform. Some prefemoral processes are more elaborate than others and have short ter- minal and subterminal spurs, but this is random variation with no geographic component. Similarly, the medial flange widens slightly basally then narrows again at the approximate location of the anterior bend in some males. Ecology. — The specimens of houstoni I collected came from dry, open, hardwood litter well removed from water. It therefore is not a cove species and has different ecological preferences from forficata and the Appalachian congeners. Distribution. — Cumberland Plateau of Grundy, Marion, and Franklin counties, Tennessee. It should also be expected in the western fringe of Se- quatchie County. Specimens were examined as follows: TENNESSEE. — Grundy Co., 4 mi. N Palmer, along co. rd. 4350, 23 May 1983 (NCSM A4122); 2.5 mi. S Palmer, along TN hwy. 108, 0.3 mi. N Marion co. line, M, 2F, 12 May 1979 (NCSM A2684); and Grundy St. For. near Tracy City, 2M, 2F, 26 April 1979 (A2606). Marion Co., 11.1 mi. N Whitwell, along TN hwy. 108, 1.8 mi. S Grundy co. line, M, 26, April 1979, A.L. Braswell, R.E. Ashton, and R. Franz (NCSM A2589); 5.8 mi. S Grundy co. line, along TN hwy. 108, 3M, 3F, 12 May 1979 (NCSM A2685) and 3.5 mi. N Whitwell, along TN hwy. 108, 0.9 mi. S. jct. co. rd. 4303, 2M, 23 May 1983 (NCSM A4121). Franklin Co., Sewanee, 2M, date unknown, S. Lazell (FSCA). Remarks. — Because the distal zone curves directly laterad and thus is coplanar with the peak, the acropodite was tilted slightly ventrad in figure 103 to show this section. The Picapa Group A single species group is proposed for picapa, which differs from the other Cumberland species in the large, bisinuate, undivided prefemoral process. The laterally directed distal zone is not reflective of a separate genus as Keeton (1960) thought when he described it. The species is not a cove inhabitant and is known only from a small area in northern Morgan County, Tennessee. Component. — picapa (Keeton). MEM. AMER. ENT. SOC., 35 130 XYSTODESMID MILLIPEDS Sigmoria (Falloria) picapa (Keeton), new combination Figs. 105-108 Hubroria picapa Keeton, 1960:2-4, Figs. 1-4. Type specimens. — One male and two female paratypes (FMNH) col- lected by B. Benesh, June 1949, at Sunbright, Morgan Co., TN. Male 105 107 | | 108 | Fics. 105-108. Sigmoria (Falloria) picapa. 105, process of 4th sternum of paratype, caudal view. 106, gonopods in situ, ventral view of topotype. 107, left gonopod of paratype, medial view. 108, telopodite of the same, lateral view. Scale line for fig. 106 = 1.00 mm; line for other figs. = 1.00 mm for 108; 1.33 mm for 105; 1.78 mm for 107. R.M. SHELLEY & D.R. WHITEHEAD 131 paratype (RLH) taken by same collector, 2 June 1952, at Burrville, Morgan. County. One male and one female topotypes (NCSM A3363) collected by R.M. Shelley and M.S. Morgan, 22 July 1980, from Morgan Co., 2.3 mi. NNE Sunbright, along US hwy. 27, 0.5 mi. N jet. co. rd. 2438. The male holotype apparently is missing from the NMNH, where Keeton (1960) reported its deposition. Two loan requests to this institution failed to pro- duce it, and its location is unknown. Diagnosis. — A large species of Sigmoria with medial flange running be- tween distal extremities of basal zone and peak and with red paranota and blue metatergal stripes; gonopods with following diagnostic characters: prefemoral process long, widest basally, tapering distad, bisinuately curved; acropodite moderately thick and heavy, arch an inverted L, extending to level of prefemoral process; basal zone long, gently curved; anterior bend well defined; peak flattened, angling slightly downward, narrowing distad; apical curve poorly defined; distal zone short, directed laterad and coplanar with peak, obscured by latter in medial view, narrowing distad and expand- ing into blunt tip; medial flange broadest distad; lateral flange located on peak opposite medial flange. Color in Life. — Paranota red; metaterga black with wide blue stripes along caudal margins connecting paranotal markings; collum with red and blue stripes along anterior and posterior edges, respectively. Male Paratype from Sunbright. — Length 42.9 mm, maximum width 10.4 mm, W/L ratio 24.2%, depth/width ratio 60.6%. Segmental widths as follows: collum 7.5mm 14th 10.2 2nd 8.7 15th 9.7 3rd 9.6 16th 8.9 4th-7th 10.0 17th 7.8 8th-13th 10.4 18th 6.0 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 5.1 mm, interantennal isthmus 1.9 mm. Antennae reaching back to caudal edge of 3rd paranota; relative lengths of antennomeres 2> 3>6>5>4> 1>7. Genae with flat central impressions. Facial setae as follows: epicranial, interantennal, frontal, and genal absent, clypeal about 12-12, labral about 16-16. Dorsum smooth, polished, moderately coriaceous on paranota. Collum broad, ends extend- ing slightly beyond those of adjacent tergite. Paranota moderately depressed, continuing slope of dorsum, caudolateral corners rounded through segment 6, blunt on 7-11, becoming pro- gressively more acute caudally. Peritremata thick and distinct, sharply elevated above paranota, ozopores located near middle of peritremata, opening dorsolaterad. Process of 4th sternum apically divided, shorter than widths of adjacent coxae (Fig. 105); knobs of 5th sternum coalesced medially into ventrally directed process, shorter than widths of adjacent coxae, flattened areas between Sth legs distinct and subconical; sternum of segment 6 convexly depressed between 7th legs to accommodate apical curvatures of acropodites, 7th legs set slightly farther apart than 6th. Postgonopodal sterna flattened and plate-like, with bicruciform impressions on segments 8-10 and variably wide, shallow, central depressions on MEM. AMER. ENT. SOC., 35 132 XYSTODESMID MILLIPEDS remaining segments. Coxae with blunt tubercles on segments 9-15; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture ovoid, 3.3 mm wide and 1.8 mm long at midpoint, strongly indented anteriomediad, sides elevated above metazonal surface. Gonopods in situ (Fig. 106, of topotype) with acropodites angling anteriomediad and overlapping each other in midline of aperture, extending forward beyond anterior margin and inserting in depression between 7th legs. Gonopod structure as follows (Figs. 107-108): Prefemoral process long, wide, and bisinuately curved, about half as long as basal zone, tapering smoothly and continuously to subacuminate tip, latter directed toward distal zone. Acropodite moderately thick and heavy, blade like, becoming progressively thinner and narrower beginning at anterior bend, con- figuration an inverted L; basal zone long, gently curved anteriad; anterior bend sharp, well defined; peak flattened, angling slightly downward, distal half narrowing strongly; apical curve broad, poorly defined; distal zone short, curving directly laterad, coplanar with peak and obscured in medial view, narrowing slightly distad and expanding apically into blunt tip; latter bent slightly downward from distal zone, with produced inner corner resembling inverted reflexed condition. Medial flange arising on distal extremity of basal zone, extending across anterior bend and terminating on distal extremity of peak, broadest distad. Lateral flange aris- ing on peak and terminating at same position as medial flange. Prostatic groove running along inner surface of basal zone, crossing to lateral side at anterior bend and continuing to opening on inner corner of tip. Female Paratype. — Length 46.7 mm, maximum width 10.8 m, W/L ratio 23.2%, depth/ width ratio 67.6%. Agreeing closely with males in somatic features, except ends of collum not produced beyond those of following tergite. Cyphopods in situ with openings of valves visible in aperture. Receptacle small, flat, located beneath medial corners of valves, surface finely granular. Valves large, equal, with broad central depression, surfaces finely granulate. Variation. — No differences are apparent between the topotype and the described paratype. In the male from Burrville the prefemoral process is a little narrower and the basal zone straighter, but otherwise it agrees with that from Sunbright. Ecology. — Keeton (1960) gave no indication of habitat in his descrip- tion, but the topotypes I collected were found under deep piles of oak leaves in shallow ravines. There are no cove habitats or rhododendron at this site or at Burrville, and investigations in such environments near Wartburg, Oakdale, and Frozen Head State Park yielded an abundance of Brachoria but none of picapa. Thus, picapa is not a cove species and has different ecological preferences. Distribution. — Known only from the type and paratype localities near Sunbright and Burrville in the Cumberland Plateau of northern Morgan County, Tennessee. Remarks. — Although the paranota of males continue the slope of the dorsum and can be described as moderately depressed, they angle more sharply ventrad than in most congeners. Males of picapa thus appear thicker and more highly arched or vaulted than most Sigmoria males. The medial view of the gonopod (Fig. 107) was prepared from more anterior and dorsal angles than usual to show the tip, which is coplanar with 2. 3>4=6=5>1>7. Genae with distinct central impressions. Facial setae as follows: epicranial, interantennal, and frontal absent, clypeal about 10-10, labral about 12-12, merging with clypeal series and continuing for short distances along genal margins. Terga smooth, polished, with only faint wrinkling. Collum broad, ends extending slightly beyond those of following tergite. Paranota moderately depressed, continuing slope of dor- sum, caudolateral corners rounded on collum but moderately acute thereafter. Peritremata low on anterior segments, becoming sharp and distinct in midbody region. Ozopores located in caudal half of peritremata, opening dorsad. Fic. 109-114. Sigmoria (Falloria) mimetica. 109, process of 4th sternum of holotype, caudal view. 110, gonopods in situ, ventral view of topotype. 111, telopodite of left gonopod of holotype, medial view. 112, the same, lateral view. 113, telopodite of left gonopod of male from 11 mi. E. Lebanon, Wilson Co., TN, medial view. 114, the same, lateral view. Scale line for fig. 110 = 1.00 mm; line for other figs. = 1.00 mm for 113-114; 1.33 mm for 109; 1.78 mm for 111-112. MEM. AMER. ENT. SOC., 35 136 XYSTODESMID MILLIPEDS Process of 4th sternum shorter than widths of adjacent coxae and divided apically (Fig. 109); of segment 5, produced into pair of small, paramedial knobs between 4th pair of legs and lower, more rounded lobes between Sth legs; of segment 6, convexly recessed between both leg pairs to accommodate gonopodal acropodites. Postgonopodal sterna relatively flat, with shallow transverse grooves originating between leg pairs. Coxae without ventrodistal spines or tubercles on all legs; prefemoral spines beginning on segment 6, becoming longer and sharper caudally. Gonopodal aperture rounded, 3.1 mm wide and 2.1 mm long at midpoint, without indenta- tions on anteriolateral margins, sides flush with metazonal surface, caudal margin slightly thickened. Gonopods in situ (Fig. 110, of topotype), with acropodites curving anteriomediad over aperture, overlapping and bending abruptly dorsad apically, extending beyond anterior margin of aperture to between 7th legs. Gonopod structure as follows (Figs. 111-112): Prefemur with short prefemoral process directed toward distal zone of acropodite, sides parallel, tapering apically to central tip. Acropodite moderately thick and heavy, arch high and rounded, overhanging and extending well beyond level of prefemoral process; basal zone relatively long and broad, situated with inner surface directed mediad, with small, rounded distomedial lobe; anterior bend broad, poorly defined; peak relatively short, high and rounded, tilted laterad exposing undersurface in medial view; apical curve broad, poorly de- fined; distal zone long, coplanar with basal zone, curving broadly into arch, extending nearly to base of acropodite just above level of prefemoral process and directed toward basal zone, twisted slightly near midlength producing additional surface on medial side; tip subacuminate, located on inner corner of distal zone, suggesting reflexed condition. Medial flange narrow and inconspicuous, located on peak but possibly incorporating lobe on basal zone. Lateral flange absent. Prostatic groove running along inner surface of acropodite, entirely visible in medial view until reaching distal zone. Male Topotype. — Length 36.2 mm, maximum width 9.6 mm, W/L ratio 26.5%, depth/ width ratio 46.9%. Segmental widths as follows: collum 7.1 mm 13th 9.2 2nd 8.0 14th 9.0 3rd 8.9 15th 8.5 4th 9.2 16th 7.0 5th-7th 9.4 17th 6.6 8th-12th 9.6 18th 4.9 The topotypes agree closely with the holotype in somatic and gonopodal features. The only noticeable difference is that the prefemoral process is slightly longer. Female Topotype. — Length 36.0 mm, maximum width 9.3 mm, W/L 25.8%, depth/ width 65.6%. Agreeing closely with holotype in somatic features, except ends of collum not produced beyond those of following segment. Cyphopods in situ with corners of receptacles visible in aperture, valves directed anteriolaterad. Receptacle large, situated along bases of valves, sur- face rugulose but smooth. Valves moderate and equal, surfaces finely granulate. Variation. — The gonopods of mimetica are quite uniform. The length of the prefemoral process varies and the arch may be broader than in the holotype, but these minor differences occur randomly throughout the range. On the eastern edge of the range, 11 mi. E Lebanon, Wilson County, occurs the different form shown in figures 113-114. The prefemoral process tapers smoothly and continuously throughout the length, rather than just apically, and the acropodite has rounded lobes on the lateral margin at the R.M. SHELLEY & D.R. WHITEHEAD 137 apical curve and another one just proximal to the tip, which is blunt. The distal zone is directed toward the prefemoral process instead of the basal zone. I believe it is a relict of a former faunal connection between mimetica and crassicurvosa, and it may be extinct, as this area has been examined three additional times without success. Since it occurs on the range periphery of mimetica, | include it under this species. Ecology. — I have taken mimetica under thin layers of leaves on relative- ly hard substrates near water sources. Some samples, for example those in Henry Horton and Cedars of Lebanon State Parks, were remote from water but still met the other ecological criteria. Distribution. — An area about 60 miles long and 50 miles wide in the Nashville Basin and adjacent fringe of the Western Highland Rim of central Tennessee. All specimens have been taken north of the Duck River, and the range traverses the Cumberland River. I have not encountered mimetica in the Eastern Highland Rim Province or in Kentucky, although the Robert- son and Macon County localities are only about 10 miles south of the state line. Specimens were examined as follows: TENNESSEE. — Robertson Co., 3 mi. E Springfield, along TN hwy. 76, 2.2 mi. E jct. TN hwy. 49, M, F, 12 June 1979 (NCSM A2753). Summer Co., 1.7 mi. S. Bransford, 3M, 3F, 20 April 1958, L. Hubricht (RLH). Macon Co., 4 mi. SW Lafayette, 2M, F, 26 April 1958, L. Hubricht (RLH). Cheatham Co., 3.6 mi. N Ashland City, along TN hwy. 49, 3.6 mi. N jct. TN hwy. 12, 4M, 6F, 12 June 1979 (NCSM A2758). Davidson Co., 6 mi. SE Ashland City, along TN hwy. 12 at Cheatham Co. line, 2F, 12 June 1979 (NCSM A2759); Glendale Hills S Nashville, M, 21 April 1917, H. Cummins (RVC) TYPE LOCALITY; and Lake Radnor St. Pk., 6M, 2F, 12 June 1979 (NCSM A2761). Wilson Co., 11 mi E Lebanon, 2M, 25 June 1957, L. Hubricht (RLH); 12.8 mi. NNE Lebanon, along US hwy. 70N, 0.8 mi. W Smith Co. line, M, 2F, 28 May 1980 (NCSM A3167); and Cedars of Lebanon St. Pk., 10M, 3F, July 1958, R.M. Sinclair (RLH) and 3M, 16 June 1976 (NCSM A876). Cannon Co., 5.0 mi. NW Wood- bury, along TN hwy. 145, 4.8 mi. N jct. US hwy. 70S, 4M, 3F, 9 May 1979 (NCSM A2653). Williamson Co., “‘beyond Glendale,’’ M, 14 October 1916, H. Cummins (RVC). Maury Co., 6.3 mi. ENE Columbia, along TN hwy. 99, 5.5 mi. E jct. US hwy. 31, M, 10 May 1979 (NCSM A2662). Marshall Co., Henry Horton St. Pk., campground area, 12M, 8F, 9 May 1979 (NCSM A2660). Remarks. — In addition to the holotype, Chamberlin (1918a) also reported a female of mimetica from Hillsboro Hills, Nashville, and a male from ‘‘beyond Glendale.’’ Both are present in the Chamberlin collection, but the female could be of Brachoria glendalea (Chamberlin), which is sym- patric with mimetica in southern Davidson County. Since the area thought to represent Glendale Hills touches the Williamson County line, the male from ‘‘beyond Glendale’’ is considered to be from this county and is so cited above. Sigmoria (F.) mimetica is the westernmost species in the genus, and the collection from Cheatham County, specifically, constitutes the range limit. MEM. AMER. ENT. SOC., 35 138 XYSTODESMID MILLIPEDS I have sampled intensively over 40 miles farther west to the Tennessee River, and sporadically another 85 or so miles to the Mississippi River, to confirm that no congeners exist in this direction. Sigmoria (Falloria) crassicurvosa Shelley, new species Figs. 115-120 Type specimens. — Male holotype (NCSM A2750) and 2 male and 3 female paratypes collected by R.M. Shelley and R.K. Tardell, 11 June 1979, from Smith Co., TN, 2.0 mi. W Carthage, along TN highway 25, 1.2 mi. W junction with TN highway 85. Male and female paratypes deposited in FSCA. Diagnosis. — A moderate-size species of Sigmoria with narrow medial flange located on peak and with pink paranota and blue transverse metatergal stripes; gonopods with following diagnostic characters: prefemoral process variable, short and acute or vestigial; acropodite thick and massive, broadly curved and subtending rectangular arch, extending well beyond level of prefemoral process; basal zone with inner surface directed mediad, with or without ledge on medial surface; anterior bend and apical curve sharp, well defined; peak gently curved, tilted laterad, exposing undersurface in medial view, with or without distal expansion on lateral sur- face; distal zone equal in length to and coplanar with basal zone, either parallel to latter and bent into arch near midlength or angling into arch, configuration variable, with or without proximal indentation on lateral sur- face, broadly rounded lobe at midlength on medial side, and terminal medial lamina; tip elongated and acuminate, with converging striae. Color in life. — Paranota pink; metaterga black with blue transverse stripes along caudal margins connecting paranotal spots; collum with concolorous blue stripes along anterior and posterior edges. Holotype. — Length 36.2 mm, maximum width 9.0 mm, W/L 24.9%, depth/width ratio 53.2%. Segmental widths as follows: collum 6.4 mm 15th 8.7 2nd 7.8 16th 7.9 3rd 8.5 17th 6.8 4th 8.9 18th 5.3 5th-14th 9.0 Fics. 115-120. Sigmoria (Falloria) crassicurvosa. 115, process of 4th sternum of holotype, caudal view. 116, gonopods in situ, ventral view of paratype. 117, telopodite of left gonopod of holotype, medial view. 118, the same, lateral view. 119, the same, subdorsal view. 120, telopodite of left gonopod of male from Dekalb Co., TN, medial view. Scale line for fig. 116 = 1.00 mn;; line for other figs. = 0.80 mm for 119; 1.00 mm for 117-118 and 120; 1.33 mm for 115. R.M. SHELLEY & D.R. WHITEHEAD MEM. AMER. ENT. SOC., 35 13) 140 XYSTODESMID MILLIPEDS Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.4 mm, interantennal isthmus 1.3mm. Antennae reaching back to middle of 3rd paranota, relative lengths of antennomeres 2>3>4=5>6>1>7. Genae with distinct impressions. Facial setae as follows: epicranial, interantennal, and genal absent, fron- tal 1-1, clypeal about 10-10, labral 12-12. Terga smooth, polished, becoming moderately coriaceous on paranota. Collum broad, ends not produced beyond those of following tergite. Paranota moderately depressed, continuing slope of dorsum, caudolateral corners rounded through segment 5, blunt on 6-12, becoming progressively more acute posteriorly. Peritremata flat on anterior segments, becoming elevated and distinct on segment 6. Ozopores located near middle of peritremata, opening dorsad. Sternum of segment 4 with small, divided process, much shorter than widths of adjacent coxae (Fig. 115); that of segment 5 with two conical paramedian knobs between 4th legs, much shorter than adjacent coxal widths, and two low elevated areas between Sth legs; that of seg- ment 6 strongly and convexly depressed between 7th legs to accommodate apical curvatures of acropodites. Postgonopodal sterna flattened and platelike, with bicruciform impressions on segments 8-9 and shallow central impression thereafter. Coxae with low, rounded tubercles on segments 10-12; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture elliptical, 4.0 mm wide and 1.3 mm long at midpoint, without indenta- tions, sides flush with metazonal surface. Gonopods in situ (Fig. 116, of paratype) with acropodites curving anteriad entirely over aperture, angling slightly toward midline and touching or slightly overlapping in apical curve regions. Gonopod structure as follows (Figs. 117-119): Prefemoral process short and acute, directed towards coxa. Acropodite thick and massive, heavily sclerotized, broadly curved but subtending rectangular arch, overhanging and extending well beyond level of prefemoral process; basal zone with inner surface directed mediad, relatively long and straight, without modifications; anterior bend sharp, well defined; peak relatively long, gently curved, tilted laterad exposing undersurface in medial view; apical curve sharp, well defined; distal zone long, equal in length and coplanar with basal zone, with broadly rounded lobe distad on medial side, bent inward into arch apically; tip broadly rounded on lateral side, with striae on both lateral and medial sides converging onto elongated, acuminate medial corner. Medial flange narrow and indistinct, located on peak. Lateral flange absent. Prostatic groove running along medial and inner surfaces of acropodite to terminal opening. Male paratypes. — The male paratypes agree with the holotype in all particulars. Female paratype. — Length 42.5 mm, maximum width 10.1 mm, W/L 23.8%, depth/ width ratio 64.4%. Cyphopods in situ with corners of valves and receptacles visible in apertures, valves directed anteriolaterad. Receptacle large, situated along bases of valves, surface rugulose but smooth. Valves moderate and equal, surfaces finely granulate. Variation. — The Dekalb County males differ considerably from the types as shown in figure 120. The prefemoral process is a vestigial nub and barely detectable on the massive prefemur. There is a ledge along the medial surface of the basal zone that is broadest distad and might represent part of the medial flange. The peak is expanded distad into a rounded lateral margin, which is indented on the proximal part of the distal zone. This sec- tion of the acropodite is straight and angles toward the basal zone instead of being parallel as in the holotype. Apically, there is no medial lobe on the distal zone, but there is a terminal lamina which obscures the proximal part of the solenomerite in medial view. R.M. SHELLEY & D.R. WHITEHEAD 141 Ecology. — Sigmoria (F.) crassicurvosa occurs under thin layers of ives on relatively hard substrates near water sources. Distribution. — Known only from the following three localities on the stern edge of the Nashville Basin. "ENNESSEE. — Smith Co., 2 mi. W Carthage, along TN hwy. 25, 1.2 mi. W jct. TN hwy. 3M, 3F, 11 June 1979 (NCSM A2750) TYPE LOCALITY; and 7 mi. SW Carthage, along 1umbered rd. off TN hwy. 141, M, F, 28 May 1980 (NCSM A3170). Dekalb Co., 5 mi. NW Smithville, along US hwy. 70, 3M, 9 May 1979 (NCSM A2648). i) \ y 123 124 Fics. 121-124. Sigmoria (Falloria) pendulata. 121, process of 4th sternum of holotype, caudal view. 122, gonopods in situ, ventral view of paratype. 123, telopodite of left gonopod of holotype, medial view. 124, the same, lateral view. Scale line for fig. 122 = 1.00 mm; line for other figs = 1.00 mm for 123-124; 1.33 mm for 121. MEM. AMER. ENT. SOC., 35 142 XYSTODESMID MILLIPEDS Sigmoria (Falloria) pendulata Shelley, new species Figs. 121-124 Type specimens. — Male holotype (NCSM A2650) and 5 male and 5 female paratypes collected by R.M. Shelley and R.K. Tardell, 9 May 1979, from Edgar Evins State Park, Dekalb Co., TN. Male and female paratypes deposited in FSCA. Diagnosis. — A moderate-size species of Sigmoria with medial flange on peak and with pink paranota and usually blue metatergal stripes; gonopods with following diagnostic characters: prefemoral process absent or varying in size to moderately long and acute; acropodite thick and massive, arch an Fic. 125. Distributions of ainsliei, dots; aphelorioides, triangles; and prolata, squares, in western North Carolina and eastern Tennessee. The dashed line is the approximate boundary between the Blue Ridge and Ridge and Valley Provinces. R.M. SHELLEY & D.R. WHITEHEAD 143 inverted U, extending well beyond level of prefemoral process; basal zone with inner surface directed mediad; anterior bend and apical curve well defined; peak high and rounded, tilted laterad; distal zone coplanar with and equal in length to basal zone, with expansion at midlength on anterior side extending onto medial and lateral surfaces and forming lobes, curved slightly inward distad and tapering to blunt tip. Color in Life. — Paranota pink; metaterga black with blue, occasionally yellow, stripes along caudal margins connecting paranotal spots; collum with pink stripe along anterior edge and blue along posterior. Holotype. — Length 33.9 mm, maximum width 9.0 mm, W/L ratio 26.5%, depth/ width ratio 69.7%. Segmental widths as follows: collum 6.4 mm 8th-14th 9.0 2nd 7.6 15th 8.5 3rd 8.3 16th 8.0 4th 8.5 17th 6.5 Sth-7th 8.8 18th 4.6 Somatic features similar to /. /atior, with following exceptions: Width across genal apices 4.7 mm, interantennal isthmus 1.4 mm. Antennae reaching back to caudal edge of 3rd paranota, relative lengths of antennomeres 2>3>6>4>1>7. Genae with shallow central impressions. Facial setae as follows: epicranial, interantennal, and genal absent, frontal 1-1, clypeal 12-12. Terga smooth, polished, with only faint wrinkling on anterior parts of paranota. Collum broad, ends extending slightly beyond those of adjacent tergite. Paranota moderately de- pressed, continuing slope of dorsum, caudolateral corners rounded through segment 5, blunt on 6-12, becoming progressively more acute posteriorly. Peritremata sharp and distinct on all segments, clearly elevated above metazonal surface, ozopores located near middle of peritremata, opening dorsolaterad. Sternum of segment 4 with minute, divided projection, much shorter than adjacent coxal widths (Fig. 121); that of segment 5 with two low rounded knobs between 4th legs, much shorter than adjacent coxal widths, and broad flattened elevations between Sth legs; that of segment 6 deeply and convexly recessed between 7th legs to accommodate apical curvatures of acropodites. Postgonopodal sterna flattened and plate-like, bicruciately impressed on segments 8-9 and with variably broad, shallow central impressions on remaining segments. Coxae with low rounded tubercles beginning on segment 8; prefemoral spines beginning on segment 5, becoming progressively longer and sharper caudally. Gonopodal aperture elliptical, 3.5 mm wide and 1.8 mm long at midpoint, without indenta- tions, sides elevated above metazonal surface. Gonopods in situ (Fig. 122, of paratype) with acropodites extending anteriomediad and peak of one overlapping apical curve of other, apical curve of former extending slightly beyond anterior margin of aperture and inserting in depres- sion between 7th legs. Gonopod structure as follows (Figs. 123-124): Prefemoral process ab- sent, with only a lowly rounded ridge at this position on prefemur. Acropodite thick and massive, heavily sclerotized, subtending symmetrical, inverted U shaped arch, overhanging and extending well beyond prefemur; basal zone moderately long, inner surface directed mediad; anterior bend sharp, well defined, continuous through peak with distal zone; peak short, high, and rounded, apex at midlength, continuing curvature of arch, tilted only slightly laterad; apical curve sharp, well defined; distal zone long, coplanar with and nearly equal in length to basal zone, with broadly rounded swelling at midlength on anterior side extending MEM. AMER. ENT. SOC., 35 144 XYSTODESMID MILLIPEDS onto medial and lateral surfaces thus producing lobe-like effect at this point, slightly curved beyond swelling and tapering to blunt termination, directed toward cannula of coxa; tip located on produced inner corner of distal zone. Medial flange narrow, located on peak, in- distinct in medial view. Lateral flange absent. Prostatic groove running along inner surface of basal zone and under surface of peak to opening at tip of distal zone. Male Paratypes. — The male paratypes agree with the holotype in all structural details. In color, however, four specimens had yellow metatergal stripes as opposed to the blue stripes of the holotype and one paratype. Female Paratype. — Length 37.9 mm, maximum width 10.3 mm, W/L ratio 27.2%, depth/ width ratio 67.9%. Cyphopods in situ with corners of valves and receptacles visible in aper- tures, valves directed anteriolaterad. Receptacle large, situated along bases of valves, surface rugulose but smooth. Valves moderate and equal, surfaces finely granulate. Variation. — Color varies only in the type series. All other specimens have blue metatergal stripes connecting pink paranotal markings. Males in the eastern edge of the range in Putnam County possess long, acute prefemoral processes that extend nearly to the levels of the distal cS fy Fic. 126. Distribution of the subgenus Falloria in central Tennessee. Dots, mimetica; half shaded dot, atypical population, possible remnant of intergrades between mimetica and crassicurvosa; squares, crassicurvosa; triangles, pendulata; asterisks, picapa; stars, forficata, ovals, abbreviata; diamonds, houstoni. The dashed lines show the approximate boundaries of the Cumberland Plateau Province (CP), and the Nashville Basin (NB) is denoted by a line of alternating dots and dashes. The intervening area, which curves northward over the Nashville Basin and continues on the western side, is the Highland Rim. The heavy line in the south- eastern (lower right) corner represents the course of the Tennessee River. R.M. SHELLEY & D.R. WHITEHEAD 145 zones, and there is a geographic trend toward a longer and larger structure from west to east. The types and the most proximal sample from western Putnam County lack the process, and it is short but distinct in sample A3176, from an intermediate locality. However, no changes are evident in the acropodites. Ecology. — Sigmoria (F.) pendulata occurs under thin layers of leaves on relatively hard substrates near water sources. Distribution. — Known only from an area about 13 miles long and 8 miles wide in the eastern Highland Rim. Specimens were examined as follows: TENNESSEE. — Dekalb Co., Edgar Evins St. Pk., 6M, 5F, 9 May 1979 (NCSM A2650-2651) TYPE LOCALITY. Putnam Co., 12 mi. W Cookeville, along US 70N at Lafayette Twp., 2M, 7F, 10 June 1979 (NCSM A2748); 14.3 mi. SW Cookeville, along co. rd. 6162, 0.3 mi. S TN hwy. 141, 5M, 4F, 28 May 1980 (NCSM A3176); and 9.5 mi. SSW Cookeville, Burgess Falls area on Falling Water Rd., 3M, 2F, 29 May 1980 (NCSM A3178). Remarks. — Approximately 17 miles separate the ranges of pendulata and mimetica, and if the swelling on the distal zone of pendulata were removed, its acropodite would be very close to that of the latter except for slightly less curvature in the distal zone. Part II: REVISION OF DELTOTARIA by Rowland M. Shelley Taxonomic Characters Hoffman (1961) cited two diagnostic features for Deltotaria, the coxal apophysis and the absence of frontal and vertigial (epicranial) facial setae. However, many apheloriine xystodesmids lack epicranial setae, and with much more material at my disposal, I can report that frontal setae are present on most individuals of both sexes of De/totaria. Thus, the only truly diagnostic generic character is the subconical coxal apophysis. Hoffman also reported that the gonocoxae were connected by a partially sclerotized sternal remnant, which would be a second such feature as this was one of the traits cited for the tribe Rhysodesmini (Hoffman 1960). However, I have carefully examined a number of males of both species and found that although the membrane joining the coxae is occasionally quite tough and difficult to tear, there is no evidence of sclerotization. Deltotaria thus con- forms to contribal genera in lacking a sternal remnant. Hoffman also regarded the form of the caudolateral corner of the paranota as significant at the specific level, since the peritremata are prolonged slightly beyond the MEM. AMER. ENT. SOC., 35 146 XYSTODESMID MILLIPEDS caudal edge in some forms to produce a small projection. However, this oc- curs widely in the Apheloriini and is insignificant. As with most apheloriine genera, the taxonomically important characters in Deltotaria involve chiefly the male gonopods. The two species are differently colored, as indicated below, but the pregonopodal sternal processes are nearly identical and in- sufficiently different from conditions in sympatric species to be useful in generic determinations. Coloration. — Both species exhibit the metatergal stripe pattern characteristic of many forms of Sigmoria s. lat., but lea Hoffman, in the Piedmont Plateau, is exclusively yellow, whereas brimleii Causey, in the southern Blue Ridge Province, displays various shades of red and orange. There is also an unusual aspect to the colors of /ea and southern specimens of brimleii philia (Chamberlin) that was first reported by Filka and Shelley (1980). Whereas many xystodesmids display a glossy, highly polished finish, there is a sparkle or glitter to the pigments of these forms of Deltotaria that is much like that of metallic paint. This sheen of /ea allows for positive iden- tifications of even females in the field and thus is diagnostic for both the species and the genus in the Piedmont Plateau. The difference is un- mistakable and readily apparent if one compares live specimens of D. lea and sympatric, yellow striped intergrades of Sigmoria (Sigmoria) latior. Gonopodal Characters..— In the terminology I have developed for “‘sigmoid’’ gonopods, what Hoffman (1961) labeled a subterminal lobe or process and identified by the letter ‘‘B’’ is really the lateral flange, and what he considered the solenomerite is actually the part of the distal zone distal to the flange. To date, all gonopodal drawings of Deltotaria have also been from a somewhat ventrolateral aspect since this view best reveals the coxal apophyses. Consequently, the medial flanges have yet to be illustrated or described. These lamellae clearly reveal the generic affinity between Deltotaria and Sigmoria s. lat., and the locations are diagnostic at the specific level. I therefore present medial views of all illustrated gonopods to highlight this as yet unrevealed feature. The coxal apophysis is the only truly diagnostic generic character, as the gonopods would otherwise be readily referrable to Sigmoria s. lat. This pro- jection arises anteriolaterad on the coxa and extends directly ventrad. The length and configuration are useful at the specific level, and as viewed in medial and lateral profiles, there are two basic types — broadly triangular with sides narrowing rapidly to a subacuminate tip, and long and slender with sides roughly parallel through most of the length except apically. Triangular apophyses obtain in /ea, where the projection is short and ter- minates below the level of the prefemur, and the nominate subspecies of brimleii, where it is moderately long and extends to the level of the R.M. SHELLEY & D.R. WHITEHEAD 147 prefemoral process. In b. philia the process is slender and much longer, stretching to or beyond the level of the tip of the acropodite and thus well beyond the prefemoral region. Despite the clear distinctions in the coxal apophyses, I do not believe the two forms of brimleii are reproductively isolated, as there is evidence of intergradation and there are no significant acropodal differences. On the acropodites, specific differences obtain in the apical thickness, the curvature of the distal zone, and the locations and shapes of the medial and lateral flanges. Deltotaria lea has a decidedly stronger, thicker, and heavier acropodite than does brimileii, particularly in the distal zone, which is broad and spatulate in the former and thin and fragile in the latter. Accordingly, lea is apically blunt whereas brimleii is subacuminate. In curvature, the distal zone extends downward from the peak in /ea and is coplanar with the basal zone. In brimleii, however, the section projects laterad to varying degrees, is not coplanar with the basal zone, and in a few individuals is near- ly coplanar with the peak. All forms of De/totaria have a medial flange, but in Jea it is broadly expanded and located on the basal zone, whereas in brimleii it is narrow and located on the peak and/or distal zone. In /ea the flange is curved to form a concave inner surface on the basal zone. In brimleii the flange is level or straight with sometimes a gently rounded margin, but it is comparatively indistinct and poorly demarcated from the acropodite stem. The lateral flange is absent from /ea and present in brimleii, where there is a general trend toward smaller flanges in a southerly direction. In northern populations the lateral flange is produced into lobes of varying lengths and shapes, but in b. philia at the southern periphery, it is narrow, indistinct, and poorly demarcated from the acropodite stem. Here, the flange imparts a hastate appearance to the acropodite and is best detected in profile by examining the distal zone from a ventral perspective. Genus DELTOTARIA Causey Deltotaria Causey, 1942:165. Chamberlin and Hoffman, 1958:29. Hoffman, 1961:22-25; 1979:158. Jeekel, 1971:258. Phanoria Chamberlin, 1949:101. Jeekel, 1971:279 Type species. — Of Deltotaria, D. brimleii Causey, 1942, by original designation; of Phanoria, P. philia Chamberlin, 1949, by original designa- tion. Diagnosis. — Characterized by the variably triangular or elongate coxal apophysis. MEM. AMER. ENT. SOC., 35 148 XYSTODESMID MILLIPEDS Color in Life. — Paranota red, orange, or yellow; metaterga black with concolorous red, orange, or yellow stripes along caudal margins connecting paranotal spots; collum with stripes along anterior or both margins; colors sometimes with sparkling, metallic sheen. Description. — Head of normal appearance, smooth, polished. Epicranial suture shallow, distinct, terminating in interantennal region; interantennal isthmus moderately wide; genae not margined laterally, with shallow central impressions, ends broadly rounded and projecting slightly beyond adjacent cranial margins. Antennae moderately slender, varying in length, becoming progressively more hirsute distally, with 4 conical sensory cones on ultimate article, no other sensory structures apparent. Facial setae reduced; epicranial and interantennal ab- sent; frontal and genal present or absent, clypeal and labral present. Terga generally smooth, polished, becoming moderately coriaceous on anterior halves of paranota. Collum variably broad, ends extending slightly beyond those of following tergite. Paranota moderately depressed, generally continuing slope of dorsum, caudolateral corners rounded on anteriormost segments, becoming blunt in midbody region and progressively more acute posteriorly. Peritremata generally indistinct, slightly elevated above paranotal surfaces; ozopores located caudal to midlength, opening dorsolaterad. Prozonites smaller than metazonites; strictures distinct, smooth. Caudal segments normal for family. Sides of metazonites smooth or irregular, with varying shallow, curved impressions. Stric- tures sharp, distinct. Pregonopodal sterna of males modified as follows: that of segment 4 with small to moderate, apically divided process between 3rd legs, shorter than widths of adjacent coxae; that of segment 5 with small projections between anterior legs and variable elevated areas between posterior ones; that of segment 6 with variable convex recession between caudal legs to accommodate apices of acropodites when body segments compressed, 7th legs sometimes set slightly farther apart than 6th. Postgonopodal sterna generally flat and un- modified, with shallow grooves on those immediately posterior to 7th segment and shallow central impressions on rest. Gonapophyses on 2nd leg pair of males distinctly elevated above coxal surfaces, with round, apical knobs. Pregonopodal legs densely hirsute; postgonopodal legs becoming progressively less hirsute caudally. Coxae with low tubercles beginning on postgonopodal legs, becoming progressively longer and sharper caudally; prefemoral spines beginning on segment 5, becoming longer and more acute caudally; tarsal claws bisinuate. Hypoproct variable, subtriangular to rounded; paraprocts with margins strongly thickened. Gonopodal aperture ovoid to elliptical, with slight indentation anteriolaterad, front flush with metazonal surface, sides and caudal edge elevated. Gonopods in situ with acropodites projecting ventrad from aperture, curving anteriomediad and overlapping at various positions in midline, extending anteriad slightly beyond anterior margin of aperture. Coxae moderate, connected by membrane only, no sternal remnant, with subconical apophyses of variable lengths arising on anterior sides and projecting ventrad. Telopodite set terminally on coxa, lat- ter not projecting distad beyond prefemoral region. Prefemora moderate, with or without variable prefemoral processes, latter occasionally vestigial. Acropodite moderately thick and heavy basally, becoming thinner and more fragile or spatulate distally, with torsion, curving through one or more vertical planes in vaguely sigmoidal configuration as seen in situ and in medial and lateral views, configurations as described for Sigmoria s. lat. Basal zone usually relatively long, around 1/3 of total acropodite length, inner surface directed anteriad, medial surface with or without broad, curved, laminate flange, latter forming convex inner surface. Anterior bend usually broad, poorly defined. Peak variable but usually relatively short and gently curved, apex near midlength, with or without variably narrow laminate flange on medial edge. Apical curve variable but usually broad, poorly defined. Distal zone of variable length and configuration, either broad and spatulate or with sides tapering smoothly and continuous- — =. — eS R.M. SHELLEY & D.R. WHITEHEAD 149 ly, projecting downward from peak and coplanar with basal zone or not coplanar and extend- ing laterad to various degrees, occasionally sharply curved into arch and lying beneath and parallel to peak; with or without variably narrow medial flange and variable lateral flange, lat- ter ranging from long and rectangular to narrowly triangular to small and gently rounded; distal zone occasionally tilted mediad or laterad exposing medial or lateral flange in opposite perspective. Termination simple, blunt to acuminate. Prostatic groove arising in pit on prefemur, running along stem of acropodite and crossing from medial to lateral sides at various locations, opening terminally on tip of distal zone. Females agreeing closely with males in somatic features, except paranota more strongly depressed, creating appearance of more highly arched body. Cyphopodal aperture broad, en- circling 2nd legs, sides slightly elevated above metazonal surfaces. Cyphopods in situ located lateral to 2nd legs, with corners of valves and receptacles visible in aperture. Receptacle small to moderate, located on medial sides of valves, surface rugulose to finely granulate. Valves small to moderate, subequal in size, directed caudolaterad, surfaces rugulose to finely granulate. Operculum minute, hidden under free end of valves. Distribution. — Piedmont Plateau of west-central North and South Carolina between the Catawba and Broad rivers; and the southern Blue Ridge Province of North Carolina, Tennessee, Georgia, and South Carolina (Fig. 146). The range in the mountains extends onto the eastern and southern escarpments, but forms have not been encountered in the Pied- mont proper. There is no tangible northern boundary in the mountains. Species. — Two, one in each of the above described regions. Other workers may interpret the montane forms differently and recognize more reproductive isolates, but I doubt if any different species await discovery. Deltotaria thus resembles Dynoria in having two allopatric species, one in the Blue Ridge Province and one in the Atlantic lowlands (see Shelley 1984b). Deltotaria brimleii Causey Diagnosis. — Color of paranota and metatergal stripes varying from red to light orange, occasionally with sparkling, metallic sheen; coxal apophysis in profile broadly triangular to slender with sides nearly parallel, moderate to long, terminating at or beyond level of prefemoral process; acropodite distally slender and fragile; medial flange located on peak or proximal part of distal zone, with only slight lobe; lateral flange present, variable. Ecology. — Deltotaria brimleii is a cove dwelling species. Remarks. — Deltotaria brimleii occurs widely in the Southern Blue Ridge Province (Fig. 146). Five names have been proposed for forms in this area, but I think none are reproductively isolated. The only possible excep- tion is the southernmost, occurring mainly in Georgia, for which the name philia was proposed by Chamberlin (1949). Its acropodite is basically the same as those of northern forms, and the narrow, indistinct lateral flange MEM. AMER. ENT. SOC., 35 150 XYSTODESMID MILLIPEDS 131 128 | 133 Fics. 127-136. Deltotaria brimleii brimleii. 127, process of 4th sternum of holotype, caudal view. 128, gonopods in situ, ventral view of male from Montreat, Buncombe Co., NC. 129, left gonopod of the same, medial view. 130, the same, lateral view. 131, distal half of acropodite of male from Bent Creek Forest Experiment Station, Buncombe Co., medial view. 132, the same, lateral view. 133, distal half of acropodite of male from Gatlinburg, Sevier Co., TN, medial view. 134, the same, lateral view. 135, distal half of acropodite of male from 8 mi. NW Brevard, Transylvania Co., NC, medial view. 136, the same lateral view. Scale line for fig. 128 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 151 continues the general north to south clinal trend toward smaller laminae. However, philia is distinguished by the form of the coxal apophysis, which in profile is narrow with nearly parallel sides and extends to or beyond the level of the distal zone, well beyond the level of the prefemur. With one ex- ception, all the other material has broadly triangular apophyses that ter- minate at the level of the prefemoral process. In this exception, from Chimney’s Picnic Area, GSNMP, Tennessee, the process is intermediate in both length and shape. Because of this sample and the absence of mean- ingful acropodal differences, I think philia is only a geographic race and therefore recognize two subspecies based on the length and configuration of the coxal process. Three names are placed in synonymy under the nominate subspecies. Deltotaria brimleii brimleii Causey, new status Figs. 127-136 Deltotaria brimleii Causey, 1942:165, Figs. 1-2. Chamberlin and Hoffman, 1958:30. Hoffman, 1961:25-28, Fig. 2c. Deltotaria brimleardia Causey, 1950a:7-8, Figs. 2-3. Chamberlin and Hoffman, 1958:29. Hoffman, 1961:30-31, Fig. 2b. NEW SYNONYMY. Deltotaria tela Causey, 1950b:38-39, Figs. 3-5. Chamberlin and Hoffman, 1958:30. Hoffman, 1961:31-33, Figs. 1b, d, 2a. NEW SYNONYMY. Deltotaria mariana Hoffman, 1961:28-30, Figs. la, e-h; 2d. NEW SYNONYMY. Type specimen. — Male holotype (ANSP) collected by C.S. Brimley, 26 May 1923, at Swannanoa, Buncombe Co., NC. According to Hoffman (1961) there is a female topoparatype in the Causey collection, now at the FSCA. Hoffman presented a detailed description of the somatic aspects of the holotype, but as the genitalia are missing, he could only quote Causey’s description (1942) and add a short interpretive comment. The following ac- count of the pregonopodal sterna and the gonopods in “‘sigmoid’’ ter- minology is therefore prepared from a nearly topotypical male. Diagnosis. — Coxal apophysis in profile moderately long and broadly triangular, extending to level of tip of prefemoral process, sides tapering rapidly to subacuminate tip; prefemoral process variable but usually pres- ent, occasionally vestigial. Color in Life. — Paranota red; metaterga black with concolorous red stripes along caudal margins connecting paranotal markings; collum with red stripes along both anterior and posterior edges. Male from Montreat, NC. — Fourth sterna with moderately long process between third legs, slightly shorter than widths of adjacent coxae (Fig. 127); segment 5 with minute paramedial knobs between 4th legs, much shorter than widths of adjacent coxae, and low rounded elevated areas between Sth legs; 6th sternum with slight recession along caudal edge to accommodate apical curvatures of acropodites. MEM. AMER. ENT. SOC., 35 152 XYSTODESMID MILLIPEDS Gonopodal aperture elliptical, 3.4 mm wide and 1.7 mm long at midpoint, indented slightly anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 128) with acropodites extending anteriolaterad from aperture and peaks overlapping in midline, curving slightly over opposite side and anterior margin with tips crossing. Gonopod structure as follows (Figs. 129-130): Coxal apophysis curving caudad at midlength, directed toward prefemoral process and terminating just below tip of latter, apically subacuminate. Prefemoral process short, wedge shaped. Acropodite moderately thick and heavy basally, becoming thin- ner and more fragile distad, arch high and gently curved, overhanging and extending just beyond level of prefemoral process; basal zone long, bowed caudad; anterior bend broad, poorly defined; peak high and gently rounded, apex at midlength; apical curve sharp, well defined; distal zone moderately long, projecting strongly laterad from peak, not coplanar with basal zone, curving sharply caudad at midlength and lying under and parallel to peak in medial view, sides tapering smoothly and continuously to acuminate tip, latter directed toward anterior bend. Medial flange clearly demarcated from peak, moderately long, arising at midlength (apex), curving gently outward to form narrow lobe, terminating at distal extremity of peak, margins smooth. Lateral flange with broad, rectangular, proximal lobe, equal in length to distal half of distal zone, with two sharp apical teeth, sides smooth, lobe terminating at midlength, flange then curving gently to terminus near tip of distal zone. Prostatic groove crossing to lateral side at apex of peak, continuing to terminal opening. Female from Polk Co., NC. — Length 34.1 mm, maximum width 7.8 mm, W/L ratio 22.9%, depth/width ratio 62.8%. Cyphopods with receptacle moderate, surface rugulose. Valves moderate, equal, surfaces rugulose. Variation. — This subspecies includes all forms with a moderate coxal process. The lateral flange is highly variable and may be relatively laminate with a curvilinear margin or be produced into variable lobes, which may be rectangular, broadly or sharply triangular, or spiniform. The longest or rec- tangular lobes obtain in the northern part of the range, in the form named brimleii by Causey (1942). In the southern form named mariana by Hoff- man (1961), the distal zone is tilted mediad exposing part of the lateral flange in medial view (Fig. 135). As noted by Causey (1950b) in the descrip- tion and later by Hoffman (1961), the form named fe/a vaguely resembles Sigmoria (Sigiria) rubromarginata because the margin of the lateral flange is slightly produced, and the distal zone projects more strongly laterad, ex- posing part of the medial flange in lateral view (Fig. 132). It occurs chiefly in the Pisgah National Forest in Buncombe County, North Carolina, and occasional males from other areas approximate this configuration, for ex- ample in Cocke County, Tennessee. I do not find the resemblance with S. rubromarginata very striking and consider it convergence; however, it might reflect an ancestral connection. Overall in b. brim/eii there is a north to south clinal progression toward smaller, less distinct lateral flanges, which culminates in the inconspicuous lamella of the other subspecies. Dif- ferences also obtain in the arch of the acropodite and the curve of the distal zone, which is directed more strongly laterad in the western part of the range. R.M. SHELLEY & D.R. WHITEHEAD 153 Distribution. — The southern Blue Ridge Province from the Linville. River to the Little Tennessee Rivers in North Carolina and the northeastern corner of Georgia, and, east-west, from the escarpment in North Carolina to the Great Smoky Mountains of Tennessee (Fig. 146). Most of the material comes from south of the Black Mountains in North Carolina. The two northernmost records represent gonopodal fragments found with paratypes of Sigmoria aberrans and S. conclusa, both synonymns of /. latior (Sigmoria), in the Chamberlin collection. They possess the lateral flange and general acropodal configurations that occur in the northern part of the range (similar to the form in Figs. 129-130) and the localities therefore seem plausible. However, confirmation with fresh material is desirable. Specimens were examined as follows: NORTH CAROLINA. — Burke Co., Linnville Falls, M (segments 7-8), 12 August 1910, R.V. Chamberlin (RVC). Mitchell Co., Altapass, M (segment 7 only), 11 August 1910, R.V. Chamberlin (RVC). Buncombe Co., Montreat, M, 5 September 1977 (NCSM A1701); Swan- nanoa, M, 26 May 1923, C.S. Brimley (SNAP) TYPE LOCALITY; 5.2 mi. NW Avery Cr., along NC hwy. 191 at Blue Ridge Parkway nr. French Broad R., M, 1 June 1977, A.L. Braswell (NCSM A1595); Bent Creek Forest Experiment Station, M, F, 30 April 1939, N.B. Causey (ANSP); 8.8 mi. SE Asheville, Lake Powhatan, 2M, 5 September 1977 (NCSM A1706); and 11 mi. S Asheville, along co. rd. 3495, 0.3 mi. E ject. I-26, M, 3F, 16 June 1979 (NCSM A2769). Polk Co., 1.5 mi. NNE Saluda, along co. rd. 1151, 0.3 mi. W jet. co. rd. 1142, M, F, 14 September 1977 (NCSM A1741). Haywood Co., along US hwy. 276 S of Waynesville, M, 24 June 1961, H.V. Weems (FSCA); and 12 miles NW Waynesville, GSMNP, nr. Cataloochee School, 3M, 3F, 8 July 1976 (NCSM A916). Swain Co., 2 mi. NE Bryson City, entrance to GSMNP nr. Deep Cr. Campground, M, F, 16 May 1978 (NCSM A1904). Tran- sylyvania Co., 8 mi. NW Brevard, Pink Beds Rec. Area, M, F, 30 July 1958, R.L. and M.S. Hoffman (NMNH); 5 mi. NW Brevard, Sycamore Flats Rec. Area, M, 26 May 1958, L. Hubricht (RLH); between Rosman and Balsam Grove, M, 19 July 1961, R.L. Hoffman (RLH); and Toxaway Gorge SE Lake Toxaway, 3M, 16 July 1961, R.L. Hoffman (RLH). Jackson Co., 4, mi. ESE Cullowhee, along Caney Fk. at John’s Cr., M, 23 October 1970, F.A. Coyle (WAS). Macon Co., 1.5 mi. SSE Highlands, Satulah Mtn., M, 4 June 1977, A.L. Braswell (NCSM A1592); Highlands, M, 1 July 1963, A. Douglas (RLH); and Coweeta Hydrologic Station, 2M, 6 July 1978 (NCSM A2318). TENNESSEE. — Cocke Co., GSMNP, Cosby Picnic Area, 3M, 8 August 1981 (NCSM A3716). Sevier Co., Gatlinburg, along Ramsey Prong, M, 10 July 1947, H. Hansen (ANSP). GEORGIA. — Rabun Co., 12 mi ENE Clayton, along GA hwy 28 at W. Fork Chatooga R., M, F, 9 June 1978 (NCSM A2057). Remarks. — The various forms of b. brimleii, which appear dissimilar because of differences in the curvature of the acropodite and the shape of the lateral flange, nevertheless connect through intermediate populations that bridge the gaps in the diagnostic characters of the nominal species reported by Hoffman (1961). In my view, none of the named forms demonstrates enough stability over a large enough area to be retained even at the subspecific level. For example, that previously known as fela (Figs. MEM. AMER. ENT. SOC., 35 154 XYSTODESMID MILLIPEDS 131-132) has virtually a point distribution, being restricted to the Bent Creek Forest Area of the Pisgah National Forest in Buncombe County. The closest males to this site only vaguely resemble its types, and to retain this name for a single population is impractical. In August 1981 I collected b. brimleii along the creek at Cosby Picnic Area, GSMNP, Tennessee, in precisely the same spots where I discovered Sigmoria (Sigiria) rubromarginata in May 1978. The latter was absent in 1981 as was the former on the previous trip. This appears to be a third ex- Fics. 137-140. Deltotaria brimelii philia. 137, process of 4th sternum of male from White Co., GA, caudal view. 138, gonopods in situ, ventral view of the same. 139, left gonopod of holotype, medial view. 140, the same, lateral view. Scale line for fig. 138 = 1.00 mm; line for other figs = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 155 ample of two ‘‘sigmoid’’ species occurring at the same site but having their - life histories adjusted so they are prevalent at different times of the year. For Sigmoria (Cheiropus) australis and S. (C.) serrata in Camden County, Georgia, I (1984a) suggested that this would minimize ecological competi- tion if the species occupied similar niches. Future investigators should therefore determine whether both species occur at Cosby in June or July and whether there is a spring to summer decline in abundance of S. (S.) rubromarginata and a concomitant increase in b. brimleii. Deltotaria brimleii philia (Chamberlin), new status Figs. 137-140 Phanoria philia Chamberlin, 1949:101, Fig. 25. Deltotaria philia: Chamberlin and Hoffman, 1958:30. Hoffman, 1961:34-35, Fig. 3b. Type specimen. — Male holotype (RVC) collected by W. Ivie, 27 April 1943, at Clarksville, Habersham Co., GA. Diagnosis. — Coxal apophysis in profile long and slender, extending to near or beyond level of tip of distal zone and well beyond level of prefemur, sides tapering gradually to subacuminate tip; prefemoral process absent or at most vestigial. Color in Life. — Paranota red or orange, or red on outer margin fading into light orange on inner margin; metaterga black with concolorous stripes, red or orange or reddish on outside and orange on inside, along caudal margins connecting paranotal markings; collum with stripe along anterior margin. Variation. — The males of b. philia agree closely with each other, and I detect no significant variation. Color, however, does vary. The specimens from Graham County, North Carolina, were uniformly orange, while those from the two Georgia locations were red. In South Carolina, the red on the outside of the paranota and stripes grades smoothly into light orange on the inside. These specimens also exhibit a sparkling, metallic sheen that is quite different from the glossy finish on other samples. The South Carolina locality is the most proximal to /Jea, which also displays this sheen, and the inner light orange hues may represent the influence of yellow pigmentation, which is the color found in /ea. Distribution. — Southern extremity of the Blue Ridge Province. The race occurs south and west of the nominate subspecies, and the Little Tennessee River appears to be a distributional boundary. Specimens were examined as follows: NORTH CAROLINA. — Graham Co., 9.5 mi. NNE Robbinsville, along Stecoah Cr. on co. rd. 1236, 0.5 mi. N jet. co. rd. 1235, 5M, 3F, 16 May 1978 (NCSM A1912). MEM. AMER. ENT. SOC., 35 156 XYSTODESMID MILLIPEDS SOUTH CAROLINA. — Oconee Co., 15.5 mi. N Walhalla, Ellicott’s Rock Wilderness Area Trail at Fish Hatchery off SC hwy. 107, Sumter Nat. For., M, F, 9 June 1978 (NCSM A2055). GEORGIA. — Habersham Co., Clarksville, M, 27 April 1943, W. Ivie (RVC) TYPE LOCALITY. White Co., 7.2 mi. E Cleveland, along GA hwy. 255 at Chattahoochee R. and Habersham Co. line, 3M, F, 16 April 1978 (NCSM A1862). Fannin Co., 11.2 mi. S Morgan- ton, Deep Hole Campground off GA hwy. 60, Chattahoochee Nat. For., M, 9 July 1978 (NCSM A2348). 144 Fics. 141-144. Deltotaria lea. 141, process of 4th sternum of holotype, caudal view. 142, gonopods in situ, ventral view of male from Gaston Co., NC. 143, left gonopod of holotype, medial view. 144, the same, lateral view. Scale line for fig. 142 = 1.00 mm; line for other figs. = 1.00 mm for each. R.M. SHELLEY & D.R. WHITEHEAD 157 Remarks. — Chamberlin (1949) and Hoffman (1961) reported that the acropodite of b. philia was apically hastate, but this appearance is actually caused by a profile view of the lateral flange. Deltotaria brimleii intergrades Males from the following locality possess coxal processess that extend beyond the level of the prefemoral process and are intermediate in length and shape between the two subspecies. The site is also somewhat in- termediate, and I consider the specimens intergrades. Sample data are as follows: TENNESSEE. — Sevier Co., 7.3 mi. S Gatlinburg, GSMNP, Chimney’s Picnic Area, 5M, F, 10 May 1978 (NCSM A1890). Deltotaria lea Hoffman Figs. 141-144 Deltotaria lea Hoffman, 1961:33-34, Figs. 1c, 3a. Filka and Shelley, 1980:29, Figs. 55-56. This species was known only from the type locality when it was described and illustrated by Hoffman (1961). Filka and Shelley (1980) added supple- mental notes on color, gonopodal variation, and reported a number of localities in North and South Carolina. No new material is available, but the pregonopodal sterna are characterized in more detail, and the gonopods are described in ‘‘sigmoid’’ terminology. Type specimen. — Male holotype (NMNH) collected by L. Hubricht, 19 April 1956, 4.5 mi. SE Lincolnton, Lincoln Co., NC. Diagnosis. — Color of paranota and metatergal stripes yellow, with sparkling metallic sheen; coxal apophysis in profile short and triangular, terminating below level of prefemoral process; acropodite distally broad and spatulate; medial flange extending nearly entire length of basal zone, greatly expanded and curved to form concave inner surface of this section; lateral flange absent. Color in Life. — Paranota yellow; metaterga black with broad yellow stripes along caudal edges connecting paranotal markings; collum with stripe along anterior margin; colors with sparkling, metallic sheen. Holotype. — Process of 4th sternum moderately long, slightly shorter than widths of ad- jacent coxae (Fig. 141); 5th sternum with low projections between anterior legs and flattened, elevated areas between 5th; 6th sternum convexly recessed along caudal margin to accom- modate apical curvature of acropodites, 7th legs set slightly farther apart than 6th. Gonopodal aperture ovoid, 3.8 mm wide and 1.4 mm long at midpoint, indented anteriolaterad, sides elevated above metazonal surface. Gonopods in situ (Fig. 142, not this specimen) with acropodites leaning mediad and apices overlapping in midline of aperture. MEM. AMER. ENT. SOC., 35 158 XYSTODESMID MILLIPEDS Gonopod structure as follows (Figs. 143-144): Coxal apophysis short, narrowly triangular, terminating below level of base of prefemoral process, sides tapering rapidly to acuminate tip. Prefemoral process short, uncinate, directed toward distal zone. Acropodite broad, medial and lateral surfaces expanded through peak (proximal 2/3 of length) into curved laminae, arch high and rounded, overhanging and extending beyond level of prefemoral process; basal zone moderately long and gently curved, inner surface concave; anterior bend broad, poorly de- fined; continuous through peak with apical curve; peak gently curved, apex at midlength; apical curve broad, poorly defined; distal zone flattened and spatulate, moderately long and directed downward from peak, coplanar with basal zone, flared outward apically; tip subacuminate. Medial flange a broad, curved lamina, located entirely on basal zone, arising proximally and terminating on distal extremity. Lateral flange absent. Prostatic groove cross- ing to lateral side at anterior bend, continuing to terminal opening. Female from Gaston Co., NC. — Length 34.2 mm, maximum width 8.8 mm, W/L ratio 25.7%, depth/width ratio 67.0%. Cyphopods with small receptacle, surface finely granulate. Valves small, equal, surfaces finely granulate. Variation. — Filka and Shelley (1980) observed that gonopodal variation included the presence or absence of a minute prefemoral process, dimension of the tip of the distal zone, and diameter of the acropodal arch. In recheck- ing all individuals of /ea, I find these features to vary randomly throughout the range. The prefemoral process is present on only four individuals: the holotype and one each from Cleveland County, North Carolina, and York and Cherokee counties, South Carolina. Similarly, occasional random males have longer distal zones that extend farther from the peak. All, however, have short coxal processes that terminate below the level of the prefemur and broad medial flanges that extend nearly the entire length of the basal zone. I also checked for facial setae in males of /ea. All lack epicranial and in- terantennal series, about half have frontal setae (1-1), and nearly all have genal setae, usually one per side but occasionally 2-2 or 3-3. Three in- dividuals have a short row of three marginal genal setae continuing from the merged clypeal and labral series. Ecology. — Deltotaria lea occurs in predominantly deciduous woodlands under thin layers of leaves on relatively hard substrates near water sources. Distribution. — An area about 70 miles long in the Piedmont Plateau of the Carolinas between the Catawba and Broad Rivers, being particularly abundant in the Kings Mountain region on the border of the two states. Localities are detailed in Filka and Shelley (1980). Remarks. — In addition to color and gonopodal differences, /ea is alsoa larger, more robust species than its congener. The paranota project more directly laterad in /ea, interrupting the slope of the dorsum and imparting a somewhat flattened appearance. In contrast, brim/eii is more vaulted, and the paranota are more continuous with the slope of the dorsum, extending sharply downward toward the substrate. R.M. SHELLEY & D.R. WHITEHEAD 159 - Because of its peripheral occurrence in the state, Filka and Shelley (1980) recommended that /ea be assigned to the Of Special Concern conservation status in North Carolina. ParT III. EcoLoGy By Rowland M. Shelley HABITATS. — The characteristics of the cove environments inhabited by species of Sigmoria s. lat. were detailed by Shelley (1981a). Theses sites also are preferred by ainsliei, aphelorioides, and prolata (Falloria), the haerens group (Cheiropus), the subgenus Dixioria (wherever the habitat oc- curs in the more widely spaced mountains north of the Nolichucky River), and Deltotaria brimleii. I have collected the nominate subspecies of trimaculata (Rudiloria) in New York and West Virginia, both times in places with cove features. However, it remains to be seen how applicable this habitat is to the subgenus Rudiloria as a whole. From what Hoffman (pers. comm.) related about the site where he discovered whiteheadi (Sigiria), it also appears to be a cove species and, along with truncata (Sigiria), one of only two that can occur in rhododendron litter. In the piedmont and Atlantic lowlands, Deltotaria lea, the subgenus Cleptoria, and persica of the planca group (Cheiropus) are found mostly in predominantly hardwood areas under thin layers of leaves on relatively firm substrates near water. This environment also is preferred by the subgenus Croatania, although these forms frequently are encountered in predominantly or exclusively pine woods. Farther south and east, australis, planca, and serrata (Cheiropus) occur in a variety of predominantly hard- wood sites, which include the best spots available in the generally un- favorable coastal environments. West of the Appalachian Mountains, the forms in the Cumberland Plateau and Nashville Basin Provinces of Tennessee occur in a variety of biotopes. Cove environments are widespread in the former, especially north of highway I-40, but surprisingly they are shunned by picapa (Falloria). My assistants and I searched in vain for around 40 man-hours in coves near Wartburg, Frozen Head State Park, and other areas in central Morgan County before finally encountering one male and one female in oak woods near the type locality. These were in a different microhabitat, under deep piles of leaves well removed from water. Clearly, picapa has ecological preferences different from those of cove dwelling congeners. However, for- ficata (Falloria), occurring north and south of picapa, does occur in coves, and the same is true for abbreviata (Falloria). Farther south, houstoni (Falloria) prefers dry, open areas but still seeks shelter under thin layers of MEM. AMER. ENT. SOC., 35 160 XYSTODESMID MILLIPEDS leaves on relatively firm substrates. Cove habitat becomes steadily rarer westward into central Tennessee, and the species of the mimetica group (Falloria) therefore opt for the best hardwood sites available, with the more secluded and moist ones being preferred. Certain areas of central Tennessee resemble those in piedmont Georgia and South Carolina, and the millipeds likewise are found in the same kind of microhabitats along banks of streams or in moist spots. Exceptional individuals occasionally are encountered in xeric environments, as in Cedars of Lebanon State Park, Wilson County. SYNTOPY. — I (1981a) previously reported that only one species of Sigmoria s. lat. occurs at a given site, and until recently I had not en- countered any syntopic apheloriine species, even among different genera. The first instance was the discovery of australis and serrata (Cheiropus) together in Camden County, Georgia (Shelley 1984a), wherein australis was reported as an undescribed species of Hubroria. These two species ap- parently avoid competition by being dominant at different times of the year. The same may be true for australis and Dynoria medialis Chamberlin at Kolomoki Mounds State Park, Early County, Georgia. A third example may involve rubromarginata (Sigiria) and Deltotaria brimleii at Cosby Pic- nic Area, GSMNP, Tennessee. I collected the former there in May 1978 and the latter in exactly the same spots in August 1981, so visits in June and July may yield both. Another example of syntopy came to my attention in July 1984, when Dr. Hoffman sent a male of whiteheadi (Sigiria) and one of /. latior (Sigmoria) which were taken within a few feet of each other at the type locality of the former. Because /atior is a ubiquitous species not restricted to cove environments, it does not occupy the same niche as whiteheadi. There still are no clear examples of syntopy between cove dwell- ing congeners, which is in accordance with Gause’s theorem that two species cannot indefinitely coexist if they have identical ecological requirements. However, a possible example of syntopy among cove species involves cor- onata and brooksi (Dixioria) in Johnson County, Tennessee. More precise knowledge of the habitats of these species is needed before this can be con- firmed. Fic. 145. Distribution of Sigmoria. Known discontinuities in the southern half of the main part of the range are shown, but a smooth curve is drawn around the range extremes north of the isthmus in southwestern Virginia. Discontinuities may exist in West Virginia, Ohio, Penn- sylvania, and other northern areas, but insufficient sampling has taken place to detect them. The ranges of the groups in the Cumberland Plateau and Nashville Basin of Tennessee, while not known to abut or overlap, are very close and hence are drawn as a single unit. From east to west, the disjunct southern areas correspond to planca and populations of australis, with one Alabama population of ri/eyi occurring on the northern periphery of the central one. The dot in north-central Alabama corresponds to the Jefferson County record of rileyi. R.M. SHELLEY & D.R. WHITEHEAD 161 MEM. AMER. ENT. SOC., 35 162 XYSTODESMID MILLIPEDS ParT IV. DiIsTRIBUTION By Rowland M. Shelley DISTRIBUTION OF SIGMORIA. — With the inclusion of the new speices and those in the synonymized genera, the distribution of Sigmoria s. lat. is enlarged to encompass the shaded areas in figure 145. Overall, the genus ranges from northern New England and southern Ontario to northern peninsular Florida, and from the Atlantic Ocean in South Carolina and Georgia as far west as central Tennessee and central Alabama. A smooth curve is drawn around range extremes in the north, where sampling has been sparse, but known discontinuities are shown from southwestern Virginia southward, where the greatest species diversity occurs and where sampling has been more intensive. The two areas are divided by an isthmus in southwestern Virginia, the northern area being occupied by Rudiloria and by one species each of Dixioria and Sigmoria s. str., and the southern by the rest of the genus. Physiographically, the main part of the generic range spans the St. Lawrence, Connecticut, Hudson, Delaware, Potomac, Ohio, New, Cape Fear, Yadkin, Catawba, Savannah, Altamaha, and St. Marys Rivers, taking in parts of the New England, Adirondack, Ridge and Valley, Blue Ridge, Piedmont Plateau, and Coastal Plain Provinces. In central Tennessee, an apparently disjunct area in the Nashville Basin and Cumberland Plateau is occupied by members of the mimetica, picapa, and translineata groups (Falloria). This area is separated from the main one by the Tennessee River and probably represents a real distributional gap. Repeated searches for Sigmoria in the northern half of the hiatus in Ander- son, Morgan, and Roane counties yielded only Brachoria. Farther south, the Tennessee River Valley in Rhea, Meigs, and Bradley counties has been chiefly cleared for pasture or other agricultural uses, and not even Brachoria is found there now. Obvious disjunctions also occur in the extreme south, involving members of the subgenera Cheiropus and Cleptoria. The most notable gap is the ap- parent absence of Sigmoria s. lat. from most of the southern half of Georgia. Both australis (Cheiropus) and rileyi (Cleptoria) occur in three dis- junct populations, two of which are mostly or completely in Alabama (Figure 68). Sigmoria (Cheiropus) planca is the southernmost species and the only one in this area wholly detached from the main generic range. DISTRIBUTIONS OF SPECIES AND SPECIES GROUPS. — Figure 125 depicts the distributions of ainsliei, aphelorioides, and prolata (Falioria), and figure 147 shows occurrences within the GSMNP. Sigmoria aphelorioides occurs south of the GSMNP and the Little Tennessee River, whereas ainsliei is in the Park and the adjacent Ridge and Valley Province. Of the ten ‘“‘sigmoid’’ species known from the Park, rubromarginata R.M. SHELLEY & D.R. WHITEHEAD 163 (Sigiria) is the dominant form in the northern section, north and east of US highway 441. Deltotaria brimleii also is abundant, and both species occur as far south as Gatlinburg and the Deep Creek area, on the west and east sides, respectively. South and west of this highway the fauna is dominated by members of the subgenus Falloria. The distributions of species of Falloria in the Tennessee disjunct area are shown in figure 126. The picapa and translineata groups are restricted to the Cumberland Plateau, whereas the mimetica group is concentrated in the Nashville Basin but spreads onto the Highland Rim on either side. The sam- ple of mimetica from Cheatham County is the westernmost of the genus Sigmoria. The fauna in the lowlands east and south of the Blue Ridge Province is so complex that, for clarity of presentation, it must be shown on two maps (for distributions of the haerens and planca groups, see Shelley 1982, 1984a). Figure 68 depicts locality records for australis and the subgenus Cleptoria, and figure 49 shows records for Croatania. Fic. 146. Distribution of Del/totaria. Dots, b. brimleii; squares, b. philia; X, brimleti intergrades; stars, /ea. The dashed line denotes the approximate boundary of the Blue Ridge Province. MEM. AMER. ENT. SOC., 35 164 XYSTODESMID MILLIPEDS Sigmoria (Cheiropus) australis and rileyi (Cleptoria) occur in three dis- junct areas. The westernmost sample of ri/eyi, from the southern extremity of the Cumberland Plateau, is some 40 miles north of the closest record of australis, in the Fall Zone. Sigmoria rileyi also occupies the northern periphery of the Chattahoochee segregate of australis. This river bisects the latter area and does not form a distributional limit as it does for Dynoria medialis (see Shelley 1984b). The eastern segregate of australis hugs the Atlantic Ocean in South Carolina and Georgia, spanning the Savannah, Ogeechee, Altamaha, and Satilla Rivers. It is unknown south of the St. Marys River, but as its distribution in coastal Georgia is similar to that of serrata, australis may also occur in northeastern Florida. Most of the records shown for the subgenus C/epforia in figure 68 are along the Savannah River, where field efforts were concentrated. Although rileyi undoubtedly is more widespread, it probably does not occur much farther west because the area around Atlanta has been sampled reasonably thoroughly. In central Georgia the ranges of rileyi, persica (Cheiropus), and Dynoria medialis are almost exclusive (see Shelley 1984a, b), suggesting that they occupy similar niches. The actual distribution of bipraesidens may parallel that of the South Carolina species, robusta, in ranging nearly to the Blue Ridge Front. Sigmoria (Cleptoria) abbotti hugs the southern side of the Savannah River in both the Piedmont Plateau and the Coastal Plain, but the South Carolina species are limited to the Piedmont south of the Enoree River. The distributions of shelfordi and arcuata are as stated previously (Shelley 1980a, 1981b), except the Oconee County male is reassigned to the former. Sigmoria macra occurs in a band north of these species, and robusta occupies a somewhat triangular area in the western tip of South Carolina. An updated distributional map for the subgenus Croatania (Figure 49) in- cludes localities listed in Shelley (1977) and new ones cited herein. The most significant additions are those for simplex, formerly known only from the type locality. This site is near the northern range limit, and the species ac- tually extends well past the Fall Zone into the central Coastal Plain. New records for saluda extend its range eastward through the Fall Zone to the in- ner edge of the Coastal Plain. Range additions to catawba are minor and areas where it was expected from previous records. Sigmoria (Croatania) yemassee, still known only from the two original collections in Jasper and Beaufort counties, and australis (Cheiropus) are the only two ‘‘sigmoid’’ species in the southern tip of South Carolina, where even /atior is absent. The subgenus Dixioria (Figure 45) occupies a small area around the con- tiguous corners of North Carolina, Tennessee, and Virginia, from Tazewell County in Virginia southward to the Nolichucky River. It is known only R.M. SHELLEY & D.R. WHITEHEAD 165 from the Blue Ridge Province in North Carolina and Tennessee, but in Virginia coronata spreads northward into the Ridge and Valley Province. Sigmoria (Dixioria) coronata, pela, watauga, and wrighti have sharp distributional limits, but only a few records are available for acuminata and brooksi. Further exploration may expand their boundaries in Tennessee and Virginia; dactylifera, however, is restricted to central Ashe County, North Carolina. Aside from the subgenus Dixioria, comparatively little is known about the forms of Sigmoria s. lat. occurring north of North Carolina. Vast areas remain to be investigated in West Virginia, Kentucky, Ohio, and Penn- sylvania. Available records, plotted in figure 17, reveal clustering of /. /atior (Sigmoria) and t. kleinpeteri (Rudiloria) in southwestern Virginia, the only region to have received much attention. Father north, the plotted ranges of guyandotta, rigida, and mohicana (Rudiloria) are subject to future expan- sion; however, the scattered records of ¢. trimaculata (Rudiloria) probably outline its area fairly accurately. The absence of the genus from piedmont Virginia is curious, and the recent discovery of whiteheadi (Sigiria) along the escarpment in Patrick County may indicate more surprises in lowlands to the east. Explorations of this area are left to future investigators, and clues to potentially productive environments are available in the ecological accounts here and in Shelley (1977; 1980a; 1981a, b, c; 1984b). Sevier Ca Cocke Co, é / ) - ia i | Ss 4 ¢ ‘ SY en y NEE i sf Tenneslseu y jp ~) > / A ~ North Caroiina N=a=— 3 o\x ke N (Po Swain ¢ in Co Haywood vo ; 147 Fic. 147. Distributions of species of Deltotaria and Sigmoria in the Great Smoky Moun- tains National Park and vicinity. Diamonds, rubromarginata; stars, tuberosa; circle, nan- tahalae; ovals, translineata; squares, lyrea; solid triangles, fumimontis; half-shaded dots, ainsliei; dots, bidens; asterisks, prolata; open triangles, D. b. brimleii; open squares, D. b. philia; X, D. brimleii intergrades. MEM. AMER. ENT. SOC., 35 XYSTODESMID MILLIPEDS 166 a Lt suapisap.iaiq a 97 1710qqD a + 4 idajid a dG aasspwuadk + oF €7 xajduis a (ZG ppnyps a IZ pqmnjpo a 0Z DSOAANIIJD] a 6I pipiponb a 81 pqojouajs + + + + + LI-vl 401)0] a el psa fn djo0p a a IsyOosq a Il Yysidm a Ol psnvjom + + 6 pIJDUOLOD aq 8 pjoulwinoD a L pied a 9 ppisit a ¢ pupodiyour a V pyjopupdns + + + €-1 DIDINIDUY sneoje[d sneole[d Kaye aspre nesjeld ule|d yoepuolpy (ZSI-IST “Sst uOXxe | MOT JoUojUT = ueyoeeddy 29 o3ply ong JUOWIpsld [eiseoD purjsuq MeN ‘€ 2192.1) “ON -goutaoid ay} 0} stwWepug J “gouIAOId oY} UTYIIM soueTINIIOQ + sQoUTAoIg otydessorshyd AQ DUOWSIG JO Setadsg JO uoNNgINsIq “VY EGCOAS 167 R.M. SHELLEY & D.R. WHITEHEAD Omowmm + aro oaw wy Dawa wa Qawawa + OY gs vs €s (6S igs 0s 6r-9V SP Ag 4 (4 Iv OV 6€ 8€ LE 9€ ce ve cf cE Te O€ 67 87 pjountsip uosouajs DIDIOaID SOP1OLM 31S pIpIUNA] DIDULOUL SIJUOW1I31U DIDjNsuv14y Ippayojlym SIJUOW1LJSND DIDUIBADWIOAGNA Do1ssad pounjd DJDIIOS S1]SALBD xDUuly] DIDIIIDAIP suasany Suas.adAlp Dijoydosnqis S1]D4jSND DADDUL ip4sofjays DIDNIAD DISNqOL MEM. AMER. ENT. SOC., 35 XYSTODESMID MILLIPEDS 168 (saoutaoid €) nypjnoDUI147 ‘(SaoulAodd ¢) 400] :Soroeds poinqtsisip Ajapim Apso %9'V % 8° OT = % 9° VV %o°8T %79 = snus jo Jusd19q € L 67 a V = saioads o1wiepug jo Joquiny] 9° %P ST %LL BES %L LT %8 OT %S'T snueD jo JUsdI9g € Ol ¢ ce 81 L I satoodg jo Joquinn a €L DSOAANIISSDAD a ZL pjojnpuad a IL payawuiuU a OL pdpoid a 69 pIpIAadqqno a 89 1uojsnoy a L9 pon fiof a 99 psoiaqn] a 69 pjpjoid a v9 suapiq 4 ar £9 1a1)SUID a (42) pad] a 19 syuoumuny a 09 DIDaUI|SUDA] a 6S sapio1sojaydp + ar gs pyjcydosax a LS DIDIAJSOMNA] a 9¢ apjpynjunu sneo}e|d sneoje|d Kaye A. a3ply nesield uleId yoepuompy (ZSI-IST “Sst uOXxe |, MOT JO119}U] ueiyorjeddy 9 O3plEy ong yUOWIpald eqiseoD puelsuq MON ‘€ 91G2L) “ON ‘g0uIAOId ay} 0} StWapugq J “Q0uTAOId OY) UIYIIM doUeLINIIO + (panurjuog) seouraoig s1ydeisorshyd 4q DIOUISIg yO soads Jo uONNgINSIq “p ATAVL R.M. SHELLEY & D.R. WHITEHEAD 169 PHYSIOGRAPHIC PROVINCES AND ENDEMISM. — Table 4 lists the species of Sigmoria s. lat. by physiographic province (Hunt 1967), in- cluding also the extent of endemism. Immediately apparent is the preponderance of total taxa and endemics in the Blue Ridge Province, in both cases more than those of the two adjacent regions combined. Slightly over half the total species are represented at least partly in the Blue Ridge, and well over 1/3 of the endemics occur there. The Piedmont and Ap- palachian Plateaus rank second and third in both categories, and there are no endemics in the Ridge and Valley Province. All three species in the In- terior Low Plateaus of central Tennessee are endemic. The southeastern Coastal Plain has four endemics, which are of interest in view of the flat topography and generally unfavorable habitats. Although disjuncta (Sigiria) is listed from two provinces, it occurs along the base of the escarp- ment in South Carolina and Georgia and does not really span the Blue Ridge and Piedmont Plateau regions. In contrast, simplex (Croatania) clearly spans the Fall Zone boundary between the Piedmont Plateau and Coastal Plain in South Carolina, inhabiting sizeable portions of both regions. Table 4 also shows the great adaptability of /atior (Sigmoria), which occurs in five provinces of vastly different environments; the large distribution of trimaculata (Rudiloria) does not cover nearly as great a range of biotopes. DISTRIBUTION OF DELTOTARIA. — The distributions of the species of Deltotaria are shown in figure 146. The strict correlation of brimleii with the boundaries of the Blue Ridge is clearly evident, although southern populations occur along the edge of the escarpment in Georgia in a manner similar to S. (Sigiria) disjuncta. Approximately 55 miles separate the two allopatric species, both of which are sympatric with several species of Sigmoria s. lat. Part IV. COLORS AND COLOR PATTERNS By Rowland M. Shelley and Donald R. Whitehead One intriguing aspect of Sigmoria s. lat. is the complex variation in color and color pattern. The first observation a collector makes is the color of his specimens, and Table 1 lists the diversity of hues and arrangements. Colors and color patterns tend to be ignored in xystodesmid systematics, but we find their geographic patterns extremely informative. COLOR PATTERNS. — The three general color patterns cluster geographically as shown in figure 148. The northern 2/3 of the range is mostly comprised of the middorsal spot pattern, which overlaps that of the uniformly black metaterga (paranotal spots only) in southwestern Virginia. MEM. AMER. ENT. SOC., 35 170 XYSTODESMID MILLIPEDS 4 fe | re) ee les a au: evan i ny any? ay cave ei 35 aes (ey TT Hf TA in aa vs * il ae Se] ] Sip a Wve wees SS 0 aes naa \ NN SBcr aS is N) es RAN Ct ae JZEES Ave EN yy aes vA ay ] A$ im thi Wj YJ NN Seas ay jamin eel aa Ly i ERS lig | 1 wed ae a ili middorsal metatergal spots i Hg Ons sy is Sle x ; -(\ uniformly black metaterga metatergal stripes Fic. 148. Approximate distributions of the color patterns of Sigmoria. R.M. SHELLEY & D.R. WHITEHEAD 171 The latter pattern extends broadly into North Carolina, narrows into South Carolina where it overlaps forms with metatergal stripes, then widens across the Savannah River into piedmont Georgia, where it overlaps a finger of the red stripe pattern and abuts on an isolated population of the latter. The metatergal stripe pattern, occurring in Tennessee and throughout the southern section of the Blue Ridge Province, extends southeastward into South Carolina where it overlaps forms with uniformly black metaterga. The distribution of the striped pattern widens in the Coastal Plain, extend- ing northward back into North Carolina and southward to the southern ex- tremity of the main generic range. Except for the area in northern Alabama and the northern and southern extremities of that bisected by the Chat- tahoochee River, the striped pattern is also the one found in the southern disjuncts. The peripheral discontinuities in the Chattahoochee segregate represent the southernmost samples of australis (Cheiropus) and the Lee County population of rileyi (Cleptoria) in the north, both with uniformly black metaterga. The ‘‘unstriped’’ pattern in northern Alabama also reflects rileyi, and additional field work is needed to determine the size of the Jefferson County area. Striped forms omitted from this map include some variants of f. trimaculata (Rudiloria) from West Virginia; whiteheadi (Sigiria) from Patrick County, Virginia, at the eastern periphery of the uniformly black metatergal pattern; and mohicana and rigida (Rudiloria), which probably are striped as judged from original descriptions and faded preserved material. Other omissions result from the lack of clear information about the patterns. Sigmoria (Cleptoria) bipraesidens, a member of an otherwise unstriped subgenus, is said to be striped (Hoffman 1967) and occurs at the range periphery of the unstriped taxa, but this needs confirmation. The col- or pattern of yemassee (Croatania) is not known, but this coastal species occurs in an area occupied by known striped species and probably also displays this pattern. With the exceptions noted above, color pattern distribution in Sigmoria s. lat. can be generalized as follows: the metatergal spot pattern takes up much of the northern 2/3 of the range, and the uniformly black metaterga and metatergal stripe patterns share rather equally the southern 1/3, the area with the greatest species diversity. In the latter, the unstriped pattern tends to occur in the northern and central parts, and the striped pattern more to the west, east, and south. Sympatry between the latter two patterns is in a relatively broad band, and elsewhere the patterns tend to abut with geo- graphically proximal borders. COLORS. — The metatergal stripe pattern can be subdivided by colors (Figures 149, 150). As with the general color patterns described above, the MEM. AMER. ENT. SOC., 35 172 XYSTODESMID MILLIPEDS colors of the stripes tend to cluster and are not randomly dispersed. For ex- ample, blue stripes (Figure 149) tend to occur to the west of red stripes, predominating in central and southeastern Tennessee. Red stripes are found over a much larger area including the southern disjuncts. Sigmoria (Cheiropus) planca, the southernmost species, is omitted from figure 149 because its stripes are more orange than red (Shelley 1984a). However, it does bridge the wide red/red hiatus along the Georgia/Florida boundary. Two species of Falloria tend to have white or whitish stripes (Figure 150): the northern one, /eucostriata, has concolorous paranotal spots, whereas the southern one, nantahalae, has red paranotal markings. In the central Smokies, the red, white, and blue stripe patterns overlap, the red f : ie aS Af ey iy (\ SNe C] Co RSS tee erence ey Dh CHa : aes ea a a ye eS Plt VScmei Wepmpjend CRB ane = we vate — — Ls 149 = Lat == ted stripes blue stripes 2 : yellow stripes Gs or e: ‘ia af SHES Pig ay Fic. 149. Approximate distributions of the red, blue, and yellow metatergal stripe colors in Sigmoria. The dots in Buncombe County, North Carolina, and Patrick County, Virginia, show isolated locations of yellow stripes in areolata and whiteheadi, respectively. R.M. SHELLEY & D.R. WHITEHEAD 173 represented by rubromarginata (Sigiria) and the other two colors by species of the subgenus Falloria. South and east of nantahalae, disjuncta and stenogon (Sigiria) tend to have concolorous violet or purple paranotal spots and metatergal stripes. The separated distribution shown in figure 150 is somewhat misleading, because intervening populations of divergens (Cheiropus) also tend to be dark. Yellow stripes occur mainly in a band in the Carolinas partly occupied also by red stripes. Small pockets of yellow stripes represented by forms of Sigiria are isolated elsewhere in the red stripe area (areolata, inornata) and in the uniformly black metatergal area in Virginia (whiteheadi). Sigmoria (Sigiria) areolata and whiteheadi are known from point distributions and are not known to be polymorphic for color, but inornata is polymorphic and may display red, orange, or yellow stripes throughout its range. IMPORTANCE OF COLOR AND COLOR PATTERN. — Among the eastern Xystodesmidae, most genera of the Rhysodesmini and all genera of the Apheloriini have bold colors on a black background, whereas none of the Nannariini or Pachydesmini display striking contrasts. Some boldly pat- terned forms apparently are mimetic, as judged from the occurrence of similar colors among sympatric taxa representing distinct lineages. For in- ip ee FERS white iil — —— — = — —_— Fic. 150. Approximate distributions of the white, yellow, and violet/purple metatergal stripe colors in Sigmoria. The dotted and dashed lines denote boundaries of the blue and red stripes, respectively. The dot in Buncombe County indicates the location of the yellow striped areolata. wat MEM. AMER. ENT. SOC., 35 174 XYSTODESMID MILLIPEDS stance, blue stripes appear in the same geographic areas in both Brachoria and Sigmoria s. lat., whereas forms of Brachoria in the range of the subgenus Dixioria display the yellow paranota/black metaterga of the lat- ter. In other instances, sympatric taxa differ in colors; for example, the red striped Deltotaria brimleii overlaps various blue striped species of Falloria in the southern Blue Ridge Province. Farther north, however, it is broadly sympatric with the red striped rubromarginata (Sigiria). Geographic distributions of the various colors and patterns considered together with probable instances of convergence suggest informative transformation series useful for phyletic analysis. This information is prob- ably as valuable and as easy to interpret as gonopod structure, and thus equally important in developing a system of relationships within Sigmoria s. lat. The combination of color and geography is also useful in field iden- tifications once distributions are well mapped. Some species have distinctive patterns within their areas of distribution, so previously unassignable females may be identifiable if colors and provenance are known. Table 3 facilitates such determinations. ParT VI. RELATIONSHIPS WITHIN SIGMORIA By Donald R. Whitehead and Rowland M. Shelley In this part, we outline analytical problems encountered in the study, discuss details of the subgenera of Sigmoria, resolve subgeneric relation- ships, and close with a discussion of ‘‘mosaic’’ evolution. We use the term “‘mosaic’’ only for observed geographic patterns found in apparently natural groups of primarily allopatric and parapatric species-group taxa which seem to lack cladistic congruence in their various character states. This kind of pattern in millipeds and in other groups of comparatively non- vagile animals resembles a jigsaw puzzle. Some other terms and notations deserve explanation. Color pattern is usually denoted by a fraction, wherein ‘‘red/red’’ means red paranotal spots and red metatergal stripes, and ‘‘yellow/black’’ indicates yellow paranotal spots and no metatergal stripes. The term ‘‘eastern’’ refers to eastern North America, from the seaboard to the central Great Plains, in- cluding the Ozarks. Relationships are depicted by equations; for example, ““Croatania = simplex + (yemassee + (catawba + saluda))’’ means that within the subgenus Croatania, simplex is sister to the other three species, and yemassee is sister to catawba plus saluda. These cladistic statements, which are exactly analogous to mobiles, are given instead of pictorial trees. Use them with associated maps to visualize implied vicariance patterns. R.M. SHELLEY & D.R. WHITEHEAD 175 ANALYTICAL PROBLEMS. — With few exceptions, species diagnostic characters used in Sigmoria s. lat. are exclusively in gonopodal structures, and it is difficult to construct satisfactory transformation series. These are complex structures, yet some types of complexities may arise from comparatively simple alterations. Hence, apparently major gonopodal differences may be evidence for nothing more than species level differences. The main somatic features used herein are the relative length of the process of sternum 4 and color patterns. There is no reason to assume convergence in either gonopods or in sternum 4 if derived states of these characters occur in geographically coherent patterns of proximal taxa. We assume the reverse, that geographic coherence represents shared ancestral features. Color patterns must be viewed more critically, not only because of in- fraspecific variation, but also because we find evidence of mimetic con- vergence both within Sigmoria s. lat. and among related genera. Hence, color patterns also tend to be geographically coherent, but they less con- vincingly indicate common ancestry. The genus Sigmoria is a cohesive geographic entity (Figure 145) having properties resembling those of a single, rapidly differentiating or fragmen- ting species. This implies that the history of Sigmoria is quite brief, perhaps post-Pleistocene. We infer a continuous, widespread, geographically varied ancestor in rather recent times, which colonized habitats as they became available and subsequently fragmented. Such fragmentation is easy to envi- sion, since most species now occupy specialized cove habitats. Where the allopatric/parapatric mosaic pattern is retained, distributional shifts must reflect ancestral proximity patterns, with range expansion by one segregate achieved only by displacement of another. Present day distribution patterns directly reflect vicariance. We use the term “‘incomplete synapomorphy”’ for a peculiar form of homoplasy occurring in the ‘‘mosaic’’, when the same plesiomorphous and apomorphous states of a character are shared by both members of a sister pair and, by inference, their common ancestor. If characters in the ancestor varied along different geographic gradients, then fragmentation must have cut across character gradients in such a way that derived states of some characters were shared by parts of two or more segregates. Given low vagili- ty in these organisms, there is no reason to assume that the derived character states diffused among the segregates before further fragmentation occurred. These geographically coherent but noncongruent apomorphous character states (‘‘incomplete synapomorphies’’) suggest common ancestries, but they do not define lineages in diagnostic terms. Consider what happens when a pepperoni pizza is sliced into wedges, with the pepperoni representing apomorphies. Some pieces of pepperoni will be MEM. AMER. ENT. SOC., 35 176 XYSTODESMID MILLIPEDS divided between two adjacent wedges, some slices will have contiguous boundaries, and still other slices will share a point of contact. In Sigmoria, the subgenera represent the pie wedges and character state distributions represent the pepperoni. Radial slicing is represented by contact in the southern Blue Ridge by proximal, relatively plesiomorphous members of Cheiropus, Sigmoria, Sigiria, and Falloria. As the Sigmoria pie was sliced by nature, the simplest type of slicing is the most probable. Thus, the first slices should have been cut across the pie rather than in a wedge by wedge sequence. Ideally, a cladistic hypothesis based solely on morphological character states should be developed for Sigmoria, similar to the generic and tribal hypotheses given in chapter VIII. However, the numerous species, com- bined with apparent convergences, clines, and instances where variation crosses cladistic lines, create a geographic mosaic of reproductive isolates in which the taxonomic or geographic distributions of many character states lack congruence. Our initial attempts at cladistic analyses, which were made without considering geographic distributions, met with failure. We here describe a solution for this mosaic problem, which should apply also to similar problems in other organisms. We consider a character-based cladistic analysis premature at this time. Future investigations should focus on development of other character systems. Our approach to analysis is based on the observation that the genus seems to have some properties of a single species. We simply test as a null hypothesis the premise that the pieces of the mosaic interlock in a pattern consistent with the properties of a biological species. Hence, the species and subspecies are treated as pieces of a largely two dimensional jigsaw puzzle. Their geographic fit is conspicuous in most instances, thus allowing a search for geographic and character congruence. We did detailed mapping of the species and subspecies taxa, searched for both continuities and major discontinuities in the characters of geographically proximal taxa, and iden- tified as major components (i.e., subgenera) those clusters of taxa that are most strongly discontinuous from proximal taxa. With one notable exception, Sigiria, we are satisfied that our eight allopatric/parapatric subgenera are monophyletic. Remaining is the prob- lem of finding the one (out of 135, 135 possible dichotomous combinations) cladistic arrangement for the subgenera that best fits known facts of geographic and character distributions. The solution, again, is based on a combination of geographic and cladistic inference. In terms of vicariance partitioning, the subgenera are distributed in a pat- tern resembling an irregularly cut, eight slice pizza pie. Most slices are adja- cent to at least two others, and residual commonality exists in the boundary R.M. SHELLEY & D.R. WHITEHEAD 177 areas. The pie is sliced but virtually intact — there is little obvious extinction pattern except for gaps between conspecific populations. The slicing pattern itself may be obvious but incomplete. As in the pizza, this might not be noticed until one attempts to isolate the slice. Maps, figures, and tables both in this paper and in previous papers by Shelley form the basis for character analysis. Out-groups to determine polarity of character states are iterative, and hence polarities may differ among different groups and at different levels. THE SUBGENERA. — Each subgenus differs in structural diversity, diagnosis, and geography. Thus, some have single species groups, whereas others are more speciose. Some have satisfactory autapomorphies, while others are linked by transformation series or are grouped only by linkage patterns or by ‘‘incomplete synapomorphies”’ which do not apply to all in- cluded taxa. Most have only allopatric/parapatric taxa, but others have dis- junct and/or sympatric taxa. Except perhaps for Sigiria, we think that each subgenus is monophyletic, regardless of its internal composition and despite the lack of complete independence with respect to all structural features. Those features termed ‘‘incomplete synapomorphies’’ at the generic level, here are the opposite; they are plesiomorphies within the subgenus. We discuss each subgenus in turn, based on a preliminary cladistic hypothesis (Figure 154): genus Sigmoria = subgenera ((Rudiloria + Dix- ioria) + (Sigmoria + (Croatania + Cleptoria))) + Cheiropus + (Sigiria + Falloria). Undesirable features of this hypothesis are summarized and resolved in our review of subgeneric relationships. Figure 156 is the revised cladistic hypothesis. Rudiloria. — This subgenus includes only the trimaculata group, with five allopatric specific and subspecific taxa (t. trimaculata, #1; t. kleinpeteri, #2; trimaculata intergrades, #3; guyandotta, #4; mohicana, #5; rigida, #6) as mapped in Figure 151, areas 1-6. Detailed distributions are not yet available, and the species and subspecies hypotheses may not survive. The subgenus might even prove to be monotypic. This is the northernmost subgenus, grouped with the partly sympatric, more southern Dixioria by apomorphous gonopod orientation. Although divergent in details of gonopodal structure, notably the enlarged prefemoral processes and ac- cessory teeth in Dixioria, these two subgenera share other features perhaps indicative of common ancestry. Aside from sharper acropodal bends in rigida, there is resemblance be- tween it and pela (Dixioria) (compare Figures 15 and 20), although the range of ridiga is separated from that of Dixioria by more divergent forms of Rudiloria. Both species have acropodites of about the same thickness, and both have a similarly shaped tooth at about the same distance from the MEM. AMER. ENT. SOC., 35 178 XYSTODESMID MILLIPEDS tip of the distal zone. We consider this a shared ancestral condition between Rudiloria and Dixioria. Various features of rigida (metatergal stripes; prefemoral process present; acropodite short, not looped, without flanges, with tooth), but not the short process of sternum 4, represent a probable ground plan for Rudiloria. The thin, fragile acropodites of various forms of Fic. 151. Generalized species distributions in Sigmoria; numbers as in Table 3. ] TT R.M. SHELLEY & D.R. WHITEHEAD 179 TaBLE 5. Apomorphous character states in Rudiloria Character Taxon 1,WV 1 2 3 4 5 6 Acropodite looped x x x xX — — — lobed — — x _— x = — bent — — — — — x XX not toothed x x x x x x — Prefemoral process absent — — — — Ke x — Process of sternum 4 short — — —_— — Xi x x Metatergal spots (not stripes) — x x x x — — Taxa numbered as in Table 3: WV indicates West Virginia samples. Character states range from plesiomorphous (—) to most apomorphous (xx). Rudiloria and Dixioria are perhaps convergent in this respect with those of such taxa as sigirioides (which is geographically proximal to pela), stenogon, and triangulata (Sigiria). Some characteristics of members of Rudiloria are summarized in table 5. Presence of dorsal metatergal spots rather than stripes in guyandotta, t. kleinpeteri, trimaculata intergrades, and all but a few western ¢. trimaculata is unique within the genus. This color pattern may reflect relationship with Dixioria in which neither dorsal spots nor stripes are present. However, since the two subgenera are partly sympatric, the color pattern might repre- sent mimetic convergence. The process of sternum 4 is moderately long in both subspecies of trimaculata, an ancestral condition shared with most members of Dixioria. These two subspecies also differ from other Rudiloria, as well as from Dixioria, in having a strongly looped acropodite, a derived state. Two species, guyandotta and mohicana, lack a prefemoral process, another derived state. In the nominate subspecies of trimaculata the looped acropodite tapers smoothly and continuously to the tip, whereas ¢. kleinpeteri has expansions or lobes at the lowest point in the loop or midlength of the distal zone. In West Virginia, adjacent to the latter’s range, is guyandotta, in which the distal zone is expanded and extends a short distance below the peak. The lamella is smaller than the lobes of t. kleinpeteri but has the same general contours, and guyandotta may be merely a form of the latter with the distal zone shortened and the lobes reduced. However, the transition is sudden, with no intermediate forms such as occur between the two subspecies of trimaculata, and guyandotta also lacks the prefemoral process found in trimaculata. In Ohio, mohicana has nearly the same acropodal configura- MEM. AMER. ENT. SOC., 35 180 XYSTODESMID MILLIPEDS tion as guyandotta but with slightly sharper angles and without the distal zone expansion. In West Virginia, rigida has basically the same acropodite as mohicana but with still sharper bends, and it also has a tooth and a prefemoral process. Thus, there is a circular gradient from a looped acropodite without lobes or expansions (and with a prefemoral process), to one with flanges, to a shorter, non-looped one with flanges (and without a prefemoral process), to a short one with slight angles and without flanges, to a short one with strong angles (and with a tooth and a prefemoral process). Between ¢. trimaculata and ¢. kleinpeteri the character change is bridged by intermediates, whereas between ¢. kleinpeteri and guyandotta it is abrupt. More collecting is needed in Kentucky, West Virginia, and Ohio to determine the nature of other tran- sitions. The acropodal configuration in mohicana could connect directly with that in ¢. trimaculata, and the possibility of a clinal continuum is enhanced by the fact that West Virginia samples of ¢. trimaculata agree with mohicana in having metatergal stripes rather than dorsal spots. Just as the acropodite of guyandotta may be shortened from that of ¢. kleinpeteri, the structure in mohicana may be a reduction of that of ¢. trimaculata. The looped acropodites of trimaculata are convergent with those of ainsliei and aphelorioides (Falloria), approximately 125 miles to the south, and with Apheloria, whose range overlaps all three species. Other aspects of the acropodites of these forms do not exhibit such similarities. Dixioria. — This subgenus includes only the pe/a group, with seven closely spaced species taxa as mapped in figures 151-152, areas 7-13. These taxa share a distinctive body form, and they share with Rudiloria a distinc- tive gonopod orientation. The distribution of Dixioria overlaps that of Rudiloria, is adjacent to that of Sigiria, is overlaid by that of the ecological generalist /atior (Sigmoria), and is separated by a small gap (ca. 45 miles) from that of Croatania. Dix- ioria and Rudiloria share with Sigmoria s. str., Croatania, and Cleptoria some non-gonopodal features, including the tendencies to lack metatergal stripes and to have relatively long processes on sternum 4. Furthermore, the enlarged prefemoral processes of Dixioria and the more southern Croatania occur in close geographic proximity. An alternative relationship (Rudiloria, Dixioria, Sigmoria s. str., Sigiria, and Falloria) is suggested by geographic proximity of forms having relatively delicate acropodites. However, as we show later in this chapter, we now regard Dixioria and Rudiloria as a lineage sister to the rest of the genus. The most differentiated member of the pela group is dactylifera (#13), which has the process of sternum 4 short (a derived state) and lacks an ac- cessory tooth (polarity undetermined). If lack of an accessory tooth is R.M. SHELLEY & D.R. WHITEHEAD 181 ancestral, dactylifera is sister to the rest of the pe/a group. Otherwise, it. might be a derived element. In Dixioria the species are mostly parapatric with abrupt transitions, a prime reason for considering them reproductively isolated along with the apparent syntopy between brooksi (#12) and coronata (#9) in Tennessee. More sampling is needed in the transition areas between the forms in this state and Virginia to determine if changes occur as abruptly as in the heavily sampled areas in North Carolina. A trend towards a longer tooth and a longer distal zone from coronata (#9) to wrighti (#10) occurs in steps involving the geographically inter- mediate watauga (#11), which has an apomorphous, more distal accessory tooth. Thus, a clinal continuity in one respect is matched by a discontinuity in another, and recognition of watauga as a separate species is the most con- servative action consistent with the present knowledge. Dixioria may actual- ly contain as few as three species, and we treat it as a superspecies in which a detailed cladistic hypothesis is unwarranted. Sigmoria. — This subgenus (Figures 151-152, areas 14-20) includes the quadrata group with two adjacent piedmont species (quadrata, #19, and laticurvosa, #20) and the /atior group with two ecologically differentiated species (the widespread, polytypic /atior, #14-17, and the bipopulational stenoloba, #18). The two groups are associated by the flange on the peak of 7 7 7 WAR cher ALE 4 essere alee eee wiz, sa | Jo Ka aie Re ee Fic. 152. Generalized distributions of Sigmoria species in the southern Blue Ridge Province; numbers as in Table 3. MEM. AMER. ENT. SOC., 35 182 XYSTODESMID MILLIPEDS the acropodite and by geography: Sigmoria = (latior + stenoloba) + (quadrata + laticurvosa). A clear and convincing pattern of geographic races and intergrades is evi- dent in /atior, far more so than elsewhere in the genus wherein fragmenta- tion patterns are more complete. With its broad ecological tolerances, /atior is able to spread between localities and is found moving on the substrate more often than any congener. The ecologically generalist nature of /atior may represent an ancestral feature, if the ancestor of the genus was widespread and geographically varied. Sigmoria (Sigmoria) latior overlays the ranges of numerous other taxa. Its distribution clarifies the relationship between the quadrata and /atior groups, which otherwise would be disjunctive. It also helps explain the curious, disrupted range of stenoloba, which lacks metatergal stripes. If this condition were viewed as synapomorphous, the only possible ancestor for stenoloba would be /. /atior; thus, a species would be derived from a sym- patric subspecies. However, lack of metatergal stripes in /. /atior and stenoloba is shared with various members of Rudiloria, Dixioria, Croatania, and Cleptoria. Also, the gonopodal flanges of /atior (s. lat.) are more developed than those of stenoloba, an apomorphic condition arguing against shared ancestry between /. /atior and stenoloba. Hence, we accept a sister relationship between /atior (s. lat.) and stenoloba, with the disrupted range of the latter resulting from displacement by the more successful /atior (Shelley 1981a). Sigmoria (Sigmoria) stenoloba is known from Wilkes and Catawba counties, North Carolina, its two allopatric populations separated by about 40 miles and distinguished only by minor gonopodal differences. Certain taxa of Sigiria (notably, whiteheadi, areolata, and some in- dividuals of inornata) have yellow paranotal spots and yellow metatergal stripes as occur in some populations of /. munda (Sigmoria), but they other- wise most closely resemble various red/red forms of Sigiria. We consider yellow/yellow as an ‘‘incomplete synapomorphy’’, given a probable trans- formation series of red/red to yellow/yellow to yellow/black, because red/red and yellow/yellow occur in both Sigmoria s. lat. and Sigiria. Mimetic convergence is unlikely. For example, whiteheadi structurally fits best with austrimontis (Sigiria) but is syntopic with yellow/black /. /atior. It seems to share a distant common ancestry with /atior (s. lat.) but to have re- tained an intermediate, ancestral yellow/ yellow coloration. It might seem to represent a range extrusion from austrimontis which failed to achieve the /. latior color pattern, but a range extrusion is incompatible both with a sim- ple fragmentation pattern and with the major range disjunction between austrimontis and whiteheadi. The disjunctive whiteheadi represents a rem- nant of an ancestral yellow/ yellow continuum. —————— eS e R.M. SHELLEY & D.R. WHITEHEAD 183 Both members of the quadrata group occur within the range of saluda (Croatania). This is one of rather few examples of secondary sympatry among ecologically restricted species. Both species of the quadrata group have metatergal stripes, and we infer that their ancestry was to the southeast of the range of the ancestral component which iacked stripes. The more widespread saluda is thus the incursive element. If the ancestry of the quadrata group was to the southeast of that of Croatania, then the subgenus Sigmoria itself was ancestrally eastern. The subgenus Sigmoria is characterized by the reflexed acropodite tip, a feature also found in some samples of arcuata, which we think is a sister ele- ment within Cleptoria and which therefore might represent ancestral condi- tions found in Cleptoria and Croatania. However, a reflexed tip also is found in stenogon (Sigiria) and fumimontis and aphelorioides (Falloria). Aside from the lack of metatergal stripes in some members of the /atior group as well as in various species of Cleptoria and Croatania, a relatively long process of sternum 4 occurs in the latter two subgenera and in the quadrata group. Both conditions are “‘incomplete synapomorphies’’, homoplasious with respect to one another and to various gonopodal features yet geographically coherent. Sigmoria (s. str.) is a discrete component of Sigmoria (s. lat.), but it is less clear how it relates to other subgenera. It appears to be a geographic sub- division with residual connections to other proximal subgenera along what are now disconnected clinal gradients. Geographic considerations and a general similarity of gonopods with those of arcuata (Cleptoria) seem to suggest a relationship of Sigmoria s. str. with Croatania and Cleptoria (see figure 154). In turn these subgenera might seem related to Rudiloria and Dixioria because of the presence in some members of a relatively long proc- ess on sternum 4 or the absence of metatergal stripes. However, there also are gonopodal similarities between species of Sigmoria s. str. and certain species of Sigiria and Cheiropus (notably areolata and stibarophalla, previously assigned to the /atior group by Shelley 198la). For reasons developed later in the chapter, we now consider Sigmoria s. str. to share common ancestry with Sigiria and Falloria (see Figure 156). Croatania. — This subgenus contains only the catawba group, with four species taxa (Figure 151, areas 21-24). Croatania has an autapomorphy, the irregularly notched expansion of the basal zone. Possibly also autapomor- phous, though shared with the more northern Dixioria, is the large prefemoral process. The spine on the basal zone in some but not all Clep- toria is considered homologous with the enlarged basal projection on the notched expansion of Croatania. This, in combination with the generally robust gonopods of both subgenera, specifies sister relationship. ‘‘In- MEM. AMER. ENT. SOC., 35 184 XYSTODESMID MILLIPEDS complete synapomorphies’’ in color and the form of the process of sternum 4 seem to link Croatania and Cleptoria with Sigmoria, Dixioria, and Rudiloria, but we provide a more satisfactory set of subgeneric relation- ships later in the chapter. Relationships within Croatania are clear. In simplex (#23), the process of sternum 4 is apomorphous in being short rather than long, and in catawba (#21), saluda (#22), and yemassee (#24) the prefemoral process is apomor- phous in being more greatly enlarged. This represents an obvious east / west vicariance. Over 100 miles downstream in the Savannah River Valley, yemassee differs from the other taxa in having smaller teeth in the expan- sion and a decidedly more proximal medial flange, which occurs on the distal extremity of the basal zone. This represents a second, south/north vicariance. In summary, Croatania = simplex + (yemassee + (catawba + saluda)). This simplicity is confused by the geographic distribution of color pat- terns. The northernmost taxon, catawba, is yellow/black, as are proximal populations of simplex. Farther south, approaching the range of primarily red/black Cleptoria, saluda and central populations of simplex also are red/black. Farther east, populations of simplex in the Fall Zone and Coastal Plain are red/red, and we predict that yemassee also is red/red as are all coastal congeners. Color patterns are not uniformly distributed by taxa, but they do form geographically coherent distributions. In some in- stances, they might be explained by mimetic convergence, but this explana- tion should apply only to sympatric taxa, not to such parapatric taxa as the geographically proximal species of Cleptoria and Croatania. Herein lies the basis for interpreting the color combination yellow/black or red/black as an “‘incomplete synapomorphy”’: its geographic distribution links various allopatric but geographically proximal taxa of these subgenera with Sigmoria and Dixioria. Where a similar color combination is found elsewhere, as in the southernmost population of australis (Cheiropus), it evidently is disconnected and may be a random convergence. Shelley (1977) proposed dispersal hypotheses to account for distributions of species of Croatania, but they are inappropriate because the animals are not sufficiently vagile. Although the species inhabit a greater diversity of biotopes than those of the proximal or sympatric rileyi (Cleptoria) and quadrata (Sigmoria) groups, their habitat requirements are still too narrow to allow them to spread through an area with so many biotopes. Cleptoria. — This subgenus contains only the rileyi group (Figure 151: rileyi, 25; abbotti, 26; bipraesidens, 27; robusta, 28; arcuata, 29; shelfordi, 30; macra, 31). Various differently distributed ‘‘incomplete synapomor- phies’’, including a ‘‘bird’s head’’ acropodite and/or presence of a basal R.M. SHELLEY & D.R. WHITEHEAD 185 spine, confirm monophyly. The basal spine is homologous to the enlarged basal projection on the notched expansion in proximal taxa of Croatania, hence these subgenera are sister groups. Puzzles within the rileyi group in- clude the resemblance of arcuata in some gonopodal features to Sigmoria s. str., the complex distributions and variations of macra and shelfordi, and the disjunctive populations of ri/eyi and shelfordi. Sigmoria (Cleptoria) arcuata is sister to the rest of the group, having a relatively slender acropodite, a laminate rather than thickened lateral flange, no distal lobe (“‘bird’s head’’), and, in the southernmost population (in Abbeville County, South Carolina), a reflexed tip (see table 6). Whether these conditions are ancestral or secondary is unclear, but they suggest features found in Sigmoria s. str. and perhaps represent other instances of ““ancomplete synapomorphy’’. Presence of the basal spine confirms place- ment of arcuata in Cleptoria. There is obvious circular reasoning in the postulate that gonopods are ancestrally stocky in Croatania/Cleptoria, if the relatively slender gonopods of arcuata are used to suggest a still more distant common ancestry with Sigmoria. Once again, ‘‘incomplete synapomorphy”’ provides a geographically plausible explanation. A curious puzzle is the isolated western, apparently relictual population of shelfordi, which is separated from the main range of the species by ar- cuata and which lies adjacent to the range of robusta. This form lacks the prefemoral process, perhaps an ancestral ‘‘incomplete synapomorphy’”’ shared with arcuata. We believe that the absence of a prefemoral process in rileyi and bipraesidens represents convergence with arcuata and the western form of shelfordi, because of the disrupted geographic distribution of this character state. Proximal populations of shelfordi and macra resemble one another, and although there are no clear intergrades, the taxa may even be conspecific. The proximal populations, the easternmost of shelfordi and southernmost of macra, tend to have a spur on the peak of the acropodite, and the former also tends to lack the basal spine as does macra throughout its range. Thus, although there is no synapomorphy for shelfordi and macra, the “‘incomplete synapomorphies”’ link them as sister species. Together, they are sister to the remaining four species, which are grouped by an upward shift of the medial flange from the basal zone toward the peak. The presence of the isolated western population of shelfordi is ac- cordingly explained by vicariance partitioning: arcuata separated first; shel- fordi + macra separated next, in an arcuate band to the south of arcuata; then, the range of shelfordi became subdivided. Next, robusta became isolated, as the ‘‘bird’s head’’ became more sinuate in the southern taxa. Later, the medial flange shifted still farther onto the peak and the prefemoral processes were lost in rileyi and bipraesidens. Thus, the cladistic MEM. AMER. ENT. SOC., 35 186 XYSTODESMID MILLIPEDS TABLE 6. Apomorphous character states in Cleptoria Character Taxon 25 27 426 28 3in 31s 30e 30w 30W 29 Acropodite massive DOS OR YOK SK CNOK CTO x x x — thickened, not laminate XX XX XX XX XX XX xX x x — “‘bird’s head’”’ XXX XXX XXX XX XX XX x x x — medial flange on peak XX XX x x — — — — — = with spur on peak — — — — — x x — —- — without basal spine _ x — — x x x — —_ — Prefemoral process absent Ke x — = a x x Without metatergal stripes — x — — — — — — — — Taxa numbered as in Table 3, with letters to indicate north, south, east, or west populations; W represents the isolated population of shelfordi. Character states range from plesiomorphous (—) to most apomorphous (xxx). hypothesis: Cleptoria = arcuata + ((shelfordi + macra) + (robusta + (abbotti + (rileyi + bipraesidens)))). Except for bipraesidens, the color pattern of Cleptoria is red/black. In this peripheral species, however, it reputedly is red/red (Hoffman 1967). The disjunct Chattahoochee population of ri/eyi lies just north of red/red forms of australis (Cheiropus), in which the southernmost population also is red/black. Thus, there may be two disjunctions here, that in ri/eyi and that in the red/black pattern. If the latter disjunction exists it may be a relict of a once more widespread distribution of this character state, but it might also be convergent. The disjunct populations of ri/eyi represent one of three instances in Sigmoria s. lat. of non-divergence among conspecific allopatric popula- tions; the others obtain in australis (Cheiropus) and stenoloba (Sigmoria). The main population of rileyi is in piedmont Georgia, and satellite popula- tions occur about 100 and 200 miles west in Lee and Jefferson counties, Alabama. Hoffman (1967) considered that in Lee County to be a distinct geographic race and predicted eventual discovery of material in the gap. However, the hiatus has been explored reasonably thoroughly at the time of year known populations are extant, and the gap appears to be real. Gonopods of all populations are practically identical, and subspecific status is unwarranted. R.M. SHELLEY & D.R. WHITEHEAD 187 The acropodites of rileyi, abbotti, bipraesidens, and robusta in Georgia and South Carolina are massive and heavily sclerotized, and in this respect they are convergent with those of crassicurvosa and pendulata (Falloria) in central Tennessee. Individuals of the latter species even display thick subter- minal acropodal lobes similar to those on the outer surfaces of the peaks and distal zones in rileyi and abbotti. The lobes are not homologous, however, as those in ri/eyi and abbotti represent modified lateral flanges, and the ones in the Tennessee species are extraneous and located on the medial surfaces. Sigmoria (Falloria) fumimontis, in the Great Smoky Mountains of eastern Tennessee, also has a heavy acropodite but the general configuration is more reminiscent of /atior (Sigmoria). This also is convergence, because the narrow medial flanges, medial tilting of the peak, and divided prefemoral process clearly assign fumimontis to the translineata group (Falloria). Hoffman (1967) and Whitehead (1972) proposed dispersal hypotheses to account for distributions of species of Cleptoria, but as with Croatania they are inappropriate because the animals are insufficiently vagile. These hypotheses did not account for shelfordi, arcuata, and robusta, but they in- correctly included divergens (Cheiropus). Cheiropus. — This major component does not form a neat, definable group and is difficult to relate to the other subgenera. However, an undeniable transformation series links stibarophalla to the progressively more derived taxa formerly assigned to the separate genera Prionogonus and Cheiropus (Shelley 1982, 1984a), and the overall distribution of the subgenus provides an explanation for the relationships and distribution of australis. Both distributions and characters of the planca group, formerly the sole constituent of the genus Cheiropus (Shelley 1984a), seem superim- posed upon rather than part of the generic mosaic, and our interpretations may need perceptive scrutiny in the future. We recognize ten species arrayed among the australis (#32), divergens (#33-34), haerens (#35-37), and planca (#38-41) species groups: Cheiropus = australis + (divergens + (haerens + planca)) groups, mapped in figure 151. The monobasic australis group has one of the three species in the genus with disjunct populations — one in coastal South Carolina and Georgia, one in south-central Alabama, and an intermediate one spanning the Chat- tahoochee River in southwestern Georgia, southeastern Alabama, and adja- cent Florida. The latter two populations are approximately 40 miles apart and may prove to be connected, although appropriate habitats have been in- vestigated without success. However, the central and eastern populations are so far apart as to virtually preclude extant linkage. Gonopods in all three areas vary within the same limits, and the populations are clearly con- MEM. AMER. ENT. SOC., 35 188 XYSTODESMID MILLIPEDS specific. The only curious character anomaly is in color pattern, with ex- treme southern representatives of the central population red/black rather than red/red. Superficially, australis most resembles stibarophalla, at the opposite range extreme of Cheiropus. If australis is sister to the rest of the subgenus, this similarity may be explained as ancestral, and so may be the geographic separation of the similar elements. Moreover, an explanation for the dis- junct populations arises from the same reasoning. They once were con- nected in the north and subsequently displaced by other forms of Cheiropus and/or Cleptoria. However, there is no good synapomorphy for the sister lineage, unless comparatively massive gonopods in stibarophalla and divergens is so considered. We adopt this scenario for explanatory purposes and erect a separate species group for australis. The next vicariance in Cheiropus occurred in the north, separating the conservative divergens group (stibarophalla, #33, and divergens, #34) from those that developed acropodal solenomerites. Again, there is no precise autapomorphy for the divergens group, formed for vicariance reasons. If stibarophalla is the conservative gonopodal type from which other Cheiropus arose, geographic connections to them make sense only if stibarophalla and divergens shared common ancestry because of the relative geographic position of stibarophalla. Otherwise, the range of divergens represents a discontinuity between those of the haerens and planca groups. One interesting feature of the divergens group is that the species divergens separates the ranges of stenogon and disjuncta (Sigiria), and proximal populations of all three taxa tend to be dark or purplish in markings. These are the only congeners so marked, and we consider this still another exam- ple of ‘‘incomplete synapomorphy’’. Mimetic convergence is unlikely, since the ranges of the three taxa are allopatric rather than sympatric. The gonopods of divergens are sufficiently unlike those of the other two species to suggest close relationship. The haerens group (Prionogonus sensu Shelley 1982) has a relatively feebly developed solenomerite intermediate between the absence of such a structure in the divergens group and the well developed one in the planca group. It also is defined by the autapomorphous row of spurs on the outer surface of the basal zone. This feature is unique among xystodesmids and not homologous to modifications of this region in the tuberosa (Falloria), rileyi (Cleptoria), and catawba (Croatania) groups. The species geo- graphically most proximal to the planca group is thrinax, which in general gonopodal form also is structurally most similar to the most proximal member of the planca group, agrestis. The haerens group includes three parapatric species (haerens, #35; divaricata, #36; and thrinax, #37) with —— SS R.M. SHELLEY & D.R. WHITEHEAD 189 small geographic ranges, and, like the australis and divergens groups, it is part of the cohesive generic mosaic. The planca group (Cheiropus sensu Shelley 1984a) includes four allopatric, more or less disjunct species so strongly differentiated from one another that three generic names have been associated with them. However, they lack obviously ‘‘sigmoid’’ gonopods, clearly an apomorphic condition. Also, an overall west to east clinal trend toward increased acropodal serra- tion which persists independent of speciation is manifested in proximal populations of agrestis (#38), serrata (#39) and planca (#40) (Shelley 1984a). Moreover, planca is structurally intermediate between serrata and persica (#41). Thus the planca group = agrestis + (serrata + (planca + persica)). This species group weakens our generic hypothesis for Sigmoria. Not only is it differentiated by having secondarily curvilinear rather than ‘‘sigmoid’’ gonopods, but its distribution broadly overlaps that of various other southern members of the genus. Sympatry occurs within Cheiropus itself, between serrata and the syntopic eastern population of australis, and also between agrestis and two species of Cleptoria. Sigiria. — This subgenus (Figures 151-152, areas 42-55) has two major subunits which are geographically contiguous but structurally disconnected by divergent forms, the rubromarginata (#42-52) and the stenogon (#53-55) species groups. The stenogon group includes three geographically associated species, stenogon (#54), areolata (#53), and disjuncta (#55). Of these, disjuncta is anatomically similar to nigrimontis (#46-49) and formerly was grouped with it (Shelley 1981a). However, the course of the prostatic groove differs, the color is purple/purple as in the most proximal populations of stenogon, and its range is separated from that of nigrimontis by stenogon and areolata, and species of the rubromarginata group. The adjacent species areolata and stenogon diverge in gonopodal features from disjuncta. Thus, the stenogon group = disjuncta + (stenogon + areolata). The polytypic nigrimontis is structurally and geographically intermediate between the forms related to rubromarginata and inornata. Specifically, one may trace gradients in gonopod configurations from nigrimontis via sigirioides (#50) to inornata (#52) and truncata (#51), and via triangulata (#45), rubromarginata (#42), and austrimontis (#43) to whiteheadi (#44). In the Toe River Valley fauna, inornata occurs in the headwaters and therefore at a geographic extreme, and although the geographically intermediate trun- cata is more highly derived in gonopodal features, it shows clearer evidence of past continuity with inornata than with sigirioides. We have found no satisfactory dichotomous resolution of relationships among members of the rubromarginata group. MEM. AMER. ENT. SOC., 35 190 XYSTODESMID MILLIPEDS An instance of geographic discontinuity occurs in the rubromarginata group among taxa which should be clinally continuous. Sigmoria rubromarginata is restricted to the Blue Ridge Province of North Carolina and Tennessee, and the main population of austrimontis is found in the South Mountains, an inselberg chain in the western Piedmont Plateau. Their chief difference involves the distal zones, that of austrimontis being twisted mediad to reveal the surfaces of the medial and lateral flanges in medial view, while that of rubromarginata faces anteriad to reveal only the edge of the medial flange in this perspective. The range of the South Moun- tain form is contiguous in the western Piedmont and Blue Ridge escarpment with more variable ones that bridge the anatomical gaps. The latter popula- tions of austrimontis have features that suggest intergradation with rubromarginata, but the taxa are disjunct by about 30 miles and are con- sidered reproductively isolated. On the Blue Ridge Front in Virginia, whiteheadi occurs some 120 miles north-northeast of the nearest population of austrimontis, at Morganton, Burke County, North Carolina. Also, whiteheadi has yellow stripes instead of red, and the gonopodal resemblance between it and austrimontis is not as close as that between the latter and rubromarginata. Sigmoria (Sigiria) whiteheadi and austrimontis might seem to represent an eastward extrusion into an area between the ranges of Dixioria and Croatania that should otherwise be occupied by Sigmoria s. str., but the supposed range extrusion violates the underlying hypothesis of simple vicariance partitioning. Rather than Sigiria plus Falloria as the basal sister element in the genus (Figure 154), that element should be either Rudiloria plus Dixioria, or Croatania plus Cleptoria. Resulting vicariance schemes appropriately place Sigiria between the ranges of Dixioria and Sigmoria. Sigiria includes three yellow/yellow taxa — areolata, inornata, and whiteheadi — that are neither geographically nor structurally contiguous with one another. This coloration is found among members of several other eastern subgenera and is more probably an ‘“‘incomplete synapomorphy”’ than an instance of convergence. Within Sigiria, however, this coloration is plesiomorphous. In whiteheadi, it is relictual in comparison to the syntopic yellow/black /. /atior (Sigmoria). In inornata, the condition is variable, and, though we lack details, we predict that the populations most proximal to truncata are the ones that are red/red. Sigmoria (Sigiria) inornata occurs in headwater areas of the Toe River Valley, and its westernmost, downstream populations most resemble truncata in gonopodal characters by having longer and flatter peaks, and shorter distal zones. In areo/ata, an obvious sister to stenogon, the color is stable so far as known, but the R.M. SHELLEY & D.R. WHITEHEAD 19] nearest populations of stenogon tend to be red/red unlike the purple/ purple ones farther south, so that there is a stepped clinal continuum. Although the stenogon group is closely allied geographically and struc- turally with the rubromarginata group, there is no obvious synapomorphy. Therefore, these may not truly be sister groups. Shelley (1981a) allied nan- tahalae (here placed in Falloria) with stenogon, as the two species then con- sidered to represent the stenogon group. We assume that the resemblances were ancestral. This satisfies our premise that nantahalae is the sister ele- ment within Falloria, which despite diversity apparently is truly monophyletic. However, we are not certain that the stenogon group is not directly sister to Falloria rather than to the rest of Sigiria. Thus of the eight subgenera, Sigiria is the one least likely to be monophyletic. Falloria. — This subgenus, (Figures 151-152, areas 56-73) consists of 18 species arrayed among eight species groups. The lineage is defended by geography (proximal only to Sigiria), linkages via transformation series, and apomorphous red/blue coloration. No other congeners have this bicolored pattern, which we presume is energetically expensive to acquire and therefore unlikely to be convergent. However, sympatric species of Brachoria also display the pattern, which we think represents mimetic con- vergence rather than ‘‘incomplete synapomorphy.”’ In either case the col- oration appears to be conserved by mimetic pressure. Two geographically peripheral species, white/ white /eucostriata (#57) and red/ white nantahalae (#56), have color patterns unique in the genus. Both occur in areas not oc- cupied by red/blue Falloria or Brachoria, and the absence of mimetic pressure may explain the absence of blue metatergal stripes. The postulated sister group to Falloria is Sigiria, particularly the stenogon group, which is the basis for out-group comparisons. Northward, where species of Falloria are parapatric with those of the rubromarginata group, there is abrupt structural discontinuity. Our hypothesis of species group relationships is Falloria = nantahalae + (leucostriata + (aphelorioides + (translineata + ((tuberosa + (bidens + picapa)) + mimetica)))). This hypothesis is derived from the following considerations. Aside from derived color pattern, the monobasic nantahalae group (#56) lacks gonopodal flanges. The ancestor of its sister group is postulated to have had a sort of hood on the peak of the acropodite. The /eucostriata group (xerophylla, #58, and leucostriata, #57) retains a short process on sternum 4, and its species are disjunct but gonopodally similar. The disjunction is explicable via vicariance partitioning. Its sister group is postulated to have had a medium process of sternum 4 ancestrally. The monobasic aphelorioides group (#59) has a looped acropodite. Its sister group is postulated to have had an enlarged prefemoral process ancestrally. MEM. AMER. ENT. SOC., 35 192 XYSTODESMID MILLIPEDS The ancestor of the franslineata group had a forked prefemoral process, whereas the ancestor of its sister lineage is postulated to have had the peak of the acropodite dentate ancestrally. The more eastern, Blue Ridge species (translineata, lyrea, fumimontis, and ainsliei, #60-63) had a more massive acropodite. The more western, Cumberland Plateau ones (forficata, houstoni, and abbreviata, #67-69) diverged into an easterly form with the process of sternum 4 lengthened and westerly forms in which it was shortened. Therefore, the translineata group = ((fumimontis + trans- lineata) + (lyrea + ainsliei)) + (forficata + (houstoni + abbreviata)). Relationships among the bidens, tuberosa, picapa, and mimetica groups are more obscure. In the bidens group, the prefemoral process is more or less swollen, with the process of sternum 4 moderate in bidens (#64) but shortened in prolata (#65). In the Cumberland Plateau of Tennessee, the monobasic picapa group (#70) has the process of sternum 4 shortened and shares certain acropodal features with prolata of the bidens group, so the two groups are sisters. In the Blue Ridge, the bidens and tuberosa groups are not linked by a satisfactory synapomorphy, but by geographic in- ference. In the monobasic tuberosa group (#66), the prefemoral process is short and the process of sternum 4 is lengthened, in addition to various other distinctive features. In the Nashville Basin, the mimetica group has the process of sternum 4 shortened, and it is divergent in having greatly reduced prefemoral processes and massive acropodites in two species. Despite their divergences and disjunctions, the tuberosa, bidens, picapa, and mimetica groups appear readily interpreted as a once continuous northern vicariad. The mimetica group (mimetica, #71, crassicurvosa, #73, and pendulata, #72) includes a curious population. One sample resembling the two eastern species, pendulata and crassicurvosa, was found in the eastern part of the range of the westernmost species, mimetica. The gonopods of mimetica are considerably thinner than those of the other two species (compare figures 111, 117, 123) and are homogeneous throughout the range. However, the above exception has a thicker acropodite with slightly swollen areas sug- gestive of lobes (Figures 113-114). It may represent a relict of a former con- nection between mimetica and crassicurvosa, and it also suggests that the three taxa of the mimetica group may be conspecific. The specimens were collected in 1957, and since then the eastern range periphery of mimetica and the area between the species has been thoroughly searched without find- ing more individuals. Consequently, the status of this population is uncer- tain, and it may be extinct. The massive gonopods of crassicurvosa and pen- dulata are synapomorphous. Thus, the mimetica group = mimetica + (crassicurvosa + pendulata.) ee EE eee R.M. SHELLEY & D.R. WHITEHEAD 193 We propose the following vicariance sequence within Falloria. The first vicariance occurred on the southeastern front of the Blue Ridge, separating nantahalae. The second separated an eastern /eucostriata group. The third segregated aphelorioides at the southern end of the Blue Ridge, and the fourth caused a south/north separation of the translineata group from the tuberosa, bidens, picapa, and mimetica groups. This vicariance scheme produces some geographic quirks, notably in dis- junctions between the Nashville Basin, Cumberland Plateau, and Blue Ridge. However, it also supplies a plausible explanation for the disjunct ranges of xerophylla and /eucostriata. They are so similar in gonopodal characters that they must once have been continuous, but they now occur at opposite ends of the rugged southern Blue Ridge Mountains and are separated not only by distance and terrain but also by numerous congeners. Here, extinction of intermediates may have been caused by range expan- sions of other taxa, as suggested by the relatively large ranges of two in- tervening eastern species, nantahalae (Falloria) and rubromarginata (Sigiria). In prolata, isolated in a small pocket of the GSMNP near Gatlinburg, Tennessee, the distal zone curves sublaterad from the peak and is not coplanar with the basal zone. It is similar to that of picapa in the Cumberland Plateau and markedly different from the curvatures in bidens and the more southerly forms of the franslineata group, in which the two zones are coplanar. The boundary between prol/ata and bidens in the Roar- ing Fork Nature Trail section of the GSMNP is sharp, as the latter occurs along the more southern entrance road and the former along the more northern exit road. This southern population of prolata has a mildly globose prefemoral process characteristic of bidens (compare figure 83 with Shelley 1981a, figure 80), unlike the northern, Greenbrier populations (Figures 79-82). The resemblance of the prefemoral process with bidens links only the most proximal populations, whereas the acropodal resemblance with picapa occurs throughout the range. Thus, the distal zone curvature of prolata and the bifurcate prefemoral process of forficata form evidence of former links between the Blue Ridge and Cumberland faunas. Geographic factors indicate that the long sternal processes of forficata in the Cumberland Plateau, tuberosa in the Great Smoky Mountains of North Carolina, and various piedmont South Carolina species of Croatania and Cleptoria, represent convergence (see Table 2). Sigmoria (Falloria) forficata is separated from tuberosa by about 90 miles of the most rugged mountains in the southeast, and some 70 miles and more mountains segregate the latter from the piedmont forms. MEM. AMER. ENT. SOC., 35 194 XYSTODESMID MILLIPEDS In the Blue Ridge, aphelorioides has a circular acropodite similar to that of ainsliei, but it is placed in a separate species group because the prefemoral process is undivided. Their distributions, moreover, are separated by other species of the ¢translineata group. Hence, their looped acropodites probably are convergent, as they also are with trimaculata (Rudiloria) and the genus Apheloria. SUBGENERIC RELATIONSHIPS. — Here, we attempt to relate the subgenera of Sigmoria, discussed above and mapped in figure 153. We have no satisfactory character state nor transformation series to define the genus Sigmoria as monophyletic, but it is defined by its geographic cohesiveness. It has geographic properties similar to those of a single species. When species diverge, there is no theoretical need for both members of a sister pair to be defined by autapomorphies. Here, this applies at the generic level be- tween the sister genera Sigmoria and Deltotaria, wherein the latter is de- fined by a satisfactory synapomorphy and represents a now independent overlay in the southern part of the former’s range. The sister relationship of these two genera, as discussed under generic relationships later, is postulated on both geographic and structural grounds. We consider slender gonopodal form in the montane D. brimleii and in sympatric species of Falloria and Sigiria, and robust gonopodal form in the piedmont D. /Jea and in species of Croatania and Cleptoria in that region, evidence of past con- nections. Problems with the preliminary subgeneric hypothesis. — The subgeneric arrangement used throughout this paper reflects a clockwise, radial ar- rangement from Rudiloria in the north to Falloria, based on the preliminary cladistic hypothesis shown in figure 154. This unsatisfactory, basally trichotomous arrangement specifies an east/west separation between the Croatania/Cleptoria and Falloria/Sigiria lines, with the geographically in- termediate and more southern Cheiropus unassigned. The eastern lineage occupied the northern Appalachians and eastern piedmont, and the western lineage occupied the southern Appalachians and Cumberland Plateau. The eastern lineage divided into northern (Rudiloria/Dixioria) and southern components, the latter next into eastern (Sigmoria) and western (Croatania/Cleptoria) lineages. Superficial gonopodal similarities between sympatric members of Deltotaria and Sigmoria formed one reason for a basal dichotomy separating Croatania and Cleptoria to the east from Falloria and Sigiria in the west. Cheiropus, which is geographically intermediate between these presumed sister pairs, does not bear conspicuous relationship to either. Reasons for linking Rudiloria, Dixioria, and Sigmoria s. str. to Croatania and Cleptoria rather than to Falloria and Sigiria arose principally from the R.M. SHELLEY & D.R. WHITEHEAD 195 geographic distributions of certain character states and postulated ‘‘in- complete synapomorphies”’ in the lengthened process of sternum 4 and the loss of metatergal stripes. In turn, Sigmoria s. str. seemed better tied to Croatania and Cleptoria than to Rudiloria and Dixioria, as judged prin- Fic. 153. Distributions of the subgenera of Sigmoria. 1, Rudiloria; 2, Dixioria; 3, Sigmoria; 4, Croatania; 5, Cleptoria; 6, Cheiropus; 7, Sigiria; 8, Falloria. MEM. AMER. ENT. SOC., 35 196 XYSTODESMID MILLIPEDS cipally from some gonopodal similarities in arcuata (Cleptoria) and quadrata (Sigmoria). This scenario is faulty in the following respects: 1. Falloria and Sigiria are allied, but it is not clear that Sigiria is monophyletic. Falloria might be directly allied to the stenogon group of Sigiria, but we found no compelling reason why the rubromarginata group might not be linked with Sigmoria rather than with Falloria. Nonetheless, we provisionally accept Falloria and Sigiria as true sister groups. 2. We accept Croatania and Cleptoria as sister groups based on synapomorphies. The argument for linking Sigmoria with Croatania/Clep- toria contradicts the character argumentation for recognizing basal Rudiloria Dixioria Sigmoria Croatania Cleptoria Cheiropus Sigiria Falloria 154 Fic. 154. Relationships among the subgenera of Sigmoria, preliminary hypothesis. R.M. SHELLEY & D.R. WHITEHEAD 197 Falloria/Sigiria and Croatania/Cleptoria lineages, because it specifies a reversal in gonopodal form from robust to slender, convergent upon that in Sigiria. 3. There are evident structural linkages of Sigmoria with Sigiria and/or Cheiropus, and either or both are geographically plausible. The position of Cheiropus is enigmatic, yet there are reasons to link it with Sigmoria. However, the ensuing hypothesis would have the combined distributions of Cheiropus and Sigmoria encircling those of Croatania and Cleptoria, an im- probable vicariance scheme. 4. We accept Rudiloria and Dixioria as sister groups based on synapomorphy. However, proximal taxa of Dixioria and Sigiria have gonopods sufficiently similar that convergence is implausible. This presents a circular argument, if Rudiloria and Dixioria are said to have arisen from a lineage which ancestraily had massive gonopods. 5. The northeastward extension of Sigiria via austrimontis and whiteheadi, which separates the ranges of Rudiloria/Dixioria from those of Croatania/Cleptoria, represents an unnecessary argument for range disper- sal made still more implausible by the large gap between austrimontis and whiteheadi. It can be avoided if Rudiloria/Dixioria is considered a basal sister element or otherwise separated from a Croatania/Cleptoria lineage. Consider the pizza pie analogy. No matter how irregular the slices, the pie does not actively slice itself. Thus, the postulated range extrusion of the rubromarginata group involving austrimontis and whiteheadi would be un- satisfactory even if there were a clear, extant continuum between the two. Resolution of the problems. —\f we accept Falloria/Sigiria, Rudiloria/Dixioria, and Croatania/Cleptoria as monophyletic groups, we need to select an appropriate dichotomous vicariance scheme for just five slices. There are 105 possible schemes, but many are geographically untenable as shown by figure 155. For instance, the sister of Rudiloria/Dix- ioria must either be or at least include Falloria/Sigiria to account for geographic proximity. Neither Cheiropus, Croatania/Cleptoria, Falloria/Sigiria, nor Sigmoria alone represent plausible first cuts, because their distributions are wedge shaped. Hence, the first cut must have been Rudiloria/Dixioria, if we assume that one group is sister to the other four. If two groups are sister to the other three, the ones that need consideration are the following. For Rudiloria/Dixioria, the only possible pairing is with Falloria/Sigiria, and the only plausible triplet arrangement would include Sigmoria. For Croatania/Cleptoria, linkage could be with Cheiropus and/or Sigmoria. An initial separation of Cheiropus and Croatania/Cleptoria from the rest of the genus based on ancestrally massive versus slender gonopods is geo- MEM. AMER. ENT. SOC., 35 198 XYSTODESMID MILLIPEDS graphically plausible, but the groups are divergent from one another and Cheiropus shows affinity elsewhere in the genus. In an initial separation of Rudiloria/Dixioria and Falloria/Sigiria from the rest of the genus, neither 14+2 Fic. 155. Vicariance map and cladogram for the subgenera of Sigmoria, numbers as in Fig. 153. The Jefferson Co., AL, record of rileyi (Cleptoria) and areas of secondary sympatry of latior (Sigmoria) are omitted. 1 + 2, Rudiloria + Dixioria; 3 +4, Croatania + Cleptoria; 5, Cheiropus [(5), planca group]; 6, Sigmoria; 7 + 8, Sigiria + Falloria. R.M. SHELLEY & D.R. WHITEHEAD 199 lineage in the basal dichotomy is supported by a satisfactory autapomor- phy. We reject these hypotheses. We propose Rudiloria/Dixioria as sister to the rest of the genus. This is the most probable, geographically simple arrangement and is based on the most satisfactory autapomorphy, although for one lineage only: gonopod orientation in Rudiloria/Dixioria. Slender gonopods in Rudiloria/Dixioria and proximal Sigiria represent the ancestral condition, not convergence, and there are no detailed residual linkages with other groups such as are im- plied between Croatania/Cleptoria and Sigmoria. This arrangement also implies convergence rather than ‘‘incomplete synapomorphy’’ in the lengthened process of sternum 4 and color pattern. Thus, the continuum of the rubromarginata group of Sigiria via austrimontis to whiteheadi divides the lengthened process of sternum 4 of Rudiloria/Dixioria from those of Croatania/Cleptoria and southern members of Sigmoria. Moreover, the color pattern involving lack of metatergal stripes may represent mimetic convergence, as the distribution of /. /atior (Sigmoria) completely overlaps that of Dixioria rather than being contiguous. If Rudiloria/Dixioria is sister to the rest of the genus, then just four groups and 15 dichotomous arrangements remain. Only Croatania/Clep- toria is a geographically and structurally suitable basal element. As in Rudiloria/Dixioria, there is an acceptable ancestral autapomorphy for Croatania/Cleptoria, the spine or irregular notching on basal zone, but not for the sister lineage. This leaves just three groups and three arrangements. Our resolution of these is based partly on geography and partly on divergence. Cheiropus has elements which are the least similar, perhaps in part reflecting ancestral geographic variations in the genus. Where geographically proximal to Falloria/Sigiria, there is little structural resemblance. Geographic connec- tions with Sigmoria are in part dissected by the range of Croatania/Clep- toria, and where the ranges overlap, as with /. munda (Sigmoria) and the divergens and haerens groups (Cheiropus), gonopods are too dissimilar to imply close common ancestry. In contrast, gonopodal similarities of prox- imal elements of Sigmoria and Falloria/Sigiria are much stronger. Relationships among the subgenera. — Our revised cladistic hypothesis (Figure 156) is: genus Sigmoria = (Rudiloria + Dixioria) + ((Croatania + Cleptoria) + (Cheiropus + (Sigmoria + (Sigiria + Falloria)))). This hypothesis represents our best estimate of vicariance partitioning of the subgenera. In the absence of a suitable synapomorphy scheme, which we think does not exist, it is perforce based principally on the geographic distributions of the taxa and their character states considered together. It does not resolve the problems of undefined monophyly of Sigiria nor defini- MEM. AMER. ENT. SOC., 35 200 XYSTODESMID MILLIPEDS tion of the lineage Sigiria/Falloria. However, the only plausible geographic linkage for Falloria is with Sigiria (or a part of Sigiria, the stenogon group), and if Sigiria is not a natural group the only plausible alternative arrange- ment is to shift the stenogon group to Falloria. Features of the hypothesis are the following: 1. The partitioning pattern (Figure 157) is coherent and sequential, re- quiring only minimal range alteration and secondary sympatry to achieve the present distributional pattern. Rudiloria Dixioria Croatania Cleptoria Cheiropus Sigmoria Sigiria Falloria 156 Fic. 156. Relationships among the subgenera of Sigmoria, revised hypothesis. R.M. SHELLEY & D.R. WHITEHEAD 201 2. The virtual encirclement of Croatania/Cleptoria by Cheiropus and Sigmoria s. str. is geographically illogical and thus probably secondary. Aside from the clearly secondary distribution of the planca group, the southern area of Cheiropus is represented only by the now disjunct popula- tions of australis. That these may be recent immigrants is consistent with their lack of differentiation. 3. The gonopodal similarities between Rudiloria/Dixioria and Sigiria reflect ancestral similarity, not convergence. Length of the process of ster- num 4 is not quite continuous from Dixioria to Sigmoria. The ancestral similarity features implied by items 4-6, both somatic and gonopodal, are ones that cross indicated cladistic lines. We conclude that they are plesiomorphous, not apomorphous. DISCUSSION. — Traditional cladistic analysis assumes that lineages are independent and definable by autapomorphies, and where apparent con- flicts occur they are resolved by parsimony techniques. The traditional analysis is based solely on character information, independent from geography. Our first assumption must always be that a character based cladistic hypotheses is attainable, regardless of initial impressions. Even after realizing that Sigmoria was a geographic mosaic, we still attempted traditional analysis. It failed. Our initial attempts involved the taxa first reviewed by Shelley (1981a) and were based on color, body size, details of gonopodal structure, and even ecology. We found virtually complete non- congruence between all characters studied and could only conclude that Sigmoria as it then stood could not be monophyletic, despite patterns of geographic contiguity among character states. Parts of the puzzle seemed to be missing. The conclusions about relationships reached in this paper need to be tested by rigorous cladistic analysis, but we contend that such analysis will require substantially more critical information such as might derive from facial chaetotaxy and comparative body chemistry. Nearly all of the characters and character conditions that we presently have for analysis are homoplasious. Therefore, the key to solving relationships within Sigmoria is the realiza- tion that the genus is a geographic mosaic. Its autapomorphy is geographic cohesiveness; Sigmoria has the geographic properties of a species. Its minimal divisions, the species and subspecies, also are geographic entities, with allopatric and/or parapatric relations to other such divisions. Character states map in cohesive geographic patterns but are noncongruent with one another. The on/y way to analyze the mosaic based on presently available character data is to include geography. MEM. AMER. ENT. SOC., 35 202 XYSTODESMID MILLIPEDS In our analysis of major components, we combined geographic and cladistic inference to refute the null hypothesis that the genus Sigmoria was acontinuum. We tested the subgenera as stationary vicariads in our analysis of relationships. In both cases, spatial configurations are the critical factor. Fic. 157. Partitioning system and cladogram for the subgenera of Sigmoria. A, Rudiloria; B, Dixioria; C, Croatania; D, Cleptoria; E. Cheiropus; F, Sigmoria; G, Sigiria; H, Falloria. R.M. SHELLEY & D.R. WHITEHEAD 203 In traditional cladistic analyses, both lineages of a sister pair are defined by autapomophies. However, in Sigmoria, rarely are both members of geographic pairs thus defined. This situation merely reflects the evolu- tionary history of the mosaic, which was created by simple vicariance parti- tioning. Both divergence and extinction are largely absent from the system. Though we were forced to use geographic distributions alone to determine sequencing, we are satisfied with the accuracy of the method although it lacks the persuasive force of traditional character analyses. The method should pertain to all tightly fitted allopatric/parapatric mosaics. For their resolution, major component analysis using both geographic and character distribution inference is the only recourse available. In Sigmoria it is easier to analyze cladistic (character) relationships within the subgenera than be- tween them. These units are segments of the geographic whole, and thus their constituent parts are less divergent and demonstrate more obvious linkage patterns. It is ironic that a thorough analysis of Sigmoria lends credence to the judgement by Hoffman (1958a) that it had become a “‘catch-all’’ genus. Now, it is much more so, because it has ‘‘caught’’ all relevant pieces of a biological mosaic. All of the subgenera, previously considered genera, represent only fragments of a single geographic entity. Sigmoria thus is larger than anyone previously imagined. What we now call Sigmoria is an enormous apheloriine mosaic complex that blankets eastern North America. We might speculate on its origin, how it came to cover so large an area, and the nature of the vicariance mechanisms, but the important fact is that vicariance did take place. Various ancestral character clines were ir- regularly fragmented to create the present picture. Extinctions and disper- sals seem to have played a negligible role. What seems to be active, disper- salist speciation is only the passive result of vicariance. The animals do not actively explore new areas and environments. Instead, they remain in one place and eventually become isolated. The resultant localized populations are reflected in the phenotypes that we interpret as reproductive isolates. The story of Sigmoria still may be incomplete. Other chapters may unfold when Apheloria and Brachoria are reexamined. Future workers may con- clude that Sigmoria must be submerged under one of these older names, particularly Brachoria, although we think they form separate mosaics. The forms assigned to Brachoria have ‘‘sigmoid’’ acropodites that fall within the range of curvatures in Sigmoria. Published drawings show apparent medial flanges, teeth on the medial and undersurfaces of the peaks and distal zones, and various other acropodal modifications comparable to con- ditions found in Sigmoria. The cingulum on the outer surface of the acropodite, which we regard as autapomorphous for Brachoria, might real- MEM. AMER. ENT. SOC., 35 204 XYSTODESMID MILLIPEDS ly be of polyphyletic origin as it occurs at different relative positions. However, a test for homology may be possible by doing the same kind of major component analysis as we have done for Sigmoria. We think Sigmoria can serve as an example in diplopodology of a geographic mosaic, a highly partitioned but still cohesive entity in which differentiation must be rapid. Problem groups, in which accurate delinea- tion of species or genera is difficult, exist throughout the class. In the Polydesmida the xystodesmid genera Rhysodemus and Nannaria may have as many as 100 and 200 species respectively (Hoffman 1964, 1966, 1970), and the chelodesmid genus Chondrodesmus has over 40 weakly differen- tiated ones (Hoffman 1978b). In the Spirobolida, the Rhinocricidae is renowned for its internal confusion. Hoffman (1969, 1978b, 1979) sug- gested that these groups are in expanding phases of evolution and actively speciating. We agree that speciation is taking place, but we think it is ac- complished passively through vicariance. These and such polydesmoid taxa as Amplinus and Pycnotropis (Platyrhacidae) and the phenotypically similar California xystodesmids Xystocheir, Paimokia, Motyxia, and Amplocheir (Xystocheirini) probably are other examples of large, complex mosaics with clines and gradients that have been randomly split by vicariance events, some of which are incomplete. In poorly vagile millipeds, dispersal probably plays only a minor role in the speciation process. The insight on mosaics derived from Sigmoria should apply also to other animal groups with allopatric/parapatric patterns. For example, one may think of systems of isolated mountain peaks, caves, or sand dunes and their specialized beetle faunas. As with the Diplopoda, these mosaic patterns reflect inherent nonvagility. In more highly vagile animals, mosaic patterns may be dictated by extrinsic biotic factors, such as critical host plant associations or mimetic complexes. In such situations, the mosaic must be identified and its pieces analyzed, requiring detailed distributional informa- tion as has been compiled for Sigmoria. We have identified the shapes of its constituent basic pieces, the species and subspecies; analyzed the relative fits of these pieces to one another; and provided an internally consistent, work- ing hypothesis of relationships for future workers to test. It remains for them to complete the analysis of the Apheloriini by applying these pro- cedures to Apheloria and Brachoria and then to reexamine Sigmoria to determine if it should be submerged under one of these older names. To facilitate this process, we conclude with a review of the apheloriine genera, an analysis of their relationships as far as possible, and an analysis of the af- finities among the four principle eastern xystodesmid tribes. i R.M. SHELLEY & D.R. WHITEHEAD 205 Part VII: THE GENERA OF THE TRIBE APHELORIINI By Rowland M. Shelley Tribe APHELORIINI Hoffman Apheloriini Hoffman, 1979:158, 187. Components: Apheloria Chamberlin, 1921; Brachoria Chamberlin, 1939; Dynoria Chamberlin, 1939; Sigmoria Chamberlin, 1939; Deltotaria Causey, 1942; Furcillaria Shelley, 1981. Diagnosis. — Relatively flexible Xystodesminae of moderate to large size; midbody sterna broad and flat or with variable grooves or impressions, without elevations, caudal edges straight or only slightly produced, without lobes or spines; anterior tarsal claws of males normal, bisinuate; gonopods small to large; sterna connected by membrane only, without sternal rem- nant; coxae with or without variable apophyses arising on anterior sides, ex- tending ventrad; prefemoral process present or absent, when present variably curved and elongate, often enlarged or subglobose basally, never acicular; telopodite usually entire, occasionally divided or with separate solenomerite, usually curving anteriad, anteriomediad, or mediad in sigmoidal, semicircular, or circular configurations, occasionally linear, usually twisted or with torsion at 1/4 to 1/3 of total length, resulting in crossing of prostatic groove from medial to lateral surfaces; cyphopods with variable receptacle. Distribution (Fig. 158). — Eastern North America from northern New England to south-central Wisconsin (passing through southern Ontario and the lower peninsula of Michigan) south to Baton Rouge, Louisiana, southern Alabama and Mississippi, and just north of Tampa, Florida. Longitudinally, the tribe is known from the Atlantic seaboard from northern Florida to Virginia, New Jersey to Connecticut, and near the coast in Maine to southeastern Nebraska and eastern Oklahoma. The shaded area in figure 158 was determined from published records in Chamberlin and Hoffman (1958) and Keeton (1959, 1965) plus specimens I have personally collected or seen in museum and private collections. The contrast between the smooth northern and irregular southern boundaries is striking, but the northern distribution has not been as thoroughly studied and its evenness may reflect insufficient sampling. I also slightly extended the northwestern corner into southeastern South Dakota, since the presence of Apheloria at Omaha, Nebraska (see Apheloria account), suggests potential occurrence in riparian habitats farther north along the Missouri River. Remarks. — This revised diagnosis incorporates no real changes from the original by Hoffman (1979), but it is substantially expanded to better MEM. AMER. ENT. SOC., 35 XYSTODESMID MILLIPEDS 206 with the other major eastern tribes — iini contrast the Aphelor . > (see Hoffman 1960, 1964 smini , and Pachyde Shelley 1984c). The date of proposal of Deltotaria is also corrected to 1942. nl , Nannar ini Rhysodesm Just as colors and color patterns tend to cluster within Sigmoria s. lat., they also do within the Apheloriini as a whole and between tribes, and this probably represents mimetic convergence. For example, populations of /. latior and stenoloba (Sigmoria) with uniformly black metaterga occur sym- Ikes and Ashe counties, North Carolina, with forms of Pleuroloma flavipes having the same markings, and as mentioned earlier patrically in Wi d patterns in Sigmoria s. lat. tend to overlap areas where ious stripe the var OK? “a, ho ees Ja COOSA ererecaneratereneneen cid, 0,6 ,0, 0,4, $5 SM £52505 "sf yo Sy KX 20) Vesocospeacone , , ees War areons Sate taete anatase SO RY SREP $000 etaten », rx HH 5 O05. S505 OSB SCO) o0509 SSSR KK ROC RK OS M SO SBS SERRE K KR RRND oe. Q SOR KKK ODO MD WererererereroZArererees OrOne re, rOrene, SS Secosoescococodecocococog nr ococesohes : SOCK RRR RR KKK EPCOS Z\ LAK SKK KR ARR KKM KKK KD SRP RRO K OOS KS 2 pS RR ILC I 90 ROC 90 eosta Nchcatetatehotaventetetete’s tetetate % RCSL ROK MKD SWS MRR RRR K EC ORC SoS 55050 SE SCEPC RII ORK RRR KOR TN) ESS REPS LSS SOF Lod 79 PPP 6252 COLO COENG AM eetaletccteneee ra Perera Vorerarerereener: SIS 25 05050505 OOO SSSR EGE SLR 005 SPOS So LOR OH 909 CSRS Os , 555 SHOR IOC g erneraten Weare ene ae Onenes 05250505 : O05 5255295555255 erarererieerere erereerere Oren LOO am Os K O452504 of 6 0% KS OOOO K WX AV. @, PPQQR PSSA SCP 52S RIG SCS CRAY So Distribution of the Apheloriini. 158. Fic. R.M. SHELLEY & D.R. WHITEHEAD 207 they are also exhibited by Dynoria, Furcillaria, and Brachoria. I was repeatedly frustrated in searches for the subgenus Falloria in the Cumberland Plateau of Tennessee by the recovery of large numbers of red/blue forms of Brachoria and none or only a few of Falloria. For exam- ple, in a large series of red/blue forms from Rainbow Lake, Hamilton County, there were seven males of B. hubrichti Keeton and only one of S. (F.) forficata. This color pattern in Brachoria extends into Alabama ap- proximately to Birmingham, but knowledge of its complete range must await revision of this genus and more field work in Kentucky. However, the extent of the red/red pattern can be reported and is shown in Figure 159. There are two large areas separated by a narrow gap in Piedmont, Georgia. One extends from eastern Tennessee through the Carolinas to piedmont and coastal Georgia, with a branch swinging northward into southeastern North Carolina, and the other begins in central and western Georgia, curves into southern Alabama, widens in a northerly direction to the northwestern corner and adjacent Mississippi, then angles to the Mississippi River. The actual northern and western limits in Alabama and Mississippi are uncertain because of inadequate knowledge of Brachoria, but from my limited per- sonal experience these colors are displayed by forms in northern and central Alabama and northern and southwestern Mississippi. The area is extended to Baton Rouge, Louisiana, as Brachoria is now known to occur there (see Brachoria account) and these were the colors of a female I collected in 1980 in Claiborne County, Mississippi, along the Mississippi River between Vicksburg and Natchez, about 100 miles north of Baton Rouge. Genus APHELORIA Chamberlin Apheloria Chamberlin, 1921:232 Leptocircus Attems, 1931:67, preoccupied by Leptocircus Swainson, 1833; (Jeekel 1971). Type species. — Of Apheloria, Fontaria montana Bollman, 1887, by original designation; of Leptocircus, L. inexpectatus Attems, 1931, by original designation. Diagnosis (adapted from Hoffman 1978a). — Telopodite set subter- minally on coxa, latter projecting distad in situ beyond base of prefemoral region; acropodite with torsion, in configuration of tight coil of around 340°, bent abruptly dorsad apically. Ecology. — The species of Apheloria appear to have less specific habitat requirements than most of those of Sigmoria s. lat. and are somewhat MEM. AMER. ENT. SOC., 35 208 XYSTODESMID MILLIPEDS similar to /atior (Sigmoria) in being ecological generalists. They occur under logs, rocks, and leaves in mixed deciduous forests as well as in many cove and stream environments. Thus in the Blue Ridge, Ridge and Valley, and Cumberland Plateau Provinces, one is more likely to encounter Apheloria on wooded hillsides than any other apheloriine genus, and it is often syn- topic with the rhysodesmine genus Cherokia. With the exception of J/atior, Apheloria is encountered in disturbed and urban habitats to a much greater degree than any other member of the tribe. Distribution. — The northern 2/3 of eastern North America from a line extending from central New York through southern Ontario, southern Wis- consin, and northern Iowa south to central Arkansas, northern Alabama, and northern South Carolina; longitudinally from along the Atlantic Ocean from southeastern North Carolina to central Connecticut west to eastern Oklahoma and southeastern Nebraska. A recent collection has established its presence on Bald Head Island off the coast of North Carolina. Species. — Exact total unknown. Hoffman (1978a) listed 14 names that were based on specimens having gonopods identical or similar to those of montana and indicated that most were synonyms or at best had subspecific rank. However, another name has since been added to the list, as I reported (1980b) that Julus virginiensis Drury, 1770, was based on a form of oS ~ SSS BS a cotacetatstutccete SRO QOS ie x5 LEER ERK sonecocens $ sag iegoceees S555 ecesttetuts esteent SSE SRR ES SRK KREIS SOROS 500,04 0.00.00) x OOOO) BOOP estates SRR RRS SRK SS x SRR OS HOSS SRR KREIS “S RRR HSCS ox Pm” <> SKF CIS OPO SOR RKRP RS $ ‘ sees erate onananecerecoranetetecetenststsi oes S S05 ones S55 KS r] o <2 $5 °, SG 5 wotes O 9% °, KS KS Fic. 159. Approximate distribution of the red paranota/red metatergal stripe color pattern in the Apheloriini. eee R.M. SHELLEY & D.R. WHITEHEAD 209 Apheloria. It is obviously much older than the other names and thus has taxonomic priority. Remarks. — Since it is the oldest name in the tribe, taxonomic decisions in the Apheloriini may need modification when Apheloria is revised. However, I think it and Brachoria comprise separate mosaic complexes which overlie each other and Sigmoria s. lat. to varying degrees and thus are generically distinct. The species of Apheloria also have different ecological properties, a fact that seems taxonomically significant. Another factor that may warrant consideration is the size, shape, and variation of the process of the 4th sternum in males. As shown in Table 2, it varies greatly in Sigmoria s. lat. and knowledge of its condition in Apheloria and Brachoria may clarify affinities among the genera. As a final note, I suggest that Fontaria luminosa Kenyon (1893) is refer- rable to Apheloria. Chamberlin and Hoffman (1958) listed the binomial under ‘‘uncertain generic position’’, and its identity has been an enigma. The type specimen(s) are lost, and the type locality is Omaha, Douglas County, Nebraska. According to Kenyon, the male genitalia curve ‘‘in- ward, forward, outward, and downward,’’ a characterization applicable to Apheloria and how a non-specialist might describe the circular acropodite when viewed in situ. In August 1983 I found a xystodesmid sample from Omaha in the NMNH consisting of one male and three females labeled, “‘Fontaria luminosa (Type?).’’ Unfortunately, the gonopods of the male were missing, but from knowledge of xystodesmid distributions, only three species are potential inhabitants of eastern Nebraska — Pleuroloma flavipes Rafinesque (Shelley 1980b), Semionellus placidus (Wood) (Chamberlin and Hoffman 1958), and a form of Apheloria (Chamberlin and Hoffman 1958, plus knowledge I have derived from museum holdings). The specimens are not P. flavipes because they lack the diagnostic sternal lobes (Shelley 1980b), and they are much too large to be Semionellus placidus. By elimination they must be a form of Apheloria if the locality is correct, and the question mark concerned the designation as type, not the place of collection. I think Omaha must be considered accurate, and assign- ment to Apheloria is compatible with Kenyon’s description. The matter can be finalized by collection of an adult male in the general vicinity of Omaha or Lincoln, where Kenyon (1893) reported a sight record. With the occur- rence of Apheloria in eastern Nebraska now fairly certain, its discovery can be reasonably predicted in riparian habitats along the Missouri River in southeastern South Dakota. The name, /uminosa, also has priority for midwestern forms of Apheloria, being 38 years older than Attems’ name, inexpectatus. MEM. AMER. ENT. SOC., 35 210 XYSTODESMID MILLIPEDS Genus BRACHORIA Chamberlin Brachoria Chamberlin, 1939:3. Tucoria Chamberlin, 1943b:17. Anfractogon Hoffman, 1948:94. Type species. — Of Brachoria, B. initialis Chamberlin, 1939, by original designation; of Tucoria, Fontaria kentuckiana Causey, 1942, by original designation; of Anfractogon, A. tenebrans Hoffman, 1948, by original designation. Diagnosis. — Acropodite with torsion, bisected at various positions by distinct cingulum. Ecology. — I have collected forms referrable to Brachoria in a variety of habitats. In western North Carolina they can be found in mixed deciduous forest and cove environments. In the Cumberland Plateau of Tennessee and Alabama they occur in the latter or along streams in the most moist spots available. Farther south in Alabama and in Mississippi, they occur only in deciduous environments along streams. Distribution. — As shown by Keeton (1959) and Hoffman (1971), Brachoria extends in a northeast-southeast direction from a center of abun- dance in the Cumberland Plateau Province. It is known as far north as Fayette County, Pennsylvania (Carnegie Museum of Natural History), and southeastern Indiana; as far east as the Blue Ridge Front in McDowell County, North Carolina (Shelley 1979b); and as far south as Lee County, Alabama, and East Baton Rouge Parish, Louisiana (FSCA). Species. — Twenty nine are now recognized, one with three subspecies. Remarks. — | indicated (1979b) that a second revision of Brachoria was needed to consider the nomenclature changes since Keeton’s revision (1959) and include newly collected material. Such a study is as important as a revi- sion of Apheloria, and Brachoria could be assigned top priority because it has more species. Although defined by the acropodal cingulum, presumably an autapomorphy, I am not certain Brachoria is monophyletic because the location of the cingulum varies, suggesting independent evolution from two or more ancestral stocks. Thus, a synonym may have to be revived or a new genus proposed to accommodate some forms. These are matters that should be addressed in a second revision. Genus DYNORIA Chamberlin Dynoria Chamberlin, 1939:7. Type species. — D. icana, Chamberlin, 1939, by original designation. R.M. SHELLEY & D.R. WHITEHEAD 211 Diagnosis. — Acropodite without torsion, prostatic groove running en- tirely along medial surface; divided at 2/3-3/4 length into lateral tibial proc- ess and medial solenomerite branch. Ecology. — My only field encounter with D. icana was in Oconee Coun- ty, South Carolina, where I found it under a thick layer of leaves in a slight depression in an oak-hickory forest near Hartwell Reservoir (Shelley 1984b). It also occurs on the edge of the Blue Ridge Province, but there are no ecological notes on collecting labels. In Georgia, D. medialis is found in typical piedmont habitat, under thin layers of leaves on relatively hard substrates near water sources. Distribution. — Southeastern extremity of the Blue Ridge Province in North Carolina and Georgia, extending onto the Piedmont Plateau of western South Carolina, and the Piedmont and Coastal Plain of central and southwestern Georgia (Shelley 1984b). Species. — Two, one in each of the above described regions. Remarks. — Dynoria is the only apheloriine genus lacking torsion, and along with Furcillaria, one of only two with a distally divided acropodite. I indicated (1984b) that they comprise a separate evolutionary branch in the Apheloriini, but as shown in Figure 160, they actually represent distinct lineages because torsion constitutes a synapomorphy between Furcillaria and the other tribal components. I also proposed a dispersal mechanism for evolution of Dynoria and Furcillaria, but this scheme is inoperative. The animals are not sufficiently vagile for dispersal to be more than a minor fac- tor in their evolution; instead, vicariance is operative. Genus FURCILLARIA Shelley Furcillaria Shelley, 1981c:953-95S. Type species. — F. aequalis Shelley, 1981, by original designation. Diagnosis. — Acropodite with torsion, prostatic groove crossing to lateral surface near 1/3 length; divided at 2/3-3/4 length into lateral tibial process and medial solenomerite branch. Ecology. — The species of Furcillaria occur under thin layers of leaves on relatively hard substrates near water sources. Distribution. — A triangular area in the Piedmont Plateau of South Carolina between the Pacolet and Savannah Rivers, extending to the western tip of the state in Oconee County. The species are allopatric and oc- cupy narrow bands within the generic range. Species. — Three. MEM. AMER. ENT. SOC., 35 212 XYSTODESMID MILLIPEDS ParT VIII. RELATIONSHIPS IN THE TRIBE APHELORIINI By Rowland M. Shelley and Donald R. Whitehead Shelley (1981a) deferred discussion of generic affinities of Sigmoria until taxonomic knowledge of the other apheloriine genera could be brought to comparable levels. This has since been done for all except Apheloria and Brachoria, so we now provide preliminary assessments of relationships among the genera and eastern xystodesmid tribes. Important recent works on the Rhysodesmini and Pachydesmini are those by Shelley on Pleuroloma (1980b) and Dicellarius (1984c). None has appeared on the Nannariini, but Hoffman (1964) presented a modern tribal diagnosis. Out-groups used to infer plesiomorphic and apomorphic character states in the Apheloriini are the three other eastern xystodesmid tribes: Rhysodesmini, Pachydesmini, and Nannariini. For tribal comparisons, Rhysodesmini serves as the out-group; although well represented in the east, it also occurs from southern Texas and New Mexico to El Salvador, where it is the sole xystodesmid tribe. The only eastern species not included in these comparisons is Semionellus placidus, which is not a member of the endemic eastern tribal complex of Nannariini, Pachydesmini, and Apheloriini. A member of the Chonaphini and related to species in the Pacific Northwest (Hoffman 1979), it has a much more slender form than do other eastern xystodesmids, and its gonopods are divergent in having the acropodite sim- ple and the prefemoral process strongly modified. The Generic Hypothesis. — The following character statements docu- ment the generic hypothesis in figure 160: Apheloriini = Dynoria + (Fur- cillaria + (Apheloria + Brachoria + (Deltotaria + Sigmoria))). 1. Presence or absence of gonopodal torsion. In all apheloriine genera except Dynoria, the acropodite is twisted at 1/3 to 1/2 length, causing the prostatic groove to cross from the medial to the lateral surfaces. This is true even for the planca group of Sigmoria, wherein the structure is basically curvilinear. Conversely, all pachydesmines except Dicellarius okefenokensis and all rhysodesmines and nannariines lack torsion, and the entire course of the prostatic groove or nearly so is visible in medial view. Presence of tor- sion is therefore considered autapomorphous. 2. Acropodite divided or undivided. The acropodite is divided in Dynoria, Furcillaria, the Pachydesmini, and many taxa of the Rhysodesmini and Nannariini. An undivided acropodite is autapomor- phous in other Apheloriini. An autapomorphous, lateral solenomerite in the haerens and planca groups of Sigmoria is not homologous with the apically divided acropodite. R.M. SHELLEY & D.R. WHITEHEAD 213 From this point on, we have found no clear synapomorphies to specify sister group relationships. The following are generic features. 3. Acropodal ground plan circular or “‘sigmoid.’’ The acropodite in Apheloria forms a tight, nearly circular coil. The acropodites of Furcillaria, Brachoria, Deltotaria, and most species of Sigmoria are basically “‘sigmoid.’’ The nearly circular loops of Sigmoria ainsliei, aphelorioides, and trimaculata differ from those of Apheloria and are considered con- vergent. The only other known xystodesmid with coiled acropodites is the oriental genus Parafontaria, placed because of other characters in a separate subfamily by Hoffman (1979). The coiled acropodite of Apheloria is apomorphous if the genus belongs to a group excluding Furcillaria as specified by character 2 and if the ‘‘sigmoid’”’ acropodite of Furcillaria is homologous to that of Brachoria, Deltotaria, and Sigmoria. Apheloria may be sister to Brachoria + (Deltotaria + Sigmoria), but we found no synapo- morphy for them. Deltotaria Ss — = @ = o << Cs = oO = >_> [—} Sigmoria 160 Fic. 160. Generic relationships in the Apheloriini. The numbers refer to apomorphies dis- cussed in the text. MEM. AMER. ENT. SOC., 35 214 XYSTODESMID MILLIPEDS 4. Presence or absence of a gonopodal cingulum. A cingulum, or trans- verse depression on the outer surface of the acropodite stem, is unique to Brachoria and is considered autapomorphous pending review of this genus. The acropodite is set off from the prefemur by a cingulum in the rhysodesmine genus Cherokia (Hoffman 1960), but we know of no other xystodesmid genus with one on the acropodite blade in each component species. 5. Presence or absence of a coxal apophysis. An apophysis, or ventrally directed projection on the gonopodal coxa, is independently apomorphous in Deltotaria, Pachydesmus, and Xystodesmus (Hoffman 1956b, 1958b; Shelley 1984c). Deltotaria is considered sister to Sigmoria, partly from overall geography and partly from geographic congruence in general gonopodal form between members of each genus. We are aware of no autapomorphy for Sigmoria or for Sigmoria + Deltotaria. Apheloria, Brachoria, Deltotaria, and Sigmoria are partly sympatric, overlying mosaics. If each is viewed as having geographic properties of a species and thus originally allopatric, Apheloria may be considered northern, Brachoria western, Deltotaria southern, and Sigmoria eastern. Aside from congruence in gonopodal features, Deltotaria fits best with Sigmoria geographically. The Tribal Hypothesis. —.The following character statements document the tribal hypothesis in figure 161, eastern Xystodesmidae = Rhysodesmini + (Nannariini + (Pachydesmini + Apheloriini)). 6. Retention or loss of a sternal remnant between the gonopodal coxae. Gonopods are specialized ambulatory appendages, primitively joined by a sclerotized sternum. Retention of a sternal remnant is one of the characteristics of the Rhysodesmini (Hoffman 1960). Sclerotization is replaced by membrane in the other three tribes, an autapomorphous condi- tion. 7. Condition of pregonopodal tarsal claws in males. The Nannariini is characterized by apomorphous twisted, spatulate, pregonopodal tarsal claws in males. In other genera the claws are bisinuately curved or uncinate, the plesiomorphic condition. 8. Presence or absence of lateral subcoxal spines on postgonopodal ster- na of males and midbody sterna of females. These spines or lobes are found in the Nannariini and most rhysodesmine genera. In the Pachydesmini and Apheloriini, the caudal sternal edge is straight except for segments 8-10 where bicruciform grooves create a lobed appearance in some species. Absence of subcoxal spines or lobes is tentatively considered to be apomor- phous. R.M. SHELLEY & D.R. WHITEHEAD 215 9. Body size. The Nannariini and many of the eastern Rhysodesmini tend to be relatively small, whereas most species of Pachydesmini and Apheloriini are large (Dicellarius okefenokensis, Sigmoria sigirioides, and Rhysodesmini Nannarlini Pachydesmini Apheloriini ] 8.9 : 16] Fic. 161. Relationships among the four major xystodesmid tribes in eastern North America. The numbers refer to apomorphies discussed in the text. MEM. AMER. ENT. SOC., 35 216 XYSTODESMID MILLIPEDS S. truncata being exceptionally small). Large body size thus appears to be apomorphous and in combination with characters 7 and 8 joins the tribes Pachydesmini and Apheloriini as a monophyletic group. 10. Condition of the prefemoral process. Except for occasional in- dividuals of the pachydesmine species Dicellarius okefenokensis, the prefemoral process is long, slender, and acicular in the Rhysodesmini, Nan- nariini, and Pachydesmini and therefore is considered plesiomorphous. In the Apheloriini, it is present or absent, but never acicular. If monophyly in the Apheloriini is based on nonacicular prefemoral processes, we lack an autapomorphy for the Pachydesmini. Other known structural distinctions between the Pachydesmini and Apheloriini all in- volve exceptions, and tribal distinction may be unwarranted. Placement of Dicellarius and Thrinaxoria in the Pachydesmini by Shelley (1984c) was based on the similarity in character 10. The acropodite is divided in the Pachydesmini and undivided in the Apheloriini except in Dynoria and Fur- cillaria. All pachydesmines except D. okefenokensis lack acropodal torsion, and all apheloriines except Dynoria have it. Likewise, the acropodite is linear in all pachydesmines except D. okefenokensis, and curved or ‘*sigmoid”’ in all apheloriines except Dynoria (secondarily curvilinear in the planca group of Sigmoria). Among genera of the Pachydesmini, Pachydesmus has coxal apophyses (convergent with those of Deltotaria in the Apheloriini) and large postgonopodal sternal elevations or podosterna (Hoffman 1958b), both conditions being autapomorphous. Dicellarius and Thrinaxoria have reduced cyphopodal receptacles, also autapomorphous. We have found no autapomorphous feature for all three genera, and this plus the exceptions regarding divided acropodites and torsion suggest that the two tribes should be combined. However, colors and color patterns may be as useful for characterizing tribes as they are for species within Sigmoria, but we have insufficient out-group information to assess polarity. The basic color pattern in the Apheloriini is a black body with bold paranotal spots and metatergal stripes. In contrast, the color pattern is generally pale among the Pachydesmini. Perhaps this lack of pigmentation in the Pachydesmini represents the missing autapomorphy. Discussion. — These are not character rich hypotheses for either tribal or generic relationships. Other traits, such as facial chaetotaxy and meristic characters of the head and body, have not been analyzed sufficiently for in- terpolation. Our comments on Sigmoria as a cohesive generic mosaic with a geographic structure virtually that of a single species led to speculation that the genus might be recent. It follows that much of the eastern xystodesmid fauna also may be recent. The endemic tribes Nannariini, Pachydesmini, ————— eee" R.M. SHELLEY & D.R. WHITEHEAD 217 and Apheloriini, respectively with 2, 3, and 6 genera, form a monophyletic group. The putative sister group, Rhysodesmini, is presumed monophyletic, but we think it probably is not. Points of interest are these: 1. There is very little somatic differentiation among the endemic tribes. Distinctions between them and their genera and species are primarily gonopodal. Occasionally, species with very different gonopods are linked by intermediates (as in the Sigmoria mimetica group, for instance), so that gonopodal differences may be less significant than thought. Lack of somatic differentiation is consistent with suggested lack of age. 2. Failure to generate satisfactory cladistic hypotheses at the generic and tribal levels implies lack of extinctions needed to create structural gaps for definitions of lineages. Lineages are ill defined because of retention of ancestral features. Lack of profound structural gaps also is consistent with suggested lack of age. 3. We have reservations about the tribal classification proposed by Hoff- man (1979), but the information at hand is insufficient to effect changes. Certainly, the distinction between the Pachydesmini and the Apheloriini does not appear to withstand scrutiny, and we do not think the picture among the entire eastern xystodesmid fauna is yet clear enough to determine where tribal limits should be set. Again, the absence of intensive analyses of Apheloria and Brachoria is a major gap in our knowledge, but comprehen- sive study is also needed in the Rhysodesmini. This tribe is both somatically and gonopodally diverse, and may need division. Neither Hoffman (1978c) nor Shelley (1980b) could adequately relate Caralinda to other rhysodesmines, and two recently discovered, small-bodied southeastern genera, Parvulodesmus and Gonoessa (Shelley 1983b, 1984d), are unique in having extremely long acropodites that overlap two or more segments in situ. We cite large body size as apomorphous for Apheloriini plus Pachydesmini, but small body size applies to the Nannariini and an increas- ing number of Rhysodesmini even though the largest xystodesmid of all, Rhysodesmus dasypus (Gervais), in Veracruz, Mexico, belongs to this tribe. Perhaps some of the eastern North American rhysodesmines combine with the other eastern tribes to form a monophyletic group. We can only speculate on this, but it becomes increasingly doubtful that all of the eastern rhysodesmine genera are contribal with Rhysodesmus and Stenodesmus in Mexico and Central America. 4. In conclusion, we again emphasize the importance of color, of which two basic patterns are displayed by the eastern xystodesmid fauna. In the Apheloriini and the Rhysodesmini (except for Gyalostethus and Gonoessa, which are comparatively dull; the color is unknown for Parvulodesmus and most species of Caralinda) the body is generally black and boldly marked by MEM. AMER. ENT. SOC., 35 218 XYSTODESMID MILLIPEDS stripes or spots. In the Pachydesmini and Nannariini, body color is neither black nor boldly marked. Instead, it is brown to olive to reddish orange, with or without paler stripes. We have no phyletic interpretation; bold pat- terns may be mimetic and therefore convergent in the Rhysodesmini and Apheloriini, as appears to be true for certain species of the latter. However, the basic color patterns do apply at the tribal level, and coloration in xystodesmids is far more interesting and informative than is generally recognized. It should become a standard and integral part of future systematic studies in the family. LITERATURE CITED ATTEMS, CARL G. 1931. Die Familie Leptodesmidae und andere Polydesmiden. Zoologica, Stuttgart, 30:1-149. 1938. Polydesmoidea II, Fams., Leptodesmidae, Platyrhacidae, Oxydes- midae, Gomphodesmidae. Das Tierreich, Lief., 69:1-487. BOLLMAN, CHARLES H. 1888. Notes upon some myriapods belonging to the U.S. National Museum. Proc. U.S. Natl. Mus., //: 343-350. CAuSEY, NELL B. 1942. Six new diplopods of the family Xystodesmidae. Ent. News, 53: 165-170. 1950a. A collection of xystodesmid millipeds from Kentucky and Tennes- see. Ent. News, 6/:5-8. 1950b. Two new polydesmoid diplopods. Ent. News, 6/:37-39. 1955. New records and descriptions of polydesmoid millipeds (Order Poly- desmida) from the eastern United States. Proc. Biol. Soc. Wash., 68:21-30. CHAMBERLIN, RALPH V. 1918a. Myriapods from Nashville, Tennessee. Psyche, 25:23-30. 1918b. New polydesmoid diplopods from Tennessee and Mississippi. Psyche, 25:122-127. 1921. On some chilopods and diplopods from Knox Co., Tennessee. Can. Ent., 53:230-233. 1939. On some diplopods of the Family Fontariidae. Bull. Univ. Utah. 30(2)[Biol. Ser., 5(3)]:1-19. 1943a. Some records and descriptions of American diplopods. Proc. Biol. Soc. Wash., 56:143-152. 1943b. On some genera and species of American millipeds. Bull. Univ. Utah, 34(6)[Biol. Ser., 8(2)]:1-20. 1947. Some records and descriptions of diplopods chiefly in the collection of the Academy. Proc. Acad. Nat. Sci. Phila, 99:21-58. 1949. Some millipeds of the families Polydesmidae and Xystodesmidae. J. Wash. Acad. Sci., 39:94-102. AND RICHARD L. HOFFMAN. 1958. Checklist of the millipeds of North Amer- ica. Bull. No. 212, U.S. Natl. Mus., 236 pp. FILKA, MARIANNE E. AND ROWLAND M. SHELLEY. 1980. The milliped fauna of the Kings Mountain Region of North Carolina (Arthropoda: Diplopoda). Brimleyana, 4:1-42. Gray, JOHN E. 1832. Myriapods, in Griffith, The animal kingdom arranged in conformity with its organization by the Baron Cuvier, 7/5 (Class Insecta, 2):787. Le R.M. SHELLEY & D.R. WHITEHEAD 219 HOFFMAN, RICHARD L. 1948. Two new genera of xystodesmid millipeds from eastern United States. Proc. Biol. Soc. Wash., 6/:93-96. 1949. Nine new xystodesmid millipeds from Virginia and West Virginia, with records of established species. Proc. U.S. Natl. Mus., 99:371-389. 1950. American polydesmoid millipeds of the genus Sigmoria, with notes on distribution. Amer. Mus. Nov. No. 1462, 7 pp. 1951. Subspecies of the milliped Apheloria trimaculata (Wood) (Polydes- mida: Xystodesmidae). Chicago Acad. Sci. Nat. Hist. Misc. No. 81, 6 pp. 1956a. Revision of the milliped genus Dixioria (Polydesmida: Xystodes- midae). Proc. U.S. Natl. Mus., /06:1-19. .1956b. Studies on some oriental xystodesmine millipeds. Proc. Ent. Soc. Wash., 58:95-104. .1958a. A new milliped of the genus Sigmoria from western North Carolina (Polydesmida: Xystodesmidae). Proc. Ent. Soc. Wash., 60:281-284. 1958b. Revision of the milliped genus Pachydesmus (Polydesmida: Xys- todesmidae). Proc. U.S. Natl. Mus., /08:181-218. 1960. Revision of the milliped genus Cherokia (Polydesmida: Xystodes- midae). Proc. U.S. Natl. Mus., 112:227-264. 1961. Revision of the milliped genus Deltotaria (Polydesmida: Xystodes- midae). Proc. U.S. Natl. Mus., 1/3:15-35. 1962. A new genus and species in the diplopod family Xystodesmidae (Polydesmida). Proc. Biol. Soc. Wash., 75:181-188. 1964. The status of Fontaria pulchella Bollman, with the proposal of a new genus and tribe in the diplopod family Xystodesmidae. Proc. Biol. Soc. Wash., 77:25-34. 1966. The Mexican genera of Xystodesmidae (Diplopoda: Polydesmida). Trans. Amer. Ent. Soc., 92:1-16. 1967. Revision of the milliped genus Cleptoria (Polydesmida: Xystodes- midae). Proc. U.S. Natl. Mus., /24:1-27. 1969. The origins and affinities of the southern Appalachian diplopod fauna. pp. 221-246, In: Perry C. Holt, ed., 1969. The Distributional History of the Biota of the Southern Appalachians, Part I: Invertebrates. Research Division Monograph I, Virginia Polytechnic Institute, Blacksburg, VA, 295 pp + i-vi. 1970. Random studies on Rhysodesmus. 1. Notes and redescriptions of miscellaneous species. Radford Rev., 24:143-162. 1971. Millipeds of the genus Brachoria in Virginia (Polydesmida: Xystodes- midae). Radford Rev., 25:83-99. 1978a. Revalidation of the generic name Rudiloria Causey, 1955 (Poly- desmida: Xystodesmidae). Myriapodologica, /(J):1-6. 1978b. Chelodesmid studies. XI. A new genus and species from Venezuela referrable to the new tribe Chondrodesmini. Rev. Suisse Zool., 85:543-551. 1978c. A new genus and species of rhysodesmine milliped from southern Georgia (Polydesmida: Xystodesmidae). Proc. Biol. Soc. Wash., 91:365-373. 1979. Classification of the Diplopoda. Museum d’ Histoire Naturelle, Gen- eva, Switzerland, 237 pp. Hunt, CHARLES B. 1967. Physiography of the United States. W.H. Freeman Co., San Francisco, 480 pp. JacoT, ARTHUR P. 1938. Four new arthropods from New England. Amer. Midl. Nat., 20:571-574. MEM. AMER. ENT. SOC., 35 220 XYSTODESMID MILLIPEDS JEEKEL, C.A.W. 1971. Nomenclator generum et familiarum Diplopodorum: A list of the genus and family group names in the class Diplopoda from the 10th edition of Linnaeus, 1758, to the end 1957. Monog. Nederl. Entom. Vereng. No. 5, 412 pp. KEETON, WILLIAM T. 1959. A revision of the milliped genus Brachoria (Polydesmida: Xystodesmidae). Proc. U.S. Natl. Mus., /09:1-58. 1960. A new genus and species of xystodesmid milliped from Tennessee. Bull. Brooklyn Ent. Soc., 55:42-45. 1965. Descriptions of three new species of Brachoria, with notes on estab- lished species (Diplopoda, Polydesmida, Xystodesmidae). Proc. Biol. Soc. Wash., 78:225-240. KENYON, FREDERICK C. 1893. A preliminary list of the Myriapoda of Nebraska with descriptions of new species. Publ. Nebraska Acad. Sci., 3714-18. KEVAN, D.K. McK. 1983. A preliminary survey of known and potentially Canadian milli- pedes (Diplopoda). Can. J. Zool., 67:2956-2975. Loomis, HAROLD F. 1943. New cave and epigean millipedes of the United States, with notes on some established species. Bull. Mus. Comp. Zool., 92:373-410. 1944. Millipeds principally collected by professor V.E. Shelford in the eastern and southeastern states. Psyche, 5/:166-177. SHEAR, WILLIAM A. 1972. A new milliped in the genus Apheloria from southern West Virginia, and the taxonomic position of Rudiloria mohicana (Diplopoda: Polydesmida: Xystodesmidae). Proc. Biol. Soc. Wash., 85:493-498. SHELLEY, ROWLAND M. 1976. Miéillipeds of the Sigmoria latior complex (Polydesmida: Xystodesmidae). Proc. Biol. Soc. Wash., 89:17-38. 1977. The milliped genus Croatania (Polydesmida: Xystodesmidae). Proc. Biol. Soc. Wash., 90:302-325. 1979a. A new milliped of the genus Caralinda from north Florida (Poly- desmida: Xystodesmidae). Fla. Ent., 62:183-187. 1979b. The status of Sigmoria gracilipes Chamberlin, with remarks on the milliped genus Brachoria (Polydesmida: Xystodesmidae). Fla. Ent., 62:220-223. 1980a. The status of Cleptoria shelfordi Loomis, with the proposal of a new genus in the milliped family Xystodesmidae (Polydesmida). Brimleyana, 3:31-42. 1980b. Revision of the milliped genus Pleuroloma (Polydesmida: Xystodes- midae). Can. J. Zool., 58:129-168. 198la. Revision of the milliped genus Sigmoria (Polydesmida: Xystodes- midae). Mem. Amer. Entomol. Soc. No. 33, 140 pp. 1981b. A new milliped of the genus Brevigonus from South Carolina, with comments on the genus and B. shelfordi (Loomis) (Polydesmida: Xystodesmidae). Brimleyana, 6:51-60. 1981c. A new xystodesmid milliped genus and three new species from pied- mont South Carolina (Polydesmida). Proc. Biol. Soc. Wash., 94:949-967. 1982. A new xystodesmid milliped genus and three new species from the eastern Blue Ridge Mountains of North Carolina (Polydesmida). Proc. Biol. Soc. Wash., 95:458-477. 1983a. The status of Cleptoria divergens (Chamberlin) (Polydesmida: Xystodesmidae). Myriapodologica, /(/2):81-89. 1983b. Parvulodesmus prolixogonus, a new genus and species of xystodesmid milliped from South Carolina (Polydesmida). Proc. Biol. Soc. Wash., 96:121-126. 1984a. Revision of the milliped genus Cheiropus (Polydesmida: Xystodes- midae). Proc. Biol. Soc. Wash., 97:263-284. R.M. SHELLEY & D.R. WHITEHEAD 221 1984b. Revision of the milliped genus Dynoria (Polydesmida: Xystodes- midae). Proc. Biol. Soc. Wash., 97:90-102. 1984c. A revision of the milliped genus Dicellarius, with a revalidation of the genus Thrinaxoria (Polydesmida: Xystodesmidae). Proc. Biol. Soc. Wash., 97:473-512. 1984d. A new xystodesmid milliped genus and five new species from the Coastal Plain of Alabama (Polydesmida). Fla. Ent., 67:453-464. WHITEHEAD, DoNALD R. 1972. Classification, phylogeny, and zoogeography of Schizo- genius Putzeys (Coleoptera: Carabidae: Scaritini). Quaest. Entomol., 8:131-348. WILLIAMS, STEPHEN R. AND ROBERT A. HEFNER. 1928. The millipedes and centipedes of Ohio. Bull. Ohio Biol. Surv. No. 18, 4(3)[Ohio State Univ. Bull., 33(7)]:93-146. Woop., Horatio C. 1864. Descriptions of new species of North American Polydes- midae. Proc. Acad. Nat. Sci. Phila., /6:6-10. 1865. The Myriapoda of North America. Trans. Amer. Philosoph. Soc., 13:137-248. Wray, Davip L. 1950. Insects of North Carolina, Second Supplement. NC Dept. Agric. Raleigh, 181 pp. 1967. Insects of North Carolina, Third Supplement. NC Dept. Agric., Raleigh, 181 pp. MEM. AMER. ENT. SOC., 35 222 INDEX TO TAXA Synonyms in italics, new species in boldface abbotti, Cleptoria — 81 abbotti, Sigmoria (Cleptoria) — 81 abbreviata, Sigmoria (Falloria) — 123 acuminata, Sigmoria (Dixioria) — 50 ainsliei, Apheloria — 117 ainsliei, Sigmoria (Falloria) — 117 antrostomicola, Apheloria — 31 aphelorioides, Sigmoria (Falloria) — 111 arcuatus, Brevigonis — 94 arcuata, Sigmoria (Cleptoria) — 94 australis, Sigmoria (Cheiropus) — 95 austrimontis, Sigmoria (Sigiria) - 102 bipraesidens, Cleptoria — 84 bipraesidens, Sigmoria (Cleptoria) — 84 Brevigonus — 76 brimleardia, Deltotaria — 151 brimleii, Deltotaria — 151 brimleii brimleii, Deltotaria — 151 brimleii philia, Deltotaria — 155 brooksi, Dixioria — 59 brooksi, Sigmoria (Dixioria) — 59 catawba, Croatania — 69 catawba, Sigmoria (Croatania) — 69 Cheiropus — 94 (Cheiropus) Sigmoria — 94 Cleptoria — 76 (Cleptoria) Sigmoria — 76 coriacea, Apheloria — 31 coronata, Deltotaria — 52 coronata, Sigmoria (Dixioria) — 52 crassicurvosa, Sigmoria (Falloria) — 138 Croatania — 68 (Croatania) Sigmoria — 68 dactylifera, Dixioria — 63 dactylifera, Sigmoria (Dixioria) — 63 Deltotaria — 147 dentifer, Dixioria — 46 Dixioria — 45 (Dixioria) Sigmoria — 45 Falloria — 106 (Falloria) Sigmoria — 106 Fontaria — 94 forficata, Sigmoria (Falloria) — 119 guyandotta, Apheloria — 35 guyandotta, Sigmoria (Rudiloria) — 35 houstoni, Sigmoria — 126 houstoni, Sigmoria (Falloria) — 126 Hubroria — 106 inornata, Sigmoria (Sigiria) — 105 keuka, Apheloria — 31 kleinpeteri, Apheloria — 34 lea, Deltotaria — 157 lutzi, Fontaria — 31 Lyrranea — 94 macra, Cleptoria — 89 macra, Sigmoria (Cleptoria) — 89 mariana, Deltotaria — 151 mimetica, Fontaria — 133 mimetica, Sigmoria — 133 mimetica, Sigmoria (Falloria) — 133 mohicana, Apheloria — 38 mohicana, Rudiloria — 38 mohicana, Sigmoria (Rudiloria) — 38 mohincana, Rudiloria — 38 pela, Apheloria — 46 pela, Fontaria - 46 pela, Sigmoria (Dixioria) — 46 pela acuminata, Dixioria — 50 pela brooksi, Dixioria — 59 pela coronata, Dixioria — 52 pela fowleri, Dixioria — 52 pela pela, Dixioria — 46 pela wrighti, Dixioria - 55 pendulata, Sigmoria (Falloria) — 142 Phanoria — 147 Philia, Deltotaria — 155 Dhilia, Phanoria — 155 picapa, Hubroria — 130 picapa, Sigmoria (Falloria) — 130 picta, Apheloria — 31 Prionogonus — 94 prolata, Sigmoria (Falloria) — 107 a tT rigida, Sigmoria (Rudiloria) — 41 rileyi, Cleptoria — 79,89 rileyi, Fontaria — 79 rileyi, Sigmoria (Cleptoria) — 79 rileyi alabama, Cleptoria — 79 rileyi rileyi, Cleptoria — 79 robusta, Sigmoria (Cleptoria) — 86 rubromarginata austrimontis, Sigmoria — 102 Rudiloria — 25 (Rudiloria) Sigmoria — 25 saluda, Croatania — 70 saluda, Sigmoria (Croatania) — 70 shelfordi, Brevigonus — 92 shelfordi, Cleptoria — 92 shelfordi, Sigmoria (Cleptoria) — 92 Sigiria — 101 (Sigiria) Sigmoria — 101 Sigmoria — 14 (Sigmoria) Sigmoria — 66 simplex, Croatania — 72 simplex, Sigmoria — 105 simplex, Sigmoria (Croatania) — 72 Stelgipus — 94 MEM. AMER. ENT. SOC., 35 223 tela, Deltotaria — 151 tortua, Apheloria — 31 trimaculata, Apheloria — 31,34 trimaculata, Fontaria — 38 trimaculata, Polydesmus (Fontaria) — 31 trimaculata, Rudiloria — 31 trimaculata, Sigmoria (Rudiloria) — 30 trimaculata antrostomicola, Apheloria — 31 trimaculata incarnata, Apheloria — 31 trimaculata kleinpeteri, Sigmoria (Rudiloria) — 34 trimaculata tortua, Apheloria — 31 trimaculata trimaculata, Apheloria — 31 trimaculata trimaculata, Apheloria (Rudiloria) — 31 trimaculata trimaculata, Sigmoria (Rudiloria) — 31 watauga, Sigmoria (Dixioria) — 58 whiteheadi, Sigmoria (Sigiria) — 102 wrighti, Sigmoria (Dixioria) — 55 yemassee, Croatania — 75 yemassee, Sigmoria (Croatania) — 75 or) a® sh yolony M4 alt li *% 7 T ate neces or THE AMERICAN Seaensaci: a ‘The Cresson Types of Hymenoptera. Ezra T. Cresson. ‘1916. $12.00. A Venational Study of the Suborder Zygoptera (Odonatay, with Keys for the Identifica- tion of Genera, Philip A, Munz. Munz. a 1949, $10.00, . The Blattidae of Panama, Morgan Hebard, 1920, $8.00. : The Types of Hymenoptera in the Academy of Natural Sciences of Pia other ‘than those of Ezra T, Cresson. Ezra T. Cresson. 1928. $8.00. ub The Eumastacinae of Southern Mexico and Central America. tok AG. Rehn and John W.H,. Rehn. 1934, $8.00. The Generic Names of the sncone Taide ond Their iia Species, W.S.L. Pate. ee. $8.00. ; Huckett. 1941, “$6.00. é a s and Reclassification of ‘the Nearctic Ichneumonidae, Henry kK. Townes, Se 1944 00. _ Blachistidae of North America (Microlepidoptere). Annette F. Braun, 1948, $1000 Classification of the Blattaria as Indicated by their Wings (Orthoptera). John W.H Rehn, 1951. $10.00. . A Taxonomic Study of the North American Licinini with Notes on the Old World Species of the Genus Diplocheila Brulle (Coleoptera). George E. Ball. 1959. $15.00. Scarabaeidae). ; . A Monograph of the Ithomiidae (Lepidoptera) Part Ill, The ‘Tribe Mechanitini Fox. | "The Shield-backed Katydids of the Genus Idiostatus, David C. Rentz, ; . The Systematic and Morphology of the Nearctic Species of Dees * Le Me oe pee - (Diptera: Chironomidae). Dean Cyrus Hansen and Edwin F. Cook. “1976, $10.00, note ee ee ees aie : ad of the ete Genus Sigmoria (Polydesmida: Xystod . Proceedings of he ms International Symposium on Ch . A reconsideration of the Milliped Genus Sigmoria, with a revision of J Deltor _M, Shelley and David R. Whitehead, 1985. $16.00 . A Taxonomic Study of the Milliped Family Spirobolidae (Diplopoda: Speen. ‘William T. Keeton, 1960. $15.00, — . The Genus Bucculatrix in- America North of Markco (Microlepidoptera). ‘Annette Fr. Braun. 1964, $15.00, > : ' The Butterflies of Libera. Richard M. Fox, Arthur W. Lindsey, Jn . Harry K. Clench — & and Lee D, Miller, 1965. $12.50. S --Pee . A Revision of the Mexican and Central American Spider Wasps of the Subfamily Pom- Le < jpilisaé (Hymenoptora: Pompllidee), Howard E. Evans, 1966. $12.50. sf a . A Taxonomic and Zoogeographic Survey of the Scarabacinae of the Antilles (Coleoptera: ais ‘Eric G. Matthews, 1966. $5.00 Oran Richard M, Fox, 1967. $9.00, ae . A List of New North American Spiders, 1940-1966. Beatrice R, Vogel, 1967, $9.00. The Higher Classification, Phylogeny and Zoogeography of the Satyridae (Lepidoptera). Lee D. Miller, 1968, $7.00, a = i 7 : . The Schizopteridae (Hemiptera: Heteroptera) With the Description of New Species from ee 0 Trinidad, Michael G, Emsley, 1969, $6.50, ‘ ey: _ A Taxonomic Revision of the Aquatic Beetle Genus Laccophilus (Dytiseidae) of North pS ey ‘America, James: R. Zimmerman. 1970, $12.00, in * A Revision of the Nearctic Species of the Tribe Parydriai (Diptera: Ephydridae). Philip . z Clausen and Edwin F, Cook. 1971, $7.00. (Tote oss Tischeriidae of America, North of Mexico (Microlepidoptera). Annetic F. _ Oe aly 1972. $7.00, ‘The Stoneflies (Plecoptera) of the Rocky Mountains, Richard W. Ns ah eal Gaufin and Rebecca F. Surdick, 1977, $10.50. The Genus Isoperla (Plecoptera) of Western North America; Holomorph Systematics, and a New Stonefly Genus Cascadoperla, Stanley W. alia: W, Stewart, 1979, $7.50, Shelley, 1981, Florida, July 28-28, 1982, 1983. $25.00. “analysis of the genera in the tribe Apheloriini (Polydesmida; ace In making inquiries relative to publications, address: | THE AMERICAN ENTOMOLOGICAL SOCIETY | 1900 Race Street ana, PA 19103 Ps ee v iy ; e rh if oi hikns ata tk he ‘i S lear ‘\ - ‘ ik + us ‘o : , F 5 i“ +4) ¥ * ee « ‘ a t mm ne Te LINS SaIuVua 17 BRARIES_ SMITHSONIAN _INSTITUTION ANOILNLILSNI_NVINOSHLIP cS ee fo) —_ P = S w = o 2 o = = x0 - a =). =) a x = > ey > = = is Pa mies a = eo) or = 2 m Z m Sirs AB = wn = on & NIAN INSTITUTION NOLMNIIESNE™ yet ldW¥d!T_LIBRARIES SMITHSONIAN z Xe: 2 < = x NE = aa : = z 2 Gh fp &® WN. 2 Om 3 5 2 EZH ANE 3 = z Ee 2° li = NY z = ES = > a ae wo a > 7p) z wo : - = , LINS S3INYVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIW Ww = wo = no = 6, 8 & uw «Go wl & aS a a a a = < } a < = . = S a = e = . _ = faa) — ae ° Se ae 5 = bg —| NIAN_ INSTITUTION NOILALILSNI_NVINOSHLINS S3IYVYGIT LIBRARIES i= = Tears rs = = , = o) 2) — Ly, 2 = 3 = a = Vee = 2 = > F Te E S ie a - a Z O z he 2 LIWS Saluvugi7 LIBRARI ES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLIM w x : £ ; 3 = Z 2 vz z a = z = 3 2 Vy. 5 i 2 g 2 8 Yh? z : : : 2.0K E > > = 7p) = 7) Ze 7) NIAN INSTITUTION = NMINCSHITING © _LIBRARIES SMITHSONIA we E a 2 2 2 L, = _ = Lp =a ing = Qs o a a YY pe. od W fas = x Cc % om <= ID US fac =7 RRS load (= o ete co 5 WY = a sap S = z ony 5 = a lINS SalyuVv¥a yl BRARIES _ SMITHSONIAN _INSTITUTION =NOILALILSNI _NVINOSHLIW = ae —_ : —_ . a av bs) \ = > = s > = Sain ne F z B 2 a IAN NOILALILSNI NVINOSHLINS S31YWYAIT_ LIBRARIES SMITHSONIA Z g z g Ge | << ass ite Ak = fe) = 5 = this a. \\ a : S : 9 Yigg = WW 8 LINS SJIYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLIV Tp) — w = (7p) = se ae & WwW 7 Ww 2 «. G > x 4 ae Z nf Ay Be a SS + SS 3 \ = =F Vf. Y = = a4 WS \ S « Uy 5 MY EI ya S a “ip 5 2 S ‘= See) = ki 2 4 z ° et NIAN” INSTITUTION ~ NOILMLILSNINVINOSHLINS. S31YVUGI7_ LIBRARIES SMITHSONIAN INS i z ee NS = tan S & uy ‘ y = fo) = = i c hy wo = o 3 = wo ad YY wD X ly, : = Ye 5 : 5 £43 fi? = = Ws e = = Jif? E C4, 2 —~ D WS E x = YF his oe bY te w er \s 7) = ae ae - } 4 z 4A Sal cs La Z ao a \ 7 LIBRARIES SMITHSONIAN CC ey S Wash) = e NO! NO! INSTITUTION Sa1uvugdia LIB INSTITUTION SMITHSONIAN NVINOSHLIWS SMITHSONIAN LIBRARIES SMITHSONIAN _INSTITUTION NOILNLILSNI LIBRARIES ; NOILNLILSNI_NVINOSHLINS S3I1YVUdit 2 ~ SAS1uYVYuYdIT_LIBRARIES ith fp gy INSTITUTION Saiyvudii NOILNLILSN! NVINOSHLINS S31YVYaII a 2) re = = /s z 4 ie ie) Be oO a Yo 5 ”n = O ¥ = Ee 2 E = > = 2) Fas w LIBRARIES SMITHSONIAN INSTITUTION = (7p) = (a9) 3 ul 1a a ay ©. aa fea 2 < 5; = =< = x GY: Ss oc _— ma os rua] fe —_ Ge: (eo) = 4 oud me =f N NOILALILSNI NVINOSHLINS S3SIYVYUSIT iz Bits z c = wo = wo iS a = eo) - es = a fw im a” ra INSTITUTION NOILNLILSNI NOILNLILSNI NVINOSHLINS S31YVYEIT LIBRARIES LIBRARIES SMITHSONIAN INSTITUTION ~~ n Sek n = < = e e Ng : AW 2 2 WAN & YS Zz E WW" Z. WS E e OW: NOILALILSNI_NVINOSHLIWS Sa 1yVveaiT_ B. uw % = =| = SS: 5 $ xy ye we (= = SS NS co = 2 eS _ fe) — =a LIBRARIES SMITHSONIAN _ INSTITUTION 5 = 5 = ‘oO oa 2) ch 5 Ee E = e caus _ ow m (¢p) Z D Z eno © Saas os oS) 7 = NVINOSHLINS SaI1uYVuGIT LIBRA < ° = s \\ -; = 0] > ra > = = es 22) yi Wc o SMITHSONIAN INSTITUTION NOILN AS = z = NESS F fo) Runes NS Zz EE z > = > pees no ie nla F NOILALILSNI_NVINOSHLIWS S31YWVYaI1_LIBRS = Sea Aw = = = = fe) ~ fe) z Se ive Fe LIBRARIES_ SMITHSONIAN INSTITUTION _ NOILN = ly 2 EO 2 2 n* m no Sie 7) = NOILALILSNI_ NVINOSHLIWS, Sa 1yVydII art BRI < = y < »® z 4 z \ ENG) a5 2) . ! = 2 Ey : Ss >" = te n za ” LIBRARIES SMITHSONIAN INSTITUTION eNeity a x uw = rm a = jaa < = oc Ss = = NOILALILSNI_NVINOSHLIWS Saidved!1_ LIBR. é z a w = o a 2) = = > A = a Be P te yA ya o LIBRARIES SMITHSONIAN euiive z g NY 5 2 = WW as 2 = . N e) x O » 2 = Z baie = > 2 7) Zz NOILNLILSNI NVINOSHLIWS it x AWN fe = 2M = = A NN a = fe) 6S e) ra Fe ) Fa LIBRARIES SMITHSONIAN INSTITUTION _ NOIL( fe) ne fe) S LE : - YE fii > = = “i Zz = n : m ” Ry Wend 4 Ha ense e | (ey eee POU AORaS a